commit cb0bc64c16b3a38cbf0c622830c18ac9ea6e2ffe Author: Chris Wilson Date: Mon Sep 6 17:15:51 2010 +0100 version: 1.10.0 *release* /me faints with excitement. cairo-version.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 87fad104645c481ab027afbc9fa45c54cacbc9a7 Author: Chris Wilson Date: Mon Sep 6 17:14:41 2010 +0100 NEWS: Sum up 2 years of development for 1.10.0 in 60 lines NEWS | 61 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 61 insertions(+) commit 04623ef9df7274d411a3b7f03129bf43e53e7086 Author: Chris Wilson Date: Mon Sep 6 13:15:55 2010 +0100 test: Add subsurface-outside-target refs to REFERENCE_IMAGES test/Makefile.am | 2 ++ 1 file changed, 2 insertions(+) commit d5b2c2a12ebea33f4d108d4deec2e9111c8c7e12 Author: Chris Wilson Date: Mon Sep 6 13:10:08 2010 +0100 test: Rename clip-fill-*-unbounded.image16 ref to match REFERENCE_IMAGES ...rgb24.ref.png => clip-fill-eo-unbounded.image16.ref.png} | Bin ...rgb24.ref.png => clip-fill-nz-unbounded.image16.ref.png} | Bin 2 files changed, 0 insertions(+), 0 deletions(-) commit cab9682f9dc4c99dcd51ddb66abd756e2ac7a63c Author: Chris Wilson Date: Mon Sep 6 12:24:47 2010 +0100 check: Ignore the documentation errors. Currently it is complaining about having documentation for unconfigured sections, ignore it for now. doc/public/check-doc-coverage.sh | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) commit 8127f1f5d52bb716dbd70efe096894f2a542f4a9 Author: Chris Wilson Date: Mon Sep 6 12:24:14 2010 +0100 Bump pixman requirements to 0.18.4 for assorted fixes. README | 2 +- configure.ac | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) commit 21a2f90cdef3d592c1d1614852fe3929ed2aa765 Author: Chris Wilson Date: Thu Sep 2 13:19:26 2010 +0100 check: Make the actual error stand out doc/public/check-doc-coverage.sh | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) commit 8258532794d0579f6c1b825b8cfb9c8f7ced4369 Author: Chris Wilson Date: Thu Sep 2 13:15:07 2010 +0100 Reduce the cairo_device_type_t to only hold the used entries. This should help clarify the documentation by not mentioning non-existent devices and prevent any assumed coupling between surface-type values and devices. src/cairo.h | 49 ++++++-------------------------- util/cairo-gobject/cairo-gobject-enums.c | 24 +++------------- 2 files changed, 12 insertions(+), 61 deletions(-) commit e3afbc1875d618c6d7e893971372be8f9889fc8b Author: Chris Wilson Date: Thu Sep 2 13:07:53 2010 +0100 check: Fix for link-time-optimisation src/check-def.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit a9e055ff481303428cd9684dca826bc3f5d868c5 Author: M Joonas Pihlaja Date: Thu Sep 2 14:51:30 2010 +0100 freelist: Separate freelist inlines from the datatype in headers. The cairo-freelist-private.h header has a number of static inline functions which call hidden functions in libcairo. This poses a problem on Solaris where the native compiler compiles inline functions whether they are used or not, thereby adding the link time requirements on hidden functions from whatever code that includes cairo-freelist-private.h. Unfortunately the boilerplate code includes cairo-private headers and indirectly the freelist header, so linking the boilerplate helper library fails on Solaris. This patch separates the structure definitions from the function prototypes and static inlines in cairo-freelist-private.h by moving the datatypes to a new cairo-freelist-type-private.h. src/Makefile.sources | 1 + src/cairo-freelist-private.h | 28 +------------------- src/cairo-freelist-type-private.h | 54 +++++++++++++++++++++++++++++++++++++++ src/cairo-xlib-display.c | 2 +- src/cairo-xlib-private.h | 2 +- 5 files changed, 58 insertions(+), 29 deletions(-) commit f5632d641c44a143386ca3b413e90ed3b1a85f89 Author: M Joonas Pihlaja Date: Thu Sep 2 13:55:55 2010 +0300 build: Remove stray backslash. A stray backslash bwetween two assignment lines apparently invokes a GNU make extension for conditional assignments. This patch fixes the build when using the native Solaris make. src/Makefile.sources | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 3a8b8fcc2a8a48cef49211c45723862f9fa8db23 Author: Benjamin Otte Date: Wed Sep 1 21:14:40 2010 +0200 gobject: Fix header defines util/cairo-gobject/cairo-gobject.h | 64 +++++++++++++++++++------------------- 1 file changed, 32 insertions(+), 32 deletions(-) commit 54a651564cdd2510d0a66f6720c4a45733ff3285 Author: Benjamin Otte Date: Wed Sep 1 18:47:16 2010 +0200 build: Work around caching not reevaluating uncached variables This is a somewhat crude hack to ensure that GOBJECT_CFLAGS/LIBS are properly cached. configure.ac | 4 ++++ 1 file changed, 4 insertions(+) commit 28ba78a9716f3c105cdfb5f7519f2435dd9cb271 Author: Benjamin Otte Date: Wed Sep 1 18:46:46 2010 +0200 gobject: Fix pkgconfig file variable I confused % and $. I blame it on spec files. configure.ac | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 61a1835e40d0b6d786951143779077d16ed2486e Author: Benjamin Otte Date: Wed Sep 1 03:17:12 2010 +0200 subsurface: Fix acquire_source_surface for out-of-bounds surfaces See subsurface-outside-target test. src/cairo-surface-subsurface.c | 41 +++++++++++++++++++++++------------------ 1 file changed, 23 insertions(+), 18 deletions(-) commit f54c51e75c7abe872127d559e029bf4383439bca Author: Benjamin Otte Date: Wed Sep 1 02:10:51 2010 +0200 test: Add subsurface-outside-target This test defines the behavior of subsurfaces when they are created to be bigger than their target. test/Makefile.sources | 1 + test/subsurface-outside-target.argb32.ref.png | Bin 0 -> 2063 bytes test/subsurface-outside-target.c | 178 ++++++++++++++++++++++++++ test/subsurface-outside-target.rgb24.ref.png | Bin 0 -> 1648 bytes 4 files changed, 179 insertions(+) commit 6eab5ee10c111cf87dcf62e3f639f44cfd1e5cf6 Author: Maarten Bosmans Date: Sat Aug 28 13:39:21 2010 +0200 build: prefer libpng14 over older versions Also move libpng to the front of the list, so that the default version of the system is first checked. configure.ac | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 9f98ec6eb915204d59820cf57d32f347f9933517 Author: M Joonas Pihlaja Date: Sun Aug 29 00:05:51 2010 +0300 device: Refer the user to per-backend docs about devices. The exact semantics of the device API can't be spelled out out in the device API docs since the effects are so tied to the specific backend. To use a particular device in practice the user should refer to the backend docs. src/cairo-device.c | 6 ++++++ 1 file changed, 6 insertions(+) commit b0f65054b810d432238ff4986bee3f63e1fc46df Author: M Joonas Pihlaja Date: Sun Aug 29 00:03:07 2010 +0300 subsurface: Add a note to the docs about usage restrictions. Make our assumptions about constructed subsurfaces explicit while we work out the correct semantics for the edge cases. src/cairo-surface-subsurface.c | 7 +++++++ 1 file changed, 7 insertions(+) commit 92fe2d370d120f4d9bc95e92209d1c1cec7e900b Author: Benjamin Otte Date: Sat Aug 28 17:12:45 2010 +0200 surface-offset: Use right variable in mask Use source/mask instead of their copies. The copies only get initialized if we actually need to translate things. src/cairo-surface-offset.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit df4b23ff74f3d22fb306f59cdba4dce6b0857cd2 Author: M Joonas Pihlaja Date: Fri Aug 27 20:26:06 2010 +0300 subsurface: Fix crash when acquiring a source image. The code is passing in the extra info received from the target surface's acquire_source() method to _cairo_surface_get_extents() rather than the acquired source surface itself. src/cairo-surface-subsurface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 2d146867dc4d4e17da2fd1e4776f62aaa999d90f Author: Andrea Canciani Date: Thu Aug 26 14:50:34 2010 +0200 subsurface: Correct cloning condition 72644c6df2c2da1d77ca10a755dbb44bbd960c0d was using the wrong (inverted) condition to check if the subsurface extents are contained in the target extents. src/cairo-surface-subsurface.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) commit 72644c6df2c2da1d77ca10a755dbb44bbd960c0d Author: Andrea Canciani Date: Thu Aug 26 14:24:17 2010 +0200 subsurface: Avoid invalid accesses Whenever subsurface extents are not contained in the target extents, using the source image given by the target (with origin corrected by using an appropriate offset in the data pointer) is not a valid operation. Fallback to cloning in that case. src/cairo-surface-subsurface.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) commit df4465bdd11e25a1ed942bc0fc8e7dc5d43f951f Author: Andrea Canciani Date: Thu Aug 26 14:23:04 2010 +0200 subsurface: Silence compiler warnings Commit d1c35f5e65d53f6b8b9060455335424ca0f44bd0 introduced two unused variable warnings. src/cairo-surface-subsurface.c | 2 -- 1 file changed, 2 deletions(-) commit d1c35f5e65d53f6b8b9060455335424ca0f44bd0 Author: Benjamin Otte Date: Thu Aug 26 13:55:02 2010 +0200 subsurface: Don't clip size to target size This clipping would modify the origin of the subsurface and then the creator has no easy way to figure this out. src/cairo-surface-subsurface.c | 3 --- 1 file changed, 3 deletions(-) commit 174fb5f16f93887e2f319edd5b46efae3e7e93ad Author: Benjamin Otte Date: Thu Aug 26 13:21:04 2010 +0200 subsurface: Don't apply device transform twice for source/mask src/cairo-surface-offset.c | 12 ------------ 1 file changed, 12 deletions(-) commit e79acf47a57b60c83d1f9491aac354161986b29b Author: Benjamin Otte Date: Thu Aug 26 12:54:25 2010 +0200 subsurface: Make CAIRO_SURFACE_TYPE_SUBSURFACE public src/cairo-analysis-surface.c | 2 +- src/cairo-image-surface.c | 2 +- src/cairo-pdf-surface.c | 4 ++-- src/cairo-ps-surface.c | 4 ++-- src/cairo-script-surface.c | 2 +- src/cairo-surface-subsurface.c | 4 ++-- src/cairo-types-private.h | 3 +-- src/cairo-xcb-surface-render.c | 4 ++-- src/cairo-xlib-surface.c | 2 +- src/cairo.h | 7 ++++++- src/drm/cairo-drm-i915-shader.c | 6 +++--- src/drm/cairo-drm-i965-shader.c | 6 +++--- 12 files changed, 25 insertions(+), 21 deletions(-) commit 40459dacb52587cc784faf2635e54797d3989408 Author: Benjamin Otte Date: Thu Aug 26 11:45:28 2010 +0200 subsurface: Don't crash when the device transform isn't integer Apply the transform before rounding to integer coordinates, so we can handle doubles easily. src/cairo-surface-subsurface.c | 10 ++++------ 1 file changed, 4 insertions(+), 6 deletions(-) commit e19b213e84ab8b8ead5e0c700e5b906590e22aa1 Author: Benjamin Otte Date: Thu Aug 26 11:18:29 2010 +0200 subsurface: Make subsurfaces their own type Do not try to hide being a subsurface. This would cause confusion when users where calling surface-type specific getters on subsurfaces. FIXME: We still need public API to access a subsurface's target though. src/cairo-surface-subsurface.c | 1 - 1 file changed, 1 deletion(-) commit e65bf64c4a6e2c7ff673449b8459b3150e438e21 Author: Benjamin Otte Date: Thu Aug 26 11:17:06 2010 +0200 subsurface: Properly account for device offsets in the target src/cairo-surface-subsurface.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) commit 8e9c4ea5890a4df3f463e592dc3aa429bb423608 Author: Chris Wilson Date: Wed Aug 11 12:18:53 2010 +0100 Mark the context as invalid on the final cairo_destroy() We initialise the context to a non-error state upon creation, but after destroy there is a window of opportunity where the object is kept alive inside the context pool and the user could mistakenly keep on passing the zombie context into cairo functions. As all entry points need to check error status, flagging the context as an error object upon the final unreference prevents such misuse (until such as time as the context is reallocated). src/cairo.c | 3 +++ 1 file changed, 3 insertions(+) commit cd52378ae60ab2075d7e9ec7335e990b2f46059e Author: Andrea Canciani Date: Tue Aug 10 19:38:59 2010 +0200 test: Update linear-gradient-large ref image Make linear-gradient-large ref image match xlib gradients (both on nvidia and on macosx) and pixman gradients (when patched to compute the linear gradients using floating point math). test/linear-gradient-large.ref.png | Bin 3753 -> 4079 bytes 1 file changed, 0 insertions(+), 0 deletions(-) commit a7762e3d6570a6981d8fb74ecc6ae30391c39b89 Author: Krzysztof Kosiński Date: Tue Aug 10 17:55:03 2010 +0200 tests: Add linear-gradient-large test Shows large rounding errors in Pixman when computing gradients. Added as a test to Cairo to catch similar issues in other backends. https://bugs.freedesktop.org/show_bug.cgi?id=29470 test/Makefile.am | 5 +-- test/Makefile.sources | 1 + test/linear-gradient-large.c | 67 +++++++++++++++++++++++++++++++++++++ test/linear-gradient-large.ref.png | Bin 0 -> 3753 bytes 4 files changed, 71 insertions(+), 2 deletions(-) commit 56ea51fdcc273531b5e86b921aad19237a1c9415 Author: Andrea Canciani Date: Mon Aug 9 20:23:50 2010 +0200 Replace insertion sort with mergesort in the scan converter Insertion sort can take up to O(n^2), mergesort is guaranteed to run in O(n*log(n)). An example showing bad performance for insertion sort is: https://bugs.freedesktop.org/show_bug.cgi?id=28067 The mergesort has been engineered to be fast even when working on cases where the insertion sort would have performed well and as expected it shows no changes in the benchmark cairo traces. src/cairo-tor-scan-converter.c | 138 ++++++++++++++++++++++++++++++----------- 1 file changed, 103 insertions(+), 35 deletions(-) commit a150371a5d10e03d6c0d781c6fac950a9ac6be18 Author: Nicolaus L Hepler Date: Tue Aug 10 09:34:39 2010 +0200 ft-font: Make alpha mapping consistent Vertical RGB mapping previously forced opaque pixels. To be consistent with horizontal RGB/BGR and vertical BGR it should use an alpha equal to the mid channel (green). src/cairo-ft-font.c | 7 ------- 1 file changed, 7 deletions(-) commit 73470862cbc5895147d5c58ca2647f539e8fe67c Author: David Schleef Date: Tue Aug 10 02:37:38 2010 +0200 perf: Fix README to say cairo-perf-micro perf/README | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) commit df8f96423429915ef1648310fce8b5aed0917c0a Author: Benjamin Otte Date: Tue Aug 10 01:24:30 2010 +0200 build: Fudge the build system some more so it handles extra libs Add NONPKGCONFIG_EXTRA_LIBS that are only put into the generated pc file but not used when linking. This is now used to add -lcairo-gobject to the cairo-gobject.pc file. build/configure.ac.features | 3 +++ configure.ac | 2 +- src/cairo-features-uninstalled.pc.in | 2 +- src/cairo-features.pc.in | 2 +- 4 files changed, 6 insertions(+), 3 deletions(-) commit 01325ad767acf6213135a0bad895de233324f745 Author: Benjamin Otte Date: Tue Aug 10 00:11:25 2010 +0200 gobject: Include the extra library in the pkgconfig file configure.ac | 1 + 1 file changed, 1 insertion(+) commit 8fb91dc3f6b691f91c322cec4b1d0ee4dbebd69c Author: M Joonas Pihlaja Date: Tue Aug 10 00:53:53 2010 +0300 script: Fix script backend build. The flight data recorders were missing an include of cairo-tee.h ever since cairo-tee.h became an optional backend. util/cairo-fdr/fdr.c | 1 + util/cairo-sphinx/fdr.c | 1 + 2 files changed, 2 insertions(+) commit 766832364904dbf5b8a67ebc1600d2ec45d2734f Author: Andrea Canciani Date: Mon Aug 9 18:47:13 2010 +0200 perf: Improve calibration Make the loops count depend on the actual calibration_loops/calibration_time instead of calibration_loops/calibration_max_time. This avoids having some tests take much less/more than the wanted time per iteration (I was having some tests taking about 1 second, other taking about 7 seconds when the ms_per_iteration was 2000) Spend 0.5-1 times the time wanted for each iteration in calibration, increase the accuracy of loops count. Just making the loops count be the correct ratio doesn't guarantee that the iteration time is accurate. By actually measuring iteration times until it gets greater than 1/4 of the wanted time, the total sum is bound to be <= the wanted iteration time and last calibration time is between 1/4 and 1/2 of the wanted time, so it should give a very accurate loop count. perf/cairo-perf-micro.c | 17 ++++++++--------- 1 file changed, 8 insertions(+), 9 deletions(-) commit 046b642db0782ab5e2a5c82988b21e05afe8e716 Author: Rich Walsh Date: Sat Aug 7 16:30:59 2010 +0300 os2: Restore surface type checking in the get_extents method. src/cairo-os2-surface.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) commit 78bdd87b7545f8e85632ac301a69da145727fcec Author: Rich Walsh Date: Sat Aug 7 16:22:26 2010 +0300 os2: Fix clipping to bounds when blitting pixels. The bounds checks on the rectangle to blit were done after converting target/source to inclusive/noninclusive coordinates rather than before, potentially causing an off-by-one in the sizes, since the same logic was applied to both inclusive and non-inclusive coordinates. This patch enforces bounds up front. Thread "OS/2 surface fixes" on cairo-l: http://lists.cairographics.org/archives/cairo/2010-July/020343.html src/cairo-os2-surface.c | 62 +++++++++++++++++++++---------------------------- 1 file changed, 27 insertions(+), 35 deletions(-) commit d0284687b32161212d6bc52309e2d5bb516dc3e8 Author: Rich Walsh Date: Sat Aug 7 03:53:24 2010 +0300 os2: Tweak an inline declaration. C99 and cairo prefers "static inline " rather than "static inline". Thread "OS/2 surface fixes" on cairo-l: http://lists.cairographics.org/archives/cairo/2010-July/020343.html src/cairo-os2-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit cb30340064a2ff24dc408e185c5a309a14f6c78c Author: Rich Walsh Date: Sat Aug 7 03:33:10 2010 +0300 os2: Consolidate error paths of cairo_os2_surface_create(). Use a single code path to release the resources acquired for a surface that's been partially constructed, rather than have multiple error exits which each release the resources acquired so far. Thread "OS/2 surface fixes" on cairo-l: http://lists.cairographics.org/archives/cairo/2010-July/020343.html src/cairo-os2-surface.c | 120 ++++++++++++++++++++++++------------------------ 1 file changed, 60 insertions(+), 60 deletions(-) commit 66e3d650fe90754c811195c75579ac7a3512b7be Author: Rich Walsh Date: Sat Aug 7 03:12:12 2010 +0300 os2: Document ownership of OS/2 objects passed to surfaces. Clarify the documentation for cairo_os2_surface_create() and cairo_os2_surface_set_hwnd() to note that the ownership of the presentation space and window respectively remains with the caller. Thread "OS/2 surface fixes" on cairo-l: http://lists.cairographics.org/archives/cairo/2010-July/020343.html src/cairo-os2-surface.c | 23 ++++++++++++----------- 1 file changed, 12 insertions(+), 11 deletions(-) commit 01525271544d5b88f87027e470fa85038d59bf8d Author: Rich Walsh Date: Sat Aug 7 03:08:21 2010 +0300 os2: Don't fake an Anchor Block when one isn't needed. The code was being tricky about passing in a HAB to DevOpenDC() even though one isn't needed at all under OS/2. Pass in NULL instead. Thread "OS/2 surface fixes" on cairo-l: http://lists.cairographics.org/archives/cairo/2010-July/020343.html src/cairo-os2-surface.c | 16 +++------------- 1 file changed, 3 insertions(+), 13 deletions(-) commit 883d9725e0f844e7d5cc4c50c75a812b5435702b Author: Rich Walsh Date: Sat Aug 7 03:01:56 2010 +0300 os2: Fix buffer allocator behaviour on arithmetic overflow. On arithmetic overflow the buffer allocator would attempt to allocate zero bytes instead of failing the allocation. Depending on the implementation of the underlying allocator being proxied, this may result in an allocation of a zero length buffer, causing all kinds of grief. This patch causes arithmetic overflows to explicitly fail the allocation. Thread "OS/2 surface fixes" on cairo-l: http://lists.cairographics.org/archives/cairo/2010-July/020343.html src/cairo-os2-surface.c | 54 +++++++++++++++++++++---------------------------- 1 file changed, 23 insertions(+), 31 deletions(-) commit b54037828c5b338b2f1dcbe48201225b258c1e1e Author: Rich Walsh Date: Sat Aug 7 02:40:20 2010 +0300 os2: Fix blitting 24 bpp pixel data. There's a format mismatch between cairo's pixel formats and OS/2 24 bpp data: the individual pixels in OS/2 take only three bytes per pixel, but there needs to be padding between lines to make the start of a line always be aligned at a DWORD boundary. This patch fixes the format conversion code to include the required padding between rows. Thread "OS/2 surface fixes" on cairo-l: http://lists.cairographics.org/archives/cairo/2010-July/020343.html src/cairo-os2-private.h | 1 + src/cairo-os2-surface.c | 116 ++++++++++++++++++++++++++++-------------------- 2 files changed, 68 insertions(+), 49 deletions(-) commit 5aadecb3ac03d379c2c8113566fa8062843df42e Author: Rich Walsh Date: Sat Aug 7 02:30:01 2010 +0300 os2: New API to create surfaces without a persistent HPS. It's not always necessary to create a presentation space (HPS) for surface tied to a window, as the user may have one already. This patch introduces three new functions to create a surface without an HPS and let the user manage an HPS associated with the surface. Thread "OS/2 surface fixes" on cairo-l: http://lists.cairographics.org/archives/cairo/2010-July/020343.html Mozilla Bugzilla: https://bugzilla.mozilla.org/show_bug.cgi?id=557159 src/cairo-os2-surface.c | 142 +++++++++++++++++++++++++++++++++++++++++++++++- src/cairo-os2.h | 14 +++++ 2 files changed, 153 insertions(+), 3 deletions(-) commit 9669faa936fe38da3615d7831e423befa5ecdfad Author: Benjamin Otte Date: Sun Aug 8 17:23:05 2010 +0200 gl: Handle errors from _cairo_gl_gradient_render() Thanks Andrea for finding this. src/cairo-gl-gradient.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) commit e31b56e020df806aa6180cd2efa0a34e55e5adaf Author: Benjamin Otte Date: Sun Aug 8 14:45:24 2010 +0200 configure: Fix gobject handling 1) Explicitly depend on glib, too. 2) Don't link gobject into libcairo proper (thanks Joonas for noticing). configure.ac | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit ae2b7b13cd5fdeaee44496056bb99f497346e262 Author: Andrea Canciani Date: Wed Aug 4 14:22:23 2010 +0200 pattern: Simplify degenerate linear pattern to solid colors Degenerate linear patterns are considered clear if they have EXTEND_NONE, the average of the first and the last stop if they are EXTEND_PAD, the weighted average of the stops (based on the size of the interpolation range in which they are active, just like integrating over the whole interpolation range and taking the average) if they are EXTEND_REPEAT or EXTEND_REFLECT. Fixes degenerate-linear-gradient src/cairo-pattern.c | 119 +++++++++++++++++++++++++++++++- test/degenerate-linear-gradient.ref.png | Bin 232 -> 322 bytes 2 files changed, 117 insertions(+), 2 deletions(-) commit d54a5a9fc90857448954d6cbe725d4665efa3d45 Author: Andrea Canciani Date: Wed Aug 4 14:16:34 2010 +0200 pattern: Add gradient degeneracy testing functions Introduce two new (private) functions for testing the degeneracy of linear and radial gradients and use them instead of verbose (and error-prone) tests. src/cairo-pattern.c | 21 +++++++++++++++++---- 1 file changed, 17 insertions(+), 4 deletions(-) commit 63434bc43819e2c7d4db568e82f5a9a7dfb0418d Author: Andrea Canciani Date: Wed Aug 4 14:22:01 2010 +0200 gstate: Correct comment src/cairo-gstate.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 28cd511fb974f886ae65e588413afcfc220a1a7c Author: Andrea Canciani Date: Sat Aug 7 22:55:46 2010 +0200 Correct color stops comparison function Two clear stops with different "underlying" colors must be considered different, since when interpolated with non-clear colors they will produce different results. src/cairo-color.c | 25 ++++--------------------- 1 file changed, 4 insertions(+), 21 deletions(-) commit f71a618bd2604c8f0beaf4816a38a54f6f271bd9 Author: Andrea Canciani Date: Sat Aug 7 21:32:09 2010 +0200 Fix configuration if gobject is absent Autodetect gobject and only use it if present. Its absence is not to be considered a failure since it's not required. build/Makefile.win32.features | 2 +- configure.ac | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) commit 661f4859cdbc0394ed0a9db34ced0f4e2bbe78ff Author: Benjamin Otte Date: Sat Aug 7 20:05:42 2010 +0200 Add cairo-gobject library This library is supposed to contain GType declarations for Cairo types. It is also meant to support gobject-introspection binding creation. boilerplate/Makefile.win32.features | 12 + build/Makefile.win32.features | 1 + build/Makefile.win32.features-h | 3 + configure.ac | 10 + src/Makefile.win32.features | 16 + util/Makefile.am | 4 + util/cairo-gobject/Makefile.am | 15 + util/cairo-gobject/cairo-gobject-enums.c | 519 +++++++++++++++++++++++++++++ util/cairo-gobject/cairo-gobject-structs.c | 87 +++++ util/cairo-gobject/cairo-gobject.h | 186 +++++++++++ 10 files changed, 853 insertions(+) commit fc4e8f036326da1839ec5320eba60823e7efc2db Author: Søren Sandmann Pedersen Date: Sat Aug 7 12:10:49 2010 -0400 Update documentation to reflect that rgb565 is no longer deprecated. src/cairo.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) commit de606af5d673df193fc07fbd047ba2cec82d2f1c Author: Andrea Canciani Date: Tue Jul 27 18:04:36 2010 +0200 quartz: improve backend detection Quartz previously crashed with NULL backends and didn't check for the backend type when getting a CGContext from a quartz surface, returning meaningless data for subsurfaces. src/cairo-quartz-surface.c | 24 +++++++++++++++--------- 1 file changed, 15 insertions(+), 9 deletions(-) commit 056ce165c4377e9cf93fd7a06b6b0d5d4d98cb8c Author: Andrea Canciani Date: Tue Jul 27 18:50:55 2010 +0200 quartz: fix solid masking fallback Falling back when painting would ignore the alpha value (which is needed to have the correct mask opacity). _cairo_quartz_surface_paint_cg doesn't fallback, so the usual mask fallback path is now taken, as expected. src/cairo-quartz-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit bdeade14ea86fb5f3ff17accd71711d2c40b1b33 Author: Andrea Canciani Date: Wed Jul 28 19:33:53 2010 +0200 quartz: fix stroking with unbounded operators Self-intersecting strokes were drawn incorrectly when an unbounded operator was used, since the fixup operation also cleared the intersection. Fixes clip-stroke-unbounded. src/cairo-quartz-surface.c | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-) commit 8f38d2aa29bc0d69690cbf542c4a2b497b7a0c0a Author: Andrea Canciani Date: Thu Jul 29 10:45:59 2010 +0200 test: improve clip-*-unbounded tests Merge clip-*-unbounded tests and add self-intersections to the paths that are drawn. This exposes a bug in the unbounded fixup code in quartz. Signed-off-by: Chris Wilson test/Makefile.am | 35 ++-- test/Makefile.sources | 3 +- test/clip-draw-unbounded.c | 184 +++++++++++++++++++++ test/clip-fill-eo-unbounded.argb32.ref.png | Bin 0 -> 3668 bytes test/clip-fill-eo-unbounded.image16.rgb24.ref.png | Bin 0 -> 2887 bytes test/clip-fill-eo-unbounded.quartz.argb32.ref.png | Bin 0 -> 3366 bytes test/clip-fill-eo-unbounded.quartz.rgb24.ref.png | Bin 0 -> 3061 bytes test/clip-fill-eo-unbounded.rgb24.ref.png | Bin 0 -> 3201 bytes test/clip-fill-eo-unbounded.svg12.rgb24.xfail.png | Bin 0 -> 3636 bytes ...p-fill-eo-unbounded.xlib-fallback.rgb24.ref.png | Bin 0 -> 3208 bytes test/clip-fill-eo-unbounded.xlib.argb32.ref.png | Bin 0 -> 3693 bytes test/clip-fill-eo-unbounded.xlib.rgb24.ref.png | Bin 0 -> 3198 bytes test/clip-fill-nz-unbounded.argb32.ref.png | Bin 0 -> 3668 bytes test/clip-fill-nz-unbounded.image16.rgb24.ref.png | Bin 0 -> 2887 bytes test/clip-fill-nz-unbounded.quartz.argb32.ref.png | Bin 0 -> 3366 bytes test/clip-fill-nz-unbounded.quartz.rgb24.ref.png | Bin 0 -> 3061 bytes test/clip-fill-nz-unbounded.rgb24.ref.png | Bin 0 -> 3201 bytes test/clip-fill-nz-unbounded.svg12.rgb24.xfail.png | Bin 0 -> 3636 bytes ...p-fill-nz-unbounded.xlib-fallback.rgb24.ref.png | Bin 0 -> 3208 bytes test/clip-fill-nz-unbounded.xlib.argb32.ref.png | Bin 0 -> 3693 bytes test/clip-fill-nz-unbounded.xlib.rgb24.ref.png | Bin 0 -> 3198 bytes test/clip-fill-unbounded.argb32.ref.png | Bin 1607 -> 0 bytes test/clip-fill-unbounded.c | 105 ------------ test/clip-fill-unbounded.image16.ref.png | Bin 1204 -> 0 bytes test/clip-fill-unbounded.pdf.argb32.ref.png | Bin 1244 -> 0 bytes test/clip-fill-unbounded.ps.argb32.xfail.png | Bin 1244 -> 0 bytes test/clip-fill-unbounded.quartz.argb32.ref.png | Bin 1436 -> 0 bytes test/clip-fill-unbounded.quartz.rgb24.ref.png | Bin 1368 -> 0 bytes test/clip-fill-unbounded.rgb24.ref.png | Bin 1304 -> 0 bytes test/clip-fill-unbounded.svg12.argb32.xfail.png | Bin 1601 -> 0 bytes test/clip-fill-unbounded.svg12.rgb24.xfail.png | Bin 1601 -> 0 bytes ...clip-fill-unbounded.xlib-fallback.rgb24.ref.png | Bin 1438 -> 0 bytes test/clip-fill-unbounded.xlib.argb32.ref.png | Bin 1759 -> 0 bytes test/clip-fill-unbounded.xlib.rgb24.ref.png | Bin 1430 -> 0 bytes test/clip-stroke-unbounded.argb32.ref.png | Bin 1694 -> 3545 bytes test/clip-stroke-unbounded.c | 102 ------------ test/clip-stroke-unbounded.image16.ref.png | Bin 1255 -> 0 bytes test/clip-stroke-unbounded.image16.rgb24.ref.png | Bin 0 -> 2778 bytes test/clip-stroke-unbounded.pdf.argb32.xfail.png | Bin 1351 -> 0 bytes test/clip-stroke-unbounded.ps.argb32.ref.png | Bin 1351 -> 0 bytes test/clip-stroke-unbounded.quartz.argb32.ref.png | Bin 1493 -> 3362 bytes test/clip-stroke-unbounded.quartz.rgb24.ref.png | Bin 1356 -> 3061 bytes test/clip-stroke-unbounded.rgb24.ref.png | Bin 1372 -> 3063 bytes test/clip-stroke-unbounded.svg12.argb32.xfail.png | Bin 1697 -> 0 bytes test/clip-stroke-unbounded.svg12.rgb24.xfail.png | Bin 1697 -> 3569 bytes ...ip-stroke-unbounded.xlib-fallback.rgb24.ref.png | Bin 1437 -> 3123 bytes test/clip-stroke-unbounded.xlib.argb32.ref.png | Bin 1785 -> 3620 bytes test/clip-stroke-unbounded.xlib.rgb24.ref.png | Bin 1442 -> 3109 bytes 48 files changed, 204 insertions(+), 225 deletions(-) commit 142df6f9638e10d0712e092d902f7670965a88f9 Author: Benjamin Otte Date: Fri Jul 30 22:04:44 2010 +0200 configure: Disable xml backend by default We don't want to enable it by default when nobody uses it. boilerplate/Makefile.win32.features | 2 +- build/Makefile.win32.features | 2 +- configure.ac | 2 +- src/Makefile.win32.features | 2 +- 4 files changed, 4 insertions(+), 4 deletions(-) commit 9f33f8453b4949cfdc63169d3acd7238f89379c2 Author: Benjamin Otte Date: Fri Jul 30 21:48:14 2010 +0200 configure: Disable tee backend by default We don't want to enable it by default when nobody uses it. boilerplate/Makefile.win32.features | 4 +++- build/Makefile.win32.features | 1 + build/Makefile.win32.features-h | 2 ++ configure.ac | 2 +- src/Makefile.win32.features | 8 +++++++- src/cairo-surface.c | 4 ++++ test/api-special-cases.c | 6 ++++++ 7 files changed, 24 insertions(+), 3 deletions(-) commit 9f4d677e2a6dd0bfbbbe20de39f92177bfdfc0e8 Author: Benjamin Otte Date: Fri Jul 30 22:15:55 2010 +0200 tee: Move definitions into separate header src/Makefile.sources | 6 +++-- src/cairo-tee-surface.c | 2 ++ src/cairo-tee.h | 66 ++++++++++++++++++++++++++++++++++++++++++++++++ src/cairo.h | 17 ------------- test/api-special-cases.c | 3 +++ 5 files changed, 75 insertions(+), 19 deletions(-) commit 14639e6b8525ada2cb93abb1d70dde827b7bf9e0 Author: Kristian Høgsberg Date: Thu Jun 17 18:29:48 2010 -0400 egl: Use EGL_KHR_surfaceless_opengl extension when available This lets us avoid creating a throwaway pbuffer just to make the context current. src/cairo-egl-context.c | 58 ++++++++++++++++++++++++++++++++----------------- 1 file changed, 38 insertions(+), 20 deletions(-) commit b6fd673d37e1c8962d15c0e0e59322a2002fc399 Author: Chris Wilson Date: Mon Jul 26 13:15:38 2010 +0100 version: 1.9.15 post-snapshot bump cairo-version.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit ed8c6f42540f03a62fd64ad83b7c397528092232 Author: Chris Wilson Date: Mon Jul 26 12:53:44 2010 +0100 version: 1.9.14 snapshot cairo-version.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 81094cf4622302ebaa9b3f328f5d2a56a1c07a64 Author: Chris Wilson Date: Mon Jul 26 12:52:56 2010 +0100 NEWS: 1.9.14 NEWS | 27 +++++++++++++++++++++++++++ 1 file changed, 27 insertions(+) commit c1fcbd0322f86327ddd89c2eef2facbc74045629 Author: Andrea Canciani Date: Mon Jul 26 11:21:48 2010 +0200 quartz: fix REFLECT radial gradients The interpolation range of repeating radial gradients can safely be reflected around any integer (previously 0), but for reflect-extended radial gradients can only be reflected around odd integers if the appearance is to be the same, thus reflecting around 1 is correct for both. Fixes radial-gradient. src/cairo-quartz-surface.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 3898e46d1de19848efeddb442dcf7f7663530256 Author: Chris Wilson Date: Fri Jul 23 09:41:48 2010 +0100 image: Compensate mask origins for glyph clipping Fixes the regression in bitmap-font and large-font pointed out by Andrea Canciani since ef067933. src/cairo-image-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 53fcfc781eeee949abc03cf90048a2228f7e0a63 Author: M Joonas Pihlaja Date: Wed Jul 21 22:39:40 2010 +0300 build: Typos in comments. Fix inline "dnl" strings left over from a string substitution "--" -> "dnl". build/configure.ac.pthread | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) commit fd1e731c9d0af307ebde4d888866c95bd6ebe66d Author: Hib Eris Date: Wed Jul 21 22:34:05 2010 +0300 build: Fix pthread detection code when there isn't one. The code was testing the output variable $use_pthread rather than the input variable $enable_pthread when checking that pthreads were actually found when requested. build/configure.ac.pthread | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 108b1c7825116ed3f93aa57384bbd3290cdc9181 Author: Karl Tomlinson Date: Sat Jul 17 13:08:53 2010 +1200 clip: consider gstate target extents in _cairo_gstate_copy_clip_rectangle_list Fixes https://bugs.freedesktop.org/show_bug.cgi?id=29125 To be consistent with _cairo_gstate_clip_extents, the context's clip should be intersected with the target surface extents (instead of only using them when there is no clip). Reviewed-by: Chris Wilson src/cairo-clip.c | 72 ++++++++++++++++++++---------------------------------- src/cairo-gstate.c | 14 ++++++++++- test/get-clip.c | 16 ++++++++++++ 3 files changed, 55 insertions(+), 47 deletions(-) commit b79ea8a6cab8bd28aebecf6e1e8229d5ac017264 Author: Karl Tomlinson Date: Sat Jul 17 11:46:25 2010 +1200 clip: consider all_clipped in _cairo_clip_get_extents If the gstate clip in _cairo_gstate_int_clip_extents() has all_clipped set (and path NULL), then it returns the gstate target extents instead of an empty rectangle. If the target is infinite, then it says the clip is unbounded. Fixes https://bugs.freedesktop.org/show_bug.cgi?id=29124 Tested-by test/get-clip Reviewed-by: Chris Wilson src/cairo-clip.c | 5 +++++ test/get-clip.c | 5 ++++- 2 files changed, 9 insertions(+), 1 deletion(-) commit 6a5115901381fe9c81a8a311e267c2a00e2a1d04 Author: Karl Tomlinson Date: Sat Jul 17 10:47:42 2010 +1200 test: return CAIRO_TEST_FAILURE from get-clip preamble on failure https://bugs.freedesktop.org/show_bug.cgi?id=29120 test/get-clip.c | 5 +++++ 1 file changed, 5 insertions(+) commit f2fa15680ec3ac95cb68d4957557f06561a7dc55 Author: Karl Tomlinson Date: Sat Jul 17 10:39:50 2010 +1200 clip: return empty clip from _cairo_clip_copy_rectangle_list when all_clipped Fixes https://bugs.freedesktop.org/show_bug.cgi?id=29122 Tested by test/get-clip Reviewed-by: Chris Wilson src/cairo-clip.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) commit 33ac2035876aeb9215ca0bdaaeea5265fcee61f8 Author: Karl Tomlinson Date: Sat Jul 17 10:13:22 2010 +1200 test: use the surface size expected in get-clip phase "No clip set" https://bugs.freedesktop.org/show_bug.cgi?id=29121 test/get-clip.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) commit bac5c28a18af5d43c96f0b64d1c7235da3f02779 Author: Chris Wilson Date: Sat Jul 17 10:09:14 2010 +0100 boilerplate: Offset the xlib window An idea courtesy of Benjamin Otte is to stress the drivers a bit more by not rendering to 0,0. Under various circumstances rendering to a Window is mapped onto the Screen root Window with a collection of offsets and clips by X. It is easy for the driver to make a mistake and misrender... boilerplate/cairo-boilerplate-xlib.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 51b0c21f02a8f0f5d7f44da05e49d7f8f3bf8bc7 Author: Benjamin Otte Date: Fri Jul 16 19:49:53 2010 +0200 build: Add explicit dependcy to cxx lib Otherwise the build will fail because it doesn't add the dependency manually. https://bugs.freedesktop.org/show_bug.cgi?id=29114 boilerplate/Makefile.am | 3 +++ src/Makefile.am | 2 +- 2 files changed, 4 insertions(+), 1 deletion(-) commit 6b92d5a0f9fce28f99b07177454da00f984fdb43 Author: Oleg Romashin Date: Fri Jul 16 09:47:23 2010 +0100 qt: Use native glyph rendering. Remove the hack of using the xlib surface to do glyph rendering and use the new interface introduced with Qt4.7, or else fallback to image rendering with freetype. Reviewed-by: Chris Wilson [2 files changed, 35 insertions(+), 140 deletions(-) \o/] src/cairo-qt-surface.cpp | 169 ++++++++++------------------------------------- src/cairo-xlib-surface.c | 6 -- 2 files changed, 35 insertions(+), 140 deletions(-) commit f53f072a223c3fbdeba315f5cba755cba498e121 Author: Chris Wilson Date: Mon Jul 12 18:53:29 2010 +0100 clip: Do the NULL check before the dereference. Breakage introduced in the commit earlier today. src/cairo-clip.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit ef0679333da881bd83b0bb4db546ea9c68f81f89 Author: Chris Wilson Date: Mon Jul 12 16:07:27 2010 +0100 image: Manually clip against bounds when rendering directly This is path is slightly peculiar in that it explicitly avoid the intermediate mask and the geometry is not pre-clipped. This in conjunction with the previous commit fixes: Clip doesn't work for text https://bugs.freedesktop.org/show_bug.cgi?id=29008 which is captured in test/partial-clip-text. src/cairo-image-surface.c | 23 +++++++++++++++++++---- 1 file changed, 19 insertions(+), 4 deletions(-) commit 679e5a600bd7dae6cbde83070fb1f7355f8025a7 Author: Chris Wilson Date: Mon Jul 12 16:06:32 2010 +0100 Differentiate between reducing clip to composite extents and a rectangle This is required for handling glyphs when rendering directly to the surface. src/cairo-clip-private.h | 4 ++++ src/cairo-clip.c | 22 +++++++++++++++------- src/cairo-gl-surface.c | 4 ++-- src/cairo-image-surface.c | 10 +++++----- src/cairo-surface-fallback.c | 10 +++++----- src/cairo-xcb-surface-render.c | 10 +++++----- src/drm/cairo-drm-i915-glyphs.c | 2 +- src/drm/cairo-drm-i915-surface.c | 8 ++++---- src/drm/cairo-drm-i965-glyphs.c | 2 +- src/drm/cairo-drm-i965-surface.c | 8 ++++---- 10 files changed, 46 insertions(+), 34 deletions(-) commit 8546a877889bfafc056c867bc9aea25e9fdcdef0 Author: Chris Wilson Date: Mon Jul 12 16:27:29 2010 +0100 test: Add partial-clip-text This exercises a bug found by Igor Nikitin: https://bugs.freedesktop.org/show_bug.cgi?id=29008 test/Makefile.am | 3 +++ test/Makefile.sources | 1 + test/partial-clip-text.c | 50 +++++++++++++++++++++++++++++++++++++ test/partial-clip-text.ps.ref.png | Bin 0 -> 107 bytes test/partial-clip-text.ref.png | Bin 0 -> 175 bytes test/partial-clip-text.svg.ref.png | Bin 0 -> 173 bytes 6 files changed, 54 insertions(+) commit 53ee897449dca497d6a3d6dc90b87fea8027cc44 Author: Chris Wilson Date: Mon Jul 12 12:37:31 2010 +0100 version: Post-snapshot bump to 1.9.13 cairo-version.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit aa4cd8287f47b4538e765e1b48dcbac19813a8a2 Author: Chris Wilson Date: Mon Jul 12 12:12:12 2010 +0100 version: 1.9.12 snapshot cairo-version.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 53b129dd8aecf4e2bd8b7d2cd27b91370c6705e9 Author: Chris Wilson Date: Mon Jul 12 12:11:42 2010 +0100 NEWS for 1.9.12. NEWS | 56 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 56 insertions(+) commit 01380f3a56b28a388d2876780f33516ecd16ee16 Author: Chris Wilson Date: Mon Jul 12 11:49:53 2010 +0100 test: Missing REFERENCE_IMAGES test/Makefile.am | 3 +++ 1 file changed, 3 insertions(+) commit 459b8ae660eb49a87ffb9da31f305718a2b4f11f Author: Chris Wilson Date: Mon Jul 12 11:05:57 2010 +0100 Make distcheck happy. src/cairo-fontconfig-private.h | 2 ++ src/cairo-skia.h | 2 +- src/glew/GL/glew.h | 2 ++ src/glew/GL/glxew.h | 2 ++ test/api-special-cases.c | 5 +++-- 5 files changed, 10 insertions(+), 3 deletions(-) commit d56b1718025f8b2354b5b1eb99af41df70a6c590 Author: Andrea Canciani Date: Mon Jul 12 10:36:24 2010 +0200 test: Add halo quartz reference images When quartz-font is enabled, halo and halo-transform fonts have a different shape. test/halo-transform.quartz.ref.png | Bin 0 -> 14360 bytes test/halo.quartz.ref.png | Bin 0 -> 8601 bytes 2 files changed, 0 insertions(+), 0 deletions(-) commit 9107895789fc721ee6ea2ccb2ce3ab99fb978caa Author: M Joonas Pihlaja Date: Mon Jul 12 11:03:15 2010 +0300 test: Add comments explaining the bug in bug-seams. Where two rasterization methods meet, havoc occurs. test/bug-seams.c | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+) commit 0152db263b9b12547fadb5d080913fcbb696d34e Author: M Joonas Pihlaja Date: Mon Jul 12 10:36:00 2010 +0300 build: Sync up autogenerated win32 build files. boilerplate/Makefile.win32.features | 9 +++++++++ build/Makefile.win32.features | 1 + build/Makefile.win32.features-h | 3 +++ src/Makefile.win32.features | 9 +++++++++ 4 files changed, 22 insertions(+) commit 06ffb7df716652baba96960cd59e1a59f4bd1ac1 Author: Eric Anholt Date: Sat Jul 10 14:19:50 2010 -0700 gl: Only use a mat3 for the transformation for gradients coordinates. No point in passing in a column of zeroes that we multiply by zero, or the row that we drop on the floor. We could do even better by just passing in a 3x2, at the expense of adding a language version dependency (or just pass in 2 vectors). No significant performance difference. src/cairo-gl-shaders.c | 28 +++++++++++++--------------- 1 file changed, 13 insertions(+), 15 deletions(-) commit 1e7485f8e5b6fd9b26f06838d8e7a35e073c394b Author: M Joonas Pihlaja Date: Mon Jul 12 00:23:11 2010 +0300 trace: Avoid stdbool.h C99ism. There is no on old Solaris without invoking c99 mode explicitly. util/cairo-trace/trace.c | 184 ++++++++++++++++++++++++----------------------- 1 file changed, 94 insertions(+), 90 deletions(-) commit 97529a3713d899468fb3ca7a1c3f75a09bd1b215 Author: M Joonas Pihlaja Date: Sun Jul 11 23:54:21 2010 +0300 build: Avoid shifting shell parameters that aren't there. The shift primitive of the Solaris 9 Bourne shell doesn't like it if we try to shift more formal arguments than are there. This causes the compare_versions function in autogen.sh to croak when the actual and compared versions differ in the number of components. autogen.sh | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit dd6026b613aa5414a128727995f2a4f79284072a Author: Benjamin Otte Date: Sun Jul 11 20:44:40 2010 +0200 doc: Make the necessity of flush/mark_dirty more obvious src/cairo-surface.c | 32 ++++++++++++++++++++++++++++++++ 1 file changed, 32 insertions(+) commit 8a1944f45b9c9056088246d4f2e72d60a59fd5c4 Author: Benjamin Otte Date: Sun Jul 11 19:11:15 2010 +0200 check: Allow unprefix type names in the title Some documentation sections use the type name as the title, and we don't want to use that title as a link. src/check-doc-syntax.sh | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) commit f0466ecf7db859a80b779b7fb4381e3c9409c39b Author: M Joonas Pihlaja Date: Sun Jul 11 21:57:40 2010 +0300 build: Fix typo in pthread test program. Missing a semicolon after pthread_cleanup_push. Incredibly only openbsd 4.5 minded. build/configure.ac.pthread | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 3f1d7de8e1620ad385930477b63454107dd8ffd3 Author: M Joonas Pihlaja Date: Sun Jul 11 17:29:07 2010 +0300 build: Fix regression provoked by newer autoconf and dodgy configuring. We're not supposed to be redefining PACKAGE_VERSION, PACKAGE_... from the configure generated confdefs.h. This patch rudely adds paper over the problem. The compiler warnings are a problem for us since our checking of various compiler flags assumes that no news is good news, and that any warning messages are due to the flags under test. The regression appears when using an autoconf >= 2.64, at least, but not with 2.61. The same issue appears in the pthread test because our conftest unconditionally #defines _GNU_SOURCE, but autoconf ends up doing that in the confdefs.h. build/configure.ac.noversion | 13 ++++++++++++- build/configure.ac.pthread | 2 ++ 2 files changed, 14 insertions(+), 1 deletion(-) commit fca8977219b857e2e61dd86ac39ae9f40086f306 Author: M Joonas Pihlaja Date: Sun Jun 27 03:03:17 2010 +0300 build: Rework pthread detection. Use two levels of pthread support: a minimal level used to build cairo itself, and a full level to build threaded apps which want to use cairo. The minimal level tries to use pthread stubs from libc if possible, but falls back to the full level if that's not possible. We use CFLAGS=-D_REENTRANT LIBS=-lpthread to find a real pthread library since that seems to work on every unix-like test box we can get our hands on. build/ax-pthread.m4 | 283 ------------------------------------------ build/configure.ac.pthread | 251 +++++++++++++++++++++++++++++++++++++ configure.ac | 47 +++---- test/Makefile.am | 16 +-- test/cairo-test-trace.c | 10 +- test/cairo-test.c | 6 +- util/cairo-sphinx/Makefile.am | 3 +- util/cairo-sphinx/sphinx.c | 4 + util/cairo-trace/Makefile.am | 4 +- 9 files changed, 291 insertions(+), 333 deletions(-) commit 47c35e5e86a3c99fc39afe2e13a7c92d5247ee1e Author: M Joonas Pihlaja Date: Sun Jun 27 01:14:03 2010 +0300 build: Purge CAIRO_CC_TRY_LINK_FLAG. Use CAIRO_CC_TRY_FLAG instead in the one place where we used CAIRO_CC_TRY_LINK_FLAG and make the build system one macro smaller. build/aclocal.cairo.m4 | 21 --------------------- build/configure.ac.warnings | 2 +- 2 files changed, 1 insertion(+), 22 deletions(-) commit d2f5592e0e0d316cfc40ec676ee6e7f4e2a699fb Author: M Joonas Pihlaja Date: Sun Jun 27 01:11:13 2010 +0300 build: Factor out a more pliable CAIRO_CC_TRY_LINK_WITH_ENV_SILENT. Introduce a new CAIRO_CC_TRY_LINK_WITH_ENV_SILENT macro for running generic link tests with arbitrary CFLAGS/LIBS/LDFLAGS and no muttering of autoconf messages. Rewrite the previous CAIRO_CC_TRY_FLAG in terms of it. build/aclocal.cairo.m4 | 31 +++++++++++++++++++++++++------ 1 file changed, 25 insertions(+), 6 deletions(-) commit 120b78ada7eed6424f07a688854657d0c3ad68f8 Author: Benjamin Otte Date: Sun Jul 11 18:48:38 2010 +0200 build: Split boilerplate lib generation into two so's Same thing as what was previously done for libcairo.so: Build an intermediate lib for the (potential) C++ stuff. boilerplate/Makefile.am | 31 ++++++++++++++++++++++++------- boilerplate/Makefile.sources | 7 ++----- 2 files changed, 26 insertions(+), 12 deletions(-) commit d6362a0d76753cb6f85f0dbd567a3ec7b7b08696 Author: Benjamin Otte Date: Sun Jul 11 16:01:44 2010 +0200 qt: Define cairo-qt's API as C symbols That way, the C linker doesn't mess up when linking libcairo.so. src/cairo-qt.h | 12 ++++-------- 1 file changed, 4 insertions(+), 8 deletions(-) commit 5d4a4775c8b091b947b52cc8f7e1e48a5eaea007 Author: Benjamin Otte Date: Sun Jul 11 15:41:56 2010 +0200 build: Fix typo that makes C++ libs not build src/Makefile.am | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 9b9952ab4f64a476bbadfd3e6cf1e66c1137ccd7 Author: Chris Wilson Date: Sat Jul 10 11:13:07 2010 +0100 Convert mime data length to use unsigned long What we want to use is size_t, but we don't want the implied POSIX dependency. However, POSIX does say that size_t is an unsigned integer that is no longer than a long, so it would appear safe to use an unsigned long as a replacement. Safer at least than unsigned int. src/cairo-image-info-private.h | 6 +++--- src/cairo-image-info.c | 4 ++-- src/cairo-output-stream-private.h | 2 +- src/cairo-output-stream.c | 2 +- src/cairo-pdf-surface.c | 18 +++++++++--------- src/cairo-png.c | 2 +- src/cairo-ps-surface.c | 2 +- src/cairo-script-surface.c | 6 +++--- src/cairo-surface.c | 4 ++-- src/cairo-svg-surface.c | 6 +++--- src/cairo-types-private.h | 2 +- src/cairo-win32-printing-surface.c | 10 +++++----- src/cairo.h | 4 ++-- util/cairo-trace/trace.c | 6 +++--- 14 files changed, 37 insertions(+), 37 deletions(-) commit eb85ab998b606144daf025285968a5d6031b9aa6 Author: Chris Wilson Date: Sat Jul 10 10:43:40 2010 +0100 subsurface: Don't assume we ever own the target. Andrea pointed out that snapshots can be shared, and so the subsurface should not be calling finish even on a snapshot. src/cairo-surface-subsurface-private.h | 1 - src/cairo-surface-subsurface.c | 17 ++++++++--------- 2 files changed, 8 insertions(+), 10 deletions(-) commit 59c83117d77f58e9cb64aaa6b572fab62fb31c6b Author: Benjamin Otte Date: Fri Jul 9 18:55:11 2010 +0200 build: Commit new versions of autogenerate Windows build files The recent commits to the build system changed a bunch of stuff in there, so update the files. boilerplate/Makefile.win32.features | 92 +++++++++++++++++++++++++++++++++---- build/Makefile.win32.features-h | 1 - src/Makefile.win32.features | 92 +++++++++++++++++++++++++++++++++---- 3 files changed, 164 insertions(+), 21 deletions(-) commit bf117e4ea7c89dadcb5a358bb47c0f213cc67461 Author: Benjamin Otte Date: Fri Jul 9 18:43:38 2010 +0200 build: fix glew include Currently wether internal glew is built is dependant on wether the tarball in use was built with internal glew or not. And that's not how it should be. configure.ac | 8 ++++---- src/Makefile.sources | 11 +++++++---- 2 files changed, 11 insertions(+), 8 deletions(-) commit 0bd704bd1e92b274209aeda6de88c7e75ffa81be Author: Benjamin Otte Date: Fri Jul 9 17:15:02 2010 +0200 build: Remove wrong flags from libcairo_cxx.la They're useful for the final so, but not for an intermediate noinst lib. src/Makefile.am | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit faceffd5beb4297299175945660522d8f1f5f2cb Author: Benjamin Otte Date: Fri Jul 9 16:58:18 2010 +0200 build: copy/paste failure: Don't build libcxx.la src/Makefile.am | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit dbb386d3c0aa9e7534d333cf3dbcfc2aaeac05d3 Author: Benjamin Otte Date: Fri Jul 9 12:49:10 2010 +0200 gl: Use unsigned int instead of GLuint GLuint is not defined in cairo-gl.h. If unsigned int is not compatible enough to GLuint, we might consider to use #include instead. src/cairo-gl-surface.c | 2 +- src/cairo-gl.h | 3 ++- 2 files changed, 3 insertions(+), 2 deletions(-) commit f7fc8569a797356d5e93ad67aae4eca31e6835cd Author: Benjamin Otte Date: Fri Jul 9 12:29:35 2010 +0200 build: Fix C++ issues with automake This is an attempt to fix the broken situation we've been in where automake links libcairo.la with c++ because it might potentially maybe include C++ files. Those potential files only exist in Chris' throwaway backends (skia, qt) and the BeOS backend, so for 99.99% of cases, these backends are not needed and linking with c++ is overkill. Also, no one wants to have libcairo.so link to libstdc++. This patch fixes that in mutliple steps: 1) Add build infrastructure to distinguish between C and C++ backends. This is done by allowing to specify backend_sources as well as backend_cxx_sources variables in Makefile.sources. 2) Optionally build a libcairo_cxx.la noinst library This intermediate library is built for C++ backends only and therefor linked using c++. It is then linked into the final libcairo.la. This does not require c++, so the linking of libcairo.la is done with cc. This also works around various weirdnesses that the current build system exposes, where it assumes cisms when in fact using c++ semantics, like not detecting c++ properly or: https://bugzilla.redhat.com/show_bug.cgi?id=606523 build/configure.ac.features | 7 ++++++- configure.ac | 10 ++++++++++ src/Makefile.am | 20 +++++++++++++++++++- src/Makefile.am.analysis | 4 ++-- src/Makefile.sources | 6 +++--- 5 files changed, 40 insertions(+), 7 deletions(-) commit df6d49f6eaf334d5a2de8bdd90919278071ab868 Author: Benjamin Otte Date: Fri Jul 9 10:29:30 2010 +0200 build: Get rid of $(foo_files) Makefile.am variables They're unused, and less variables make the build system easier to understand. build/configure.ac.features | 1 - src/Makefile.am | 1 - 2 files changed, 2 deletions(-) commit ce530b2802683fe6c5289367ef03018f4081a2da Author: Benjamin Otte Date: Fri Jul 9 04:24:51 2010 +0200 doc: Add "@Since: 1.10" tags to device docs src/cairo-device.c | 14 ++++++++++++++ src/cairo.h | 2 ++ 2 files changed, 16 insertions(+) commit c1fed693020c96126cdc697ddd22567398369b04 Author: Benjamin Otte Date: Fri Jul 9 04:20:42 2010 +0200 gl: Remove double ** from some comments gtk-doc doesn't like that unless it's a real gtk-doc comment. And it wasn't. src/cairo-gl-composite.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 5c73b371ae31f230210987941eec2db49d2e02be Author: Benjamin Otte Date: Fri Jul 9 04:20:06 2010 +0200 doc: Document remaining cairo-device functions src/cairo-device.c | 32 ++++++++++++++++++++++++++++++++ 1 file changed, 32 insertions(+) commit 9877c1932a44f7e5300d6fcd99c19748aa5a4828 Author: Benjamin Otte Date: Fri Jul 9 03:52:45 2010 +0200 doc: Add more documentation for devices Add general docs and document acquire/release. I'm not happy with the documentation yet. In particular, I have 2 issues: 1) The threading guarantees Cairo provides are missing. 2) There's no docs on which Cairo functions do acquire devices. I guess I'll have to fix them later. src/cairo-device.c | 86 ++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 86 insertions(+) commit 739d6e35fa5632b80b888d6475f2fdb6be6054a6 Author: Benjamin Otte Date: Fri Jul 9 03:04:27 2010 +0200 doc: Clarify that cairo_surface_get_device() can return NULL src/cairo-surface.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit 65d0590b041835fb466f4a7073910ef1383d9e77 Author: Benjamin Otte Date: Fri Jul 9 02:23:13 2010 +0200 doc: Copy docs for standard device APIs Copied from the surface equivalents src/cairo-device.c | 42 ++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 42 insertions(+) commit 8a08b6fa5288f376688730c66ae9219d01180f42 Author: Benjamin Otte Date: Fri Jul 9 02:21:00 2010 +0200 doc: Remove the declarations This information was duplicated. It exists as the "@Title" of the SECTION inline documentation already. doc/public/cairo-sections.txt | 28 ---------------------------- 1 file changed, 28 deletions(-) commit 5e76bc6deb9fd4dc7a0f9181d3941ba55ce89c56 Author: Benjamin Otte <otte@redhat.com> Date: Fri Jul 9 01:10:24 2010 +0200 doc: Add missing region documentation src/cairo-region.c | 25 +++++++++++++++++++++++++ src/cairo.h | 13 +++++++++++++ 2 files changed, 38 insertions(+) commit 57da730d286f643c8f68b445907f7cdd1f53e5de Author: Benjamin Otte <otte@redhat.com> Date: Thu Jul 8 23:44:35 2010 +0200 doc: Fix a syntax issue in quartz-font gtkdoc A missing empty line caused the long description to be part of @See_Also. Ooops. src/cairo-quartz-font.c | 1 + 1 file changed, 1 insertion(+) commit fb03eba44f1ae7dd76dd3db82bfb55b06b0932ae Author: Andrea Canciani <ranma42@gmail.com> Date: Thu Jul 8 20:06:08 2010 +0200 test: Add quartz functions to api-special-cases test test/api-special-cases.c | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+) commit 59aab0bdc24838252850034cee36ed756f3c3e2a Author: Andrea Canciani <ranma42@gmail.com> Date: Thu Jul 8 20:05:13 2010 +0200 test: fix typo test/api-special-cases.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit b870cc030d8e0eb6309695e5a101c802f688cf40 Author: Benjamin Otte <otte@redhat.com> Date: Thu Jul 8 13:05:18 2010 +0200 doc: Move tmpl/ docs to inline docs I did this manually so I could review the docs at the same time. If anyone finds typos or other mistakes I did, please complain to me (or better: fix them). doc/public/cairo-sections.txt | 4 +- doc/public/tmpl/.gitignore | 3 - doc/public/tmpl/cairo-atsui.sgml | 21 - doc/public/tmpl/cairo-beos.sgml | 22 -- doc/public/tmpl/cairo-font-face.sgml | 116 ------ doc/public/tmpl/cairo-font-options.sgml | 199 ---------- doc/public/tmpl/cairo-ft.sgml | 87 ----- doc/public/tmpl/cairo-image.sgml | 127 ------ doc/public/tmpl/cairo-matrix.sgml | 169 -------- doc/public/tmpl/cairo-paths.sgml | 263 ------------- doc/public/tmpl/cairo-pattern.sgml | 350 ----------------- doc/public/tmpl/cairo-pdf.sgml | 101 ----- doc/public/tmpl/cairo-png.sgml | 96 ----- doc/public/tmpl/cairo-ps.sgml | 144 ------- doc/public/tmpl/cairo-quartz-fonts.sgml | 52 --- doc/public/tmpl/cairo-quartz.sgml | 64 --- doc/public/tmpl/cairo-scaled-font.sgml | 227 ----------- doc/public/tmpl/cairo-status.sgml | 98 ----- doc/public/tmpl/cairo-surface.sgml | 326 --------------- doc/public/tmpl/cairo-svg.sgml | 91 ----- doc/public/tmpl/cairo-text.sgml | 319 --------------- doc/public/tmpl/cairo-transforms.sgml | 132 ------- doc/public/tmpl/cairo-types.sgml | 43 -- doc/public/tmpl/cairo-user-fonts.sgml | 169 -------- doc/public/tmpl/cairo-version.sgml | 205 ---------- doc/public/tmpl/cairo-win32-fonts.sgml | 106 ----- doc/public/tmpl/cairo-win32.sgml | 101 ----- doc/public/tmpl/cairo-xcb-xrender.sgml | 26 -- doc/public/tmpl/cairo-xcb.sgml | 26 -- doc/public/tmpl/cairo-xlib-xrender.sgml | 60 --- doc/public/tmpl/cairo-xlib.sgml | 147 ------- doc/public/tmpl/cairo.sgml | 674 -------------------------------- src/cairo-beos-surface.cpp | 10 + src/cairo-font-face.c | 17 + src/cairo-font-options.c | 12 + src/cairo-ft-font.c | 25 ++ src/cairo-image-surface.c | 21 + src/cairo-matrix.c | 23 ++ src/cairo-misc.c | 20 + src/cairo-path.c | 9 + src/cairo-pattern.c | 17 + src/cairo-pdf-surface.c | 17 + src/cairo-png.c | 18 + src/cairo-ps-surface.c | 17 + src/cairo-quartz-font.c | 17 + src/cairo-quartz-surface.c | 17 + src/cairo-scaled-font.c | 10 + src/cairo-surface.c | 47 +++ src/cairo-svg-surface.c | 17 + src/cairo-types-private.h | 8 + src/cairo-user-font.c | 22 ++ src/cairo-version.c | 156 ++++++++ src/cairo-win32-font.c | 17 + src/cairo-win32-surface.c | 24 ++ src/cairo-xcb-surface-render.c | 13 + src/cairo-xcb-surface.c | 13 + src/cairo-xlib-surface.c | 40 ++ src/cairo.c | 57 +++ 58 files changed, 666 insertions(+), 4566 deletions(-) commit 65a1e351330106ee0af9f6f0df06b82954341ee6 Author: Benjamin Otte <otte@redhat.com> Date: Thu Jul 8 03:28:05 2010 +0200 doc: Remove CAIRO_FORMAT_RGB16_565 from hidden symbols list It's part of the cairo_format_t documentation now. doc/public/cairo-sections.txt | 1 - 1 file changed, 1 deletion(-) commit e7594aebce6007bfa2027ba2655875db6c036e73 Author: Benjamin Otte <otte@redhat.com> Date: Thu Jul 8 03:21:45 2010 +0200 doc: Fix some consistency issues that confuse gtk-doc src/cairo-region.c | 16 +++++++++++++--- src/cairo.h | 16 ++++++++++++++-- 2 files changed, 27 insertions(+), 5 deletions(-) commit da2663a0aa1e3f0a6b2f6ae2e9952ac27dec3f5d Author: Benjamin Otte <otte@redhat.com> Date: Thu Jul 8 03:14:42 2010 +0200 doc: Add list of new symbol in Cairo 1.10 doc/public/cairo-docs.xml | 3 +++ 1 file changed, 3 insertions(+) commit a106aa86a3267b4886f7eabdc4b07bb394e11531 Author: Benjamin Otte <otte@redhat.com> Date: Thu Jul 8 03:13:20 2010 +0200 doc: add --name-space option to gtkdoc-mkdb This sorts function names in new symbols lists way better. doc/public/Makefile.am | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 71d17e4b1f1cb83bc605d72d8419d4bfb824c828 Author: Benjamin Otte <otte@redhat.com> Date: Thu Jul 8 02:42:39 2010 +0200 drm: Fix Makefile for drm_xr Use cairo_drm_xr_sources/headers for the drm_xr stuff. This makes it work correctly in gtk-doc instead of conditionally adding it to cairo_drm_sources/headers. src/Makefile.sources | 10 ++++------ 1 file changed, 4 insertions(+), 6 deletions(-) commit 4b3e488cebeeace02eae1164ed12caa93d7c5847 Author: Benjamin Otte <otte@redhat.com> Date: Thu Jul 8 02:42:06 2010 +0200 skia: Add cairo-skia.h as a public header src/Makefile.sources | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 2a546286da5fe7203846d99a4a3d732c73e62db5 Author: Benjamin Otte <otte@redhat.com> Date: Thu Jul 8 02:20:17 2010 +0200 doc: Add a section about regions doc/public/cairo-docs.xml | 1 + doc/public/cairo-sections.txt | 30 ++++++++++++++++++++++++++++++ 2 files changed, 31 insertions(+) commit ab1f6d1af7c0e47077013edd58e93fa9fc6f2b5c Author: Benjamin Otte <otte@redhat.com> Date: Thu Jul 8 01:41:27 2010 +0200 doc: Add missing surface APIs doc/public/cairo-sections.txt | 2 ++ 1 file changed, 2 insertions(+) commit 90bf273a38856d178e90721168b8b076f79fbf38 Author: Benjamin Otte <otte@redhat.com> Date: Thu Jul 8 01:39:37 2010 +0200 doc: Add a section for cairo_device_t doc/public/cairo-docs.xml | 1 + doc/public/cairo-sections.txt | 18 ++++++++++++++++++ 2 files changed, 19 insertions(+) commit 60e4f28d091df8ac6b898448eb52fae9ca920d5d Author: Benjamin Otte <otte@redhat.com> Date: Thu Jul 8 01:39:07 2010 +0200 doc: Add cairo_in_clip() to cairo_t section doc/public/cairo-sections.txt | 1 + 1 file changed, 1 insertion(+) commit 6843186b729b20acddccd36184d2c5a7b5525d2e Author: Benjamin Otte <otte@redhat.com> Date: Thu Jul 8 01:38:43 2010 +0200 doc: Add cairo_rectangle_int_t to the general types list doc/public/cairo-sections.txt | 1 + 1 file changed, 1 insertion(+) commit c54dff82d3055529b7663dcff7a970cae919a9e0 Author: Benjamin Otte <otte@redhat.com> Date: Thu Jul 8 01:33:42 2010 +0200 glew: The headers are private, not public src/Makefile.sources | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 4252e3b8f915fc23611f1a667bf7bc1bb6a21dfd Author: Benjamin Otte <otte@redhat.com> Date: Thu Jul 8 01:32:25 2010 +0200 doc: Add drm and glew to the ignored directories This reduces the number of unused symbols to a reasonable size. ;) doc/public/Makefile.am | 2 ++ 1 file changed, 2 insertions(+) commit 729d00341c26d46d82359014df97472a469e49a4 Author: Benjamin Otte <otte@redhat.com> Date: Thu Jul 8 13:57:43 2010 +0200 gl: Fix compile error after rebase failure src/cairo-gl-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 292423c73376a7f9b84c9e3e578c242f486e921b Author: Colin Walters <walters@verbum.org> Date: Wed Jul 7 15:46:33 2010 -0400 Support NOCONFIGURE environment variable This is taken from gnome-common; it can be convenient for builders to separate the autogen from the configure stage. autogen.sh | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) commit 9e6c151af810886e94372ac8e4878915059c6c14 Author: Kristian Høgsberg <krh@bitplanet.net> Date: Thu Jun 17 18:19:39 2010 -0400 gl: Add new cairo-gl surface constructor to create surface for texture src/cairo-gl-private.h | 2 + src/cairo-gl-surface.c | 111 +++++++++++++++++++++++++++++++++++++++++++------ src/cairo-gl.h | 4 ++ 3 files changed, 105 insertions(+), 12 deletions(-) commit 4c91bb9a221bc8e3d65a96365bbd1157b3f4e612 Author: Benjamin Otte <otte@redhat.com> Date: Tue Jul 6 17:27:18 2010 +0200 region: Add cairo_region_xor() and cairo_region_xor_rectangle() gdk_region_xor() was a quite often used function in GDK and now that GDKe uses cairo regions, it seems like a worthwhile addition to Cairo. src/cairo-region.c | 80 ++++++++++++++++++++++++++++++++++++++++++++++++++++++ src/cairo.h | 6 ++++ src/cairoint.h | 2 ++ 3 files changed, 88 insertions(+) commit 82de6336d88be43de759b94634e87b9e4a8391b1 Author: Benjamin Otte <otte@redhat.com> Date: Tue Jul 6 00:48:36 2010 +0200 test: Add note about how to fix failures to api-special-cases test/api-special-cases.c | 6 ++++++ 1 file changed, 6 insertions(+) commit 75f269b14e1e4cdadd8c08566a07b2460ae3364b Author: Benjamin Otte <otte@redhat.com> Date: Tue Jul 6 00:41:12 2010 +0200 test: Add documentation to the api-special-cases test Hopefully people add other backend APIs to the testsuite. test/api-special-cases.c | 27 +++++++++++++++++++++++++++ 1 file changed, 27 insertions(+) commit 398870e64a18f32f1854cb19f64c34e2b1987140 Author: Benjamin Otte <otte@redhat.com> Date: Tue Jul 6 00:21:38 2010 +0200 xlib: Setters should set finished surfaces into an error status Caught by api-special-cases test src/cairo-xlib-surface.c | 18 ++++++++++++++---- 1 file changed, 14 insertions(+), 4 deletions(-) commit 6e562d1448eb70b7269f8f2f280719aa2aeb3ce7 Author: Benjamin Otte <otte@redhat.com> Date: Tue Jul 6 00:20:57 2010 +0200 xcb: Setters should set finished surfaces into an error status Caught by api-special-cases test src/cairo-xcb-surface.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) commit fa070704966f10e7264778a3ffb401054a68e8f7 Author: Benjamin Otte <otte@redhat.com> Date: Tue Jul 6 00:20:36 2010 +0200 test: Add xcb and xlib functions to api-special-cases test test/api-special-cases.c | 91 ++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 91 insertions(+) commit c4068fe9414bad663f77c030ac96fa4b4cba87ba Author: Benjamin Otte <otte@redhat.com> Date: Mon Jul 5 23:14:33 2010 +0200 test: Add ps, pdf and svg to the api-special-cases test test/api-special-cases.c | 103 +++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 103 insertions(+) commit 3e008e8ab0a1c5254bd1d8a58c9e9a89846fe7ac Author: Benjamin Otte <otte@redhat.com> Date: Mon Jul 5 23:14:19 2010 +0200 svg: Set error when extracting from finished surface Caught by api-special-cases test src/cairo-svg-surface.c | 10 ++++++++++ 1 file changed, 10 insertions(+) commit 2e66cbaee2efc4a8eb2117f945d68e2816349cfe Author: Benjamin Otte <otte@redhat.com> Date: Mon Jul 5 23:10:12 2010 +0200 ps: Allow specifying if errors on extract_surface should be set Getters should not set errors on passed in input. So we don't. Caught by api-special-cases test src/cairo-ps-surface.c | 39 ++++++++++++++++++++++----------------- 1 file changed, 22 insertions(+), 17 deletions(-) commit f5dfc3ce6eab3bf836f2d7f7cf72d2834c42d72f Author: Benjamin Otte <otte@redhat.com> Date: Mon Jul 5 23:08:08 2010 +0200 ps: Set error when extracting from finished surface Caught by api-special-cases test src/cairo-ps-surface.c | 10 ++++++++++ 1 file changed, 10 insertions(+) commit 106357f6e28467f6db02e8d69c4aea459aca781c Author: Benjamin Otte <otte@redhat.com> Date: Mon Jul 5 23:07:24 2010 +0200 pdf: Set error when extracting from finished surface Caught by api-special-cases test src/cairo-pdf-surface.c | 10 ++++++++++ 1 file changed, 10 insertions(+) commit 702aa19fdae8bd511c4374f580dde6503afd3341 Author: Benjamin Otte <otte@redhat.com> Date: Mon Jul 5 22:50:44 2010 +0200 gl: Fail properly when called with finished functions Caught by api-special-cases test src/cairo-gl-surface.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) commit b87be76bdfcbd329c79a9f11b24d22010ad15dde Author: Benjamin Otte <otte@redhat.com> Date: Mon Jul 5 22:50:18 2010 +0200 test: Add gl functions to the tests test/api-special-cases.c | 39 ++++++++++++++++++++++++++++++++++++++- 1 file changed, 38 insertions(+), 1 deletion(-) commit 448653e7b9938100c3f2e30e026fbc4d9de7d0d5 Author: Benjamin Otte <otte@redhat.com> Date: Mon Jul 5 22:41:43 2010 +0200 Call _cairo_error() when setting the FINISHED error on surfaces src/cairo-surface-subsurface.c | 2 +- src/cairo-surface.c | 12 ++++++------ 2 files changed, 7 insertions(+), 7 deletions(-) commit 19da8e138d317c1770d8bbd21a49c657d954f126 Author: Benjamin Otte <otte@redhat.com> Date: Mon Jul 5 22:40:54 2010 +0200 tee: Add proper error handling Caught by api-special-cases src/cairo-tee-surface.c | 15 +++++++++++++++ 1 file changed, 15 insertions(+) commit a9f30fc1f0c3e50715468b09707477973bf5268b Author: Benjamin Otte <otte@redhat.com> Date: Mon Jul 5 22:30:55 2010 +0200 test: Add recording and tee surface functions to api-special-cases test/api-special-cases.c | 44 ++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 44 insertions(+) commit 41abc0ca37855751ed06051a30c0d866977ed2bd Author: Benjamin Otte <otte@redhat.com> Date: Mon Jul 5 22:19:05 2010 +0200 test: add PNG functions test/api-special-cases.c | 36 ++++++++++++++++++++++++++++++++++++ 1 file changed, 36 insertions(+) commit 897eaa3e97f865ad2c28442c26efaf4ba391cf03 Author: Benjamin Otte <otte@redhat.com> Date: Mon Jul 5 22:11:23 2010 +0200 test: Add a non-error surface check In that check, we call functions that require a specific surface type with surfaces of a different type and check that they don't crash and set the correct error. test/api-special-cases.c | 36 ++++++++++++++++++++++++++++++++++++ 1 file changed, 36 insertions(+) commit 986ddf565215dc01065eade46921abe8a4b7912e Author: Benjamin Otte <otte@redhat.com> Date: Mon Jul 5 22:03:42 2010 +0200 test: add image surface functions to api-special-cases test test/api-special-cases.c | 90 +++++++++++++++++++++++++++++++++++------------- 1 file changed, 66 insertions(+), 24 deletions(-) commit 72878d70ea75be03f47fa9d53f8305e76162054e Author: Benjamin Otte <otte@redhat.com> Date: Mon Jul 5 20:47:19 2010 +0200 test: Add a new test variant to api-special-cases test This variant tests a (hopefully) inert error surface. test/api-special-cases.c | 35 +++++++++++++++++++++++++++++++++++ 1 file changed, 35 insertions(+) commit 9de25a0297c141a0776835920168afaa3ca77a87 Author: Benjamin Otte <otte@redhat.com> Date: Mon Jul 5 18:33:40 2010 +0200 configure: remove dolt libtool should be fast by now Makefile.am | 5 +- build/aclocal.dolt.m4 | 178 -------------------------------------------------- configure.ac | 1 - 3 files changed, 2 insertions(+), 182 deletions(-) commit 6744b78de17930285569da2158a334763e2bd333 Author: Benjamin Otte <otte@redhat.com> Date: Mon Jul 5 18:25:31 2010 +0200 subsurface: Fail if created from a finished surface Caught by api-special-cases test src/cairo-surface-subsurface.c | 2 ++ 1 file changed, 2 insertions(+) commit 8fd1ecea12f15c2bda710065909a14415ba92bd8 Author: Benjamin Otte <otte@redhat.com> Date: Mon Jul 5 18:25:16 2010 +0200 surface: Fail if set_mime_data() is called on a finished surface Caught by api-special-cases test src/cairo-surface.c | 2 ++ 1 file changed, 2 insertions(+) commit e388ff1ae732970d6166780817ec121cf58ccd0e Author: Benjamin Otte <otte@redhat.com> Date: Mon Jul 5 18:24:23 2010 +0200 surface: Fail early if create_similar() is called on a finished surface Caught by api-special-cases test src/cairo-surface.c | 2 ++ 1 file changed, 2 insertions(+) commit c1689ef6f27f99c36e1a5f4de658363ca0aad25a Author: Benjamin Otte <otte@redhat.com> Date: Mon Jul 5 18:23:37 2010 +0200 surface: don't call begin_modification on finished surfaces Fixes https://bugs.launchpad.net/ubuntu/+source/cairo/+bug/600622 Caught by api-special-cases test src/cairo-surface.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit cf0245c1201d4884ff19e881e9c118784fa14884 Author: Benjamin Otte <otte@redhat.com> Date: Mon Jul 5 18:17:09 2010 +0200 tests: Add a testcase that checks APIs work consistently In particular, it checks that finished and surfaces in an error state do not do bad things, like cause crashes. So far this test only checks surface APIs, but it should be extended to cover all APIs. Please update this test when new APIs get added. Motivated by https://bugs.launchpad.net/ubuntu/+source/cairo/+bug/600622 test/Makefile.sources | 1 + test/api-special-cases.c | 360 +++++++++++++++++++++++++++++++++++++++++ test/api-special-cases.ref.png | Bin 0 -> 95 bytes 3 files changed, 361 insertions(+) commit eb63284fc3ae9377003ca312be58ae8f5dbbfdbd Author: Benjamin Otte <otte@redhat.com> Date: Mon Jul 5 00:37:10 2010 +0200 configure: Use automake silent rules This replaces shave usage. If silent rules are not available, don't use them. configure.ac | 1 + 1 file changed, 1 insertion(+) commit 68bbb6b4a0a3b9bc923ccf130b1ca4221a5e54bd Author: Benjamin Otte <otte@redhat.com> Date: Mon Jul 5 00:35:39 2010 +0200 configure: remove shave build/.gitignore | 2 -- build/aclocal.shave.m4 | 77 ----------------------------------------------- build/shave-libtool.in | 69 ------------------------------------------ build/shave.in | 82 -------------------------------------------------- configure.ac | 4 --- 5 files changed, 234 deletions(-) commit 9ce87c67ff64083ec48edfefe83a1df3756c364c Author: Benjamin Otte <otte@redhat.com> Date: Sat Jul 3 02:54:55 2010 +0200 perf: print comment describing backend Use the descibe string to output information about the backend we're testing. perf/cairo-perf-trace.c | 31 ++++++++++++++++++++++++++++++- 1 file changed, 30 insertions(+), 1 deletion(-) commit fcfe7c67248ea171dc40c0dbd861e489a2df5941 Author: Benjamin Otte <otte@redhat.com> Date: Sat Jul 3 02:54:28 2010 +0200 boilerplate: Add describe string Print the pixman version we're using boilerplate/cairo-boilerplate.c | 21 ++++++++++++++++++--- 1 file changed, 18 insertions(+), 3 deletions(-) commit 0a7135148a634ea6692a4edd83193870a5d526b3 Author: Benjamin Otte <otte@redhat.com> Date: Sat Jul 3 02:54:08 2010 +0200 gl: Add describe string boilerplate/cairo-boilerplate-glx.c | 29 +++++++++++++++++++++++++---- 1 file changed, 25 insertions(+), 4 deletions(-) commit a1c4b001a548611b7d77e9280fdb5a82da275830 Author: Benjamin Otte <otte@redhat.com> Date: Wed Jun 30 18:27:54 2010 +0200 boilerplate: Add a describe vfunc This function is supposed to describe the backend in use. The describe function is optional - and therefore initialized as NULL everywhere. Note: It is well known that the xlib backend uses X. What is not known is what version the server supports or what graphics card it is running on. That is the information the describe vfunc is supposed to provide. boilerplate/Makefile.win32.features | 8 ++++++++ boilerplate/cairo-boilerplate-directfb.c | 4 ++-- boilerplate/cairo-boilerplate-drm.c | 2 ++ boilerplate/cairo-boilerplate-egl.c | 1 + boilerplate/cairo-boilerplate-glx.c | 4 ++++ boilerplate/cairo-boilerplate-pdf.c | 4 ++-- boilerplate/cairo-boilerplate-ps.c | 8 ++++---- boilerplate/cairo-boilerplate-quartz.c | 5 +++-- boilerplate/cairo-boilerplate-script.c | 2 +- boilerplate/cairo-boilerplate-skia.c | 4 ++-- boilerplate/cairo-boilerplate-svg.c | 8 ++++---- boilerplate/cairo-boilerplate-test-surfaces.c | 16 ++++++++-------- boilerplate/cairo-boilerplate-vg.c | 4 ++++ boilerplate/cairo-boilerplate-wgl.c | 2 ++ boilerplate/cairo-boilerplate-win32-printing.c | 4 ++-- boilerplate/cairo-boilerplate-win32.c | 4 ++-- boilerplate/cairo-boilerplate-xcb.c | 7 +++++++ boilerplate/cairo-boilerplate-xlib.c | 4 ++++ boilerplate/cairo-boilerplate.c | 10 +++++----- boilerplate/cairo-boilerplate.h | 4 ++++ build/Makefile.win32.features-h | 1 + src/Makefile.win32.features | 8 ++++++++ 22 files changed, 80 insertions(+), 34 deletions(-) commit 4c039b1cdfac5b5d6aa0c38f0779213fd14ec7de Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jul 2 17:01:12 2010 +0100 user-font: Check the status before destroying the context src/cairo-user-font.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 1f84f5682c6ff246b3d28c75c8731504f31c1ee1 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jul 2 13:19:17 2010 +0100 xlib: Apply translation to image surface upload. Fixes: Bug 28888 - cairo_paint with rgb images does not work correctly https://bugs.freedesktop.org/show_bug.cgi?id=28888 src/cairo-xlib-surface.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) commit d2d6c96c24501d888422ea42d3c90d3c8f3647a0 Author: Andrea Canciani <ranma42@gmail.com> Date: Tue Jun 29 09:58:54 2010 +0200 user-font: correct handling of size 0 text Text with size 0 has a singular scale matrix, thus requires special handling to avoid invalidating the context where it is used. Fixes pthread-show-text and text-zero-len (they failed with assertion when ran using the user font backend). src/cairo-matrix.c | 9 ++++++++ src/cairo-scaled-font.c | 5 ++--- src/cairo-user-font.c | 60 +++++++++++++++++++++++++++++-------------------- src/cairoint.h | 3 +++ 4 files changed, 50 insertions(+), 27 deletions(-) commit 6db247e13c6fa61879029b335c6fa649d810c11b Author: Andrea Canciani <ranma42@gmail.com> Date: Tue Jun 29 09:28:04 2010 +0200 user-font: ignore translation in scale matrix Currently the translation component of the scaled_font scale matrix is applied by gstate and only linear components should be applied in font backends. Fixes font-matrix-translation (when using user-font as font backend). src/cairo-user-font.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) commit 549431e98ed0be17b80f24f0e0c167c7ad632f4f Author: Andrea Canciani <ranma42@gmail.com> Date: Mon Jun 28 22:40:28 2010 +0200 ft-font: fix typo src/cairo-ft-font.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 97a88c5adb08a3181bad1f63a7bbdfe5784b4207 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jun 28 17:10:59 2010 +0100 xlib: Compile without fontconfig Create a cairo-fontconfig-private.h where we can add the missing defines in a common location rather than in the body of the code. src/Makefile.sources | 1 + src/cairo-fontconfig-private.h | 76 ++++++++++++++++++++++++++++++++++++++++++ src/cairo-ft-font.c | 5 +-- src/cairo-surface-subsurface.c | 1 + src/cairo-xlib-screen.c | 20 +---------- 5 files changed, 80 insertions(+), 23 deletions(-) commit 7d6910014310344e5d16c97ffd1f1c9fbc01b143 Author: Andrea Canciani <ranma42@gmail.com> Date: Mon Jun 28 17:12:41 2010 +0200 quartz: remove unused declarations src/cairo-quartz-surface.c | 17 ----------------- 1 file changed, 17 deletions(-) commit bd4c14b94eeb386b30439929e8e54a5d9b5d5fca Author: Andrea Canciani <ranma42@gmail.com> Date: Thu Jun 24 23:20:41 2010 +0200 quartz: remove unused os version info It was previously used to activate a workaround for text transformation whenrunning on some os versions. Now the workaround is not needed anymore. src/cairo-quartz-surface.c | 7 ------- 1 file changed, 7 deletions(-) commit aa7e9c43b6381930455a68258dcc3e14dce2b0e4 Author: Andrea Canciani <ranma42@gmail.com> Date: Thu Jun 24 14:18:02 2010 +0200 quartz-font: correct and explain matrix computations glyph_path was taking into account the translation (which currently is already applied in gstate) and the sign of the elements of the matrices was not explained. src/cairo-quartz-font.c | 13 ++++++------- 1 file changed, 6 insertions(+), 7 deletions(-) commit b26f72fef99e9869a62cbb8e44a82837f87cf2c6 Author: Andrea Canciani <ranma42@gmail.com> Date: Thu Jun 24 22:34:36 2010 +0200 quartz: check return status If the font backend is unable to tell the extents of the operation, consider it unbound. src/cairo-quartz-surface.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) commit 9c0d761bfcdd28d52c83d74f46dd3c709ae0fa69 Author: Andrea Canciani <ranma42@gmail.com> Date: Thu Jun 24 15:26:03 2010 +0200 quartz: improve text transform handling Use scale instead of manually compositing font_matrix and ctm and composite it with the context ctm, so that no workaround for clipping is needed anymore. src/cairo-quartz-surface.c | 46 ++++++++++++---------------------------------- 1 file changed, 12 insertions(+), 34 deletions(-) commit 9068b5768b8560fbf095f1c0eecb5c805232f794 Author: Andrea Canciani <ranma42@gmail.com> Date: Thu Jun 24 15:15:37 2010 +0200 quartz-font: silence compiler warnings Remove an unused function, explicitly ignore or check return values, don't define unused variables. src/cairo-quartz-font.c | 84 ++++++++++++++++++------------------------------- 1 file changed, 30 insertions(+), 54 deletions(-) commit ca161a585a7ce09cfcd468ea7591c5959077a907 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 26 14:50:57 2010 +0100 version: 1.9.9 Post-snapshot bump. cairo-version.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 0f1ff0daab7259ec16076f788760da4f35cb0cdc Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 26 14:20:25 2010 +0100 test: Missing ref image. test/Makefile.am | 1 + 1 file changed, 1 insertion(+) commit 45d1f461ce2d2e3af16a5a90ffd8980bffc5eaa8 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 26 14:17:23 2010 +0100 version: 1.9.10 cairo-version.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 7a68a7bfd550b5e4b6635495e7de49fe306fbad4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 26 14:16:27 2010 +0100 NEWS for 1.9.10 snapshot NEWS | 28 ++++++++++++++++++++++++++++ 1 file changed, 28 insertions(+) commit 8df41361196c7ba1b8d272507cc3d523cdcf5632 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 26 09:52:11 2010 +0100 Regenerate default win32 headers. boilerplate/Makefile.win32.features | 8 -------- build/Makefile.win32.features-h | 1 - src/Makefile.win32.features | 8 -------- 3 files changed, 17 deletions(-) commit 1373675ef7602363aeca0d9ec12be79719ea066d Author: Benjamin Otte <otte@redhat.com> Date: Fri Jun 25 14:10:12 2010 +0200 pop-group: Translate existing path properly This fixes the reverse case of the push_group case fixed a few commits ago. src/cairo.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) commit 59076776674e3ae47099f20e1d86064fbd856008 Author: Benjamin Otte <otte@redhat.com> Date: Fri Jun 25 14:08:49 2010 +0200 test: improve push-group-path-offset to also catch the pop_group() case cairo_pop_group() also trasnforms the path wrong, so we use cairo_fill() to copy data instead of cairo_paint(). (Also, unbust the testcase.) test/push-group-path-offset.c | 45 +++++++++++++++---------------------------- 1 file changed, 16 insertions(+), 29 deletions(-) commit 2e8571d95ca36f8ad9e20796990cdc0b3f763082 Author: Benjamin Otte <otte@redhat.com> Date: Fri Jun 25 13:41:22 2010 +0200 push-group: Fix path translation when pushing a group with device offsets With the previous code, the parent's device offset wasn't undone before applying the new device offset. Tested by push-group-path-offset test. src/cairo.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) commit a9b8d1a94ea624ea8b12a125bdf70c6f248e7ca7 Author: Benjamin Otte <otte@redhat.com> Date: Fri Jun 25 13:38:40 2010 +0200 test: Add a new test for a bug with device-offset tests All the credit for this one goes to make distcheck test/Makefile.sources | 1 + test/push-group-path-offset.c | 90 ++++++++++++++++++++++++++++++++++++ test/push-group-path-offset.ref.png | Bin 0 -> 186 bytes 3 files changed, 91 insertions(+) commit 1099a3515b6a69e2ff360d8e6b1df71e3ea5c213 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jun 24 17:22:34 2010 +0100 drm/intel: Fix common off-by-one by rewriting the [XY]MAX macro. Many rectangles in the command stream are specified as inclusive rects, i.e. the max values are (width-1, height-1), which is easy to neglect. src/drm/cairo-drm-i915-private.h | 2 +- src/drm/cairo-drm-i915-surface.c | 24 ++++++------------------ src/drm/cairo-drm-i965-shader.c | 12 ++++++------ src/drm/cairo-drm-intel-command-private.h | 4 ++-- 4 files changed, 15 insertions(+), 27 deletions(-) commit fb6a00571c818c738d66bc04dc75e79491878f78 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jun 14 10:19:37 2010 +0100 drm: _cairo_color_get_content() compilation fixes src/drm/cairo-drm-i915-shader.c | 2 +- src/drm/cairo-drm-i965-shader.c | 5 +---- src/drm/cairo-drm-intel.c | 3 +-- 3 files changed, 3 insertions(+), 7 deletions(-) commit 42956b46436b9704be7619e1b61ab222923376d0 Author: Zoxc <zoxc32@gmail.com> Date: Sun Jun 20 21:07:57 2010 +0000 build: Added cairo deflate stream as a requirement for the script surface. src/Makefile.sources | 3 +++ 1 file changed, 3 insertions(+) commit b036a09972a19e63a85a4661a457525e2abe4026 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Thu Jun 24 14:59:18 2010 +0300 whitespace: Fixup formal arguments and tabs in boilerplate/ and perf/. Ran a script to align the formal parameters of functions and collapse spaces to tabs in code. boilerplate/cairo-boilerplate-beos.cpp | 42 +++++----- boilerplate/cairo-boilerplate-directfb.c | 86 ++++++++++----------- boilerplate/cairo-boilerplate-drm.c | 2 +- boilerplate/cairo-boilerplate-egl.c | 20 ++--- boilerplate/cairo-boilerplate-getopt.c | 12 +-- boilerplate/cairo-boilerplate-glx.c | 60 +++++++-------- boilerplate/cairo-boilerplate-pdf.c | 24 +++--- boilerplate/cairo-boilerplate-private.h | 2 +- boilerplate/cairo-boilerplate-ps.c | 49 ++++++------ boilerplate/cairo-boilerplate-qt.cpp | 18 ++--- boilerplate/cairo-boilerplate-quartz.c | 18 ++--- boilerplate/cairo-boilerplate-scaled-font.h | 2 +- boilerplate/cairo-boilerplate-script.c | 30 ++++---- boilerplate/cairo-boilerplate-skia.c | 18 ++--- boilerplate/cairo-boilerplate-svg.c | 59 +++++++------- boilerplate/cairo-boilerplate-system.c | 10 ++- boilerplate/cairo-boilerplate-system.h | 10 ++- boilerplate/cairo-boilerplate-test-surfaces.c | 100 ++++++++++++------------ boilerplate/cairo-boilerplate-vg.c | 36 ++++----- boilerplate/cairo-boilerplate-wgl.c | 58 +++++++------- boilerplate/cairo-boilerplate-win32-printing.c | 29 +++---- boilerplate/cairo-boilerplate-win32.c | 18 ++--- boilerplate/cairo-boilerplate-xcb.c | 85 +++++++++++---------- boilerplate/cairo-boilerplate-xlib.c | 90 +++++++++++----------- boilerplate/cairo-boilerplate.c | 102 +++++++++++++------------ boilerplate/cairo-boilerplate.h | 44 ++++++----- perf/cairo-perf-chart.c | 38 ++++++--- perf/cairo-perf-compare-backends.c | 30 ++++---- perf/cairo-perf-diff-files.c | 37 +++++---- perf/cairo-perf-graph-files.c | 73 ++++++++++-------- perf/cairo-perf-graph-widget.c | 60 +++++++++------ perf/cairo-perf-graph.h | 8 +- perf/cairo-perf-micro.c | 82 ++++++++++---------- perf/cairo-perf-os2.c | 10 +-- perf/cairo-perf-posix.c | 24 +++--- perf/cairo-perf-print.c | 5 +- perf/cairo-perf-report.c | 62 +++++++++------ perf/cairo-perf-trace.c | 82 +++++++++++--------- perf/cairo-perf-win32.c | 6 +- perf/cairo-perf.h | 38 ++++----- perf/cairo-stats.c | 11 +-- perf/cairo-stats.h | 6 +- 42 files changed, 850 insertions(+), 746 deletions(-) commit 616e35d1fae703dc118ce503d25d4f57cfd16dfd Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Thu Jun 24 09:44:07 2010 +0300 boilerplate: Remove the xlib-window target from Renderless builds. I don't know why _cairo_boilerplate_xlib_window_create_surface() is in the Render-only section of the code, but since it is, this patch makes the boilerplate at least compile if we don't have Render support built in. boilerplate/cairo-boilerplate-xlib.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit aed744008dbaab49f28f9c54ad0850074ece5d66 Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk> Date: Thu Jun 24 05:38:14 2010 -0400 test: Fix typo in copyright statement test/bug-seams.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 5c7e1dd690af21943435676bfac320cfe863cf55 Author: Soeren Sandmann <sandmann@daimi.au.dk> Date: Thu Jun 24 10:11:48 2010 +0100 test: Add an example of seams. A pair of perfectly conjoint quads should not produce a seam when accumulating into a temporary mask. Yet here is an example of where they do! test/Makefile.am | 3 ++ test/Makefile.sources | 1 + test/bug-seams.c | 98 +++++++++++++++++++++++++++++++++++ test/bug-seams.ref.png | Bin 0 -> 1604 bytes test/bug-seams.xlib-fallback.ref.png | Bin 0 -> 2133 bytes test/bug-seams.xlib.ref.png | Bin 0 -> 1606 bytes 6 files changed, 102 insertions(+) commit 413ec4b708a511eb2f3866165a35d3a8d97bd326 Author: Benjamin Otte <otte@redhat.com> Date: Tue Jun 22 23:19:32 2010 +0200 gl: Return a surface from _cairo_surface_create_in_error() On error, surface creation functions should always return a surface created with _cairo_surface_create_in_error() instead of a new surface in an error state. This simplifies internal code as no refcounting has to be done. src/cairo-gl-surface.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit a4e292507cf7c2f960d040edd57b56a976c73da6 Author: Benjamin Otte <otte@redhat.com> Date: Tue Jun 22 22:01:25 2010 +0200 gl: Propagate surface creation error instead of crashing src/cairo-gl-glyphs.c | 22 ++++++++++++++++------ 1 file changed, 16 insertions(+), 6 deletions(-) commit faa4e6761c8f74a1acaa7ccc5bc8bb23b2f5cdb1 Author: Benjamin Otte <otte@redhat.com> Date: Mon Jun 21 14:01:34 2010 +0200 egl: Copy glx code to detect the proper surface in acquire src/cairo-egl-context.c | 9 +++++++++ 1 file changed, 9 insertions(+) commit 76a721f7e94fb040f106b200c09c1a7b9ae83d36 Author: Benjamin Otte <otte@redhat.com> Date: Mon Jun 21 13:30:06 2010 +0200 egl: Remove unused prev_foo usage src/cairo-egl-context.c | 7 ------- 1 file changed, 7 deletions(-) commit b8051ee985ed46c69f5fba1216ead1ebacecfd44 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 19 18:09:27 2010 +0100 script: Pass the buffer length back to freetype. After querying the required size, inform the backend of the allocated size of the buffer. src/cairo-script-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 8ac9a9df24eefc5f1125db87eb8eda2e94f54070 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 19 13:23:50 2010 +0100 xcb: Return the status upon flushing the device. The vfunc was updated to expect the status return, so supply it. src/cairo-xcb-connection.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) commit 9fffc0e120df8947c34d9e423c69ab66b763d875 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 19 11:15:13 2010 +0100 xlib: Precise and Imprecise were reversed. The intention was to switch to imprecise mode by default, using the exact rendering mode only upon request. src/cairo-xlib-surface.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) commit b322b2e61984c2e9e00fb44343ebe5e423246445 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 19 10:26:05 2010 +0100 xlib: Perform image uploads inplace. Under the restrictive conditions where we are doing an untransformed contained upload of an image into a matching surface, perform it in place. The drivers will attempt to stream such uploads efficiently, far more so than our code. src/cairo-xlib-surface.c | 112 +++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 112 insertions(+) commit c3eb95bf6670bbc06ef908481da95f3504c7dc4d Author: Benjamin Otte <otte@redhat.com> Date: Fri Jun 18 19:26:07 2010 +0200 egl: Fail properly if no configs are available instead of crashing boilerplate/cairo-boilerplate-egl.c | 4 ++++ 1 file changed, 4 insertions(+) commit 3908d80f57bc0abef47721a5b8b9afd5041118e3 Author: Benjamin Otte <otte@redhat.com> Date: Fri Jun 18 16:11:20 2010 +0200 glx: Remove useless optimization trying to avoid glXMakeCurrent() The optimization is not performance-relevant. And having less code is always a good idea. src/cairo-glx-context.c | 19 ++----------------- 1 file changed, 2 insertions(+), 17 deletions(-) commit 64f90322f73c37ac5667292949bb45b0279239d9 Author: Benjamin Otte <otte@redhat.com> Date: Fri Jun 18 12:40:48 2010 +0200 gl: Refactor status handling in _cairo_gl_context_release() Previously, the code returned a status and required the caller to mangle this status with his own status. Now, the function takes the previous status ass an argument and does the mangling itself. Also contains fixes for all the callers to actually check the return value - which is now rather trivial as it just requires passing through the status variable. src/cairo-gl-composite.c | 6 ++---- src/cairo-gl-device.c | 2 +- src/cairo-gl-glyphs.c | 14 +++++--------- src/cairo-gl-gradient.c | 3 ++- src/cairo-gl-private.h | 14 ++++++++------ src/cairo-gl-surface.c | 47 +++++++++++++++++++++++++---------------------- 6 files changed, 43 insertions(+), 43 deletions(-) commit fc3d521c121da237974e486f1b7735009764b441 Author: Benjamin Otte <otte@redhat.com> Date: Fri Jun 18 12:23:37 2010 +0200 gl: Inline the check_error() function src/cairo-gl-private.h | 23 +++++++---------------- 1 file changed, 7 insertions(+), 16 deletions(-) commit 9d7fa289132650261e546b39af2371c262f46d3c Author: Benjamin Otte <otte@redhat.com> Date: Fri Jun 18 12:18:05 2010 +0200 gl: Don't assert if there used to be a GL error When acquiring the GL context, do not assert that the GL context is not in an error state. Do not even call _cairo_error(). Handling GL errors in other code is not Cairo's responsibility. Instead just clear all previous errors so we don't accidentally set surfaces into error states to unrelated errors. src/cairo-gl-private.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit 0f9a8cd18f14b1f23aaefe14db3b5ad07c84cff7 Author: Benjamin Otte <otte@redhat.com> Date: Fri Jun 18 12:17:36 2010 +0200 gl: Add a custom glGetError() function This function clears all errors and returns the first one that happened. src/cairo-gl-private.h | 19 ++++++++++++++++--- 1 file changed, 16 insertions(+), 3 deletions(-) commit 8048d3aa0a11ab1c054887682b8b2a899a87da0e Author: Benjamin Otte <otte@redhat.com> Date: Fri Jun 18 10:50:21 2010 +0200 gl: Remove custom fprintf fudging on GL errors Now that we probably call _cairo_error() on every GL error, there is no need to use custom methods to catch those errors. The usual breakpoint is enough. src/cairo-gl-private.h | 7 +------ src/cairo-gl-surface.c | 20 -------------------- 2 files changed, 1 insertion(+), 26 deletions(-) commit f4da5048bf87df0651ec83ca3f1ad6a5af7eb16d Author: Benjamin Otte <otte@redhat.com> Date: Fri Jun 18 10:44:15 2010 +0200 gl: cairo_warn about return value from cairo_gl_context_release() src/cairo-gl-private.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit d9179f480a43c7171806e2c33804aaae70a2cfa6 Author: Benjamin Otte <otte@redhat.com> Date: Fri Jun 18 00:43:09 2010 +0200 boilerplate: Reinstate glFinish() for wgl It was accidentally disabled. boilerplate/cairo-boilerplate-wgl.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 67e84d61230a2012c18d8bff60e8835aaa57be07 Author: Kristian Høgsberg <krh@bitplanet.net> Date: Thu Jun 17 17:38:00 2010 -0400 egl: Implement acquire and release for EGL backend src/cairo-egl-context.c | 46 +++++++++++++++++++++++++++++++++++++--------- 1 file changed, 37 insertions(+), 9 deletions(-) commit 7ab9ce1b9130dcd63b2b2b2516b77ef1ae531144 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jun 17 15:47:04 2010 +0100 image: Make a local copy of the trap coordinates as the boxes alias. As we transform the array of trapezoids into an array of boxes in-place, we must take local copies of the coordinates before writing into the boxes otherwise we may inadvertently modify the trapezoidal coordinates. Fixes test/a1-bug. src/cairo-image-surface.c | 15 +++++++++++---- 1 file changed, 11 insertions(+), 4 deletions(-) commit f2645fa361003636d1da6ca38c3b61fef1b1ce4d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jun 17 15:38:53 2010 +0100 test: Add a1-bug Capture a bug report from "cairouser" where a1 rasterisation is now broken. test/Makefile.am | 3 +++ test/Makefile.sources | 1 + test/a1-bug.c | 61 ++++++++++++++++++++++++++++++++++++++++++++ test/a1-bug.image16.ref.png | Bin 0 -> 3329 bytes test/a1-bug.ref.png | Bin 0 -> 3736 bytes test/a1-bug.xlib.ref.png | Bin 0 -> 3362 bytes 6 files changed, 65 insertions(+) commit 800e977c2ea1e097c4b7332b18a5fc03089603c1 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Thu Jun 17 14:27:13 2010 +0300 perf: Allow comparing perf reports with only one difference. The cairo-perf-diff-files tool would ignore perf reports with just one test for no apparent reason. The traces take so long that it's useful to be able to compare runs with just one trace. perf/cairo-perf-diff-files.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 29b8cfd2b6cf590fcfe64e5ba5b4479c2748ac06 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jun 17 10:11:31 2010 +0100 test/halo: Remove copied code. How did that get there? I obviously ran the test to generate the output, so where did it spring as it is the old version of the function? From 16364768d79570e7201a87f0a985c65acaff5560. Mysteries will never cease. test/halo.c | 36 ------------------------------------ 1 file changed, 36 deletions(-) commit a9a22649e2ffa1cd3720c8555dc3c6aff2040e3d Author: Zoxc <zoxc32@gmail.com> Date: Wed Jun 16 21:33:30 2010 +0200 wgl: Only reset GL context on the dummy window. This changes the WGL context destruction code to only reset the GL context for the dummy window, since doing it on <nothing> is an invalid operation. src/cairo-wgl-context.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) commit ca35e09ece5107c3ebc9f111d4a3d96d69aa3d03 Author: Zoxc <zoxc32@gmail.com> Date: Mon Jun 14 20:34:52 2010 +0200 gl: Added wgl boilerplate and moved the glx and egl boilerplates into their own files. boilerplate/Makefile.sources | 4 +- boilerplate/Makefile.win32.features | 8 + boilerplate/cairo-boilerplate-egl.c | 141 ++++++++++++ ...ro-boilerplate-gl.c => cairo-boilerplate-glx.c} | 91 -------- boilerplate/cairo-boilerplate-wgl.c | 236 +++++++++++++++++++++ build/Makefile.win32.features-h | 1 + src/Makefile.win32.features | 8 + 7 files changed, 397 insertions(+), 92 deletions(-) commit 2af3262895d74492a0f856f049d1524a527257bb Author: Zoxc <zoxc32@gmail.com> Date: Mon Jun 14 20:33:17 2010 +0200 wgl: Renamed hwnd field to wnd. src/cairo-wgl-context.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) commit 7a023a62f7517ad0d54f4d59c99909fadcc05e82 Author: Nicolaus L Helper <nlhepler@gmail.com> Date: Thu Jun 17 08:56:30 2010 +0100 ft,fc,xlib: LCD filtering patch. This adds internal API to retrieve the LCD filtering parameters from fontconfig, or as set on the Screen, and feed them to FreeType when rendering the glyph. References: Bug 10301 - LCD filtering patch https://bugs.freedesktop.org/show_bug.cgi?id=10301 Tested-by: Brandon Wright <bearoso@gmail.com> Forward-ported-by: Robert Hooker <sarvatt@gmail.cm> ickle: The API is clearly not ready for public consumption, the enum are poorly named, however this stands by itself as enabling system wide properties. src/cairo-font-options.c | 51 +++- src/cairo-ft-font.c | 675 +++++++++++++++++++++++++++++++++------------- src/cairo-surface.c | 1 + src/cairo-types-private.h | 25 ++ src/cairo-xlib-screen.c | 28 ++ src/cairoint.h | 7 + 6 files changed, 597 insertions(+), 190 deletions(-) commit 36b4b0631cc220d01c411b596a4eef839338cd7c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jun 17 08:50:31 2010 +0100 image: Use consistent rounding modes for a1 rasterisation. When checking if the traps are equivalent to a set of rectangles, we need to use the same rounding mode as when converting the traps to a set of boxes and then filling them. Failure to do leads to a situation where (-127,-128) was thought to be equivalent to (0,0) but we attempted to fill from 0 to -1 instead. References: http://lists.cairographics.org/archives/cairo/2010-June/020115.html src/cairo-image-surface.c | 58 ++++++++++++++++++++++++++++------------------- 1 file changed, 35 insertions(+), 23 deletions(-) commit 3306bcb1d91265d60c460aa64d3ee4a4acb430a1 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jun 17 08:47:48 2010 +0100 fixed: Refactor code to use more inlines and less duplication of logic. This also has the side-effect of fixing the types in the problematic functions which hid Andrea's true fix for the unsigned FRAC_MASK. src/cairo-fixed-private.h | 18 +++++++++++++++--- 1 file changed, 15 insertions(+), 3 deletions(-) commit c0dee7964c4394b7963041f246855fd3b01f4ebb Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jun 16 18:49:58 2010 +0100 test: Add a1-rasterisation Check the rounding criteria when rasterising. test/Makefile.am | 2 + test/Makefile.sources | 1 + test/a1-rasterisation-rectangles.ref.png | Bin 0 -> 1729 bytes test/a1-rasterisation-triangles.ref.png | Bin 0 -> 1729 bytes test/a1-rasterisation.c | 101 +++++++++++++++++++++++++++++++ 5 files changed, 104 insertions(+) commit 16364768d79570e7201a87f0a985c65acaff5560 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jun 16 15:09:22 2010 +0100 test/halo: Transformation seems key to the bug... test/halo.c | 36 ++++++++++++++++++++++++++++++++++++ 1 file changed, 36 insertions(+) commit cb28e0bfeb47a72f14ee42f161ebf6b5eb61e76b Author: Andrea Canciani <ranma42@gmail.com> Date: Thu Jun 17 08:42:39 2010 +0200 test: correct options in getopt call 'k' was missing and 't' was present but unhandled. Options have been reordered according to the "USAGE" ordering. test/cairo-test-runner.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 3cd07559328b60e3da85debb805cb4a3fc4abc22 Author: Andrea Canciani <ranma42@gmail.com> Date: Wed Jun 16 22:54:08 2010 +0200 correct rounding computation cairo_fixed_integer_round[_down] were adding an unsigned mask value before shifting its result, causing the shift to be computed as logical (unsigned) right shift, thus producing incorrect values for negative inputs. Making the mask value signed fixes this issue. Bug report by cu: http://lists.cairographics.org/archives/cairo/2010-June/020115.html src/cairo-fixed-private.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 8689d75caaa9c6ee42976873d21684fa9f48834f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jun 16 15:09:22 2010 +0100 test/halo: Transformation seems key to the bug... Only it is actually transformation of the result that is required in order to magnify the effect to a detectable level. test/Makefile.am | 5 +++ test/halo-transform.image16.ref.png | Bin 0 -> 10983 bytes test/halo-transform.ps.ref.png | Bin 0 -> 8691 bytes test/halo-transform.ref.png | Bin 0 -> 15086 bytes test/halo-transform.xlib.ref.png | Bin 0 -> 15346 bytes test/halo.c | 64 ++++++++++++++++++++++++++++++++++-- test/halo.image16.ref.png | Bin 5328 -> 6172 bytes test/halo.ps.ref.png | Bin 0 -> 5220 bytes test/halo.ref.png | Bin 7524 -> 8839 bytes test/halo.xlib.ref.png | Bin 7404 -> 8835 bytes 10 files changed, 67 insertions(+), 2 deletions(-) commit 71936a1f414807a51c68de111362c2d5499bdfa7 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jun 15 21:36:15 2010 +0100 test/aliasing: Add a few more colour combinations. In particular, compare red against its inverse, cyan. Etc. test/aliasing.c | 6 +++++- test/aliasing.image16.ref.png | Bin 20810 -> 97251 bytes test/aliasing.ref.png | Bin 22980 -> 104672 bytes test/aliasing.xlib.ref.png | Bin 24018 -> 104315 bytes 4 files changed, 5 insertions(+), 1 deletion(-) commit e2220a1f3a7b483e615f5bb993a014d20b0f788a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jun 15 11:45:54 2010 +0100 configure: Try an alternate name for system glew. configure.ac | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) commit 25ec0c1e9b5af2ba68200b79b252de5c1b371d0a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jun 15 10:59:01 2010 +0100 build: Enable system GLEW on non-UNIX platforms configure.ac | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) commit 52624446636bb5941ca2180afa2c0be275414cec Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jun 15 10:40:36 2010 +0100 glew: Define NULL src/glew/glew.c | 4 ++++ 1 file changed, 4 insertions(+) commit 56d06b088b9c355ca76a38191717a8e03aa5f59a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jun 14 22:58:38 2010 +0100 test: Add an aliasing test case. Just a simple test in drawing neighbouring edges and looking for the effects of aliasing. test/Makefile.am | 3 ++ test/Makefile.sources | 1 + test/aliasing.c | 98 ++++++++++++++++++++++++++++++++++++++++++ test/aliasing.image16.ref.png | Bin 0 -> 20810 bytes test/aliasing.ref.png | Bin 0 -> 22980 bytes test/aliasing.xlib.ref.png | Bin 0 -> 24018 bytes 6 files changed, 102 insertions(+) commit 4b39f02bf7e8b4921518db4a2d8564514cb323a9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jun 14 20:38:26 2010 +0100 test: Add halo halo uses text_path + stroke + (fill | show_text) to generate a "halo" around the string. This is to try to replicate a bug described by Ian Britten in the PDF backend where the rendering of the show_text looked distorted compared to the stroke. test/Makefile.am | 3 ++ test/Makefile.sources | 1 + test/halo.c | 98 ++++++++++++++++++++++++++++++++++++++++++++++ test/halo.image16.ref.png | Bin 0 -> 5328 bytes test/halo.ref.png | Bin 0 -> 7524 bytes test/halo.xlib.ref.png | Bin 0 -> 7404 bytes 6 files changed, 102 insertions(+) commit 072a200ec502bf5ac3e62702e6f14ed18171b1e0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jun 14 16:05:14 2010 +0100 build: Add wgl to Makefile.sources and regenerate features. boilerplate/Makefile.win32.features | 10 ++++++++++ build/Makefile.win32.features | 1 + build/Makefile.win32.features-h | 3 +++ src/Makefile.sources | 6 ++++-- src/Makefile.win32.features | 14 ++++++++++++++ 5 files changed, 32 insertions(+), 2 deletions(-) commit 4a0bd91ff77c8afddde6d09495bc9ca7242f953d Author: Dmitri Vorobiev <dmitri.vorobiev@movial.com> Date: Mon Jun 14 16:40:39 2010 +0300 scaled-font: fine-tune caching This patch implements the ideas outlined by Behdad Esfahbod in the following mailing list message: http://lists.cairographics.org/archives/cairo/2010-June/020065.html Specifically, two things have been adjusted. First, the size of the look-up table was reduced to 64. Second, cache codepath is now bypassed for strings that are shorter than 16, not only for one-character strings. This allowed us to reduce the LUT initialization overhead while still retaining the advantage of caching for common-case string sizes. We have experimented with different LUT sizes, and it came out that the size of 64 is the best one in view of speed, at least for our language-neutral benchmark, which generated random strings of printable ASCII characters. Below is a table presenting benchmark results for different values of LUT size: =============================================================================== Benchmark | [1] | [2] | [3] | [4] | [5] | [6] | [7] =============================================================================== 8px text, 1 chars | 0.41 | 0.41 | 0 | 0.41 | 0 | 0.41 | 0 8px text, 10 chars | 2.13 | 2.21 | 3.76 | 2.19 | 2.82 | 2.09 | -1.88 8px text, 20 chars | 2.97 | 3.04 | 2.36 | 3.01 | 1.35 | 2.98 | 0.34 12px text, 1 chars | 0.94 | 0.94 | 0 | 0.95 | 1.06 | 0.94 | 0 12px text, 10 chars | 4.73 | 4.89 | 3.38 | 4.9 | 3.59 | 4.82 | 1.9 12px text, 20 chars | 6.32 | 6.42 | 1.58 | 6.46 | 2.22 | 6.32 | 0 16px text, 1 chars | 1.75 | 1.76 | 0.57 | 1.77 | 1.14 | 1.76 | 0.57 16px text, 10 chars | 8.13 | 8.45 | 3.94 | 8.43 | 3.69 | 8.44 | 3.81 16px text, 20 chars | 10.41 | 10.69 | 2.69 | 10.64 | 2.21 | 10.65 | 2.31 24px text, 1 chars | 3.3 | 3.3 | 0 | 3.32 | 0.61 | 3.3 | 0 24px text, 10 chars | 14.68 | 14.97 | 1.98 | 14.97 | 1.98 | 14.87 | 1.29 24px text, 20 chars | 17.93 | 18.01 | 0.45 | 18.06 | 0.73 | 17.81 | -0.67 96px text, 1 chars | 23.65 | 23.38 | -1.14 | 23.74 | 0.38 | 23.65 | 0 96px text, 5 chars | 50.52 | 51.34 | 1.62 | 51.48 | 1.9 | 51.41 | 1.76 96px text, 10 chars | 57.5 | 58.11 | 1.06 | 58.27 | 1.34 | 58.04 | 0.94 =============================================================================== [1]: Git head, Mpix/s [2]: {GLYPH_LUT_SIZE = 32, CACHING_THRESHOLD = 16} [3]: Gain of {32, 16} w.r.t. Git head [4]: {GLYPH_LUT_SIZE = 64, CACHING_THRESHOLD = 16} [5]: Gain of {64, 16} w.r.t. Git head [6]: {GLYPH_LUT_SIZE = 128, CACHING_THRESHOLD = 16} [7]: Gain of {128, 16} w.r.t. Git head The benchmark itself can be found from this mailing list message: http://lists.cairographics.org/archives/cairo/2010-June/020064.html src/cairo-scaled-font.c | 61 ++++++++++++++++++++++++++++++++++++------------- 1 file changed, 45 insertions(+), 16 deletions(-) commit 505a0456d2498112155db7e7f275a14dc98f643e Author: Zoxc <zoxc32@gmail.com> Date: Mon Jun 14 01:20:54 2010 +0200 gl: Added WGL context and surface. build/configure.ac.features | 1 + configure.ac | 9 ++ src/cairo-gl.h | 16 +++ src/cairo-wgl-context.c | 253 ++++++++++++++++++++++++++++++++++++++++++++ 4 files changed, 279 insertions(+) commit fd6c38b9e006feefa20ce4f54d3108dad51c828d Author: Zoxc <zoxc32@gmail.com> Date: Mon Jun 14 01:04:10 2010 +0200 win32: Fixed compile errors in Windows backend. src/cairo-win32-printing-surface.c | 9 ++++----- src/cairo-win32-surface.c | 4 ++-- 2 files changed, 6 insertions(+), 7 deletions(-) commit 5f10139cd92383c657e328ffad41874c4f668551 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Jun 13 12:33:10 2010 +0100 NEWS: A couple more tweaks I only spotted these in the release announcement. NEWS | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) commit 4932aaa3173799c8f3ac30e4c9a594c20c83a121 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Jun 13 12:02:53 2010 +0100 version: 1.9.9 Post-snapshot bump. cairo-version.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 3a20b10cd0d94406fbd5fe3bb3d4820a95364537 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Jun 13 10:27:03 2010 +0100 version: 1.9.8 Snapshot time! cairo-version.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit b612a230865a1a62f9cdcb051de9b57b509646b2 Author: Andrea Canciani <ranma42@gmail.com> Date: Sun Jun 13 12:29:24 2010 +0200 test: update quartz ref images in makefile test/Makefile.am | 143 ++++++++++++++++++++++++++++++++++++++++++------------- 1 file changed, 109 insertions(+), 34 deletions(-) commit 0693af2328606a4c7e4c888d2a1cd346e6bef405 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Jun 13 10:22:37 2010 +0100 test: Move all output below test/output/ make distcheck complains of remanents being left under test/ after a clean, notably the files used to check the capabilities of a similar surface and the fallback-resolution output. test/cairo-test.c | 53 ++++++++++++++++++++++++++-------------------- test/fallback-resolution.c | 30 ++++++++++++++++++++++---- 2 files changed, 56 insertions(+), 27 deletions(-) commit ffc3a39c95e918fa2fcd14c1b1063b9a9acceebf Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 12 20:04:36 2010 +0100 NEWS: Minor tweaks. NEWS | 46 +++++++++++++++++++++++----------------------- 1 file changed, 23 insertions(+), 23 deletions(-) commit b6e16b8d2950cdadf2e2b14b0e7502d4a9ebdc09 Author: Andrea Canciani <ranma42@gmail.com> Date: Sat Jun 12 10:33:35 2010 +0200 test: quartz ref image refresh test/a1-image-sample.quartz.xfail.png | Bin 0 -> 972 bytes test/a1-mask-sample.quartz.xfail.png | Bin 0 -> 972 bytes test/a1-traps-sample.quartz.xfail.png | Bin 0 -> 122 bytes test/big-line.quartz.ref.png | Bin 0 -> 993 bytes test/caps-joins-curve.quartz.ref.png | Bin 0 -> 5199 bytes test/clear.quartz.argb32.ref.png | Bin 0 -> 691 bytes test/clear.quartz.rgb24.ref.png | Bin 0 -> 606 bytes test/clip-disjoint.quartz.ref.png | Bin 0 -> 5476 bytes test/clip-fill-rule.quartz.rgb24.ref.png | Bin 0 -> 363 bytes test/clip-fill-unbounded.quartz.argb32.ref.png | Bin 0 -> 1436 bytes test/clip-fill-unbounded.quartz.rgb24.ref.png | Bin 0 -> 1368 bytes test/clip-fill.quartz.ref.png | Bin 0 -> 875 bytes test/clip-group-shapes-circles.quartz.ref.png | Bin 0 -> 1518 bytes ....ref.png => clip-nesting.quartz.argb32.ref.png} | Bin 1067 -> 1048 bytes test/clip-nesting.quartz.rgb24.ref.png | Bin 956 -> 937 bytes test/clip-operator.quartz.argb32.ref.png | Bin 0 -> 9586 bytes test/clip-operator.quartz.ref.png | Bin 9539 -> 0 bytes test/clip-operator.quartz.rgb24.ref.png | Bin 4412 -> 4104 bytes test/clip-push-group.quartz.ref.png | Bin 203 -> 166 bytes test/clip-shape.quartz.ref.png | Bin 0 -> 3229 bytes test/clip-stroke-unbounded.quartz.argb32.ref.png | Bin 0 -> 1493 bytes test/clip-stroke-unbounded.quartz.rgb24.ref.png | Bin 0 -> 1356 bytes test/clip-stroke.quartz.ref.png | Bin 0 -> 1305 bytes test/clip-text.quartz.ref.png | Bin 0 -> 854 bytes test/clip-twice.quartz.argb32.ref.png | Bin 0 -> 1171 bytes test/clip-twice.quartz.ref.png | Bin 1171 -> 0 bytes test/clip-twice.quartz.rgb24.ref.png | Bin 1111 -> 1095 bytes test/clipped-group.quartz.ref.png | Bin 0 -> 308 bytes test/culled-glyphs.quartz.ref.png | Bin 0 -> 493 bytes test/dash-caps-joins.quartz.ref.png | Bin 4727 -> 0 bytes test/dash-caps-joins.quartz.xfail.png | Bin 0 -> 4446 bytes test/dash-curve.quartz.ref.png | Bin 47965 -> 0 bytes test/dash-curve.quartz.xfail.png | Bin 0 -> 42238 bytes test/dash-scale.quartz.ref.png | Bin 8079 -> 7614 bytes test/dash-state.quartz.ref.png | Bin 8698 -> 0 bytes test/dash-state.quartz.xfail.png | Bin 0 -> 6957 bytes test/degenerate-arc.quartz.ref.png | Bin 0 -> 552 bytes test/degenerate-curve-to.quartz.ref.png | Bin 0 -> 247 bytes test/degenerate-dash.quartz.xfail.png | Bin 0 -> 1560 bytes test/degenerate-path.quartz.argb32.xfail.png | Bin 0 -> 200 bytes test/degenerate-path.quartz.ref.png | Bin 232 -> 0 bytes test/degenerate-path.quartz.rgb24.ref.png | Bin 198 -> 0 bytes test/degenerate-path.quartz.rgb24.xfail.png | Bin 0 -> 172 bytes test/degenerate-pen.quartz.ref.png | Bin 1144 -> 0 bytes test/degenerate-rel-curve-to.quartz.ref.png | Bin 0 -> 246 bytes test/extend-pad-border.quartz.ref.png | Bin 0 -> 432 bytes test/extend-pad-similar.quartz.xfail.png | Bin 0 -> 270 bytes test/extend-pad.quartz.xfail.png | Bin 0 -> 270 bytes test/fill-alpha-pattern.quartz.ref.png | Bin 0 -> 7943 bytes test/fill-alpha.quartz.ref.png | Bin 0 -> 2704 bytes test/fill-and-stroke-alpha-add.quartz.ref.png | Bin 603 -> 542 bytes test/fill-and-stroke-alpha.quartz.ref.png | Bin 550 -> 0 bytes test/fill-and-stroke.quartz.argb32.ref.png | Bin 0 -> 273 bytes test/fill-and-stroke.quartz.ref.png | Bin 286 -> 0 bytes test/fill-and-stroke.quartz.rgb24.ref.png | Bin 264 -> 249 bytes ...ill-degenerate-sort-order.quartz.argb32.ref.png | Bin 0 -> 2451 bytes test/fill-degenerate-sort-order.quartz.ref.png | Bin 2470 -> 0 bytes ...fill-degenerate-sort-order.quartz.rgb24.ref.png | Bin 2098 -> 2079 bytes test/fill-image.quartz.ref.png | Bin 0 -> 1297 bytes ...rtz.ref.png => fill-rule.quartz.argb32.ref.png} | Bin 2079 -> 2060 bytes test/fill-rule.quartz.rgb24.ref.png | Bin 1790 -> 1771 bytes test/filter-bilinear-extents.quartz.xfail.png | Bin 0 -> 308 bytes test/filter-nearest-offset.quartz.xfail.png | Bin 0 -> 2130 bytes test/filter-nearest-transformed.quartz.xfail.png | Bin 0 -> 534 bytes test/finer-grained-fallbacks.quartz.argb32.ref.png | Bin 0 -> 988 bytes test/finer-grained-fallbacks.quartz.rgb24.ref.png | Bin 0 -> 763 bytes test/font-matrix-translation.quartz.ref.png | Bin 1090 -> 996 bytes test/ft-show-glyphs-table.quartz.xfail.png | Bin 0 -> 827 bytes .../ft-text-vertical-layout-type1.quartz.xfail.png | Bin 0 -> 3635 bytes test/ft-text-vertical-layout-type3.quartz.ref.png | Bin 0 -> 3573 bytes test/glyph-cache-pressure.quartz.ref.png | Bin 3654 -> 3284 bytes test/gradient-constant-alpha.quartz.argb32.ref.png | Bin 0 -> 214 bytes test/group-unaligned.quartz.ref.png | Bin 0 -> 363 bytes test/huge-radial.quartz.ref.png | Bin 0 -> 84690 bytes test/inverse-text.quartz.ref.png | Bin 0 -> 2565 bytes test/joins.quartz.ref.png | Bin 0 -> 5833 bytes test/leaky-dash.quartz.ref.png | Bin 241 -> 195 bytes test/leaky-dashed-rectangle.quartz.ref.png | Bin 0 -> 339 bytes test/leaky-dashed-stroke.quartz.ref.png | Bin 0 -> 9229 bytes test/line-width-scale.quartz.ref.png | Bin 6434 -> 5623 bytes test/linear-gradient-reflect.quartz.ref.png | Bin 878 -> 780 bytes test/linear-gradient-subset.quartz.ref.png | Bin 0 -> 887 bytes test/linear-gradient.quartz.ref.png | Bin 1164 -> 1060 bytes test/long-dashed-lines.quartz.ref.png | Bin 2313 -> 2068 bytes test/mask-alpha.quartz.argb32.ref.png | Bin 641 -> 622 bytes test/mask-transformed-image.quartz.ref.png | Bin 0 -> 3909 bytes test/mask-transformed-similar.quartz.ref.png | Bin 0 -> 3909 bytes test/mask.quartz.argb32.ref.png | Bin 0 -> 10575 bytes test/mask.quartz.ref.png | Bin 10682 -> 0 bytes test/mask.quartz.rgb24.ref.png | Bin 8418 -> 8354 bytes test/new-sub-path.quartz.ref.png | Bin 383 -> 370 bytes test/new-sub-path.quartz.rgb24.ref.png | Bin 357 -> 0 bytes test/operator-clear.quartz.argb32.ref.png | Bin 0 -> 1228 bytes test/operator-clear.quartz.ref.png | Bin 1459 -> 0 bytes test/operator-clear.quartz.rgb24.ref.png | Bin 1293 -> 1096 bytes test/operator-source.quartz.argb32.ref.png | Bin 0 -> 4754 bytes test/operator-source.quartz.ref.png | Bin 4637 -> 0 bytes test/operator-source.quartz.rgb24.ref.png | Bin 3533 -> 3535 bytes test/over-above-source.quartz.argb32.ref.png | Bin 0 -> 511 bytes test/over-above-source.quartz.ref.png | Bin 530 -> 0 bytes test/over-above-source.quartz.rgb24.ref.png | Bin 456 -> 437 bytes ...ng => over-around-source.quartz.argb32.ref.png} | Bin 612 -> 593 bytes test/over-around-source.quartz.rgb24.ref.png | Bin 491 -> 0 bytes ...g => over-between-source.quartz.argb32.ref.png} | Bin 570 -> 551 bytes test/over-between-source.quartz.rgb24.ref.png | Bin 467 -> 0 bytes test/path-append.quartz.ref.png | Bin 0 -> 6395 bytes test/pthread-same-source.quartz.xfail.png | Bin 0 -> 957 bytes test/pthread-show-text.quartz.ref.png | Bin 0 -> 42996 bytes test/push-group-color.quartz.ref.png | Bin 0 -> 3127 bytes test/push-group.quartz.argb32.ref.png | Bin 0 -> 3932 bytes test/push-group.quartz.rgb24.ref.png | Bin 0 -> 3659 bytes test/radial-gradient.quartz.ref.png | Bin 92334 -> 0 bytes test/random-intersections-curves-eo.quartz.ref.png | Bin 0 -> 326366 bytes test/random-intersections-curves-nz.quartz.ref.png | Bin 0 -> 324940 bytes test/random-intersections-eo.quartz.ref.png | Bin 197778 -> 179179 bytes test/random-intersections-nonzero.quartz.ref.png | Bin 0 -> 180871 bytes ...recording-surface-pattern.quartz.argb32.ref.png | Bin 0 -> 4570 bytes test/recording-surface-pattern.quartz.ref.png | Bin 4572 -> 0 bytes .../recording-surface-pattern.quartz.rgb24.ref.png | Bin 4063 -> 4058 bytes test/rectilinear-dash.quartz.xfail.png | Bin 0 -> 494 bytes test/rectilinear-stroke.quartz.xfail.png | Bin 0 -> 209 bytes test/reflected-stroke.quartz.ref.png | Bin 5293 -> 0 bytes test/rel-path.quartz.ref.png | Bin 212 -> 0 bytes test/rel-path.quartz.rgb24.ref.png | Bin 228 -> 0 bytes test/rotate-image-surface-paint.quartz.ref.png | Bin 257 -> 356 bytes test/rotated-clip.quartz.ref.png | Bin 0 -> 3923 bytes test/rounded-rectangle-fill.quartz.ref.png | Bin 0 -> 684 bytes test/scale-offset-image.quartz.ref.png | Bin 0 -> 9054 bytes test/scale-offset-similar.quartz.ref.png | Bin 0 -> 9054 bytes test/select-font-face.quartz.ref.png | Bin 2970 -> 2645 bytes test/self-intersecting.quartz.xfail.png | Bin 0 -> 233 bytes test/show-text-current-point.quartz.ref.png | Bin 3093 -> 2773 bytes test/smask-fill.quartz.ref.png | Bin 0 -> 2076 bytes test/smask-mask.quartz.ref.png | Bin 0 -> 3357 bytes test/smask-paint.quartz.ref.png | Bin 0 -> 3499 bytes test/smask-stroke.quartz.ref.png | Bin 0 -> 1697 bytes test/smask-text.quartz.ref.png | Bin 0 -> 1904 bytes test/smask.quartz.ref.png | Bin 0 -> 4067 bytes test/source-clip-scale.quartz.ref.png | Bin 148 -> 126 bytes test/spline-decomposition.quartz.xfail.png | Bin 0 -> 19560 bytes test/stroke-ctm-caps.quartz.ref.png | Bin 1109 -> 899 bytes test/stroke-image.quartz.ref.png | Bin 1802 -> 1350 bytes test/subsurface-pad.quartz.xfail.png | Bin 0 -> 197 bytes test/surface-pattern-big-scale-down.quartz.ref.png | Bin 0 -> 258 bytes ...urface-pattern-operator.quartz.argb32.xfail.png | Bin 0 -> 5917 bytes ...surface-pattern-operator.quartz.rgb24.xfail.png | Bin 0 -> 2546 bytes ...e-pattern-scale-down-extend-none.quartz.ref.png | Bin 0 -> 391 bytes ...-pattern-scale-down-extend-pad.quartz.xfail.png | Bin 0 -> 397 bytes ...attern-scale-down-extend-reflect.quartz.ref.png | Bin 0 -> 475 bytes ...pattern-scale-down-extend-repeat.quartz.ref.png | Bin 0 -> 397 bytes test/surface-pattern-scale-down.quartz.ref.png | Bin 3188 -> 2276 bytes test/surface-pattern-scale-up.quartz.xfail.png | Bin 0 -> 3294 bytes test/surface-pattern.quartz.xfail.png | Bin 0 -> 9216 bytes test/text-antialias-gray.quartz.ref.png | Bin 1041 -> 956 bytes test/text-antialias-none.quartz.ref.png | Bin 300 -> 270 bytes test/text-antialias-subpixel.quartz.ref.png | Bin 1065 -> 949 bytes test/text-glyph-range.quartz.ref.png | Bin 0 -> 2191 bytes test/text-pattern.quartz.argb32.ref.png | Bin 0 -> 2148 bytes test/text-pattern.quartz.ref.png | Bin 2158 -> 0 bytes test/text-pattern.quartz.rgb24.ref.png | Bin 1874 -> 1867 bytes test/text-rotate.quartz.ref.png | Bin 17820 -> 16655 bytes test/trap-clip.quartz.argb32.ref.png | Bin 0 -> 6060 bytes test/trap-clip.quartz.ref.png | Bin 6041 -> 0 bytes test/trap-clip.quartz.rgb24.ref.png | Bin 5385 -> 5694 bytes test/unantialiased-shapes.quartz.ref.png | Bin 4394 -> 3868 bytes test/unbounded-operator.quartz.ref.png | Bin 4066 -> 0 bytes test/unbounded-operator.quartz.rgb24.ref.png | Bin 1918 -> 0 bytes test/user-font-proxy.quartz.ref.png | Bin 0 -> 19795 bytes test/user-font-rescale.quartz.ref.png | Bin 0 -> 15982 bytes test/user-font.quartz.ref.png | Bin 0 -> 5960 bytes test/xcomposite-projection.quartz.ref.png | Bin 0 -> 639 bytes 171 files changed, 0 insertions(+), 0 deletions(-) commit 0d5a15f23812575c80db44ca23075293cc0dbfad Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 12 20:00:24 2010 +0100 check: Disable running cairo-perf-micro Since this takes days to run now and should not find any bugs that are not covered by the test-suite it seems like a pointless exercise. Especially as I am trying to make a release! perf/Makefile.am | 15 ++++----------- 1 file changed, 4 insertions(+), 11 deletions(-) commit e9fb39daee9ef3f80e4aedbda68027b36ce94fe3 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 12 19:40:42 2010 +0100 NEWS: take two. NEWS | 81 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++----- 1 file changed, 75 insertions(+), 6 deletions(-) commit 62899e5e23a798bb3e9b60c9e793555e973b7823 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 12 18:51:31 2010 +0100 NEWS: Boring stub for 1.9.10. NEWS | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+) commit fc372b1d9dea1ed4cb4229eaaad62a2d8e112a03 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 12 16:50:30 2010 +0100 test: Missing ref images test/Makefile.am | 8 ++++---- test/coverage-triangles.xlib.xfail.png | Bin 0 -> 15859 bytes 2 files changed, 4 insertions(+), 4 deletions(-) commit 8737bc8b17ff0c437bb57df018bb7c540a394a7c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 12 16:49:46 2010 +0100 gl: start returning the failure status aftern an invalid GL op. src/cairo-gl-composite.c | 1 + src/cairo-gl-device.c | 26 +++++++++++++++++++------- src/cairo-gl-glyphs.c | 10 +++++++--- src/cairo-gl-private.h | 45 ++++++++++++++++++++++++++++++++++----------- src/cairo-gl-surface.c | 40 +++++++++++++++++++++++++--------------- 5 files changed, 86 insertions(+), 36 deletions(-) commit 5b2f90bf532184e46829de4b2e2f9d0b10c1ff81 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 12 16:49:11 2010 +0100 test: Forgotten ref image. test/clipped-group.xlib-fallback.ref.png | Bin 0 -> 344 bytes 1 file changed, 0 insertions(+), 0 deletions(-) commit cdd6f11716e3fbe579d68161caec220c7852fdff Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 12 15:44:29 2010 +0100 test: Record the expected failure mode for partial-coverage-intersecting-quads As the diagonals of the quad intersect below the limits of Cairo's fixed point, we fail to take heed of the intersection. test/Makefile.am | 1 + test/partial-coverage-intersecting-quads.xfail.png | Bin 0 -> 262 bytes 2 files changed, 1 insertion(+) commit 3d72b4aa70d106b5da093e6405c7e3890fa343cf Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 12 15:32:31 2010 +0100 test/coverage: The comment was still wrong. Writing comments is hard. I give in. test/coverage.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit e6d1b3a8ce0b516aed7715b57d9ca5ab4fad6c2f Author: Andrea Canciani <ranma42@gmail.com> Date: Sat Jun 12 16:22:54 2010 +0200 test: update font family name Update the name of the test font family to DejaVu, so that backends unable to rely on fontconfig to match font names can find the correct font anyway. test/README | 10 +++++----- test/cairo-test.h | 2 +- 2 files changed, 6 insertions(+), 6 deletions(-) commit 3cb882d01b53cb3cec9be62d678dd4ef7ea2c1fd Author: Andrea Canciani <ranma42@gmail.com> Date: Sat Jun 12 16:20:37 2010 +0200 test: make test font family a define In order to make it easier to change the test font family, add it as a define in the global header and always reference the macro in the tests. test/cairo-test.h | 2 ++ test/clip-all.c | 2 +- test/clip-empty-save.c | 2 +- test/clip-empty.c | 2 +- test/clip-operator.c | 2 +- test/font-face-get-type.c | 2 +- test/font-matrix-translation.c | 2 +- test/ft-show-glyphs-positioning.c | 2 +- test/ft-show-glyphs-table.c | 2 +- test/ft-text-antialias-none.c | 2 +- test/ft-text-vertical-layout-type3.c | 2 +- test/get-path-extents.c | 2 +- test/glyph-cache-pressure.c | 2 +- test/inverse-text.c | 6 +++--- test/large-font.c | 2 +- test/operator-clear.c | 2 +- test/operator-source.c | 2 +- test/overlapping-glyphs.c | 2 +- test/pdf-features.c | 2 +- test/ps-features.c | 2 +- test/select-font-face.c | 6 +++--- test/select-font-no-show-text.c | 2 +- test/show-text-current-point.c | 2 +- test/smask-text.c | 2 +- test/smask.c | 2 +- test/text-antialias-gray.c | 2 +- test/text-antialias-none.c | 2 +- test/text-antialias-subpixel.c | 2 +- test/text-glyph-range.c | 2 +- test/text-pattern.c | 2 +- test/text-rotate.c | 2 +- test/text-transform.c | 2 +- test/text-zero-len.c | 2 +- test/unbounded-operator.c | 2 +- test/user-font-rescale.c | 6 +++--- 35 files changed, 42 insertions(+), 40 deletions(-) commit b1e4df928fb59ee295830d5c0fa26330714f88f2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 12 15:28:28 2010 +0100 test/coverage: Refine comments Joonas didn't like me putting incorrect labels in my comments. Said they were misleading and worse than useless. Harsh. test/coverage.c | 20 +++++++++++++------- 1 file changed, 13 insertions(+), 7 deletions(-) commit 968374b6335199cbf02c8c4733537388da2a45f5 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 12 15:14:17 2010 +0100 test/coverage: Add a comment describing the column_triangle geometry() test/coverage.c | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+) commit 299263f8c912d3468d51ed22bcb6ab96275709f0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 12 13:48:12 2010 +0100 test: Disable tests that are too slow to run by default. Introduce a slow mode [-s] to the runner, and add "slow" to the requirements for the very long running tests like the coverage stress tests. test/cairo-test-runner.c | 17 +++++++++++++++-- test/coverage.c | 12 ++++++------ test/half-coverage.c | 4 ++-- test/mask-glyphs.c | 2 +- test/partial-coverage.c | 16 ++++++++-------- 5 files changed, 32 insertions(+), 19 deletions(-) commit 53139261c7ba1a4cc0b869cb48c136b0362a3b0c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 12 13:40:17 2010 +0100 test: Use exact matching on test names by default. Specifying individual tests to run is more common than using a group chosen from a keyword, so change the default matching mode and introduce '-k' to select by keyword. test/cairo-test-runner.c | 45 ++++++++++++++++++++++----------------------- 1 file changed, 22 insertions(+), 23 deletions(-) commit 97288b0859bae6cea1e3bff84b632e00f1d10917 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 12 13:34:27 2010 +0100 test: More minute geometry exercised by partial-coverage. Moving beyond the capabilities of cairo... test/Makefile.am | 2 + test/partial-coverage-intersecting-quads.ref.png | Bin 0 -> 189 bytes ...partial-coverage-intersecting-triangles.ref.png | Bin 0 -> 202 bytes test/partial-coverage.c | 131 +++++++++++++++++++++ 4 files changed, 133 insertions(+) commit 4e3ef57bc892b0b046c486390adc7164a1de64de Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 12 13:18:42 2010 +0100 test: More coverage coverage. A couple of different shapes that aim to test the tessellation side of the rasterisers more... And worryingly there does seem to be an unexpected systematic error. test/Makefile.am | 8 + test/coverage-column-triangles.ref.png | Bin 0 -> 208 bytes test/coverage-column-triangles.xlib.xfail.png | Bin 0 -> 12745 bytes test/coverage-intersecting-quads.ref.png | Bin 0 -> 253 bytes test/coverage-intersecting-quads.xlib.xfail.png | Bin 0 -> 15296 bytes test/coverage-intersecting-triangles.ref.png | Bin 0 -> 209 bytes .../coverage-intersecting-triangles.xlib.xfail.png | Bin 0 -> 16728 bytes test/coverage-row-triangles.ref.png | Bin 0 -> 208 bytes test/coverage-row-triangles.xlib.xfail.png | Bin 0 -> 12671 bytes test/coverage.c | 249 ++++++++++++++++++++- 10 files changed, 251 insertions(+), 6 deletions(-) commit 9b7cc7641b691a3b9e3d5edd51053c9a095d9c5a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 12 10:41:09 2010 +0100 cairo: Create error objects for cairo_t Perform an early check for error status and prevent creation of a full object. This means that we do not pass down error objects to the initialisation routines and so can survive without paranoia inside the library. It also has brings consistency that like the other constructors, no object is created in error and we can skip the cairo_destroy() if we choose (and we don't waste one of the precious zero-alloc context slots. Fixes crash in test/a8-mask introduced by 1a544361e845. src/cairo-debug.c | 2 ++ src/cairo-mutex-list-private.h | 1 + src/cairo.c | 76 ++++++++++++++++++++++++++++++++++++------ src/cairoint.h | 3 ++ 4 files changed, 71 insertions(+), 11 deletions(-) commit 9b6617a3b3b31d52e6646400b16ec56c754bfebd Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 11 21:58:34 2010 +0100 image: Apply component alpha to composite masks. If we need to pattern requires component alpha, then we must take a copy of the image and enable component alpha for pixman. Fixes test/text-antialias-subpixel on xlib-fallback -- i.e. we will finally render subpixel antialiased text on ancient XServers. src/cairo-image-surface.c | 40 ++++++++++++++++++++++++---------------- 1 file changed, 24 insertions(+), 16 deletions(-) commit 7440fc07f77b112b00957f5459129dba23a5683f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 11 21:29:02 2010 +0100 test: xlib ref image refresh test/Makefile.am | 26 +++++++++++++++------ ...clip-fill-unbounded.xlib-fallback.rgb24.ref.png | Bin 1429 -> 1438 bytes test/clip-fill-unbounded.xlib.argb32.ref.png | Bin 1758 -> 1759 bytes test/clip-fill-unbounded.xlib.rgb24.ref.png | Bin 1426 -> 1430 bytes test/clip-group-shapes-circles.xlib.ref.png | Bin 0 -> 1534 bytes ...-group-shapes-unaligned-rectangles.xlib.ref.png | Bin 0 -> 431 bytes ...ip-stroke-unbounded.xlib-fallback.rgb24.ref.png | Bin 1429 -> 1437 bytes test/clip-stroke-unbounded.xlib.argb32.ref.png | Bin 1792 -> 1785 bytes test/clip-stroke-unbounded.xlib.rgb24.ref.png | Bin 1426 -> 1442 bytes test/clip-stroke.xlib-fallback.ref.png | Bin 1490 -> 1491 bytes test/clip-stroke.xlib.ref.png | Bin 1490 -> 1491 bytes test/clipped-group.xlib.ref.png | Bin 0 -> 289 bytes test/coverage-rectangles.xlib.xfail.png | Bin 0 -> 15981 bytes test/finer-grained-fallbacks.xlib-fallback.ref.png | Bin 0 -> 919 bytes test/finer-grained-fallbacks.xlib.ref.png | Bin 1111 -> 1110 bytes test/finer-grained-fallbacks.xlib.rgb24.ref.png | Bin 1114 -> 856 bytes test/ft-text-vertical-layout-type1.xlib.ref.png | Bin 3639 -> 3625 bytes test/ft-text-vertical-layout-type3.xlib.ref.png | Bin 3605 -> 3597 bytes test/leaky-dashed-rectangle.xlib.ref.png | Bin 347 -> 344 bytes test/linear-gradient-subset.xlib.ref.png | Bin 0 -> 782 bytes test/pthread-show-text.xlib-fallback.ref.png | Bin 0 -> 42073 bytes ...m-intersections-curves-nz.xlib-fallback.ref.png | Bin 264288 -> 264277 bytes test/random-intersections-curves-nz.xlib.ref.png | Bin 264288 -> 264277 bytes test/recording-surface-pattern.xlib.argb32.ref.png | Bin 0 -> 3055 bytes test/recording-surface-pattern.xlib.rgb24.ref.png | Bin 0 -> 3124 bytes test/scale-offset-image.xlib-fallback.xfail.png | Bin 8706 -> 8686 bytes test/scale-offset-image.xlib.ref.png | Bin 0 -> 9136 bytes test/scale-offset-image.xlib.xfail.png | Bin 9132 -> 0 bytes test/scale-offset-similar.xlib-fallback.xfail.png | Bin 8706 -> 8686 bytes test/scale-offset-similar.xlib.xfail.png | Bin 8696 -> 8698 bytes test/self-intersecting.xlib.ref.png | Bin 252 -> 208 bytes test/smask-text.xlib-fallback.ref.png | Bin 0 -> 1874 bytes test/smask.xlib-fallback.ref.png | Bin 0 -> 3423 bytes test/subsurface.xlib.ref.png | Bin 0 -> 1782 bytes test/text-rotate.xlib-fallback.ref.png | Bin 0 -> 16592 bytes 35 files changed, 19 insertions(+), 7 deletions(-) commit a049889c64131b170b66ed8906309f6adee80fa5 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 11 21:26:26 2010 +0100 pattern: Remove incorrect optimisations from _cairo_pattern_aquire_surface() Safe reduction of patterns is performed in gstate, so not only are the extra checks in _cairo_pattern_acquire_surface redundant there are also unsafe. Simply remove them. Fixes test/radial-gradient-extend [xlib-fallback] src/cairo-pattern.c | 110 ++++++++-------------------------------------------- 1 file changed, 17 insertions(+), 93 deletions(-) commit 00bc1d1578dc2640db64315270b00073697fb9fa Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 11 20:31:29 2010 +0100 pattern: Remove extraordinary _cairo_pattern_fini_snapshot(). Miraculously the circular references from self-copy have disappeared and the forced finish within _cairo_pattern_fini_snapshot() now quite explosive. By replacing them with an ordinary _cairo_pattern_fini() the crash from test/smask-image-mask disappear along and valgrind remains happy. Fixes test/smask-image-mask and similar. src/cairo-pattern.c | 17 ----------------- src/cairo-recording-surface.c | 24 ++++++++++++------------ src/cairoint.h | 3 --- 3 files changed, 12 insertions(+), 32 deletions(-) commit 1098e203dc21529f138bfabfa7558f61b8427ea0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 11 18:07:28 2010 +0100 test: pdf ref image refresh test/Makefile.am | 11 +++++++++++ test/clear-source.pdf.xfail.png | Bin 0 -> 974 bytes test/clip-fill-unbounded.pdf.argb32.ref.png | Bin 0 -> 1244 bytes test/clip-stroke-unbounded.pdf.argb32.xfail.png | Bin 0 -> 1351 bytes test/clip-unbounded.pdf.argb32.xfail.png | Bin 0 -> 95 bytes test/clip-unbounded.pdf.rgb24.xfail.png | Bin 0 -> 95 bytes test/filter-nearest-offset.pdf.xfail.png | Bin 2152 -> 4522 bytes test/filter-nearest-transformed.pdf.xfail.png | Bin 532 -> 532 bytes test/pthread-show-text.pdf.ref.png | Bin 0 -> 47015 bytes test/radial-gradient-source.pdf.argb32.ref.png | Bin 0 -> 235290 bytes test/rotate-image-surface-paint.pdf.xfail.png | Bin 358 -> 442 bytes test/scale-source-surface-paint.pdf.argb32.xfail.png | Bin 0 -> 157 bytes test/scale-source-surface-paint.pdf.rgb24.xfail.png | Bin 0 -> 152 bytes test/surface-pattern-operator.pdf.argb32.xfail.png | Bin 0 -> 5485 bytes test/surface-pattern-operator.pdf.rgb24.xfail.png | Bin 0 -> 2095 bytes 15 files changed, 11 insertions(+) commit 1c91d019c54afde0edb022312491b086588ce6c4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 11 17:47:08 2010 +0100 test: ps ref image refresh. Update to gs-8.71 and some additional reference images for the new tests. test/Makefile.am | 116 ++++++++++++--------- test/README | 2 +- test/big-line.ps.ref.png | Bin 0 -> 946 bytes test/bug-bo-rectangular.ps.xfail.png | Bin 0 -> 945 bytes test/caps-joins-curve.ps.ref.png | Bin 3728 -> 3715 bytes test/caps-joins.ps.ref.png | Bin 2282 -> 2268 bytes test/caps-sub-paths.ps.ref.png | Bin 0 -> 163 bytes test/caps.ps.ref.png | Bin 1466 -> 1478 bytes test/clear-source.ps.xfail.png | Bin 0 -> 552 bytes test/clip-disjoint.ps.ref.png | Bin 0 -> 4680 bytes test/clip-fill-rule.ps.argb32.ref.png | Bin 0 -> 251 bytes test/clip-fill-rule.ps.rgb24.ref.png | Bin 0 -> 234 bytes test/clip-fill-rule.ps2.argb32.ref.png | Bin 310 -> 0 bytes test/clip-fill-rule.ps2.rgb24.ref.png | Bin 302 -> 0 bytes test/clip-fill-rule.ps3.argb32.ref.png | Bin 310 -> 0 bytes test/clip-fill-rule.ps3.rgb24.ref.png | Bin 302 -> 0 bytes test/clip-fill-unbounded.ps.argb32.xfail.png | Bin 0 -> 1244 bytes test/clip-image.ps.ref.png | Bin 0 -> 2498 bytes test/clip-nesting.ps.argb32.ref.png | Bin 0 -> 441 bytes test/clip-nesting.ps.rgb24.ref.png | Bin 0 -> 492 bytes test/clip-nesting.ps2.argb32.ref.png | Bin 651 -> 0 bytes test/clip-nesting.ps2.rgb24.ref.png | Bin 636 -> 0 bytes test/clip-nesting.ps3.argb32.ref.png | Bin 651 -> 0 bytes test/clip-nesting.ps3.rgb24.ref.png | Bin 636 -> 0 bytes test/clip-push-group.ps.ref.png | Bin 0 -> 120 bytes test/clip-push-group.ps2.argb32.ref.png | Bin 179 -> 0 bytes test/clip-push-group.ps2.rgb24.ref.png | Bin 179 -> 0 bytes test/clip-push-group.ps3.argb32.ref.png | Bin 179 -> 0 bytes test/clip-push-group.ps3.rgb24.ref.png | Bin 179 -> 0 bytes test/clip-stroke-unbounded.ps.argb32.ref.png | Bin 0 -> 1351 bytes test/clip-stroke.ps.xfail.png | Bin 0 -> 662 bytes test/clip-twice.ps.argb32.ref.png | Bin 0 -> 492 bytes test/clip-twice.ps.rgb24.ref.png | Bin 0 -> 409 bytes test/clip-twice.ps2.argb32.ref.png | Bin 575 -> 0 bytes test/clip-twice.ps2.rgb24.ref.png | Bin 541 -> 0 bytes test/clip-twice.ps3.argb32.ref.png | Bin 575 -> 0 bytes test/clip-twice.ps3.rgb24.ref.png | Bin 541 -> 0 bytes test/clipped-group.ps2.ref.png | Bin 343 -> 214 bytes test/clipped-group.ps3.ref.png | Bin 343 -> 214 bytes test/copy-path.ps.ref.png | Bin 0 -> 379 bytes test/copy-path.ps2.ref.png | Bin 474 -> 0 bytes test/copy-path.ps3.ref.png | Bin 474 -> 0 bytes test/dash-caps-joins.ps.ref.png | Bin 0 -> 3669 bytes test/dash-caps-joins.ps2.argb32.ref.png | Bin 3870 -> 0 bytes test/dash-caps-joins.ps2.rgb24.ref.png | Bin 3870 -> 0 bytes test/dash-caps-joins.ps3.argb32.ref.png | Bin 3870 -> 0 bytes test/dash-caps-joins.ps3.rgb24.ref.png | Bin 3870 -> 0 bytes test/dash-curve.ps2.ref.png | Bin 27293 -> 24355 bytes test/dash-curve.ps3.ref.png | Bin 27293 -> 24355 bytes test/dash-infinite-loop.ps.ref.png | Bin 0 -> 617 bytes test/dash-scale.ps.ref.png | Bin 0 -> 5965 bytes test/dash-scale.ps2.argb32.ref.png | Bin 6318 -> 0 bytes test/dash-scale.ps2.rgb24.ref.png | Bin 6316 -> 0 bytes test/dash-scale.ps3.argb32.ref.png | Bin 6318 -> 0 bytes test/dash-scale.ps3.rgb24.ref.png | Bin 6316 -> 0 bytes test/dash-state.ps2.ref.png | Bin 10219 -> 8740 bytes test/dash-state.ps3.ref.png | Bin 10219 -> 8740 bytes test/degenerate-arc.ps2.ref.png | Bin 585 -> 509 bytes test/degenerate-arc.ps3.ref.png | Bin 585 -> 509 bytes test/degenerate-curve-to.ps.xfail.png | Bin 202 -> 221 bytes test/degenerate-dash.ps.xfail.png | Bin 1835 -> 1837 bytes test/degenerate-path.ps.argb32.xfail.png | Bin 236 -> 234 bytes test/degenerate-path.ps.rgb24.xfail.png | Bin 186 -> 184 bytes test/degenerate-pen.ps.ref.png | Bin 0 -> 759 bytes test/degenerate-pen.ps2.ref.png | Bin 865 -> 0 bytes test/degenerate-pen.ps3.ref.png | Bin 865 -> 0 bytes test/degenerate-rel-curve-to.ps.ref.png | Bin 0 -> 225 bytes test/degenerate-rel-curve-to.ps.xfail.png | Bin 202 -> 0 bytes test/fill-and-stroke.ps.argb32.ref.png | Bin 0 -> 257 bytes test/fill-and-stroke.ps.rgb24.ref.png | Bin 0 -> 225 bytes test/fill-and-stroke.ps2.argb32.ref.png | Bin 310 -> 0 bytes test/fill-and-stroke.ps2.rgb24.ref.png | Bin 319 -> 0 bytes test/fill-and-stroke.ps3.argb32.ref.png | Bin 310 -> 0 bytes test/fill-and-stroke.ps3.rgb24.ref.png | Bin 319 -> 0 bytes .../fill-degenerate-sort-order.ps.argb32.xfail.png | Bin 1901 -> 1903 bytes test/fill-degenerate-sort-order.ps.rgb24.xfail.png | Bin 1645 -> 1638 bytes test/fill-image.ps.ref.png | Bin 1799 -> 1645 bytes test/fill-rule.ps2.argb32.ref.png | Bin 2253 -> 1878 bytes test/fill-rule.ps2.rgb24.ref.png | Bin 2031 -> 1524 bytes test/fill-rule.ps3.argb32.ref.png | Bin 2253 -> 1878 bytes test/fill-rule.ps3.rgb24.ref.png | Bin 2031 -> 1524 bytes test/finer-grained-fallbacks.ps2.argb32.ref.png | Bin 1173 -> 1168 bytes test/finer-grained-fallbacks.ps2.rgb24.ref.png | Bin 1154 -> 936 bytes test/finer-grained-fallbacks.ps3.argb32.ref.png | Bin 1173 -> 1168 bytes test/finer-grained-fallbacks.ps3.rgb24.ref.png | Bin 1154 -> 936 bytes test/ft-text-vertical-layout-type1.ps.ref.png | Bin 1634 -> 2088 bytes test/ft-text-vertical-layout-type3.ps.ref.png | Bin 1636 -> 2134 bytes test/group-unaligned.ps.ref.png | Bin 332 -> 321 bytes test/group-unaligned.ps.rgb24.xfail.png | Bin 0 -> 371 bytes test/implicit-close.ps.ref.png | Bin 0 -> 252 bytes test/joins.ps.ref.png | Bin 5496 -> 4230 bytes test/leaky-dashed-rectangle.ps.ref.png | Bin 444 -> 358 bytes test/leaky-dashed-stroke.ps.ref.png | Bin 0 -> 5293 bytes test/leaky-dashed-stroke.ps2.ref.png | Bin 5511 -> 0 bytes test/leaky-dashed-stroke.ps3.ref.png | Bin 5511 -> 0 bytes test/leaky-polygon.ps.ref.png | Bin 0 -> 289 bytes test/leaky-polygon.ps2.ref.png | Bin 354 -> 0 bytes test/leaky-polygon.ps3.ref.png | Bin 354 -> 0 bytes test/line-width-scale.ps2.ref.png | Bin 3911 -> 3431 bytes test/line-width-scale.ps3.ref.png | Bin 3911 -> 3431 bytes test/linear-gradient-subset.ps3.ref.png | Bin 0 -> 619 bytes test/long-dashed-lines.ps2.ref.png | Bin 1516 -> 1329 bytes test/long-dashed-lines.ps3.ref.png | Bin 1516 -> 1329 bytes test/new-sub-path.ps2.argb32.ref.png | Bin 556 -> 398 bytes test/new-sub-path.ps3.argb32.ref.png | Bin 556 -> 398 bytes test/operator-alpha-alpha.ps.xfail.png | Bin 1338 -> 3429 bytes test/over-around-source.ps2.argb32.ref.png | Bin 522 -> 508 bytes test/over-around-source.ps3.argb32.ref.png | Bin 522 -> 508 bytes test/over-below-source.ps2.argb32.ref.png | Bin 503 -> 368 bytes test/over-below-source.ps2.rgb24.ref.png | Bin 0 -> 362 bytes test/over-below-source.ps3.argb32.ref.png | Bin 503 -> 368 bytes test/over-below-source.ps3.rgb24.ref.png | Bin 0 -> 362 bytes test/path-append.ps.ref.png | Bin 4516 -> 4525 bytes test/pixman-rotate.ps.argb32.ref.png | Bin 0 -> 355 bytes test/push-group-color.ps3.ref.png | Bin 0 -> 2561 bytes test/random-intersections-curves-eo.ps.ref.png | Bin 151676 -> 151668 bytes test/random-intersections-curves-nz.ps.ref.png | Bin 134732 -> 134746 bytes test/random-intersections-eo.ps.ref.png | Bin 78806 -> 78037 bytes test/random-intersections-nonzero.ps.ref.png | Bin 77299 -> 76450 bytes test/recording-surface-pattern.ps.argb32.ref.png | Bin 0 -> 3064 bytes test/recording-surface-pattern.ps.rgb24.ref.png | Bin 0 -> 3147 bytes test/recording-surface-pattern.ps2.argb32.ref.png | Bin 3257 -> 0 bytes test/recording-surface-pattern.ps2.rgb24.ref.png | Bin 3873 -> 0 bytes test/recording-surface-pattern.ps3.argb32.ref.png | Bin 3257 -> 0 bytes test/recording-surface-pattern.ps3.rgb24.ref.png | Bin 3873 -> 0 bytes test/reflected-stroke.ps.ref.png | Bin 0 -> 3368 bytes test/reflected-stroke.ps2.ref.png | Bin 3898 -> 0 bytes test/reflected-stroke.ps3.ref.png | Bin 3898 -> 0 bytes test/rounded-rectangle-fill.ps.ref.png | Bin 0 -> 551 bytes test/rounded-rectangle-stroke.ps.ref.png | Bin 0 -> 568 bytes test/scale-offset-image.ps.ref.png | Bin 7289 -> 7697 bytes test/spline-decomposition.ps.ref.png | Bin 9028 -> 8939 bytes test/stroke-image.ps.ref.png | Bin 1641 -> 1682 bytes test/subsurface.ps.ref.png | Bin 0 -> 1651 bytes test/surface-pattern-big-scale-down.ps.ref.png | Bin 0 -> 235 bytes test/surface-pattern-big-scale-down.ps.xfail.png | Bin 238 -> 0 bytes test/text-glyph-range.ps.ref.png | Bin 0 -> 1253 bytes test/text-glyph-range.ps2.ref.png | Bin 1389 -> 0 bytes test/text-glyph-range.ps3.ref.png | Bin 1389 -> 0 bytes test/text-rotate.ps.ref.png | Bin 0 -> 8765 bytes test/text-rotate.ps2.ref.png | Bin 6851 -> 0 bytes test/text-rotate.ps3.ref.png | Bin 6851 -> 0 bytes test/text-transform.ps.ref.png | Bin 0 -> 3430 bytes test/text-transform.ps2.ref.png | Bin 4154 -> 0 bytes test/text-transform.ps3.ref.png | Bin 4154 -> 0 bytes test/trap-clip.ps2.argb32.ref.png | Bin 4849 -> 4839 bytes test/trap-clip.ps2.rgb24.ref.png | Bin 5617 -> 4729 bytes test/trap-clip.ps3.argb32.ref.png | Bin 5591 -> 4751 bytes test/trap-clip.ps3.rgb24.ref.png | Bin 5512 -> 4553 bytes test/twin.ps.ref.png | Bin 2142 -> 2136 bytes test/user-font-proxy.ps.ref.png | Bin 7766 -> 7837 bytes test/user-font-rescale.ps.ref.png | Bin 0 -> 6781 bytes test/user-font-rescale.ps2.ref.png | Bin 8041 -> 0 bytes test/user-font-rescale.ps3.ref.png | Bin 8041 -> 0 bytes test/xlib-expose-event.ps.ref.png | Bin 0 -> 39035 bytes 155 files changed, 65 insertions(+), 53 deletions(-) commit 1a544361e845e4881990624a597f9dc2b82d1c73 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 11 16:04:41 2010 +0100 gstate: Update cached matrix state after device transform changes on the target Commit 8d67186cb291cb877e52b987e2ac18c2a1175a57 caches whether the device transform is identity on context creation. However, the api is quite lax and allows the user to modify the device transform *after* he has started to use the surface in a context, as apparently WebKit does. Since this is not the only instance where we may need to invalidate caches if the user modifies state, introduce a simple mechanism for hooking into notifications of property changes. Fixes test/clip-device-offset. src/Makefile.sources | 1 + src/cairo-gstate-private.h | 3 +++ src/cairo-gstate.c | 28 +++++++++++++++++++++---- src/cairo-observer.c | 50 +++++++++++++++++++++++++++++++++++++++++++++ src/cairo-surface-private.h | 1 + src/cairo-surface.c | 6 ++++++ src/cairo-types-private.h | 6 ++++++ src/cairoint.h | 5 +++++ 8 files changed, 96 insertions(+), 4 deletions(-) commit f74b11415a1f7682dd50c222baa8815ef93681dc Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 11 14:06:15 2010 +0100 test: Use a group to render PDF correctly using poppler. In order to support blend modes, we need to use a clear background whilst rendering the PDF, which inspired the use of DEST_OVER to avoid having to render to an intermediate surface. However, the adjunct of using subpixel text means that we cannot simply using the alpha channel as the sole mask and so need to render to an intermediate and paint it over an opaque background. test/any2ppm.c | 9 ++++++--- test/pdf2png.c | 9 ++++++--- 2 files changed, 12 insertions(+), 6 deletions(-) commit 7f227ec8d3f981dbbf2ca1914bf7fd29d9cdfd05 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 11 12:54:15 2010 +0100 Revert "test/pdf2png: Don't use DEST_OVER as poppler starts using subpixel text." This reverts commit 506b2ebe714d61a64972b607a42a55e48d1c722a. We have conflicting requirements here. In order to support extended blend modes correctly we need to use DEST_OVER, at the expense of supporting subpixel text. More thought required. test/any2ppm.c | 9 +++++---- test/pdf2png.c | 10 +++++----- 2 files changed, 10 insertions(+), 9 deletions(-) commit c761e0c5bb55e8869521ad5b4a2447ea9abcb4d1 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 11 12:40:36 2010 +0100 test: Restrict the coverage tests to raster targets. Typo inside the requirements failed to prevent the tests running on the vector backends. test/coverage.c | 4 ++-- test/half-coverage.c | 4 ++-- test/partial-coverage.c | 18 +++++++++--------- 3 files changed, 13 insertions(+), 13 deletions(-) commit b49c944fe4abb9f3c70fac02a71061bf8d07259a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 11 12:38:17 2010 +0100 test: Add a convenience script to disable the screensaver when testing If the screensaver activates, it can cause failures when attempting to render into an xlib window, so inhibit if we know how. test/Makefile.am | 2 +- test/run-cairo-test-suite.sh | 18 ++++++++++++++++++ 2 files changed, 19 insertions(+), 1 deletion(-) commit 4e4724d48c63d0cd4ec507ce72dcc55fb6e5f584 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 11 12:19:56 2010 +0100 gl: make check insists "cairoint.h" is first. src/cairo-gl-gradient.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) commit 4edbcf1b1d046c9c626a0c0ad10d3c71f17984f2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 11 12:17:19 2010 +0100 color: Mark _cairo_color_get_content() as private. src/cairoint.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 506b2ebe714d61a64972b607a42a55e48d1c722a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 11 12:04:39 2010 +0100 test/pdf2png: Don't use DEST_OVER as poppler starts using subpixel text. Using DEST_OVER in this form to paint the background white fails in the presence of subpixel geometry (particular text), so remove the hack and simply paint the background white prior to passing the surface to poppler. KISS. test/any2ppm.c | 9 ++++----- test/pdf2png.c | 10 +++++----- 2 files changed, 9 insertions(+), 10 deletions(-) commit edb73b6dcf1adce40d6c10c492e3f78556e22b85 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 11 11:16:42 2010 +0100 xlib: Adjust trapezoid precision based on antialias. Render supports two modes of precision when rendering trapezoids. Precise specifies points sampling on a 15x17 grid, ala pixman. Imprecise allows the driver more freedom in the methods used, which may be more amenable to acceleration. Choose to use the imprecise mode by default, but still allow users to force the more rigidly specified precision by changing the antialias mode. src/cairo-xlib-surface-private.h | 1 + src/cairo-xlib-surface.c | 31 +++++++++++++++++++++++++++++++ 2 files changed, 32 insertions(+) commit 290749bdb5c634c030db81722214661c34344403 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 11 10:59:17 2010 +0100 polygon: Reorder conditionals based on likelihood. The vast majority of edges will be unclipped, so process those first. src/cairo-polygon.c | 30 +++++++++++++++--------------- 1 file changed, 15 insertions(+), 15 deletions(-) commit 55037bfb2454a671332d961e061c712ab5471580 Author: Karl Tomlinson <karlt+@karlt.net> Date: Fri Jun 11 10:35:43 2010 +0100 xlib: Find matching Visual for XRenderFormat Not only is this useful for users to know which Visual matches any particular Cairo surface, it should also close a few obscure bugs of not converting images correctly on upload. Fixes: Bug 28492 - cairo_xlib_surface_create_with_xrender_format does not create visual for resulting surface https://bugs.freedesktop.org/show_bug.cgi?id=28492 Mozilla Bug 567065 - Try to create offscreen Xlib surface from existing visual if possible https://bugzilla.mozilla.org/show_bug.cgi?id=567065 Mozilla Bug 445250 - cairo_draw_with_xlib should provide a non-NULL visual to callback https://bugzilla.mozilla.org/show_bug.cgi?id=445250 Reported-by: Oleg Romashin <romaxa@gmail.com> src/cairo-xlib-surface.c | 74 ++++++++++++++++++++++++++++++++++++++++++------ 1 file changed, 65 insertions(+), 9 deletions(-) commit cf426bce8ebd3aff5d50c6a54bbd731132b3d312 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 11 10:05:09 2010 +0100 test: Add a simple coverage test. This test case should be much clearer than half-coverage over what it purports to actually test. In each pixel, it draws a single geometric primitive that coverages a known percentage of the pixel and then we measure how close the rasterisers are to that ideal. test/Makefile.am | 2 + test/Makefile.sources | 1 + test/coverage-rectangles.ref.png | Bin 0 -> 259 bytes test/coverage-triangles.ref.png | Bin 0 -> 253 bytes test/coverage.c | 135 +++++++++++++++++++++++++++++++++++++++ 5 files changed, 138 insertions(+) commit bd3d9ef3d1dbc5364e79e6afb47d9e124cb61ca4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 11 09:12:16 2010 +0100 test: Distinguish tests that throw an error from a normal fail. Hitting an error in a test case is almost as bad as crashing, and the severity may be lost amidst "normal" failures. So introduce a new class of ERROR so that we can immediately spot these during a test run, and appropriately log them afterwards. test/cairo-test-runner.c | 89 ++++++++++++++++++++++++++++++++++++++++++++++-- test/cairo-test.c | 20 ++++++++++- test/cairo-test.h | 1 + 3 files changed, 107 insertions(+), 3 deletions(-) commit eeafeebd2ec8ad8a9e7053aaaa0f845b58563b3b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 11 09:06:20 2010 +0100 path: Exponentially grow buffer based on populated points and ops. Instead of simply doubling the buffer size every time we overflow a point or an op, enlarge the buffer to fit twice the number of used points and ops. We expect paths to be fairly consistent in the mix of operations, and this allows the buffer size to tune itself to actual usage and reduce wastage. src/cairo-path-fixed-private.h | 3 ++- src/cairo-path-fixed.c | 39 +++++++++++++++++++-------------------- 2 files changed, 21 insertions(+), 21 deletions(-) commit b0056a86b4f6c92fa7eb89baf8483a46979c8cf1 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jun 10 18:27:22 2010 +0100 test: Exercise partial coverage some more. Add some overlapping triangles (coarse enough for them to be resolved using Cairo's 24.8 fixed point without loss of precision) to exercise handling of overlapping subpixel geometry. test/Makefile.am | 4 + ...tial-coverage-overlap-half-triangles-eo.ref.png | Bin 0 -> 189 bytes ...partial-coverage-overlap-half-triangles.ref.png | Bin 0 -> 189 bytes ...overage-overlap-three-quarter-triangles.ref.png | Bin 0 -> 186 bytes ...artial-coverage-three-quarter-reference.ref.png | Bin 0 -> 186 bytes test/partial-coverage.c | 284 +++++++++++++++++++-- 6 files changed, 269 insertions(+), 19 deletions(-) commit 836f6166597eaa687b10e5f26fe71e099c434aa1 Author: Andrea Canciani <ranma42@gmail.com> Date: Wed Jun 9 19:09:22 2010 +0200 gl: support single stop gradients src/cairo-gl-gradient-private.h | 2 +- src/cairo-gl-gradient.c | 3 +-- 2 files changed, 2 insertions(+), 3 deletions(-) commit d17e2c5e23eac302b7463a08205088147da1172a Author: Andrea Canciani <ranma42@gmail.com> Date: Thu Jun 10 10:52:16 2010 +0200 ps: support single stop gradients src/cairo-ps-surface.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) commit eb7fc351156bd66921ac3b3ffbd1fba6ba743ffb Author: Andrea Canciani <ranma42@gmail.com> Date: Thu Jun 10 10:51:50 2010 +0200 pdf: support single stop gradients src/cairo-pdf-surface.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) commit e2660a0eac5f6081bf052db16b8bd85c814d19d8 Author: Andrea Canciani <ranma42@gmail.com> Date: Wed Jun 9 16:59:29 2010 +0200 pattern: improve single stop gradients handling None-extended single stop gradients are now explicitly made clear. src/cairo-pattern.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) commit 7a024bc206aae790c7357be83474126037335ab1 Author: Andrea Canciani <ranma42@gmail.com> Date: Wed Jun 9 17:59:47 2010 +0200 test: exercise ps crash with single stop gradients PS crashes only when using the source operator, as it will otherwise fallback to rasterization of the gradients. test/linear-gradient-one-stop.argb32.ref.png | Bin 0 -> 225 bytes test/linear-gradient-one-stop.c | 2 +- test/linear-gradient-one-stop.ref.png | Bin 344 -> 0 bytes test/linear-gradient-one-stop.rgb24.ref.png | Bin 0 -> 174 bytes test/radial-gradient-one-stop.argb32.ref.png | Bin 0 -> 1392 bytes test/radial-gradient-one-stop.c | 2 +- test/radial-gradient-one-stop.ref.png | Bin 1422 -> 0 bytes test/radial-gradient-one-stop.rgb24.ref.png | Bin 0 -> 1153 bytes 8 files changed, 2 insertions(+), 2 deletions(-) commit a0f8cfe64660c0043db15f9b9cbf1298bf1838d1 Author: Andrea Canciani <ranma42@gmail.com> Date: Tue Jun 8 17:04:00 2010 +0200 pattern: improve degenerate gradients handling Degenerate radial gradients are now considered clear. src/cairo-pattern.c | 35 +++++++++++++++++++++++++---------- 1 file changed, 25 insertions(+), 10 deletions(-) commit bccd89b4177b5f5bd05def9bee5f0927a9f25e4a Author: Andrea Canciani <ranma42@gmail.com> Date: Tue Jun 8 15:45:09 2010 +0200 gstate: correct optimizations Gradient were previously hand-optimized (without properly checking for extend modes). By properly using _cairo_pattern functions we avoid code duplication and bugs. Fixes linear-gradient-extend, radial-gradient-extend. src/cairo-gstate.c | 85 +++++++++++------------------------------------------- 1 file changed, 17 insertions(+), 68 deletions(-) commit 06c6207ad4205f211be70e324c6d32ea7d28dca8 Author: Andrea Canciani <ranma42@gmail.com> Date: Tue Jun 1 19:40:14 2010 +0200 pattern: add gradient_is_solid function It contains in a single place the logic needed to check if a gradient pattern is solid (within a specified region). src/cairo-pattern.c | 53 +++++++++++++++++++++++++++++++++++++++++++++++++++++ src/cairoint.h | 5 +++++ 2 files changed, 58 insertions(+) commit 561625ee3bd2732457eaaf28937edf557ee7661d Author: Andrea Canciani <ranma42@gmail.com> Date: Tue Jun 1 19:39:02 2010 +0200 pattern: improve clear/opaque check functions _cairo_pattern_is_opaque was missing some checks about the extend type. Conversely _cairo_pattern_is_clear was being too strict about gradients. src/cairo-pattern.c | 112 ++++++++++++++++++++++++++++++++++++++++++++++++---- src/cairoint.h | 4 ++ 2 files changed, 108 insertions(+), 8 deletions(-) commit baaf312e047a9bea6f54e63cd6534c2ed7448523 Author: Andrea Canciani <ranma42@gmail.com> Date: Mon Apr 19 09:53:00 2010 +0200 pattern: remove content field from solid patterns The content field in solid patterns had ill-defined semantic (or no semantic at all), thus it can be removed. src/cairo-color.c | 16 ++++++++++++++++ src/cairo-gl-surface.c | 2 +- src/cairo-gstate.c | 14 +++++--------- src/cairo-pattern.c | 37 ++++++++++--------------------------- src/cairo-script-surface.c | 9 +++------ src/cairo-surface.c | 4 ++-- src/cairo-types-private.h | 1 - src/cairo-xlib-surface.c | 6 +++--- src/cairo.c | 4 ++-- src/cairoint.h | 9 +++++---- 10 files changed, 47 insertions(+), 55 deletions(-) commit 7461947eb1a5f002a5931303843a5a2deca0e383 Author: Andrea Canciani <ranma42@gmail.com> Date: Mon Apr 19 09:51:26 2010 +0200 surface: remove content argument from is_similar The content argument was basically unuses. Xlib change extracted from ickle's wip/compositor branch. src/cairo-directfb-surface.c | 2 +- src/cairo-pattern.c | 5 ++--- src/cairo-surface.c | 5 ++--- src/cairo-win32-surface.c | 3 +-- src/cairo-xlib-surface.c | 30 ++---------------------------- src/cairoint.h | 6 ++---- 6 files changed, 10 insertions(+), 41 deletions(-) commit ef5f9b5c61750207947587173d21b46e2d299f33 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jun 10 14:13:53 2010 +0100 test: Update partial coverage. Gah, no wonder the output looked wrong for the triangles, they only covered half the pixel. So separate triangles into two cases. test/Makefile.am | 2 + test/partial-coverage-half-reference.ref.png | Bin 0 -> 189 bytes test/partial-coverage-half-triangles.ref.png | Bin 0 -> 189 bytes test/partial-coverage.c | 98 +++++++++++++++++++++++++-- 4 files changed, 96 insertions(+), 4 deletions(-) commit 8d67186cb291cb877e52b987e2ac18c2a1175a57 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jun 10 11:36:58 2010 +0100 gstate: Track whether the combination of ctm * device is identity. In the fairly common condition that both the ctm and the device transforms are identity, the function overhead of calling the matrix multiplication on the point overwhelmingly dominates. src/cairo-clip.c | 1 + src/cairo-gstate-private.h | 310 +++++++++++++++++++++++++++++++++++++++++++++ src/cairo-gstate.c | 18 ++- src/cairoint.h | 295 ------------------------------------------ 4 files changed, 327 insertions(+), 297 deletions(-) commit 1349728d1ef63ed562a3fd0ee3c0a89aaba77616 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jun 10 11:26:12 2010 +0100 test: Add partial coverage. Another experiment in measuring consistency of rasterisation stratagems across the backends. test/Makefile.am | 3 + test/Makefile.sources | 1 + test/partial-coverage-rectangles.ref.png | Bin 0 -> 202 bytes test/partial-coverage-reference.ref.png | Bin 0 -> 202 bytes test/partial-coverage-triangles.ref.png | Bin 0 -> 202 bytes test/partial-coverage.c | 213 +++++++++++++++++++++++++++++++ 6 files changed, 217 insertions(+) commit 486118361ad0c2e68fa9f6b44ddb10e84d237eca Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jun 10 11:28:26 2010 +0100 test: Fix half-coverage colouration. Hmm, red over red makes the test pointless. The test still remains of highly dubious quality, as it primarily serves as a rendercheck more than a test of Cairo. The best hope for this test is that it inspires a better one. And incorporate the notes made by Joonas. test/half-coverage.c | 28 +++++++++++++++++++++++++++- 1 file changed, 27 insertions(+), 1 deletion(-) commit 5cb764850f7c405085739647bac2809045e7cdf3 Author: Dmitri Vorobiev <dmitri.vorobiev@movial.com> Date: Wed Jun 9 19:20:27 2010 +0300 scaled-font: optimize cairo_scaled_font_text_to_glyphs() This patch serves two purposes. First, it factors out the heavy part of the cairo_scaled_font_text_to_glyphs() routine thus allowing GCC to better optimize the cache cleanup loop. Keeping the look-up table indices in a separate array speeds up array initialization even further. Second, this patch introduces a shortcut for the case when the string to be rendered consists of a single character. In this case, caching is not necessary at all. We have a benchmark that uses Cairo to render a large amount of random strings of consisting of printable ASCII characters. Below are Oprofile results collected while running this benchmark. It is easy to see that the heavy part becomes noticeably lighter. Before: Profiling through timer interrupt samples % app name symbol name 198755 13.5580 libcairo.so.2.10907.0 cairo_scaled_font_text_to_glyphs 88580 6.0424 libcairo.so.2.10907.0 _cairo_scaled_glyph_lookup 81127 5.5340 libcairo.so.2.10907.0 _cairo_hash_table_lookup 68186 4.6513 libcairo.so.2.10907.0 cairo_scaled_font_glyph_extents 47145 3.2160 libcairo.so.2.10907.0 _composite_glyphs_via_mask 46327 3.1602 libcairo.so.2.10907.0 _cairo_scaled_font_glyph_device_extents 44817 3.0572 libcairo.so.2.10907.0 _composite_glyphs 40431 2.7580 libcairo.so.2.10907.0 .plt After (note that cairo_scaled_font_text_to_glyphs_internal_single() was inlined): Profiling through timer interrupt samples % app name symbol name 107264 7.6406 libcairo.so.2.10907.0 cairo_scaled_font_text_to_glyphs_internal_multiple 87888 6.2604 libcairo.so.2.10907.0 _cairo_scaled_glyph_lookup 79011 5.6281 libcairo.so.2.10907.0 _cairo_hash_table_lookup 71723 5.1090 libcairo.so.2.10907.0 cairo_scaled_font_glyph_extents 48084 3.4251 libcairo.so.2.10907.0 _composite_glyphs_via_mask 46636 3.3220 libcairo.so.2.10907.0 _cairo_scaled_font_glyph_device_extents 44740 3.1869 libcairo.so.2.10907.0 _composite_glyphs 42472 3.0254 libc-2.8.so _int_malloc 39194 2.7919 libcairo.so.2.10907.0 _cairo_gstate_transform_glyphs_to_backend 38614 2.7506 libcairo.so.2.10907.0 .plt 37063 2.6401 libcairo.so.2.10907.0 _cairo_ft_ucs4_to_index 36856 2.6253 libc-2.8.so random 36376 2.5911 libcairo.so.2.10907.0 _cairo_scaled_glyphs_equal 34545 2.4607 libcairo.so.2.10907.0 cairo_matrix_transform_point 31690 2.2573 libc-2.8.so malloc 29395 2.0939 libcairo.so.2.10907.0 _cairo_matrix_is_identity 26142 1.8621 libcairo.so.2.10907.0 _cairo_utf8_to_ucs4 24406 1.7385 libc-2.8.so free 24059 1.7138 libcairo.so.2.10907.0 cairo_scaled_font_text_to_glyphs [ickle: slightly amended for stylistic consistency.] src/cairo-scaled-font.c | 168 ++++++++++++++++++++++++++++++++---------------- 1 file changed, 114 insertions(+), 54 deletions(-) commit c43399fa68696aa62599fb71752850dd46c15c82 Author: Andrea Canciani <ranma42@gmail.com> Date: Wed Jun 9 17:53:09 2010 +0200 gl: fix compilation on MacOS X MacOS X uses different defines to avoid multiple inclusion of GL header files. Adding them to glew.h fixes the compilation when GL is enabled. src/glew/GL/glew.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit 6eb5f859f1d38783dc43541e7e87b0c5f91691d8 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jun 9 10:40:32 2010 +0100 bo: And disable DEBUG_TRAPS again. Meh. I'm going back to bed. Thanks Joonas for catching this. src/cairo-bentley-ottmann.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 85af0ec37456b4eb47c39326b24370a9ccc43092 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jun 9 10:33:29 2010 +0100 test: Add half-coverage to exercise the rasteriser with subpixel geometry. test/Makefile.sources | 1 + test/half-coverage.c | 129 ++++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 130 insertions(+) commit 56c081bdc6d29df932e6993109586ff275d8de12 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jun 9 10:33:01 2010 +0100 bo: Fix debugging for changes in internal traps api. src/cairo-bentley-ottmann.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) commit a946d39555136adc4eff3b68a62ba37f6d9a6299 Author: Benjamin Otte <otte@redhat.com> Date: Tue Jun 8 21:09:51 2010 +0200 gl: Add support for clip regions to the span renderer Clip surface support is still missing, but i suppose that'd need a tiny bit more code... src/cairo-gl-surface.c | 16 +++++++++++++--- 1 file changed, 13 insertions(+), 3 deletions(-) commit f61b3f25aff454800a9d6013012d41633139d896 Author: Benjamin Otte <otte@redhat.com> Date: Tue Jun 8 12:03:36 2010 +0200 gl: Add an assertion that we always have a texture When painting, the sources must be textures and not windows, and we did that wrong previously. This assertion makes sure that never happens again. src/cairo-gl-composite.c | 1 + 1 file changed, 1 insertion(+) commit c6c9a24a1df1826cab78bd8ef1c5a1074bafa0d7 Author: Benjamin Otte <otte@redhat.com> Date: Tue Jun 8 04:11:33 2010 +0200 gl: Use CAIRO_COLOR_BLACK ... instead of creating black on our own - and wrong, too. src/cairo-gl-surface.c | 8 +------- 1 file changed, 1 insertion(+), 7 deletions(-) commit 19bc6793d1d9b209d5a03a727288504b1e71f7c4 Author: Benjamin Otte <otte@redhat.com> Date: Tue Jun 8 04:00:02 2010 +0200 gl: Only clone texture surfaces Using non-texture surfaces as source or mask will fail, so we need to fallback. Caught by the subsurface-modify-child test. src/cairo-gl-surface.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) commit 44483d843e4fe1f5db5cf213f76913d683265bb4 Author: Benjamin Otte <otte@redhat.com> Date: Tue Jun 8 03:59:10 2010 +0200 gl: Fix argument order oops... src/cairo-gl-surface.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 72bf1a536775138d79787729994ec9cfe2193be7 Author: Benjamin Otte <otte@redhat.com> Date: Tue Jun 8 01:50:12 2010 +0200 Update my copyright notices to preferred version No more copyright headers containing my name. test/clear-source.c | 35 ++++++++++++++++++----------------- test/clip-fill-no-op.c | 35 ++++++++++++++++++----------------- test/clip-stroke-no-op.c | 35 ++++++++++++++++++----------------- test/pthread-same-source.c | 35 ++++++++++++++++++----------------- test/pthread-similar.c | 35 ++++++++++++++++++----------------- test/xcomposite-projection.c | 35 ++++++++++++++++++----------------- test/zero-mask.c | 35 ++++++++++++++++++----------------- 7 files changed, 126 insertions(+), 119 deletions(-) commit 10d95c7020ca94903d474beb9c8ede89c186cbe3 Author: Andrea Canciani <ranma42@gmail.com> Date: Tue Jun 8 20:48:25 2010 +0200 test: add/update reference images Add missing reference images and update one-stop references according to current subset policy (outside the defined range, the gradient is clear). test/degenerate-linear-gradient.ref.png | Bin 0 -> 232 bytes test/degenerate-radial-gradient.ref.png | Bin 0 -> 428 bytes test/linear-gradient-one-stop.ref.png | Bin 468 -> 344 bytes test/radial-gradient-one-stop.ref.png | Bin 3772 -> 1422 bytes 4 files changed, 0 insertions(+), 0 deletions(-) commit dd0f8872711d9896416cca2027f578f7d58d228f Author: Andrea Canciani <ranma42@gmail.com> Date: Mon Jun 7 17:34:12 2010 +0200 test: fix typo test/linear-gradient-extend.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 9b88b5dce5327cbd13ea05de949c502fe41f91f7 Author: Andrea Canciani <ranma42@gmail.com> Date: Tue Jun 8 10:45:06 2010 +0200 test: add degenerate gradient and single stop tests Add tests for degeneratate linear gradients (with start point equal to the end point), degenerate radial gradients (start radius and end radius equal to zero, same start and end circle) and gradients (both linear and radial) with just a single stop. test/Makefile.sources | 6 +- test/degenerate-linear-gradient.c | 81 +++++++++++++++++++++++++ test/degenerate-radial-gradient.c | 93 +++++++++++++++++++++++++++++ test/linear-gradient-one-stop.c | 90 ++++++++++++++++++++++++++++ test/linear-gradient-one-stop.ref.png | Bin 0 -> 468 bytes test/radial-gradient-one-stop.c | 107 ++++++++++++++++++++++++++++++++++ test/radial-gradient-one-stop.ref.png | Bin 0 -> 3772 bytes 7 files changed, 376 insertions(+), 1 deletion(-) commit 7d8359721b581845260c0442a174e48b061484a7 Author: Benjamin Otte <otte@redhat.com> Date: Mon Jun 7 16:46:46 2010 +0200 gl: Fix span renderer doing bad stuff for CLEAR and SOURCE SOURCE will fallback now, CLEAR is identical to DEST_OUT with white. src/cairo-gl-surface.c | 7 +++++++ 1 file changed, 7 insertions(+) commit ef8fd1355ea03b8566ef31642b32133d3543e4e8 Author: Benjamin Otte <otte@redhat.com> Date: Mon Jun 7 15:03:37 2010 +0200 gl: Fix span renderer for unbounded spans The span renderer used to not output rects for the top and bottom rows when they didn't contain any spans. src/cairo-gl-surface.c | 72 ++++++++++++++++++++++++++++++++++---------------- 1 file changed, 49 insertions(+), 23 deletions(-) commit 1d11af083f5368c88032a33ff40d0e21c139db18 Author: Benjamin Otte <otte@redhat.com> Date: Sat Jun 5 15:22:44 2010 +0200 gl: Add a simple spans renderer for stroke/fill It's very simple as clipped polygons or ANTIALIAS_NONE still return UNSUPPORTED. Also, no optimizations are done, so even pixel-aligned rectangles use the full span rendering. Still, there are no performance regressions in the benchmark traces and firefox-talos-svg and swfdec-giant-steps both got ~15% faster. src/cairo-gl-surface.c | 178 ++++++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 175 insertions(+), 3 deletions(-) commit 550335efeddf813c1bec493513254fdbd34a9ace Author: Benjamin Otte <otte@redhat.com> Date: Sat Jun 5 15:19:44 2010 +0200 Remove _cairo_surface_composite_trapezoids_as_polygon() The function computed the composite rectangles wrong and was only used in a gl fallback anyway. So instead of trying to fix it, just remove it and make sure gl doesn't fallback. src/cairo-gl-surface.c | 14 --------- src/cairo-spans-private.h | 12 -------- src/cairo-spans.c | 75 ----------------------------------------------- src/cairoint.h | 12 -------- 4 files changed, 113 deletions(-) commit 1e003fce8f7239aaf4608747d63287e0c715ee51 Author: Benjamin Otte <otte@redhat.com> Date: Sat Jun 5 14:58:32 2010 +0200 gl: Fix vertex size changes not causing updates of the operands Check vertex size stays identical when setting up vertices. src/cairo-gl-composite.c | 17 ++++++++++------- 1 file changed, 10 insertions(+), 7 deletions(-) commit 39143400ddd0d4e63a72ce91c423f764e466e405 Author: Benjamin Otte <otte@redhat.com> Date: Fri Jun 4 12:25:52 2010 +0200 gl: Add a gradient texture cache For firefox-planet-gnome, 19135 times a gradient gets rendered using only 10 different gradients. So we get a 100% hit rate in the cache. Unfortunately, texture upload is not the biggest problem of this test, as the performance increase is only moderate - at least on i965: 34.3s => 33.5s src/cairo-gl-device.c | 12 ++++++++- src/cairo-gl-gradient-private.h | 18 +++++++++---- src/cairo-gl-gradient.c | 58 ++++++++++++++++++++++++++++++++++++++--- src/cairo-gl-private.h | 2 ++ 4 files changed, 81 insertions(+), 9 deletions(-) commit 932ab2641ea3183171c3b5699c96c4709fc6bd49 Author: Benjamin Otte <otte@redhat.com> Date: Thu Jun 3 18:54:18 2010 +0200 device: flush before setting finished Otherwise APIs critical for flushing - in particular acquiring the device - do not work. src/cairo-device.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 35e219d08f1c202399655db183ca0105a1c2166f Author: Benjamin Otte <otte@redhat.com> Date: Thu Jun 3 17:50:59 2010 +0200 gl: Make gradient textures a separate object This is necessary so we can do proper refcounting and don't delete the gradient texture prematurely. src/Makefile.sources | 4 +- src/cairo-gl-composite.c | 156 +++------------------------- src/cairo-gl-gradient-private.h | 80 ++++++++++++++ src/cairo-gl-gradient.c | 223 ++++++++++++++++++++++++++++++++++++++++ src/cairo-gl-private.h | 9 +- 5 files changed, 327 insertions(+), 145 deletions(-) commit 9c17a477d29f633a4850ef8b458cbca82a19afc4 Author: Benjamin Otte <otte@redhat.com> Date: Thu Jun 3 17:40:56 2010 +0200 gl: Use the generic functions for filter/extend in gradients src/cairo-gl-composite.c | 26 ++++++++------------------ src/cairo-gl-private.h | 2 ++ 2 files changed, 10 insertions(+), 18 deletions(-) commit df93802765afcf883abef5549bd5dff3465e2d79 Author: Benjamin Otte <otte@redhat.com> Date: Thu Jun 3 17:38:28 2010 +0200 gl: Create separate functions for setting extend and filter src/cairo-gl-composite.c | 47 ++++++++++++++++++++++++++++------------------- 1 file changed, 28 insertions(+), 19 deletions(-) commit 10e71806d2e1929aa127642a397fa6ccef434b5b Author: Benjamin Otte <otte@redhat.com> Date: Wed Jun 2 20:04:57 2010 +0200 gl: Switch to deferred rendering 1) call _cairo_gl_composite_flush() or cairo_surface_flush() where needed 2) Destroy texture operands when necessary 3) get rid of _cairo_gl_composite_end() With this patch, vertices are not flushed immediately anymore, but only when needed or when a new set of vertices is emitted that requires an incompatible setup. This improves performance a lot in particular for text. (gnome-terminal-vim gets 10x faster) src/cairo-gl-composite.c | 40 ++++++++++++++++++++++------------------ src/cairo-gl-glyphs.c | 4 ---- src/cairo-gl-private.h | 4 ---- src/cairo-gl-surface.c | 12 ++++++------ 4 files changed, 28 insertions(+), 32 deletions(-) commit f2f79ca1b3455000df4138ab500ae03b6584250c Author: Benjamin Otte <otte@redhat.com> Date: Wed Jun 2 19:57:47 2010 +0200 gl: Make using shaders an explicit argument This is preparation for a followup patch src/cairo-gl-composite.c | 15 ++++++++------- 1 file changed, 8 insertions(+), 7 deletions(-) commit 19c1d8316e22fc37c7cdaf6c3e56519c06d10561 Author: Benjamin Otte <otte@redhat.com> Date: Wed Jun 2 18:14:28 2010 +0200 gl: Special case blend mode for CAIRO_CONTENT_COLOR This ensures that alpha stays at 1 for RGB in all cases. src/cairo-gl-composite.c | 2 ++ 1 file changed, 2 insertions(+) commit 1f249064cc5c19a39ffe0faaa8825c8f8b6a2175 Author: Benjamin Otte <otte@redhat.com> Date: Wed Jun 2 18:11:04 2010 +0200 gl: rework _cairo_gl_set_operator() 1) store the current operator. This will be useful later to check if the operator changed. 2) pass the context instead of the destination as first argument. The destination is known to be the current target. src/cairo-gl-composite.c | 16 +++++++++------- src/cairo-gl-device.c | 3 +++ src/cairo-gl-private.h | 1 + 3 files changed, 13 insertions(+), 7 deletions(-) commit f66500d8b052251ea3ce06f06d2fa4c8dec2ff3b Author: Benjamin Otte <otte@redhat.com> Date: Wed Jun 2 15:53:48 2010 +0200 gl: Only resetup textures if we need to src/cairo-gl-composite.c | 45 +++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 45 insertions(+) commit d9dcafd61ad6aeecb4538e9fd44e3443a8a50bf1 Author: Benjamin Otte <otte@redhat.com> Date: Wed Jun 2 15:26:45 2010 +0200 gl: Do not reset the current target on release This now also requires a call to cairo_device_flush(). src/cairo-gl-device.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit ce36be5c67fa5b574eb11eb98cdd8c9cd7e09335 Author: Benjamin Otte <otte@redhat.com> Date: Wed Jun 2 15:23:57 2010 +0200 glx: When acquiring device, use the correct drawable src/cairo-glx-context.c | 16 +++++++++++++--- 1 file changed, 13 insertions(+), 3 deletions(-) commit 5819bb07eb7ac64f0e2fbe6a46cc3db0de1d92ce Author: Benjamin Otte <otte@redhat.com> Date: Wed Jun 2 15:23:08 2010 +0200 gl: Store the vertex offset in the operand src/cairo-gl-composite.c | 1 + src/cairo-gl-private.h | 1 + 2 files changed, 2 insertions(+) commit c75460c54d88ce9a50ac59c15a10684043ca4b0e Author: Benjamin Otte <otte@redhat.com> Date: Sun May 30 22:01:10 2010 +0200 gl: Introduce a temporary texture unit ...and use it for image uploads. This makes sure that the texture units used for SOURCE and MASK get not clobbered when images get uploaded to textures. src/cairo-gl-composite.c | 2 ++ src/cairo-gl-device.c | 17 +++++++++++++++++ src/cairo-gl-private.h | 8 +++++++- src/cairo-gl-surface.c | 2 ++ 4 files changed, 28 insertions(+), 1 deletion(-) commit 1ddff8aba35466fe6689536bc5f2d7eeefaea02c Author: Benjamin Otte <otte@redhat.com> Date: Sun May 30 21:25:20 2010 +0200 gl: Only allocate a framebuffer if we need one This way, we don't clobber the current target when creating textures for image surfaces. src/cairo-gl-device.c | 26 ++++++++++++++++++++++++++ src/cairo-gl-private.h | 2 +- src/cairo-gl-surface.c | 20 ++------------------ 3 files changed, 29 insertions(+), 19 deletions(-) commit 1c15510c3d00553aba9ec4fd4e6764da3cdb0a41 Author: Benjamin Otte <otte@redhat.com> Date: Sun May 30 20:12:31 2010 +0200 Call cairo_surface_flush() before setting finished With the current code, the surface will never be flushed as the flush function checks if the surface is finished, and if so, doesn't call the vfunc. Ooops. src/cairo-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 467d5e7f90b9edfb62ca71e67608704424a10aca Author: Benjamin Otte <otte@redhat.com> Date: Fri May 28 21:56:57 2010 +0200 gl: Flush properly in set_destination src/cairo-gl-device.c | 33 +++++++++++++++++---------------- 1 file changed, 17 insertions(+), 16 deletions(-) commit e15348d8d0bbff6968b1f640e63315d159f143f5 Author: Benjamin Otte <otte@redhat.com> Date: Fri May 28 21:55:26 2010 +0200 gl: Don't reset vertex size src/cairo-gl-composite.c | 10 ++++++---- src/cairo-gl-device.c | 1 + 2 files changed, 7 insertions(+), 4 deletions(-) commit a8d6ac1efc3f1f5ff4f92f4e7a3492e62518f3f7 Author: Benjamin Otte <otte@redhat.com> Date: Fri May 28 21:51:08 2010 +0200 gl: Remove check if we're between begin/end This will be useless soon, because we will keep the current state alive as long as possible. src/cairo-gl-composite.c | 1 - src/cairo-gl-private.h | 7 ------- 2 files changed, 8 deletions(-) commit 8f675fb801f7864aa52778d4d53d7ab68f4e664b Author: Benjamin Otte <otte@redhat.com> Date: Fri May 28 21:46:52 2010 +0200 gl: Stop disabling the vertex array and the array buffer src/cairo-gl-composite.c | 12 ++++-------- src/cairo-gl-device.c | 4 ++++ 2 files changed, 8 insertions(+), 8 deletions(-) commit bef0b541497eb5a621df0b5528e08adb0beff961 Author: Benjamin Otte <otte@redhat.com> Date: Fri May 28 21:08:17 2010 +0200 gl: Do not reset shaders on _end() Instead, keep the shader around until cairo_device_flush() is called. src/cairo-gl-composite.c | 35 ++++++++++++++++++++++------------- src/cairo-gl-device.c | 3 +++ src/cairo-gl-private.h | 6 ++++++ 3 files changed, 31 insertions(+), 13 deletions(-) commit 534c14729921cbc2f8e24575eb792de9ddb96e23 Author: Benjamin Otte <otte@redhat.com> Date: Fri May 28 21:06:58 2010 +0200 region: clarify docs of cairo_region_equal() src/cairo-region.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) commit 0fc6f08981604ffdd0109362dcbbd9773501cd9a Author: Benjamin Otte <otte@redhat.com> Date: Fri May 28 20:45:52 2010 +0200 gl: Make _shader_by_type() a getter And refactor code to do the current_shader assigning itself. src/cairo-gl-composite.c | 16 +++++++++++----- src/cairo-gl-private.h | 5 +++-- src/cairo-gl-shaders.c | 13 ++++++++----- 3 files changed, 22 insertions(+), 12 deletions(-) commit 2971ca049881298cd171cf27cd3beb49d2c0572f Author: Benjamin Otte <otte@redhat.com> Date: Fri May 28 20:39:22 2010 +0200 gl: bind all shader variables at the same place Avoids the need to bind the shader when creating it. src/cairo-gl-composite.c | 26 ++++++++++++++++++++------ src/cairo-gl-shaders.c | 11 ----------- 2 files changed, 20 insertions(+), 17 deletions(-) commit e11d8370e0503f085beaa355d85fa6991cd497da Author: Benjamin Otte <otte@redhat.com> Date: Fri May 28 17:37:19 2010 +0200 gl: Set GL_BLEND and GL_SCISSOR_TEST globally Note that we do set them unconditionally in _begin() because a flush might have cleared them. src/cairo-gl-composite.c | 11 +++-------- src/cairo-gl-device.c | 8 ++++++++ src/cairo-gl-surface.c | 1 + 3 files changed, 12 insertions(+), 8 deletions(-) commit 7747f6d9146c19419fdc4fbbdf72eed42dfa731d Author: Benjamin Otte <otte@redhat.com> Date: Fri May 28 17:19:56 2010 +0200 gl: Introduce flush functions The flush functions will clear the state that will in the future be kep unconditionally. src/cairo-gl-composite.c | 6 +++--- src/cairo-gl-device.c | 22 +++++++++++++++++++++- src/cairo-gl-private.h | 4 ++++ src/cairo-gl-surface.c | 36 ++++++++++++++++++++++++++++++++---- 4 files changed, 60 insertions(+), 8 deletions(-) commit 5c74beaaa5dedd82f891f1cc109142f7b6e222a8 Author: Benjamin Otte <otte@redhat.com> Date: Fri May 28 17:19:45 2010 +0200 docs: fix typo src/cairo-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 92d7b1eee96e6b1448aaf3c95a44238fa0eeff48 Author: Benjamin Otte <otte@redhat.com> Date: Fri May 28 16:19:12 2010 +0200 device: Make flush vfunc return a cairo_status_t Mirror the behavior of the surface flush vfunc: Make it return a status and if it's an error set it on the device. src/cairo-device-private.h | 2 +- src/cairo-device.c | 9 +++++++-- src/cairo-xml-surface.c | 4 +++- 3 files changed, 11 insertions(+), 4 deletions(-) commit 63e3cf3888d5b55295a04c4af28e876c04245b85 Author: Benjamin Otte <otte@redhat.com> Date: Fri May 28 15:09:23 2010 +0200 gl: Move unsetting the state into the operand destroy function src/cairo-gl-composite.c | 43 +++++++++++++++++++++++++++++-------------- 1 file changed, 29 insertions(+), 14 deletions(-) commit b9b85dbf3706fe9e50f40b5fda655e71931c2d7f Author: Benjamin Otte <otte@redhat.com> Date: Fri May 28 13:42:11 2010 +0200 gl: get rid of the operand's pattern variable Now someone just needs to teach cairo-gl to not use cairo_pattern_acquire_surface() and we could avoid the nasty hack in cairo_gl_operand_destroy(). src/cairo-gl-composite.c | 21 ++++++++++----------- src/cairo-gl-private.h | 2 -- 2 files changed, 10 insertions(+), 13 deletions(-) commit ae9517a39fdddec389553a64116140e222970ef0 Author: Benjamin Otte <otte@redhat.com> Date: Fri May 28 13:18:28 2010 +0200 gl: Check mask's attributes for component alpha ... and don't duplicate that information into the cairo_gl_composite_t src/cairo-gl-composite.c | 14 ++++++++------ src/cairo-gl-private.h | 1 - 2 files changed, 8 insertions(+), 7 deletions(-) commit f72afc09d0b99030f5766e7bbbb05f6db606e195 Author: Benjamin Otte <otte@redhat.com> Date: Fri May 28 13:14:09 2010 +0200 gl: Merge fixed function setup into general texture setup src/cairo-gl-composite.c | 84 ++++++++++++++++++++---------------------------- 1 file changed, 34 insertions(+), 50 deletions(-) commit dd1faaf02e2a75ca8b9c484abd15726f5c6d3ac4 Author: Benjamin Otte <otte@redhat.com> Date: Fri May 28 12:54:54 2010 +0200 gl: Move function I wanna use it elsewhere and keep the diff small. src/cairo-gl-composite.c | 58 ++++++++++++++++++++++++------------------------ 1 file changed, 29 insertions(+), 29 deletions(-) commit d03800313623fb2b9e7d15c6a659ceaaee2882e8 Author: Benjamin Otte <otte@redhat.com> Date: Fri May 28 12:51:08 2010 +0200 gl: Simplify mask operand setup code Component alpha just has one different value src/cairo-gl-composite.c | 40 +++++++--------------------------------- 1 file changed, 7 insertions(+), 33 deletions(-) commit e20261dddaa194ab4e980da8681e9378d7fd7ef1 Author: Benjamin Otte <otte@redhat.com> Date: Fri May 28 12:48:59 2010 +0200 gl: Get rid of operand argument to emit functions src/cairo-gl-composite.c | 34 +++++++++++++++------------------- src/cairo-gl-glyphs.c | 2 +- src/cairo-gl-private.h | 2 -- src/cairo-gl-surface.c | 11 +++-------- 4 files changed, 19 insertions(+), 30 deletions(-) commit 8062fb352e6cd1d961f0fbc660779f4ad94c7508 Author: Benjamin Otte <otte@redhat.com> Date: Fri May 28 11:03:09 2010 +0200 gl: Add an enum for tex units Instead of hardcoding 0 for source and 1 for mask, use an enum. It's also clearer when functions take a cairo_gl_tex_t argument instead of a GLuint. src/cairo-gl-composite.c | 20 ++++++++++---------- src/cairo-gl-private.h | 5 +++++ src/cairo-gl-shaders.c | 34 ++++++++++++---------------------- 3 files changed, 27 insertions(+), 32 deletions(-) commit 5db362dd6792af20282e0e9a2805ee6f1b921e44 Author: Benjamin Otte <otte@redhat.com> Date: Fri May 28 10:24:33 2010 +0200 gl: Keep a copy of the current operand in the context Note that they are currently only valid as long as the cairo_gl_composite_t exists, but that will be changed soon. src/cairo-gl-composite.c | 20 ++++++++++++++++---- src/cairo-gl-private.h | 2 ++ 2 files changed, 18 insertions(+), 4 deletions(-) commit a05f062c806ac014d0daffa1d74bdb0dd90a8a26 Author: Benjamin Otte <otte@redhat.com> Date: Wed May 26 19:23:18 2010 +0200 gl: Move struct definitions in header src/cairo-gl-private.h | 66 +++++++++++++++++++++++++------------------------- 1 file changed, 33 insertions(+), 33 deletions(-) commit 85d4c6c55b85c52d8f86247f00659e37354abd49 Author: Benjamin Otte <otte@redhat.com> Date: Wed May 26 21:23:40 2010 +0200 gl: operand->texture.surface cannot be NULL src/cairo-gl-composite.c | 10 +++------- 1 file changed, 3 insertions(+), 7 deletions(-) commit 9486ad5fbdad047b5194f6bed68a08bb4c594610 Author: Benjamin Otte <otte@redhat.com> Date: Wed May 26 19:16:41 2010 +0200 gl: Remove composite_t argument from _flush() function src/cairo-gl-composite.c | 17 +++++++---------- src/cairo-gl-glyphs.c | 2 +- src/cairo-gl-private.h | 3 +-- 3 files changed, 9 insertions(+), 13 deletions(-) commit 16e420b4dca5325ab295a27103299a916f334b05 Author: Benjamin Otte <otte@redhat.com> Date: Wed May 26 19:13:53 2010 +0200 gl: Remove dependency on composite_t in draw function src/cairo-gl-composite.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit e46215ec6b94355b741acf5dfdd264d6396b4bc8 Author: Benjamin Otte <otte@redhat.com> Date: Wed May 26 18:44:20 2010 +0200 gl: Remove unneeded and broken workaround We now ensure the values are properly zeroed when setting the operator. src/cairo-gl-composite.c | 14 +------------- 1 file changed, 1 insertion(+), 13 deletions(-) commit b2d709b5c7e463437189b8e4a2c53a3317437825 Author: Benjamin Otte <otte@redhat.com> Date: Wed May 26 17:41:01 2010 +0200 gl: Switch src alpha handling to a tiny on/off function Instead of doing setup every time, we just set this one flag. Much nicer. src/cairo-gl-composite.c | 22 +++++++--------------- 1 file changed, 7 insertions(+), 15 deletions(-) commit 53fd451f85974d2ed0d6f29a0a0549fd4c7035fe Author: Benjamin Otte <otte@redhat.com> Date: Wed May 26 17:32:14 2010 +0200 gl: Avoid redundant mask setup The mask doesn't change, so no need to set it up twice src/cairo-gl-composite.c | 2 -- 1 file changed, 2 deletions(-) commit 1b3675fc4b9c7f6ad4d9edd9b27bee10c2502b81 Author: Benjamin Otte <otte@redhat.com> Date: Wed May 26 17:28:05 2010 +0200 gl: Split fixed-function operand setup into own function src/cairo-gl-composite.c | 151 +++++++++++++---------------------------------- 1 file changed, 42 insertions(+), 109 deletions(-) commit 02da871050f6be02ba0231e545d6613a42c3a7e6 Author: Benjamin Otte <otte@redhat.com> Date: Wed May 26 17:02:29 2010 +0200 gl: Unify fixed function setup code src/cairo-gl-composite.c | 15 ++++----------- 1 file changed, 4 insertions(+), 11 deletions(-) commit 74f9094162fba54c29859e25ad5be085604aa0b6 Author: Benjamin Otte <otte@redhat.com> Date: Wed May 26 15:46:39 2010 +0200 gl: switch OPERAND0 and OPERAND1 in the fixed function path This should not change anything as we multiply the values. But it allows unifying parts of the code, which will happen in another patch. src/cairo-gl-composite.c | 42 +++++++++++++++++++++--------------------- 1 file changed, 21 insertions(+), 21 deletions(-) commit c6aac5d0134f1473d3867c1b6a4b40753700ffcb Author: Benjamin Otte <otte@redhat.com> Date: Wed May 26 15:42:56 2010 +0200 gl: Move the pre shader to the context, too It's required when flushing, too. src/cairo-gl-composite.c | 10 +++++----- src/cairo-gl-private.h | 2 +- 2 files changed, 6 insertions(+), 6 deletions(-) commit 61bc3067ca5fa4aeaef537c397c37826facc7621 Author: Benjamin Otte <otte@redhat.com> Date: Wed May 26 15:28:32 2010 +0200 gl: Store the clip region in the context This is necessary because we want to get rid of keeping the composite stuff around. src/cairo-gl-composite.c | 15 ++++++++++----- src/cairo-gl-device.c | 2 ++ src/cairo-gl-private.h | 1 + 3 files changed, 13 insertions(+), 5 deletions(-) commit db79880d9a836d50e9202caa5c73e22a2b6b97b3 Author: Benjamin Otte <otte@redhat.com> Date: Wed May 26 13:03:12 2010 +0200 gl: Make gl_composite_begin() acquire the context Also remove the previous places for context acquisition. This completes fixing the potential ABBA deadlock I talked about a few commits ago. src/cairo-gl-composite.c | 21 +++++++++++++++++---- src/cairo-gl-glyphs.c | 7 ++++++- src/cairo-gl-private.h | 4 ++-- src/cairo-gl-surface.c | 26 +++++--------------------- 4 files changed, 30 insertions(+), 28 deletions(-) commit 1320a168fa5b73fd42438040a4bf8ee9ad222a0e Author: Benjamin Otte <otte@redhat.com> Date: Wed May 26 12:26:49 2010 +0200 gl: Remove unused code We don't need to acquire/release the context, we're not using it. src/cairo-gl-glyphs.c | 11 +---------- 1 file changed, 1 insertion(+), 10 deletions(-) commit 8f5c3b706b5bad6b0851a3e27752f7d1a3ca8a72 Author: Benjamin Otte <otte@redhat.com> Date: Wed May 26 12:16:01 2010 +0200 gl: Make composite_set_clip_region not require the context src/cairo-gl-composite.c | 3 +-- src/cairo-gl-glyphs.c | 2 +- src/cairo-gl-private.h | 3 +-- src/cairo-gl-surface.c | 2 +- 4 files changed, 4 insertions(+), 6 deletions(-) commit ac6abcbac7f79e5717c44e251135dcba8ab00b27 Author: Benjamin Otte <otte@redhat.com> Date: Wed May 26 12:14:24 2010 +0200 gl: Make composite_set_source/mask/mask_spans not require a context src/cairo-gl-composite.c | 77 +++++++++++++++++++++++++----------------------- src/cairo-gl-glyphs.c | 5 ++-- src/cairo-gl-private.h | 9 ++---- src/cairo-gl-surface.c | 13 ++++---- 4 files changed, 51 insertions(+), 53 deletions(-) commit 88a70d66fcdbca7e489baa48cd145bb69c3e28e0 Author: Benjamin Otte <otte@redhat.com> Date: Wed May 26 11:59:36 2010 +0200 gl: Add API for querying if we need GL_TEXTURE_RECTANGLE_EXT We need to do that querying outside of taking the lock, so we want to be able to query the device, not the context. src/cairo-gl-composite.c | 11 +++++------ src/cairo-gl-glyphs.c | 2 +- src/cairo-gl-private.h | 6 ++++++ 3 files changed, 12 insertions(+), 7 deletions(-) commit 9c6e1338a1c7d61b4cc84bcdc5c38573a34cd561 Author: Benjamin Otte <otte@redhat.com> Date: Wed May 26 11:36:22 2010 +0200 gl: Make cairo_gl_composite_t not require a gl context anymore First patch in a series. The idea is to make all gl operations in Cairo work like this: 1) prepare cairo_gl_composite_t 2) acquire device 3) begin composite operation 4) emit vertexes 5) end operation 6) release device 7) cleanup composite operation This both reduces the time required with the device locked, cleans the API and more importantly avoids ABBA deadlocks, as we currently call acquire_pattern() with the device lock held. src/cairo-gl-composite.c | 6 ++---- src/cairo-gl-glyphs.c | 7 +++---- src/cairo-gl-private.h | 6 ++---- src/cairo-gl-surface.c | 19 ++++++++----------- 4 files changed, 15 insertions(+), 23 deletions(-) commit 60f0cb9c60556f94af903dd6f9da0628a05a1f7f Author: Benjamin Otte <otte@redhat.com> Date: Sat May 22 18:42:44 2010 +0200 gl: Flush surface at the right times src/cairo-gl-composite.c | 2 ++ src/cairo-gl-device.c | 2 ++ src/cairo-gl-surface.c | 2 ++ 3 files changed, 6 insertions(+) commit ffeeae462ea0f8c7af326f3bcc66b88881bd8fdd Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Jun 6 16:50:34 2010 +0100 Revert "gstate: reduce a couple more operators" This reverts commit a85c6c1e36273d534f01ade0714b0592b6bed5ab. Soeren found that this introduced a regression when displaying pages in evince, and since I'm not overly enthralled at the added complexity here for very little gain, remove the reduction and leave it clear for a more complete solution in the future. src/cairo-gstate.c | 15 +-------------- 1 file changed, 1 insertion(+), 14 deletions(-) commit 44e628a6f4e6b9b21ad67b08869186a779d0db5a Author: Makoto Kato <m_kato@ga2.so-net.ne.jp> Date: Mon May 31 18:22:40 2010 +0900 openvg: fix build break src/cairo-vg-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit cc080a1e2731a42045709d789587070da944145c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu May 27 20:51:55 2010 +0100 gl: Unmap the vertex buffer prior to using in DaawArrays() "It is an INVALID_OPERTION error to source data from a buffer object that is currently mapped." mesa is currently lax in this regard, but the Nvidia libGL conforms to the spec and thus was causing failures. src/cairo-gl-composite.c | 23 ++++++++++++----------- 1 file changed, 12 insertions(+), 11 deletions(-) commit b3e15bd5c307352065555c6c3ae5f0bd15b9a791 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu May 27 18:43:03 2010 +0100 gl: _cairo_set_shader_by_type() cannot return UNSUPPORTED So remove the superfuous test. src/cairo-gl-composite.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) commit 59bd34e6a3661483f60bb01be8b406a5c38e1131 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu May 27 18:35:40 2010 +0100 gl: Check for an error device when creating surfaces. src/cairo-gl-surface.c | 3 +++ 1 file changed, 3 insertions(+) commit 53f086105975c1c2bd4c43a1b0613ebde3580e95 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu May 27 18:27:53 2010 +0100 gl: Make _cairo_gl_shader_(init|compile) static Only used within _cairo_gl_shaders.c, so remove the private symbol. src/cairo-gl-private.h | 10 ---------- src/cairo-gl-shaders.c | 41 +++++++++++++++++++++++++---------------- 2 files changed, 25 insertions(+), 26 deletions(-) commit dd0866cdc878a0be438a117051ccc64f12f81970 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu May 27 17:52:11 2010 +0100 boilerplate: Exercise RENDER capable windows. Ensure that the drivers actually work when rendering to a normal window. boilerplate/cairo-boilerplate-xlib.c | 107 +++++++++++++++++++++++++++++++++++ 1 file changed, 107 insertions(+) commit 2dc42af9422a15f902934a53c074e4194991a9e4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon May 17 16:32:20 2010 +0100 script: Update csi-exec to match new interface. util/cairo-script/csi-exec.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit 2f0f4ed0e2ae5657dedfe180b7be3a2496753ba8 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu May 20 21:54:51 2010 +0100 freed-pool: Don't access beyond the end of the array. Argh. This bug has been here for quite some time and only showed itself with a corrupt pointer on ppc32. Since the erroneous write is inside the block, it remained undetected by valgrind. src/cairo-freed-pool-private.h | 8 +++++--- src/cairo-freed-pool.c | 4 +++- 2 files changed, 8 insertions(+), 4 deletions(-) commit 97b4aeba44e5f4b23b3ed417b991570b46d4736e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu May 20 21:54:16 2010 +0100 atomic: Add memory barriers if required for intel primitives src/cairo-atomic-private.h | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) commit 24e684c267404528b40fa49b55272d73bd44167d Author: Benjamin Otte <otte@gnome.org> Date: Thu May 20 22:37:18 2010 +0200 gl: spans need the dummy texture Thanks Eric. That took me half a day. src/cairo-gl-composite.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit d7a564d8eed32b4417fdb4c8c5d91467e4ae1019 Author: Benjamin Otte <otte@redhat.com> Date: Thu May 20 22:10:06 2010 +0200 gl: Don't set is_clear 1) It's wrong. 2) The core functions set the flag correctly already. src/cairo-gl-surface.c | 2 -- 1 file changed, 2 deletions(-) commit f9cf07b126293d00df085d4fbfa337e1132e0258 Author: Benjamin Otte <otte@redhat.com> Date: Thu May 20 16:08:38 2010 +0200 gl: Fix coverage passing for spans not being endian safe. src/cairo-gl-composite.c | 29 ++++++++++++++++------------- src/cairo-gl-private.h | 2 +- src/cairo-gl-surface.c | 4 ++-- 3 files changed, 19 insertions(+), 16 deletions(-) commit 2a0f34c6da2eb4f10b198d04894fc1413352e041 Author: Benjamin Otte <otte@redhat.com> Date: Thu May 20 11:23:18 2010 +0200 gl: Allow NONE and SPANS sampler for source, too src/cairo-gl-shaders.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) commit 1586f7740ad436447ac75c6b869da7446172a63c Author: Benjamin Otte <otte@gnome.org> Date: Thu May 20 13:17:52 2010 +0200 gl: Refactor fixed function setup code In particular get rid of the _cairo_gl_set_tex_combine_constant_color() function by inlining it and move common glTexEnv calls out of the switch statement. src/cairo-gl-composite.c | 130 +++++++++++++++++------------------------------ 1 file changed, 46 insertions(+), 84 deletions(-) commit 4889090c76bdab5e4eb887593caf5d5a34f90864 Author: Benjamin Otte <otte@gnome.org> Date: Thu May 20 12:13:43 2010 +0200 gl: set the active texture before fiddling with with glTexEnv fixes the fixed function pipeline src/cairo-gl-composite.c | 8 ++++++++ 1 file changed, 8 insertions(+) commit 3cff2da912150868390bf4b9df166c88b3dfcf2e Author: Benjamin Otte <otte@gnome.org> Date: Thu May 20 12:03:18 2010 +0200 gl: Remove unused variables src/cairo-gl-composite.c | 9 --------- 1 file changed, 9 deletions(-) commit 4c7dd1be4526c85b47f1fe1d4663e7fbec0926b1 Author: Benjamin Otte <otte@redhat.com> Date: Wed May 19 22:03:18 2010 +0200 gl: Move vertex array setup code into the texture setup code src/cairo-gl-composite.c | 35 ++++++++++++++--------------------- 1 file changed, 14 insertions(+), 21 deletions(-) commit d9d5976bdf333a5a7ea7707bc3076696f3c079d2 Author: Benjamin Otte <otte@redhat.com> Date: Wed May 19 21:54:32 2010 +0200 gl: Add an assertion that a context cannot call _begin() twice Evert _begin() call must be followed by an _end() call to avoid any nastiness. Just like GL. :) src/cairo-gl-composite.c | 2 ++ src/cairo-gl-private.h | 7 +++++++ 2 files changed, 9 insertions(+) commit 5b8b1fe1cb692530b40db330c3040cd85a8b6bd7 Author: Benjamin Otte <otte@redhat.com> Date: Wed May 19 21:10:26 2010 +0200 gl: Move vertex buffer handling from setup into context We can only emit one vertex buffer at a time anyway. src/cairo-gl-composite.c | 34 ++++++++++++++++++---------------- src/cairo-gl-private.h | 8 ++++---- 2 files changed, 22 insertions(+), 20 deletions(-) commit 0d5f2eab77f055b063538dcd345a1e0db480f3bc Author: Benjamin Otte <otte@redhat.com> Date: Wed May 19 20:49:49 2010 +0200 gl: Actuall do fall back to fixed function paths. Previously we were just returning UNSUPPORTED all the time. I guess that explains why my code got so slow... src/cairo-gl-composite.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) commit bb7ca71c1af416ca084f03089f5fe762403635dd Author: Benjamin Otte <otte@redhat.com> Date: Wed May 19 20:47:31 2010 +0200 gl: Don't store the shader in the composite_t anymore Now that we have ctx->current_shader, there's no need to keep it in the composite setup code. src/cairo-gl-composite.c | 30 ++++++++++++++---------------- src/cairo-gl-private.h | 10 ++++------ src/cairo-gl-shaders.c | 16 ++++++---------- 3 files changed, 24 insertions(+), 32 deletions(-) commit 6c477b4a945a8f8b866bb2f5e125d4a0328d05f4 Author: Benjamin Otte <otte@redhat.com> Date: Wed May 19 20:10:19 2010 +0200 gl: Move shader checking in the TexEnv setup code As the previously generic setup code only contains non-shader code now, we can check in advance that no shader is in use and do all the other stuff without checks. src/cairo-gl-composite.c | 208 +++++++++++++++++++++++------------------------ 1 file changed, 101 insertions(+), 107 deletions(-) commit 73262edd8201de8ab8a01e01f83af5b5d6470f9e Author: Benjamin Otte <otte@redhat.com> Date: Wed May 19 19:25:03 2010 +0200 gl: Move texture setup code into its own file This ensures textures don't need to be (re)set when doing component alpha and running 2 shaders. src/cairo-gl-composite.c | 208 +++++++++++++++++++++-------------------------- 1 file changed, 92 insertions(+), 116 deletions(-) commit 970c879dd897019f90b0c8ff5a18781be4935c55 Author: Benjamin Otte <otte@redhat.com> Date: Wed May 19 18:34:00 2010 +0200 gl: Binding variables only works for the current shader So make the code reflect that. src/cairo-gl-composite.c | 27 ++++++++++++--------------- src/cairo-gl-private.h | 6 ------ src/cairo-gl-shaders.c | 22 ++++++++-------------- 3 files changed, 20 insertions(+), 35 deletions(-) commit 065969d78f23c634ebcb2b931cbb1fe184f2cedf Author: Benjamin Otte <otte@redhat.com> Date: Wed May 19 18:21:06 2010 +0200 gl: Separate binding to shaders from setting operands That way we only need to bind to shaders when creating them, but not everytime we switch shaders on text output. src/cairo-gl-composite.c | 156 +++++++++++++++++++++++------------------------ src/cairo-gl-shaders.c | 2 +- 2 files changed, 79 insertions(+), 79 deletions(-) commit c53f7778bcf274f64f801ef9ee1641a088842f60 Author: Benjamin Otte <otte@redhat.com> Date: Wed May 19 16:50:57 2010 +0200 gl: Use the mask, not the source Not that we'd ever have constant color component-alpha masks, but that's an entirely different topic. src/cairo-gl-composite.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) commit 26e8e506b66022910c262da0618ef5be8c5a080e Author: Benjamin Otte <otte@redhat.com> Date: Wed May 19 13:19:56 2010 +0200 gl: Add _cairo_gl_surface_is_texture() It's more explicit then just checking surface->fb != 0 Also, it makes improving that check easier. src/cairo-gl-device.c | 4 ++-- src/cairo-gl-private.h | 6 ++++++ src/cairo-gl-surface.c | 11 ++++++----- 3 files changed, 14 insertions(+), 7 deletions(-) commit e4cb46b50015f792b0e8c7b49f3b42f2c3c0ad00 Author: Benjamin Otte <otte@redhat.com> Date: Wed May 19 12:10:10 2010 +0200 gl: Implement texture upload to window with composite() Instead of custom code that does basically the same thing, use a temporary surface and _cairo_gl_surface_composite() to upload images. src/cairo-gl-shaders.c | 2 +- src/cairo-gl-surface.c | 129 ++++++++++++++++++------------------------------- 2 files changed, 48 insertions(+), 83 deletions(-) commit 7c8759e279f51ea722c07e4b0c70f4a49f34393d Author: Benjamin Otte <otte@redhat.com> Date: Wed May 19 11:46:12 2010 +0200 gl: cache the current shader in the context src/cairo-gl-private.h | 4 +++- src/cairo-gl-shaders.c | 7 +++++++ 2 files changed, 10 insertions(+), 1 deletion(-) commit 6864ca6c25aa8147a562d7f693a95025dde807a8 Author: Benjamin Otte <otte@redhat.com> Date: Wed May 19 11:10:11 2010 +0200 gl: Move _cairo_gl_set_operator() to the file it's used in Also make it a private function. src/cairo-gl-composite.c | 57 ++++++++++++++++++++++++++++++++++++++++++++++++ src/cairo-gl-private.h | 4 ---- src/cairo-gl-surface.c | 57 ------------------------------------------------ 3 files changed, 57 insertions(+), 61 deletions(-) commit 0d2d4c59026c31da084e6797d109a230341b396d Author: Benjamin Otte <otte@redhat.com> Date: Tue May 18 21:56:55 2010 +0200 gl: Detect color-pattern clears of surfaces ... and use glClear() there. The common case here is a solid color and OPERATOR_OVER. src/cairo-gl-private.h | 3 --- src/cairo-gl-surface.c | 38 +++++++++++++++++++++++++++++--------- 2 files changed, 29 insertions(+), 12 deletions(-) commit b70eb275c4aa54f7c90a985b5de67da12ac08a30 Author: Benjamin Otte <otte@redhat.com> Date: Tue May 18 21:56:18 2010 +0200 gl: detect image uploads and fast-path them src/cairo-gl-surface.c | 23 +++++++++++++++++++++++ 1 file changed, 23 insertions(+) commit 8adbf3aa19fd9fb8200973ecd72cdebfd9a5364d Author: Benjamin Otte <otte@redhat.com> Date: Tue May 18 20:12:44 2010 +0200 gl: Use _cairo_gl_surface_draw_image() when caching glyphs src/cairo-gl-glyphs.c | 47 ++++++++--------------------------------------- 1 file changed, 8 insertions(+), 39 deletions(-) commit 915aa64ef077fe3b6094d0f1c8bd87ab5acc4868 Author: Benjamin Otte <otte@redhat.com> Date: Tue May 18 13:01:30 2010 +0200 gl: Actually check errors for _cairo_gl_composite_begin() src/cairo-gl-glyphs.c | 2 ++ 1 file changed, 2 insertions(+) commit 0222cc7d76b0dc724630b660687f98a2460eb0b9 Author: Benjamin Otte <otte@redhat.com> Date: Tue May 18 13:00:34 2010 +0200 gl: Get rid of _cairo_gl_composite_set_mask_texture() Now that the glyph cache contains real surfaces, there's no longer a need to have a special function for it. src/cairo-gl-composite.c | 16 ---------------- src/cairo-gl-glyphs.c | 15 ++++++++------- src/cairo-gl-private.h | 6 ------ 3 files changed, 8 insertions(+), 29 deletions(-) commit 208d9f2a7e1f66ff2764c41fe67f65e27279b2ab Author: Benjamin Otte <otte@redhat.com> Date: Tue May 18 12:49:59 2010 +0200 gl: Make glyph cache a real surface This has the huge advantage that we can use real surface functions on it. src/cairo-gl-glyphs.c | 67 +++++++++++++++++++------------------------------- src/cairo-gl-private.h | 3 +-- 2 files changed, 26 insertions(+), 44 deletions(-) commit 050117996339cfe35add1f2cd44d0e5578d4a981 Author: Benjamin Otte <otte@redhat.com> Date: Tue May 18 12:02:54 2010 +0200 surface: Add _cairo_surface_release_device_reference() API See the API documentation for that function about its intended purpose. src/cairo-surface-private.h | 1 + src/cairo-surface.c | 26 +++++++++++++++++++++++++- src/cairoint.h | 3 +++ 3 files changed, 29 insertions(+), 1 deletion(-) commit c489f67d1d1d6079f9b58b726ed42beea1dfc44e Author: Benjamin Otte <otte@redhat.com> Date: Mon May 17 20:45:42 2010 +0200 gl: Make a bunch of functions static They are no longer used outside of cairo_gl_composite_t. src/cairo-gl-composite.c | 8 ++++---- src/cairo-gl-private.h | 16 ---------------- 2 files changed, 4 insertions(+), 20 deletions(-) commit c8008e118a54d62690d6555c9ffabf734c39e69f Author: Benjamin Otte <otte@redhat.com> Date: Mon May 17 19:16:20 2010 +0200 gl: Don't name the union in cairo_gl_operand_t Gets rid of ultra-long variable names. src/cairo-gl-composite.c | 152 +++++++++++++++++++++++------------------------ src/cairo-gl-private.h | 2 +- 2 files changed, 77 insertions(+), 77 deletions(-) commit 53260148ea9abacdc510826765c1d63f486add6c Author: Andrea Canciani <ranma42@gmail.com> Date: Fri May 14 21:04:22 2010 +0200 quartz: fix repeating radial gradients on 64 bits Quartz API uses the type CGFloat, which is double on 64 bits, float on 32 bits architectures src/cairo-quartz-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 0104976dc853b60de14080a749a6f729852f1b22 Author: Andrea Canciani <ranma42@gmail.com> Date: Fri May 14 21:03:19 2010 +0200 quartz: silence compiler warnings All the possible enum values have to be explicitly handled in switch statements. src/cairo-quartz-surface.c | 2 ++ 1 file changed, 2 insertions(+) commit ac94a00a4d3acb36f011330b78d1311534fe90e0 Author: Andrea Canciani <ranma42@gmail.com> Date: Fri May 14 19:33:15 2010 +0200 quartz: fallback for A1 images on little endian archs Quartz uses a bit ordering that is consistent with cairo on big endian architectures, but isn't on little endian architectures. Fixes a1-mask and large-source-roi (on little endian) src/cairo-quartz-surface.c | 2 ++ 1 file changed, 2 insertions(+) commit f40e3b09a130a9ccbe310120fd5cfc7ad7828a4d Author: Andrea Canciani <ranma42@gmail.com> Date: Thu May 13 12:15:19 2010 +0200 quartz-font: Silence 0x0 CGContext warning Silence Quartz complaints about operations on empty contexts: <Error>: CGContextSetFont: invalid context 0x0 <Error>: CGContextSetFontSize: invalid context 0x0 <Error>: CGContextSetTextMatrix: invalid context 0x0 <Error>: CGContextSetAlpha: invalid context 0x0 <Error>: CGContextShowGlyphsAtPoint: invalid context 0x0 src/cairo-quartz-font.c | 77 +++++++++++++++++++++++++++---------------------- 1 file changed, 42 insertions(+), 35 deletions(-) commit 1687e6169463947554f5476674a577e67e2c543b Author: Andrea Canciani <ranma42@gmail.com> Date: Thu May 13 11:54:32 2010 +0200 quartz-font: Conform context and antialias handling to quartz-surface A8 surfaces are now kAlphaOnly surfaces in quartz-font too. Subpixel font smoothing can be enabled. src/cairo-quartz-font.c | 38 ++++++++++++++++++++++++++++---------- 1 file changed, 28 insertions(+), 10 deletions(-) commit d68bbc0fe5706634e4c025bdcb311905b7f008e4 Author: Benjamin Otte <otte@gnome.org> Date: Mon May 17 20:57:08 2010 +0200 gl: Add missing NULL-check in non-shader code src/cairo-gl-composite.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 7f22372edf10f5eab53ea132d9226f6d5b49eb9a Author: Benjamin Otte <otte@redhat.com> Date: Mon May 17 14:20:17 2010 +0200 gl: Get rid of src and mask argument to composite_init() This was thought of as a way to detect self-painting, but it's better to check that later in _begin() - when we check it. src/cairo-gl-composite.c | 2 -- src/cairo-gl-glyphs.c | 2 +- src/cairo-gl-private.h | 2 -- src/cairo-gl-surface.c | 9 ++++----- 4 files changed, 5 insertions(+), 10 deletions(-) commit 75f15babed8409fb1eaae514411de01d57072c80 Author: Benjamin Otte <otte@redhat.com> Date: Mon May 17 14:17:47 2010 +0200 gl: Implement fill_rectangles with composite Tiny performance improvement, but a lot less code. src/cairo-gl-surface.c | 204 ++++++++++--------------------------------------- 1 file changed, 41 insertions(+), 163 deletions(-) commit e7c5b4f79fa24b75026c93bc9244115d2ee23930 Author: Benjamin Otte <otte@redhat.com> Date: Mon May 17 13:56:22 2010 +0200 gl: make _set_mask() accept NULL That way, it can be used to unset the mask. src/cairo-gl-composite.c | 3 +++ src/cairo-gl-surface.c | 14 ++++++-------- 2 files changed, 9 insertions(+), 8 deletions(-) commit 145db7a427e594f1d68a6c0f4afe5e365e219cf0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon May 17 12:37:46 2010 +0100 gl: Check status after composite_begin Skip the unnecessary work in _cairo_gl_surface_composite() if we fail to setup the GL composite operation. src/cairo-gl-surface.c | 2 ++ 1 file changed, 2 insertions(+) commit 8695c41e80ce932fd5dd434307b92f323a937c7c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon May 17 12:36:54 2010 +0100 gl: Throw a _cairo_error when detecting a GLerror This allows us to easily set an breakpoint and inspect cairo's state when we do something illegal. src/cairo-gl-private.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) commit 1c18ab02c8d41be8e5886cbb34aeda5099189fef Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon May 17 11:42:26 2010 +0100 gl: Rename cairo_gl_shader_program_t to cairo_gl_shader_t And complete the move of the shaders into the cairo_gl_shader namespace. Of particular note, the bind_*_to_shader become _cairo_gl_shader_bind_*() and have proper types. src/cairo-gl-composite.c | 138 ++++++------- src/cairo-gl-device.c | 14 +- src/cairo-gl-private.h | 93 +++++---- src/cairo-gl-shaders.c | 508 ++++++++++++++++++++++++++--------------------- src/cairo-gl-surface.c | 54 ++--- 5 files changed, 424 insertions(+), 383 deletions(-) commit 2325d755b03958d8ead9a995c0d1d03e4f93af0c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon May 17 10:33:46 2010 +0100 gl: Pretty print the error GLenum. src/cairo-gl-private.h | 3 ++- src/cairo-gl-surface.c | 19 +++++++++++++++++++ 2 files changed, 21 insertions(+), 1 deletion(-) commit 7ef1bd22ded512f4fad3959796d7f40c4ddc5824 Author: Benjamin Otte <otte@redhat.com> Date: Mon May 17 01:12:34 2010 +0200 gl: Update copyright statements - add 2010 Red Hat copyright for my code - include myself as author - update contributors list src/cairo-gl-composite.c | 5 ++++- src/cairo-gl-device.c | 5 ++++- src/cairo-gl-glyphs.c | 5 +++++ src/cairo-gl-private.h | 6 ++++-- src/cairo-gl-shaders.c | 8 +++++++- src/cairo-gl-surface.c | 5 ++++- 6 files changed, 28 insertions(+), 6 deletions(-) commit 0b69e51b526a609a2dc5f1d2fb9c49f240feb57b Author: Benjamin Otte <otte@redhat.com> Date: Mon May 17 00:56:21 2010 +0200 gl: Assert that binding values to shaders never fails As that would be a programming error, we can use an assert and get rid of return values that we weren't checking anyway. src/cairo-gl-composite.c | 129 ++++++++++++++++++++--------------------------- src/cairo-gl-private.h | 12 ++--- src/cairo-gl-shaders.c | 114 ++++++++++++++++------------------------- src/cairo-gl-surface.c | 15 +++--- 4 files changed, 111 insertions(+), 159 deletions(-) commit 0bb11e8f5bcdc8a6f8384197b39c2ad48eaab1cd Author: Benjamin Otte <otte@redhat.com> Date: Mon May 17 00:49:01 2010 +0200 gl: Use GL_TRIANGLES instead of GL_QUADS It seems to increase performance by 1-2% on average and up to 30% in some cases (on i965). src/cairo-gl-composite.c | 16 +++++++++++----- 1 file changed, 11 insertions(+), 5 deletions(-) commit 2b771da47a87f86f378314bfe8033874d9444914 Author: Benjamin Otte <otte@redhat.com> Date: Sun May 16 23:58:47 2010 +0200 xlib: ensure variable isn't uninitialized src/cairo-xlib-surface.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) commit b768a33a602942825e5ec651ae7bafd6d5fc33ac Author: Benjamin Otte <otte@redhat.com> Date: Sun May 16 21:10:45 2010 +0200 gl: Set composite->has_component_alpha based on mask Previously it was set when initing the composite structure, but that doesn't allow resetting it when a new mask is used. src/cairo-gl-composite.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) commit 1cde4601df8a3689450e2a0e15888c53aff12983 Author: Benjamin Otte <otte@redhat.com> Date: Sun May 16 21:05:16 2010 +0200 gl: Replace _cairo_gl_operand_init() with setters for src/mask Now cairo_gl_operand_t is no longer exposed outside of cairo-gl-composite.c src/cairo-gl-composite.c | 34 +++++++++++++++++++++++++++++++++- src/cairo-gl-glyphs.c | 10 +++++----- src/cairo-gl-private.h | 25 ++++++++++++++++--------- src/cairo-gl-surface.c | 26 +++++++++++++------------- 4 files changed, 67 insertions(+), 28 deletions(-) commit 6575cb7a95e109ab39eda58603ed6a76581a41b0 Author: Benjamin Otte <otte@redhat.com> Date: Sun May 16 20:39:00 2010 +0200 gl: Convert glyphs path to use cairo_gl_composite_t src/cairo-gl-composite.c | 122 ++++++++++++----- src/cairo-gl-glyphs.c | 331 ++++++----------------------------------------- src/cairo-gl-private.h | 22 ++++ 3 files changed, 150 insertions(+), 325 deletions(-) commit 8937f2b4b5e4bcb8ada1e478df6a7329c0b80a5a Author: Benjamin Otte <otte@redhat.com> Date: Sun May 16 13:19:48 2010 +0200 gl: Require ARB_vertex_buffer_object extension unconditionally <anholt> we shouldn't bother with non-vbo support. there's no justification for a driver not supporting it. src/cairo-gl-composite.c | 1 - src/cairo-gl-device.c | 5 +++-- src/cairo-gl-glyphs.c | 3 --- src/cairo-gl-private.h | 7 ------- src/cairo-gl-surface.c | 12 ++++-------- 5 files changed, 7 insertions(+), 21 deletions(-) commit 29dfd5b488e16ec8be879b2415635df93b874afc Author: Benjamin Otte <otte@redhat.com> Date: Sun May 16 03:36:05 2010 +0200 gl: Remove unused variables src/cairo-gl-surface.c | 7 +------ 1 file changed, 1 insertion(+), 6 deletions(-) commit 4144488f516214b675d25d48add8cbd83fd17025 Author: Benjamin Otte <otte@redhat.com> Date: Sun May 16 03:29:04 2010 +0200 gl: Convert span renderer to use cairo_gl_composite_begin/end() Adds two new APIs to the composite API: - _cairo_gl_composite_set_mask_spans() Tells the composite renderer to use the color values for spans. - _cairo_gl_composite_set_clip_region() Tells the composite renderer to use GL_SCISSORS_TEST to clip when drawing. This function is very unperformant, so if it can be avoided, it probably shouldn't be used. src/cairo-gl-composite.c | 72 +++++++++++++- src/cairo-gl-private.h | 10 ++ src/cairo-gl-surface.c | 247 ++++++----------------------------------------- 3 files changed, 110 insertions(+), 219 deletions(-) commit 6e81d85e3682f14e1c5dbe97dda2cc1fcd32df3a Author: Benjamin Otte <otte@redhat.com> Date: Sun May 16 03:27:28 2010 +0200 gl: Change _cairo_gl_composite_emit_rect() Two changes: - The function now takes x1/y1,x2/y2 instead of x,y,width,height - The function takes a color uint32_t. This will be used by spans. src/cairo-gl-composite.c | 27 +++++++++++++++------------ src/cairo-gl-private.h | 9 +++++---- src/cairo-gl-surface.c | 10 ++++++---- 3 files changed, 26 insertions(+), 20 deletions(-) commit f8398cc2d83eb9ac37874f71d4f8805dbf0df412 Author: Benjamin Otte <otte@redhat.com> Date: Sun May 16 02:34:18 2010 +0200 gl: remove unused variables from span renderer src/cairo-gl-surface.c | 7 ------- 1 file changed, 7 deletions(-) commit f5c5077def47b91b57dfbc8c2622b4aaf5ca8fbb Author: Benjamin Otte <otte@redhat.com> Date: Sun May 16 02:06:43 2010 +0200 gl: Support component alpha in gl_composite_t Use that to get rid of cairo_gl_surface_composite_component_alpha() src/cairo-gl-composite.c | 165 ++++++++++++++++++++++++++-- src/cairo-gl-glyphs.c | 2 +- src/cairo-gl-private.h | 4 + src/cairo-gl-surface.c | 280 +---------------------------------------------- 4 files changed, 167 insertions(+), 284 deletions(-) commit 88c118f891ca9ba472b8acff33829de845bb507c Author: Benjamin Otte <otte@redhat.com> Date: Sun May 16 00:24:17 2010 +0200 gl: Implement _composite_emit_rect() and _composite_end() And use it in cairo_gl_surface_composite() src/cairo-gl-composite.c | 164 +++++++++++++++++++++++++++++++++++++++++++++++ src/cairo-gl-private.h | 19 ++++++ src/cairo-gl-surface.c | 107 +++---------------------------- 3 files changed, 192 insertions(+), 98 deletions(-) commit b96735f7b465a16182c4c8555878ba16b947e8f9 Author: Benjamin Otte <otte@redhat.com> Date: Sat May 15 22:12:12 2010 +0200 gl: Add _cairo_gl_composite_begin() function This is the first step at adding the following function set: _cairo_gl_composite_begin() _cairo_gl_composite_emit_*() _cairo_gl_composite_end() which will take care of proiding a unified model for emitting vertexes using vbos instead of the 5 different methods we currently have. src/cairo-gl-composite.c | 30 ++++++++++++++++++++++++++++++ src/cairo-gl-private.h | 6 ++++++ src/cairo-gl-surface.c | 22 +--------------------- 3 files changed, 37 insertions(+), 21 deletions(-) commit 211a8b8234d24dd8e00332cb0b0e625ac1ffef58 Author: Benjamin Otte <otte@redhat.com> Date: Sat May 15 21:27:58 2010 +0200 gl: make _cairo_operand_destroy private src/cairo-gl-composite.c | 2 +- src/cairo-gl-private.h | 3 --- 2 files changed, 1 insertion(+), 4 deletions(-) commit bf1060252abaa97eb5ce149f8b99055b42096c67 Author: Benjamin Otte <otte@redhat.com> Date: Sat May 15 21:26:55 2010 +0200 gl: make show_glyphs use the cairo_composite_t API src/cairo-gl-glyphs.c | 17 ++++++++++------- 1 file changed, 10 insertions(+), 7 deletions(-) commit 948290de12c5f451d361903f0f52e0cde5527180 Author: Benjamin Otte <otte@redhat.com> Date: Sat May 15 20:33:00 2010 +0200 gl: Make the span renderer use the new composite APIs src/cairo-gl-surface.c | 32 ++++++++++++++++++++------------ 1 file changed, 20 insertions(+), 12 deletions(-) commit 7ad8c3b45691ce0c8a86180bb3c1117b9a9980c0 Author: Benjamin Otte <otte@redhat.com> Date: Sat May 15 20:07:44 2010 +0200 gl: Move composite code into own file This allows designing a cleaner interface for cairo_composite_t as there will not be static functions that get called outside of the "published" interfaces. src/Makefile.sources | 3 +- src/cairo-gl-composite.c | 898 +++++++++++++++++++++++++++++++++++++++++++++++ src/cairo-gl-private.h | 8 + src/cairo-gl-surface.c | 851 +------------------------------------------- 4 files changed, 910 insertions(+), 850 deletions(-) commit 3ff32b00f0f85a853163c20b101ac5dd6a07da77 Author: Benjamin Otte <otte@redhat.com> Date: Sat May 15 18:20:56 2010 +0200 gl: cairo_gl_compsite_setup_t => cairo_gl_composite_t src/cairo-gl-glyphs.c | 4 ++-- src/cairo-gl-private.h | 26 +++++++++++++------------- src/cairo-gl-surface.c | 46 +++++++++++++++++++++++----------------------- 3 files changed, 38 insertions(+), 38 deletions(-) commit c16edee26fb9f1669ee0ce1a61f0c54d802c6c7b Author: Benjamin Otte <otte@redhat.com> Date: Sat May 15 17:33:24 2010 +0200 gl: use glBlendFuncSeparate() This way we can treat alpha-only textures correctly and don't have to emit different shaders for it. Also gets rid of GL_OPERAND_TEXTURE_ALPHA. src/cairo-gl-glyphs.c | 9 +++------ src/cairo-gl-private.h | 1 - src/cairo-gl-shaders.c | 37 ------------------------------------- src/cairo-gl-surface.c | 37 +++++++++++++------------------------ 4 files changed, 16 insertions(+), 68 deletions(-) commit 4f5221a834a883262cce607d247f607d6c200363 Author: Benjamin Otte <otte@redhat.com> Date: Wed May 12 21:12:37 2010 +0200 gl: acquire device manually in finish, as the default way will error We'll get CAIRO_STATUS_DEVICE_FINISHED, d'oh src/cairo-gl-device.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) commit 3f6300444928d12bf4a6e2f6f2c5fec1de7dba8e Author: Benjamin Otte <otte@redhat.com> Date: Wed May 12 20:53:20 2010 +0200 gl: Put shaders into a cache This allows to grow the complexity of the shaders without enlarging the static array to unreasonable sizes. The cache size of 64 is essentially random. src/cairo-gl-device.c | 2 + src/cairo-gl-private.h | 8 ++-- src/cairo-gl-shaders.c | 111 +++++++++++++++++++++++++++++++++++++++++-------- 3 files changed, 100 insertions(+), 21 deletions(-) commit 08b04caf52518f1672694341355b5f5418008e78 Author: Benjamin Otte <otte@redhat.com> Date: Wed May 12 18:10:04 2010 +0200 gl: Add cairo_gl_composite_setup_init/fini() functions Simplifies the main code paths, and can be extended to do dest copies. src/cairo-gl-private.h | 13 ++++++++ src/cairo-gl-surface.c | 91 ++++++++++++++++++++++++++++---------------------- 2 files changed, 65 insertions(+), 39 deletions(-) commit aa14df0db5e4f236a9668ceda4e148e7d3ba0e9f Author: Benjamin Otte <otte@redhat.com> Date: Wed May 12 17:01:47 2010 +0200 gl: move operator check component alpha compositing checks the operator itself src/cairo-gl-surface.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) commit a3ee0a7f113ed38df66580ff7e38a79759b0c933 Author: Benjamin Otte <otte@redhat.com> Date: Wed May 12 16:57:32 2010 +0200 gl: Programmatically generate fragment shaders src/cairo-gl-shaders.c | 428 +++++++++++++++++++++---------------------------- src/cairo-gl-surface.c | 6 +- 2 files changed, 186 insertions(+), 248 deletions(-) commit fe43b13052f7063f880aeeebda1880a6a2097922 Author: Benjamin Otte <otte@redhat.com> Date: Wed May 12 13:57:06 2010 +0200 gl: Automatically generate the vertex shader source The idea is being able to generate shaders from the given input on-demand. This allows creating more advanced shaders, such as those that include the destination (for self-painting and unsupported-by-blending operator) or painting images in weird formats (extra translation step in shader, think YUV). src/cairo-gl-device.c | 2 +- src/cairo-gl-private.h | 21 +++--- src/cairo-gl-shaders.c | 195 ++++++++++++++++++++++++++++++------------------- src/cairo-gl-surface.c | 3 +- 4 files changed, 132 insertions(+), 89 deletions(-) commit 405eee07ad9a5bc325b339b8588facb905fb28f5 Author: Benjamin Otte <otte@redhat.com> Date: Wed May 12 00:02:28 2010 +0200 gl: Get rid of cairo_gl_shader_source/mask_t Use the cairo_gl_operand_type_t instead. Those enums are pretty much identical. src/cairo-gl-glyphs.c | 13 +++-- src/cairo-gl-private.h | 67 ++++++++-------------- src/cairo-gl-shaders.c | 60 ++++++++++--------- src/cairo-gl-surface.c | 153 +++++++++++++++++++++++++++++-------------------- 4 files changed, 157 insertions(+), 136 deletions(-) commit 5149bb87d7b547766ce6ff4976217454adbab562 Author: Benjamin Otte <otte@redhat.com> Date: Tue May 11 21:35:13 2010 +0200 gl: s/_cairo_gl_set_destination/_cairo_gl_context_set_destination/ It's an operation on the context, so name it accordingly src/cairo-gl-device.c | 2 +- src/cairo-gl-glyphs.c | 2 +- src/cairo-gl-private.h | 2 +- src/cairo-gl-surface.c | 16 ++++++++-------- 4 files changed, 11 insertions(+), 11 deletions(-) commit 9f34e403f38192d711304fbde4dedbaa8e15fd6e Author: Benjamin Otte <otte@redhat.com> Date: Tue May 11 13:15:16 2010 +0200 gl: Share vertex shaders Previously, we created a new vertex shader for every shader program we used. Now the code shares identical vertex shaders between programs. src/cairo-gl-device.c | 17 ++++- src/cairo-gl-private.h | 27 ++++++-- src/cairo-gl-shaders.c | 179 ++++++++++++++++++++++++++++--------------------- src/cairo-gl-surface.c | 7 +- 4 files changed, 142 insertions(+), 88 deletions(-) commit 562c323ae8689907d7a62a7c2f5c10fb8d055608 Author: Adrian Johnson <ajohnson@redneon.com> Date: Sun May 16 17:47:13 2010 +0930 cff: Use correct glyph advance when subsetting cff fonts Previously the glyph advance in font units was used for the widths in the PDF font dictionary. This only works for cff fonts that use a [0.001 0 0 0.001 0 0] font matrix. src/cairo-cff-subset.c | 38 ++++++++++++++++++--------------- src/cairo-pdf-surface.c | 19 +++++++++-------- src/cairo-scaled-font-subsets-private.h | 6 +++--- 3 files changed, 34 insertions(+), 29 deletions(-) commit edcefa87ed0a8ff59b54ef9251182ce68f9158ba Author: Adrian Johnson <ajohnson@redneon.com> Date: Wed May 12 23:12:55 2010 +0930 type1: Use correct glyph advance when subsetting type 1 fonts Previously the glyph advance in font units was used for the widths in the PDF font dictionary. This only works for Type 1 fonts that use a [0.001 0 0 0.001 0 0] font matrix. https://bugs.freedesktop.org/show_bug.cgi?id=28061 src/cairo-pdf-surface.c | 23 ++++++++++++----------- src/cairo-scaled-font-subsets-private.h | 6 +++--- src/cairo-type1-fallback.c | 18 +++++++++--------- src/cairo-type1-subset.c | 6 +++--- 4 files changed, 27 insertions(+), 26 deletions(-) commit 34fd094b3be54138c20ea5c4aab1d9597d056f35 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun May 16 10:23:30 2010 +0100 rtree: Actually pin the node when pinning. How embarrassing. src/cairo-rtree-private.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 63062511f6755af9ddd2985250465f94ebc5a3e8 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun May 16 10:15:42 2010 +0100 boilerplate: Exercise manual double-buffering to a window Shouldn't possibly go wrong, but the mix of fbo and Window might prove interesting. boilerplate/cairo-boilerplate-gl.c | 117 +++++++++++++++++++++++++++++++++++++ 1 file changed, 117 insertions(+) commit 605be3182308ec7dfe15e9d89890c33800b1eea9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun May 16 10:11:02 2010 +0100 cairo: Missing private markup for _cairo_format_to_pixman_format src/cairoint.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 14d6c3e054da89e47fc82ea29a708290eefe1357 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun May 16 10:07:07 2010 +0100 configure: Include PTHREAD_CFLAGS in CAIRO_LDFLAGS Required for successful linking of our programs. I am not sure what the impact this has upon other users of Cairo yet, I think we need to export the PTHREAD_CFLAGS via cairo.pc as well. configure.ac | 1 + 1 file changed, 1 insertion(+) commit c3df7e6d58364f883d91567757ed999ea88e10ca Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun May 16 09:54:04 2010 +0100 rtree: Replace undeclared TRUE/FALSE with 1/0 src/cairo-rtree-private.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 465ceb330eada9937945f5c4d156169167d4bf8c Author: Andrea Canciani <ranma42@gmail.com> Date: Sun May 16 09:41:12 2010 +0200 test: improve copyright notices Conform copyright notices to a1-image-sample.c test/dash-offset.c | 48 ++++++++++++++++------------------------- test/linear-gradient-extend.c | 36 ++++++++++++++++--------------- test/radial-gradient-extend.c | 36 ++++++++++++++++--------------- test/surface-pattern-operator.c | 36 ++++++++++++++++--------------- 4 files changed, 76 insertions(+), 80 deletions(-) commit e8bc25fdb21340b0a5a25f7193fd0bc5de540a50 Author: Andrea Canciani <ranma42@valium.area51.sns.it> Date: Fri May 14 21:40:13 2010 +0200 test: add linear-gradient-extend and radial-gradient-extend Check that constant gradient to solid color optimization are only applied if the extents permit it. test/Makefile.sources | 2 + test/linear-gradient-extend.c | 90 ++++++++++++++++++++++++++++++++++++ test/linear-gradient-extend.ref.png | Bin 0 -> 371 bytes test/radial-gradient-extend.c | 90 ++++++++++++++++++++++++++++++++++++ test/radial-gradient-extend.ref.png | Bin 0 -> 466 bytes 5 files changed, 182 insertions(+) commit a9d821b7a0f582b3d1b0175adee3350c81061a94 Author: Benjamin Otte <otte@redhat.com> Date: Fri May 14 21:06:18 2010 +0200 gl: gl surfaces are only similar when they share the same device Fixes gl-surface-source test for gl target src/cairo-gl-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 435329a70f172c8cec72831dd7a2b2cfc0496342 Author: Richard Procter <richard.n.procter@gmail.com> Date: Fri May 14 14:48:46 2010 -0400 Bug 28100 - Font option CAIRO_HINT_METRICS_OFF unhonoured on win32 Fix typo. Ouch! src/cairo-win32-font.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 3a550eb2253588372b2c071b3c3152d5271c1aa2 Author: Benjamin Otte <otte@redhat.com> Date: Thu May 13 21:17:19 2010 +0200 gl: Remove gradient optimizations These optimizations are done in cairo-gstate.c these days. src/cairo-gl-surface.c | 31 ------------------------------- 1 file changed, 31 deletions(-) commit 3b1c0a4bd66660780095e6016e3db451f34503a3 Author: Benjamin Otte <otte@redhat.com> Date: Fri May 14 15:56:17 2010 +0200 fallback: Remove span renderer paths Those paths were broken, as they didn't properly translate the polygon to the destination size. And rather than adding lots of code that allows translation, it's easier to just delete this code. Note that the only user of the code was the GL backend anyway. src/cairo-surface-fallback.c | 67 -------------------------------------------- 1 file changed, 67 deletions(-) commit e3b2212bc5c4884eab356ae263b8318e10c379ac Author: Benjamin Otte <otte@redhat.com> Date: Thu May 13 15:14:29 2010 +0200 build: Fix pthread detection Incorporate the pthread detection code from the autoconf archive. Should fix build on obscure archs that pretend to be UNIXy - like Mac OS X Tiger. http://www.gnu.org/software/autoconf-archive/ http://www.gnu.org/software/autoconf-archive/ax_pthread.html build/ax-pthread.m4 | 283 ++++++++++++++++++++++++++++++++++++++++++ build/configure.ac.system | 11 -- configure.ac | 14 +-- util/cairo-sphinx/Makefile.am | 5 +- util/cairo-trace/Makefile.am | 2 +- 5 files changed, 292 insertions(+), 23 deletions(-) commit acfbf95a0c7620359fd917224b5b391702740190 Author: Benjamin Otte <otte@redhat.com> Date: Thu May 13 14:21:21 2010 +0200 test: Make global variables in new test static It seems there's no gcc flag that warns about this. test/zero-mask.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit c69143e3f9c95ab3937a7efb8e4de170c4edb648 Author: Benjamin Otte <otte@redhat.com> Date: Wed May 12 21:14:36 2010 +0200 gl: Don't keep a temporary surface in the device We get circular references that way. If it turns out later such a cache is critical, we'll need to invent a way to avoid circling references. For now, just pretend it's not important. src/cairo-gl-glyphs.c | 36 ++++++++---------------------------- src/cairo-gl-private.h | 1 - 2 files changed, 8 insertions(+), 29 deletions(-) commit ec1ca17b71b41159ea8754d42954863b62623536 Author: Benjamin Otte <otte@redhat.com> Date: Thu May 13 12:07:53 2010 +0200 gl: Don't create 0x0 surfaces We get a very annoyed libGL if we try to create 0x0 textures, so we return an image surface instead. Tested by the zero-mask test. src/cairo-gl-surface.c | 4 ++++ 1 file changed, 4 insertions(+) commit 108e3f5031fc16843197382af78af299b19ef290 Author: Benjamin Otte <otte@redhat.com> Date: Thu May 13 12:01:39 2010 +0200 Use a solid color for 0xN and Nx0 surfaces This gets around the need to check for 0-sized surfaces in the backends. src/cairo-gstate.c | 15 +++++++++++++++ 1 file changed, 15 insertions(+) commit 18c963824ebbb0e7054e682ecb5664a282a58552 Author: Benjamin Otte <otte@redhat.com> Date: Thu May 13 11:56:22 2010 +0200 Make _cairo_color_equal() work for unmultiplied colors, too Color stop colors are unmultiplied, and we use it there. src/cairo-color.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) commit 7bd8d0db1f4ac448c93d6d342726f64a0d6300e3 Author: Benjamin Otte <otte@redhat.com> Date: Thu May 13 11:23:33 2010 +0200 image: Fix wholly unbounded fixup Tested by zero-mask test. src/cairo-image-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 8aa50b5a7873b8317d0593105497607dbb2fae1d Author: Benjamin Otte <otte@redhat.com> Date: Thu May 13 10:27:37 2010 +0200 cairo_paint_with_alpha(cr, 0.0) is only a noop when bounded by mask src/cairo.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit ee32dfb3de035e73cb8c2073c0826daf244e3845 Author: Benjamin Otte <otte@redhat.com> Date: Thu May 13 10:04:49 2010 +0200 test: Add a test for various zero mask operations test/Makefile.sources | 3 +- test/zero-mask.c | 199 +++++++++++++++++++++++++++++++++++++++++++ test/zero-mask.ref.png | Bin 0 -> 402 bytes test/zero-mask.rgb24.ref.png | Bin 0 -> 382 bytes 4 files changed, 201 insertions(+), 1 deletion(-) commit 986ae4232067d4f33931d98fb6324871a1025633 Author: Owen W. Taylor <otaylor@fishsoup.net> Date: Thu Feb 25 15:18:22 2010 -0500 language bindings guide: clarify mapping for cairo_pattern_create_rgb[a] Clarify that cairo_pattern_create_rgb() and cairo_pattern_create_rgba() should be be overloaded as a single constructor for SolidPattern, but instead should be static methods. doc/public/language-bindings.xml | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) commit 18b48a6ebc044c07f08bafd0abcb68c595c5746c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu May 13 10:00:18 2010 +0100 Make cairo_color_stop_t a unique type. Hopefully reduce the occurrence of the confusion between the premultiplied shorts in cairo_color_t and the non-premultiplied shorts in cairo_color_stop_t. The existence of the two separate types is debatable and open for review. src/cairo-color.c | 4 ++-- src/cairo-gstate.c | 2 +- src/cairo-pattern.c | 2 +- src/cairo-pdf-surface.c | 9 --------- src/cairo-ps-surface.c | 32 ++------------------------------ src/cairo-svg-surface.c | 4 ++-- src/cairo-types-private.h | 17 ++++++++++++++++- src/cairoint.h | 4 ++-- 8 files changed, 26 insertions(+), 48 deletions(-) commit dfa2544f158930b002df74e676f3ef7aeeee1bd4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu May 13 09:52:39 2010 +0100 color: Special case comparing color stops. color stops are not premultiplied so we need to handle them carefully when comparing. The next step will be to make cairo_color_stop_t a unique type to prevent this mistake again. src/cairo-color.c | 30 ++++++++++++++++++++++++++++++ src/cairo-gl-surface.c | 4 ++-- src/cairo-gstate.c | 4 ++-- src/cairo-pattern.c | 6 +++--- src/cairoint.h | 4 ++++ src/drm/cairo-drm-intel.c | 2 +- 6 files changed, 42 insertions(+), 8 deletions(-) commit 4741d33cf67409a6fddbc32a5d0c8d68224cd939 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu May 13 09:31:10 2010 +0100 gstate: Don't reduce solid source and mask for SOURCE and CLEAR Not all operators are created equal, and this optimization breaks for SOURCE and CLEAR where the mask operates on the clip and not the source. src/cairo-gstate.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit ee871940ffe229ed285e618566e45c7d28c5857c Author: Andrea Canciani <ranma42@gmail.com> Date: Thu May 13 01:54:39 2010 +0200 Revert bd672d080ca0036ef1b88d26583e6af36f541461 on src/cairo-composite-rectangles.c The extent of the operation can be larger than the source if the operator is not bounded by the source. Fixes clip-unbounded, operator-source, surface-pattern-operator src/cairo-composite-rectangles.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) commit 0870c6fb5b39dcc04fa376123848adde2d06d2ce Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Apr 15 20:31:24 2010 +0100 gcc-4.5 warnings and optimisation flags. build/configure.ac.warnings | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit cf0933a05aedb66b5d8999c84b935349b25dd974 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Apr 12 10:42:14 2010 +0100 gallium: Refresh. Catch up with changes in APIs, still no substance and the integration with winsys handles needs review. boilerplate/Makefile.win32.features | 8 - build/Makefile.win32.features-h | 1 - src/Makefile.win32.features | 8 - src/drm/cairo-drm-gallium-surface.c | 514 ++++++++++++++++++++++-------------- 4 files changed, 317 insertions(+), 214 deletions(-) commit bd672d080ca0036ef1b88d26583e6af36f541461 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Mar 26 19:37:08 2010 +0000 drm: code dump, sorry. Lots upon lots of tiny fixes mixed in with experimental code. :( boilerplate/Makefile.win32.features | 18 + boilerplate/cairo-boilerplate-drm.c | 12 +- build/Makefile.win32.features | 1 + build/Makefile.win32.features-h | 4 + build/configure.ac.features | 1 + configure.ac | 13 + src/Makefile.sources | 7 + src/Makefile.win32.features | 22 + src/cairo-boxes-private.h | 4 + src/cairo-boxes.c | 31 + src/cairo-composite-rectangles.c | 7 +- src/cairo-drm-xr.h | 66 + src/cairo-drm.h | 4 +- src/cairo-fixed-private.h | 18 +- src/cairo-freelist-private.h | 1 - src/cairo-freelist.c | 7 +- src/cairo-image-surface.c | 2 +- src/cairo-list-private.h | 31 +- src/cairoint.h | 7 + src/drm/cairo-drm-i915-glyphs.c | 82 +- src/drm/cairo-drm-i915-private.h | 153 +- src/drm/cairo-drm-i915-shader.c | 866 ++++++----- src/drm/cairo-drm-i915-spans.c | 146 +- src/drm/cairo-drm-i915-surface.c | 1884 ++++++++++++++++++------ src/drm/cairo-drm-i965-glyphs.c | 36 +- src/drm/cairo-drm-i965-private.h | 2 +- src/drm/cairo-drm-i965-shader.c | 11 +- src/drm/cairo-drm-i965-spans.c | 16 +- src/drm/cairo-drm-i965-surface.c | 146 +- src/drm/cairo-drm-intel-debug.c | 1 + src/drm/cairo-drm-intel-ioctl-private.h | 25 + src/drm/cairo-drm-intel-private.h | 92 +- src/drm/cairo-drm-intel-surface.c | 90 +- src/drm/cairo-drm-intel.c | 530 ++++--- src/drm/cairo-drm-private.h | 10 +- src/drm/cairo-drm-radeon-surface.c | 60 +- src/drm/cairo-drm-surface.c | 37 +- src/drm/cairo-drm-xr.c | 2377 +++++++++++++++++++++++++++++++ 38 files changed, 5416 insertions(+), 1404 deletions(-) commit c8fba4960339797521afbf8068fcbe62d064f5dc Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Apr 3 15:38:04 2010 +0100 rtree: Remove the unused evict hook. src/cairo-gl-glyphs.c | 3 +-- src/cairo-rtree-private.h | 4 +--- src/cairo-rtree.c | 13 +------------ src/drm/cairo-drm-i915-surface.c | 3 +-- src/drm/cairo-drm-intel.c | 2 +- 5 files changed, 5 insertions(+), 20 deletions(-) commit 2a0726337368462046ef84d9be4cf59734b39806 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Apr 3 15:30:18 2010 +0100 rtree: defer propagating pinned nodes until eviction. Only during infrequent eviction do we require absolute knowledge of which graph of nodes are in use and thus pinned. So during the common use of querying the glyph cache, we just mark the leaf as used. Then we need to make space for a new glyph, we move the parents of the left nodes from the evictable list to the pinned list. src/cairo-rtree-private.h | 12 ++++++++++-- src/cairo-rtree.c | 33 +++++++++++---------------------- 2 files changed, 21 insertions(+), 24 deletions(-) commit cbe8fd0794adaccdf3eb15eef780a030e3d51784 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Apr 3 14:56:25 2010 +0100 script: Avoid the expensive of the redundant memset. As we are about to immediately fill the entire image, allocate the memory ourselves to avoid the redundant memset performed by pixman. util/cairo-script/cairo-script-operators.c | 39 ++++++++++++++++++++---------- 1 file changed, 26 insertions(+), 13 deletions(-) commit b9f7a4b5261b6d4e7bdbb5cc56d78d50ad1bd4a7 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Apr 3 14:48:36 2010 +0100 script: Don't hash the entire image. The reuse hit rate is very small, and most images are quickly distinguished in the first few bytes... Though perhaps not for video as in the swfdec-youtube case... util/cairo-script/cairo-script-operators.c | 35 ++++++++++++++++++++---------- 1 file changed, 24 insertions(+), 11 deletions(-) commit a85c6c1e36273d534f01ade0714b0592b6bed5ab Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Apr 3 14:09:56 2010 +0100 gstate: reduce a couple more operators src/cairo-gstate.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-) commit a3cb52e4037db9e565d32f85d69c364b606555e9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Apr 3 13:36:18 2010 +0100 simplify pattern extents for translation matrices src/cairo-pattern.c | 22 ++++++++++++++-------- 1 file changed, 14 insertions(+), 8 deletions(-) commit d45c7dc62d6b036bbbe85c3c1951807e055930d9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Apr 1 17:09:11 2010 +0100 xcb: discard glyph mask and use dst directly when possible. boilerplate/cairo-boilerplate-xcb.c | 12 ++++-- src/cairo-gstate.c | 18 ++++++-- src/cairo-pattern.c | 2 +- src/cairo-xcb-surface-render.c | 85 +++++++++++++++++++++++++++++-------- 4 files changed, 92 insertions(+), 25 deletions(-) commit a8b611ea8eb90f190ee3a9e7999e08e2d865ac43 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Mar 27 08:23:10 2010 +0000 clip: Fill instead of creating intermediate surfaces. src/cairo-clip.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) commit fb53d05a69d0291fe68dfe8669b5c1f7636ede1a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed May 12 20:06:34 2010 +0100 test: Remove pthread_yield() I accidentally pushed this debugging left-over from when I was tracking down a race in memfault. Remove it so that the test suite compiles on more minimal pthread setups. test/pthread-same-source.c | 2 -- test/pthread-show-text.c | 2 -- test/pthread-similar.c | 2 -- 3 files changed, 6 deletions(-) commit cdd98dba6a62360071aebb6dbbb6be2ab6be368f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed May 12 19:35:15 2010 +0100 win32-printing: compile fix. Missing arg for _cairo_recording_surface_replay_region(). src/cairo-win32-printing-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 35307fc66f649cc042ec07b7b79277d7ee5987f3 Author: Jonathan Kew <jfkthame@gmail.com> Date: Tue May 11 15:12:01 2010 -0400 win32: Grow glyph extents to account for ClearType Add extra pixel on each side of the glyph's black box returned by GetGlyphOutlineW, to avoid clipping ClearType pixels. From https://bugzilla.mozilla.org/show_bug.cgi?id=445087 src/cairo-win32-font.c | 13 +++++++++++++ 1 file changed, 13 insertions(+) commit d65e8064c0bd3f2d5ad200493bfefb89aefbba55 Author: Robert O'Callahan <robert@ocallahan.org> Date: Tue May 11 13:59:58 2010 -0400 Ensure the quartz backend returns the first stop for negative positions on the gradient line of a nonrepeating linear gradient. I discovered a small bug in cairo-quartz gradients. If you have multiple stops at position 0, then cairo-quartz pads with the *last* stop at position 0, instead of the first stop at position 0. This patch fixes that. From https://bugzilla.mozilla.org/show_bug.cgi?id=513395 src/cairo-quartz-surface.c | 24 ++++++++++++++++++++++-- 1 file changed, 22 insertions(+), 2 deletions(-) commit ce27db9a558cbf4065b18aed55cc706fc0a6d128 Author: Robert O'Callahan <robert@ocallahan.org> Date: Tue May 11 13:59:52 2010 -0400 Don't repeat a Quartz gradient more times than necessary, to avoid Quartz quality problems when there are lots of repeated color stops. -- Add a parameter to _cairo_quartz_setup_source so we can pass down the extents of the object we're drawing -- Compute fill/stroke/glyph extents and pass them down in the cases we need to (repeating/reflecting gradients) -- Pass those extents on down to where we set up the gradients -- Make _cairo_quartz_setup_linear_source fall back to pixman for the degenerate case where the linear gradient vector has no length -- In CreateRepeatingRadialGradientFunction and CreateRepeatingLinearGradientFunction, use the object extents (or surface extents, for the paint() case) instead of the clip box to calculate the parameters for the gradient -- I've changed the way CreateRepeatingLinearGradientFunction calculates the repetition count. The new approach gives much more precise bounds on the number of repetitions needed (and is very similar to what we do for radial gradients). This is important because if we specify a much larger input range than we really need for our gradient color function, Quartz samples it too coarsely over the range we actually care about, and the gradients look bad. For example, suppose start = (5,0), end = (6,10), the CTM is identity and the bounds we want to cover is (0,0)-(10,10). I think the current approach sets up the gradient to be repeated 10 times. In fact only 3 repetitions are needed. Also, using 'width' here didn't look right: - y_rep_end = (int) ceil((surface->extents.width - MAX(mstart.y, mend.y)) / dy From https://bugzilla.mozilla.org/show_bug.cgi?id=508730 src/cairo-quartz-surface.c | 261 +++++++++++++++++++++++++++++++-------------- 1 file changed, 179 insertions(+), 82 deletions(-) commit 7b4e63d5b8bda8e9ea1f6455d779070f64f07f6f Author: Robert O'Callahan <robert@ocallahan.org> Date: Tue May 11 13:59:43 2010 -0400 Clean up Quartz gradient code by moving some local variables to static const globals. From https://bugzilla.mozilla.org/show_bug.cgi?id=508730 src/cairo-quartz-surface.c | 31 +++++++++++++------------------ 1 file changed, 13 insertions(+), 18 deletions(-) commit 8302952dcff20a1d2de194152ace810c7056f994 Author: Robert O'Callahan <robert@ocallahan.org> Date: Tue May 11 13:58:10 2010 -0400 quartz: Don't fallback to pixman for repeating radial gradients. Figuring out where the outer circle should move to is tricky. I hope the algebra in there is understandable. This is a nice performance improvement, probably because we avoid painting the gradient over the entire clipBox (which is usually the entire surface). I tried to write reftests that compared a repeating radial gradient to a non-repeating gradient with manually repeated stops, but it didn't work because the rasterization was slightly different --- I'm not sure why. This patch also forces us to use pixman for all degenerate cases where the circles intersect. This at least makes us consistent across platforms. From https://bugzilla.mozilla.org/show_bug.cgi?id=508227 src/cairo-quartz-surface.c | 203 ++++++++++++++++++++++++++++++++++++++++----- 1 file changed, 180 insertions(+), 23 deletions(-) commit 1bda2334b32394a821e6286fbc76617e68da3895 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue May 11 13:51:38 2010 +0100 trace: And for the bonus round... Make it compile. util/cairo-trace/trace.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit b23f3bd20450aae764f9f9362c560b00bf68636b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue May 11 13:54:08 2010 +0100 trace: Try wrapping FT_Open_Face() We have to be careful to disambuigate an internal call by FreeType from the application. util/cairo-trace/trace.c | 38 ++++++++++++++++++++++++++------------ 1 file changed, 26 insertions(+), 12 deletions(-) commit ee446910aa1db0ca0b54791a2ef3bb24b277c9a4 Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Mon May 10 15:06:25 2010 -0400 Add an additional set parantheses around the container_of macro The non-__GNUC__ version of the container_of macro would do wrong when used like container_of(...)-> because '->' binds tighter than the cast in the container_of macro. Adding an additional set of parantheses fixes this. src/cairoint.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit e540d040bddc717f17e0e2510cffc0bc3cb41ccd Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon May 10 13:36:53 2010 +0100 xcb: trivial memfault fixes. The first fixes required to kick-start memfault testing of the xcb backend. boilerplate/cairo-boilerplate-xcb.c | 22 +++++++++++--- src/cairo-xcb-connection-render.c | 35 +++++++++++++++++------ src/cairo-xcb-connection.c | 57 ++++++++++++++++++++++++++----------- src/cairo-xcb-private.h | 2 +- src/cairo-xcb-screen.c | 15 +++++----- src/cairo-xcb-surface-core.c | 1 - src/cairo-xcb-surface-render.c | 32 +++++++++++++++++---- src/cairo-xcb-surface.c | 5 ++-- src/cairo-xlib-xcb-surface.c | 1 - 9 files changed, 123 insertions(+), 47 deletions(-) commit 8b486db9a9d74b40df296382eb1833bc40ae791a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon May 10 12:45:31 2010 +0100 region: _cairo_region_create_in_error() Avoid leaks when reporting memfault associated with constructing regions. src/cairo-path-fill.c | 7 ++---- src/cairo-region-private.h | 3 +++ src/cairo-region.c | 63 ++++++++++++++++++++++++++++++++++++++++------ 3 files changed, 60 insertions(+), 13 deletions(-) commit 9d863cd3942c3086c24d67305f7a5892604d0eeb Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat May 8 18:20:11 2010 +0100 xcb: Find the correct xcb_screen_t for faking Screen When choosing the xcb_screen_t to use for the xlib-xcb backing surface, it helps if it matches the screen used to generate similar surfaces and snapshots - otherwise we end up pulling the image back from the XServer every time we want to use the Picture. src/cairo-xlib-xcb-surface.c | 23 +++++++++++++++++++++-- 1 file changed, 21 insertions(+), 2 deletions(-) commit 448d3571088463fc61641badcdfdc8c0002ae12a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat May 8 09:59:54 2010 +0100 xcb: Upload images in place. A common operation is to store an image in a similar surface, so construct a fast path to avoid the allocation and double-blit through a temporary pixmap. src/cairo-xcb-private.h | 5 ++ src/cairo-xcb-surface-render.c | 169 ++++++++++++++++++++++++++++++++++++++++- src/cairo-xcb-surface.c | 31 ++++---- 3 files changed, 189 insertions(+), 16 deletions(-) commit e48cbd3b47a6e4f7c1f66b3085df41546460e477 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat May 8 09:56:46 2010 +0100 xcb: Reset picture after failed snapshot. Clear the local picture variable if we cannot use the snapshot so that we are forced to create a new and valid picture. src/cairo-xcb-surface-render.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 4e3c19833ef8631c1f1cd54870c0a86d88252886 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri May 7 21:30:13 2010 +0100 test: Force cairo-test-suite to return SUCCESS Set the CAIRO_TEST_FORCE_PASS environment variable to run through the test suite and ignore errors. Useful for forcing distcheck to continue past a broken test suite. test/cairo-test-runner.c | 10 ++++++++++ 1 file changed, 10 insertions(+) commit ad8abc01105f02a05497969b6b5ec2c8742daeb2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri May 7 21:28:58 2010 +0100 subsurface: Don't double apply device offset for deep subsurfaces. If we have a subsurface of a subsurface then the device offset has already been applied to the extents that we use to offset the new subsurface. src/cairo-surface-subsurface.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) commit d2a250ad33dcd1a3960b07d2f0270266a3f3f106 Author: Benjamin Otte <otte@redhat.com> Date: Fri May 7 20:36:33 2010 +0200 gl: Make the shader implementation per-context, not global src/cairo-gl-device.c | 8 +--- src/cairo-gl-glyphs.c | 4 +- src/cairo-gl-private.h | 37 +++++++++++------ src/cairo-gl-shaders.c | 105 ++++++++++++++++++++++++++----------------------- src/cairo-gl-surface.c | 89 +++++++++++++++++++++-------------------- 5 files changed, 131 insertions(+), 112 deletions(-) commit 4571055c46e82cbb6d3ab47860d63e400b338238 Author: Benjamin Otte <otte@redhat.com> Date: Wed May 5 19:36:46 2010 +0200 gl: Print GL errors when releasing the device Don't scatter calls to error printing around everywhere, instead do it in the one place where it matters. Also, convert the functions to macros, so we can use __FILE__ and __LINE__ when printing a warning src/cairo-gl-private.h | 15 ++++++++++----- src/cairo-gl-surface.c | 22 +--------------------- 2 files changed, 11 insertions(+), 26 deletions(-) commit 59b31aeed9b3ec2a1ac6da188ef5e86d20626662 Author: Benjamin Otte <otte@redhat.com> Date: Sat May 8 15:09:39 2010 +0200 gl: Make check_span_renderer() call not require a device lock Note that we didn't lock the device previously, so the function was broken. src/cairo-gl-device.c | 1 + src/cairo-gl-private.h | 7 +++++++ src/cairo-gl-surface.c | 5 +++-- 3 files changed, 11 insertions(+), 2 deletions(-) commit 81f4dd65a32efae645b826b84e8382f7bf7a9b2d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri May 7 22:35:18 2010 +0100 cairo: Special case cairo_t with NULL_POINTER Avoid allocation for the potential user error of attempting to use cairo_create(NULL). src/cairo.c | 24 +++++++++++++++++++++++- 1 file changed, 23 insertions(+), 1 deletion(-) commit a61570a55e70040ffcf8ff3cb2c7943e71a5e2a0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri May 7 22:18:52 2010 +0100 test/copy-path: memfault status checks. test/copy-path.c | 61 +++++++++++++++++++++++++++++++++++--------------------- 1 file changed, 38 insertions(+), 23 deletions(-) commit e6180d1d5e29a91f1bfc20956bb017ab74fb8b8e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri May 7 22:18:10 2010 +0100 surface-fallback: Only destroy the clip after it has been initialized. More memfault detected error path errors. src/cairo-surface-fallback.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) commit 01a208f7e3785254dc101f09a73943ec7cfb1d6e Author: timeless <timeless@bemail.org> Date: Fri May 7 10:24:49 2010 -0400 Remove useless null checks of a and b in _cairo_path_fixed_equal 'a' and 'b' are dereferenced before being checked for null. From https://bugzilla.mozilla.org/show_bug.cgi?id=555706 src/cairo-path-fixed.c | 22 ++++++++-------------- 1 file changed, 8 insertions(+), 14 deletions(-) commit 0068d9b5cd09a7aa0c5a2ab4da97ca181934ff56 Author: Jonathan Kew <jfkthame@gmail.com> Date: Fri May 7 10:16:32 2010 -0400 quartz: Work around bad glyph extents returned by CoreGraphics for empty glyphs in Al Bayan font This resolves the problem by checking for the crazy glyph bbox result from CoreGraphics, and replacing it with an empty rect. From https://bugzilla.mozilla.org/show_bug.cgi?id=534260 src/cairo-quartz-font.c | 10 ++++++++++ 1 file changed, 10 insertions(+) commit c5d9643417690380c102bf0b6ad195cb264d3feb Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri May 7 11:24:28 2010 +0100 doc: regen doc/public/tmpl/cairo-font-face.sgml | 3 +++ doc/public/tmpl/cairo-font-options.sgml | 4 ++++ doc/public/tmpl/cairo-ft.sgml | 3 +++ doc/public/tmpl/cairo-image.sgml | 5 +++++ doc/public/tmpl/cairo-matrix.sgml | 3 +++ doc/public/tmpl/cairo-paths.sgml | 3 +++ doc/public/tmpl/cairo-pattern.sgml | 3 +++ doc/public/tmpl/cairo-pdf.sgml | 3 +++ doc/public/tmpl/cairo-png.sgml | 3 +++ doc/public/tmpl/cairo-ps.sgml | 3 +++ doc/public/tmpl/cairo-quartz-fonts.sgml | 3 +++ doc/public/tmpl/cairo-quartz.sgml | 3 +++ doc/public/tmpl/cairo-scaled-font.sgml | 3 +++ doc/public/tmpl/cairo-status.sgml | 6 ++++++ doc/public/tmpl/cairo-surface.sgml | 9 +++++++++ doc/public/tmpl/cairo-svg.sgml | 3 +++ doc/public/tmpl/cairo-text.sgml | 3 +++ doc/public/tmpl/cairo-transforms.sgml | 3 +++ doc/public/tmpl/cairo-types.sgml | 3 +++ doc/public/tmpl/cairo-user-fonts.sgml | 4 ++++ doc/public/tmpl/cairo-version.sgml | 5 +++++ doc/public/tmpl/cairo-win32-fonts.sgml | 3 +++ doc/public/tmpl/cairo-win32.sgml | 3 +++ doc/public/tmpl/cairo-xlib-xrender.sgml | 3 +++ doc/public/tmpl/cairo-xlib.sgml | 3 +++ doc/public/tmpl/cairo.sgml | 3 +++ 26 files changed, 93 insertions(+) commit b9e9ff59348412a1f1f2c57d6185c84f77874c26 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri May 7 11:21:50 2010 +0100 subsurface: Include device offset in extents Adjust the subsurface extents so that the user specifies the extents in world space rather than device space. src/cairo-surface-subsurface.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) commit 5e3b28ffeac3fc7239c7ada6992e4794a27b0818 Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Thu May 6 22:27:09 2010 -0400 Remove comma from the end of the CAIRO_PATTERN_ACQUIRE enum src/cairoint.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit cc8442514ab15f08708ecb26f499c764a7d171f4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu May 6 21:50:16 2010 +0100 gl: Enable fixed-function without glsl. Through a very simple mistake we were triggering fallbacks after detecting we were not able to perform the composite using GLSL -- we were just returning the UNSUPPORTED status instead of enabling the fixed-function path. src/cairo-gl-surface.c | 56 ++++++++++++++++++++++++++++---------------------- 1 file changed, 31 insertions(+), 25 deletions(-) commit 30d358e098353af24f16cdb7efe41f10f276f559 Author: Metal Sonic <kidlinux96@gmail.com> Date: Thu May 6 16:07:43 2010 -0400 Update license blocks to use "Mozilla Foundation" instead of "Mozilla Corporation" From https://bugzilla.mozilla.org/show_bug.cgi?id=507387 src/cairo-fixed-private.h | 2 +- src/cairo-fixed-type-private.h | 2 +- src/cairo-malloc-private.h | 2 +- src/cairo-quartz-font.c | 2 +- src/cairo-quartz-image-surface.c | 2 +- src/cairo-quartz-image.h | 2 +- src/cairo-quartz-surface.c | 2 +- src/cairo-quartz.h | 2 +- 8 files changed, 8 insertions(+), 8 deletions(-) commit 506636e19edcdb656c94f61c915c9d49829dade8 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu May 6 17:21:59 2010 +0100 cairo: stash a number of contexts for use with NO_MUTEX The implementation is the same as the atomic one (bar the use of atomic primitives to manipulate the occupancy!). Patch based on the original by Jeff Muizelaar. src/cairo.c | 33 ++++++++++++++++++++++++++++++++- 1 file changed, 32 insertions(+), 1 deletion(-) commit 91dfee420c3e9d85e5cd4f2cec6dc708c0e471a2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu May 6 17:03:56 2010 +0100 device: Reorder finalization so that user data is destroyed last. As the user is likely to attach the underlying resources to the device for automatic collection upon finalization, it is important that the user data is then destroy last (so that those resources are still available in the surface and device cleanup routines). src/cairo-device.c | 9 +++++++-- src/cairo-surface.c | 4 ++-- 2 files changed, 9 insertions(+), 4 deletions(-) commit 2658d7ef5f8f7e06929f4b1cae64e5312db24ec4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu May 6 17:02:39 2010 +0100 test/gl-surface-source: Attach GLXContext to device user data. As we actually use the GLXContext to create the device, we only want to free those resources upon the final unreference of the device (and not the initial surface). test/gl-surface-source.c | 26 +++++++++++--------------- 1 file changed, 11 insertions(+), 15 deletions(-) commit f5167dc2e1a13d8c4e5d66d7178a24b9b5e7ac7a Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Thu May 6 11:43:52 2010 -0400 Add an implementation of ffs() for MSVC src/cairo-compiler-private.h | 15 +++++++++++++++ 1 file changed, 15 insertions(+) commit a2a2bd62ff37e2642f671ca7ebf26954e1856abe Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu May 6 14:39:30 2010 +0100 surface-fallback: Handle memfaults during clipping. Don't be lazy, propagate the error rather than asserting. src/cairo-surface-fallback.c | 26 +++++++++++++++----------- 1 file changed, 15 insertions(+), 11 deletions(-) commit 2e3acee410ff127d4557ca8191625338cd225313 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu May 6 14:38:55 2010 +0100 test: Add gl-surface-source Exercise using GL sources. test/Makefile.am | 7 ++ test/Makefile.sources | 3 + test/gl-surface-source.argb32.ref.png | Bin 0 -> 377 bytes test/gl-surface-source.c | 115 +++++++++++++++++++++++++++++++++ test/gl-surface-source.image16.ref.png | Bin 0 -> 305 bytes test/gl-surface-source.rgb24.ref.png | Bin 0 -> 301 bytes 6 files changed, 125 insertions(+) commit 75d8550891b8b2c09200aaaa812c6c9332161cd7 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu May 6 13:50:27 2010 +0100 test/clip-contexts: Free secondary context test/clip-contexts.c | 2 ++ 1 file changed, 2 insertions(+) commit bf83bf00a75343f7d1dcd15dc4eeec1b1a4b9428 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu May 6 10:53:48 2010 +0100 test: Add subsurface-image-repeat Test handling of source clipping with an image - principally to check handling of mime types and zero-copy image snapshotting in the vector surfaces and ensure that they work with subsurface patterns as well. test/Makefile.am | 2 + test/Makefile.sources | 1 + test/subsurface-image-repeat.c | 70 +++++++++++++++++++++++++++ test/subsurface-image-repeat.image16.ref.png | Bin 0 -> 776 bytes test/subsurface-image-repeat.ref.png | Bin 0 -> 915 bytes 5 files changed, 73 insertions(+) commit 88986fdbef795c61cd60231046663d3cd80b4947 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu May 6 10:46:49 2010 +0100 xlib: Remove reference counting for cairo_xlib_screen_t The screen is owned by the cairo_xlib_display_t device, so we can simplify and close the refleak by removing the surplus reference counting. src/cairo-xlib-display.c | 84 +++++++++++------------------------------------- src/cairo-xlib-private.h | 22 +++---------- src/cairo-xlib-screen.c | 47 ++++++--------------------- src/cairo-xlib-surface.c | 38 +++++++--------------- src/cairo-xlib-visual.c | 2 +- 5 files changed, 46 insertions(+), 147 deletions(-) commit 632fabc77d89254e2e6915148fa870f23c2a5722 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu May 6 10:10:38 2010 +0100 xlib: Convert visual array to a cairo_list_t It's simpler and more memory efficient. src/cairo-xlib-private.h | 4 ++- src/cairo-xlib-screen.c | 76 +++++++++++++++--------------------------------- 2 files changed, 26 insertions(+), 54 deletions(-) commit 20d75bdadc897be238ceb4d2d02f079e469144e4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu May 6 10:03:36 2010 +0100 xlib: Remove screen->has_render Use the info from the display instead. src/cairo-xlib-private.h | 1 - src/cairo-xlib-screen.c | 17 ++++------------- 2 files changed, 4 insertions(+), 14 deletions(-) commit 9e0b716ecf1b27d74e63f25a055d8120d8d53309 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu May 6 09:58:46 2010 +0100 xlib: Remove obsolete comment, there is no race anymore The display lock is now held for the duration of the screen creation, so there is no race. Remove the false comment. src/cairo-xlib-screen.c | 4 ---- 1 file changed, 4 deletions(-) commit 62346c500a7421e30b3054fdcc07fc78a18da96e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu May 6 09:37:02 2010 +0100 xlib: Ensure the global display mutex is initialized before use. src/cairo-xlib-display.c | 2 ++ 1 file changed, 2 insertions(+) commit fb6caf08ae5ccf8f8c2c251d32fd98aacbcebdcc Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu May 6 09:30:41 2010 +0100 image: Propagate error instead of asserting. Apparently we can get to this point without evaluating the clip surface, so return the error status rather than assert. src/cairo-image-surface.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit 5e95c62bbb0cb28ee22af8f97267ce2da72cf47a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu May 6 09:09:22 2010 +0100 clip: Propagate failure from retrieving the previous clip surface. src/cairo-clip.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit a6ee7aea8ac11a9e5a9b58e17e881db9666a2b24 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed May 5 10:52:39 2010 +0100 test: Add xlib-expose-event.image16.ref.png test/Makefile.am | 1 + test/xlib-expose-event.image16.ref.png | Bin 0 -> 30332 bytes 2 files changed, 1 insertion(+) commit ded2dc8faaa3a48c84a31422a31c64595a6eaf8a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed May 5 14:44:55 2010 +0100 scaled-font: Pluck last glyph from the most recent page. In converting to cairo_list_t, it helps to preserve the semantics of appending the page to the end of the list where it is expected to be plucked in the event of an allocation failure. src/cairo-scaled-font.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit fa937913e06bc295750538be45aa83eb42332fb4 Author: Adrian Johnson <ajohnson@redneon.com> Date: Wed May 5 21:35:11 2010 +0930 PS: Add missing 'q' when resetting clip path src/cairo-ps-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 9c594d8b405eba09b07c0a438f5311f5c5e40313 Author: Benjamin Otte <otte@redhat.com> Date: Wed May 5 12:28:59 2010 +0200 gl: Get rid of ctx variable The code used renderer->ctx and ctx interchangably, that was confusing. src/cairo-gl-surface.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) commit a354f1f92a17378f42d0b703482ae2b8168c7f68 Author: Benjamin Otte <otte@redhat.com> Date: Wed May 5 12:28:24 2010 +0200 gl: Another case of proper device acquiring This patch isn't strictly necessary, but it cleans up the code. src/cairo-gl-surface.c | 15 +++++++++------ 1 file changed, 9 insertions(+), 6 deletions(-) commit 36210ee51444979271f7ba1cc0ac452cd30df0af Author: Benjamin Otte <otte@redhat.com> Date: Wed May 5 12:24:43 2010 +0200 gl: Get rid of another cast to cairo_gl_context_t src/cairo-gl-surface.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) commit 58b03691d379979b50d74740748b8b8523e0cd0d Author: Benjamin Otte <otte@redhat.com> Date: Wed May 5 12:21:23 2010 +0200 gl: Introduce _cairo_gl_device_has_glsl() function The function takes a cairo_device_t argument as it can be called on an unacquired device. This is the first patch in a process to try to rid cairo-gl-surface.c of casts from cairo_device_t to cairo_gl_context_t. As its members should usually only be accessed when the device is locked, the only way the cairo_gl_context_t should be acquired is using _cairo_gl_context_acquire(). src/cairo-gl-glyphs.c | 2 +- src/cairo-gl-private.h | 6 ++++++ src/cairo-gl-surface.c | 9 ++++----- 3 files changed, 11 insertions(+), 6 deletions(-) commit 5223b654230d3074d0bc31eb74c415e50eced365 Author: Benjamin Otte <otte@redhat.com> Date: Wed May 5 12:07:49 2010 +0200 gl: Get rid of cairo_gl_surface_glfinish() from public API This is now done explicitly in the boilerplate code. It was not a useful public API to begin with. boilerplate/cairo-boilerplate-gl.c | 17 +++++++---------- src/cairo-gl-surface.c | 9 --------- src/cairo-gl.h | 3 --- 3 files changed, 7 insertions(+), 22 deletions(-) commit e071fa2c4f62b5ead39c6c6401f39360ea4dee1f Author: Benjamin Otte <otte@redhat.com> Date: Wed May 5 12:03:13 2010 +0200 gl: Remove unused member variable src/cairo-gl-private.h | 1 - 1 file changed, 1 deletion(-) commit 90c64f60d853f1a6ea5febd362ef3528cc9d914b Author: Benjamin Otte <otte@redhat.com> Date: Wed May 5 12:01:59 2010 +0200 gl: Move device-specific code into cairo-gl-device.c src/Makefile.sources | 6 +- src/cairo-gl-device.c | 205 +++++++++++++++++++++++++++++++++++++++++++++++++ src/cairo-gl-surface.c | 162 -------------------------------------- 3 files changed, 210 insertions(+), 163 deletions(-) commit 0ff5a18fd5623b57c7493e6b67a6b1c75af82fb3 Author: Benjamin Otte <otte@redhat.com> Date: Wed May 5 11:52:40 2010 +0200 gl: Make _gl_set_destination() take the context as an argument Keeping with the API introduced for Xlib, functions that require an acquired context take this context as the first argument. src/cairo-gl-glyphs.c | 2 +- src/cairo-gl-private.h | 2 +- src/cairo-gl-surface.c | 20 +++++++++----------- 3 files changed, 11 insertions(+), 13 deletions(-) commit e8c5b6b1c2099fc6a49d0a082f5e884d3fbd89df Author: Benjamin Otte <otte@redhat.com> Date: Wed May 5 10:05:30 2010 +0200 Fix typo in docs src/cairo-array.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 53508e6e3047b4dac6d69d5d3ff2208178998fd4 Author: Benjamin Otte <otte@redhat.com> Date: Tue May 4 19:28:45 2010 +0200 gl: Acquire the context before destroying it. src/cairo-gl-surface.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) commit c46aec489722bf09cd10c52d70bb4975eb1546d9 Author: Benjamin Otte <otte@redhat.com> Date: Tue May 4 19:28:22 2010 +0200 gl: acquire/release context during surface_finish() src/cairo-gl-surface.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) commit 88801aabb34328746a53d3d59d036dada7ced756 Author: Benjamin Otte <otte@redhat.com> Date: Tue May 4 19:23:08 2010 +0200 gl: Acquire context when initing/destroying gl operands src/cairo-gl-glyphs.c | 17 ++++++----- src/cairo-gl-private.h | 3 +- src/cairo-gl-surface.c | 80 ++++++++++++++++++++++++++------------------------ 3 files changed, 54 insertions(+), 46 deletions(-) commit c6d7d054ca44f3849ee1ab76de9032a4eaa210fe Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed May 5 10:39:49 2010 +0100 test: Add radial-gradient-mask Exercise using radial gradients as a mask. test/Makefile.am | 6 ++ test/Makefile.sources | 2 + test/radial-gradient-mask-source.argb32.ref.png | Bin 0 -> 116311 bytes test/radial-gradient-mask-source.c | 111 +++++++++++++++++++++++ test/radial-gradient-mask-source.image16.ref.png | Bin 0 -> 82800 bytes test/radial-gradient-mask-source.rgb24.ref.png | Bin 0 -> 129093 bytes test/radial-gradient-mask.argb32.ref.png | Bin 0 -> 116311 bytes test/radial-gradient-mask.c | 110 ++++++++++++++++++++++ test/radial-gradient-mask.image16.ref.png | Bin 0 -> 82800 bytes test/radial-gradient-mask.rgb24.ref.png | Bin 0 -> 129093 bytes 10 files changed, 229 insertions(+) commit 5f878f917980cde145d4c385b0c0b66336c0f454 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed May 5 10:24:14 2010 +0100 clip: Trivial compiler warning cairo-clip.c: In function ‘_cairo_clip_path_reapply_clip_path_translate’: cairo-clip.c:446: warning: suggest parentheses around assignment used as truth value src/cairo-clip.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit 840dad765febf7b807bbb75e401c159a8c5c4dfe Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed May 5 09:46:03 2010 +0100 test/mask-glyphs: Check for memfault test/mask-glyphs.c | 32 +++++++++++++++++++++++--------- 1 file changed, 23 insertions(+), 9 deletions(-) commit ad541a1ec62e18bcaf74994affbdb0503e277f9c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed May 5 09:33:36 2010 +0100 clip: Propagate memfault from translating clip region src/cairo-clip.c | 6 ++++++ 1 file changed, 6 insertions(+) commit 4bc54643f5cdeddf2145d28c317b55ca6b5949a5 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue May 4 21:36:03 2010 +0100 image: Propagate failure from pixman_image_set_clip_region() src/cairo-image-surface.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) commit 9ef33a1d8c0476f4f1eeba4f26523fdeb6f7d809 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue May 4 21:35:51 2010 +0100 test/bitmap-font: Destroy FcPattern test/bitmap-font.c | 8 ++------ 1 file changed, 2 insertions(+), 6 deletions(-) commit fba4cf1d94b3ab676e0b62ae3574bf1ee4e69de0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed May 5 09:22:03 2010 +0100 recording: Mark an empty recording surface as clear. src/cairo-recording-surface.c | 6 ++++++ 1 file changed, 6 insertions(+) commit e5329805394c94c915e43a0040b51d5019b4a90e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed May 5 09:17:51 2010 +0100 recording: Avoid refcycles by always copying the command array. Short-term solution to avoid the refleaks and to make the test suite happy. A more elegant solution would be to track the references and avoid the substantial memory overhead of copying the recording surfaces. Thanks to Benjamin Otte for pointing out the solution to avoiding refcycles. src/cairo-recording-surface-private.h | 1 - src/cairo-recording-surface.c | 23 ++++++++++++----------- 2 files changed, 12 insertions(+), 12 deletions(-) commit 6a06e0ef7b534355d0f3b4885159ffc37dbc3867 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed May 5 09:17:33 2010 +0100 test/subsurface-modify-parent: Free region. Minor refleak. test/subsurface-modify-parent.c | 1 + 1 file changed, 1 insertion(+) commit 4a46c715564c886e7434d65625fb84b5d87c0970 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue May 4 18:51:30 2010 +0100 type3: Minor tweaks to remove a redundant check. src/cairo-type3-glyph-surface.c | 8 ++------ 1 file changed, 2 insertions(+), 6 deletions(-) commit 11531b90a2317bb1113b187d6aa9c5083f5ecaa1 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue May 4 18:50:58 2010 +0100 ps: Force finish of the type3 glyph surfaces. src/cairo-ps-surface.c | 2 ++ 1 file changed, 2 insertions(+) commit 5155ebeaa25815124124db2599336ab2ddbe7413 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue May 4 18:50:27 2010 +0100 pdf: Assert that no other error can occur for an invertible matrix. src/cairo-pdf-operators.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) commit d9259bd06822c871a4c653f395cfc1f933caaeb7 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue May 4 18:49:55 2010 +0100 paginated: Mark as clear initially. src/cairo-paginated-surface.c | 2 ++ 1 file changed, 2 insertions(+) commit bc14e0e953630d47a65e701a377631e406eb080c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon May 3 14:18:33 2010 +0100 scaled-font: Convert to cairo_list_t Convert the open-coded doubly-linked list of glyph pages for a font into the common cairo_list_t. src/cairo-scaled-font-private.h | 2 +- src/cairo-scaled-font.c | 46 ++++++++++++++++++++--------------------- 2 files changed, 23 insertions(+), 25 deletions(-) commit 627527693c38c328fdb79058f5299609c0031029 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon May 3 13:41:41 2010 +0100 ps: Remove the fill stroke command This is not supported by PostScript, so eliminate it from the prologue. src/cairo-ps-surface.c | 2 -- 1 file changed, 2 deletions(-) commit 4dd92ef2bc877eb197f096c3eb8fe74cb2e7b099 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun May 2 14:35:59 2010 +0100 test: Add subsurface-pad Round out the subsurface tests with PAD. test/Makefile.am | 1 + test/Makefile.sources | 1 + test/subsurface-pad.c | 76 ++++++++++++++++++++++++++++++++++++++++++++ test/subsurface-pad.ref.png | Bin 0 -> 181 bytes 4 files changed, 78 insertions(+) commit 85e3cffde164a5af444639dfd182fde2448809ee Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Apr 30 21:09:01 2010 +0100 test: Add subsurface-modify-{child,parent} Add tests to ensure that modifications to the subsurface via the original surface are tracked by the subsurface, i.e. that any snapshots are detached upon modification of the parent as well as modification of itself and vice versa. test/Makefile.am | 2 + test/Makefile.sources | 2 + test/subsurface-modify-child.c | 98 ++++++++++++++++++++++++++++++++++ test/subsurface-modify-child.ref.png | Bin 0 -> 197 bytes test/subsurface-modify-parent.c | 77 ++++++++++++++++++++++++++ test/subsurface-modify-parent.ref.png | Bin 0 -> 197 bytes 6 files changed, 179 insertions(+) commit 19ac81f50ba353400934bf112523b4257ad9de2f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue May 4 18:25:23 2010 +0100 scaled-font: Remove MRU cache upon detection of error. src/cairo-scaled-font.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-) commit 6f2d4f5b2aba9344fee9375614dfd0c6236bfd7c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue May 4 18:09:11 2010 +0100 image: Propagate failure from pixman_image_fill_boxes(). src/cairo-image-surface.c | 55 ++++++++++++++++++++++++++--------------------- 1 file changed, 30 insertions(+), 25 deletions(-) commit ded7be0b9ce12f8d4a84c8c9dd622d92fe347a34 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue May 4 17:01:35 2010 +0100 test/user-font-rescale: Use after free and check for memfault. test/user-font-rescale.c | 16 +++++++++++----- 1 file changed, 11 insertions(+), 5 deletions(-) commit b0052c52153376400a30dca1d67f4aca9735cc09 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue May 4 17:01:17 2010 +0100 test/bitmap-font: Propagate memfault test/bitmap-font.c | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-) commit 9187c7532755dda8d2677ec792162c71029fb37d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue May 4 13:18:09 2010 +0100 test/bitmap-font: Check for memfault error test/bitmap-font.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 0fced9279227614ede6964e94ede9de49a3a3388 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue May 4 13:17:38 2010 +0100 test/user-font-proxy: Check we install user data on the scaled font test/user-font-proxy.c | 14 ++++++++++---- 1 file changed, 10 insertions(+), 4 deletions(-) commit 38188961b996c97d344195749b7bec8c93f9bad7 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue May 4 13:17:07 2010 +0100 image: Check for pixman image allocation failure during glyphs src/cairo-image-surface.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) commit 4a181ebac416ec35e412ddbe9ed1392fb8b74ab7 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue May 4 10:26:30 2010 +0100 pdf: Cleanup local reference to source on error paths. src/cairo-pdf-surface.c | 2 ++ 1 file changed, 2 insertions(+) commit ba48edcd86ee1bca06b25587e56abcd2c1b19d79 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue May 4 10:26:13 2010 +0100 ps: Always cleanup used font subsets. src/cairo-ps-surface.c | 19 ++++++++----------- 1 file changed, 8 insertions(+), 11 deletions(-) commit 9a0958be012e097c239260fac0dec0b1e5679028 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue May 4 10:25:48 2010 +0100 test: Cleanup after failed pthread_join(). test/pthread-same-source.c | 17 +++++++++-------- test/pthread-show-text.c | 17 +++++++++-------- test/pthread-similar.c | 17 +++++++++-------- 3 files changed, 27 insertions(+), 24 deletions(-) commit 08b9984da9f2119cd4ed03bbd656a349b49d1f12 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue May 4 08:19:41 2010 +0100 image: Replace assert with error return It is possible to encounter an allocation failure here since we may have tweaked the clip since the last evaluation of the clip region. src/cairo-image-surface.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit 2c68fcace4ad47ad09e9699e45afb514be1381f9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon May 3 20:39:34 2010 +0100 test: Embed the list pointer into cairo_test_t Reduce the amount of output printed by leak-detectors when a test crashes. test/cairo-test-runner.c | 31 +++++++++++++++---------------- test/cairo-test.h | 7 ++++--- 2 files changed, 19 insertions(+), 19 deletions(-) commit 7002b824729885a9d579f677ecc665b620847480 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue May 4 13:52:09 2010 +0200 xlib: Don't delete gradients in unbounded fixup code The fix is ugly, but for now it works. src/cairo-xlib-surface.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit af26560f258d93cc78782ddd0208128756874c11 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon May 3 19:21:59 2010 +0100 test: Improve memfault behaviour. Various minor tweaks to convert asserts into error returns and to improve error checking on intermediate surfaces. boilerplate/cairo-boilerplate-svg.c | 2 +- test/cairo-test.c | 21 ++++++++++++--- test/png.c | 34 +++++++++++++++++-------- test/pthread-same-source.c | 7 +++++ test/surface-source.c | 11 ++++++-- test/toy-font-face.c | 16 +++++++++--- test/user-data.c | 51 ++++++++++++++++++++++++++----------- test/xlib-surface-source.c | 7 ++++- 8 files changed, 112 insertions(+), 37 deletions(-) commit c549203c8d69474be4362037f702e4fb59c9929e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon May 3 19:21:18 2010 +0100 scaled-font: Check for an error return when retrieving the implementation src/cairo-scaled-font.c | 8 ++++++++ 1 file changed, 8 insertions(+) commit c93e6f014d9678b1aea34fd7a30a1fc2f51c6347 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon May 3 19:20:42 2010 +0100 paginated: propagate malloc failures more cleanly. src/cairo-paginated-surface.c | 26 +++++++++++++------------- 1 file changed, 13 insertions(+), 13 deletions(-) commit 8e9fd9c01732c3102c27c7dee50f6e494ba7cdd8 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon May 3 19:20:07 2010 +0100 image: A few missing tests for malloc failure src/cairo-image-surface.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) commit 87781ffbd914bca29b4d744fb48678ab06a07108 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon May 3 19:18:49 2010 +0100 ft: Convert an assert into an unlikely error return. The assert depends upon good behaviour from fontconfig, which is no guaranteed under memfault, so return an error instead. src/cairo-ft-font.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) commit 5672b7a18a34456862977a1cf678bf6575dcc3f4 Author: Benjamin Otte <otte@redhat.com> Date: Mon May 3 13:13:27 2010 +0200 fallback: get src_x/y variables right when compositing traps src/cairo-image-surface.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) commit c10a5a9fb0463f254fb607a2260b540256f888cc Author: Dave Yeo <daveryeo@telus.net> Date: Mon May 3 10:20:51 2010 +0200 os2: Fix get_extents() segfaulting. bed2701e1c89095878d549cbca8f22d84f3dda3c removed one line too much. src/cairo-os2-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 9df81fe4d3f0ccc5632d0e84e920fe1c448bd531 Author: Benjamin Otte <otte@redhat.com> Date: Sun May 2 19:59:26 2010 +0200 boilerplate: don't compile pdf test code when pdf is not tested Avoids gcc warnings when libpoppler version isn't good enough. boilerplate/cairo-boilerplate-pdf.c | 3 +++ 1 file changed, 3 insertions(+) commit 358f57c8765f823f8e6db7629ea6ca48fcd392b8 Author: Benjamin Otte <otte@redhat.com> Date: Fri Apr 30 23:29:24 2010 +0200 image: remove useless optimization The optimization is done way smarter inside _cairo_matrix_transform_bounding_box() src/cairo-image-surface.c | 11 +++-------- 1 file changed, 3 insertions(+), 8 deletions(-) commit 94aa6d24b6d5ab4528a0b0a9b1b0c6eb7218f861 Author: Benjamin Otte <otte@redhat.com> Date: Fri Apr 30 23:24:41 2010 +0200 fallback: Create pixman image for correct rectangle This was most prominently shown by the a1-image-sample test, but multiple tests exposed the problem with the xlib-fallback code. src/cairo-image-surface.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) commit 41fecf8e662c9ad2fe063754db8302f349a223ef Author: Carlos Garcia Campos <carlosgc@gnome.org> Date: Fri Apr 30 21:18:52 2010 +0200 [configure] Bump poppler dependency to 0.13.3 configure.ac | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 3294daf423e88b55aa13841dfa7c557ee9833f92 Author: Benjamin Otte <otte@redhat.com> Date: Fri Apr 30 20:14:54 2010 +0200 Create cleared surface instead of a scratch surface in clone_similar() A cleared surface is needed when playing back a recording surface that has semi-transparent operations. As this only affects surface types where scratch surfaces aren't already cleared (ie it affected xlib but not image), the effect was only visible for those backends. This should not cause any performance regressions as the other backends set surface->clear = TRUE when creating the scratch surface, so the clear operation is effectively a no-op. Covered by the existing pdf-surface-source and svg-surface-source tests. src/cairo-surface.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) commit 0aeb612d2f822e8cda1f509b6a4f2995e24053fc Author: Carlos Garcia Campos <carlosgc@gnome.org> Date: Fri Apr 30 16:18:57 2010 +0200 [test] Update pdf reference image for text-rotate test/text-rotate.pdf.ref.png | Bin 16757 -> 16744 bytes 1 file changed, 0 insertions(+), 0 deletions(-) commit fe2844464721644b837816aa4dad4ead508f0b89 Author: Carlos Garcia Campos <carlosgc@gnome.org> Date: Fri Apr 30 16:16:23 2010 +0200 [test] Update pdf reference image for surface-pattern-scale-down test/surface-pattern-scale-down.pdf.ref.png | Bin 1603 -> 2189 bytes 1 file changed, 0 insertions(+), 0 deletions(-) commit dba6f23318005a4cdbe7566519ce896278ea184a Author: Carlos Garcia Campos <carlosgc@gnome.org> Date: Fri Apr 30 16:11:52 2010 +0200 [test] Update pdf reference image for leaky-dashed-rectangle test/leaky-dashed-rectangle.pdf.ref.png | Bin 380 -> 377 bytes 1 file changed, 0 insertions(+), 0 deletions(-) commit 6b81187ac4870311a10c024af9f30c366be785cd Author: Carlos Garcia Campos <carlosgc@gnome.org> Date: Fri Apr 30 16:05:21 2010 +0200 [test] Update pdf reference images for clip-operator test/clip-operator.pdf.argb32.ref.png | Bin 9266 -> 9437 bytes test/clip-operator.pdf.rgb24.ref.png | Bin 5166 -> 5391 bytes 2 files changed, 0 insertions(+), 0 deletions(-) commit b485ae5b90bf72cde0c30c2350e03d1369ca6dcd Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Apr 30 14:14:01 2010 +0100 atomic: Add Andrea's copyright notice Note Andrea's copyright for his contribution of platform agnostic fallbacks and the implementation for MacOS/X. src/cairo-atomic-private.h | 2 ++ 1 file changed, 2 insertions(+) commit 4ee5119a3467ea086efccef2907b0c576d600b79 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Apr 30 13:00:05 2010 +0100 test: Expand pthread-show-text to cover all surfaces. In a similar fashion to pthread-same-source and pthread-similar, check that the texting handling is thread-safe for all surface and font backends. test/Makefile.am | 3 + test/pthread-show-text.c | 118 +++++++++++++++++++++------------ test/pthread-show-text.image16.ref.png | Bin 0 -> 22167 bytes test/pthread-show-text.ps.ref.png | Bin 0 -> 16558 bytes test/pthread-show-text.ref.png | Bin 0 -> 29885 bytes 5 files changed, 77 insertions(+), 44 deletions(-) commit c6dc8ad7dc46d03899cd37fff40bd7f3a60339e2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Apr 30 12:38:25 2010 +0100 Revert "paginated: Call surface finish explicitly on recording surface" This reverts commit 5fc04bba9fa8ddda8cf7d7a97015f21a21429172. Whilst this fixes the self-referential reference leak, it however introduces use-after-finish into normal behaviour. Close, but not quite. src/cairo-paginated-surface.c | 1 - 1 file changed, 1 deletion(-) commit b972669c9ac87d1c2ca922e12024ef5da394aa6b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Apr 30 12:37:16 2010 +0100 boilerplate: Fix use after free from 3ae5723 If we want to access the surface during the cleanup, we must hold our own reference to it. Make it so for the forced finish of the recording surface. boilerplate/cairo-boilerplate.c | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-) commit 905c345df1f828551af2ab0b751388428e4dfaf6 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Apr 30 10:31:50 2010 +0100 recording: Check that we do not use a finished recording surface. User paranoia - as the replay is exposed to the user it is convenient to add the finish check in the core replay function so that we gain a paranoia check for the internal paths as well. src/cairo-recording-surface.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) commit 5fe7c5842f3589efce765b3869c1dd55367a5aba Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Apr 30 10:13:09 2010 +0100 pdf; Emit subsurface patterns natively. Encode subsurface patterns into the PDF stream natively (when possible), similar to the cairo-ps backend. src/cairo-pdf-surface.c | 113 ++++++++++++++++++++++++++++++++++++++++-------- 1 file changed, 94 insertions(+), 19 deletions(-) commit 8ded35fd69ed0bbb07359c0278071a25af66571c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Apr 28 14:26:21 2010 +0100 ps: Enable native encoding of subsurface patterns. Carefully handle subsurfaces of a recording surface through the analysis and paginated surfaces so that we can generate a native pattern for the vector backends, demonstrated by the PostScript backend. Nothing remarkable, just a lot of bookkeeping to track the wrapped surface types and to apply the correct offsets when generating the subsurface pattern. src/cairo-analysis-surface.c | 13 +- src/cairo-paginated-surface.c | 1 + src/cairo-pdf-surface.c | 4 +- src/cairo-ps-surface.c | 121 +++++++++++-- src/cairo-recording-surface-private.h | 1 + src/cairo-recording-surface.c | 12 +- src/cairo-surface-private.h | 4 +- src/cairo-surface-subsurface-private.h | 4 +- src/cairo-surface-subsurface.c | 101 +++++++---- src/cairo-surface-wrapper-private.h | 7 +- src/cairo-surface-wrapper.c | 314 +++++++++++++++++++++++++++------ src/cairo-surface.c | 13 +- test/Makefile.am | 1 + test/Makefile.sources | 1 + test/subsurface-reflect.c | 76 ++++++++ test/subsurface-reflect.ref.png | Bin 0 -> 210 bytes test/subsurface-repeat.c | 8 +- test/subsurface-similar-repeat.c | 3 + test/subsurface.c | 7 +- test/subsurface.image16.ref.png | Bin 1430 -> 1643 bytes test/subsurface.ref.png | Bin 1597 -> 1811 bytes 21 files changed, 565 insertions(+), 126 deletions(-) commit 5fc04bba9fa8ddda8cf7d7a97015f21a21429172 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Apr 29 21:51:29 2010 +0100 paginated: Call surface finish explicitly on recording surface During show_page() when we destroy the recording surface after emitting the page, ensure that we actually call cairo_surface_finish() to dispose of any self-referential reference leaks. src/cairo-paginated-surface.c | 1 + 1 file changed, 1 insertion(+) commit 3ae57234644c3756785c551beffe584f837b0273 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Apr 29 21:50:22 2010 +0100 boilerplate: Ensure that the recording surfaces are finished. Be paranoid and explicitly call finish to cleanup self-referential leaks when using paginated/recording surfaces. boilerplate/cairo-boilerplate-pdf.c | 5 ++++- boilerplate/cairo-boilerplate-ps.c | 5 ++++- boilerplate/cairo-boilerplate-svg.c | 5 ++++- boilerplate/cairo-boilerplate.c | 10 +++++----- 4 files changed, 17 insertions(+), 8 deletions(-) commit f08cc311af1248b39c3e757ef192515ed8506862 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Apr 29 20:34:56 2010 +0100 boilerplate: Destroy the redundant image reference When using a script surface to record the recording surface, we replace the local reference to the image surface. boilerplate/cairo-boilerplate.c | 2 ++ 1 file changed, 2 insertions(+) commit ab2776c9a16134c50b48fd202263421ec0f466e7 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Apr 29 18:50:32 2010 +0100 snapshot: Attach the backend generated snapshot to the target Cache the result of snapshotting using the backend vfunc in the normal manner by attaching the snapshot to the target. This should reduce resource usage in these cases. src/cairo-surface-snapshot.c | 21 ++++++++++++++++++++- 1 file changed, 20 insertions(+), 1 deletion(-) commit 07122f37d11eabe62bc9c81ccbf71bbe8b7a1005 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Apr 29 15:19:18 2010 +0100 surface: Convert snapshots from an array to a double-linked list. Saves the memory allocation for the array, and the overhead of maintaining the area for both insertions and more importantly deletes. src/cairo-recording-surface.c | 6 +- src/cairo-surface-private.h | 5 +- src/cairo-surface-snapshot.c | 10 +-- src/cairo-surface.c | 130 ++++++++++++++----------------------- src/cairo-vg-surface.c | 8 +-- src/cairo-xcb-surface-core.c | 6 +- src/cairo-xcb-surface-render.c | 10 +-- src/cairo-xcb-surface.c | 12 ++-- src/cairoint.h | 2 +- src/drm/cairo-drm-i915-shader.c | 11 ++-- src/drm/cairo-drm-i965-shader.c | 11 ++-- src/drm/cairo-drm-intel-surface.c | 8 +-- src/drm/cairo-drm-radeon-surface.c | 8 +-- 13 files changed, 77 insertions(+), 150 deletions(-) commit 4cb733c28551f4a34cd4a225b8d797a55bf9b977 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Apr 29 12:04:44 2010 +0100 image: Simplify translation of pattern matrix for composite_trapezoids() src/cairo-image-surface.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) commit 3940b0e91c274de0cf2fca4b34d4025b92965c19 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Apr 29 17:29:13 2010 +0100 subsurface: s/region/rectangle/ After a renewed discussion, it was pointed out that the API in Cairo was not restrictive and by using doubles we would be consisted with the rest of the API. Thus prompting the name change to cairo_surface_create_for_rectangle() similar to cairo_rectangle(). And document the public API. src/cairo-surface-subsurface.c | 39 ++++++++++++++++++++++++------ src/cairo.h | 10 ++++---- test/subsurface-repeat.c | 4 +-- test/subsurface-similar-repeat.c | 2 +- test/subsurface.c | 4 +-- util/cairo-script/cairo-script-operators.c | 12 ++++----- util/cairo-trace/trace.c | 10 ++++---- 7 files changed, 53 insertions(+), 28 deletions(-) commit 4b6fa8d5a78731aca80f3c8c6320e7edc845c5b5 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Apr 30 09:47:51 2010 +0100 test: Remove group-unaligned.*new.png Entirely misleading, looks like PDF is behaving the same as image and this remnant was thus giving the wrong impression. test/Makefile.am | 1 - test/group-unaligned.pdf.new.png | Bin 427 -> 0 bytes 2 files changed, 1 deletion(-) commit e3f990d84131f7bd1e64e54d02bac780b1e9f85e Author: Benjamin Otte <otte@redhat.com> Date: Thu Apr 29 22:34:35 2010 +0200 test: Update reference images for last checkin Both tests now pas on all backends I tested: test, image, xlib and recording test/a1-traps-sample.ref.png | Bin 148 -> 120 bytes test/unantialiased-shapes.ref.png | Bin 3969 -> 3968 bytes 2 files changed, 0 insertions(+), 0 deletions(-) commit 6d36f06751377459e35afe1ac26c59d33b16c730 Author: Benjamin Otte <otte@redhat.com> Date: Thu Apr 29 22:12:02 2010 +0200 image: Round down when rendering antialiased boxes Matches Pixman's output when going via pixman_rasterize_trapezoid() src/cairo-fixed-private.h | 6 ++++++ src/cairo-image-surface.c | 9 +++++---- 2 files changed, 11 insertions(+), 4 deletions(-) commit 95179a5de91b5c6f99dbccc3b6c950df58d4f2a7 Author: Benjamin Otte <otte@redhat.com> Date: Thu Apr 29 20:45:23 2010 +0200 Remove excessive semicolons s/;;/;/ basically src/cairo-image-surface.c | 2 +- src/cairo-wideint.c | 2 +- src/drm/cairo-drm-i965-shader.c | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) commit 5d07307b691afccccbb15f773d5231669ba44f5a Author: Benjamin Otte <otte@redhat.com> Date: Thu Apr 29 18:20:59 2010 +0200 xlib: Don't modify variables that are needed later In the XCopyArea region code, don't modify src_x/y when they are later used in the unbounded fixup code. Exposed by composite-integer-translate-source test. src/cairo-xlib-surface.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) commit 90160deb2a84f787dff46243471e37600b24799b Author: Benjamin Otte <otte@redhat.com> Date: Thu Apr 29 18:20:44 2010 +0200 xlib: Constify some function arguments src/cairo-xlib-surface.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) commit 4da71d1a01c4d4cc84b5e31cc89dd7f297b23359 Author: Benjamin Otte <otte@redhat.com> Date: Thu Apr 29 18:11:54 2010 +0200 xlib: remove unused code src/cairo-xlib-surface.c | 12 ------------ 1 file changed, 12 deletions(-) commit 758ffadcb1d0a0478d43ca40775687a37e5920a4 Author: Benjamin Otte <otte@redhat.com> Date: Thu Apr 29 16:54:09 2010 +0200 fallback: Only do _fill/compsoite_rectangles on bounded source For unbounded sources, these optimizations fail to take into account the regions outside the source. src/cairo-surface-fallback.c | 16 +++++++++------- 1 file changed, 9 insertions(+), 7 deletions(-) commit 6288ad719da898c4a75cdc6527f5a41d4c900e57 Author: Benjamin Otte <otte@redhat.com> Date: Thu Apr 29 15:37:32 2010 +0200 fallback: translate extents regions properly The extents clip regions that were created for unbounded operators were not translated properly. This has been changed now. This is a followup to 9ce8bef9d6e6d773dd1f4b184916ed5c96c3541d. src/cairo-surface-fallback.c | 17 ++++++++++------- 1 file changed, 10 insertions(+), 7 deletions(-) commit b0760826f30f63637561b353e7eed1913036da2d Author: Benjamin Otte <otte@redhat.com> Date: Thu Apr 29 12:11:48 2010 +0200 test: Set have_result for image matches Causes the log to contain information about the reference imagery used. test/cairo-test.c | 16 +++++++++++++--- 1 file changed, 13 insertions(+), 3 deletions(-) commit 8f85c2d77cc2ca7984f9771c0096fbad798dbde8 Author: Benjamin Otte <otte@redhat.com> Date: Thu Apr 29 12:09:31 2010 +0200 image: translate source pattern in composite_trapezoids() Fixes xlib fallback paths as exposed by the clip-push-group test. src/cairo-image-surface.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) commit aa495eac5a2db9b396eb1953eb112c733d0ababc Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Apr 29 10:04:18 2010 +0100 atomic: Tweak for compilation x86. Missing definition of _cairo_atomic_ptr_get() used in the fallbacks. src/cairo-atomic-private.h | 27 ++++++++++++++++++++------- src/cairo-atomic.c | 16 ++++++++-------- 2 files changed, 28 insertions(+), 15 deletions(-) commit 248af38b3efa3f96225eea43f4ba5b94baff34a6 Author: Andrea Canciani <ranma42@gmail.com> Date: Wed Apr 28 17:23:47 2010 +0200 atomic: Add MacOSX atomic implementation Enable atomic operation on MacOS X, using the functions provided by libkern/OSAtomic.h Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> build/aclocal.cairo.m4 | 10 ++++++++++ src/cairo-atomic-private.h | 30 ++++++++++++++++++++++++++++++ 2 files changed, 40 insertions(+) commit 56a367a1626b2b7ec3d9e64e74f016867f294a34 Author: Andrea Canciani <ranma42@gmail.com> Date: Wed Apr 28 19:17:10 2010 +0200 atomic: Correct implementation of _atomic_fetch () _atomic_fetch() was expected to replace the content of a slot with NULL and return the old content, but it incorrectly returned the previous content even if it was unable to perform the exchange (because of conflicts with other threads accessing the pool). Fix suggested by Chris Wilson <chris@chris-wilson.co.uk> src/cairo-freed-pool-private.h | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) commit d66b1ca1cfda64088521581bc2bd494bff9f5d23 Author: Andrea Canciani <ranma42@gmail.com> Date: Wed Apr 28 16:23:15 2010 +0200 atomic: Separate bool and old-value compare-and-exchange Some implementations only offer one version of compare-and-exchange, thus we expose both through cairo-atomic, implementing what is missing through appropriate fallbacks. *_cmpxchg() now return a boolean (this unbreaks _cairo_atomic_uint_cmpxchg) *_cmpxchg_return_old() return the old value Code is updated everywhere to reflect this, by using *_cmpxchg() wherever the returned value was only tested to check if the exchange had really taken place. Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> src/cairo-atomic-private.h | 68 ++++++++++++++++++++++++++++++++++++++---- src/cairo-atomic.c | 4 +-- src/cairo-freed-pool-private.h | 4 +-- src/cairo-image-surface.c | 6 ++-- src/cairo.c | 4 +-- src/drm/cairo-drm.c | 2 +- 6 files changed, 72 insertions(+), 16 deletions(-) commit a0bf424b820b404947671b56f357a2cef71640fb Author: Andrea Canciani <ranma42@gmail.com> Date: Wed Apr 28 15:29:32 2010 +0200 atomic: Remove unused function _cairo_atomic_int_set() _cairo_atomic_int_set() was only used in the definition of CAIRO_REFERENCE_SET_VALUE, which was never used. Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> src/cairo-atomic-private.h | 10 ---------- src/cairo-atomic.c | 8 -------- src/cairo-reference-count-private.h | 1 - 3 files changed, 19 deletions(-) commit 9ce8bef9d6e6d773dd1f4b184916ed5c96c3541d Author: Benjamin Otte <otte@redhat.com> Date: Wed Apr 28 18:34:20 2010 +0200 fallback: Propagate extents properly Otherwise unbounded operators will clear the full surface. Improves the score for the unbounded-operator test, in particular the output for the test-fallback case. src/cairo-surface-fallback.c | 112 ++++++++++++++++++++++++++++++------------- 1 file changed, 78 insertions(+), 34 deletions(-) commit 91fd97ae7c4de23c61d481c98c9352f0639d76dd Author: Benjamin Otte <otte@redhat.com> Date: Wed Apr 28 20:21:37 2010 +0200 fallback: Pass the correct extents for unbounbded operations src/cairo-surface-fallback.c | 20 +++++++++++++------- 1 file changed, 13 insertions(+), 7 deletions(-) commit 06e9caf86199e8261a07db6d4774628fa147728d Author: Benjamin Otte <otte@redhat.com> Date: Wed Apr 28 20:05:13 2010 +0200 image: pixman_image_fill_rectangles() => pixman_image_fill_boxes() src/cairo-image-surface.c | 32 ++++++++++++++++---------------- 1 file changed, 16 insertions(+), 16 deletions(-) commit bc49df322770b3bd1797c0e153b97f1f296fbd1e Author: Benjamin Otte <otte@redhat.com> Date: Wed Apr 28 19:56:36 2010 +0200 pixman_image_composite => pixman_image_composite32 Fix up the remaining callers src/cairo-gl-surface.c | 12 ++++++------ src/cairo-image-surface.c | 12 ++++++------ 2 files changed, 12 insertions(+), 12 deletions(-) commit 393da364a7f26e696141c58d4fb6fdefb2ea245a Author: Benjamin Otte <otte@redhat.com> Date: Wed Apr 28 18:18:15 2010 +0200 fallback: Sanitize code that queries surface extents The previous code was setting extents.is_bounded, but that value has a completely different meaning. src/cairo-surface-fallback.c | 20 ++++++++++---------- 1 file changed, 10 insertions(+), 10 deletions(-) commit 13914039567d5e8269fc29c51018ce172a40f2e9 Author: Benjamin Otte <otte@redhat.com> Date: Wed Apr 28 13:05:28 2010 +0200 test: run xlib-expose-event test unconditionally test/Makefile.sources | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 2ce1afa3222145e9c4f8c74a0034e0da9e93b70b Author: Benjamin Otte <otte@redhat.com> Date: Wed Apr 28 12:54:54 2010 +0200 test-suite: add image_diff_is_failure() function This cleans the code and fixes a boolean logic error where this check was done manually. test/buffer-diff.c | 8 ++++++++ test/buffer-diff.h | 4 ++++ test/cairo-test-trace.c | 3 +-- test/cairo-test.c | 9 +++------ test/fallback-resolution.c | 3 +-- test/xlib-surface.c | 4 ++-- 6 files changed, 19 insertions(+), 12 deletions(-) commit 2a91d425088cb8a93de76f2f91f32a7f23f0aecd Author: Benjamin Otte <otte@redhat.com> Date: Wed Apr 28 12:26:29 2010 +0200 fallback: Fix clip_region handling in mask creation Fixes the xlib-expose-event test. src/cairo-surface-fallback.c | 16 +++++++++++++++- 1 file changed, 15 insertions(+), 1 deletion(-) commit bb853916f3581af13cc1abff185ef5e0218f9f1e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Apr 28 11:37:59 2010 +0100 image: Generate clear/white/black images when !HAS_ATOMIC_OPS Andrea reported that b74cc0f broke compilation without atomic ops. src/cairo-image-surface.c | 10 ++++++++++ 1 file changed, 10 insertions(+) commit 45d4c6e0b2fe62a3c6f1fdb7f359c9771111b470 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Apr 28 10:01:06 2010 +0100 script: Replace the missing " " after pop I accidentally deleted the whitespace after popping the recording surface, thus causing the scripts to be broken. src/cairo-script-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 6a99e83c50d715ba1e47aa5c7be714a423568e57 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Apr 28 09:55:20 2010 +0100 subsurface: Mark the image as !clear after copying. src/cairo-surface-subsurface.c | 22 ++++++++++++---------- 1 file changed, 12 insertions(+), 10 deletions(-) commit ca3df75e8f876991f2dc9e85c9daa3fd96e826d2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Apr 28 09:54:56 2010 +0100 script: Reconstruct subsurfaces. src/cairo-script-surface.c | 84 ++++++++++++++++++++++++++++++++-------------- 1 file changed, 58 insertions(+), 26 deletions(-) commit 0f0d349a400e097856e6d9863e66c0a041e65c35 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Apr 28 09:54:37 2010 +0100 trace: Wrap cairo_surface_create_for_region() util/cairo-script/cairo-script-operators.c | 33 +++++++++++++++++++++++++++ util/cairo-trace/trace.c | 36 ++++++++++++++++++++++++++++++ 2 files changed, 69 insertions(+) commit 80fc2a8e49aa6ab4646b14eed9a4cf348a9149b4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Apr 28 09:50:49 2010 +0100 test: Expand xlib-expose-event to cover all backends Use a similar surface to create an equivalent backing surface for any backend, thus enabling the test to run against any target. The comment about forcing fallbacks has long since been false. test/xlib-expose-event.c | 76 ++++++------------------------------------------ 1 file changed, 9 insertions(+), 67 deletions(-) commit d3c4349730be991db0c85094103c744fc2d94836 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Apr 27 21:04:52 2010 +0100 test: Mark surfaces as finished if the data goes out of scope. The issue being that as the on-stack data is being referenced via a zero-copy snapshot outside of the functions scope as the surface is only finished and the source written long after the draw() returns. The correct procedure is that the user must call cairo_surface_finish() prior to any surface becoming inaccessible. In this case, this triggers the snapshot to preserve a copy of the data whilst it is still valid. test/bilevel-image.c | 4 +++- test/filter-nearest-offset.c | 1 + test/mask-ctm.c | 4 +++- test/mask-surface-ctm.c | 1 + test/move-to-show-surface.c | 2 ++ test/paint-repeat.c | 1 + test/paint-source-alpha.c | 1 + test/paint-with-alpha.c | 1 + test/rgb24-ignore-alpha.c | 1 + test/scale-down-source-surface-paint.c | 1 + test/scale-source-surface-paint.c | 1 + test/set-source.c | 1 + test/smask-image-mask.c | 4 +++- test/smask.c | 4 +++- test/source-surface-scale-paint.c | 1 + test/translate-show-surface.c | 1 + test/zero-alpha.c | 1 + 17 files changed, 26 insertions(+), 4 deletions(-) commit a5f54e48e3136076f3c8c60fc068f6a2105d9a33 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Apr 27 21:05:22 2010 +0100 snapshot: The snapshot masquerades as the target surface type. Not wholly convinced this is a good idea, but it matches the behaviour of the other internal surface types. src/cairo-surface-snapshot.c | 2 ++ 1 file changed, 2 insertions(+) commit 4a678afdf73d6f7b2b8a532ac7024976702c8aac Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Apr 27 18:57:56 2010 +0100 snapshot: Simply memcpy if the cloned image matches the original. src/cairo-surface-snapshot.c | 16 ++++++++++------ 1 file changed, 10 insertions(+), 6 deletions(-) commit b74cc0f6d5b71295c70d4c29e24ab252175f9a5b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Apr 27 18:57:00 2010 +0100 image: Convert 1x1 samples into solid colors. If the sampled extents of the operation on an image surface is just a single pixel, we can safely convert to a solid color. src/cairo-image-surface.c | 161 +++++++++++++++++++++++++++++++++++++++------- 1 file changed, 136 insertions(+), 25 deletions(-) commit f5cf131a9d8984bd6b3403396beed2ffbc26bded Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Apr 27 18:56:23 2010 +0100 surface: skip OVER is the source is clear. If the source has no alpha, the OVER operation becomes DST, i.e. a no-op. src/cairo-surface.c | 24 ++++++++++++++++++++++++ 1 file changed, 24 insertions(+) commit 453b1cba11f71694cfe0c3d934ac1f5aabda67fd Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Apr 27 16:47:31 2010 +0100 test: Update image ref images, and used fonts in README test/Makefile.am | 2 +- test/README | 6 +++--- test/finer-grained-fallbacks.argb32.ref.png | Bin 0 -> 1068 bytes test/finer-grained-fallbacks.ref.png | Bin 1070 -> 0 bytes test/finer-grained-fallbacks.rgb24.ref.png | Bin 839 -> 837 bytes test/ft-text-vertical-layout-type1.ref.png | Bin 3644 -> 3591 bytes test/ft-text-vertical-layout-type3.ref.png | Bin 3608 -> 3616 bytes test/mask-glyphs.ref.png | Bin 1189352 -> 1189351 bytes test/smask-text.ref.png | Bin 1874 -> 1672 bytes test/smask.ref.png | Bin 3423 -> 3396 bytes 10 files changed, 4 insertions(+), 4 deletions(-) commit 70656bba81e01eaf7b48817b2ca16fdf6d0b4bd5 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Apr 27 14:19:22 2010 +0100 test: Add image16 refs test/Makefile.am | 144 +++++++++++++++++++++ test/big-line.image16.ref.png | Bin 0 -> 988 bytes test/bug-bo-rectangular.image16.ref.png | Bin 0 -> 955 bytes test/caps-joins-alpha.image16.ref.png | Bin 0 -> 2268 bytes test/caps-joins-curve.image16.ref.png | Bin 0 -> 4928 bytes test/caps-joins.image16.ref.png | Bin 0 -> 2587 bytes test/caps-sub-paths.image16.ref.png | Bin 0 -> 171 bytes test/caps.image16.ref.png | Bin 0 -> 1936 bytes test/clear-source.image16.ref.png | Bin 0 -> 909 bytes test/clip-disjoint.image16.ref.png | Bin 0 -> 3886 bytes test/clip-fill-no-op.image16.ref.png | Bin 0 -> 153 bytes test/clip-fill-rule.image16.ref.png | Bin 0 -> 356 bytes test/clip-fill-unbounded.image16.ref.png | Bin 0 -> 1204 bytes test/clip-fill.image16.ref.png | Bin 0 -> 904 bytes test/clip-image.image16.ref.png | Bin 0 -> 2032 bytes test/clip-operator.image16.ref.png | Bin 0 -> 3131 bytes test/clip-push-group.image16.ref.png | Bin 0 -> 159 bytes test/clip-shape.image16.ref.png | Bin 0 -> 2488 bytes test/clip-stroke-no-op.image16.ref.png | Bin 0 -> 153 bytes test/clip-stroke-unbounded.image16.ref.png | Bin 0 -> 1255 bytes test/clip-stroke.image16.ref.png | Bin 0 -> 1308 bytes test/clip-text.image16.ref.png | Bin 0 -> 811 bytes test/clip-twice.image16.ref.png | Bin 0 -> 1069 bytes test/clipped-group.image16.ref.png | Bin 0 -> 281 bytes test/clipped-surface.image16.ref.png | Bin 0 -> 296 bytes test/close-path-current-point.image16.ref.png | Bin 0 -> 1804 bytes ...omposite-integer-translate-over.image16.ref.png | Bin 0 -> 11570 bytes ...posite-integer-translate-source.image16.ref.png | Bin 0 -> 11570 bytes test/copy-path.image16.ref.png | Bin 0 -> 556 bytes test/culled-glyphs.image16.ref.png | Bin 0 -> 426 bytes test/dash-caps-joins.image16.ref.png | Bin 0 -> 4491 bytes test/dash-curve.image16.ref.png | Bin 0 -> 32445 bytes test/dash-scale.image16.ref.png | Bin 0 -> 7748 bytes test/dash-state.image16.ref.png | Bin 0 -> 7910 bytes test/degenerate-arc.image16.ref.png | Bin 0 -> 592 bytes test/degenerate-arcs.image16.ref.png | Bin 0 -> 120 bytes test/degenerate-curve-to.image16.ref.png | Bin 0 -> 278 bytes test/degenerate-pen.image16.ref.png | Bin 0 -> 954 bytes test/degenerate-rel-curve-to.image16.ref.png | Bin 0 -> 279 bytes test/extend-pad-border.image16.ref.png | Bin 0 -> 446 bytes test/extend-reflect-similar.image16.ref.png | Bin 0 -> 99786 bytes test/extend-reflect.image16.ref.png | Bin 0 -> 99786 bytes test/extend-repeat-similar.image16.ref.png | Bin 0 -> 83738 bytes test/extend-repeat.image16.ref.png | Bin 0 -> 83738 bytes test/extended-blend-alpha.image16.ref.png | Bin 0 -> 4626 bytes test/extended-blend.image16.ref.png | Bin 0 -> 4145 bytes test/fill-alpha-pattern.image16.ref.png | Bin 0 -> 3380 bytes test/fill-alpha.image16.ref.png | Bin 0 -> 2145 bytes test/fill-and-stroke-alpha-add.image16.ref.png | Bin 0 -> 536 bytes test/fill-and-stroke-alpha.image16.ref.png | Bin 0 -> 470 bytes test/fill-and-stroke.image16.ref.png | Bin 0 -> 251 bytes test/fill-degenerate-sort-order.image16.ref.png | Bin 0 -> 1753 bytes test/fill-image.image16.ref.png | Bin 0 -> 1163 bytes test/fill-rule.image16.ref.png | Bin 0 -> 1625 bytes test/filter-bilinear-extents.image16.ref.png | Bin 0 -> 895 bytes test/filter-nearest-transformed.image16.ref.png | Bin 0 -> 418 bytes test/finer-grained-fallbacks.image16.ref.png | Bin 0 -> 862 bytes test/font-matrix-translation.image16.ref.png | Bin 0 -> 852 bytes test/ft-show-glyphs-positioning.image16.ref.png | Bin 0 -> 2772 bytes test/ft-show-glyphs-table.image16.ref.png | Bin 0 -> 8052 bytes test/ft-text-vertical-layout-type1.image16.ref.png | Bin 0 -> 3072 bytes test/ft-text-vertical-layout-type3.image16.ref.png | Bin 0 -> 3141 bytes test/glyph-cache-pressure.image16.ref.png | Bin 0 -> 2453 bytes test/group-clip.image16.ref.png | Bin 0 -> 193 bytes test/group-unaligned.image16.ref.png | Bin 0 -> 395 bytes test/huge-linear.image16.ref.png | Bin 0 -> 1542 bytes test/huge-radial.image16.ref.png | Bin 0 -> 17893 bytes test/image-surface-source.image16.ref.png | Bin 0 -> 305 bytes test/joins.image16.ref.png | Bin 0 -> 5858 bytes test/large-font.image16.ref.png | Bin 0 -> 5713 bytes test/large-twin-antialias-mixed.image16.ref.png | Bin 0 -> 14398 bytes test/leaky-dashed-rectangle.image16.ref.png | Bin 0 -> 367 bytes test/leaky-dashed-stroke.image16.ref.png | Bin 0 -> 8089 bytes test/leaky-polygon.image16.ref.png | Bin 0 -> 329 bytes test/line-width-scale.image16.ref.png | Bin 0 -> 4721 bytes test/linear-gradient-reflect.image16.ref.png | Bin 0 -> 190 bytes test/linear-gradient-subset.image16.ref.png | Bin 0 -> 791 bytes test/linear-gradient.image16.ref.png | Bin 0 -> 941 bytes test/linear-uniform.image16.ref.png | Bin 0 -> 131 bytes test/long-dashed-lines.image16.ref.png | Bin 0 -> 1974 bytes test/mask-alpha.image16.ref.png | Bin 0 -> 560 bytes test/mask-glyphs.image16.ref.png | Bin 0 -> 1053144 bytes test/mask-transformed-image.image16.ref.png | Bin 0 -> 2748 bytes test/mask-transformed-similar.image16.ref.png | Bin 0 -> 2748 bytes test/mask.image16.ref.png | Bin 0 -> 5533 bytes test/operator-alpha-alpha.image16.ref.png | Bin 0 -> 4140 bytes test/operator-source.image16.ref.png | Bin 0 -> 2756 bytes test/over-around-source.image16.ref.png | Bin 0 -> 491 bytes test/paint-source-alpha.image16.ref.png | Bin 0 -> 253 bytes test/paint-with-alpha.image16.ref.png | Bin 0 -> 253 bytes test/path-append.image16.ref.png | Bin 0 -> 5252 bytes test/path-stroke-twice.image16.ref.png | Bin 0 -> 210 bytes test/pdf-surface-source.image16.ref.png | Bin 0 -> 305 bytes test/ps-surface-source.image16.ref.png | Bin 0 -> 305 bytes test/pthread-same-source.image16.ref.png | Bin 0 -> 1007 bytes test/push-group-color.image16.ref.png | Bin 0 -> 2277 bytes test/push-group.image16.ref.png | Bin 0 -> 2286 bytes test/radial-gradient-source.image16.ref.png | Bin 0 -> 132037 bytes test/radial-gradient.image16.ref.png | Bin 0 -> 202415 bytes .../random-intersections-curves-eo.image16.ref.png | Bin 0 -> 181565 bytes .../random-intersections-curves-nz.image16.ref.png | Bin 0 -> 204090 bytes test/random-intersections-eo.image16.ref.png | Bin 0 -> 97747 bytes test/random-intersections-nonzero.image16.ref.png | Bin 0 -> 107644 bytes test/recording-surface-pattern.image16.ref.png | Bin 0 -> 2856 bytes test/reflected-stroke.image16.ref.png | Bin 0 -> 4254 bytes test/rotated-clip.image16.ref.png | Bin 0 -> 3336 bytes test/rounded-rectangle-fill.image16.ref.png | Bin 0 -> 730 bytes test/rounded-rectangle-stroke.image16.ref.png | Bin 0 -> 732 bytes test/scale-offset-image.image16.ref.png | Bin 0 -> 7793 bytes test/scale-offset-similar.image16.ref.png | Bin 0 -> 7793 bytes test/select-font-face.image16.ref.png | Bin 0 -> 1962 bytes test/show-text-current-point.image16.ref.png | Bin 0 -> 1932 bytes test/smask-fill.image16.ref.png | Bin 0 -> 925 bytes test/smask-mask.image16.ref.png | Bin 0 -> 1358 bytes test/smask-paint.image16.ref.png | Bin 0 -> 1469 bytes test/smask-stroke.image16.ref.png | Bin 0 -> 1330 bytes test/smask-text.image16.ref.png | Bin 0 -> 1206 bytes test/smask.image16.ref.png | Bin 0 -> 2213 bytes test/spline-decomposition.image16.ref.png | Bin 0 -> 14064 bytes test/stroke-ctm-caps.image16.ref.png | Bin 0 -> 908 bytes test/stroke-image.image16.ref.png | Bin 0 -> 1167 bytes test/subsurface.image16.ref.png | Bin 0 -> 1430 bytes test/surface-pattern-operator.image16.ref.png | Bin 0 -> 1977 bytes test/surface-pattern-scale-down.image16.ref.png | Bin 0 -> 1313 bytes test/surface-pattern-scale-up.image16.ref.png | Bin 0 -> 3864 bytes test/surface-pattern.image16.ref.png | Bin 0 -> 11870 bytes test/svg-surface-source.image16.ref.png | Bin 0 -> 305 bytes test/text-antialias-gray.image16.ref.png | Bin 0 -> 895 bytes test/text-antialias-subpixel.image16.ref.png | Bin 0 -> 866 bytes test/text-glyph-range.image16.ref.png | Bin 0 -> 1731 bytes test/text-rotate.image16.ref.png | Bin 0 -> 12599 bytes test/text-transform.image16.ref.png | Bin 0 -> 4469 bytes test/transforms.image16.ref.png | Bin 0 -> 326 bytes test/trap-clip.image16.ref.png | Bin 0 -> 4344 bytes test/twin-antialias-gray.image16.ref.png | Bin 0 -> 3005 bytes test/twin-antialias-mixed.image16.ref.png | Bin 0 -> 2049 bytes test/twin-antialias-subpixel.image16.ref.png | Bin 0 -> 3005 bytes test/twin.image16.ref.png | Bin 0 -> 3005 bytes test/unbounded-operator.image16.ref.png | Bin 0 -> 1276 bytes test/user-font-mask.image16.ref.png | Bin 0 -> 4948 bytes test/user-font-proxy.image16.ref.png | Bin 0 -> 14460 bytes test/user-font-rescale.image16.ref.png | Bin 0 -> 12590 bytes test/user-font.image16.ref.png | Bin 0 -> 5814 bytes test/xcomposite-projection.image16.ref.png | Bin 0 -> 1000 bytes test/xlib-surface-source.image16.ref.png | Bin 0 -> 305 bytes 145 files changed, 144 insertions(+) commit ebe6f2ac6988991afde0d685bea9f207ed3360d8 Author: Andrea Canciani <ranma42@gmail.com> Date: Tue Apr 27 10:59:09 2010 +0200 quartz: Assert success of path creation Path creation can only fail because of the callbacks, but in quartz they all return CAIRO_STATUS_SUCCESS. Therefore we can just assert that path creation was successful and simplify calling functions (as they don't have to handle potential errors anymore). src/cairo-quartz-surface.c | 35 +++++++++++++++-------------------- 1 file changed, 15 insertions(+), 20 deletions(-) commit f67b6009278ef3dfe91ddbffb989dcfeed174352 Author: Andrea Canciani <ranma42@gmail.com> Date: Sun Apr 4 11:50:41 2010 +0200 quartz: Simplify path creation If paths are created before changing the ctm (when stroking) no multiplication is needed in the path construction code. src/cairo-quartz-surface.c | 60 ++++++++-------------------------------------- 1 file changed, 10 insertions(+), 50 deletions(-) commit 3b2ceff0502ba409c161e497ebe015e0a0a88847 Author: Andrea Canciani <ranma42@gmail.com> Date: Wed Apr 7 18:34:25 2010 +0200 quartz: Stroke without ctm_inverse multiplication If the CTM is not changed before creating the path, no multiplication needs to be made between points and the inverse of the CTM. src/cairo-quartz-surface.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) commit 514d366cde689f8200b049834bebbd421d5d8bcb Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Apr 27 11:07:09 2010 +0100 snapshot: propagate status on finish src/cairo-surface-snapshot.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) commit a505104013a1db0c8b1092c8a1848d7f0b02e6bc Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Apr 26 20:22:13 2010 +0100 image: Compute sample extents In order to reuse the original image as the pixman pattern, then the entire operation must be wholly contained within the extents of the image (including subsurfaces) and be reducible to an untransformed REPEAT_NONE. src/cairo-image-surface.c | 72 +++++++++++++++++++++++++++++++---------------- src/cairo-pattern.c | 2 +- src/cairoint.h | 4 +++ 3 files changed, 52 insertions(+), 26 deletions(-) commit b8a7f8621a84083735d0e2c8748f5fa2b7f4b36a Author: Andrea Canciani <ranma42@gmail.com> Date: Tue Apr 27 10:17:23 2010 +0200 Update FSF address I updated the Free Software Foundation address using the following script. for i in $(git grep Temple | cut -d: -f1 ) do sed -e 's/59 Temple Place[, -]* Suite 330, Boston, MA *02111-1307[, ]* USA/51 Franklin Street, Suite 500, Boston, MA 02110-1335, USA/' -i "$i" done Fixes http://bugs.freedesktop.org/show_bug.cgi?id=21356 COPYING-LGPL-2.1 | 4 ++-- boilerplate/cairo-boilerplate-drm.c | 2 +- boilerplate/cairo-boilerplate-gl.c | 2 +- boilerplate/cairo-boilerplate-qt.cpp | 2 +- boilerplate/cairo-boilerplate-vg.c | 2 +- build/aclocal.pkg.m4 | 2 +- doc/tutorial/src/include/cairo-tutorial-gtk.h | 2 +- doc/tutorial/src/include/cairo-tutorial-pdf.h | 2 +- doc/tutorial/src/include/cairo-tutorial-png.h | 2 +- doc/tutorial/src/include/cairo-tutorial-xlib.h | 2 +- doc/tutorial/src/include/cairo-tutorial.h | 2 +- src/cairo-analysis-surface-private.h | 2 +- src/cairo-analysis-surface.c | 2 +- src/cairo-arc-private.h | 2 +- src/cairo-arc.c | 2 +- src/cairo-array.c | 2 +- src/cairo-atomic-private.h | 2 +- src/cairo-atomic.c | 2 +- src/cairo-base64-stream.c | 2 +- src/cairo-base85-stream.c | 2 +- src/cairo-bentley-ottmann-rectangular.c | 2 +- src/cairo-bentley-ottmann-rectilinear.c | 2 +- src/cairo-bentley-ottmann.c | 2 +- src/cairo-beos-surface.cpp | 2 +- src/cairo-beos.h | 2 +- src/cairo-botor-scan-converter.c | 2 +- src/cairo-boxes-private.h | 2 +- src/cairo-boxes.c | 2 +- src/cairo-cache-private.h | 2 +- src/cairo-cache.c | 2 +- src/cairo-cff-subset.c | 2 +- src/cairo-clip-private.h | 2 +- src/cairo-clip.c | 2 +- src/cairo-color.c | 2 +- src/cairo-combsort-private.h | 2 +- src/cairo-compiler-private.h | 2 +- src/cairo-composite-rectangles-private.h | 2 +- src/cairo-composite-rectangles.c | 2 +- src/cairo-debug.c | 2 +- src/cairo-deflate-stream.c | 2 +- src/cairo-deprecated.h | 2 +- src/cairo-device-private.h | 2 +- src/cairo-device.c | 2 +- src/cairo-directfb-surface.c | 2 +- src/cairo-directfb.h | 2 +- src/cairo-drm.h | 2 +- src/cairo-egl-context.c | 2 +- src/cairo-error-private.h | 2 +- src/cairo-fixed-private.h | 2 +- src/cairo-fixed-type-private.h | 2 +- src/cairo-fixed.c | 2 +- src/cairo-font-face-twin.c | 2 +- src/cairo-font-face.c | 2 +- src/cairo-font-options.c | 2 +- src/cairo-freed-pool-private.h | 2 +- src/cairo-freed-pool.c | 2 +- src/cairo-ft-font.c | 2 +- src/cairo-ft-private.h | 2 +- src/cairo-ft.h | 2 +- src/cairo-gl-glyphs.c | 2 +- src/cairo-gl-private.h | 2 +- src/cairo-gl-shaders.c | 2 +- src/cairo-gl-surface.c | 2 +- src/cairo-gl.h | 2 +- src/cairo-glx-context.c | 2 +- src/cairo-gstate-private.h | 2 +- src/cairo-gstate.c | 2 +- src/cairo-hash-private.h | 2 +- src/cairo-hash.c | 2 +- src/cairo-hull.c | 2 +- src/cairo-image-info-private.h | 2 +- src/cairo-image-info.c | 2 +- src/cairo-image-surface.c | 2 +- src/cairo-list-private.h | 2 +- src/cairo-lzw.c | 2 +- src/cairo-malloc-private.h | 2 +- src/cairo-matrix.c | 2 +- src/cairo-misc.c | 2 +- src/cairo-mutex-impl-private.h | 2 +- src/cairo-mutex-list-private.h | 2 +- src/cairo-mutex-private.h | 2 +- src/cairo-mutex-type-private.h | 2 +- src/cairo-mutex.c | 2 +- src/cairo-os2-private.h | 2 +- src/cairo-os2-surface.c | 2 +- src/cairo-os2.h | 2 +- src/cairo-output-stream-private.h | 2 +- src/cairo-output-stream.c | 2 +- src/cairo-paginated-private.h | 2 +- src/cairo-paginated-surface-private.h | 2 +- src/cairo-paginated-surface.c | 2 +- src/cairo-path-bounds.c | 2 +- src/cairo-path-fill.c | 2 +- src/cairo-path-fixed-private.h | 2 +- src/cairo-path-fixed.c | 2 +- src/cairo-path-in-fill.c | 2 +- src/cairo-path-private.h | 2 +- src/cairo-path-stroke.c | 2 +- src/cairo-path.c | 2 +- src/cairo-pdf-operators-private.h | 2 +- src/cairo-pdf-operators.c | 2 +- src/cairo-pdf-surface-private.h | 2 +- src/cairo-pdf-surface.c | 2 +- src/cairo-pdf.h | 2 +- src/cairo-pen.c | 2 +- src/cairo-png.c | 2 +- src/cairo-polygon.c | 2 +- src/cairo-private.h | 2 +- src/cairo-ps-surface-private.h | 2 +- src/cairo-ps-surface.c | 2 +- src/cairo-ps.h | 2 +- src/cairo-qt-surface.cpp | 2 +- src/cairo-qt.h | 2 +- src/cairo-quartz-font.c | 2 +- src/cairo-quartz-image-surface.c | 2 +- src/cairo-quartz-image.h | 2 +- src/cairo-quartz-private.h | 2 +- src/cairo-quartz-surface.c | 2 +- src/cairo-quartz.h | 2 +- src/cairo-recording-surface-private.h | 2 +- src/cairo-recording-surface.c | 2 +- src/cairo-rectangle.c | 2 +- src/cairo-rectangular-scan-converter.c | 2 +- src/cairo-reference-count-private.h | 2 +- src/cairo-region-private.h | 2 +- src/cairo-region.c | 2 +- src/cairo-rtree-private.h | 2 +- src/cairo-rtree.c | 2 +- src/cairo-scaled-font-private.h | 2 +- src/cairo-scaled-font-subsets-private.h | 2 +- src/cairo-scaled-font-subsets.c | 2 +- src/cairo-scaled-font.c | 2 +- src/cairo-script-surface.c | 2 +- src/cairo-script.h | 2 +- src/cairo-skia-surface.cpp | 2 +- src/cairo-skia.h | 2 +- src/cairo-slope-private.h | 2 +- src/cairo-slope.c | 2 +- src/cairo-spline.c | 2 +- src/cairo-stroke-style.c | 2 +- src/cairo-surface-clipper-private.h | 2 +- src/cairo-surface-clipper.c | 2 +- src/cairo-surface-fallback-private.h | 2 +- src/cairo-surface-fallback.c | 2 +- src/cairo-surface-offset-private.h | 2 +- src/cairo-surface-offset.c | 2 +- src/cairo-surface-private.h | 2 +- src/cairo-surface-snapshot-private.h | 2 +- src/cairo-surface-snapshot.c | 2 +- src/cairo-surface-subsurface-private.h | 2 +- src/cairo-surface-subsurface.c | 2 +- src/cairo-surface-wrapper-private.h | 2 +- src/cairo-surface-wrapper.c | 2 +- src/cairo-surface.c | 2 +- src/cairo-svg-surface-private.h | 2 +- src/cairo-svg-surface.c | 2 +- src/cairo-svg.h | 2 +- src/cairo-system.c | 2 +- src/cairo-tee-surface-private.h | 2 +- src/cairo-tee-surface.c | 2 +- src/cairo-toy-font-face.c | 2 +- src/cairo-traps.c | 2 +- src/cairo-truetype-subset-private.h | 2 +- src/cairo-truetype-subset.c | 2 +- src/cairo-type1-fallback.c | 2 +- src/cairo-type1-private.h | 2 +- src/cairo-type1-subset.c | 2 +- src/cairo-type3-glyph-surface-private.h | 2 +- src/cairo-type3-glyph-surface.c | 2 +- src/cairo-types-private.h | 2 +- src/cairo-unicode.c | 2 +- src/cairo-user-font-private.h | 2 +- src/cairo-user-font.c | 2 +- src/cairo-version.c | 2 +- src/cairo-vg-surface.c | 2 +- src/cairo-vg.h | 2 +- src/cairo-wideint-private.h | 2 +- src/cairo-wideint-type-private.h | 2 +- src/cairo-wideint.c | 2 +- src/cairo-win32-font.c | 2 +- src/cairo-win32-printing-surface.c | 2 +- src/cairo-win32-private.h | 2 +- src/cairo-win32-surface.c | 2 +- src/cairo-win32.h | 2 +- src/cairo-xcb-connection-core.c | 2 +- src/cairo-xcb-connection-render.c | 2 +- src/cairo-xcb-connection-shm.c | 2 +- src/cairo-xcb-connection.c | 2 +- src/cairo-xcb-private.h | 2 +- src/cairo-xcb-screen.c | 2 +- src/cairo-xcb-shm.c | 2 +- src/cairo-xcb-surface-cairo.c | 2 +- src/cairo-xcb-surface-core.c | 2 +- src/cairo-xcb-surface-private.h | 2 +- src/cairo-xcb-surface-render.c | 2 +- src/cairo-xcb-surface.c | 2 +- src/cairo-xcb.h | 2 +- src/cairo-xlib-display.c | 2 +- src/cairo-xlib-private.h | 2 +- src/cairo-xlib-screen.c | 2 +- src/cairo-xlib-surface-private.h | 2 +- src/cairo-xlib-surface.c | 2 +- src/cairo-xlib-visual.c | 2 +- src/cairo-xlib-xcb-surface.c | 2 +- src/cairo-xlib-xrender-private.h | 2 +- src/cairo-xlib-xrender.h | 2 +- src/cairo-xlib.h | 2 +- src/cairo-xml-surface.c | 2 +- src/cairo-xml.h | 2 +- src/cairo.c | 2 +- src/cairo.h | 2 +- src/cairoint.h | 2 +- src/drm/cairo-drm-bo.c | 2 +- src/drm/cairo-drm-gallium-surface.c | 2 +- src/drm/cairo-drm-i915-glyphs.c | 2 +- src/drm/cairo-drm-i915-shader.c | 2 +- src/drm/cairo-drm-i915-spans.c | 2 +- src/drm/cairo-drm-i915-surface.c | 2 +- src/drm/cairo-drm-i965-glyphs.c | 2 +- src/drm/cairo-drm-i965-shader.c | 2 +- src/drm/cairo-drm-i965-spans.c | 2 +- src/drm/cairo-drm-i965-surface.c | 2 +- src/drm/cairo-drm-intel-ioctl-private.h | 2 +- src/drm/cairo-drm-intel-private.h | 2 +- src/drm/cairo-drm-intel-surface.c | 2 +- src/drm/cairo-drm-intel.c | 2 +- src/drm/cairo-drm-private.h | 2 +- src/drm/cairo-drm-radeon-private.h | 2 +- src/drm/cairo-drm-radeon-surface.c | 2 +- src/drm/cairo-drm-radeon.c | 2 +- src/drm/cairo-drm-surface.c | 2 +- src/drm/cairo-drm.c | 2 +- src/test-fallback-surface.c | 2 +- src/test-fallback-surface.h | 2 +- src/test-fallback16-surface.c | 2 +- src/test-fallback16-surface.h | 2 +- src/test-null-surface.c | 2 +- src/test-null-surface.h | 2 +- src/test-paginated-surface.c | 2 +- src/test-paginated-surface.h | 2 +- src/test-wrapping-surface.c | 2 +- src/test-wrapping-surface.h | 2 +- test/dash-offset.c | 2 +- test/miter-precision.c | 2 +- test/operator-alpha-alpha.c | 2 +- test/pdiff/args.c | 2 +- test/pdiff/args.h | 2 +- test/pdiff/lpyramid.c | 2 +- test/pdiff/lpyramid.h | 2 +- test/pdiff/pdiff.c | 2 +- test/pdiff/pdiff.h | 2 +- test/pdiff/perceptualdiff.c | 2 +- util/backtrace-symbols.c | 2 +- util/cairo-script/cairo-script-file.c | 2 +- util/cairo-script/cairo-script-hash.c | 2 +- util/cairo-script/cairo-script-interpreter.c | 2 +- util/cairo-script/cairo-script-interpreter.h | 2 +- util/cairo-script/cairo-script-objects.c | 2 +- util/cairo-script/cairo-script-operators.c | 2 +- util/cairo-script/cairo-script-private.h | 2 +- util/cairo-script/cairo-script-scanner.c | 2 +- util/cairo-script/cairo-script-stack.c | 2 +- 262 files changed, 263 insertions(+), 263 deletions(-) commit 9672aab6b7c3f1efc065a6e5eb69443a1ecaa723 Author: Benjamin Otte <otte@redhat.com> Date: Mon Apr 26 21:28:53 2010 +0200 tests: Integrate xlib-expose-event test into usual tests Previously the test was using the preamble vfunc and generating output itself. Now it uses the draw function and ignores any but the xlib backends. test/xlib-expose-event.c | 63 ++++++++++++------------------------------------ 1 file changed, 15 insertions(+), 48 deletions(-) commit 36e0a3d3a01f9962d97386ea2d3c60020e937a81 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Apr 26 18:18:27 2010 +0100 trace: Remove unused functions trace.c:1154: warning: ‘_has_font_face_id’ defined but not used trace.c:1196: warning: ‘_get_pattern_id’ defined but not used trace.c:1870: warning: ‘_emit_font_face’ defined but not used trace.c:1882: warning: ‘_emit_scaled_font’ defined but not used util/cairo-trace/trace.c | 24 ------------------------ 1 file changed, 24 deletions(-) commit 3a2d9ffe0333090bb31ff01048ed506595f20cf9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Apr 26 18:17:49 2010 +0100 trace: Check return value to suppress compiler warning trace.c: In function ‘get_prog_name’: trace.c:741: warning: ignoring return value of ‘fgets’, declared with attribute warn_unused_result util/cairo-trace/trace.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) commit 37be183412eb35abc11e602857602aee05839fc7 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Apr 26 18:12:05 2010 +0100 trace: Trivial compiler warning fix trace.c:1665: warning: initialization from incompatible pointer type util/cairo-trace/trace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 3c9e5d9792e101cfbd5a1116f06579539ed7b585 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Apr 26 18:13:36 2010 +0100 region: Cast to remove const to suppress compiler warnings. airo-region.c: In function ‘cairo_region_intersect’: cairo-region.c:503: warning: passing argument 3 of ‘pixman_region32_intersect’ discards qualifiers from pointer target type /usr/local/include/pixman-1/pixman.h:518: note: expected ‘struct pixman_region32_t *’ but argument is of type ‘const struct pixman_region32_t *’ cairo-region.c: In function ‘cairo_region_union’: cairo-region.c:566: warning: passing argument 3 of ‘pixman_region32_union’ discards qualifiers from pointer target type /usr/local/include/pixman-1/pixman.h:521: note: expected ‘struct pixman_region32_t *’ but argument is of type ‘const struct pixman_region32_t *’ src/cairo-region.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 8c72122df51a10a32b56cdf20e5a1037ee59a958 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Apr 25 10:30:12 2010 +0100 clip: Skip combining with solid pixel aligned boxes. src/cairo-clip.c | 6 ++++++ 1 file changed, 6 insertions(+) commit 240ebaf81a42c24c4d4267879c5ef8102e97017e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Apr 25 09:44:18 2010 +0100 clip: Fix sign reverse when combining with the clip surface. Finally, found the reversed sign in the clipping code, thanks cu! Fixes: test/clip-shape src/cairo-clip.c | 24 ++---------------------- 1 file changed, 2 insertions(+), 22 deletions(-) commit 4d3632761b928c14fb1ce257af077f45658d8537 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Apr 25 09:42:13 2010 +0100 clip: Report the surface offset when retrieving the clip mask Stop the callers from guessing the origin of the clip surface by reporting it explicitly! This enables the clip to bypass any rectangles overlaid on top of the clip surface, which is common when the backends limit the clip to the extents of the operation -- but irrelevant to the actual content of the clip mask src/cairo-clip-private.h | 2 +- src/cairo-clip.c | 26 ++++++++++++++++++++------ src/cairo-image-surface.c | 39 +++++++++++++++++++++------------------ src/cairo-surface-fallback.c | 20 ++++++++++---------- src/cairo-xcb-surface-render.c | 31 +++++++++++++++---------------- src/drm/cairo-drm-i915-shader.c | 9 ++++----- src/drm/cairo-drm-i965-shader.c | 7 ++++--- 7 files changed, 75 insertions(+), 59 deletions(-) commit 241ce933f8fa44210015dcdc4a66b3f9d8ac56e8 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Apr 25 09:37:50 2010 +0100 test: Add clip-shape cu found a bug when using clipping and fills with the image backend, but it turns out to be the lost sign bug afflicting everything... To trigger the bug requires evaluating the clip surface prior to using in with _cairo_clip_combine_with_surface(). Which is taken along a particular path when using a clip mask with non-pixel geometry on image, but more easily hit by the fallback code. Reported-by: <cairouser@yahoo.com> test/Makefile.am | 4 ++ test/Makefile.sources | 1 + test/clip-shape.c | 86 ++++++++++++++++++++++++++++++++++ test/clip-shape.ps.ref.png | Bin 0 -> 1797 bytes test/clip-shape.ref.png | Bin 0 -> 2935 bytes test/clip-shape.xlib-fallback.ref.png | Bin 0 -> 2916 bytes test/clip-shape.xlib.ref.png | Bin 0 -> 2944 bytes 7 files changed, 91 insertions(+) commit 1687c7b7d3074de8eeea1a5c339df9a12cc38da2 Author: Benjamin Otte <otte@redhat.com> Date: Fri Apr 23 22:22:28 2010 +0200 xlib: Fix display closing to work properly Previously we free()'d the cairo_device's memory which was not good at all. Now the code causes cairo_device_finish() instead. src/cairo-xlib-display.c | 37 +++++++++++++++++++++++-------------- 1 file changed, 23 insertions(+), 14 deletions(-) commit 637564c562de21c17c36d192d3ab1b3fe069754b Author: Benjamin Otte <otte@redhat.com> Date: Fri Apr 23 22:10:32 2010 +0200 xlib: Fix screen device handling Add a _cairo_xlib_device_create() function that could easily be exported as a replacement for _cairo_xlib_display_get(). This function returns a cairo_device_t instead of a cairo_xlib_display_t because the display isn't acquired. src/cairo-xlib-display.c | 33 ++++++++++++++++++--------------- src/cairo-xlib-private.h | 4 ++-- src/cairo-xlib-screen.c | 23 ++++++++++++++++------- 3 files changed, 36 insertions(+), 24 deletions(-) commit 49b52a8946cbd5f785f71069313e4a204358887b Author: Benjamin Otte <otte@redhat.com> Date: Thu Apr 22 21:56:31 2010 +0200 gl: Make drawing to windows work again The correct MIN/MAG_FILTER wasn't set. src/cairo-gl-surface.c | 2 ++ 1 file changed, 2 insertions(+) commit 3efbc0c5c850d0cb5c5af5bcabbc7293670ea355 Author: Benjamin Otte <otte@redhat.com> Date: Tue Apr 20 21:58:59 2010 +0200 gl: Only unref a surface if it exists Note: This will likely work for NULL clones, but I prefer not dereferencing NULLs. That gives people a wrong understanding of the code (i.e. me). src/cairo-gl-surface.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit 64662be4ef3851d21658e5fdb2efb6806b45eba8 Author: Benjamin Otte <otte@redhat.com> Date: Thu Apr 22 22:05:03 2010 +0200 gl: Use GLfloat instead of float in gl*Array() functions src/cairo-gl-surface.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) commit e4f84f97b2f5d37bc1fb7dd510df733053ffe624 Author: Benjamin Otte <otte@redhat.com> Date: Tue Apr 20 21:33:50 2010 +0200 gl: Don't acquire the gl context twice src/cairo-gl-surface.c | 7 ------- 1 file changed, 7 deletions(-) commit e40a2d1f5eeea139e29c27e38495b9c0bf9e39a1 Author: Benjamin Otte <otte@redhat.com> Date: Thu Apr 22 22:03:29 2010 +0200 gl: Use correct type when uploading images src/cairo-gl-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 55cf323f963bbcc11bcc290eaf71656e1ba91efd Author: Benjamin Otte <otte@redhat.com> Date: Thu Apr 22 21:58:01 2010 +0200 gl: fix typo in comment src/cairo-gl-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 887d43578c3acc7d45c8b2e3ade5da29be804eeb Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Apr 23 20:14:03 2010 +0100 xlib: Simply release the ref, not destroy, from CloseDisplay. Do not call the destroy function directly, but rely on the reference counting to call the notifier upon the last reference. Instead, simply release the reference we were holding for the cache and CloseDisplay callback. src/cairo-xlib-display.c | 22 ++++++++++++---------- 1 file changed, 12 insertions(+), 10 deletions(-) commit c63e3490a5fc2836837e7adcb5ecad62bdfd18ab Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Apr 22 20:33:06 2010 +0100 cairo: Handle the all-clipped state in cairo_push_group() Yet another bug reported by Jeff Muizelaar, thanks! Fixes: test/clip-empty-group src/cairo.c | 81 +++++++++++++++++++++++++++++++++++-------------------------- 1 file changed, 46 insertions(+), 35 deletions(-) commit 567e485f28716d7b72cbf864a0c573148be91cd8 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Apr 23 14:37:43 2010 +0100 test: Add clip-empty-group Exercise a bug reported by Jeff Muizelaar whereby cairo_push_group() was broken by everything being clipped out. test/Makefile.am | 1 + test/Makefile.sources | 1 + test/clip-empty-group.c | 65 ++++++++++++++++++++++++++++++++++++++++++ test/clip-empty-group.ref.png | Bin 0 -> 103 bytes 4 files changed, 67 insertions(+) commit 4438cc6a49e7e902dce045706f7125a2c3e2174b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Apr 3 18:10:57 2010 +0100 cairo: Use explicit device flushing. Now with the concept of a cairo_device_t and the ability to flush it, we now longer require the heuristic of automatically flushing on behalf of the user at the end of every context. src/cairo.c | 11 ----------- 1 file changed, 11 deletions(-) commit 0fc140abc3c848e72b1b2b959026355fdbd6a941 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Apr 21 12:01:49 2010 +0100 pdf: Suppress compiler warning. src/cairo-pdf-operators.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) commit 1923a33a7277009b2898840d671fdc8f8d1b3922 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Apr 21 11:59:33 2010 +0100 cairoint.h: Change ASSERT_NOT_REACHED to remove compiler warning src/cairoint.h | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) commit d366da7d8c46b08673b9e36dd130608a311ded4b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Apr 23 14:38:42 2010 +0100 Win32 features update boilerplate/Makefile.win32.features | 8 -------- src/Makefile.win32.features | 8 -------- 2 files changed, 16 deletions(-) commit c1e51a0264d204716f7706fbc66c2a7786b7947c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Apr 23 14:33:53 2010 +0100 build: Spelling correction build/Makefile.win32.features-h | 1 - build/configure.ac.features | 2 +- 2 files changed, 1 insertion(+), 2 deletions(-) commit e0306da1d73509e2cfd0fce81a1b4259e5e3fcef Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Apr 21 22:51:53 2010 +0100 configure: typo in check for xcb-shm configure.ac | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit b7fbda276df52de6a5ac67397a8d49657233fcfa Author: Andrea Canciani <ranma42@gmail.com> Date: Sun Apr 18 23:16:07 2010 +0200 Revert "script: Another RGB16_565 warning." This reverts commit a9f506493371ac91494488e5ca38e57ceb5a10cb. The change was already applied in bdf8f50adae67498d5fe36727ed2cd34c0206509. src/cairo-script-surface.c | 6 ------ 1 file changed, 6 deletions(-) commit f74ad37e66bbdcc4d727ed6d931dde870d84c2f4 Author: Benjamin Otte <otte@redhat.com> Date: Sun Apr 18 23:20:02 2010 +0200 xlib: Make display a cairo_device_t The code now uses the locking of the cairo_device_t instead of its own mutexes. The code was modified so that cairo_surface_t does no longer reference the display directly. Instead, it calls _cairo_xlib_display_acquire(). If a function assumes an already acquired display, it now requires the caller to provide a cairo_xlib_display_t parameter. Functions that do not require an acquired display will not require the display parameter or they will take the cairo_device_t instead. src/cairo-xlib-display.c | 234 +++++++--------- src/cairo-xlib-private.h | 66 +++-- src/cairo-xlib-screen.c | 161 +++-------- src/cairo-xlib-surface-private.h | 2 - src/cairo-xlib-surface.c | 557 +++++++++++++++++++++++---------------- 5 files changed, 494 insertions(+), 526 deletions(-) commit 2bffa2467ed67517419e425fc34fb5b8294a9e96 Author: Benjamin Otte <otte@redhat.com> Date: Wed Jan 27 10:56:22 2010 +0100 Use pixman_image_composite32() Replace all occurences of pixman_image_composite() src/cairo-directfb-surface.c | 16 +- src/cairo-image-surface.c | 352 ++++++++++++++++++++--------------------- src/cairo-pattern.c | 16 +- src/cairo-surface-snapshot.c | 12 +- src/cairo-surface-subsurface.c | 24 +-- src/cairo-xlib-surface.c | 16 +- src/drm/cairo-drm-intel.c | 12 +- 7 files changed, 224 insertions(+), 224 deletions(-) commit 567923d66a375259a6a8083aa4ef1051097a60c5 Author: Benjamin Otte <otte@redhat.com> Date: Wed Jan 27 09:39:20 2010 +0100 image: Make unbounded fixup code not use pixman_fill() The new pixman_image_fill_boxes() does proper fallbacks for weird pixman formats. src/cairo-image-surface.c | 123 ++++++++++++++++++---------------------------- 1 file changed, 47 insertions(+), 76 deletions(-) commit b648365cfa2d808b34b924a6b105718e0673bb9b Author: Benjamin Otte <otte@redhat.com> Date: Wed Jan 27 09:21:41 2010 +0100 Require pixman 0.17.5 We want pixman_image_composite32() and pixman_image_fill_boxes() configure.ac | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 0a19ad7c104cd81fa2b93aaf7ef9302f50b3f369 Author: Benjamin Otte <otte@redhat.com> Date: Thu Apr 15 17:58:16 2010 +0200 gl: Get rid of "return value not checked" warnings Assert that those functions only return successfully. src/cairo-gl-surface.c | 42 ++++++++++++++++++++++++------------------ 1 file changed, 24 insertions(+), 18 deletions(-) commit 7745ea21e104d0a5eec1814f19bacb7334f42185 Author: Benjamin Otte <otte@redhat.com> Date: Wed Apr 14 15:59:36 2010 +0200 gl: acquire/release properly in _cairo_gl_draw_image() Fixes various testsuite failures, in particular ones related to masks. src/cairo-gl-surface.c | 16 ++++++++++++++-- 1 file changed, 14 insertions(+), 2 deletions(-) commit 596d3017a1b11cb9844e69330cdece7441929725 Author: Benjamin Otte <otte@redhat.com> Date: Tue Apr 13 20:53:47 2010 +0200 gl: acquire/release context when getting images Avoids Mesa complaining about no current context. src/cairo-gl-surface.c | 7 +++++++ 1 file changed, 7 insertions(+) commit 46ef05c3d73af2b222602954b986832e77355a12 Author: Benjamin Otte <otte@redhat.com> Date: Mon Mar 29 12:58:18 2010 +0200 glx: Add getters for device's Display and GLXContext src/cairo-gl.h | 6 ++++++ src/cairo-glx-context.c | 30 ++++++++++++++++++++++++++++++ 2 files changed, 36 insertions(+) commit 712919223d08f8b4c43f828322fdc285560c137f Author: Benjamin Otte <otte@redhat.com> Date: Fri Mar 26 15:50:10 2010 +0100 gl: Make code safe for multithreaded access The code callss the gl device's acquire/release in cairo_device_acquire/release(). This way, external APIs can use these functions to prepare for rendering GL. Also adds code to unset the glx context if it wasn't set when acquiring the device. This allows multithreaded apps to work fine with just using cairo APIs, but might introduce penalties in single-threaded applications. src/cairo-egl-context.c | 2 ++ src/cairo-gl-private.h | 5 ++++- src/cairo-gl-surface.c | 20 +++++++++++++++++++- src/cairo-glx-context.c | 37 +++++++++++++++++++++++++++++++++++-- 4 files changed, 60 insertions(+), 4 deletions(-) commit 8f2e82cea3405fa9a6591911c0d42ea3dd0c6180 Author: Benjamin Otte <otte@redhat.com> Date: Thu Apr 15 17:48:24 2010 +0200 xcb: Make code not complain about every new format src/cairo-xcb-surface-render.c | 46 +++++++++++++++++------------------------- 1 file changed, 18 insertions(+), 28 deletions(-) commit 1d753b2f4d724c5ac3e452848ca4dda5891816a6 Author: Benjamin Otte <otte@redhat.com> Date: Tue Feb 23 21:53:31 2010 +0100 xlib: Make code not complain about every new format src/cairo-xlib-surface.c | 46 ++++++++++++++++++---------------------------- 1 file changed, 18 insertions(+), 28 deletions(-) commit ebc9cc46fe7740c6d97e02edc66bae383413764b Author: Benjamin Otte <otte@redhat.com> Date: Fri Mar 5 15:08:34 2010 +0100 gl: Use right format The internal format should always be GL_RGBA. src/cairo-gl-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 4142f0836d6719dd136a5fc6ac1adaddd681a3f2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Apr 15 09:09:31 2010 +0100 gl: Handle component-alpha ADD A simple extension to the component-alpha OVER [as ca-ADD is just the second pass in ca-OVER] to support amalgamation of glyph masks. This speeds up firefox-36 by ~5% on g45. Before: firefox-36-20090611 76.185 After: firefox-36-20090611 72.863 src/cairo-gl-surface.c | 16 +++++++++------- 1 file changed, 9 insertions(+), 7 deletions(-) commit a4bae1956bee0be98a5a22bd82d417192776e7f0 Author: Benjamin Otte <otte@redhat.com> Date: Wed Apr 14 22:46:34 2010 +0200 gl: Really don't write error status to the inert object. src/cairo-gl-surface.c | 3 +++ 1 file changed, 3 insertions(+) commit 5fed41ee2bb3097c1446c1cf2038c912d5932692 Author: Benjamin Otte <otte@redhat.com> Date: Wed Apr 14 22:43:29 2010 +0200 test: Add test checking that all setters properly check surface->status In particular, make sure that the setters when called on a const nil surface don't try to set surface->status. test/Makefile.sources | 1 + test/error-setters.c | 109 ++++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 110 insertions(+) commit 6826f020014fff566678a1ff92014211e2a21d4c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Apr 14 19:33:34 2010 +0100 gl: Don't write error status to the inert object. Reported-by: Benjamin Otte <otte@redhat.com> src/cairo-gl-surface.c | 3 +++ 1 file changed, 3 insertions(+) commit 3814855a649ffca695ce31d97c291496e32d2e26 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Apr 14 20:22:14 2010 +0100 script: Flush prior to modifying the image data. util/cairo-script/cairo-script-operators.c | 1 + 1 file changed, 1 insertion(+) commit 03f7bfc351c27ab9d0b764a21bdcf2304e8c2e81 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Apr 14 20:14:39 2010 +0100 rtree: Remove the false assert upon collapse. src/cairo-rtree.c | 2 -- 1 file changed, 2 deletions(-) commit b0383d0cd2b40185b1676d9c88d3647cecc1d7fd Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Apr 14 19:33:34 2010 +0100 svg: Don't write error status to the inert object. Reported-by: Benjamin Otte <otte@redhat.com> src/cairo-svg-surface.c | 37 +++++++++++++++++++++---------------- 1 file changed, 21 insertions(+), 16 deletions(-) commit 5274b63adb9ad808f94e6ea04707991ec1192694 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Apr 14 19:33:34 2010 +0100 ps: Don't write error status to the inert object. Reported-by: Benjamin Otte <otte@redhat.com> src/cairo-pdf-surface.c | 2 -- src/cairo-ps-surface.c | 74 ++++++++++++++++++------------------------------- 2 files changed, 27 insertions(+), 49 deletions(-) commit 45d23b8a791d5845025a1a4d93f99030ec9e0115 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Apr 14 19:33:34 2010 +0100 pdf: Don't write error status to the inert object. Reported-by: Benjamin Otte <otte@redhat.com> src/cairo-pdf-surface.c | 41 ++++++++++++++++++++++------------------- 1 file changed, 22 insertions(+), 19 deletions(-) commit 4351304b5110200b77e6851995cf533a929ba17e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Apr 14 19:29:07 2010 +0100 Win32 features update. boilerplate/Makefile.win32.features | 8 ++++++++ build/Makefile.win32.features-h | 1 + 2 files changed, 9 insertions(+) commit bd17b898b1c249aa0dc443bc303bad4fea8690bc Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Apr 14 19:11:15 2010 +0100 surface: Propagate the error status to _cairo_surface_get_device() If the surface is in error, the appropriate action is report the error using an inert error cairo_device_t. Reported-by: Benjamin Otte <otte@redhat.com> src/cairo-surface.c | 3 +++ 1 file changed, 3 insertions(+) commit 005596907fc9b62fa4bf72ec35e0d1a1a242ef93 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Apr 14 19:05:09 2010 +0100 xlib: Don't operate on an error object inside setters. Fixes: https://bugzilla.gnome.org/show_bug.cgi?id=599574 The backend API manipulated the surface prior to checking whether it was an inert error surface - and in the event of an error surface, tried to overwrite it's error status. src/Makefile.win32.features | 8 ++++++++ src/cairo-xcb-surface.c | 3 +++ src/cairo-xlib-surface.c | 6 ++++++ src/cairo-xlib-xcb-surface.c | 6 ++++++ 4 files changed, 23 insertions(+) commit 8afd4e4c3db53159d1e5b6f13d3355cb5fb1750b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Apr 13 17:52:19 2010 +0100 hull: _cairo_hull_vertex_compare() return 0 for identical elements Andreas Falkenhahn reported a bizarre situation with some implementations of qsort that actually compare the same elements and require the comparator to return 0. So be it. Reported-by: Andreas Falkenhahn <andreas@airsoftsoftwair.de> src/cairo-hull.c | 7 +++++++ 1 file changed, 7 insertions(+) commit d95037db9915033ef1eee24c2fc05e8a95af5457 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Apr 11 21:04:29 2010 +0100 cache: Tidy _cairo_cache_shrink_to_accommodate() There is no need to shrink the cache if we add an entry of size 0, so don't by moving the guards in _cairo_cache_shrink_to_accommodate() to the callers. src/cairo-cache.c | 10 +++------- 1 file changed, 3 insertions(+), 7 deletions(-) commit e6309c6307179388c5de938bffdb44b83b694f28 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Apr 11 21:03:00 2010 +0100 xcb: Use normal finish to decouple from surface cache. Hook into the standard finishing process for a more robust cache removal mechanism. firefox was able to trigger some double free asserts otherwise. src/cairo-xcb-screen.c | 4 ++- src/cairo-xcb-surface-render.c | 57 ++++++++++++++++++++---------------------- 2 files changed, 30 insertions(+), 31 deletions(-) commit e425c44e9c4c1796ccc3557a0368cf3b3a685a31 Author: Andrea Canciani <ranma42@gmail.com> Date: Fri Apr 9 08:32:22 2010 +0200 boilerplate: Compile xlib without xrender cairo_boilerplate_xlib_surface_disable_render() is not defined when compiling without XRender, but it was used nonetheless. Replace it with an empty stub when XRender is not available. boilerplate/cairo-boilerplate-xlib.c | 6 ++++++ 1 file changed, 6 insertions(+) commit 061ea78ad2838450b20d5550a7285823e0d24e1b Author: Andrea Canciani <ranma42@gmail.com> Date: Fri Apr 9 08:28:33 2010 +0200 perf: Correct cairo-perf-diff syntax { cmd-list; } seems to literally be the required syntax. Blanks and ';' are both mandatory. perf/cairo-perf-diff | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 6986970ea5197685274b1fd809d72fbd771c0bab Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Apr 10 17:05:38 2010 +0100 clip: Compile fix for previous clip. PEBKAC. src/cairo-clip.c | 3 --- 1 file changed, 3 deletions(-) commit 0899852c974099da9f8c5e493fa89b8d022646c5 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Apr 10 16:00:45 2010 +0100 clip: Compare the whole clip when testing for equality. Should fix test/clip-contexts src/cairo-clip-private.h | 4 ++++ src/cairo-clip.c | 40 +++++++++++++++++++++++++++++++++++++--- src/cairo-surface-clipper.c | 5 +---- 3 files changed, 42 insertions(+), 7 deletions(-) commit 557016a86a5a4487aeb6ab6392795eb709ee8bb5 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Apr 10 15:59:13 2010 +0100 test: Add clip-contexts This should exercise a bug found by Jeff Muizelaar that cairo-surface-clipper was mistakenly thinking that clip operations on a second context was a no-op as the topmost clip path matched that of the previous context. test/Makefile.am | 1 + test/Makefile.sources | 1 + test/clip-contexts.c | 73 +++++++++++++++++++++++++++++++++++++++++++++ test/clip-contexts.ref.png | Bin 0 -> 98 bytes 4 files changed, 75 insertions(+) commit a9f506493371ac91494488e5ca38e57ceb5a10cb Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Apr 7 22:44:02 2010 +0100 script: Another RGB16_565 warning. src/cairo-script-surface.c | 6 ++++++ 1 file changed, 6 insertions(+) commit 42b5cac7668625c9761113ff72b47af5cfd10377 Author: Adrian Johnson <ajohnson@redneon.com> Date: Wed Apr 7 19:10:15 2010 +0930 PDF-operators: ensure text operations flushed before emitting clip src/cairo-pdf-operators.c | 6 ++++++ 1 file changed, 6 insertions(+) commit c004800fae362ae907ef9d1f822d02c598572161 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Apr 7 22:46:38 2010 +0100 script: Compile without FT_FONT src/cairo-script-surface.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) commit bdf8f50adae67498d5fe36727ed2cd34c0206509 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Apr 7 22:44:02 2010 +0100 script: Another RGB16_565 warning. src/cairo-script-surface.c | 6 ++++++ 1 file changed, 6 insertions(+) commit f5f64317874a3faa4f47688848a499b1e8836cf6 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Apr 7 21:57:22 2010 +0100 perf: Switch between micro and trace benchmarks for cairo-perf-diff perf/cairo-perf-diff | 10 ++++------ 1 file changed, 4 insertions(+), 6 deletions(-) commit 70ca0040fda77b6865c20c94c1b7deaa6e63c481 Author: Andrea Canciani <ranma42@gmail.com> Date: Wed Apr 7 22:54:57 2010 +0200 Actually check for memory allocation failure Clang static analyzer signals "Dereference of null pointer" in case the malloc returns NULL, as the code is checking the wrong pointer. src/cairo-boxes.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit d7f71ac5c209376bacb45199a0eaa3b1937172e9 Author: Benjamin Otte <otte@redhat.com> Date: Wed Apr 7 21:57:38 2010 +0200 region: Make the 2nd argument to intersect and union const src/cairo-region.c | 4 ++-- src/cairo.h | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) commit ded3a9639b33b1e637e99250a463bc227c9e60f4 Author: Kristian Rietveld <kris@gtk.org> Date: Tue Dec 29 08:47:09 2009 +0100 quartz: Define cairo_quartz_float_t and use instead of float On Mac OS 10.6 and newer cairo_quartz_float_t is defined to be a CGFloat, resolving issues with the 64-bit builds. Fixes https://bugs.freedesktop.org/show_bug.cgi?id=25814 src/cairo-quartz-private.h | 6 ++++++ src/cairo-quartz-surface.c | 40 +++++++++++++++++++++------------------- 2 files changed, 27 insertions(+), 19 deletions(-) commit 7c2c3f621f9b009b97c5f269bf313be926cc51c7 Author: Andrea Canciani <ranma42@gmail.com> Date: Sat Apr 3 22:54:57 2010 +0200 Correct masking optimization In _cairo_surface_mask() there was an invaild optimization (it failed for surfaces without alpha content). Using _cairo_pattern_is_clear() is the correct way to evaluate if a pattern is clear. Fixes clear-source src/cairo-surface.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) commit 24fc36c10f5c1c34e3059a8334178743052d7ad2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Apr 2 08:44:45 2010 +0100 script: Fix endian encoding of '<|' length. Andrea found and fixed (and updated all the traces!) an endian bug where we were encoding a 32bit length inside the compressed string stream. However, this one inside the script backed escaped his notice. src/cairo-script-surface.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) commit 697094488c652ad351301696bba9a384cc70002c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Apr 2 08:40:28 2010 +0100 test: Add inverse text This is exercising a bug that I have introduced either in the xcb backend or in the driver in my experimental branches. So simple, yet so wrong. test/Makefile.am | 1 + test/Makefile.sources | 1 + test/inverse-text.c | 67 ++++++++++++++++++++++++++++++++++++++ test/inverse-text.image16.ref.png | Bin 0 -> 1863 bytes test/inverse-text.ps.ref.png | Bin 0 -> 776 bytes test/inverse-text.ref.png | Bin 0 -> 2150 bytes 6 files changed, 69 insertions(+) commit fe610d559c70ab67f782b122437908ac80788611 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Mar 31 09:33:30 2010 +0100 perf/micro: Show megapixel fill rate during paint. perf/micro/paint-with-alpha.c | 9 ++++++++- perf/micro/paint.c | 8 +++++++- 2 files changed, 15 insertions(+), 2 deletions(-) commit 8f7078eba71dcd2aa70bf97c03ed7b4341260143 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Apr 1 20:42:44 2010 +0100 boilerplate: Allox xlib-fallback to build without RENDER This surface doesn't use RENDER and should produce the same results when cairo-xlib is compiled without any support for RENDER. boilerplate/cairo-boilerplate-xlib.c | 72 ++---------------------------------- 1 file changed, 4 insertions(+), 68 deletions(-) commit d209b40c64bddeb0d830678bc2f84e1016fed879 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Apr 1 20:36:15 2010 +0100 xlib: Rearrange xrender stubs to restore compilation without xrender.h boilerplate/Makefile.win32.features | 8 -------- build/Makefile.win32.features-h | 1 - src/Makefile.win32.features | 8 -------- src/cairo-xlib-xrender-private.h | 14 +++++++------- 4 files changed, 7 insertions(+), 24 deletions(-) commit db4dbb38340e93da616319f00e0f4327d1c1d815 Author: Behdad Esfahbod <behdad@behdad.org> Date: Thu Apr 1 01:53:22 2010 -0400 Improve doc syntax check to allow inline macro docs src/check-doc-syntax.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 00101fa71614fb48de436d6618e26353cf44ef95 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Mar 31 16:19:02 2010 +0100 configure: check -lrt for shm_open() OS/X includes shm_open() in libc, and fails to link if -lrt is specified. So perform the appropriate configure time magic. configure.ac | 5 +++++ util/cairo-sphinx/Makefile.am | 2 +- 2 files changed, 6 insertions(+), 1 deletion(-) commit a81d347c2484e60ceec1d9dc81e1b88724496eaf Author: Andrea Canciani <ranma42@gmail.com> Date: Fri Feb 5 22:34:51 2010 +0100 perf: Fix timestamp computation on MacOS X The OIL routines don't work as expected on MacOS X. The operating system gives access to the timestamp counter through the function mach_absolute_time. Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> perf/cairo-perf-posix.c | 7 +++++++ 1 file changed, 7 insertions(+) commit 55ce0b774869e9fa614c6f9fec7905f7205cb5ff Author: Andrea Canciani <ranma42@gmail.com> Date: Fri Feb 5 22:30:05 2010 +0100 script: Fix script scanner endianness The script interpreter was reading the length of compressed data as an host-endian uint32_t, thus making cairo-script able to correctly read traces that were produced on the same endianness as the one they ran upon, but unsuitable for portabile cairo-scripts. Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> util/cairo-script/cairo-script-scanner.c | 30 +++++++++++++++--------------- 1 file changed, 15 insertions(+), 15 deletions(-) commit 4f617eaf77540ba2140086bd5a19fe6d62503d62 Author: Andrea Canciani <ranma42@gmail.com> Date: Fri Feb 5 22:27:28 2010 +0100 trace: Fix trace endianness Trace files were using host-endian to represent the length of compressed data, making the trace format not portable. Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> util/cairo-trace/trace.c | 15 +++++++++++++++ 1 file changed, 15 insertions(+) commit 35f19bc084792bbad42b86a399103ebfbf407d05 Author: Andrea Canciani <ranma42@gmail.com> Date: Wed Mar 31 12:18:01 2010 +0200 pattern: Fix _gradient_is_opaque() for zero stops _gradient_is_opaque() previously returned TRUE for gradient with no stops, triggering a false optimization in _cairo_gstate_mask(). Fixes test/gradient-zero-stops-mask Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> src/cairo-pattern.c | 3 +++ 1 file changed, 3 insertions(+) commit 01e182194224e0eb7493b2ae3c8367ed455cfd54 Author: Andrea Canciani <ranma42@gmail.com> Date: Wed Mar 31 12:14:25 2010 +0200 test: Add gradient-zero-stops-mask Add a test to check that gradient with no stops are not considered opaque by gstate (and thus masking with them is not optimized to paint). test/Makefile.am | 2 + test/Makefile.sources | 1 + test/gradient-zero-stops-mask.argb32.ref.png | Bin 0 -> 86 bytes test/gradient-zero-stops-mask.c | 59 +++++++++++++++++++++++++++ test/gradient-zero-stops-mask.rgb24.ref.png | Bin 0 -> 86 bytes 5 files changed, 62 insertions(+) commit ce3ad6f41edf86ed6914f4d7f364111eba42ca65 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Mar 31 08:21:17 2010 +0100 xlib: Avoid multiple evaluations inside bswap_*() macro. Fixes: Fonts are not readable if remote display to a machine with different endian. https://bugzilla.mozilla.org/show_bug.cgi?id=526977 Reported-and-tested-by: Ginn Chen <Ginn.Chen@Sun.COM> src/cairo-xlib-surface.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit be4ffa9df275513de3175415ee889b7323499a37 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Mar 28 19:40:54 2010 +0100 doc: Fix some missing '%' in gtk-doc comments src/cairo-surface.c | 4 ++-- src/cairo-svg-surface.c | 10 +++++----- 2 files changed, 7 insertions(+), 7 deletions(-) commit 7ab350378e8597e9872dbe390b8454d0a63bff28 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Mar 28 19:38:24 2010 +0100 Silence enumeration warnings following addition of RGB16_565 src/cairo-debug.c | 3 +++ src/cairo-gl-glyphs.c | 2 ++ src/cairo-png.c | 1 + src/cairo-qt-surface.cpp | 2 ++ src/cairo-scaled-font.c | 1 + src/cairo-script-surface.c | 15 +++++++++++++ src/cairo-xcb-surface-render.c | 5 +++++ src/cairo-xlib-surface.c | 5 +++++ src/cairo-xml-surface.c | 1 + src/drm/cairo-drm-i915-surface.c | 3 +++ src/drm/cairo-drm-i965-shader.c | 4 ++++ src/drm/cairo-drm-i965-surface.c | 1 + src/drm/cairo-drm-intel-surface.c | 1 + src/drm/cairo-drm-intel.c | 12 ++++++++-- src/drm/cairo-drm-radeon-surface.c | 1 + util/cairo-script/cairo-script-operators.c | 35 ++++++++++++++++++++++++++++++ util/cairo-sphinx/sphinx.c | 1 + util/cairo-trace/trace.c | 25 ++++++++++++++++++--- 18 files changed, 113 insertions(+), 5 deletions(-) commit d901692a5550c51fd3eefd307609fb800ef02a95 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Mar 30 15:31:32 2010 +0100 scaled-glyph: Allow removal info fields. Pushed a fraction of a second too soon... Update the scaled_glyph->has_info if the backend clears any field as well. src/cairo-scaled-font.c | 19 ++++++++++++++++--- 1 file changed, 16 insertions(+), 3 deletions(-) commit 5c4b6d520bc7ce7d5193a89bbda44b91f5eb0d4d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Mar 30 15:26:42 2010 +0100 xcb: Cache whether we have already check a glyph for size. Avoid repeated cairo_scaled_glyph_lookup() and checking of sizes if we have already seen this glyph in this run. src/cairo-xcb-surface-render.c | 10 ++++++++++ 1 file changed, 10 insertions(+) commit 1c76e3e9ad1437662bf91ed1d21c4bda8fd50fb4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Mar 30 15:24:08 2010 +0100 scaled-font: Optimize cairo_scaled_font_lookup() By tracking which fields of information are already available on the scaled_glyph we can more efficiently determine if we already have the requested fields. This reduces from about 6 conditionals to one, and reduces the function overhead by ~20% -- which has a measurable improvement on glyph benchmarks. src/cairo-scaled-font.c | 73 +++++++++++++---------------------------------- src/cairo-types-private.h | 1 + 2 files changed, 21 insertions(+), 53 deletions(-) commit d2ec151e490b227e7f3d6879bf7a893577dfefa5 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Mar 30 15:23:18 2010 +0100 perf/micro: Exercise different sizes and antialising modes with glyphs perf/micro/glyphs.c | 85 ++++++++++++++++++++++++++++++++++++++++++++++++----- 1 file changed, 77 insertions(+), 8 deletions(-) commit 72481acf0d1e8648c5d03ce51f84c639132475a9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Mar 30 04:10:28 2010 +0100 perf: Compute ops per second Provide a hook for the test to be able to compute the number of ops per second. For instance, the glyphs test uses it to report the number of kiloglyph per second Cairo is able to render. perf/cairo-perf-micro.c | 47 ++++++++++++++++++++++++-------------- perf/cairo-perf.h | 9 ++++++-- perf/micro/box-outline.c | 4 ++-- perf/micro/cairo-perf-cover.c | 5 ++-- perf/micro/composite-checker.c | 2 +- perf/micro/dragon.c | 10 ++++---- perf/micro/fill.c | 6 ++--- perf/micro/glyphs.c | 35 ++++++++++++++++++++++++++-- perf/micro/intersections.c | 8 +++---- perf/micro/long-dashed-lines.c | 2 +- perf/micro/long-lines.c | 8 +++---- perf/micro/mask.c | 18 +++++++-------- perf/micro/mosaic.c | 8 +++---- perf/micro/paint-with-alpha.c | 2 +- perf/micro/paint.c | 2 +- perf/micro/pattern_create_radial.c | 2 +- perf/micro/pythagoras-tree.c | 2 +- perf/micro/rectangles.c | 6 ++--- perf/micro/rounded-rectangles.c | 6 ++--- perf/micro/spiral.c | 28 +++++++++++------------ perf/micro/stroke.c | 4 ++-- perf/micro/subimage_copy.c | 2 +- perf/micro/tessellate.c | 6 ++--- perf/micro/text.c | 2 +- perf/micro/twin.c | 2 +- perf/micro/unaligned-clip.c | 2 +- perf/micro/world-map.c | 2 +- perf/micro/zrusin.c | 4 ++-- 28 files changed, 142 insertions(+), 92 deletions(-) commit 2a98d0586c19fbb2b555f471895d73f253c4943b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Mar 15 18:23:23 2010 +0000 drm/i915: Avoid using another unpreserved temporary As we may load from a sampler, we can not use on of the unpreserved temporaries to store the intermediate colour result. src/drm/cairo-drm-i915-shader.c | 26 +++++++++++++++----------- 1 file changed, 15 insertions(+), 11 deletions(-) commit 26f963557bf065fd25b9c4f6652a252735a0fb74 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Mar 15 18:22:06 2010 +0000 drm/i915: Normalize spans opacity value On the generic path I forgot to rescale the alpha value into [0,1]. src/drm/cairo-drm-i915-spans.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) commit 97f8c20727eaeb2dfddb0d4796192cc042eb14fa Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Mar 27 21:52:16 2010 +0000 boilerplate: Create an image16 target In order to exercise the newly restored r5g6g5 support, we need to create an appropriate surface and feed it through the test and performance suites. boilerplate/cairo-boilerplate-directfb.c | 6 ++- boilerplate/cairo-boilerplate-drm.c | 6 ++- boilerplate/cairo-boilerplate-gl.c | 12 +++-- boilerplate/cairo-boilerplate-pdf.c | 4 +- boilerplate/cairo-boilerplate-ps.c | 8 +-- boilerplate/cairo-boilerplate-quartz.c | 3 ++ boilerplate/cairo-boilerplate-script.c | 2 +- boilerplate/cairo-boilerplate-skia.c | 6 ++- boilerplate/cairo-boilerplate-svg.c | 8 +-- boilerplate/cairo-boilerplate-test-surfaces.c | 21 ++++---- boilerplate/cairo-boilerplate-vg.c | 12 +++-- boilerplate/cairo-boilerplate-win32-printing.c | 4 +- boilerplate/cairo-boilerplate-win32.c | 6 ++- boilerplate/cairo-boilerplate-xcb.c | 21 +++++--- boilerplate/cairo-boilerplate-xlib.c | 12 +++-- boilerplate/cairo-boilerplate.c | 39 +++++++++++++-- boilerplate/cairo-boilerplate.h | 1 + perf/cairo-perf-micro.c | 59 +---------------------- perf/cairo-perf-trace.c | 67 +------------------------- 19 files changed, 120 insertions(+), 177 deletions(-) commit 022291be1cbddf4f6722f0bf76ebda6922780276 Author: Oleg Romashin <romaxa@gmail.com> Date: Wed Mar 24 13:12:25 2010 -0400 Add back support for CAIRO_FORMAT_RGB16_565 Due to slow memory speed, 16bpp color depth is still very popular on mobile devices. Maemo5 is also using 16bpp color depth. src/cairo-deprecated.h | 14 -------------- src/cairo-image-surface.c | 16 +++++++++++++++- src/cairo-xlib-display.c | 28 +++++++++++++++++++++++++--- src/cairo.h | 7 ++----- src/cairoint.h | 2 +- 5 files changed, 43 insertions(+), 24 deletions(-) commit 36e12b1952f9bdc52491c5a9bdf8ce155865c362 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Mar 24 12:14:20 2010 +0000 subsurface: Ignore return from _cairo_clip_init_copy() The return is just a hint that the clip is empty, however we are about to immediately apply the surface clip and so do not care for an 'optimised' path. src/cairo-surface-subsurface.c | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-) commit e356e5d18b923d9b0ff83e9460582710f9e515d9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Mar 24 11:38:19 2010 +0000 recording: Fix leak of clip from snapshots of recording surfaces. src/cairo-recording-surface.c | 1 + 1 file changed, 1 insertion(+) commit 5b4885f693c0d800da7160770b341166e3bdea61 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Mar 24 11:25:29 2010 +0000 clip: propagate the no-clip through the copy. src/cairo-clip.c | 1 + 1 file changed, 1 insertion(+) commit b3e0393161904c2fea62b0198dea551a77201c19 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Mar 24 11:21:36 2010 +0000 clip: Don't reduce all-clip to no-clip. src/cairo-clip.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit b12114f296e4e8db3207ab593a431fd0ca2ee67c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Mar 24 11:08:38 2010 +0000 snapshot: Finish the clone. Kill ref cycles from snapshots patterns by explicitly calling finish on the cloned surface. src/cairo-surface-snapshot.c | 1 + 1 file changed, 1 insertion(+) commit db913d731e50505892697bdc919f9fe146eea6ab Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Mar 24 10:16:58 2010 +0000 xcb: Ensure that we own the glyph cache before use. Sigh. One day, I will enable multiple glyph caches to be attached to a scaled font, but for today, simply avoid the crash. src/cairo-xcb-surface-render.c | 38 +++++++++++++++++++++++++++++--------- 1 file changed, 29 insertions(+), 9 deletions(-) commit 6da3cea3564bd8c5de37bf2244a2dd656202e4ec Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Mar 23 20:53:11 2010 +0000 scaled-font: Destroy the old surface when replacing scaled_glyph->recording src/cairo-scaled-font.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) commit 548092fe8c9af1423a10e9566cbc4315d2f28efc Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Mar 23 20:08:04 2010 +0000 script: Free the surface/font bitmaps. src/cairo-script-surface.c | 13 +++++++++++++ 1 file changed, 13 insertions(+) commit 83d1bd9f37da93fbdc586788b6891d0eccdb7cee Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Mar 23 20:04:10 2010 +0000 script: Manually unlink font entries upon destruction As the device is already finished, we can not lock it without raising an error, so we have to open code the destruction of the font entries. Fortunately we can make several simplifying assumptions about the required cleanup as we know the device is also being destroyed. src/cairo-script-surface.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) commit cccf6753ab68b0795351da2626f9e4ecd60c2a2e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Mar 23 19:45:35 2010 +0000 script: More acquire device fixes. We also need to acquire the device upon finish, similar surface creation and the pagination functions, i.e. the other times outside of the drawing ops that must modify the shared context/device. src/cairo-script-surface.c | 69 ++++++++++++++++++++++++++++++++++++---------- 1 file changed, 54 insertions(+), 15 deletions(-) commit 25a77b263d170265a9acf1697793cbbfa07dd852 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Mar 23 16:49:21 2010 +0000 boilerplate: Cleanup the list of backends upon shutdown. boilerplate/cairo-boilerplate.c | 13 +++++++++++++ boilerplate/cairo-boilerplate.h | 3 +++ perf/cairo-perf-micro.c | 2 ++ perf/cairo-perf-trace.c | 2 ++ test/cairo-test.c | 2 ++ 5 files changed, 22 insertions(+) commit 1ddcd5cf31bb47e9ff18ddf94d0a4648fa70a617 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Mar 23 16:43:39 2010 +0000 clip: Remove the redundant _cairo_clip_init_rectangle() As _cairo_clip_init_rectangle() is equivalent and more importantly more clearly written as: _cairo_clip_init(&clip); if (status = _cairo_clip_rectangle(&clip, &rect)) { _cairo_clip_fini(&fini); return status; } perform the transformation and in the process catch a few mistakes along error paths. src/cairo-clip-private.h | 4 ---- src/cairo-clip.c | 30 +++++++++++------------------- src/cairo-paginated-surface.c | 18 +++++++++--------- src/cairo-recording-surface.c | 11 +++++------ src/cairo-surface-wrapper.c | 14 +++++--------- 5 files changed, 30 insertions(+), 47 deletions(-) commit 61ad28fe7d334c63197ae3881d5edd074d63cfec Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Mar 23 10:44:16 2010 +0000 device: Add language binding interfaces. Damien Carbonne reported that cairo_device_t lacked the language binding hooks normally associated with cairo objects. So add the missing get_reference_count, get_user_data and set_user_data. src/cairo-device-private.h | 1 + src/cairo-device.c | 78 ++++++++++++++++++++++++++++++++++++++++++++++ src/cairo.h | 14 +++++++++ 3 files changed, 93 insertions(+) commit 8a8c2f6c282c1822dc1a638c2258c8449b1d678b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Mar 23 10:34:06 2010 +0000 cairo: Typos in docs. A couple of typos reported by Damien Carbonne. src/cairo-recording-surface.c | 10 ++++------ src/cairo.h | 4 ++-- 2 files changed, 6 insertions(+), 8 deletions(-) commit 4c55c87478a2595569f9c1f13657c3d309f31407 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Mar 23 10:05:35 2010 +0000 image: Free traps after conversion from boxes. src/cairo-image-surface.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) commit 07f7bddc30157fed8f9dced00fef44a2307b4b01 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Mar 23 10:05:35 2010 +0000 xcb: Free traps after conversion from boxes. src/cairo-xcb-surface-render.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) commit 181403fb534d0216123043bcd3ee6cff60e1e6fd Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Mar 23 09:38:59 2010 +0000 test/clear-source: Free source after use. test/clear-source.c | 38 ++++++++++++++++++++++---------------- 1 file changed, 22 insertions(+), 16 deletions(-) commit 35432e37c264e1ace764b4a1393d9b8579eb52c0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Mar 23 09:35:51 2010 +0000 xcb: Destroy reference to clip surface after use. src/cairo-xcb-surface-render.c | 3 +++ 1 file changed, 3 insertions(+) commit 98e3dfbb899f0228fc9987a56af93b012989a27a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Mar 23 09:35:31 2010 +0000 xcb: Relinquish the xcb connection on finish. src/cairo-xcb-connection.c | 7 +++++++ 1 file changed, 7 insertions(+) commit 638cae3bdeaf0b10d1fd59a519f5d7a05c5b179c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Mar 23 08:25:13 2010 +0000 xcb: Fix leak of clip rectangle during show-glyphs. src/cairo-xcb-surface-render.c | 15 ++++++--------- 1 file changed, 6 insertions(+), 9 deletions(-) commit 1d4ad787747b5c92c3b062afde5b98c72ac4cc95 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Mar 23 08:20:49 2010 +0000 test/pthread-same-source: Free source after use. test/pthread-same-source.c | 6 ++++++ 1 file changed, 6 insertions(+) commit bfc027ac6d05f489d0d26110c225d2871be0971c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Mar 23 08:16:57 2010 +0000 test/clip-image: Free image after use. test/clip-image.c | 1 + 1 file changed, 1 insertion(+) commit e214f09d633093ce9b2ca0bffce10bc68a6e30b0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Mar 22 22:20:58 2010 +0000 xcb: Consume the implicit reference for the cached connection on finish src/cairo-xcb-connection.c | 5 +++++ 1 file changed, 5 insertions(+) commit c235543bdfd8b1495cb7ae8f2a82b6267c4d9b84 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Mar 22 22:15:44 2010 +0000 xcb: Destroy reference to local source picture in show-glyphs. src/cairo-xcb-surface-render.c | 14 +++++++++++--- 1 file changed, 11 insertions(+), 3 deletions(-) commit c42cdd2c9d883ef359ac57b65eba4ed15441181d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Mar 22 20:14:09 2010 +0000 xcb: Gracefully destroy any lingering fallback during finish. A fallback should never persist beyond the flush into the finish, but yet one remains in test/clip-shapes-unaligned-rectangles. For the time been, simply clean up the rogue surface. src/cairo-xcb-surface.c | 19 ++++++++++++------- 1 file changed, 12 insertions(+), 7 deletions(-) commit fd96aa3de2218dcc6671636f35a24738e3cae996 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Mar 22 19:07:31 2010 +0000 gl: Hook in glyph cache finalisation. src/cairo-gl-glyphs.c | 23 ++++++++++++----------- src/cairo-gl-private.h | 7 ++++--- src/cairo-gl-surface.c | 4 ++++ 3 files changed, 20 insertions(+), 14 deletions(-) commit c7511fa7ad932675c8cebba504d3c538ead3c44c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Mar 22 17:21:38 2010 +0000 drm/intel: Revoke glyph caches on device finish src/drm/cairo-drm-i915-glyphs.c | 2 +- src/drm/cairo-drm-i965-glyphs.c | 2 +- src/drm/cairo-drm-intel-private.h | 1 + src/drm/cairo-drm-intel.c | 18 ++++++++++++------ 4 files changed, 15 insertions(+), 8 deletions(-) commit 9ba73bdfc451d205f411e3dbc8073ae10d58b4f3 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Mar 22 17:17:08 2010 +0000 gl: Decouple cache from scaled-font on context destruction src/cairo-gl-glyphs.c | 8 +++++++- src/cairo-gl-private.h | 4 ++++ src/cairo-gl-surface.c | 14 +++++++++++++- src/cairo-scaled-font-private.h | 2 ++ src/cairo-scaled-font.c | 1 + 5 files changed, 27 insertions(+), 2 deletions(-) commit 7812d095ab7657400be470af8fdc7f20eaedd6c5 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Mar 22 16:46:58 2010 +0000 drm/intel: Replace open-coding of _cairo_rtree_node_remove() src/drm/cairo-drm-intel.c | 17 +++-------------- 1 file changed, 3 insertions(+), 14 deletions(-) commit 48d847162558dc09388942a96ff443c253068d68 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Mar 22 13:05:23 2010 +0000 recording: Cleanup the clip on error. Whilst recording the commands, remember to cleanup the local clip after detecting an error. src/cairo-recording-surface.c | 5 +++++ src/cairo-surface-snapshot.c | 2 ++ 2 files changed, 7 insertions(+) commit c87737c2bf983a8555b08c526050423a5eab5642 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Mar 22 13:04:19 2010 +0000 ps: Skip surface emission if in error state. During finish, if we encountered an error during the construction of the surface, simply proceed to cleanup rather than emit the incomplete and potentially subtly broken output. src/cairo-ps-surface.c | 4 ++++ 1 file changed, 4 insertions(+) commit e961cdf568f4551bbd6bca6b7505fc9f9b0805f9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Mar 22 10:48:48 2010 +0000 image: Special case wholly unbounded fixups. In the event of an empty bounded rectangle, the computation of the unbounded - bounded rectangles leads to negative areas, integer overflow and death. [And similarly for the derived surfaces.] src/cairo-image-surface.c | 24 +++++++++++ src/cairo-xcb-surface-render.c | 62 +++++++++++++++------------ src/drm/cairo-drm-i915-surface.c | 62 +++++++++++++++------------ src/drm/cairo-drm-i965-surface.c | 93 +++++++++++++++++++++------------------- 4 files changed, 144 insertions(+), 97 deletions(-) commit 844d8ea57d69c9a68fbec64f4438953850f7657f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Mar 22 10:37:18 2010 +0000 surface: Check pattern for error status. Ensure that the error status from patterns is propagated when used by doing so in the surface layer. Similarly check that a surface pattern has not been finished. src/cairo-surface.c | 59 +++++++++++++++++++++++++++++++++++++++++++++++++---- 1 file changed, 55 insertions(+), 4 deletions(-) commit 213093f43770d078dd146d5a10b9a5803097b8a8 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Mar 22 10:36:47 2010 +0000 ps: Free page on error path. src/cairo-ps-surface.c | 1 + 1 file changed, 1 insertion(+) commit c8a8e57d6af7261d70103ddd46f8da409896b5e7 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Mar 21 20:42:34 2010 +0000 script: Use a stack to push/pop recursed line numbers. Still not entirely helpful in the event of recursive parsing without a reference to the file as well as the line number in the event of an exception. util/cairo-script/cairo-script-scanner.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) commit 4b4de940ee6e76ec9677b5468ebf709c6d2db982 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Mar 21 20:41:18 2010 +0000 script: Remove the version check. When compiling we can depend on whatever version of cairo we need, but we should be wary of checking for runtime compatibility when building standalone. util/cairo-script/cairo-script-operators.c | 2 -- 1 file changed, 2 deletions(-) commit a7cb314b060f6ada855ec75b283ef64bf835824a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Mar 21 20:40:54 2010 +0000 test/any2ppm: Check for errors after executing script. test/any2ppm.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) commit b73a33b89e43d63a437035a05cf91df5c3b4d186 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Mar 21 20:40:19 2010 +0000 script: Make the test suite run again! src/cairo-script-surface.c | 194 ++++++++++++++++++++++++--------------------- 1 file changed, 103 insertions(+), 91 deletions(-) commit 2a59f0af6aab5fe0ebb19195d4c018fc5221016c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Mar 21 20:38:27 2010 +0000 wrapper: Apply device transform inverse as appropriate. Fixes many failures of the paginated surfaces as they replayed through the recording surfaces. src/cairo-surface-wrapper.c | 89 ++++++++++++++++----------------------------- 1 file changed, 31 insertions(+), 58 deletions(-) commit 5720fff827e5149b201a9366179db93ce258b120 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Mar 21 20:37:38 2010 +0000 ps: Destroy the local surface on error. Minor leak of the intermediate surface when converting to an opaque source following an error. src/cairo-ps-surface.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) commit 35f318a59cde9c635036cc14c90878400528882d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Mar 21 20:36:26 2010 +0000 paginated: Use common is_clear. As equivalent processing to the page_is_blank optimisation done inside the paginated surface is applied by the gstate, remove the redundant code. src/cairo-paginated-surface-private.h | 1 - src/cairo-paginated-surface.c | 34 +--------------------------------- 2 files changed, 1 insertion(+), 34 deletions(-) commit b101c7dab8cdbf7b9321355a8d2311b2f863f011 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Mar 11 01:48:43 2010 +0000 gl: Add EGL interface Enable the EGL backend for GL. boilerplate/Makefile.win32.features | 8 ++ boilerplate/cairo-boilerplate-gl.c | 98 +++++++++++++++++++++ build/Makefile.win32.features-h | 1 + configure.ac | 52 ++++++----- src/Makefile.sources | 1 + src/Makefile.win32.features | 8 ++ src/cairo-egl-context.c | 168 ++++++++++++++++++++++++++++++++++++ src/cairo-gl.h | 14 +++ 8 files changed, 327 insertions(+), 23 deletions(-) commit ca6e4e1f91e0a90666551d2fa74cfc04484d4e80 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Mar 11 01:48:08 2010 +0000 spans: Initialise rects->is_bounded for polygon compat. src/cairo-spans.c | 2 ++ 1 file changed, 2 insertions(+) commit 6754f1db1716a6570c82e207e21b09a10de120ba Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Mar 10 15:13:00 2010 +0000 gstate: Account for undropped non-transformed glyphs. In the simplest case of pass-through glyphs we made a mistake in propagating the number of glyphs. Reported-by: Christophe de Dinechin <christophe@taodyne.com> src/cairo-gstate.c | 1 + 1 file changed, 1 insertion(+) commit 5fb36fe5ab2a2b30da213557936122a829493906 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Mar 8 15:32:15 2010 +0000 drm/i965: Acquire device around commit. src/drm/cairo-drm-i915-surface.c | 4 +- src/drm/cairo-drm-i965-shader.c | 13 +++-- src/drm/cairo-drm-i965-surface.c | 106 ++++++++++++++++++++++++++------------- 3 files changed, 82 insertions(+), 41 deletions(-) commit 4083f40fbd085dc2039fe62592cf1239373c7fca Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Mar 7 16:59:57 2010 +0000 glew: don't pull in GL/glu.h src/glew/GL/glew.h | 10 ---------- 1 file changed, 10 deletions(-) commit 473e006ad6ef0959c8a7059aba0689a16a937732 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Mar 6 18:12:36 2010 +0000 test: Add a1-mask-sample Test sampling positions with a NEAREST mask. test/Makefile.am | 1 + test/Makefile.sources | 1 + test/a1-mask-sample.c | 83 ++++++++++++++++++++++++++++++++++++++++++++ test/a1-mask-sample.ref.png | Bin 0 -> 148 bytes 4 files changed, 85 insertions(+) commit f79b2ceea4722b5059563be1eb55de617fecc004 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Mar 4 19:52:58 2010 +0000 drm/i915: Discard redundant solitary clips during fill src/drm/cairo-drm-i915-surface.c | 10 ++++++++++ 1 file changed, 10 insertions(+) commit 7cea9ae290ba24e4d2f87cef4228cce5fcb3181b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Mar 4 17:23:55 2010 +0000 drm/i915: Apply more micro-optimisations when targetting CONTENT_ALPHA src/drm/cairo-drm-i915-shader.c | 51 ++++++++++++++++++++++++++++++----------- 1 file changed, 37 insertions(+), 14 deletions(-) commit dce8b028cd5b249054e8197a536c29f6ba4110da Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Mar 4 17:13:53 2010 +0000 drm/i915: Adjust clip matrix for clip origin. src/drm/cairo-drm-i915-shader.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) commit 8e702044b3816002011488c56f1ec9c6b9682eee Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Mar 4 17:13:02 2010 +0000 drm/i915: Handle clip/combine coords for spans. src/drm/cairo-drm-i915-spans.c | 71 +++++++++++++++++++++++++----------------- 1 file changed, 43 insertions(+), 28 deletions(-) commit e1f0c2e73fa8cda9e38ab182288201fa27a3b363 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Mar 4 17:11:58 2010 +0000 perf: Check for and include unistd.h isatty() and access() require unistd.h, so include it! perf/cairo-perf-trace.c | 8 ++++++++ 1 file changed, 8 insertions(+) commit 271752b944ef0898b5d1be57f26de83c8e40ae3a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Mar 4 08:24:08 2010 +0000 gl: Pad the gradient texture. Fixes test/linear-gradient-subset src/cairo-gl-surface.c | 1 + 1 file changed, 1 insertion(+) commit afe6f4f0519606c4bc7e9b705b0cae75692d7af2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Mar 3 23:06:43 2010 +0000 drm/i915: Use a preserved temporary to store the radial texcoord. src/drm/cairo-drm-i915-shader.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) commit 4b2e0478822cb8415be4bac21950dbf03c0f01c8 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Mar 3 21:45:31 2010 +0000 drm/i915: Eliminate redundant radial vertex shader. It is always equivalent to a simple texture transformation, so reduce it to the texture shader. src/drm/cairo-drm-i915-glyphs.c | 13 ------------- src/drm/cairo-drm-i915-private.h | 1 - src/drm/cairo-drm-i915-shader.c | 8 +------- src/drm/cairo-drm-i915-spans.c | 42 ++-------------------------------------- 4 files changed, 3 insertions(+), 61 deletions(-) commit 51610517982a4cf70ca8d396d0ad4c58a477c5e9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Mar 3 21:56:56 2010 +0000 drm/i915: Fix encoding of negated operand channels Long ago when converting the pixel shader structs into macros and reducing the code size by ~100k (the inlines were too depth for constant propagation and CSE), I broke the encoding of negated channels. So instead use a single bit to indicate a negation rather than 2s complement (with sign extension into neighbouring channels, oops). The disadvantage is that expressing the negated channel is a little more ugly. src/drm/cairo-drm-i915-private.h | 22 ++++++++++++++-------- src/drm/cairo-drm-i915-shader.c | 16 ++++++++-------- 2 files changed, 22 insertions(+), 16 deletions(-) commit 5b973c670807bda2bb355c90ea817e2a2f092f0b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Mar 3 21:25:11 2010 +0000 drm/i915: Fix crash with linear-step-function Avoid the potential divide by zero by falling back to pixman rendered gradient textures for such troublesome input. src/drm/cairo-drm-i915-shader.c | 17 +++++++++++------ 1 file changed, 11 insertions(+), 6 deletions(-) commit 404ead7eebafa7ca077e1d012dde0a598b64245a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Mar 3 21:01:49 2010 +0000 drm/i915: Ensure simple alpha is written to green on alpha-only surfaces. The intricacies of h/w make convoluted s/w. src/drm/cairo-drm-i915-shader.c | 20 ++++++++++++++------ 1 file changed, 14 insertions(+), 6 deletions(-) commit 1f2daf652480ef768f847467d9171b5e4a7f01a3 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Mar 3 20:06:08 2010 +0000 drm/i915: Avoid double multiplication of linear alpha. src/drm/cairo-drm-i915-shader.c | 8 ++------ 1 file changed, 2 insertions(+), 6 deletions(-) commit cdaef58c19a191632473fcbbe688e6231c02b9b2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Mar 3 18:52:40 2010 +0000 drm/i915: Acquire device before shader commit. src/drm/cairo-drm-i915-shader.c | 2 + src/drm/cairo-drm-i915-surface.c | 111 ++++++++++++++++++++++++++------------- 2 files changed, 77 insertions(+), 36 deletions(-) commit dd6f96dd4aeb48262d5af8a3006e6404521c151b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Mar 3 16:06:25 2010 +0000 drm/i965: Ignore flushing on device finish. src/drm/cairo-drm-i965-surface.c | 3 +++ 1 file changed, 3 insertions(+) commit 3692597567f62dd300e8ce8c43d43ac449616041 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Mar 3 13:56:48 2010 +0000 drm/i965: Merge the constants into the surface stream. As we can use an offset from the surface base for the constant buffers, we need to fold the constants into the surface stream. src/drm/cairo-drm-i965-private.h | 11 ++----- src/drm/cairo-drm-i965-shader.c | 26 ++++----------- src/drm/cairo-drm-i965-surface.c | 70 +++------------------------------------- 3 files changed, 14 insertions(+), 93 deletions(-) commit cc585947574f6e17d299a5987ee69162f63d5a52 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Mar 3 23:06:23 2010 +0000 drm/intel: Set the 1D gradient texture to be untiled. src/drm/cairo-drm-intel.c | 1 + 1 file changed, 1 insertion(+) commit 543cf073643b259096ee2a95d91b9067395a7e09 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Mar 3 21:19:41 2010 +0000 drm/intel: Set default extend mode for rendering gradient subsets. src/drm/cairo-drm-intel.c | 1 + 1 file changed, 1 insertion(+) commit 521d1c1d2c51bca07db1aed3c7f62d860cece3f4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Mar 3 20:25:54 2010 +0000 surface: Simplify is_clear logic. As we pre-check whether the operation is a no-op on an already cleared surface, then for all but a paint with CLEAR and no clip, the result is a non-clear surface. src/cairo-surface.c | 22 ++++++++-------------- 1 file changed, 8 insertions(+), 14 deletions(-) commit 017db956644fa1da22033653490f362d1e3e47e0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Mar 3 20:06:30 2010 +0000 gstate: Use _cairo_pattern_clear for CLEAR. src/cairo-gstate.c | 100 +++++++++++++++++++++++++++++++++-------------------- 1 file changed, 62 insertions(+), 38 deletions(-) commit 70412bba0e9fd0ac6e7343d7aebe953355207aab Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Mar 3 17:34:39 2010 +0000 test: Avoid derefencing a NULL xcb connection test/xcb-surface-source.c | 23 ++++++++++++++--------- 1 file changed, 14 insertions(+), 9 deletions(-) commit 5bfea8447b1b2333218295f51a9474d061192d49 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Mar 3 21:32:13 2010 +0000 test: Add references for rounded-rectangle-{fill,stroke} test/Makefile.am | 2 ++ test/rounded-rectangle-fill.ref.png | Bin 0 -> 787 bytes test/rounded-rectangle-stroke.ref.png | Bin 0 -> 871 bytes 3 files changed, 2 insertions(+) commit 4126d580d8b9db9217ed17aadcce20b14e77a00b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Mar 2 14:38:36 2010 +0000 surface-fallback: Free traps on composite_trapezoids() error src/cairo-surface-fallback.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit 8a59522bbdc61d5c90f1ae55111b5408865755b3 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Mar 2 14:36:16 2010 +0000 surface-fallback: Propagate NOTHING_TO_DO NOTHING_TO_DO is converted to SUCCESS by the surface layer, so clean up the code slightly by reducing the number of checks and conversions. src/cairo-surface-fallback.c | 15 +++------------ 1 file changed, 3 insertions(+), 12 deletions(-) commit f07195860620959c27d43080a7b987e28222735a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Mar 2 13:57:02 2010 +0000 xlib: Handle a1 image uploads through converter Fixes test/large-source [xlib] src/cairo-xlib-surface.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) commit f979dd22d8f48e6ac7c50372c7d180c3b590dc74 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Mar 2 11:20:11 2010 +0000 image: Don't rely on clip regions being clipped to surface extents. Fixes a crash in test/clip-fill-unbounded [xlib-fallback]. src/cairo-image-surface.c | 32 ++++++++++++++++++++------------ 1 file changed, 20 insertions(+), 12 deletions(-) commit 15d9c5fcd8881b4593ec23633450a4f629b6d42b Author: Alexander Shulgin <alex.shulgin@gmail.com> Date: Sat Feb 20 20:59:32 2010 +0200 Added new MIME type CAIRO_MIME_TYPE_URI and it's support in SVG backend The rationale behind this change is that when someone is trying to draw on a SVG surface using image surface patterns, the resulting SVG file can take up to ridiculous 20 megabytes for 3-4 typical photographic images in a single SVG file. This also can take significant amount of CPU time to complete. The reason for this behaviour is that currently whenever SVG backend needs to emit an image tag for a surface pattern it takes a snapshot of the subject surface, encodes it in PNG, then Base64-encodes and emits the (huge) resulting string. With use of cairo_surface_set_mime_data API this can be somewhat improved by associating JPEG image contents with the corresponding surfaces. Still this doesn't allow for post-processing of involved photographic images without regenerating the SVG file. As SVG specification allows URIs in the image tag's xlink:href attribute, it is possible instead of embedding encoded image data to simply link image files residing physically on the same medium as the generated SVG file: files on disk under common directory, files on a web server at common base URI, etc. To make this happen we add new (unofficial) MIME type "text/x-uri" and let users associate URIs with surfaces through cairo_surface_set_mime_data() API. When SVG backend needs to emit surface contents and it sees "text/x-uri" attached to the surface, it emits this data instead of taking snapshot. The URI data is emitted as is, so correctness check is left solely to the client code. doc/public/cairo-sections.txt | 1 + doc/public/tmpl/cairo-surface.sgml | 7 ++++ src/cairo-surface.c | 16 ++++++++- src/cairo-svg-surface.c | 72 +++++++++++++++++++++++++++++++++++--- src/cairo.h | 1 + 5 files changed, 92 insertions(+), 5 deletions(-) commit d91bab565c54205f8dd7aa8a9de819dc4bc3551c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Mar 2 08:12:52 2010 +0000 test: Remove the double-clip from clip-group-shapes A push/pop is an implicit save/restore, so a clip applied inside the group is temporary, whereas a clip applied outside affects both the geometry inside *and* the application of the group when painted. So reset the clip afterwards to get the desired behaviour. test/clip-group-shapes.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) commit e8c819c5227d29dfdf503022339276eb725a3e97 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Mar 1 10:25:01 2010 +0000 Warnings cleanup for CAIRO_FORMAT_INVALID. A few more fixes required case statements to handle the addition of CAIRO_FORMAT_INVALID. src/cairo-qt-surface.cpp | 2 ++ src/drm/cairo-drm-i915-surface.c | 2 ++ src/drm/cairo-drm-i965-shader.c | 2 ++ src/drm/cairo-drm-i965-surface.c | 1 + src/drm/cairo-drm-intel-surface.c | 1 + src/drm/cairo-drm-intel.c | 3 +++ src/drm/cairo-drm-radeon-surface.c | 1 + 7 files changed, 12 insertions(+) commit 1a7ba1a8061c373cd66d15b3a1c8dfada92745ef Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Mon Mar 1 01:23:17 2010 +0200 xcb: Update minimum required versions of libxcb. The new xcb surface uses xcb_writev() and xcb_take_socket() which were introduced in libxcb 1.1.92. The boilerplate in turn uses the major_code and minor_code fields in xcb_generic_error_t, which were introduced in 1.4. boilerplate/cairo-boilerplate-xcb.c | 6 ++++++ configure.ac | 5 ++++- 2 files changed, 10 insertions(+), 1 deletion(-) commit 5b7f4bb241f3191c1589cd714f373719efded56e Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Sun Feb 28 23:31:06 2010 +0200 api: Introduce CAIRO_FORMAT_INVALID formally in the API. We were exposing the actual value of CAIRO_FORMAT_INVALID through API functions already, so it makes sense to just go ahead and put it in the cairo_format_t enum. boilerplate/cairo-boilerplate.c | 2 +- src/cairo-debug.c | 1 + src/cairo-gl-glyphs.c | 6 ++++++ src/cairo-image-surface.c | 6 +++++- src/cairo-png.c | 1 + src/cairo-scaled-font.c | 1 + src/cairo-script-surface.c | 22 ++++++++++++++-------- src/cairo-win32-surface.c | 6 ++++++ src/cairo-xcb-surface-render.c | 3 +++ src/cairo-xlib-display.c | 1 + src/cairo-xlib-surface.c | 3 +++ src/cairo-xml-surface.c | 28 +++++++++------------------- src/cairo.h | 10 ++++++---- src/cairoint.h | 4 ++-- src/drm/cairo-drm-gallium-surface.c | 1 + src/drm/cairo-drm-i915-surface.c | 1 + src/drm/cairo-drm-intel-surface.c | 2 +- src/drm/cairo-drm-intel.c | 11 ++++++++--- src/drm/cairo-drm-radeon-surface.c | 2 +- test/any2ppm.c | 1 + test/png.c | 1 + util/cairo-script/cairo-script-operators.c | 5 +++++ util/cairo-sphinx/sphinx.c | 9 ++++++--- util/cairo-trace/trace.c | 4 ++++ 24 files changed, 88 insertions(+), 43 deletions(-) commit 620cd9c2be4a6bef790e6818652470a5c53d578d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Feb 28 09:09:20 2010 +0000 image: Allow reduction of repeat modes for unity scaled patterns. src/cairo-image-surface.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) commit 9ecb0f3cec491264041d0d3718595ad95164980a Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Sun Feb 28 01:42:10 2010 +0200 image: Fix bug optimizing surface pattern extend modes. The image surface tries to convert surface pattern's extend modes to EXTEND_NONE, if it can, when converting a cairo_pattern_t to a pixman_image_t. The check was not taking into account the transformation matrix on the pattern, so it was possible to trick it into using EXTEND_NONE by downscaling the source pattern enough. This patch changes the optimization to only take if the pattern has no transformation. Fixes surface-pattern-scale-down-extend-{pad,reflect,repeat} failures in the test suite for the image backend. src/cairo-image-surface.c | 4 ++++ 1 file changed, 4 insertions(+) commit 260d7f8ace9dea8ff6b3a70f481e433cf399a3d2 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Sat Feb 27 17:41:28 2010 +0200 test: Test downscaling and extending a surface pattern. Franz Schmid reported a regression in 1.9.6 when downscaling and using EXTEND_REPEAT for an image surface pattern. This patch adds such tests for every extend mode. test/Makefile.sources | 1 + .../surface-pattern-scale-down-extend-none.ref.png | Bin 0 -> 329 bytes test/surface-pattern-scale-down-extend-pad.ref.png | Bin 0 -> 320 bytes ...rface-pattern-scale-down-extend-reflect.ref.png | Bin 0 -> 328 bytes ...urface-pattern-scale-down-extend-repeat.ref.png | Bin 0 -> 330 bytes test/surface-pattern-scale-down-extend.c | 107 +++++++++++++++++++++ 6 files changed, 108 insertions(+) commit ebadc2ed0810d9941a2f44586016073fea90115b Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Sat Feb 27 03:24:55 2010 +0200 test: Add tests covering more clip fast paths in groups. The clip-group-shapes-* tests check that it doesn't matter whether the clip path is set before or after pushing a group using specific types of clip paths (aligned/unaligned rectangles and general paths.) test/Makefile.sources | 1 + test/clip-group-shapes-aligned-rectangles.ref.png | Bin 0 -> 378 bytes test/clip-group-shapes-circles.ref.png | Bin 0 -> 1510 bytes .../clip-group-shapes-unaligned-rectangles.ref.png | Bin 0 -> 415 bytes test/clip-group-shapes.c | 189 +++++++++++++++++++++ 5 files changed, 190 insertions(+) commit e8d0b4fb39eba799e8649df6dd609f6100efb95e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Feb 26 10:28:23 2010 +0000 gl: Markup internal functions with cairo_private src/cairo-gl-private.h | 22 +++++++++++----------- 1 file changed, 11 insertions(+), 11 deletions(-) commit ef807119ff8ee23c0444f1197f92b06a5caf6024 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Thu Feb 25 13:58:53 2010 +0200 compiler: Explain the slim hidden macros. The slim_hidden_* macro definitions are rather cryptic at first sight and I keep needing to rederive how they work just find out what they're supposed to do. This patch adds a comment explaining how they're used and work. src/cairo-compiler-private.h | 30 ++++++++++++++++++++++++++++++ 1 file changed, 30 insertions(+) commit dcf897967d54f579c737bbcc10af7fa295b586e5 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Sun Feb 21 01:45:30 2010 +0200 test: Fix a wrongly typed return value in scaled-font-zero-matrix. The test was returning a cairo_status_t, but should be returning a cairo_test_status_t instead. When the test failed it was being reported as having crashed, rather than merely failed, because the enum value of CAIRO_TEST_CRASHED happened to be same as the cairo_status_t value of the cairo context at the end of the failing test. test/scaled-font-zero-matrix.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) commit 8bb06915ed6628c6d8978b6c2fec474bbf08d7e9 Author: Benjamin Otte <otte@redhat.com> Date: Tue Feb 23 21:01:13 2010 +0100 image: split cairo_image_surface_coerce() Split into a general cairo_image_surface_coerce() that coerces to one of the 3 supported formats (ARGB32, RGB24, A8) based on content and the more general cairo_image_surface_coerce_to_format() that coerces to a specified format. src/cairo-gl-glyphs.c | 3 +-- src/cairo-gl-surface.c | 3 +-- src/cairo-image-surface.c | 14 ++++++++++++-- src/cairo-png.c | 3 +-- src/cairo-scaled-font.c | 2 +- src/cairo-script-surface.c | 4 +--- src/cairo-svg-surface.c | 4 ++-- src/cairo-type3-glyph-surface.c | 2 +- src/cairo-xcb-surface-render.c | 21 ++++----------------- src/cairoint.h | 8 ++++++-- 10 files changed, 30 insertions(+), 34 deletions(-) commit b7c42b6aaa2dba09ed3ff9ae85eae27bec268713 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Feb 23 19:27:16 2010 +0000 drm: Add missing private headers Reported by: Thomas Jones <thomas.jones@utoronto.ca> src/Makefile.sources | 2 ++ 1 file changed, 2 insertions(+) commit 6ab5f89571b0252496d58242ed3060a9486d68d4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Feb 23 08:54:58 2010 +0000 bo-rectangular: Fix incorrect skipping of colinear eo edges Fixes test/bug-bo-rectangular After skipping edges, we need to bd careful to only terminate the box on a closing edge. src/cairo-bentley-ottmann-rectangular.c | 31 ++++++++++++++----------------- 1 file changed, 14 insertions(+), 17 deletions(-) commit 1897156d96d77ff28f585794c1eb1baffa105488 Author: Benjamin Otte <otte@gnome.com> Date: Tue Feb 23 08:41:27 2010 +0000 test: Add bug-bo-rectangular Exercises a bug found in the special case rectangular tessellator with colinear eo edges. test/Makefile.am | 1 + test/Makefile.sources | 1 + test/bug-bo-rectangular.c | 67 ++++++++++++++++++++++++++++++++++++++++ test/bug-bo-rectangular.ref.png | Bin 0 -> 950 bytes 4 files changed, 69 insertions(+) commit 21b2457aabf7df2a9f040751cd48e396443262f1 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Feb 22 15:26:37 2010 +0000 fallback: Initialize rects.is_bounded for span renderer. src/cairo-surface-fallback.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit 6dd22b767808f721de5d00fe384711ffaa269453 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Feb 22 15:24:08 2010 +0000 gl: Initialise glyphs mask to ensure allocation in absence of stash. src/cairo-gl-glyphs.c | 1 + 1 file changed, 1 insertion(+) commit 668ac047e6c790b0f8f58c52f169c688caa81678 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Feb 22 14:57:01 2010 +0000 gl: Avoid attempting to create a program on GLSL-incapable h/w for spans src/cairo-gl-shaders.c | 17 ++++++++++++----- src/cairo-gl-surface.c | 6 ++++++ 2 files changed, 18 insertions(+), 5 deletions(-) commit 29df5c91d02276211962a083284feb9a424f0d97 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Feb 22 14:46:18 2010 +0000 gl: Return unsupported for h/w that cannot handle shaders. Remove the ASSERT_NOT_REACHED and propagate the unsupported status so that we fallback to fixed-function gracefully. src/cairo-gl-shaders.c | 28 +++++++++++++++------------- 1 file changed, 15 insertions(+), 13 deletions(-) commit 9f46bad301eff967dc0790fc271d872f8a45cedb Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Feb 22 14:20:34 2010 +0000 bo-rectangular: Prevent invalid read during edge traversal Benjamin Otte tracked down an invalid read triggered by WebKit. The cause is that we attempt to dereference the list_head as an edge as we failed to check that during the skipping of colinear edges we advanced to the end, under the false assumption that there would always be a closing edge in a rectangle. This assumption is broken if the tail rectangles having colinear right edges. src/cairo-bentley-ottmann-rectangular.c | 6 ++++++ 1 file changed, 6 insertions(+) commit 75db4f0ece194b7aa0455509785a300784778ec4 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Mon Feb 22 02:26:47 2010 +0200 text: Fix assert failure from compositing by SOURCE with a mask internally. Change the operator used to upgrade the format of a glyph mask from CAIRO_OPERATOR_SOURCE to CAIRO_OPERATOR_ADD. The _cairo_scaled_font_show_glyphs() function upgrades the pixel format of a glyph mask by calling _cairo_surface_composite() to copy the mask to an image surface of the upgraded destination. The way it was doing it however was to use CAIRO_OPERATOR_SOURCE, a white source pattern and the glyph's rasterised glyph as the mask pattern. This combination isn't supported by _cairo_surface_composite(), which asserts that no mask is present when the operator is SOURCE or CLEAR. Reported by Mikael Magnusson to #cairo on irc.freenode.net. src/cairo-scaled-font.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 20f8f17f0e494a57a6b413be76f9333b42d88014 Author: Andrea Canciani <ranma42@gmail.com> Date: Sat Feb 20 21:42:23 2010 +0100 xcb: Fix compilation without dri2 XCB should be able to compile without xcb-drm. To do so it can include xcb/dri2.h only if xcb-drm is enabled. src/cairo-xcb-surface.c | 2 ++ 1 file changed, 2 insertions(+) commit fc7be0014d91c8536db8a15ae98b5e78c354c160 Author: Carl Worth <cworth@cworth.org> Date: Fri Feb 19 17:39:50 2010 -0800 Update version to 1.9.7. Following the 1.9.6 snapshot. cairo-version.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit cb090136b2b0b89bde69d9575f2f592b46e144e8 Author: Carl Worth <cworth@cworth.org> Date: Fri Feb 19 17:28:29 2010 -0800 Update REFERENCE_IMAGES list for "make distcheck". As usual, "make distcheck" does not work until we update this list. test/Makefile.am | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) commit 20cf7c92b332a5fd28227f0cdaa50e3f1080c038 Author: Carl Worth <cworth@cworth.org> Date: Fri Feb 19 17:24:33 2010 -0800 Update version to 1.9.6 For a new cairo 1.9.6 snapshot. cairo-version.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 1c3d417c30bd60c51512d23440896a6a7f51bf23 Author: Carl Worth <cworth@cworth.org> Date: Fri Feb 19 17:23:16 2010 -0800 NEWS: Add notes for the 1.9.6 snapshot. Based on a very brief reading of the git log. There's certainly a lot of interesting stuff missing here. NEWS | 114 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++-- 1 file changed, 112 insertions(+), 2 deletions(-) commit 9057c4b6be5f5cc6f8905e017dde534402ddedda Author: Eric Anholt <eric@anholt.net> Date: Wed Feb 17 12:37:07 2010 -0800 test: Add an option to explicitly specify a test name to the runner. test/cairo-test-runner.c | 14 +++++++++++--- 1 file changed, 11 insertions(+), 3 deletions(-) commit e0a948bd2ff3504f5d8b3b3f2136cd9b86cf8ec6 Author: Eric Anholt <eric@anholt.net> Date: Wed Feb 17 11:48:41 2010 -0800 gl: Be sure to do linear interpolation on gradients. src/cairo-gl-surface.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) commit f1e794956f60a935bfc38ae1c2c2d51d37aadee3 Author: Eric Anholt <eric@anholt.net> Date: Wed Feb 17 11:03:52 2010 -0800 gl: Stop trying to handle first/last stop offset in the gradient shaders. Since moving to pixman for the gradient textures, first=0 and last=1, so there's no need to rescale to 0,1 any more. Shaves 6 Gen4 ISA instructions, including 2 inverses, in the radial fragment shader. src/cairo-gl-private.h | 4 ---- src/cairo-gl-shaders.c | 12 ------------ src/cairo-gl-surface.c | 51 +++----------------------------------------------- 3 files changed, 3 insertions(+), 64 deletions(-) commit f813a0243b90ea4151f60f3d13bb6c84d21270df Author: Eric Anholt <eric@anholt.net> Date: Wed Feb 17 10:58:16 2010 -0800 gl: Remove the old unused radial/linear gradients shaders. They're dynamically generated now for source/mask. src/cairo-gl-private.h | 6 --- src/cairo-gl-shaders.c | 117 ------------------------------------------------- 2 files changed, 123 deletions(-) commit 8da843e996bda7152abb217b3f461a8091dbc880 Author: Eric Anholt <eric@anholt.net> Date: Wed Feb 17 10:54:39 2010 -0800 gl: Reduce complexity of radial fragment shaders. This shaves 1 Mesa IR instruction, and 6 Gen4 ISA instructions. src/cairo-gl-shaders.c | 10 ++-------- 1 file changed, 2 insertions(+), 8 deletions(-) commit 89bdc2f8d55d951e15b77e6737c57b208d984b0a Author: Eric Anholt <eric@anholt.net> Date: Thu Feb 4 23:17:59 2010 -0800 gl: Implement draw_image for window targets. Creates a texture and draws with it instead of doing TexSubImage. Open question is whether this wouldn't be better in general. Fixes several failures with ARB_texture_rectangle path due to fallbacks to window drawing. src/cairo-gl-surface.c | 157 ++++++++++++++++++++++++++++++++++++++----------- 1 file changed, 124 insertions(+), 33 deletions(-) commit fcd29473ff71b74bf541199293a966df2232fd63 Author: Eric Anholt <eric@anholt.net> Date: Tue Feb 16 13:47:41 2010 -0800 gl: Fix glyphs texture coordinates for ARB_texture_rectangle. Fixes most of the text testcases to match the ARB_texture_non_power_of_two results. src/cairo-gl-glyphs.c | 16 ++++++++++------ 1 file changed, 10 insertions(+), 6 deletions(-) commit e845450905f373e4fcb9ae4717cfd2c3e6472423 Author: Eric Anholt <eric@anholt.net> Date: Tue Feb 16 13:18:33 2010 -0800 gl: Add GLSL support for ARB_texture_rectangle, and repeat/reflect fallbacks. Most testcases are now passing like the ARB_texture_non_power_of_two case. EXT_texture_rectangle support is dropped in favor of ARB_texture_non_power_of_two. If we have issues with drivers not having that but having EXT (which just lacks the GLSL part of the spec), we can split it out. Right now non-GLSL support in cairo-gl is probably in bad shape anyway and will require someone that cares for it in order to get fixed up. src/cairo-gl-shaders.c | 44 ++++++++++++++++++++++++++++++++++++++++++++ src/cairo-gl-surface.c | 21 +++++++++++++++------ 2 files changed, 59 insertions(+), 6 deletions(-) commit 40294377cb26fab7bcf17ee98b8e4f5bf9778ca2 Author: Dave Airlie <airlied@redhat.com> Date: Sat Jan 30 19:24:31 2010 +1000 gl: initial support for EXT_texture_rectangle extension The only issue is the repeat modes aren't supported for rectangular textures. In any case even with ARB_npot radeon's pre-r600 lie and fail to do repeats anyway. src/cairo-gl-glyphs.c | 25 +++++----- src/cairo-gl-private.h | 1 + src/cairo-gl-surface.c | 124 ++++++++++++++++++++++++++++++------------------- 3 files changed, 89 insertions(+), 61 deletions(-) commit 6542a515f04d52d10fdc89ee9abc76e54282a2d8 Author: Eric Anholt <eric@anholt.net> Date: Sun Feb 7 20:26:56 2010 +0100 gl: Reduce the size of the gradient texture for small numbers of stops. This code is stolen straight from cairo-drm-intel.c. This saves a bunch of time calculating interpolated points when we just do interpolation between points at sampling time anyway. Reduces firefox-talos-svg from 47 seconds back to the 42 it was at before the pixman change. This regresses the reported result of huge-radial, but there's no visible difference. src/cairo-gl-surface.c | 63 +++++++++++++++++++++++++++++++++++--------------- 1 file changed, 45 insertions(+), 18 deletions(-) commit 3b678a88b0d026bb765fd9cdff92fe7235e19848 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Feb 12 16:44:25 2010 +0000 drm/i915: Set a source for unbounded fixups. src/drm/cairo-drm-i915-surface.c | 52 +++++++++++++++++++++++++++++++--------- 1 file changed, 41 insertions(+), 11 deletions(-) commit 45a275ee29095b69c620ece868411885f4574be9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Feb 12 14:12:24 2010 +0000 drm/i915: Use correct linear mode for mask. src/drm/cairo-drm-i915-shader.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 37dbf157f354234d04cd4f0c5d069ff0d535e862 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Feb 12 14:08:32 2010 +0000 clip: Avoiding derefencing NULL clip_region src/cairo-image-surface.c | 26 ++++++++++++++------------ src/cairo-xcb-surface-render.c | 26 ++++++++++++++------------ 2 files changed, 28 insertions(+), 24 deletions(-) commit 6995e1d982a3ad84a1b74730ce557840745fcd8e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Feb 12 14:08:14 2010 +0000 boilerplate: Include X11.h for GL/GLX boilerplate/cairo-boilerplate-gl.c | 3 +++ 1 file changed, 3 insertions(+) commit 5390df961f6dff8e25e5aac21062026a81710d88 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Feb 12 12:20:35 2010 +0000 clip: Restrict composite extents to clip extents Fixes test/clip-rectangle-twice. src/cairo-clip-private.h | 6 +++ src/cairo-clip.c | 84 ++++++++++++++++++++++++++++++++++++++++ src/cairo-image-surface.c | 74 +++++++++-------------------------- src/cairo-surface-fallback.c | 71 ++------------------------------- src/cairo-xcb-surface-render.c | 78 ++++++++++--------------------------- src/drm/cairo-drm-i915-surface.c | 47 ++-------------------- src/drm/cairo-drm-i965-surface.c | 47 ++-------------------- 7 files changed, 137 insertions(+), 270 deletions(-) commit 51047483f462a905567b42275ae061ead4df0a07 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Feb 12 12:18:30 2010 +0000 test: Add clip-twice-rectangle Jeff Muizeelar found another bug with clipping whereby the clip was been incorrectly discarded. test/Makefile.am | 1 + test/Makefile.sources | 1 + test/clip-twice-rectangle.c | 70 ++++++++++++++++++++++++++++++++++++++ test/clip-twice-rectangle.ref.png | Bin 0 -> 323 bytes 4 files changed, 72 insertions(+) commit 22b5f78c1c1ae7d289a99a3c52354bba3b9e817f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Feb 11 10:38:51 2010 +0000 gl: Remove eagle support Eagle is no more, Kristian has superseded it with true EGL support. He is so happy... boilerplate/Makefile.win32.features | 10 -- build/Makefile.win32.features | 1 - build/Makefile.win32.features-h | 3 - build/configure.ac.features | 1 - configure.ac | 12 --- src/Makefile.sources | 2 - src/Makefile.win32.features | 14 --- src/cairo-eagle-context.c | 185 ------------------------------------ src/cairo-gl.h | 12 --- 9 files changed, 240 deletions(-) commit ed4a30b38311e4ce0730ecd6026432f4a1e8ddf7 Author: Benjamin Otte <otte@redhat.com> Date: Wed Feb 10 22:10:23 2010 +0100 build: Add -Wdeclaration-after-statement ... and fix the compile errors from it I get on my build. It's Cairo style to put declarations before the code, so better warn about it. Besides, it eases porting to old compilers like MSVC. build/configure.ac.warnings | 2 +- src/cairo-gl-shaders.c | 8 ++++---- src/cairo-gl-surface.c | 3 ++- src/cairo-xcb-surface-core.c | 3 ++- 4 files changed, 9 insertions(+), 7 deletions(-) commit 4113e455a317785b83324215a6b0156995665989 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Feb 10 21:09:56 2010 +0000 drm/i965: fix use of uninitialized variable whilst dumping errors Cut'n'paste bug from i915 error path. src/drm/cairo-drm-i965-surface.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) commit 5d34902c0cb9712179292112fbdb0f06be327fd8 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Feb 10 12:10:16 2010 +0000 script: Compile without mmap Should fix: Bug 26509 - Cairo fails to compile without mmap http://bugs.freedesktop.org/show_bug.cgi?id=26509 As reported by Hib Eris, Cairo files to compile under a mingw32 cross-compiler as we use a structure only defined if HAVE_MMAP unconditionally. util/cairo-script/cairo-script-operators.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) commit fb5af4ae407044d1c0d0144ec13f5951075426a3 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Feb 10 12:00:59 2010 +0000 drm/intel: An interrupted SET_TILING modifies the input parameters If a signal interrupts the SET_TILING ioctl, the tiling and stride values are updated to reflect the current condition of the buffer, so we need to restore those to the desired values before repeating the ioctl. src/drm/cairo-drm-intel.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) commit 3c66c721d1bb1db6052acf722f609dcbe56431f5 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Feb 9 08:48:42 2010 +0000 qt: Fix compile error. Update after prototype changed for create_similar(), as pointed out by augzilla. src/cairo-qt-surface.cpp | 1 - 1 file changed, 1 deletion(-) commit a1e5b07fa834819b8d9df9fa56758639a08262d7 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Feb 7 20:26:27 2010 +0000 ps: Initialise page bbox to page size src/cairo-ps-surface.c | 5 +++++ 1 file changed, 5 insertions(+) commit 44b6370cb04d27e1ae3e50558a9085d318990938 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Feb 7 19:36:29 2010 +0000 ps: Propagate NOTHING_TO_DO The upper layers should know what to do if there is nothing to be done, so pass on that knowledge. src/cairo-ps-surface.c | 36 ++++-------------------------------- 1 file changed, 4 insertions(+), 32 deletions(-) commit a4793d1331845b85a9c1810035119dfbcffa5082 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Feb 7 19:27:32 2010 +0000 test: Add linear-gradient-subset It was observed that we never actually test the condition that the gradient is only defined for a portion of its range, i.e. the starting offset is >0 and the ending offset is <0. By definition the colour between 0 and start offset is the start color, so check that this behaviour is followed by all backends. test/Makefile.am | 1 + test/Makefile.sources | 1 + test/linear-gradient-subset.c | 127 ++++++++++++++++++++++++++++++++++++ test/linear-gradient-subset.ref.png | Bin 0 -> 825 bytes 4 files changed, 129 insertions(+) commit 0f99303cd6d68df79070200a843194a20ad1c597 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Feb 7 12:40:19 2010 +0000 gl: Use pixman to create the gradient texture This is just for consistency as we know that pixman has to handle all the corner cases anyway... src/cairo-gl-surface.c | 145 +++++++++++++++++++++++++++---------------------- 1 file changed, 80 insertions(+), 65 deletions(-) commit b028d287e6effe30833963f4df40db3846e7f4da Author: Eric Anholt <eric@anholt.net> Date: Fri Feb 5 07:25:04 2010 -0800 [gl] Make the VBO once at startup instead of recreating per glyphs/spans. This shaves 2% off of firefox-talos-gfx. src/cairo-gl-glyphs.c | 10 ++-------- src/cairo-gl-private.h | 1 + src/cairo-gl-surface.c | 8 +++----- 3 files changed, 6 insertions(+), 13 deletions(-) commit 696a715702ed18bbe3f7d8b97654a055fa37444e Author: Eric Anholt <eric@anholt.net> Date: Wed Feb 3 21:34:24 2010 -0800 [gl] Add radial gradients acceleration. This is significantly cribbed from Zach Laine's work, but reworked so that gradients can be plugged in as either source or mask operands for any of the paths. src/cairo-gl-private.h | 13 +++++ src/cairo-gl-shaders.c | 77 +++++++++++++++++++++++++++-- src/cairo-gl-surface.c | 129 +++++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 216 insertions(+), 3 deletions(-) commit 297b0ab47fa63ef99e65b6834b731c260ea3e941 Author: Eric Anholt <eric@anholt.net> Date: Thu Feb 4 23:57:43 2010 -0800 [gl] When making a boilerplate GLX window, ensure it has alpha. cairo_gl_surface_create_for_window assumes CONTENT_COLOR_ALPHA, so make sure the fbconfig we choose is good enough. Fixes gl-window testcase results to basically match the non-window testcases. boilerplate/cairo-boilerplate-gl.c | 17 +++++++++-------- 1 file changed, 9 insertions(+), 8 deletions(-) commit 6708bc0593a649d083ede429ae73f06691edc018 Author: Eric Anholt <eric@anholt.net> Date: Thu Feb 4 23:47:33 2010 -0800 [gl] When filling a gradient texture, multiply alpha after interpolating. Fixes gradient-alpha testcase. src/cairo-gl-surface.c | 20 +++++++++----------- 1 file changed, 9 insertions(+), 11 deletions(-) commit e426cdd56973ff1114dac5e8b1130180d43631b0 Author: Eric Anholt <eric@anholt.net> Date: Wed Feb 3 16:03:52 2010 -0800 [gl] Implement linear gradients acceleration. This is significantly cribbed from Zach Laine's work, but reworked so that gradients can be plugged in as either source or mask operands for any of the paths. This cuts the runtime of firefox-talos-svg in half on my GM45, at the expense of gradient-alpha. surface-pattern-operator also now fails due to small rasterization differences. src/cairo-gl-glyphs.c | 1 + src/cairo-gl-private.h | 11 ++++ src/cairo-gl-shaders.c | 39 ++++++++++++-- src/cairo-gl-surface.c | 139 +++++++++++++++++++++++++++++++++++++++++++++---- 4 files changed, 176 insertions(+), 14 deletions(-) commit 05b18fc09af0e0887c652bd482d68f00bec6d4d7 Author: Eric Anholt <eric@anholt.net> Date: Thu Feb 4 00:12:58 2010 -0800 [gl] Fix create_gradient_texture to premultiply alpha. src/cairo-gl-surface.c | 3 +++ 1 file changed, 3 insertions(+) commit 31e706454b7f0fa510047f582250a7a2db24dd31 Author: Eric Anholt <eric@anholt.net> Date: Wed Feb 3 15:48:40 2010 -0800 [gl] Pull the gradient operand setup out to its own function. Hopefully this clarifies what's going on, including the fallthrough to texture setup on failure. src/cairo-gl-surface.c | 77 ++++++++++++++++++++++++++++---------------------- 1 file changed, 43 insertions(+), 34 deletions(-) commit 745f263c1c2d263f7c5e78964bf7181f650301cc Author: T. Zachary Laine <whatwasthataddress@gmail.com> Date: Wed Feb 3 15:38:02 2010 -0800 [gl] Add function for creating a 1D texture to use for gradient lookups. Rather than have fragment shaders try to walk a variable-length set of stops and try to compute the interpolation between the nearest stops per fragment, we make a little texture representing the gradient along a line and have the shaders compute texture coordinates along the line and sample. This should be a good tradeoff between CPU work and GPU work. [anholt: I pulled this set of code out of Zach Laine's tree to use for gradient acceleration, and wanted to make sure attribution was appropriate. I applied only minor tweaking] src/cairo-gl-private.h | 1 + src/cairo-gl-surface.c | 117 +++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 118 insertions(+) commit 88c8304e9072e5a46890a0eb32fcdbc857eb1a12 Author: Eric Anholt <eric@anholt.net> Date: Wed Feb 3 17:06:45 2010 -0800 [gl] Sanity check the sampler uniform binding. src/cairo-gl-shaders.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) commit 42597170f2303475bb0507f4c6e948eead28c7bd Author: Eric Anholt <eric@anholt.net> Date: Wed Feb 3 16:55:45 2010 -0800 [gl] Convert the spans code to using GLSL when available. src/cairo-gl-private.h | 1 + src/cairo-gl-shaders.c | 35 ++++++++++++++++++++++++++++++--- src/cairo-gl-surface.c | 52 +++++++++++++++++++++++++++++++------------------- 3 files changed, 65 insertions(+), 23 deletions(-) commit 8d16fb10ea9940c0c9f311ff08c51d572c068c17 Author: Eric Anholt <eric@anholt.net> Date: Wed Feb 3 15:31:57 2010 -0800 [gl] Test for required EXT_bgra and explain what it's used for. src/cairo-gl-surface.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) commit b2d8c9cbf2faec570450d163da140c54e64bbdc8 Author: Eric Anholt <eric@anholt.net> Date: Wed Feb 3 15:25:47 2010 -0800 [gl] Use the shader source choice from operand setup instead of DIY. src/cairo-gl-glyphs.c | 15 +-------------- 1 file changed, 1 insertion(+), 14 deletions(-) commit 0fceac4d52d4b574b9c1f911b83d15dc7e01691d Author: Eric Anholt <eric@anholt.net> Date: Wed Feb 3 12:24:41 2010 -0800 [gl] Move the component alpha composite to using GLSL when available. src/cairo-gl-surface.c | 165 ++++++++++++++++++++++++++++++++----------------- 1 file changed, 108 insertions(+), 57 deletions(-) commit 8357336ed7be91ad359d205b1f223486d13215f2 Author: Eric Anholt <eric@anholt.net> Date: Wed Feb 3 11:54:24 2010 -0800 [gl] Move normal composites to using GLSL when available. src/cairo-gl-private.h | 2 ++ src/cairo-gl-surface.c | 55 ++++++++++++++++++++++++++++++++++++-------------- 2 files changed, 42 insertions(+), 15 deletions(-) commit 106bc158bafa4eecfd7f79422fc19d267c20dcdb Author: Eric Anholt <eric@anholt.net> Date: Wed Feb 3 15:15:02 2010 -0800 [gl] Move the glyphs shader construction to a more general location. src/cairo-gl-glyphs.c | 213 ++++++++++--------------------------------------- src/cairo-gl-private.h | 48 +++++++---- src/cairo-gl-shaders.c | 200 ++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 273 insertions(+), 188 deletions(-) commit f475351f75a3ec079628a2eb2643ebc09f0e5cc8 Author: Eric Anholt <eric@anholt.net> Date: Tue Feb 2 19:23:17 2010 -0800 [gl] Use GLSL when available for glyph rendering. This reduces the CPU work in translating fixed function state to shaders, but currently is a slight cost on GM45 because we end up changing shaders more frequently since other parts of the pipeline are doing fixed function still. src/cairo-gl-glyphs.c | 189 ++++++++++++++++++++++++++++++++++++++++++++++--- src/cairo-gl-private.h | 19 +++++ src/cairo-gl-shaders.c | 10 ++- src/cairo-gl-surface.c | 56 +++++++++------ 4 files changed, 239 insertions(+), 35 deletions(-) commit 672973caa0c3877b3b9f1c5f5966449ce0d7e239 Author: Eric Anholt <eric@anholt.net> Date: Tue Feb 2 11:17:43 2010 -0800 [gl] Convert fill_rectangles to the new GLSL core/arb wrapper. src/cairo-gl-private.h | 12 +++--- src/cairo-gl-shaders.c | 106 +++++++++++-------------------------------------- src/cairo-gl-surface.c | 39 +++++++++--------- 3 files changed, 48 insertions(+), 109 deletions(-) commit 416273db34a715a6fa180f52ddfd11b82b6e8b6c Author: Eric Anholt <eric@anholt.net> Date: Tue Feb 2 10:37:42 2010 -0800 [gl] Cut out the abstraction of GL_VERTEX_SHADER vs GL_VERTEX_SHADER_ARB. They're the same number, so just use the core 2.0 name. src/cairo-gl-shaders.c | 38 ++------------------------------------ 1 file changed, 2 insertions(+), 36 deletions(-) commit 48a5a6eefc923db47c35ad7da3b635d205736716 Author: Eric Anholt <eric@anholt.net> Date: Tue Feb 2 10:36:16 2010 -0800 [gl] Avoid macro obfuscation of the shader impl structures. src/cairo-gl-shaders.c | 52 ++++++++++++++++++++++++++++++-------------------- 1 file changed, 31 insertions(+), 21 deletions(-) commit 25ccc5dcb036554430b66fb1bf5f74766daa5612 Author: Zach Laine <whatwasthataddress@gmail.com> Date: Thu Jan 14 11:49:04 2010 -0600 [gl] Add shader support code for GL versions < 3.0. Adds cairo_gl_shader_program_t, and functions to manipulate same. Multiple GL entry points for shaders are provided -- one for the pre-GL 2.0 extenstions entry points, and one for GL 2.0. This code is well tested, but currently unused in the GL backend. src/cairo-gl-private.h | 49 ++++ src/cairo-gl-shaders.c | 623 ++++++++++++++++++++++++++++++++++++++++++++++++- 2 files changed, 671 insertions(+), 1 deletion(-) commit a6897ad3867611bd74f8ab7de4b1e8d8d1c12c71 Author: Eric Anholt <eric@anholt.net> Date: Tue Feb 2 11:24:36 2010 -0800 [gl] Check for GLSL support once at context init instead of per draw. src/cairo-gl-private.h | 1 + src/cairo-gl-surface.c | 11 ++++++++++- 2 files changed, 11 insertions(+), 1 deletion(-) commit b1829a47bff2b03a850d4049f228d2a41e14d6bd Author: Andrea Canciani <ranma42@gmail.com> Date: Wed Feb 3 11:19:26 2010 +0100 [quartz] Unbreak mask fallback path _cairo_quartz_surface_paint will never fail as it falls back to image when CoreGraphics can't handle the requested operation. This means that upon fallback the ClipImage set by the masking code gets ignored, thus the mask fallback path is broken. Using the _cg function ensures that masking is either completely done by CoreGraphics, or the fallback path passes the mask operation to image. src/cairo-quartz-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 597dd24ddcc763281c82790b7677e54dc2db8f92 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Feb 3 08:53:44 2010 +0000 drm/i915: Free potential leak of cloned textures. src/drm/cairo-drm-i915-shader.c | 6 ++++++ 1 file changed, 6 insertions(+) commit 913a31494ec4bd000dfa303e362d8273b6857cfe Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Feb 2 17:49:09 2010 +0000 drm/i915: Increase batch buffer space. src/drm/cairo-drm-i915-private.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 1315323b20398c7c4efa68e4e1c7f6fd6c1d8d91 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Feb 2 17:38:06 2010 +0000 drm/i915: check for batch buffer overflow. Oops, the check for sufficient batch space mysteriously disappeared during the rearrangement. src/drm/cairo-drm-i915-private.h | 3 --- src/drm/cairo-drm-i915-shader.c | 44 ++++++++++++++++++++++++++++++++++++++++ src/drm/cairo-drm-i915-surface.c | 35 -------------------------------- 3 files changed, 44 insertions(+), 38 deletions(-) commit e2be50c722347734801468d7d9568a18dbbaaa17 Author: Andrea Canciani <ranma42@gmail.com> Date: Fri Jan 29 12:06:13 2010 +0100 [quartz] Fix UNSUPPORTED operations The fallback path shouldn't be used anymore, thus fallbacks are now handled by passing unsupported ops to the image backend. Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> src/cairo-quartz-surface.c | 234 ++++++++++++++++++++++++++++++++++++++------- 1 file changed, 198 insertions(+), 36 deletions(-) commit 3e69c38fe642be467fee0cad166b83006741d55c Author: Andrea Canciani <ranma42@gmail.com> Date: Sun Jan 24 21:59:32 2010 +0100 [quartz] Fix surface to CGImage conversion Snapshotting a surface doesn't produce a cairo_image_surface_t. Acquiring (and later releasing) the surface is needed to access its image data. src/cairo-quartz-surface.c | 75 +++++++++++++++++++++++----------------------- 1 file changed, 38 insertions(+), 37 deletions(-) commit 7aba47e9d87bf09a5fa7e41a3c76c775f28b08f3 Author: Andrea Canciani <ranma42@gmail.com> Date: Mon Jan 25 15:09:51 2010 +0100 [quartz] Silence compiler warnings src/cairo-quartz-surface.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) commit e9b07d194630f3f5e31eef3606f8e4c188a01d5f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Feb 2 16:28:30 2010 +0000 drm/i915: Don't tile page sized bo. Minor tweak from < to <= so that a page sized bo is also not tiled for those 32x32 pixmaps... src/drm/cairo-drm-i915-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 70ca9dd170ce1cf8941311ec1f3dc117aef141b6 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Feb 2 16:28:09 2010 +0000 drm/i915: Pass the correct length to dump the batchbuffer. src/drm/cairo-drm-i915-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 3266a1a72be9c0a6b6c0014c4b64098507b6d385 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Feb 2 16:27:52 2010 +0000 drm: Handle absence of PCI_ID src/drm/cairo-drm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 1ecefc53a1d370d9fffb323952dcabeef5b872c6 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Feb 2 16:26:54 2010 +0000 pattern: Zero-length gradients are not necessary empty Fixes: test/linear-step-function If the extend mode is unbounded, then the gradient is also unbound. src/cairo-pattern.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) commit 2a59220aa448cd7416f1851c4c62ca6ac396302d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Feb 2 16:24:30 2010 +0000 test: Add linear-step-function Another bug identified by Jeff Muizelaar was that the bounds for a zero-length (i.e. it started and stopped at the same point) were miscomputed. This test case exercises that bug. Note: I believe the output is wrong here when padding a zero-length gradient. On the left it should be red, and on the right it should be blue. test/Makefile.am | 1 + test/Makefile.sources | 1 + test/linear-step-function.c | 60 ++++++++++++++++++++++++++++++++++++ test/linear-step-function.xfail.png | Bin 0 -> 116 bytes 4 files changed, 62 insertions(+) commit 7b37ba8a49dca39a42b5395e0715d3d0a5a89972 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Feb 2 16:17:13 2010 +0000 clip: Propagate all-clipped when copying Fixes test/clip-empty-save. The all-clipped flag was not being copied to the new clip, which occurs for instance when an empty clip is pushed via a gstate save. src/cairo-clip.c | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) commit 7dc464bb2bd64d1469b7b03afaf2191ee6d9e28b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Feb 2 16:15:32 2010 +0000 test: Add clip-empty-save Jeff Muizelaar found a bug in _cairo_clip_init_copy() which was not correctly propagating the all-clipped status when an empty clip was saved in the gstate. This test case exercises that bug. test/Makefile.am | 1 + test/Makefile.sources | 1 + test/clip-empty-save.c | 68 +++++++++++++++++++++++++++++++++++++++++++ test/clip-empty-save.ref.png | Bin 0 -> 118 bytes 4 files changed, 70 insertions(+) commit ab3dc7bb31f7f8b8c4d87c1ae62bd946b260c77e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Jan 31 16:59:07 2010 +0000 csi-replay: compile fix util/cairo-script/csi-replay.c | 32 +++++++++++++++++++++----------- 1 file changed, 21 insertions(+), 11 deletions(-) commit ab605214d151098ad153e8bf74ae0ca71a34c963 Author: Andrea Canciani <ranma42@gmail.com> Date: Mon Feb 1 09:27:16 2010 +0100 Partially revert "Improve stroking of densely dashed styles" This reverts commit 26e9f149063b9e1fdb54fc54fccbefdf04a68190 on cairo-path-stroke. The changes in cairo-path-stroke are not needed anymore since dash pattern approximation is now done in gstate before passing the dash pattern to the backend. src/cairo-path-stroke.c | 24 ++++++------------------ 1 file changed, 6 insertions(+), 18 deletions(-) commit 8ffe0fc3825f0f6bc4d06607f6819ea8c2c2040b Author: Andrea Canciani <ranma42@gmail.com> Date: Sat Jan 30 09:29:48 2010 +0100 Revert "[quartz] Approximate dense dashing patterns" This reverts commit 27701ed8447d4c21b7bf9709a2fc21690b3cdc96. Approximation is now done in gstate. src/cairo-quartz-surface.c | 32 +++++++++++--------------------- 1 file changed, 11 insertions(+), 21 deletions(-) commit 2e91648bcba06832dfa92bd742c32b2f5c1ad989 Author: Andrea Canciani <ranma42@gmail.com> Date: Fri Jan 15 15:46:20 2010 +0100 Approximate dash patterns in gstate Since approximation is only done when the pattern is sub-tolerance, it is acceptable to apply it even for vector backends. By doing it once and for all backends, backends are guaranteed to have non-degenerate input. Acked-by: Adrian Johnson <ajohnson@redneon.com> src/cairo-gstate.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-) commit 67f666896cbfd0ac2728c7aa1325a7d0b2c8282d Author: Andrea Canciani <ranma42@gmail.com> Date: Sun Jan 24 22:17:18 2010 +0100 Improve comments style consistency Each line of a multi-line comment should begin with '*'. src/cairo-stroke-style.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 45c795a1820e0317b123e7782cefed6ade8996d7 Author: Andrea Canciani <ranma42@gmail.com> Date: Fri Jan 15 10:44:58 2010 +0100 Improve dash pattern approximation Dash pattern approximation was taking the caps style into account only for coverage computation, but not when computing the new pattern, thus the computed approximation had a higher coverage if the caps style was SQUARE or ROUND. Reviewed-by: M. Joonas Pihlaja <jpihlaja@cc.helsinki.fi> src/cairo-stroke-style.c | 56 ++++++++++++++++++++++++++++++++++++++++++++++-- 1 file changed, 54 insertions(+), 2 deletions(-) commit 8d7841048b079ce2a582ff17c90e82e0081e5f42 Author: Andrea Canciani <ranma42@gmail.com> Date: Sun Jan 17 19:00:47 2010 +0100 Round caps coverage extimate explanation Comment on how the round caps coverage has been computed, explaining the complete procedure. The comments doesn't contain intermediate (verbose and ugly) results, but when executed in a symbolic math program (sage, for example) computes the expected results. Reviewed-by: M. Joonas Pihlaja <jpihlaja@cc.helsinki.fi> src/cairo-stroke-style.c | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) commit 8558cd579059974758bda696c33f08399267d33e Author: Eric Anholt <eric@anholt.net> Date: Thu Jan 28 17:39:34 2010 -0800 [gl] Use GL_MESA_pack_invert to flip get_image of windows around. This doesn't have any fallback for a GL that fails to implement the extension. Fixes 120 testcases for gl-window. src/cairo-gl-surface.c | 4 ++++ 1 file changed, 4 insertions(+) commit 11a120fb8477910f5ac63dc1ebde45c276764783 Author: Andrea Canciani <ranma42@gmail.com> Date: Fri Jan 29 14:21:06 2010 +0100 Silence autoconf on Mac OS X Autoconf was complaining that AC_COMPILE_IFELSE and AC_RUN_IFELSE were called before AC_GNU_SOURCE. Moving AC_GNU_SOURCE right after AC_INIT solves this. configure.ac | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 05d4f9b461749f129b17cc97d6e3e6b49fe8e8d9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 29 11:30:49 2010 +0000 xlib: Compile fix for no fontconfig src/cairo-xlib-screen.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit e3e74641450481e7ad393bbc6fd1a7428be53de1 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 29 11:22:26 2010 +0000 atomic: Fix up compile on PPC with libatomic-ops src/cairo-atomic-private.h | 2 -- src/cairo-atomic.c | 3 ++- 2 files changed, 2 insertions(+), 3 deletions(-) commit e11d2d0b4d646757fcba11631b6db169e2c183b0 Author: Alexander Shulgin <alex.shulgin@gmail.com> Date: Fri Jan 29 09:17:01 2010 +0000 surface: Fix detach mime-data after ecda633f887a10da650b4cd3efb17c861a6f519f Hi, while browsing the git log I've noticed a small mistake in the above commit, where we reversed the order of init/fini required to reset the mime-data array. src/cairo-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 5f0a0088af784157d8efe149ed073c8fa010e648 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jan 28 13:23:39 2010 +0000 drm-intel: Unmap a full glyph cache. When we swap a full glyph cache for a fresh one, be sure to unmap the old one prior to release. src/drm/cairo-drm-i915-glyphs.c | 2 +- src/drm/cairo-drm-intel.c | 3 +++ 2 files changed, 4 insertions(+), 1 deletion(-) commit 78cd9471200c69c8b3737e859260f4552145d958 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jan 27 23:45:17 2010 +0000 drm: Paranoid assert that the fallback is released upon finish. src/drm/cairo-drm-surface.c | 2 ++ 1 file changed, 2 insertions(+) commit 5aa1191204e3eed6e3e97e23f37a52bbdfffcbce Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jan 27 23:42:26 2010 +0000 drm-intel: flush in memory modifications to texture. If the surface was written to using a fallback, and so is mapped, we need to flush those modifications by relinquishing the map. So the next time the application tries to write to the surface, those writes are correctly serialised with our reads. src/drm/cairo-drm-i915-shader.c | 14 ++++++++++++++ src/drm/cairo-drm-i965-shader.c | 12 ++++++++++++ 2 files changed, 26 insertions(+) commit a38accb694452fc5dfac63535648656885cd47eb Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jan 27 23:27:46 2010 +0000 drm-intel: The accumulation bo may be NULL, beware. If we are not accumulating but rendering directly to the output vbo, then the tail bo may not even be allocated, let alone mapped. src/drm/cairo-drm-i915-spans.c | 2 +- src/drm/cairo-drm-i965-glyphs.c | 2 +- src/drm/cairo-drm-i965-spans.c | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) commit 44734b9d3ba60fb45abb0268a916e40cd688ef6a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jan 27 23:17:44 2010 +0000 drm-intel: Assert that the bo is unmapped upon release. Check through error paths to catch a few more places where the mapped bo may have been leaked, and add an assert to abort in case we do leak a mapping. src/drm/cairo-drm-i915-spans.c | 3 +++ src/drm/cairo-drm-i965-glyphs.c | 4 ++++ src/drm/cairo-drm-i965-spans.c | 3 +++ src/drm/cairo-drm-intel.c | 6 +++++- 4 files changed, 15 insertions(+), 1 deletion(-) commit 6848c7c850aa0a710c1409d1fc2cb441d02a2b28 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jan 27 23:01:37 2010 +0000 drm-i915: Don't tile temporary image textures. As uploading to a tiled buffer is much slower than linear memory, don't unless we expect to reuse the texture. This is not true for sub-image clones, which are single shot affairs. src/drm/cairo-drm-i915-shader.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit f75b7c4c110e880f4bdd578d9180f9afbb1d0034 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jan 27 19:20:57 2010 +0000 drm/i915: Ensure surface is initialized. Let's try that again, this time remembering to place it on the conjunction of the two branches, so *both* paths are initialized. src/drm/cairo-drm-i915-shader.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) commit 02f20458d1dd48fc4589734fc4b3e0c8bea1ce5c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jan 27 19:02:21 2010 +0000 drm-i915: Ensure surface size for clone is initialized. If the surface had a snapshot, then we would have not initialized the surface size and so compute a bogus normalization matrix. src/drm/cairo-drm-i915-shader.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) commit c82ff8cbcf27dc20d2b44a988eed1b676d996e8c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jan 27 18:48:32 2010 +0000 configure: Enable AC_SYS_LARGEFILE for drm drm makes use of 64bit offsets when mmaping, and so we require the 64bit file interfaces. I had presumed that it would be pulled in with AC_GNU_SOURCE, but no. configure.ac | 1 + 1 file changed, 1 insertion(+) commit 6d2d338ac04b02200e009c9e97d3ce030d07eb6f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jan 25 22:47:10 2010 +0000 drm: compile fix for cairo_always_inline s/always_inline/cairo_always_inline/ src/drm/cairo-drm-i965-private.h | 8 +-- src/drm/cairo-drm-intel-brw-eu.h | 132 +++++++++++++++++++------------------- src/drm/cairo-drm-intel-private.h | 2 +- src/drm/cairo-drm-private.h | 2 +- 4 files changed, 72 insertions(+), 72 deletions(-) commit 58980b4ad60a60923a59370c81d59bb0457a7922 Author: Benjamin Otte <otte@redhat.com> Date: Wed Jan 27 09:56:14 2010 +0100 build: Remove glitz surface glitz is unmaintained and the GL surface is far superior anyway. README | 8 +- boilerplate/Makefile.am | 13 - boilerplate/Makefile.sources | 5 - boilerplate/Makefile.win32.features | 10 - boilerplate/cairo-boilerplate-glitz-agl.c | 191 --- boilerplate/cairo-boilerplate-glitz-glx.c | 265 ---- boilerplate/cairo-boilerplate-glitz-wgl.c | 188 --- build/Makefile.win32.features | 1 - build/Makefile.win32.features-h | 3 - build/configure.ac.features | 1 - configure.ac | 38 - doc/public/cairo-docs.xml | 1 - doc/public/cairo-sections.txt | 1 - doc/public/tmpl/cairo-glitz.sgml | 22 - src/Makefile.sources | 4 - src/Makefile.win32.features | 14 - src/cairo-deprecated.h | 1 - src/cairo-glitz-private.h | 41 - src/cairo-glitz-surface.c | 2450 ----------------------------- src/cairo-glitz.h | 57 - test/Makefile.am | 8 - test/Makefile.sources | 2 - test/glitz-surface-source.argb32.ref.png | Bin 377 -> 0 bytes test/glitz-surface-source.c | 293 ---- test/glitz-surface-source.ps2.ref.png | Bin 376 -> 0 bytes test/glitz-surface-source.ps3.ref.png | Bin 376 -> 0 bytes test/glitz-surface-source.rgb24.ref.png | Bin 301 -> 0 bytes util/cairo.modules | 10 +- 28 files changed, 3 insertions(+), 3624 deletions(-) commit 17ddb1598b3df374b6275d907441f34318b4257f Author: Benjamin Otte <otte@redhat.com> Date: Mon Jan 25 20:17:12 2010 +0100 docs: appease make check src/cairo-gl-surface.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit bac513a85dc8ef2915b2e60d034cf3dbc82a8e0c Author: Benjamin Otte <otte@redhat.com> Date: Mon Jan 25 20:05:56 2010 +0100 build: Fix pthread detection THe pthread.h detection code didn't pass -lpthread or -pthread to Cairo. configure.ac | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) commit 582604f7b6b5b2ab6e681245bc7cde64d4995307 Author: Benjamin Otte <otte@redhat.com> Date: Mon Jan 25 19:22:33 2010 +0100 [png] Simplify coercion code Call _cairo_image_surface_coerce() unconditionally to ensure coercion to one of the standard formats happens even when it's a format we support. src/cairo-png.c | 16 ++++++---------- 1 file changed, 6 insertions(+), 10 deletions(-) commit 002e02715cac9a285fd97e1bb6b068ceebfae3d8 Author: Benjamin Otte <otte@redhat.com> Date: Mon Jan 25 19:42:54 2010 +0100 Remove useless pixman version checks We require pixman >= 0.15.16 in configure.ac, so checks for older versions aren't required anymore. src/cairo-gl-surface.c | 2 -- src/cairo-image-surface.c | 10 ++-------- 2 files changed, 2 insertions(+), 10 deletions(-) commit 5a690ad4acb3687c45c201b3ce92cd3e41bf3b30 Author: Benjamin Otte <otte@gnome.org> Date: Fri Jan 15 13:44:04 2010 +0100 Give cairo_surface_mark_dirty() the slim hidden treatment Fixes test suite failure src/cairo-surface.c | 1 + src/cairoint.h | 1 + 2 files changed, 2 insertions(+) commit 571ab1007571cb8d78fcb5206b43d6b964aa0df1 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jan 25 18:22:03 2010 +0000 xcb: Fix linking with xcb-drm src/cairo-xcb-surface.c | 2 ++ 1 file changed, 2 insertions(+) commit ecda633f887a10da650b4cd3efb17c861a6f519f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jan 25 18:17:28 2010 +0000 mime-data: Detach existing alternate representations upon modification A major sore point in the current mime-data API is the choice to keep the alternate mime-data representations after the user has actually modifying the primary representation (i.e. has made it a target of a cairo_t and drawn to it). This has been rightly chastised as unexpected behaviour, and given that we already have the mechanism to detach snapshots upon modification, there is no reason not to behave as expected. src/cairo-surface.c | 21 ++++++++++++++++++++- 1 file changed, 20 insertions(+), 1 deletion(-) commit 123bdb086ad9c5725a51791458d31309c69a3f41 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jan 25 16:45:03 2010 +0000 xcb: Make shm optional Trying to build xcb on a system without SHM wrapped by xcb. The right answer would be to build libxcb-shm. The quick answer is to compile out shm support. boilerplate/Makefile.win32.features | 10 ++++++++++ build/Makefile.win32.features | 1 + build/Makefile.win32.features-h | 3 +++ build/configure.ac.features | 1 + configure.ac | 14 +++++++++++++- src/Makefile.sources | 8 ++++++-- src/Makefile.win32.features | 14 ++++++++++++++ src/cairo-xcb-connection.c | 22 ++++++++++++++++++---- src/cairo-xcb-private.h | 21 +++++++++++++++++++++ src/cairo-xcb-surface-core.c | 2 ++ src/cairo-xcb-surface-render.c | 2 ++ src/cairo-xcb-surface.c | 14 ++++++++++++++ 12 files changed, 105 insertions(+), 7 deletions(-) commit e1b3330376c173633e146eb6cf578592d424e4d5 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jan 25 14:48:44 2010 +0000 configure: dri2 is only required for xcb-drm Move the optional dependency on dri2 to the xcb-drm integration where it is required. configure.ac | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) commit 8151c72307e8cb369fbb524de56f87f805efff2b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jan 25 12:14:39 2010 +0000 xcb: compile errors with lockdep After moving the mutex to the superclass, I forgot to update the references from the lockdep asserts. src/cairo-xcb-connection.c | 6 +++--- src/cairo-xcb-screen.c | 16 ++++++++-------- 2 files changed, 11 insertions(+), 11 deletions(-) commit be81d656c53c76032fd3fe11c81402d17ec56327 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jan 25 11:54:24 2010 +0000 clip: Create a fallback clip surface If the destination cannot support a clip surface, then we will need to use an intermediate image surface and trigger fallbacks upon use. src/cairo-clip.c | 5 +++++ 1 file changed, 5 insertions(+) commit c453d8ab704e5f41cb010241efe9bcb9ff886024 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jan 25 11:45:01 2010 +0000 test: Erroneous ref image for big-line xlib-fallback. test/big-line.xlib-fallback.ref.png | Bin 723 -> 0 bytes 1 file changed, 0 insertions(+), 0 deletions(-) commit 180aea76adbb6d5c5909652b7a8cafb7a693e982 Author: Adrian Johnson <ajohnson@redneon.com> Date: Mon Jan 25 20:48:02 2010 +1030 Fix printing of rotated user fonts b7550bf0 introduced a regression when user fonts are rendered to a rotated context. src/cairo-type3-glyph-surface.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) commit abcb9d2c39178620b1847fe6ea6e144602464ecc Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jan 25 09:35:09 2010 +0000 image: Prevent deref of NULL clip_region Another stray from surface-fallback, spotted by Carlos: *INT_cairo_region_get_extents (region=0x0, extents=0xbfffce90) at cairo-region.c:385 385 if (region->status) { (gdb) bt 0 *INT_cairo_region_get_extents (region=0x0, extents=0xbfffce90) at cairo-region.c:385 1 0x0013ba46 in _cairo_image_surface_composite_trapezoids (op=CAIRO_OPERATOR_OVER, pattern=0xbfffdda8, abstract_dst=0x837de10, antialias=CAIRO_ANTIALIAS_DEFAULT, src_x=486, src_y=45, dst_x=0, dst_y=0, width=15, height=25, traps=0x837f100, num_traps=30, clip_region=0x0) at cairo-image-surface.c:4102 2 0x0015d846 in _cairo_surface_composite_trapezoids (op=CAIRO_OPERATOR_OVER, pattern=0xbfffdda8, dst=0x837de10, antialias=CAIRO_ANTIALIAS_DEFAULT, src_x=486, src_y=45, dst_x=0, dst_y=0, width=15, height=25, traps=0x837f100, num_traps=30, clip_region=0x0) at cairo-surface.c:2125 3 0x001607ce in _cairo_surface_fallback_composite_trapezoids (op=CAIRO_OPERATOR_OVER, pattern=0xbfffdda8, dst=0x837db00, antialias=CAIRO_ANTIALIAS_DEFAULT, src_x=486, src_y=45, dst_x=486, dst_y=45, width=15, height=25, traps=0x837fd78, num_traps=30, clip_region=0x0) at cairo-surface-fallback.c:1641 4 0x0015d8a4 in _cairo_surface_composite_trapezoids (op=CAIRO_OPERATOR_OVER, pattern=0xbfffdda8, dst=0x837db00, antialias=CAIRO_ANTIALIAS_DEFAULT, src_x=486, src_y=45, dst_x=486, dst_y=45, width=15, height=25, traps=0x837fd78, num_traps=30, clip_region=0x0) at cairo-surface.c:2137 5 0x001623b0 in _composite_traps_draw_func (closure=0xbfffd350, op=CAIRO_OPERATOR_OVER, src=0xbfffdda8, dst=0x837db00, dst_x=0, dst_y=0, extents=0xbfffdc78, clip_region=0x0) at cairo-surface-fallback.c:546 6 0x00161234 in _clip_and_composite (clip=0x0, op=<value optimized out>, src=0xbfffdda8, draw_func=0x162310 <_composite_traps_draw_func>, draw_closure=0xbfffd350, dst=0x837db00, extents=0xbfffdc78) at cairo-surface-fallback.c:463 7 0x0016205b in _clip_and_composite_trapezoids (src=<value optimized out>, op=CAIRO_OPERATOR_OVER, dst=0x837db00, traps=0xbfffd7bc, antialias=CAIRO_ANTIALIAS_DEFAULT, clip=0x0, extents=0xbfffdc78) at cairo-surface-fallback.c:853 8 0x001626ef in _cairo_surface_fallback_stroke (surface=0x837db00, op=CAIRO_OPERATOR_OVER, source=0xbfffdda8, path=0x1b164c, stroke_style=0x1b1524, ctm=0x1b15b0, ctm_inverse=0x1b15e0, tolerance=0.10000000000000001, antialias=CAIRO_ANTIALIAS_DEFAULT, clip=0x0) at cairo-surface-fallback.c:1173 9 0x0015f2f4 in _cairo_surface_stroke (surface=0x837db00, op=CAIRO_OPERATOR_OVER, source=0xbfffdda8, path=0x1b164c, stroke_style=0x1b1524, ctm=0x1b15b0, ctm_inverse=0x1b15e0, tolerance=0.10000000000000001, antialias=CAIRO_ANTIALIAS_DEFAULT, clip=0xbfffde88) at cairo-surface.c:2041 10 0x00134629 in _cairo_gstate_stroke (gstate=0x1b1514, path=0x1b164c) at cairo-gstate.c:1133 11 0x0012a706 in *INT_cairo_stroke_preserve (cr=0x1b13c0) at cairo.c:2234 12 0x0012a732 in *INT_cairo_stroke (cr=0x1b13c0) at cairo.c:2207 src/cairo-image-surface.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) commit 38ba696c483c289c2522b124ca628ef602c12126 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jan 25 09:20:11 2010 +0000 surface-fallback: We no longer own a reference to the clip surface _cairo_clip_get_surface() now returns a borrowed reference to the cached surface on the clip, so we must not destroy it - as Carlos pointed out when he hit the assert: 12:55 < KaL> ickle: cairo-surface.c:595: cairo_surface_destroy: Assertion `((*&(&surface->ref_count)->ref_count) > 0)' failed. 12:56 < KaL> ickle: trying to render any pdf file with poppler glib demo after installing cairo from git master 13:00 < KaL> ickle: well, it seems it has nothing ot do with poppler, since it crashes in clearlooks src/clearlooks_draw.c:347 src/cairo-surface-fallback.c | 3 --- 1 file changed, 3 deletions(-) commit b3ba610ec05f863a0830a1f30ed130649fbadfbb Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jan 25 09:09:18 2010 +0000 win32: Trivially fix the compile error with printing. _cairo_pattern_is_opaque() now takes the extents over which the operation is defined so that we make exclude the clear pixels that surround EXTEND_NONE surfaces when determining opacity. In order to take full advantage of this we need to start performing an extents query on the operation and pass that down to the analysis... This patch however is just the quick compile fix to pass a NULL and restore the old behaviour. Fixes (with previous commit): Bug 26197 - Cairo doesn't build on windows http://bugs.freedesktop.org/show_bug.cgi?id=26197 src/cairo-win32-printing-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 386ca53f280652983870316593e510cfef59c559 Author: Haakon Sporsheim <haakon.sporsheim@gmail.com> Date: Sun Jan 24 16:43:42 2010 +0100 win32: Compile fix, rename render_row to render_rows. src/cairo-win32-surface.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit a9a73e5df42840148fd4db60cc0d1ba5d137b764 Author: Andrea Canciani <ranma42@gmail.com> Date: Sun Jan 24 23:26:28 2010 +0100 Constrain subsurface creation Force the subsurface extents to be inside the target extents and compose subsubsurfaces offsets correctly. Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> src/cairo-surface-subsurface.c | 36 +++++++++++++++--------------------- 1 file changed, 15 insertions(+), 21 deletions(-) commit b7550bf0f1cafb5cc5df13da50b0cec80313038f Author: Adrian Johnson <ajohnson@redneon.com> Date: Sun Jan 24 22:23:26 2010 +1030 Fix incorrect font scale and glyph position in Type 3 fonts The PDF file referenced by bug 26186 contains a Type 3 font with non identity font matrix and a "1/2" glyph created by drawing the "1" and "2" from a Type 1 font. This combination exposed a bug in the font scale and glyph position in _cairo_type3_glyph_surface_show_glyphs when printing user font glyphs. src/cairo-type3-glyph-surface.c | 21 +++++---------------- 1 file changed, 5 insertions(+), 16 deletions(-) commit ea7fda442c169a159b2e3c51894a0f5e70a13984 Author: Eric Anholt <eric@anholt.net> Date: Fri Jan 22 16:26:38 2010 -0800 [gl] Release the context acquired in surface_create. Fixes assertion failures in many testcases since cf9ea9f2f0ea1e1b42dc1507ed0dac4dfbf5d387 src/cairo-gl-surface.c | 2 ++ 1 file changed, 2 insertions(+) commit 20990365f3885457d66b4a8bef865f027a98333f Author: Eric Anholt <eric@anholt.net> Date: Fri Jan 22 16:13:02 2010 -0800 [gl] Emit rectanges a whole primitive at a time. Since the spans rework, we were emitting half a primitive at a time, and if we flushed our VBO full of quads out halfway through, we could end up dropping the primitive and then out of phase. src/cairo-gl-surface.c | 25 +++++++++++-------------- 1 file changed, 11 insertions(+), 14 deletions(-) commit c8a61a3267ffd53d4b4d61e2e2bcb70448ea4d41 Author: Andrea Canciani <ranma42@gmail.com> Date: Sat Jan 23 18:25:11 2010 +0100 Fix compilation on MacOS X Some functions declared in MacOS X headers use the always_inline attribute, thus defining it to be a different expression breaks the build. src/cairo-botor-scan-converter.c | 8 ++++---- src/cairo-compiler-private.h | 4 ++-- src/cairo-gl-private.h | 4 ++-- 3 files changed, 8 insertions(+), 8 deletions(-) commit 90ba1242b61e7b8e5ae2c2ca028d0058f6dabd88 Author: Andrea Canciani <ranma42@gmail.com> Date: Sat Jan 23 18:13:35 2010 +0100 Fix image backend compilation When HAS_ATOMIC_OPS is not defined, cairo-image-surface.c does not compile because _pixman_white_image calls _pixman_image_for_solid which gets defined only later in the code. src/cairo-image-surface.c | 3 +++ 1 file changed, 3 insertions(+) commit 2a466ba07072cb394cfc955cfb0633b0cc278044 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 22 22:26:52 2010 +0000 script: Encode octal values correctly. Oops, we were emitting the octal numerals in reverse order. src/cairo-script-surface.c | 21 +++++++++++++-------- 1 file changed, 13 insertions(+), 8 deletions(-) commit 87529198aa37f5eb5943eb80cc1aae858fdaadd0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 22 22:21:44 2010 +0000 scaled-font: Mark an error site with _cairo_error() src/cairo-scaled-font-subsets.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 6544f3b25d2120b7bd04f6c956c2ea7b9c101c76 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 22 22:15:25 2010 +0000 ps: Wean off the low-level image composite interface The low-level surface composite interface will disappear in the near future and results in much more ugly code than calling the high level interface - so use it when flattening images into the page background. src/cairo-ps-surface.c | 77 ++++++++++++++++++-------------------------------- 1 file changed, 27 insertions(+), 50 deletions(-) commit f57cb45fa9274351b36f96eb932a931493b5db94 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 22 22:18:35 2010 +0000 ps: Arguments to fill_extents() were reversed. src/cairo-ps-surface.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit 58540405d9bc6581c1d07f17bd5aee330f06d4a2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 22 22:05:37 2010 +0000 pdf: Arguments to fill_extents() were reversed. src/cairo-pdf-surface.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit 34dd80632e4671f0ca245be9db4280db62bdb1f9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jan 21 18:39:26 2010 +0000 glew: Silence compiler. src/glew/GL/glew.h | 2 +- src/glew/glew.c | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) commit cf9ea9f2f0ea1e1b42dc1507ed0dac4dfbf5d387 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jan 21 18:39:12 2010 +0000 gl: Acquire the context whilst creating surfaces. src/cairo-gl-private.h | 5 ++--- src/cairo-gl-surface.c | 37 ++++++++++++++++++++++++------------- 2 files changed, 26 insertions(+), 16 deletions(-) commit a7e05bd602e7f8bee63f5d32cb95d65628e6e715 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jan 21 18:22:04 2010 +0000 glew: Do not assume that UNIX == GLX src/glew/glew.c | 5 +++++ 1 file changed, 5 insertions(+) commit c0008242b0f162d8c7717009ba792ed61b2924d1 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 22 21:35:23 2010 +0000 misc: Only use custom lround() under DISABLE_SOME_FLOATING_POINT On my Core2, the library version of lround() is faster than our hand-rolled non-floating point implementation. So only enable our code if we are trying to minimise the number of floating point operations -- even then, it would worth investigating the library performance first. [Just a reminder that optimisation choices will change over time as our hardware and software evolves.] src/cairo-misc.c | 2 ++ src/cairo-ps-surface.c | 4 ++-- src/cairoint.h | 5 +++++ 3 files changed, 9 insertions(+), 2 deletions(-) commit 1236c41072a7966eda7db48a381fd0508e5289be Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 22 21:26:26 2010 +0000 xcb: Refresh. Still an experimental backend, it's now a little too late to stabilise for 1.10, but this should represent a major step forward in its feature set and an attempt to catch up with all the bug fixes that have been performed on xlib. Notably not tested yet (and expected to be broken) are mixed-endian connections and low bitdepth servers (the dithering support has not been copied over for instance). However, it seems robust enough for daily use... Of particular note in this update is that the xcb surface is now capable of subverting the xlib surface through the ./configure --enable-xlib-xcb option. This replaces the xlib surface with a proxy that forwards all operations to an equivalent xcb surface whilst preserving the cairo-xlib API that is required for compatibility with the existing applications, for instance GTK+ and Mozilla. Also you can experiment with enabling a DRM bypass, though you need to be extremely foolhardy to do so. boilerplate/Makefile.win32.features | 20 + boilerplate/cairo-boilerplate-xcb.c | 522 ++- build/Makefile.win32.features | 2 + build/Makefile.win32.features-h | 6 + build/configure.ac.features | 2 + configure.ac | 26 +- src/Makefile.sources | 22 +- src/Makefile.win32.features | 28 + src/cairo-list-private.h | 7 + src/cairo-mutex-list-private.h | 4 + src/cairo-pattern.c | 41 +- src/cairo-xcb-connection-core.c | 482 +++ src/cairo-xcb-connection-render.c | 969 +++++ src/cairo-xcb-connection-shm.c | 194 + src/cairo-xcb-connection.c | 867 ++++ src/cairo-xcb-private.h | 760 ++++ src/cairo-xcb-screen.c | 518 +++ src/cairo-xcb-shm.c | 576 +++ src/cairo-xcb-surface-cairo.c | 94 + src/cairo-xcb-surface-core.c | 613 +++ ...o-xcb-xrender.h => cairo-xcb-surface-private.h} | 42 +- src/cairo-xcb-surface-render.c | 4471 ++++++++++++++++++++ src/cairo-xcb-surface.c | 3382 ++++----------- src/cairo-xcb.h | 41 +- src/cairo-xlib-xcb-surface.c | 515 +++ src/cairoint.h | 16 + test/xcb-surface-source.c | 1 - 27 files changed, 11679 insertions(+), 2542 deletions(-) commit 77afe8491ed7038a8399c01f10d8f062a7239225 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 22 20:48:54 2010 +0000 drm: Add backends for i915 and i965. As proof-of-principle add the nearly working demonstrations of using DRM to render directly with the GPU bypassing both RENDER and GL for performance whilst preserving high quality rendering. The basis behind developing these chip specific backends is that this is the idealised interface that we desire for this chips, and so a target for cairo-gl as we continue to develop both it and our GL stack. Note that this backends do not yet fully pass the test suite, so only use if you are brave and willing to help develop them further. boilerplate/cairo-boilerplate-drm.c | 2 +- src/Makefile.sources | 17 + src/cairo-drm.h | 31 +- src/cairo-misc.c | 57 + src/cairoint.h | 2 + src/drm/cairo-drm-bo.c | 13 +- src/drm/cairo-drm-gallium-surface.c | 3 +- src/drm/cairo-drm-i915-glyphs.c | 534 ++++++ src/drm/cairo-drm-i915-private.h | 1169 ++++++++++++ src/drm/cairo-drm-i915-shader.c | 2674 +++++++++++++++++++++++++++ src/drm/cairo-drm-i915-spans.c | 708 +++++++ src/drm/cairo-drm-i915-surface.c | 1996 ++++++++++++++++++++ src/drm/cairo-drm-i965-glyphs.c | 500 +++++ src/drm/cairo-drm-i965-private.h | 742 ++++++++ src/drm/cairo-drm-i965-shader.c | 2852 +++++++++++++++++++++++++++++ src/drm/cairo-drm-i965-spans.c | 408 +++++ src/drm/cairo-drm-i965-surface.c | 1949 ++++++++++++++++++++ src/drm/cairo-drm-intel-brw-defines.h | 824 +++++++++ src/drm/cairo-drm-intel-brw-eu-emit.c | 1089 +++++++++++ src/drm/cairo-drm-intel-brw-eu-util.c | 121 ++ src/drm/cairo-drm-intel-brw-eu.c | 250 +++ src/drm/cairo-drm-intel-brw-eu.h | 1043 +++++++++++ src/drm/cairo-drm-intel-brw-structs.h | 1328 ++++++++++++++ src/drm/cairo-drm-intel-command-private.h | 909 +++++++++ src/drm/cairo-drm-intel-debug.c | 1208 ++++++++++++ src/drm/cairo-drm-intel-ioctl-private.h | 417 +++++ src/drm/cairo-drm-intel-private.h | 427 ++++- src/drm/cairo-drm-intel-surface.c | 421 +++-- src/drm/cairo-drm-intel.c | 1240 +++++++++---- src/drm/cairo-drm-private.h | 101 +- src/drm/cairo-drm-radeon-private.h | 3 - src/drm/cairo-drm-radeon-surface.c | 281 +-- src/drm/cairo-drm-radeon.c | 5 +- src/drm/cairo-drm-surface.c | 235 +-- src/drm/cairo-drm.c | 232 +-- 35 files changed, 22675 insertions(+), 1116 deletions(-) commit b9407af6a4bc792c1bcb52c90aa8a618627bb618 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 22 17:57:41 2010 +0000 image: Implement high level interface. Write a dedicated compositor for pixman so that we avoid the middle-layer syndrome of surface-fallback. The major upshot of this rewrite is that the image surface is now several times quicker for glyph compositing, which dramatically improves performance for text rendering by firefox and friends. It also uses a couple of the new scan convertors, such as the rectangular scan converter for rectilinear paths. Speedups ======== image-rgba firefox-talos-gfx-0 342050.17 (342155.88 0.02%) -> 69412.44 (69702.90 0.21%): 4.93x speedup ███▉ image-rgba vim-0 97518.13 (97696.23 1.21%) -> 30712.63 (31238.65 0.85%): 3.18x speedup ██▏ image-rgba evolution-0 69927.77 (110261.08 19.84%) -> 24430.05 (25368.85 1.89%): 2.86x speedup █▉ image-rgba poppler-0 41452.61 (41547.03 2.51%) -> 21195.52 (21656.85 1.08%): 1.96x speedup █ image-rgba firefox-planet-gnome-0 217512.61 (217636.80 0.06%) -> 123341.02 (123641.94 0.12%): 1.76x speedup ▊ image-rgba swfdec-youtube-0 41302.71 (41373.60 0.11%) -> 31343.93 (31488.87 0.23%): 1.32x speedup ▍ image-rgba swfdec-giant-steps-0 20699.54 (20739.52 0.10%) -> 17360.19 (17375.51 0.04%): 1.19x speedup ▎ image-rgba gvim-0 167837.47 (168027.68 0.51%) -> 151105.94 (151635.85 0.18%): 1.11x speedup ▏ image-rgba firefox-talos-svg-0 375273.43 (388250.94 1.60%) -> 356846.09 (370370.08 1.86%): 1.05x speedup src/cairo-image-surface.c | 3987 ++++++++++++++++++++++++----- src/cairo-mutex-list-private.h | 2 + src/cairo-xcb-surface.c | 10 +- src/cairoint.h | 21 +- test/clip-fill-unbounded.argb32.ref.png | Bin 1615 -> 1607 bytes test/clip-fill-unbounded.rgb24.ref.png | Bin 1312 -> 1304 bytes test/clip-stroke-unbounded.argb32.ref.png | Bin 1703 -> 1694 bytes test/clip-stroke-unbounded.rgb24.ref.png | Bin 1383 -> 1372 bytes test/clip-stroke.ref.png | Bin 1451 -> 1442 bytes test/clipped-group.ref.png | Bin 289 -> 289 bytes test/leaky-dashed-rectangle.ref.png | Bin 347 -> 357 bytes test/scale-offset-image.xfail.png | Bin 9960 -> 9961 bytes test/scale-offset-similar.xfail.png | Bin 9960 -> 9961 bytes test/self-intersecting.ref.png | Bin 213 -> 168 bytes 14 files changed, 3342 insertions(+), 678 deletions(-) commit ec7cc9dea16f94e1e56720be548cf5520ec4708b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 22 19:27:45 2010 +0000 ft: Tidy computation of transformed bitmap, and mark if CA Discard a redundant clear as the image surface is guaranteed to return a cleared surface that meets pixman/xlib requirements for alignment, and more importantly add the ComponentAlpha flag on the pixman image generated as appropriate. src/cairo-ft-font.c | 31 +++++++++++-------------------- 1 file changed, 11 insertions(+), 20 deletions(-) commit b713510f213d0a83cb0575d6870416a0f13786c9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 22 17:56:26 2010 +0000 surface-fallback: Convert to composite rectangles src/cairo-surface-fallback.c | 396 ++++++++++++++----------------------------- src/cairo-surface.c | 3 + src/cairo-traps.c | 47 ++--- src/cairoint.h | 3 +- 4 files changed, 158 insertions(+), 291 deletions(-) commit 498c10032ea3f8631a928cd7df96766f2c8ddca4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 22 16:39:29 2010 +0000 clip: Implement clipping using only the high-level operators. Revamp clipping in preparation for the removal of the low-level interface and promote backend to use the higher levels. The principle here is that the higher level interface gives the backend more scope for choosing better performing primitives. src/cairo-clip-private.h | 6 +- src/cairo-clip.c | 257 +++++++++++++++++++++---------------------- src/cairo-surface-fallback.c | 2 +- 3 files changed, 131 insertions(+), 134 deletions(-) commit 05bb43a00770f17566c80f28faf684597d1b6afb Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 22 16:14:53 2010 +0000 types: Move cairo_scaled_glyph_t to cairo-types-private.h Another step in bringing some sanity to our headers. src/cairo-types-private.h | 25 +++++++++++++++++++++++++ src/cairoint.h | 26 -------------------------- 2 files changed, 25 insertions(+), 26 deletions(-) commit 9c98656fb1ac10a6085c5db5731b6b986af26165 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 22 16:11:48 2010 +0000 color: Add enum value for the largest stock color. Useful if we wish to keep a cache of patterns for the very common stock colors... src/cairo-color.c | 13 +++++++------ src/cairo-types-private.h | 3 ++- 2 files changed, 9 insertions(+), 7 deletions(-) commit b7b65b5315836297f86c54121ea45e469f514312 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 22 16:07:25 2010 +0000 stroke: Convert rectilinear paths to boxes. A quick and efficient method for extracting the set of boxes that represent the path when stroked. src/cairo-path-stroke.c | 170 ++++++++++++++++++++++++++++++++++++++---------- src/cairoint.h | 7 ++ 2 files changed, 142 insertions(+), 35 deletions(-) commit fe6a7f4c9e69c86588eca4925a9d65f3086f98bf Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 22 16:02:07 2010 +0000 fill: Convert rectilinear paths to boxes. A quick and efficient method for extracting the set of boxes that represent the path when filled. src/cairo-path-fill.c | 76 ++++++++++++++++++++++++++++++++++++++++++++++++++- src/cairoint.h | 5 ++++ 2 files changed, 80 insertions(+), 1 deletion(-) commit 6b77567b6ef28710c7707ab82c7fa95c810152d1 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 22 15:54:45 2010 +0000 path: Compute coarse bounds upon construction. Frequently we only need the coarse path bounds, so avoid walking over the list of points once more as we can cheaply track the extents during construction. src/cairo-debug.c | 6 ++ src/cairo-gstate.c | 21 ++++-- src/cairo-path-bounds.c | 159 ++++++++++++++++++++--------------------- src/cairo-path-fixed-private.h | 4 +- src/cairo-path-fixed.c | 83 +++++++++++++++++++-- src/cairo.c | 2 + src/cairoint.h | 9 ++- 7 files changed, 184 insertions(+), 100 deletions(-) commit 6bfcf3ea55964fee1c9b73818c3bb7cdacbc4f82 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 22 15:37:25 2010 +0000 matrix: Do not need to iteratively refine conversion for unity matrices. If the scale factors on the matrix are unity, then the conversion from a cairo_matrix_t to a pixman_matrix_t is exact (within numerical restrictions). src/cairo-matrix.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 13c56800ca6c32a6d8251edec5d3976399b70ea4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 22 16:33:34 2010 +0000 gstate: Reduce operators Some operators may be equivalent to simpler operations, so make the transformation in the gstate, before calling down into the surface backends. src/cairo-gstate.c | 86 +++++++++++++++++++++++++++++++++++++++++++++--------- 1 file changed, 72 insertions(+), 14 deletions(-) commit 4d52be39bf36a3557f4e7c61dba764b1c6542c8f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 22 16:09:11 2010 +0000 gstate: Skip ops with a clear mask. As pointed out by Benjamin Otte, these are expensive no-ops that we can trivially detect, just so long as we remember the semantics of extend modes. src/cairo-gstate.c | 6 ++++++ src/cairo-pattern.c | 23 +++++++++++++++++++++++ src/cairoint.h | 3 +++ 3 files changed, 32 insertions(+) commit c2ef45226d6801acb15854de55a5f5bc728ed65d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 22 14:33:53 2010 +0000 gstate: Check for an invalid matrix before cairo_transform() Similar to the other checks that will generate an invalid matrix prior to performing the operation. src/cairo-gstate.c | 3 +++ 1 file changed, 3 insertions(+) commit 22316a0c17cd3b49ec625814f4e737b3fdc6ca63 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 22 14:33:16 2010 +0000 gstate: White space. A couple of minor coding style violations. src/cairo-gstate.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) commit 29bedde904824098f73b978d73d36b3b5420ac32 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 22 16:19:31 2010 +0000 pattern: Add convenience patterns for stock colours By preallocating in our data segment a couple of solid patterns for the stock colours, it becomes more convenient when using those in surface operations, such as when clearing. src/cairo-clip.c | 29 +++++++++++------------------ src/cairo-gl-glyphs.c | 10 +++------- src/cairo-gstate.c | 9 +++------ src/cairo-pattern.c | 24 ++++++++++++++++++++++++ src/cairo-scaled-font.c | 9 ++------- src/cairo-surface-fallback.c | 14 +++----------- src/cairo-vg-surface.c | 5 ++--- src/cairo-win32-printing-surface.c | 8 ++------ src/cairo-xcb-surface.c | 7 +------ src/cairo-xlib-surface.c | 6 +----- src/cairoint.h | 3 ++- 11 files changed, 54 insertions(+), 70 deletions(-) commit 9eb98e1dc5af7eac896bf469e491063777e678be Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 22 14:14:51 2010 +0000 scaled-font: Use tight glyph bounds when determining overlap. An issue that we currently have is that we have a pessimistic false-positive rate when determining whether glyphs within a string overlap. By using the tight bounds, the overlap detection is arguably less accurate presuming pixel-aligned opacity masks but we make the trade-off for performance. src/cairo-scaled-font.c | 55 ++++++++++++++++++++++--------------------------- 1 file changed, 25 insertions(+), 30 deletions(-) commit 005b0c4eb2a1fe2c4de31ebe2a3244bea8bd2625 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 22 19:47:58 2010 +0000 test: Add explicit device management to xlib-surface-source Mostly pedagogical example. test/xlib-surface-source.c | 5 +++++ 1 file changed, 5 insertions(+) commit ca02b51ef6d058b6fb492258177f0396a5819e26 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 22 21:57:58 2010 +0000 test: Add an xcb surface source test test/Makefile.am | 6 ++ test/Makefile.sources | 3 + test/xcb-surface-source.argb32.ref.png | Bin 0 -> 377 bytes test/xcb-surface-source.c | 145 +++++++++++++++++++++++++++++++++ test/xcb-surface-source.rgb24.ref.png | Bin 0 -> 301 bytes 5 files changed, 154 insertions(+) commit f0678fb70ceec5db1b7c3f0eb1c5603039daf307 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 22 19:02:39 2010 +0000 test: Tweak aligned of text-rotate Reset the rotation for every quadrant so that the starting rectangles are pixel-aligned. test/text-rotate.c | 68 ++++++++++++++++++++++++++++++++++------------- test/text-rotate.ref.png | Bin 16738 -> 16722 bytes 2 files changed, 49 insertions(+), 19 deletions(-) commit 2edd2adafc471f4aa9c417d4bc76cc38466a9ed0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jan 19 22:45:59 2010 +0000 test: Exercise scan converters with large rounded rectangles. Having added a specialised scan converter on the premise that it should be better at handling rounded rectangles, ensure that they are indeed rendered correctly. test/Makefile.am | 2 + test/Makefile.sources | 2 + test/rounded-rectangle-fill.c | 65 +++++++++++++++++++++++++++++ test/rounded-rectangle-fill.xlib.ref.png | Bin 0 -> 890 bytes test/rounded-rectangle-stroke.c | 64 ++++++++++++++++++++++++++++ test/rounded-rectangle-stroke.xlib.ref.png | Bin 0 -> 860 bytes 6 files changed, 133 insertions(+) commit 5187501e997d6b8c0d135542d21accde9e311281 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jan 19 22:41:45 2010 +0000 test: Add --track-origins=yes to valgrind flags. Enable origin tracking by default for make check-valgrind. This is slower and consumes more memory than regular valgrind, but the additional information provided about the source of the uninitialised data is often invaluable. test/Makefile.am | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) commit 3a5d71c431dfb251308c3e4e02d2ea4acec90a91 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jan 19 19:18:54 2010 +0000 pattern: An EXTEND_NONE surface is not opaque if we sample outside Fixes test/clear-source as proposed by Benjamin Otte. src/cairo-gstate.c | 2 +- src/cairo-pattern.c | 33 ++++++- src/cairo-pdf-surface.c | 219 ++++++++++++++++++++++++++++------------------- src/cairo-ps-surface.c | 186 ++++++++++++++++++++++++---------------- src/cairo-xlib-surface.c | 8 +- src/cairoint.h | 3 +- 6 files changed, 284 insertions(+), 167 deletions(-) commit b07de014eba1a482747ac235670011972a82d19d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jan 19 18:44:34 2010 +0000 spans: Add a rectangular scan converter This is a highly specialised scan converter for the relatively common case of where the input geometry is known to be a series of rectangles. Generally not device aligned (or else we would most likely have chosen an even higher performance path that does not require a coverage mask), this optimised converter can simply compute the analytical coverage by utilising a special case Bentley-Ottmann intersection finder. src/Makefile.sources | 1 + src/cairo-rectangular-scan-converter.c | 723 +++++++++++++++++++++++++++++++++ src/cairo-spans-private.h | 25 ++ 3 files changed, 749 insertions(+) commit e49855497e7214c21c85b03d7ab94e9e008f821b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jan 19 18:36:20 2010 +0000 spans: Add a Bentley-Ottmann variant on the Tor scan converter This variant uses the Bentley-Ottmann algorithm to only maintain the active edge list upon edge events and so can efficiently skip areas where no change occurs. This means that it can be much quicker than the Tor algorithm (which is still used to compute the coverages from the active edges) for geometries consisting of long straight lines with few intersections. However due to the computational overhead of the Bentley-Ottmann event processing, for dense curvy paths, simply updating the active edge list in sync with computing the coverages is a win. Due to advantageous adaptive step size, the scan converter can be run at a much higher subsampling with little extra overhead compared with Tor, currently it uses a 256x256 subsampling grid to avoid any impedance mismatch with path precision. Given the current status of implementations, this scan converter [botor] is likely to be advantage where detecting large regions of unchanged span data will result in improved performance, for instance the drm backends which convert the scan data into rectangles. src/Makefile.sources | 1 + src/cairo-botor-scan-converter.c | 2199 ++++++++++++++++++++++++++++++++++++++ src/cairo-compiler-private.h | 11 + src/cairo-fixed-private.h | 22 +- src/cairo-freelist-private.h | 16 + src/cairo-freelist.c | 38 +- src/cairo-spans-private.h | 23 + src/cairoint.h | 9 - 8 files changed, 2299 insertions(+), 20 deletions(-) commit 424dcf296378a107286a164aaf135a34e40c42ac Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jan 19 19:11:31 2010 +0000 boxes: Enable generation of boxes from rectilinear polygons. Extend the special case rectilinear tessellator to also produce boxes. src/cairo-bentley-ottmann-rectilinear.c | 122 +++++++++++++++++++++++++++----- src/cairoint.h | 5 ++ 2 files changed, 108 insertions(+), 19 deletions(-) commit b83f1c347dfd77139e9485745d43da946b086b74 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jan 19 19:09:10 2010 +0000 boxes: Enable tessellation Extend the special case rectangular tessellator to handle generation of cairo_boxes_t. src/cairo-bentley-ottmann-rectangular.c | 778 +++++++++++++++++--------------- src/cairoint.h | 5 + 2 files changed, 416 insertions(+), 367 deletions(-) commit 43beaa5873b9ad10620bfe7ed5f9212a3c44effd Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jan 19 18:11:26 2010 +0000 boxes: Efficient storage for an array of cairo_box_t. Currently we use cairo_traps_t to also pass around arrays of boxes. This is woefully inefficient in terms of storage, but also means that we repeatedly have to verify that the traps are a set of boxes. By explicitly passing around a cairo_boxes_t we avoid the semantic loss. This will be heavily used in pending commits. src/Makefile.sources | 2 + src/cairo-boxes-private.h | 80 ++++++++++++++ src/cairo-boxes.c | 269 ++++++++++++++++++++++++++++++++++++++++++++++ src/cairo-types-private.h | 1 + 4 files changed, 352 insertions(+) commit ae25f1c360b79f0b7b1bb73e9ebc47eb794d8007 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jan 19 17:11:55 2010 +0000 Alter definition of cairo_composite_rectangles_t This is a more useful definition that is able to individually track the rectangles that compose the composite operation. This will be used by the specialist compositors as a means to perform the common extents determination for an operation. src/Makefile.sources | 2 + src/cairo-composite-rectangles-private.h | 105 ++++++++++++++++ src/cairo-composite-rectangles.c | 197 +++++++++++++++++++++++++++++++ src/cairo-gl-surface.c | 19 +-- src/cairo-image-surface.c | 31 ++--- src/cairo-misc.c | 41 +++++++ src/cairo-rectangle.c | 21 ---- src/cairo-spans.c | 29 +++-- src/cairo-surface-fallback.c | 13 +- src/cairo-types-private.h | 24 +--- src/cairo-win32-surface.c | 21 ++-- src/cairoint.h | 14 +-- 12 files changed, 417 insertions(+), 100 deletions(-) commit 9cd9137843f8f1c3d32bedb6510259ab3638a2c5 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jan 19 17:11:11 2010 +0000 spans: Pass multiple rows of identical spans to the renders. It is quite common amongst our geometry to have rows of repeated span data, for example a rounded rectangle will have repeating data between the top and bottom rounded corners. By passing the repeat length to the renderers, they may be able to use that information more efficiently, and the scan converters can avoid recomputing the same span data. src/cairo-gl-surface.c | 141 +++++++++-------- src/cairo-image-surface.c | 79 ++++------ src/cairo-spans-private.h | 24 ++- src/cairo-spans.c | 6 +- src/cairo-tor-scan-converter.c | 343 +++++++++++++++++++++++++---------------- src/cairo-win32-surface.c | 13 +- src/cairoint.h | 4 +- 7 files changed, 349 insertions(+), 261 deletions(-) commit a04873c0770df5eaed078493df5216ca28322de7 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jan 18 23:40:00 2010 +0000 qt: Trivial compile fix for boilerplate. boilerplate/cairo-boilerplate-qt.cpp | 2 ++ 1 file changed, 2 insertions(+) commit 0f09349dd85b15d5de14248e2f3f17a91fbcbb5e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jan 18 23:37:36 2010 +0000 gl: Exercise Window destinations with boilerplate Add a gl-window boilerplate target to exercise using GL to render to a visible Drawable -- for instance, a window has a different coordinate system to a framebuffer... boilerplate/cairo-boilerplate-gl.c | 99 ++++++++++++++++++++++++++++++++++++++ 1 file changed, 99 insertions(+) commit ccea7fd7c1a2d7144e892c19615488e368529bc0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jan 18 23:37:11 2010 +0000 gl: Port to cairo_device_t boilerplate/cairo-boilerplate-gl.c | 28 +++-- src/cairo-eagle-context.c | 25 +++-- src/cairo-gl-glyphs.c | 46 +++++--- src/cairo-gl-private.h | 44 +++++--- src/cairo-gl-surface.c | 222 ++++++++++++++++--------------------- src/cairo-gl.h | 22 ++-- src/cairo-glx-context.c | 25 +++-- 7 files changed, 211 insertions(+), 201 deletions(-) commit 3acd520c9dec89e72d7ff61adb1ae30bab12e256 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jan 18 23:11:19 2010 +0000 xml: Port to cairo_device_t src/cairo-xml-surface.c | 210 ++++++++++++++++++++++++------------------------ src/cairo-xml.h | 13 +-- util/trace-to-xml.c | 7 +- 3 files changed, 115 insertions(+), 115 deletions(-) commit 49ab86772a44d7a044464d875324bd0af96af728 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jan 18 22:43:24 2010 +0000 script: Port cairo_script_context_t to cairo_device_t Use the unifying cairo_device_t for cairo_script_context_t and replace. boilerplate/cairo-boilerplate-script.c | 6 +- boilerplate/cairo-boilerplate.c | 6 +- src/cairo-script-surface.c | 491 ++++++++++++++++++--------------- src/cairo-script.h | 35 +-- test/cairo-test-trace.c | 7 +- util/cairo-fdr/fdr.c | 10 +- util/cairo-script/csi-trace.c | 11 +- util/cairo-sphinx/fdr.c | 4 +- util/cairo-sphinx/sphinx.c | 17 +- util/cairo-trace/trace.c | 8 +- 10 files changed, 312 insertions(+), 283 deletions(-) commit f617d5fc982f749d0981c81c1de1be8dc3632717 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jan 18 21:53:42 2010 +0000 Add cairo_device_t The device is a generic method for accessing the underlying interface with the native graphics subsystem, typically the X connection or perhaps the GL context. By exposing a cairo_device_t on a surface and its various methods we enable finer control over interoperability with external interactions of the device by applications. The use case in mind is, for example, a multi-threaded gstreamer which needs to serialise its own direct access to the device along with Cairo's across many threads. Secondly, the cairo_device_t is a unifying API for the mismash of backend specific methods for controlling creation of surfaces with explicit devices and a convenient hook for debugging and introspection. The principal components of the API are the memory management of: cairo_device_reference(), cairo_device_finish() and cairo_device_destroy(); along with a pair of routines for serialising interaction: cairo_device_acquire() and cairo_device_release() and a method to flush any outstanding accesses: cairo_device_flush(). The device for a particular surface may be retrieved using: cairo_surface_get_device(). The device returned is owned by the surface. src/Makefile.sources | 2 + src/cairo-analysis-surface.c | 9 +- src/cairo-beos-surface.cpp | 5 +- src/cairo-device-private.h | 85 ++++++++++++ src/cairo-device.c | 265 ++++++++++++++++++++++++++++++++++++ src/cairo-directfb-surface.c | 2 + src/cairo-gl-surface.c | 1 + src/cairo-glitz-surface.c | 4 +- src/cairo-image-surface.c | 4 +- src/cairo-misc.c | 4 + src/cairo-os2-surface.c | 1 + src/cairo-paginated-surface.c | 4 +- src/cairo-pdf-surface.c | 4 +- src/cairo-ps-surface.c | 4 +- src/cairo-qt-surface.cpp | 7 +- src/cairo-quartz-image-surface.c | 1 + src/cairo-quartz-surface.c | 6 +- src/cairo-recording-surface.c | 9 +- src/cairo-script-surface.c | 1 + src/cairo-skia-surface.cpp | 1 + src/cairo-spans.c | 4 + src/cairo-surface-private.h | 1 + src/cairo-surface-snapshot.c | 1 + src/cairo-surface-subsurface.c | 1 + src/cairo-surface.c | 29 ++++ src/cairo-svg-surface.c | 4 +- src/cairo-tee-surface.c | 1 + src/cairo-type3-glyph-surface.c | 4 +- src/cairo-types-private.h | 1 + src/cairo-vg-surface.c | 1 + src/cairo-win32-printing-surface.c | 1 + src/cairo-win32-surface.c | 8 +- src/cairo-xcb-surface.c | 4 +- src/cairo-xlib-surface.c | 4 +- src/cairo-xml-surface.c | 1 + src/cairo.h | 117 ++++++++++++++++ src/cairoint.h | 1 + src/drm/cairo-drm-gallium-surface.c | 2 + src/drm/cairo-drm-intel-surface.c | 5 +- src/drm/cairo-drm-radeon-surface.c | 5 +- src/test-fallback-surface.c | 4 +- src/test-fallback16-surface.c | 1 + src/test-null-surface.c | 5 +- src/test-paginated-surface.c | 4 +- src/test-wrapping-surface.c | 1 + util/cairo-trace/trace.c | 2 + 46 files changed, 607 insertions(+), 24 deletions(-) commit 82f8aa548d70acf51b319000d7a5c176fc73da64 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jan 18 22:02:09 2010 +0000 Recursive mutex. Implement a recursive mutex which will be needed for cairo_device_t. In particular only pthreads by default is a non-recursive mutex (to my knowledge) - both win32 critical sections and mutexes on Quartz are recursive. src/cairo-mutex-impl-private.h | 28 ++++++++++++++++++++++++++++ src/cairo-mutex-type-private.h | 9 ++++++++- 2 files changed, 36 insertions(+), 1 deletion(-) commit d2f251f84b326c5f8825c9a631ea729f21891580 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jan 18 22:12:23 2010 +0000 configure: Globally define AC_GNU_SOURCE We were using _GNU_SOURCE throughout the codebase, so simply define it once during configure. This is the easiest method to enable recursive mutexes using pthreads, as required in a pending patch. boilerplate/cairo-boilerplate-pdf.c | 4 ++++ boilerplate/cairo-boilerplate-ps.c | 4 ++++ configure.ac | 3 +++ src/cairo-gstate.c | 2 -- src/cairo-matrix.c | 2 -- src/cairo-scaled-font.c | 2 -- 6 files changed, 11 insertions(+), 6 deletions(-) commit 934d0d0d6585eb7638c7db597c40dd821092c034 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jan 18 19:05:34 2010 +0000 Real zero-copy cow snapshotting The first iteration of COW snapshotting always made an initial copy when the snapshot was requested (and reused that copy until the surface was modified). However, in a few circumstances we can avoid even that copy so long as the surface is still alive and unmodified between the snapshotting and its use. In order to do so, we need a new proxy surface that can automatically perform the copy if the target should disappear prior to use. src/Makefile.sources | 2 + src/cairo-paginated-surface.c | 7 +- src/cairo-surface-snapshot-private.h | 48 ++++++++ src/cairo-surface-snapshot.c | 225 +++++++++++++++++++++++++++++++++++ src/cairo-surface.c | 101 ++-------------- src/cairo-types-private.h | 2 + 6 files changed, 291 insertions(+), 94 deletions(-) commit 6b3e19aa434d159db7878b2a7fb28a026b2d1756 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jan 18 19:12:25 2010 +0000 Remove content matching from snapshot lookup. This was never used, so remove the complexity from the interface. src/cairo-recording-surface.c | 3 +-- src/cairo-surface.c | 23 +++++++---------------- src/cairo-vg-surface.c | 3 +-- src/cairoint.h | 3 +-- src/drm/cairo-drm-intel-surface.c | 3 +-- src/drm/cairo-drm-radeon-surface.c | 3 +-- 6 files changed, 12 insertions(+), 26 deletions(-) commit dc8290814c0e98b38828953bac6dd0893c31c9ad Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jan 18 18:06:51 2010 +0000 Add subsurface. A subsurface is a region of another surface that may be used either to restrict the writable area of a context or the readable extents of a source. Whilst writing, access to the exterior of the subsurface is prevented via clipping and when used as a source reads from the exterior of the subsurface are governed via the extend mechanism of the pattern. src/Makefile.sources | 2 + src/cairo-surface-subsurface-private.h | 48 ++++ src/cairo-surface-subsurface.c | 464 +++++++++++++++++++++++++++++++++ src/cairo-types-private.h | 4 +- src/cairo.h | 7 + test/Makefile.am | 3 + test/Makefile.sources | 3 + test/subsurface-repeat.c | 72 +++++ test/subsurface-repeat.ref.png | Bin 0 -> 197 bytes test/subsurface-similar-repeat.c | 83 ++++++ test/subsurface-similar-repeat.ref.png | Bin 0 -> 197 bytes test/subsurface.c | 80 ++++++ test/subsurface.ref.png | Bin 0 -> 1597 bytes 13 files changed, 765 insertions(+), 1 deletion(-) commit b8eacbfae1c155f412d84120411103cb610e383b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jan 18 18:09:11 2010 +0000 Add surface-offset internal API. This is a simplified version of the wrapping surface where the target surface is just a subsurface onto which we wish to draw the current operation. In particular this is useful for the subsurface API as well as fallbacks. src/Makefile.sources | 2 + src/cairo-surface-offset-private.h | 95 ++++++++++ src/cairo-surface-offset.c | 354 +++++++++++++++++++++++++++++++++++++ 3 files changed, 451 insertions(+) commit f2c32d01836379766bc287edf77381123767daeb Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jan 19 18:06:54 2010 +0000 Unify the two freed object pools Discard some duplicate code and shared a single freed object pointer pool between the pattern and clip. src/Makefile.sources | 2 + src/cairo-clip.c | 88 +----------------------------- src/cairo-freed-pool-private.h | 121 +++++++++++++++++++++++++++++++++++++++++ src/cairo-freed-pool.c | 91 +++++++++++++++++++++++++++++++ src/cairo-pattern.c | 115 ++++++--------------------------------- 5 files changed, 234 insertions(+), 183 deletions(-) commit cfd204824fada7d2b4bcf4994c4200ae9b5a8b26 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jan 18 18:20:16 2010 +0000 Constify stroke style and matrices. As a simple step to ensure that we do not inadvertently modify (or at least generate compiler warns if we try) user data, mark the incoming style and matrices as constant. src/cairo-analysis-surface.c | 12 ++++---- src/cairo-paginated-surface.c | 6 ++-- src/cairo-path-bounds.c | 6 ++-- src/cairo-path-stroke.c | 60 ++++++++++++++++++------------------ src/cairo-pdf-operators-private.h | 24 +++++++-------- src/cairo-pdf-operators.c | 38 +++++++++++------------ src/cairo-pdf-surface-private.h | 11 +++---- src/cairo-pdf-surface.c | 40 ++++++++++++------------ src/cairo-ps-surface.c | 6 ++-- src/cairo-qt-surface.cpp | 14 ++++----- src/cairo-quartz-surface.c | 6 ++-- src/cairo-recording-surface.c | 6 ++-- src/cairo-script-surface.c | 6 ++-- src/cairo-stroke-style.c | 2 +- src/cairo-surface-fallback-private.h | 6 ++-- src/cairo-surface-fallback.c | 6 ++-- src/cairo-surface-wrapper-private.h | 12 ++++---- src/cairo-surface-wrapper.c | 12 ++++---- src/cairo-surface.c | 14 ++++----- src/cairo-svg-surface.c | 40 ++++++++++++------------ src/cairo-tee-surface.c | 6 ++-- src/cairo-type3-glyph-surface.c | 6 ++-- src/cairo-win32-printing-surface.c | 6 ++-- src/cairo-xml-surface.c | 6 ++-- src/cairoint.h | 44 +++++++++++++------------- src/drm/cairo-drm-private.h | 6 ++-- src/drm/cairo-drm-surface.c | 6 ++-- src/test-null-surface.c | 6 ++-- src/test-paginated-surface.c | 6 ++-- src/test-wrapping-surface.c | 6 ++-- 30 files changed, 212 insertions(+), 213 deletions(-) commit 3e17c559b04e2df208f28125746d7e0acfb4b476 Author: Benjamin Otte <otte@gnome.org> Date: Wed Dec 9 19:46:28 2009 +0100 [test] Add pthread-same-source test Multiple threads render using the same source surface, but with different extend and filter. test/Makefile.am | 1 + test/Makefile.sources | 1 + test/pthread-same-source.c | 160 +++++++++++++++++++++++++++++++++++++++ test/pthread-same-source.ref.png | Bin 0 -> 1000 bytes 4 files changed, 162 insertions(+) commit 30a12f802f4ff46f13dc36e60936cd857585a9ea Author: Benjamin Otte <otte@gnome.org> Date: Wed Nov 25 18:55:05 2009 +0100 [test] Add a simple multithreading test test/Makefile.am | 1 + test/Makefile.sources | 5 +- test/pthread-similar.c | 106 +++++++++++++++++++++++++++++++++++++++++++ test/pthread-similar.ref.png | Bin 0 -> 170 bytes 4 files changed, 111 insertions(+), 1 deletion(-) commit 038829eb84e53532a6e3c75225933ad85da50cfc Author: Benjamin Otte <otte@gnome.org> Date: Sun Dec 6 17:47:50 2009 +0100 [test] Add new test checking is_clear is handled properly This test does various operations with a clear source surface with various content types. The idea is to make sure optimizations done when surface->is_clear == TRUE only happen in the proper cases. test/Makefile.sources | 1 + test/clear-source.c | 163 ++++++++++++++++++++++++++++++++++++++++++++++ test/clear-source.ref.png | Bin 0 -> 881 bytes 3 files changed, 164 insertions(+) commit bf7fb4e0e09d05f42425cc7969c64992d4dc9ff9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jan 18 17:18:08 2010 +0000 script: Handle cache allocation failure more gracefully Instead of bailing out if we cannot store the glyph cache on the font, simply do not store the glyph cache on the font... util/cairo-script/cairo-script-operators.c | 74 +++++++++++++----------------- 1 file changed, 33 insertions(+), 41 deletions(-) commit 2b312806f138da6f88d03bb3c0c3852b342eb220 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jan 18 17:14:24 2010 +0000 script: Free the correct pattern after failure. The error path attempted to free the resolved pattern which it had just discovered was NULL and not the locally allocated pattern... util/cairo-script/cairo-script-operators.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit bc2d0ad114f8fc0b0579851154fe9d39c0aa0fc9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Nov 27 17:30:51 2009 +0000 script: Permit surface operations on the context By implicitly reference the target of the context instead, i.e. this reduces the use of: /target get (example.png) write-to-png pop as a common idiom where the context is kept on the stack and the surface forgotten. util/cairo-script/cairo-script-operators.c | 14 +++++++++++--- 1 file changed, 11 insertions(+), 3 deletions(-) commit 411c09eed7ebff7335948a7f96d2b29d82ed817a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Nov 1 08:40:41 2009 +0000 perf: Enable a surface cache for perf-trace Real applications that control their Drawable externally to Cairo are 'disadvantaged' by cairo-perf-trace when it creates a similar surface for each new instance of the same Drawable. The difficulty in maintaining one perf surface for every application surface is that the traces do not track lifetimes for the application surfaces, so we would just accumulate stale surfaces. The surface cache takes a different approach and returns the same surface for each active Drawable, and maintains a hold-over of the MRU 16 surfaces. This achieves 60-80% hit rate with firefox, which is probably as good as can be expected. Obviously for double-buffered applications we only every draw to freshly created surfaces (and Gtk+ bypasses cairo to do the final copy -- the ideal application would just use a push-group for double buffering, in which case we would capture and replay the entire expose event). To enable use of the surface cache whilst replaying use -c: ./cairo-perf-trace -c firefox-talos-gfx perf/Makefile.am | 3 +- perf/cairo-perf-trace.c | 141 ++++++++++++++++++++++++++- test/any2ppm.c | 4 +- util/cairo-script/cairo-script-interpreter.h | 3 +- util/cairo-script/cairo-script-operators.c | 13 ++- util/cairo-sphinx/sphinx.c | 3 +- 6 files changed, 157 insertions(+), 10 deletions(-) commit c50c8b90c0033686b465e6766fba4772c1d698c6 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jan 18 16:58:40 2010 +0000 Move _cairo_error() to a standalone header A pending commit will want to include some utility code from cairo and so we need to extricate the error handling from the PLT symbol hiding. src/Makefile.sources | 1 + src/cairo-analysis-surface.c | 1 + src/cairo-array.c | 1 + src/cairo-base64-stream.c | 1 + src/cairo-base85-stream.c | 1 + src/cairo-bentley-ottmann-rectangular.c | 1 + src/cairo-bentley-ottmann-rectilinear.c | 1 + src/cairo-bentley-ottmann.c | 1 + src/cairo-beos-surface.cpp | 2 ++ src/cairo-cache.c | 1 + src/cairo-cff-subset.c | 1 + src/cairo-clip.c | 1 + src/cairo-deflate-stream.c | 1 + src/cairo-directfb-surface.c | 1 + src/cairo-eagle-context.c | 1 + src/cairo-error-private.h | 60 +++++++++++++++++++++++++++++++++ src/cairo-font-face-twin.c | 1 + src/cairo-font-face.c | 1 + src/cairo-font-options.c | 1 + src/cairo-freelist.c | 1 + src/cairo-ft-font.c | 1 + src/cairo-gl-glyphs.c | 2 ++ src/cairo-gl-shaders.c | 1 + src/cairo-gl-surface.c | 1 + src/cairo-glitz-surface.c | 2 ++ src/cairo-glx-context.c | 2 ++ src/cairo-gstate.c | 1 + src/cairo-hash.c | 1 + src/cairo-hull.c | 1 + src/cairo-image-surface.c | 1 + src/cairo-lzw.c | 1 + src/cairo-matrix.c | 1 + src/cairo-misc.c | 1 + src/cairo-os2-surface.c | 1 + src/cairo-output-stream.c | 1 + src/cairo-paginated-surface.c | 1 + src/cairo-path-fill.c | 1 + src/cairo-path-fixed.c | 1 + src/cairo-path-stroke.c | 1 + src/cairo-path.c | 1 + src/cairo-pattern.c | 1 + src/cairo-pdf-operators.c | 1 + src/cairo-pdf-surface.c | 1 + src/cairo-pen.c | 1 + src/cairo-png.c | 2 ++ src/cairo-polygon.c | 1 + src/cairo-ps-surface.c | 1 + src/cairo-qt-surface.cpp | 6 ++-- src/cairo-quartz-font.c | 2 ++ src/cairo-quartz-image-surface.c | 2 ++ src/cairo-quartz-surface.c | 2 ++ src/cairo-recording-surface.c | 3 +- src/cairo-region.c | 1 + src/cairo-rtree.c | 1 + src/cairo-scaled-font-subsets.c | 1 + src/cairo-scaled-font.c | 1 + src/cairo-script-surface.c | 1 + src/cairo-stroke-style.c | 1 + src/cairo-surface-fallback.c | 1 + src/cairo-surface-wrapper.c | 1 + src/cairo-surface.c | 1 + src/cairo-svg-surface.c | 1 + src/cairo-tee-surface.c | 1 + src/cairo-tor-scan-converter.c | 1 + src/cairo-toy-font-face.c | 1 + src/cairo-traps.c | 1 + src/cairo-truetype-subset.c | 1 + src/cairo-type1-fallback.c | 1 + src/cairo-type1-subset.c | 1 + src/cairo-type3-glyph-surface.c | 1 + src/cairo-unicode.c | 1 + src/cairo-user-font.c | 1 + src/cairo-vg-surface.c | 3 +- src/cairo-win32-font.c | 1 + src/cairo-win32-printing-surface.c | 1 + src/cairo-win32-surface.c | 1 + src/cairo-xcb-surface.c | 3 +- src/cairo-xlib-display.c | 1 + src/cairo-xlib-screen.c | 1 + src/cairo-xlib-surface.c | 1 + src/cairo-xlib-visual.c | 2 ++ src/cairo-xml-surface.c | 1 + src/cairo.c | 1 + src/cairoint.h | 12 ------- src/drm/cairo-drm-bo.c | 1 + src/drm/cairo-drm-intel-surface.c | 1 + src/drm/cairo-drm-intel.c | 1 + src/drm/cairo-drm-radeon-surface.c | 1 + src/drm/cairo-drm-radeon.c | 1 + src/drm/cairo-drm-surface.c | 1 + src/drm/cairo-drm.c | 1 + src/test-fallback-surface.c | 1 + src/test-fallback16-surface.c | 1 + src/test-null-surface.c | 2 ++ src/test-paginated-surface.c | 1 + src/test-wrapping-surface.c | 1 + 96 files changed, 170 insertions(+), 17 deletions(-) commit 558f9501700ba569f04f345f7c64d16b1870e37c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jan 18 22:47:34 2010 +0000 surface-wrapper: Avoid copying patterns and clips unless transformed. An older variant of the fixes for moving the device transformation out of the surface layer, but languished in a side branch. The only benefit of this patch is that it avoids the copy where possible. src/cairo-surface-wrapper.c | 134 +++++++++++++++++++++++++------------------- 1 file changed, 75 insertions(+), 59 deletions(-) commit 8f694817227366f4a4c6bbbf46e6cdddb4a065df Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jan 19 17:12:47 2010 +0000 gitignore: refresh Add forgotten local targets to .gitignore .gitignore | 1 + perf/.gitignore | 1 + util/.gitignore | 2 ++ util/cairo-script/.gitignore | 1 + 4 files changed, 5 insertions(+) commit 5914e995723682e0b1f685d9ff5ea5abe05ae024 Author: Eric Anholt <eric@anholt.net> Date: Fri Jan 22 09:21:36 2010 -0800 [gl] Cache a temporary glyph compositing mask for reuse later. Cuts firefox-talos-gfx time from 56 seconds to 43 seconds. src/cairo-gl-glyphs.c | 29 ++++++++++++++++++++++------- src/cairo-gl-private.h | 4 ++++ src/cairo-gl-surface.c | 4 +++- 3 files changed, 29 insertions(+), 8 deletions(-) commit 1bf0f64ee7783742fa777a496567e4da48e300dd Author: Eric Anholt <eric@anholt.net> Date: Thu Jan 21 14:08:11 2010 -0800 [gl] Composite component-alpha glyphs directly instead of through a mask. Cuts gnome-terminal-vim time from 58 seconds to 20. src/cairo-gl-glyphs.c | 190 +++++++++++++++++++++++++++++++------------------ src/cairo-gl-private.h | 4 ++ src/cairo-gl-surface.c | 2 +- 3 files changed, 127 insertions(+), 69 deletions(-) commit 12d521df8acc483b2daa844d4f05dc2fe2765ba6 Author: Eric Anholt <eric@anholt.net> Date: Thu Jan 21 13:19:17 2010 -0800 [gl] Use GL_RGBA textures even for CAIRO_CONTENT_COLOR. When the texture is GL_RGB, GL_CLAMP_TO_BORDER (EXTEND_NONE) fills the border color alpha channel with 1, when the whole reason we were using the border color was to get a color and alpha of 0. We're forced to use GL_RGBA textures and do extra work to fill in the alpha channel of them to 1 to get cairo's desired behavior. This fixes a failure in rotate-image-surface-paint and 4 other testcases. No performance difference in firefox-talos-gfx. src/cairo-gl-surface.c | 131 ++++++++++++++++++++++++++++--------------------- 1 file changed, 75 insertions(+), 56 deletions(-) commit e316cb9db57094359c28f6b740e73674e8dcc134 Author: Eric Anholt <eric@anholt.net> Date: Wed Jan 20 13:33:56 2010 -0800 [gl] Add support for component-alpha glyph rendering. This is done using a mask and the two-pass trick I did for EXA. Fixes text-antialias-subpixel. src/cairo-gl-glyphs.c | 60 ++++++-- src/cairo-gl-private.h | 3 +- src/cairo-gl-surface.c | 384 +++++++++++++++++++++++++++++++++++++++++++++++-- 3 files changed, 424 insertions(+), 23 deletions(-) commit ceeb7e7e703213a83ca43d80962519074a19364d Author: Eric Anholt <eric@anholt.net> Date: Tue Jan 19 13:43:16 2010 -0800 [gl] Use the UNSUPPORTED() macro in more places for fallback debugging. src/cairo-gl-surface.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) commit 5d9e2fd84b4f66e5b121d95805ce34e2f482d1b0 Author: Andrea Canciani <ranma42@gmail.com> Date: Fri Jan 15 16:28:12 2010 +0100 [quartz] Leftbehind change from a39075c88f9b2e12e19348ea11c9b3f739818265 This commit completes the change in a39075c88f9b2e12e19348ea11c9b3f739818265. It removes unused variables and corrects the extents rect computation. src/cairo-quartz-surface.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) commit 6950b233e41a0c926465728c37f9bc5512f2844b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 15 14:25:16 2010 +0000 qt: Make flush() robust. Hanno Meyer-Thurow reported in http://bugs.freedesktop.org/show_bug.cgi?id=26063 [PATCH] various fixes to cairo-qt a few issues related to _cairo_qt_surface_flush() where firefox was crashing after being built to use cairo-qt. This is an amalgam of those patches that hopefully address the issue. src/cairo-qt-surface.cpp | 33 ++++++++++++++++++++++++++------- 1 file changed, 26 insertions(+), 7 deletions(-) commit 4e315d84723b27746a6a012ab7de429b2f2ef90b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 15 12:21:52 2010 +0000 pattern: Premultiply color-stop when converting to solid Fixes: test/linear-uniform https://bugzilla.mozilla.org/show_bug.cgi?id=539165 Bug 539165 - gradients with a single stop do not have their color multiplied by the alpha. As reported by Jeff Muizelaar, we regressed in 2d790daa as the color-stops are not premultiplied and so could not be treated as an ordinary cairo_color_t. Instead we have to create a intermediate cairo_color_t from the original values in order for the premultiplication to be performed. src/cairo-gstate.c | 26 +++++++++++++++++++------- src/cairo-pattern.c | 32 +++++++++++++++++++++----------- 2 files changed, 40 insertions(+), 18 deletions(-) commit f40560a9bb5b12d2584ed297bf4114baf2aa9cc1 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 15 12:17:30 2010 +0000 test: Add linear-uniform Jeff reported a regression found by Mozilla whereby a uniform gradient was not being premultiplied on conversion to a solid pattern. https://bugzilla.mozilla.org/show_bug.cgi?id=539165 [Bug 539165] gradients with a single stop do not have their color multiplied by the alpha. test/Makefile.am | 1 + test/Makefile.sources | 1 + test/linear-uniform.c | 63 ++++++++++++++++++++++++++++++++++++++++++++ test/linear-uniform.ref.png | Bin 0 -> 128 bytes 4 files changed, 65 insertions(+) commit e22c02dbaeda6941083fc89b517508dc0c4386a0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 15 12:02:58 2010 +0000 test: Add degenerate-arcs A simple test to ensure that using degenerate-arcs such as when drawing a rounded rectangle with radii=0 construct a proper path. test/Makefile.am | 1 + test/Makefile.sources | 1 + test/degenerate-arcs.c | 57 +++++++++++++++++++++++++++++++++++++++++++ test/degenerate-arcs.ref.png | Bin 0 -> 120 bytes 4 files changed, 59 insertions(+) commit a39075c88f9b2e12e19348ea11c9b3f739818265 Author: Andrea Canciani <ranma42@gmail.com> Date: Wed Jan 6 18:53:54 2010 +0100 [quartz] Fix generic masking Masking with unbounded operators was broken because an incorrect transformation was applied to inverted mask used for the fixup. src/cairo-quartz-surface.c | 59 +++++++++++++++++----------------------------- 1 file changed, 21 insertions(+), 38 deletions(-) commit 27701ed8447d4c21b7bf9709a2fc21690b3cdc96 Author: Andrea Canciani <ranma42@gmail.com> Date: Tue Jan 5 19:42:23 2010 +0100 [quartz] Approximate dense dashing patterns Quartz doesn't try to approximate dash pattern even if they are well beyond suface sampling frequency. Approximating dash patterns when their density is too high avoids looping indefinitely in these degenerate cases. src/cairo-quartz-surface.c | 32 +++++++++++++++++++++----------- 1 file changed, 21 insertions(+), 11 deletions(-) commit ad7ab00c10fc1adcf2ef4574fdd00b25dfcdafdd Author: Andrea Canciani <ranma42@gmail.com> Date: Sat Jan 2 13:23:53 2010 +0100 [quartz] Improve A8 and A1 support Use A8 images as alpha-only masks intread of greyscale images. Add support to A1 images (they were ignored before). src/cairo-quartz-surface.c | 59 +++++++++++++++++++++++++++------------------- 1 file changed, 35 insertions(+), 24 deletions(-) commit f3b3a5c43f247f3e9019fa6c1fd2abb3b00dd9ea Author: Adrian Johnson <ajohnson@redneon.com> Date: Fri Jan 15 07:47:30 2010 +1030 Fix bug in _cairo_ft_index_to_ucs4 Previously it would miss the first character causing the space glyph to return 0x00A0 instead of 0x0020. src/cairo-ft-font.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit e26e2c8cc41b4e60462eacbd87894dc82beca883 Author: Eric Anholt <eric@anholt.net> Date: Thu Jan 14 10:05:02 2010 -0800 [gl] Fix the color bits of CAIRO_CONTENT_ALPHA surface patterns to be 0. Fixes surface-pattern-operator and alpha-similar. src/cairo-gl-private.h | 1 - src/cairo-gl-surface.c | 22 +++++++++++----------- 2 files changed, 11 insertions(+), 12 deletions(-) commit 520945603cdbacb90b3067f24a37b6dbd2ea6808 Author: Eric Anholt <eric@anholt.net> Date: Thu Jan 14 10:00:37 2010 -0800 [gl] Remove the special-casing of acquiring a pattern surface. _cairo_pattern_acquire_surface() should do fine at it, and matches what cairo-drm-i915 does. src/cairo-gl-surface.c | 127 ------------------------------------------------- 1 file changed, 127 deletions(-) commit dbf9faf8232e7bf279377b5f3fc08ed210ca4e51 Author: Eric Anholt <eric@anholt.net> Date: Wed Jan 13 10:33:32 2010 -0800 [gl] Use GLSL for fill_rectangles when available. Ultimately, we want all of our paths to use shaders when they are exposed -- it brings us closer to GL 3.0 compatibility and it should reduce the work that GL drivers have to do per operation to compute the required hardware state. src/Makefile.sources | 2 +- src/cairo-gl-private.h | 6 +++ src/cairo-gl-shaders.c | 117 +++++++++++++++++++++++++++++++++++++++++++++ src/cairo-gl-surface.c | 125 +++++++++++++++++++++++++++++++++++++++++++++++-- 4 files changed, 244 insertions(+), 6 deletions(-) commit 57341cf5ac4d16222da27856e3a1092a473597e8 Author: Eric Anholt <eric@anholt.net> Date: Wed Jan 13 14:18:06 2010 -0800 [gl] Use ADD instead of SOURCE when making a mask for overlapping glyphs. The whole point of making the mask was to avoid overwriting one glyph with the data from another. Whoops. Fixes ft-show-glyphs-positioning, text-rotate, overlapping-glyphs, clip-text. src/cairo-gl-glyphs.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 6e0a63102e7e2d3fa3d5c993f2b94a2d6e40bdd8 Author: Eric Anholt <eric@anholt.net> Date: Wed Jan 13 14:06:23 2010 -0800 [gl] Clear the current_target cache when binding a framebuffer in creation. Failing to do so meant that we might not re-bind the framebuffer to the correct one if we didn't follow up surface creation with drawing to it. Fixes many testcases. Thanks to T. Zachary Laine for tracking down that the current_target was the source of these issues. src/cairo-gl-surface.c | 1 + 1 file changed, 1 insertion(+) commit ad0f4bee19b214b96fb07a45d1f059b362afee0f Author: T. Zachary Laine <whatwasthataddress@gmail.com> Date: Thu Jan 7 17:24:31 2010 -0600 [gl] Clear the is_clear flag on the temporary glyph mask we create. Normally is_clear is cleared by the caller of one of the 5 basic entrypoints, but in this temporary surface usage nobody else can do it for us. Fixes ft-show-glyphs-table, user-font-rescale, and select-font-face. src/cairo-gl-glyphs.c | 1 + 1 file changed, 1 insertion(+) commit 384e50d858e9eec670f764c758fb3aa09b56caed Author: T. Zachary Laine <whatwasthataddress@gmail.com> Date: Wed Jan 13 15:06:24 2010 -0600 [gl] Removed repeated lines of source The two removed lines are redundant with code a few lines above them. src/cairo-gl-surface.c | 2 -- 1 file changed, 2 deletions(-) commit 515ccb63a67c88ddb2bc5adfd018cae55bc93185 Author: Eric Anholt <eric@anholt.net> Date: Mon Jan 11 16:04:21 2010 -0800 [gl] Fix the glyph cache full flush to really try again. Previously, the initial error handling would dump through to software fallback instead of retrying in the following code. src/cairo-gl-glyphs.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) commit 4aae4bfa8f28a470ab18575670563ad28374e445 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jan 12 17:10:04 2010 +0000 test: Fix typo in path-stroke-twice reference image Sigh. I committed the reference image with the name path-stroke-twice-ref.png. ...{path-stroke-twice-ref.png => path-stroke-twice.ref.png} | Bin 1 file changed, 0 insertions(+), 0 deletions(-) commit 74ea4c908067f5579b51d3dbaea387da8f343671 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jan 12 17:00:25 2010 +0000 path: Do not remove anti-parallel line segments in case we are stroking Bug 26010 - cairo_line_to optimizes away path segments http://bugs.freedesktop.org/show_bug.cgi?id=26010 As exercised by path-stroke-twice, we incorrectly optimise away a line segment if the path doubled back upon itself. This is very reminiscent of the optimisation bug for replacing curve-to with line-to. src/cairo-path-fixed.c | 5 ++++- src/cairo-slope-private.h | 7 +++++++ 2 files changed, 11 insertions(+), 1 deletion(-) commit a5dd5a6069c0d7435260a954d8d98483f5fbcc1c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jan 12 16:32:09 2010 +0000 test: Add path-stroke-twice Exercises a bug found by alois@astro.ch, whereby we inadvertently remove a line segment when the path doubles back upon itself. Bug 26010 - cairo_line_to optimizes away path segments http://bugs.freedesktop.org/show_bug.cgi?id=26010 test/Makefile.am | 2 ++ test/Makefile.sources | 1 + test/path-stroke-twice-ref.png | Bin 0 -> 206 bytes test/path-stroke-twice.c | 54 ++++++++++++++++++++++++++++++++++++++ test/path-stroke-twice.ps.ref.png | Bin 0 -> 197 bytes 5 files changed, 57 insertions(+) commit 4204605813d93e7e2f0d8761e90a6dd6b8089dcb Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 8 11:58:15 2010 +0000 type1: Destroy the output stream. mpsuzuki spotted that the output stream for the subset was not being destroyed on the error path - on closer inspection, it wasn't destroyed along the normal path either and may never have been correctly initialised to NULL on an earlier error. Reported-by: mpsuzuki@hiroshima-u.ac.jp src/cairo-type1-subset.c | 28 +++++++++++++++------------- 1 file changed, 15 insertions(+), 13 deletions(-) commit 41a24d822ac094f0841ba0c92ca86448ad062def Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jan 7 11:58:01 2010 +0000 trace: Correctly encode octal values. How embarrassing. util/cairo-trace/trace.c | 21 ++++++++++++++------- 1 file changed, 14 insertions(+), 7 deletions(-) commit 3ae9d04c6ddd311ffab91170fb9342e37c5530a8 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Fri Jan 1 20:13:33 2010 +0200 [stroker] Fix off-by-one memory allocation in _tessellate_fan(). The number of points in a triangle fan was miscomputed because it was computing the number of line segments rather than points in the fan. Now we include the final point of the fan correctly in the count. This fixes https://bugs.webkit.org/show_bug.cgi?id=33071 as reported by Benjamin Otte. A derived test case was not added to the cairo test suite since the bug is difficult to trigger in a reliable way which causes visible results (as opposed to silent heap corruption.) The easiest way of triggering the bug is to stroke a line using a large line width and round caps or joins. src/cairo-path-stroke.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit a0ea0b63fdd38a73d6696da5cd4800d9a7289240 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Dec 30 16:21:50 2009 +0000 arc: Just emit a line-to if radius is zero. In order to be consistent with the semantics of cairo_arc() using a radius of 0 should simply degenerate into a line-to. Reported-by: Benjamin Berg src/cairo.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) commit 21560b3ec6d015004867207c5bdd90061b8a5d80 Author: Andrea Canciani <ranma42@gmail.com> Date: Fri Dec 11 10:29:12 2009 +0100 [trace] Fix compilation when FreeType is disabled cairo-trace needs some headers which are missing if freetype is not enabled. Adding them explicitly fixes compilation. util/cairo-trace/trace.c | 3 +++ 1 file changed, 3 insertions(+) commit 393d2d9c9199c368916fb5412abaac163f57ff86 Author: Tim Janik <timj@gtk.org> Date: Mon Dec 28 13:34:43 2009 +0100 Fixed documentation typo. src/cairo-image-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit e2d75203c515f9740a8e0e9779391de0a1a54b07 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Sun Nov 29 15:56:26 2009 +0200 [dirty] Add more missing surface dirtying notifications. Now that the image surface actually cares about cairo_surface_mark_dirty() we're hitting cases where we've forgotten to mark surfaces dirty. boilerplate/cairo-boilerplate.c | 1 + src/cairo-xlib-surface.c | 1 + util/cairo-script/cairo-script-operators.c | 2 ++ 3 files changed, 4 insertions(+) commit b76565d2f4a2510aabada4cef968b9b597bbef09 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Sun Nov 29 15:40:30 2009 +0200 [test] Don't use uninitialised data in surface-source.c The draw_pattern() function assumed the incoming surface was clear, but it's not. Explicitly clear the surface first. test/surface-source.c | 4 ++++ 1 file changed, 4 insertions(+) commit e09b754fdd43206e1668812be4ff11d25188148c Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Sun Nov 29 01:11:29 2009 +0200 [gstate] Change dash offset normalisation to preserve offsets in range. We have a test case get-and-set which wants to see whatever it puts into a cairo_t come back out again, but at the same time cairo-gstate wants to range reduce the dash offset it's given to a sane range. This patch changes the range reduction algorithm to always normalize to a non-negative dash offset and not touch dash offsets which are already in range. src/cairo-gstate.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) commit b394240941dffa263776a62cf42dc9b7e477f7a2 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Sun Nov 29 02:48:23 2009 +0200 [test] Add missing mark_dirty() calls to test cases. A recent optimisation has added a flag to the image surface which is used to track whether the surface is clear or not. This makes it imperative that clients call cairo_surface_mark_dirty() if they use cairo to allocate their pixel buffers and then proceed to initialize them without telling cairo about it. test/a1-mask.c | 1 + test/large-source.c | 2 ++ test/user-font-mask.c | 1 + 3 files changed, 4 insertions(+) commit 9ddf14cfd5c3bf0cf507cfc82b192f5c46f66f9a Author: Adrian Johnson <ajohnson@redneon.com> Date: Sun Nov 29 10:40:18 2009 +1030 PDF: Don't specify colorspace in JPEG2000 image dictionary PDF uses the colorspace specified in the JPEG2000 image. src/cairo-pdf-surface.c | 1 - 1 file changed, 1 deletion(-) commit d58560bda8361bd25f1b2207ecfab4b331601045 Author: Benjamin Otte <otte@gnome.org> Date: Fri Nov 27 21:04:55 2009 +0100 Satisfy make check src/cairo-stroke-style.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 16fffc5c05857896c03decdcebff7f1b5e582390 Author: Benjamin Otte <otte@gnome.org> Date: Fri Nov 27 18:13:45 2009 +0100 Apply device transform in surface wrapper src/cairo-surface-wrapper.c | 56 +++++++++++++++++++++++++++++++++++++-------- 1 file changed, 46 insertions(+), 10 deletions(-) commit 02d7e123d640d62db46ef03d7faecbc7598444e1 Author: Adrian Johnson <ajohnson@redneon.com> Date: Fri Nov 27 23:11:56 2009 +1030 PDF: Remove extraneous \n from end of jpeg/jp2 data PDF requires a '\n' between the end of stream data and the "endstream" that is not included in the stream length. Ensure this is always added in _close_stream where it is not included in the stream length. Previously the jpeg/jp2 embedding functions were adding the '\n'. This resulted in the '\n' becoming part of the stream data. src/cairo-pdf-surface.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) commit 4fc7bdaed651a5c19eb89dddd88808468e0e7eb8 Author: Adrian Johnson <ajohnson@redneon.com> Date: Thu Nov 26 21:14:20 2009 +1030 PS: Add some missing pdf-operators flushes src/cairo-ps-surface.c | 15 +++++++++++++++ 1 file changed, 15 insertions(+) commit 3f0c9afa9760a90065c67927c63770f9bb8abe99 Author: Benjamin Otte <otte@gnome.org> Date: Wed Nov 25 15:57:46 2009 +0100 [clip] Don't ignore outermost clip when it clips everything Fixes testcases clip-stroke-no-op and clip-fill-no-op added in 0d8a04ef47426d87c7932f308a831e36b4ef7683 src/cairo-clip.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) commit 0d8a04ef47426d87c7932f308a831e36b4ef7683 Author: Benjamin Otte <otte@gnome.org> Date: Wed Nov 25 10:48:39 2009 +0000 test: Add cases to exercise clipping after no-ops Whilst investigating: Bug 31788: SVG <line> with y1 == y2 and both with fractional part don't render https://bugs.webkit.org/show_bug.cgi?id=31788 I found that it can be reduced to a no-op clip following a no-op stroke or fill. These conditions are exercised here and are shown to be a bug introduced in the 1.9.x development series. test/Makefile.am | 2 ++ test/Makefile.sources | 2 ++ test/clip-fill-no-op.c | 66 +++++++++++++++++++++++++++++++++++++++++ test/clip-fill-no-op.ref.png | Bin 0 -> 152 bytes test/clip-stroke-no-op.c | 66 +++++++++++++++++++++++++++++++++++++++++ test/clip-stroke-no-op.ref.png | Bin 0 -> 152 bytes 6 files changed, 136 insertions(+) commit 26e9f149063b9e1fdb54fc54fccbefdf04a68190 Author: Andrea Canciani <ranma42@gmail.com> Date: Tue Nov 10 23:09:56 2009 +0100 Improve stroking of densely dashed styles Add some auxiliary functions to cairo-stroke-style to compute properties of the dashed patterns (period, "on" coverage) and to generate approximations when the dash pattern is sub-tolerance. These functions are used in cairo-path-stroke to simplify dash patterns stroked by cairo. Fixes dash-infinite-loop See http://bugs.freedesktop.org/show_bug.cgi?id=24702 src/cairo-path-stroke.c | 24 +++++++--- src/cairo-stroke-style.c | 119 +++++++++++++++++++++++++++++++++++++++++++++++ src/cairoint.h | 20 ++++++++ 3 files changed, 157 insertions(+), 6 deletions(-) commit 9c24288c820069e80b0feb5e99ece4c89e92c0c6 Author: Andrea Canciani <ranma42@gmail.com> Date: Wed Nov 11 00:09:08 2009 +0100 Revert "[test] Reorder dash-infinite-loop to not hit a runaway allocation." The infinite loop problem in _cairo_stroker_dash_start is solved by commit ee02f3484899527380df94c00f40da87f41660ea, so hitting that problem is not possible anymore and dash-infinite stroke always hit the memory intensive loops. This reverts commit 29432d3d32bc84ec4a2e1815a84e4ac2089138fe. test/dash-infinite-loop.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit cc2d2630669b084ec43e415d2806d94af00cf56c Author: Andrea Canciani <ranma42@gmail.com> Date: Tue Nov 10 22:58:59 2009 +0100 Improve stroke offset computation The stroke offset was forced to be positive because stroking code wants to be able to decrement it until it finds the first dash to be drawn. This can lead to long (almost infinite) loops if the offset is positive but huge compared to the total length of the dash pattern. Computing the offset this way guarantees that it's always the smallest non-negative equivalent offset. src/cairo-gstate.c | 10 ++++------ 1 file changed, 4 insertions(+), 6 deletions(-) commit e436a57c22f2c2a87404cac27e2d5e3c404f8bf9 Author: Andrea Canciani <ranma42@gmail.com> Date: Tue Nov 10 13:29:28 2009 +0100 Fix odd length dashing with negative offset When computing an equivalent offset, a wrong total dash length was used when the dash elements were odd and more than 1. Doubling the total dash length whenever they are odd is needed to correctly compute the new offset. Fixes dash-offset. src/cairo-gstate.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit b1a76394655793fd698a1281b00a3d049f9e70f5 Author: Andrea Canciani <ranma42@gmail.com> Date: Tue Nov 10 13:09:50 2009 +0100 Add dash-offset test Stroking a dash pattern of odd length with a negative offset is broken (except when the pattern is composed by a single dash). test/Makefile.am | 1 + test/Makefile.sources | 1 + test/dash-offset.c | 95 +++++++++++++++++++++++++++++++++++++++++++++++ test/dash-offset.ref.png | Bin 0 -> 794 bytes 4 files changed, 97 insertions(+) commit 7d8788b4ba5ae335a7bb77bd09e7ebd20376f632 Author: Robert O'Callahan <robert at ocallahan.org> Date: Mon Nov 9 18:58:59 2009 +0100 [quartz] Cleanup fallback source Remove the transformation code as it's useless and confusing. The clipbox was transformed in a wrong way if the CTM was not the base one, otherwise the transformation did nothing. As long as _cairo_quartz_setup_fallback_source is only called with the base CTM, no transformation needs to be performed on the clipbox. Patch by Robert O'Callahan, see Mozilla bug #507939. src/cairo-quartz-surface.c | 9 --------- 1 file changed, 9 deletions(-) commit 248090e52b02f2b2dc424c6073489df32632b9d1 Author: Andrea Canciani <ranma42@gmail.com> Date: Mon Nov 9 21:38:00 2009 +0100 [quartz] Fix compositing operators Fallback on SATURATE (since quartz doesn't support it) and on blending operators (to be implemented using CGContextSetBlendMode, after its availability has been assessed). Workaround the different behaviour of quartz operators on alpha masks. Fixes text in clip-operator and clip-text, blend modes in extended-blend, extended-blend-alpha, saturate in operator, operator-alpha and surface-pattern-operator. Changes the output of operator-alpha-alpha, but to really fix it pure-alpha surfaces needs to be treated as alpha masks and not as greyscale images. src/cairo-quartz-surface.c | 182 ++++++++++++++++++++++++++++++++++----------- 1 file changed, 140 insertions(+), 42 deletions(-) commit d44414191a685a23fd48bb32826366dfd113e1e5 Author: Andrea Canciani <ranma42@gmail.com> Date: Mon Nov 9 14:21:40 2009 +0100 [quartz] Fix source-unbounded compositing of surface sources Surface sources were not respecting EXTEND_NONE when composited because the part outside of the surface rectangle was not changed. To correctly extend the source, it's sufficient to composite a transparent rectangle with a hole in the region covered by the surface. Fixes clip-unbounded and operator-source. Changes the output of surface-pattern-operator. src/cairo-quartz-surface.c | 62 +++++++++++++++++++++------------------------- 1 file changed, 28 insertions(+), 34 deletions(-) commit 7ef611a67a2a68d0a50e2191804a8ef137c78b33 Author: Andrea Canciani <ranma42@gmail.com> Date: Mon Nov 9 20:17:35 2009 +0100 [quartz] Fix stroking with unbounded operator Stroking was not fixed up correctly when the stroke trasform was not the identity and when the stroked path was dashed. Instead of building a mask image and clearing through it (forgetting to use the CTM and the dash setting), the code now simply fills the outside of the stroke with tranparent black. Additionally this commit moves the CTM composition so that the invocation of _cairo_quartz_setup_source is always done with the base CTM. This commit changes the result of clip-stroke-unbounded (the clip mask is now correctly aligned to the stroked path). src/cairo-quartz-surface.c | 36 +++++++++++++----------------------- 1 file changed, 13 insertions(+), 23 deletions(-) commit 577ac4fc6d810cae7ff6d460fdea1886cd97258c Author: Andrea Canciani <ranma42@gmail.com> Date: Mon Nov 9 14:24:13 2009 +0100 [quartz] Fix implied colour of alpha-only surface sources If a fill colour was set, an alpha-only surface source would have used that colour instead of black. src/cairo-quartz-surface.c | 2 ++ 1 file changed, 2 insertions(+) commit 8b25e4a7fd839ac399d14460d2596037134c9801 Author: Andrea Canciani <ranma42@gmail.com> Date: Mon Oct 19 18:38:57 2009 +0200 [quartz] Improve unbounded operators fixup Don't allocate a colorspace and use a 16-bytes aligned stride and an alpha-only image for the rendering of the fixup mask. src/cairo-quartz-surface.c | 16 +++++++--------- 1 file changed, 7 insertions(+), 9 deletions(-) commit ed8bdcd59863964a3eb9c49ca1a9bfd5f544350c Author: Andrea Canciani <ranma42@gmail.com> Date: Mon Oct 19 17:33:10 2009 +0200 [quartz] Improve A8 image handling A8 images were cosidered grayscale, whereas they should behave like transparency mask. They are now alpha-only images and the stride is 16-bytes aligned, as suggested. Fixes alpha-similar and mask-transformed-similar. Makes the patterns behave (as expected) as masks in surface-pattern-operator. Affects the output of operator-alpha-alpha, but the result is not correct because of wrong compositing. src/cairo-quartz-surface.c | 21 +++++++++------------ 1 file changed, 9 insertions(+), 12 deletions(-) commit b32695c248b08289ce431d5f2fdab5205219c203 Author: Robert O'Callahan <robert at ocallahan.org> Date: Tue Oct 20 16:07:52 2009 +0200 [quartz] Fix gradient pattern painting CTM was changed and not restored when painting gradients. Fixes the alignment of surface patterns in surface-pattern-operator. Patch by Robert O'Callahan, see Mozilla bug #507939. src/cairo-quartz-surface.c | 2 ++ 1 file changed, 2 insertions(+) commit 1ce096657b2d38956e3229042f87aea42b0d0d8c Author: Andrea Canciani <ranma42@gmail.com> Date: Tue Nov 3 22:51:29 2009 +0100 Fix compilation on MacOS X Enable "ld_preload" (DYLD_INSERT_LIBRARIES + DYLD_FORCE_FLAT_NAMESPACE) and build cairo-trace as a library instead of a module to make it work also on MacOS X. configure.ac | 17 ++++++++++++++++- util/cairo-trace/Makefile.am | 22 +++++++++++----------- util/cairo-trace/cairo-trace.in | 6 +++++- util/cairo-trace/trace.c | 2 +- 4 files changed, 33 insertions(+), 14 deletions(-) commit c181f9e9aeac375eed07aaab58d177e3476d5117 Author: Andrea Canciani <ranma42@gmail.com> Date: Sat Nov 7 19:29:30 2009 +0100 Add new operator-alpha-alpha test Compositing of alpha-only sources is not throughly tested and as a consequence the misbehaviour of some backends is not shown by the test suite. [ickle: Added XFAILs for pdf, ps, svg.] src/cairo-xcb-surface.c | 4 - test/Makefile.am | 4 + test/Makefile.sources | 1 + test/operator-alpha-alpha.c | 166 ++++++++++++++++++++++++++++++++ test/operator-alpha-alpha.pdf.xfail.png | Bin 0 -> 3749 bytes test/operator-alpha-alpha.ps.xfail.png | Bin 0 -> 1338 bytes test/operator-alpha-alpha.ref.png | Bin 0 -> 3441 bytes test/operator-alpha-alpha.svg.xfail.png | Bin 0 -> 838 bytes 8 files changed, 171 insertions(+), 4 deletions(-) commit cee3a6169bdf7f64c126c487b443dcb7a9f0586a Author: Adrian Johnson <ajohnson@redneon.com> Date: Sun Nov 8 21:52:06 2009 +1030 Type1-subset: Append "cleartomark" operator for binary fonts that don't include it Type 1 fonts embedded in PDF may omit the fixed-content portion of the font that contains the "cleartomark" operator. src/cairo-type1-subset.c | 22 +++++++++++++++------- 1 file changed, 15 insertions(+), 7 deletions(-) commit d962d6bae2681ea1f3c9533a83c8902d76e81c6a Author: Adrian Johnson <ajohnson@redneon.com> Date: Sun Nov 8 21:46:16 2009 +1030 PDF: Include Type 1 fixed-content portion in the embedded font Some fonts may contain additional PostScript code after the cleartomark. src/cairo-pdf-surface.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) commit ac59c7580894fc5fd424f7f6f8c1532d15048566 Author: Adrian Johnson <ajohnson@redneon.com> Date: Sun Nov 8 21:38:58 2009 +1030 Type1-subset: Check for binary eexec data Type 1 fonts embedded in PDF have the the encrypted portion in binary but the existing check for binary only works for Type 1 fonts in PFB format. Add an additional check based on the first 4 characters of eexec data. The Type 1 specification gurantees that at least one of the first 4 bytes of ciphertext is not an ASCII Hex character. src/cairo-type1-subset.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) commit a4a59be600cbe17f43afa20422f9d05f3a5d5b52 Author: Adrian Johnson <ajohnson@redneon.com> Date: Sat Nov 7 14:36:49 2009 +1030 Type 1 subset: Don't add a new line when in binary mode At the end of the private dictionary a new line is appended so that the trailing zeros will not exceed 80 columns. In hex mode (used for PS embedding) extra white space is harmless as it is ignored when parsing the ascii hex encoded data. But in binary mode (used for PDF embedding) the new line must not be added as it corrupts the binary data. https://bugs.launchpad.net/ubuntu/+source/cups/+bug/419143 src/cairo-type1-subset.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit 9775f44436ffaf54931614d1867bcb7d147acd06 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Nov 5 10:13:34 2009 +0000 AUTHORS: Add Andrea Canciani AUTHORS | 1 + 1 file changed, 1 insertion(+) commit d51915f873da8d40740a908c86f02e972e358be9 Author: Andrea Canciani <ranma42@gmail.com> Date: Wed Nov 4 16:15:48 2009 +0100 xlib: Fix double free Both _cairo_xlib_call_close_display_hooks and _cairo_xlib_screen_destroy invoke _cairo_xlib_screen_close_display, which frees all the cached GCs, but (if HAS_ATOMIC_OPS is undefined) doesn't mark them as freed, thus freeing them agin upon the second invocation. src/cairo-xlib-screen.c | 1 + 1 file changed, 1 insertion(+) commit 7c1424a315b677ca119d11b7ced1d4c969dde568 Author: Andrea Canciani <ranma42@gmail.com> Date: Sun Oct 18 08:59:33 2009 +0200 quartz: Fix compilation The function _cairo_image_surface_create_for_content does not exist. The correct name of the function is instead _cairo_image_surface_create_with_content. src/cairo-quartz-image-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 8a10c25c3f4036d17e24852e7dc970fa06fcb4af Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Nov 5 10:02:54 2009 +0000 test: Tweak Benjamin's xcomposite-projection Include Benjamin's advice on how to make the bug more visible inline with the code. test/xcomposite-projection.c | 69 ++++++++++++++++++++++++++++++-------------- 1 file changed, 48 insertions(+), 21 deletions(-) commit a9f37953c71e16f65122097462e51b035783a965 Author: Benjamin Otte <otte@gnome.org> Date: Wed Nov 4 13:59:58 2009 +0100 [xlib] Use correct number when checking if points are out of range src/cairo-xlib-surface.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) commit aa1f245ea313a2f1f0dda3994f5b48073b0c385d Author: Benjamin Otte <otte@gnome.org> Date: Wed Nov 4 13:56:15 2009 +0100 [test] Add "target=raster" to test The test isn't useful on vector backends and fails there due to antialiasing issues. test/xcomposite-projection.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 52afe9c77f2f54372fefeca50321a27e8dda0f63 Author: Benjamin Otte <otte@gnome.org> Date: Wed Nov 4 11:15:57 2009 +0100 [xlib] Fix new Composite test When the reference point was tirggering _line_exceeds_16_16() and got adjusted, the code failed to compute the srcX and srcY arguments for the call to XRenderCompositeTrapezoids() correctly and caused the resulting source image to be misaligned. src/cairo-fixed-private.h | 9 +++++++++ src/cairo-xlib-surface.c | 22 +++++++++++----------- 2 files changed, 20 insertions(+), 11 deletions(-) commit 84bbf179c375622d2c7b4e21b1b8ce189b5a18f2 Author: Benjamin Otte <otte@gnome.org> Date: Wed Nov 4 11:06:57 2009 +0100 [test] Add a test exposing bugs in XRenderComposite This test fills a slightly rotated surface slightly above the 0 line. This hits some corner cases in the XRenderComposite path. I discovered these issues while playing with video rendering onto the canvas in HTML5 (both Webkit and Mozilla have this problem). I used CAIRO_ANTIALIAS_NONE and a single-color source in the test to get rid of aliasing issues in the output images. This makes some issues slightly less visible, but still fails for all of them. If you want to get a clearer view, disable it and use romedalen.png instead - it has the same size as the red surface. (At least) 3 bugs are at work here: - if _line_exceeds_16_16() triggers for the reference point, the source surface will be misaligned. - the intel driver seems to have an off-by-one bug on my i945 when positioning the source surface, causing black seams at the top (not visible in the test unless using romedalen.png) and on the left of the image. - My Xvfb fails completely in picture up/download in the xlib-fallback path. test/Makefile.sources | 1 + test/xcomposite-projection.c | 81 +++++++++++++++++++++++++++++++++++++ test/xcomposite-projection.ref.png | Bin 0 -> 1112 bytes 3 files changed, 82 insertions(+) commit 9b42b6156d9916e64e19f2ccb0b43f4d1df72e79 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Nov 2 12:50:45 2009 +0000 build: Add -pthread to real_pthread_CFLAGS build/configure.ac.system | 12 +++++++++++ configure.ac | 51 ++++++++++++++++++++--------------------------- 2 files changed, 34 insertions(+), 29 deletions(-) commit 3eecf2f60c85e0195360ff3c3a54d58ceaa4a6f1 Author: Benjamin Otte <otte@gnome.org> Date: Mon Nov 2 13:20:25 2009 +0100 sphinx: Add required pthread libs Try again, this time for real hopefully. configure.ac | 2 +- util/cairo-sphinx/Makefile.am | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) commit 837eb577b4707e447fafb0dfca378ecf4ea3fda5 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Nov 2 11:56:29 2009 +0000 sphinx: Add required pthread libs Switching to pthread-stubs broke the build for sphinx on unbroken platforms [those that do not unnecessarily link against -lpthreads]. configure.ac | 7 +++++++ util/cairo-sphinx/Makefile.am | 5 +++-- 2 files changed, 10 insertions(+), 2 deletions(-) commit 942a15ba9e85fa3297ad6e65270928ad3c1c0a42 Author: Benjamin Otte <otte@gnome.org> Date: Mon Nov 2 12:51:14 2009 +0100 [test] Update upscale test to expose an Xlib failure By filling the background with grey, we can see that Xlib fills areas with black that it shouldn't. Xlib therefore now fails this test. test/surface-pattern-scale-up.c | 4 +++- test/surface-pattern-scale-up.pdf.ref.png | Bin 2793 -> 3482 bytes test/surface-pattern-scale-up.ps2.ref.png | Bin 943 -> 913 bytes test/surface-pattern-scale-up.ps3.ref.png | Bin 943 -> 913 bytes test/surface-pattern-scale-up.ref.png | Bin 4061 -> 4017 bytes 5 files changed, 3 insertions(+), 1 deletion(-) commit 19c411a5b8b84d31516d9c85642ad55ef5d29aba Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 30 07:58:45 2009 +0000 path: Skip any secondary degenerate line-to segments. Only the very first line-to following a move-to can have any significance if degenerate whilst stroking, so skip all others. In other words, 0 0 m 0 0 l stroke produces a capped degenerate path (i.e a dot), 0 0 m 0 0 l 0 0 l stroke produces the same degenerate stroke, and 0 0 m 0 0 l 1 0 l stroke produce a horizontal line. src/cairo-path-fixed.c | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-) commit 23bcf91748c4bb04c16e503b913da3bfc237463f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 30 07:49:56 2009 +0000 path: Misclassification of rectilinear after degenerate line-to Malte Nuhn reported hitting an assertion: cairo-path-stroke.c:1816: _cairo_rectilinear_stroker_line_to: Assertion `a->x == b->x || a->y == b->y' failed. http://bugs.freedesktop.org/show_bug.cgi?id=24797 when stroking an apparently simple path: 0 8.626485 m 0 8.626485 l 5.208333 2.5 l 10.416667 2.5 l 15.625 2.5 l 20.833333 2.5 l 26.041667 2.5 l 31.25 2.5 l 36.458333 2.5 l 41.666667 2.5 l 46.875 2.5 l 52.083333 2.5 l 57.291667 2.5 l 62.5 2.5 l 67.708333 2.5 l 72.916667 2.5 l 78.125 2.5 l 83.333333 2.5 l 88.541667 2.5 l 93.75 2.5 l 98.958333 2.5 l 104.166667 2.5 l 109.375 2.5 l 114.583333 2.5 l 119.791667 2.5 l 125 2.5 l stroke which upon reduction becomes: 0.000000 8.625000 m 5.207031 2.500000 l 125.000000 2.500000 l stroke The bug is that after spotting a repeated line-to we move the previous end-point without reclassifying the path, hence we miss the non-rectilinear step. src/cairo-path-fixed.c | 101 +++++++++++++++++++++++++------------------------ 1 file changed, 52 insertions(+), 49 deletions(-) commit 600dd833983bd85bf8aea487393ce156abb622bd Author: Adrian Johnson <ajohnson@redneon.com> Date: Mon Sep 28 19:14:46 2009 +0930 Provide a builtin implementation of isspace() and isdigit() The glibc versions are slow in utf8 locales. src/cairo-output-stream.c | 7 +++---- src/cairo-pdf-operators.c | 6 ++---- src/cairo-type1-subset.c | 16 +++++++--------- src/cairoint.h | 16 ++++++++++++++++ 4 files changed, 28 insertions(+), 17 deletions(-) commit f571356d346565cae34f08d12c584c4bd40921d6 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Oct 27 20:27:09 2009 +0000 xlib: Set has_component_alpha = FALSE for gradient Pictures Whilst creating the xlib surface wrapper for gradient Pictures, the corresponding attributes require ComponentAlpha disabled. src/cairo-xlib-surface.c | 1 + 1 file changed, 1 insertion(+) commit 1a80cd2aa00b232a9a2192a85d6264e44ffcbd69 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Oct 26 11:05:38 2009 +0000 scaled-font: Fix documentation examples with incorrect derefs Bug 24730 -- New: misleading use cases in cairo_scaled_font_text_to_glyphs's docs http://bugs.freedesktop.org/show_bug.cgi?id=24730 An apparent copy'n'paste error from the code which has to deal with indirect pointers crept into the documentation for the same functions. src/cairo-scaled-font.c | 20 ++++++++++---------- 1 file changed, 10 insertions(+), 10 deletions(-) commit f648fcd2f91ea5c9ec318c2c0573ee82ac8c0ac9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Oct 26 11:02:16 2009 +0000 gitignore: Add forgotten cairo-perf-print perf/.gitignore | 1 + 1 file changed, 1 insertion(+) commit 6911ba2f76d4ccdf228c029e2719461a72848824 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Oct 26 11:01:01 2009 +0000 perf: Fix comparisons The sort order was unstable resulting in incorrect ordering dependent upon libc version. perf/cairo-perf-diff-files.c | 22 +++++++++++++++------- 1 file changed, 15 insertions(+), 7 deletions(-) commit f955a4cba7784f97ef1dd978158e98176e03baf0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Oct 25 11:04:17 2009 +0000 wideint: Protect 64_32_div for non-HAVE_UINT16_T systems Peter Clifton spotted that we failed to properly wrap _cairo_int64_32_div() for systems without a native 64 bit type. src/cairo-wideint-private.h | 5 +++++ 1 file changed, 5 insertions(+) commit c701d7813b6d116c9db53f63f791928a407499c7 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 23 14:42:48 2009 +0100 [pattern] Compute zero extents for empty patterns If the pattern is for example a repeating 0x0 image, then treat it as having zero extents. This should workaround the bug presented here: https://bugs.freedesktop.org/show_bug.cgi?id=24693 Attached PDF crashes evince with a Floating point exception src/cairo-pattern.c | 27 ++++++++++++++++++++++++--- 1 file changed, 24 insertions(+), 3 deletions(-) commit 43a775f60da57206b194039cd5d8b6a735284c45 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Thu Oct 22 02:13:36 2009 +0300 [meta] Rename cairo_meta_surface_t to cairo_recording_surface_t. The new name is more descriptive than the rather opaque meta surface. Discussed with vigour on the mailing list and #cairo: http://lists.cairographics.org/archives/cairo/2009-July/017571.html NEWS | 7 + boilerplate/Makefile.win32.features | 14 +- boilerplate/cairo-boilerplate-pdf.c | 6 +- boilerplate/cairo-boilerplate-ps.c | 8 +- boilerplate/cairo-boilerplate-svg.c | 8 +- boilerplate/cairo-boilerplate-test-surfaces.c | 2 +- boilerplate/cairo-boilerplate-win32-printing.c | 2 +- boilerplate/cairo-boilerplate.c | 46 +-- boilerplate/cairo-boilerplate.h | 2 +- build/Makefile.win32.features-h | 2 +- configure.ac | 2 +- doc/public/tmpl/cairo-surface.sgml | 2 +- perf/cairo-perf-micro.c | 4 +- perf/cairo-perf-trace.c | 4 +- src/Makefile.sources | 4 +- src/Makefile.win32.features | 14 +- src/cairo-analysis-surface.c | 48 +-- src/cairo-paginated-private.h | 6 +- src/cairo-paginated-surface-private.h | 4 +- src/cairo-paginated-surface.c | 92 ++--- src/cairo-pattern.c | 2 +- src/cairo-pdf-surface.c | 56 +-- src/cairo-ps-surface.c | 42 +- ...private.h => cairo-recording-surface-private.h} | 54 +-- ...ro-meta-surface.c => cairo-recording-surface.c} | 460 ++++++++++----------- src/cairo-scaled-font.c | 28 +- src/cairo-script-surface.c | 42 +- src/cairo-script.h | 4 +- src/cairo-surface.c | 58 +-- src/cairo-svg-surface.c | 46 +-- src/cairo-tee-surface.c | 6 +- src/cairo-type3-glyph-surface.c | 14 +- src/cairo-types-private.h | 2 +- src/cairo-user-font.c | 42 +- src/cairo-vg-surface.c | 10 +- src/cairo-win32-printing-surface.c | 60 +-- src/cairo-xml-surface.c | 32 +- src/cairo-xml.h | 4 +- src/cairo.c | 2 +- src/cairo.h | 20 +- src/cairoint.h | 12 +- test/Makefile.am | 32 +- test/Makefile.sources | 2 +- test/cairo-test-trace.c | 48 +-- test/cairo-test.c | 6 +- ...rface-pattern.c => recording-surface-pattern.c} | 8 +- ...=> recording-surface-pattern.gl.argb32.ref.png} | Bin ...> recording-surface-pattern.pdf.argb32.ref.png} | Bin ...=> recording-surface-pattern.pdf.rgb24.ref.png} | Bin ...> recording-surface-pattern.ps2.argb32.ref.png} | Bin ...=> recording-surface-pattern.ps2.rgb24.ref.png} | Bin ...> recording-surface-pattern.ps3.argb32.ref.png} | Bin ...=> recording-surface-pattern.ps3.rgb24.ref.png} | Bin ...ng => recording-surface-pattern.quartz.ref.png} | Bin ...recording-surface-pattern.quartz.rgb24.ref.png} | Bin ...n.ref.png => recording-surface-pattern.ref.png} | Bin ...png => recording-surface-pattern.rgb24.ref.png} | Bin ...> recording-surface-pattern.svg.argb32.ref.png} | Bin ...=> recording-surface-pattern.svg.rgb24.ref.png} | Bin ...ng => scale-offset-similar.recording.xfail.png} | Bin util/cairo-fdr/fdr.c | 8 +- util/cairo-sphinx/sphinx.c | 38 +- util/cairo-trace/trace.c | 10 +- util/trace-to-xml.c | 4 +- 64 files changed, 718 insertions(+), 711 deletions(-) commit df357f26ff72571acb840715efa4930054d4fdbe Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 16 16:48:54 2009 +0100 Support component-alpha. Within our code base we carried a few hacks to utilize the component alpha capabilities of pixman, whilst not supporting the concept for our own masks. Thus we were setting it upon the pixman_image_t that we passed around through code that was blissfully unaware and indeed the component-alpha property was forgotten (e.g. upgrading glyph masks). The real issue is that without explicit support that a pattern carries subpixel masking information, that information is lost when using that pattern with composite. Again we can look at the example of compositing a sub-pixel glyph mask onto a remote xlib surface for further failure. src/cairo-ft-font.c | 5 --- src/cairo-gl-glyphs.c | 23 ++++++++++--- src/cairo-gl-surface.c | 18 ++++++---- src/cairo-gstate.c | 17 ++++++---- src/cairo-image-surface.c | 55 +++++++++++++++++++++++-------- src/cairo-pattern.c | 22 ++++++++++--- src/cairo-pdf-surface.c | 10 ++++-- src/cairo-scaled-font.c | 14 ++++---- src/cairo-svg-surface.c | 10 ++++-- src/cairo-types-private.h | 71 ++++++++++++++++++++++++++++++++++++++++ src/cairo-win32-font.c | 10 +++--- src/cairo-xlib-surface-private.h | 1 + src/cairo-xlib-surface.c | 43 +++++++++++++++++++----- src/cairoint.h | 69 +------------------------------------- 14 files changed, 234 insertions(+), 134 deletions(-) commit f0cd20e6cec445eb627c2708c2230c8bad1b64ce Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Oct 21 10:36:27 2009 +0100 [gl] Do not clear internal surfaces. Create a scratch surface that will be initialised as required for internal use. External surfaces, i.e. those returned to the user, are cleared as normal. src/cairo-gl-surface.c | 46 +++++++++++++++++++++++++++++++--------------- 1 file changed, 31 insertions(+), 15 deletions(-) commit 7c9ebd4a852ad709e57ed48b9610db00de79de7e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Oct 21 09:04:30 2009 +0100 Make the surface->is_clear logic common A nasty surprise whilst profiling is that performing redundant clear operations is extremely painful. We can mitigate this somewhat by tracking the cleared state of surfaces and skipping repeated attempts to clear a surface. src/cairo-gl-surface.c | 2 ++ src/cairo-image-surface.c | 47 +++----------------------------- src/cairo-surface-private.h | 6 +++-- src/cairo-surface.c | 66 ++++++++++++++++++++++++++++++++++++++++----- src/cairoint.h | 1 - 5 files changed, 68 insertions(+), 54 deletions(-) commit 723055722f1d9a133fe9e78bb19165f2c7be720e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Oct 21 10:40:51 2009 +0100 [doc] Add comments on when surfaces are cleared on construction. Or perhaps, more importantly, when the contents are left unmodified. src/cairo.h | 9 +++++++++ 1 file changed, 9 insertions(+) commit cef8194178b01eaa2d10c1ba8291a9a4c5a6d302 Author: Andrea Canciani <ranma42@gmail.com> Date: Tue Oct 20 14:37:28 2009 +0200 [test] Add surface-pattern-operator Test the results of compositing ALPHA and COLOR_ALPHA surfaces with different operators. test/Makefile.am | 4 + test/Makefile.sources | 1 + test/surface-pattern-operator.argb32.ref.png | Bin 0 -> 5217 bytes test/surface-pattern-operator.c | 117 ++++++++++++++++++++++ test/surface-pattern-operator.rgb24.ref.png | Bin 0 -> 1942 bytes test/surface-pattern-operator.xlib.argb32.ref.png | Bin 0 -> 5171 bytes test/surface-pattern-operator.xlib.rgb24.ref.png | Bin 0 -> 1913 bytes 7 files changed, 122 insertions(+) commit 710303ef15b9a1bf73b2b07b7aa51ec816d0aa07 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Oct 19 10:42:51 2009 +0100 [test] Update rotate-image-surface-paint Update this test case so the background is not the default border colour so that we spot when the backend pads out the transformed image surface. test/Makefile.am | 5 +- test/rotate-image-surface-paint.c | 78 ++++++++++++++------------ test/rotate-image-surface-paint.pdf.xfail.png | Bin 217 -> 358 bytes test/rotate-image-surface-paint.ps.ref.png | Bin 0 -> 236 bytes test/rotate-image-surface-paint.ps2.ref.png | Bin 212 -> 0 bytes test/rotate-image-surface-paint.ps3.ref.png | Bin 212 -> 0 bytes test/rotate-image-surface-paint.ref.png | Bin 191 -> 220 bytes test/rotate-image-surface-paint.svg.xfail.png | Bin 0 -> 387 bytes 8 files changed, 45 insertions(+), 38 deletions(-) commit ea39f0302a936711043aa6834ba0ab487736572b Author: Luo Jinghua <sunmoon1997@gmail.com> Date: Mon Oct 19 22:21:00 2009 +0800 atomic: correct libatomic_ops version of _cairo_atomic_int_cmpxchg AO_compare_and_swap_full() doesn't have the same semantic as _cairo_atomic_int_cmpxchg(). The former returns a TRUE on success FALSE otherwise while the later returns the atomic value. src/cairo-atomic-private.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit a2254e56474fcdf5f25232812cdbef30a2b5ea60 Author: Benjamin Otte <otte@gnome.org> Date: Sun Oct 18 22:12:13 2009 +0200 [gstate] Apply target device offsets to patterns immediately Previously target device offsets were applied in cairo-surface.c which could cause bugs when paths were taken as fallbacks, as for example pointed out by ade55037ffa596b690c6a1051394589f76eb1f48 and quick-fixed by 79190d89853958ee8252440d35a662fabf122afd. The quick-fix is now unnecessary and was removed. src/cairo-gstate.c | 4 +++ src/cairo-surface-fallback.c | 41 +++++------------------------- src/cairo-surface.c | 59 -------------------------------------------- 3 files changed, 10 insertions(+), 94 deletions(-) commit 4b720d56db72d314b3b7cedb05a53321b1d0adbe Author: Benjamin Otte <otte@gnome.org> Date: Sun Oct 18 21:32:10 2009 +0200 [clip] Remove non-existing functions from header src/cairo-clip-private.h | 9 --------- 1 file changed, 9 deletions(-) commit cacf9ccedffa0c804a56c76c8ab17bf81831b621 Author: Benjamin Otte <otte@gnome.org> Date: Sun Oct 18 21:30:22 2009 +0200 [path] Fix compiler warnings src/cairo-path-fill.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) commit a70705b2ede8c0dc401f28405641e70f8a6d8d92 Author: Benjamin Otte <otte@gnome.org> Date: Sun Oct 18 21:30:06 2009 +0200 [xlib] Fix compiler warnings src/cairo-xlib-screen.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) commit 96117c1212be2bf39a300a212dec85466f49f31e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Oct 17 10:36:10 2009 +0100 [test] Add radial-gradient-source ranma42 found a bug in compositing with a radial gradient using the SOURCE operator with EXTEND_NONE, as exercised here. test/Makefile.am | 2 + test/Makefile.sources | 1 + test/radial-gradient-source.argb32.ref.png | Bin 0 -> 272428 bytes test/radial-gradient-source.c | 115 +++++++++++++++++++++++++++++ test/radial-gradient-source.rgb24.ref.png | Bin 0 -> 193364 bytes 5 files changed, 118 insertions(+) commit 54df07a3da2bfbbdbe2dde92ca298ac0d7b8d255 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 16 10:11:41 2009 +0100 [surface] Don't AND in the desired content. Gah, that was a horrible mistake. It was a flawed hack to create Pixmaps of the correct depth when cloning patterns for blitting to the xlib backend. However, it had the nasty side-effect of discarding alpha when targeting Window surfaces. The correct solution is to simply correct the Pixmap of the desired depth and render a matching pattern onto the surface - i.e. a reversal the current acquire -> clone. See the forthcoming revised xcb backend on how I should have done it originally. src/cairo-directfb-surface.c | 2 -- src/cairo-gl-surface.c | 4 +--- src/cairo-glitz-surface.c | 3 --- src/cairo-image-surface.c | 4 ---- src/cairo-pattern.c | 15 +++------------ src/cairo-quartz-surface.c | 1 - src/cairo-surface-fallback-private.h | 1 - src/cairo-surface-fallback.c | 3 +-- src/cairo-surface.c | 14 ++++---------- src/cairo-xlib-surface.c | 18 +----------------- src/cairoint.h | 4 ---- src/test-fallback-surface.c | 1 - src/test-fallback16-surface.c | 2 -- 13 files changed, 10 insertions(+), 62 deletions(-) commit 6a19a82efd3afe8fb8bd30e5362b247de8efe159 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 16 08:43:21 2009 +0100 [image] Clone the format for a similar surface with identical content Honour the incoming surface format when we are asked to create a similar surface with identical content. The goal of cairo_surface_create_similar() is to create an intermediate with similar characteristics to the original that can be used in place of the original and be quick to copy to the original. Matching the format for the same content, ensures that the blits between the two need only be a memcpy. src/cairo-image-surface.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) commit ed94d7caad6b851f6f80fa4d67441ad0387a8cd4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 16 08:38:24 2009 +0100 [fallback] Create intermediate surface with same content. The goal is to create a similar surface with an identical format to maximise performance in the subsequent blit, e.g. the xlib backend could make the similar surface with an identical depth and so use the core protocol, or the image surface could indeed make an identical copy so that pixman only has to do a fast memcpy. As there is no direct method to specify such a clone, we ask the backend for a similar surface of identical content, and trust that the semantics are clear enough for the intent to obvious. src/cairo-surface-fallback.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) commit c739940137c4ffc6cc727ea30afc0e92f7ab1e80 Author: Carl Worth <cworth@cworth.org> Date: Thu Oct 15 16:24:44 2009 -0700 Increment version to 1.9.5 Just after the 1.9.4 snapshot. cairo-version.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit ad6334b9870c63e641b531d6e910c084b587d0f9 Author: Carl Worth <cworth@cworth.org> Date: Thu Oct 15 15:33:20 2009 -0700 Add mising images to REFERENCE_IMAGES list. Again, the kind of thing that only gets fixed at release time. But at least we have a nice test for this. test/Makefile.am | 8 ++++++++ 1 file changed, 8 insertions(+) commit cdad5e10768978b11088f82858a1e6e0037dc0a7 Author: Carl Worth <cworth@cworth.org> Date: Thu Oct 15 14:58:39 2009 -0700 Mark in-fill-empty-trapezoid as an XFAIL test. Otherwise, it was disrupting my attempts to release a snapshot, (since it doesn't respect CAIRO_TEST_TARGET=" " and not run like most of the tests do). test/Makefile.am | 5 +++++ test/in-fill-empty-trapezoid.c | 4 ++-- 2 files changed, 7 insertions(+), 2 deletions(-) commit b0f309abb33fb7e068199c941835af0217d8d1fc Author: Carl Worth <cworth@cworth.org> Date: Thu Oct 15 14:40:22 2009 -0700 Update version to 1.9.4 In preparation for the 1.9.4 snapshot. cairo-version.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 3b113275f8b081685227fc073c03f64bb1a229d3 Author: Carl Worth <cworth@cworth.org> Date: Thu Oct 15 14:17:30 2009 -0700 Remove stale references to non-existant reference images. This is just one of those things that we don't get right until someone tries running "make distcheck". test/Makefile.am | 7 ------- 1 file changed, 7 deletions(-) commit 37078b262a268179f8b7a3dfdc263e1a403daa7d Author: Carl Worth <cworth@cworth.org> Date: Thu Oct 15 14:11:04 2009 -0700 NEWS: Update with correct data for 1.9.4 snapshot NEWS | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) commit d5609ca04ecc5f5e3d5cef8ca0e67fa98cd653de Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Thu Oct 15 17:47:33 2009 +0300 [build] Check compiler flags using AC_TRY_LINK instead of AC_TRY_COMPILE. Turns out we were passing suncc warning suppression flags to gcc by mistake since -e<entry point> is actually a valid option for gcc. This caused the -erroff=E_ENUM_TYPE_MISMATCH_ARG and other -erroff options to be passed to the linker. In the normal case of a GNU ld linker this doesn't matter since it ignores bogus entry points, but the GNU gold linker replaces a bogus entry point with NULL. This patch makes the CAIRO_CC_TRY_FLAG() check stricter by testing that the flag doesn't interfere with linking executables. build/aclocal.cairo.m4 | 19 ++++++++----------- 1 file changed, 8 insertions(+), 11 deletions(-) commit 79190d89853958ee8252440d35a662fabf122afd Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Oct 15 14:05:37 2009 +0100 [surface] Avoid double application of device offset when calling fill() _cairo_surface_fallback_paint() attempts to avoid a clipped operation if we can convert the paint into a fill of the clipmask. However by calling _cairo_surface_fill() we incur a double application of device offset to the source, triggering various failures. Company spotted this and managed to extract an excellent minimal test case, test/clip-device-offset. This commit fixes that failure. src/cairo-surface-fallback.c | 41 +++++++++++++++++++++++++++++++++++------ 1 file changed, 35 insertions(+), 6 deletions(-) commit ade55037ffa596b690c6a1051394589f76eb1f48 Author: Benjamin Otte <otte@gnome.org> Date: Thu Oct 15 13:19:11 2009 +0100 [test] Add clip-device-offset Exercises a bug demonstrated by WebKit with clipping + device offsets. test/Makefile.am | 2 + test/Makefile.sources | 1 + test/clip-device-offset.argb32.ref.png | Bin 0 -> 175 bytes test/clip-device-offset.c | 79 +++++++++++++++++++++++++++++++++ test/clip-device-offset.rgb24.ref.png | Bin 0 -> 168 bytes 5 files changed, 82 insertions(+) commit dac73d260a9b9a848bb97436ad84081c51629511 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Oct 15 12:18:47 2009 +0100 [build] Link against pthread-stubs Avoid pulling in the real pthread library if the application is single threaded and not using pthreads, by linking against pthread-stubs instead. build/configure.ac.features | 1 + configure.ac | 8 +++++++- src/cairo-mutex-impl-private.h | 2 +- test/Makefile.am | 4 ++++ test/cairo-test-trace.c | 10 +++++----- test/cairo-test.c | 6 +++--- util/cairo-trace/Makefile.am | 2 +- 7 files changed, 22 insertions(+), 11 deletions(-) commit ef9286751d8346ecb803bfb5916581ee4dfc84db Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Oct 13 16:10:39 2009 +0100 [atomic] Fallback to libatomic-ops-dev Use libatomic-ops-dev in preference to mutex-based atomics, if we do not have the builtin atomic intrinsics available. build/aclocal.cairo.m4 | 10 ++++++++++ src/cairo-atomic-private.h | 41 ++++++++++++++++++++++++++++++++++++----- src/cairo-surface.c | 4 ++-- src/cairo-xlib-private.h | 2 +- src/cairo-xlib-screen.c | 16 +++++++++------- src/cairo.c | 10 +++++----- 6 files changed, 63 insertions(+), 20 deletions(-) commit d85eda97dd9116f51e0255b29652f4b52ba4f8e9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Oct 13 14:14:55 2009 +0100 [pattern] Compute extents for gradients As noted in the comments we could also compute the pattern extents for gradients with CAIRO_EXTEND_NONE under certain circumstances, i.e. radial gradients and device axis aligned linear gradients. src/cairo-pattern.c | 184 +++++++++++++++++++++++++++++++++++----------------- 1 file changed, 126 insertions(+), 58 deletions(-) commit 3a83bc8617dae3cc2bfea862f252338463871648 Author: Adrian Johnson <ajohnson@redneon.com> Date: Fri Oct 2 23:28:10 2009 +0930 PDF: Fix regression caused by 1ae5a419 Bug 24240 acroread prints error message when printing src/cairo-pdf-surface.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) commit 91a4f59f135dd42a28e14be55d1f08712c0c3097 Author: Adrian Johnson <ajohnson@redneon.com> Date: Fri Oct 2 23:11:54 2009 +0930 PDF: Fix regression caused by 1ae5a419 Bug 24240 first char missing when printing a PDF. src/cairo-scaled-font-subsets.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 39284159234d5318c90fa3a72178cd3d14fe0fbb Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Sep 30 20:11:12 2009 +0100 [boilerplate/xcb] Fix pixmap depth All the error checking, finally pointed out that I was creating a pixmap with the wrong depth! Oops. boilerplate/cairo-boilerplate-xcb.c | 19 ++++++++++++++++--- 1 file changed, 16 insertions(+), 3 deletions(-) commit e9bc2180d73acc7133d715ed4380ed20fb4364c0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Sep 30 19:50:44 2009 +0100 [xcb] Deferred error checking. XCB avoids the dreaded abort on XError mechanism by forcing the client to perform deferred error checking. So do so. This allows us to combine the fire-and-forget rendering model with accurate error checking, without killing the client or mixing our errors with theirs. XCB for the win! boilerplate/cairo-boilerplate-xcb.c | 17 + src/cairo-freelist-private.h | 5 + src/cairo-freelist.c | 33 ++ src/cairo-xcb-surface.c | 712 +++++++++++++++++++++++------------- 4 files changed, 521 insertions(+), 246 deletions(-) commit 8e4e0aa7ee5b4e0963409cda825705a09aae61e1 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Sep 30 17:50:09 2009 +0100 [boilerplate/xcb] Check for connection errors during test boilerplate/cairo-boilerplate-xcb.c | 51 +++++++++++++++++----- test/cairo-test.c | 86 ++++++++++++++++++------------------- 2 files changed, 83 insertions(+), 54 deletions(-) commit 395555b116a497c99b9e8365a202377c344c0c45 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Sep 30 00:37:36 2009 +0100 [xlib] Suppress warning that should never have been. src/cairo-xlib-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 42ea356c0393288eea16eb5c591fa51009f85cb9 Author: Adrian Johnson <ajohnson@redneon.com> Date: Wed Sep 30 21:21:42 2009 +0930 Fix truetype subset crash Bug 24213 src/cairo-truetype-subset.c | 1 + 1 file changed, 1 insertion(+) commit 6809e1216119e16c5be236418dc560d2e9bf854d Author: Benjamin Otte <otte@gnome.org> Date: Wed Sep 30 08:54:19 2009 +0200 [API] Make _cairo_meta_surface_replay() private again Replaying a meta surface can be achieved by using it as a source for a cairo_paint() so exporting a separate API is unnecesary and confusing. So after consulting Chris and Carl, we decided to remove the function again. NEWS | 1 - src/cairo-meta-surface-private.h | 5 ++++- src/cairo-meta-surface.c | 39 +++++++++++++++++---------------------- src/cairo-paginated-surface.c | 4 ++-- src/cairo-script-surface.c | 6 +++--- src/cairo-surface.c | 4 ++-- src/cairo-svg-surface.c | 2 +- src/cairo-type3-glyph-surface.c | 4 ++-- src/cairo-user-font.c | 2 +- src/cairo-vg-surface.c | 2 +- src/cairo-xml-surface.c | 2 +- src/cairo.h | 4 ---- test/cairo-test-trace.c | 8 ++++++-- util/cairo-trace/trace.c | 22 ---------------------- 14 files changed, 40 insertions(+), 65 deletions(-) commit f8c49b4187a32b60408b23b2c2abce1778a8dc95 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Sep 29 03:02:40 2009 +0100 [xlib] XRequest bread crumbing Add bread crumbs in the form of NoOperations that can be parsed by xtrace and very useful when debugging protocol/server errors. src/cairo-xlib-surface.c | 57 ++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 57 insertions(+) commit 1c4f61ec5068a5935eb6d5dd1d6f2c0275b70529 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Sep 29 03:01:37 2009 +0100 [tessellator] Invalid conversion of list head to edge. When scanning for collinear right edges, we need to check that we do not go beyond the end of the array. src/cairo-bentley-ottmann-rectangular.c | 2 ++ 1 file changed, 2 insertions(+) commit f23ae97e307f00a79cbf2e01f9ca20da29ea87c3 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Sep 24 16:30:57 2009 +0100 [perf] Support parsing reports from stdin perf/cairo-perf-compare-backends.c | 22 ++++++++++++---------- perf/cairo-perf-report.c | 29 +++++++++++++++++++---------- 2 files changed, 31 insertions(+), 20 deletions(-) commit 50163a9fafa77dd1135ca5b6b03c7ec5619687c1 Author: Benjamin Otte <otte@gnome.org> Date: Tue Sep 29 00:23:12 2009 +0200 [xlib] Set proper bpp when computing format from masks Fixing omission in caa9eed4646e78a62ffb49d9c4f48c0b351c7a7f src/cairo-xlib-surface.c | 1 + 1 file changed, 1 insertion(+) commit a160c8c4009e069db53bec79abd8926231de9a7f Author: Benjamin Otte <otte@gnome.org> Date: Mon Sep 28 23:52:14 2009 +0200 [xlib] Handle 24bpp formats during dithering src/cairo-xlib-surface.c | 10 ++++++++++ 1 file changed, 10 insertions(+) commit 9e89de813516c48baed67a4d18325971cd42cd4a Author: Benjamin Otte <otte@gnome.org> Date: Mon Sep 28 23:51:14 2009 +0200 [image] Fix masks computation for BGRx format Patch originally by Mark Kettenis src/cairo-image-surface.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) commit 4b2c74e28531e403a9538f20a3ace7b114f9bbae Author: Adrian Johnson <ajohnson@redneon.com> Date: Mon Sep 28 22:57:10 2009 +0930 PS: Use standard page size names for some common sizes Some PS viewers like gv use the label in the %PageMedia DSC comment as the displayed page size. The page names and sizes were obtained from the list at: http://www.gnu.org/software/gv/manual/html_node/Paper-Keywords-and-paper-size-in-points.html src/cairo-ps-surface.c | 66 +++++++++++++++++++++++++++++++++++++++++++++++--- 1 file changed, 62 insertions(+), 4 deletions(-) commit db5e3db0a9b218bebad6921a95d5244047050348 Author: Adrian Johnson <ajohnson@redneon.com> Date: Fri Sep 25 23:07:56 2009 +0930 PS: Add DocumentMedia/PageMedia DSC comments Provide default DocumentMedia and PageMedia DSC comments if the user does not specify them using cairo_ps_surface_dsc_comment(). This is required so that PostScript viewers such as gv use the correct page size. src/cairo-ps-surface-private.h | 1 + src/cairo-ps-surface.c | 88 ++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 89 insertions(+) commit caa9eed4646e78a62ffb49d9c4f48c0b351c7a7f Author: Benjamin Otte <otte@gnome.org> Date: Thu Sep 24 23:42:05 2009 +0200 [xlib] Handle case in draw_image() where pixman format is not RGB When an image cannot be expressed as masks, its data cannot be used in an XImage. In that case, we use pixman to copy the image to the desired format. src/cairo-xlib-surface.c | 50 ++++++++++++++++++++++++++++++++++++++++++------ 1 file changed, 44 insertions(+), 6 deletions(-) commit 9f5fe6711566ee34afe210f271f1c5a061efa745 Author: Benjamin Otte <otte@gnome.org> Date: Thu Sep 24 22:57:39 2009 +0200 [image] Return cairo_bool_t from _pixman_format_to_masks Return FALSE when the pixman format does not support ARGB masks. In particular, return FALSE for YUV formats. src/cairo-image-surface.c | 21 ++++++++++----------- src/cairoint.h | 2 +- 2 files changed, 11 insertions(+), 12 deletions(-) commit 120247ac55d92d9c9b24fad45c765d16e692f355 Author: Adrian Johnson <ajohnson@redneon.com> Date: Thu Sep 24 23:02:38 2009 +0930 Add EPS test This test demonstrates a regression in the EPS output since 1.8.8. The ps-eps.ref.eps was created with 1.8.8. test/Makefile.am | 2 ++ test/Makefile.sources | 1 + test/ps-eps.c | 85 +++++++++++++++++++++++++++++++++++++++++++++++++++ test/ps-eps.ref.eps | 84 ++++++++++++++++++++++++++++++++++++++++++++++++++ 4 files changed, 172 insertions(+) commit c8a4f486712eba4ad0ccbf5ae3b18920137d5b1b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Sep 23 18:08:34 2009 +0100 [image] Do not trust user supplied data to be clear Gah, more important than the whitespace was the reversal of is_clear... src/cairo-image-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 03bce099eae68643979e5742b2703b2b2e835cca Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Sep 23 17:56:00 2009 +0100 [image] Tweak coding style Whitespace spam. src/cairo-image-surface.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) commit ab558fa843accc440686f7b8eda197277a6a4eb2 Author: Benjamin Otte <otte@gnome.org> Date: Wed Sep 23 17:40:17 2009 +0200 [image] Move setting of is_clear to _create_with_pixman_format() Simplifies code and ensures other users of that function don't run into the same bug. src/cairo-image-surface.c | 18 ++++++------------ 1 file changed, 6 insertions(+), 12 deletions(-) commit 07ee9cfbf07deba228e73d26b419a4e130de24e3 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Sep 23 15:33:23 2009 +0100 [image] Do assumption initial user data is cleared. Benjamin Otte pointed out the error of my ways that a clear on a cairo_image_surface_create_for_data() was not working. This is because I modified the image surface to skip clears when it knows the target data has been cleared. This flag must be reset when the user interacts with the surface, such as providing the initial surface data. src/cairo-image-surface.c | 26 ++++++++++++++++++++++---- 1 file changed, 22 insertions(+), 4 deletions(-) commit 9948b5180aa4312b3184d9ba46e9ad6f0d8e700c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Sep 22 18:20:03 2009 +0100 [test] Revamp surface-source Include a translucent region in the source that exercises the xlib bug that I'm trying to fix. Hmm. test/Makefile.am | 18 +++++++++----- test/glitz-surface-source.argb32.ref.png | Bin 0 -> 377 bytes test/glitz-surface-source.ref.png | Bin 268 -> 0 bytes test/glitz-surface-source.rgb24.ref.png | Bin 0 -> 301 bytes test/image-surface-source.argb32.ref.png | Bin 0 -> 377 bytes test/image-surface-source.ref.png | Bin 332 -> 0 bytes test/image-surface-source.rgb24.ref.png | Bin 0 -> 301 bytes test/pdf-surface-source.argb32.ref.png | Bin 0 -> 377 bytes test/pdf-surface-source.ref.png | Bin 268 -> 0 bytes test/pdf-surface-source.rgb24.ref.png | Bin 0 -> 301 bytes test/ps-surface-source.argb32.ref.png | Bin 0 -> 377 bytes test/ps-surface-source.ref.png | Bin 268 -> 0 bytes test/ps-surface-source.rgb24.ref.png | Bin 0 -> 301 bytes test/quartz-surface-source.argb32.ref.png | Bin 0 -> 377 bytes test/quartz-surface-source.ref.png | Bin 332 -> 0 bytes test/quartz-surface-source.rgb24.ref.png | Bin 0 -> 301 bytes test/surface-source.c | 37 ++++++++++++++++++++++++++-- test/svg-surface-source.argb32.ref.png | Bin 0 -> 377 bytes test/svg-surface-source.ref.png | Bin 268 -> 0 bytes test/svg-surface-source.rgb24.ref.png | Bin 0 -> 301 bytes test/test-fallback16-surface-source.ref.png | Bin 268 -> 0 bytes test/xlib-surface-source.argb32.ref.png | Bin 0 -> 377 bytes test/xlib-surface-source.ref.png | Bin 268 -> 0 bytes test/xlib-surface-source.rgb24.ref.png | Bin 0 -> 301 bytes 24 files changed, 47 insertions(+), 8 deletions(-) commit 941d3693fac831c4ce8c61cbac7c77b566b97611 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Sep 22 14:43:45 2009 +0100 [test] Exercise push-group-color. Hunting for a known bug in the xlib backend where it invalidly converts an argb32 source to rgb24. However, this does not appear to be that bug, but still a useful exercise nevertheless. test/Makefile.am | 2 + test/Makefile.sources | 1 + test/push-group-color.c | 141 +++++++++++++++++++++++++++++++++++++ test/push-group-color.ref.png | Bin 0 -> 2902 bytes test/push-group-color.xlib.ref.png | Bin 0 -> 2964 bytes 5 files changed, 144 insertions(+) commit 30f45ce5f7d639dd5a0b60f544b3535e3bc2105d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Sep 22 15:15:47 2009 +0100 [gl] Handle an absent visual. If we cannot find the correct visual for the fbconfig, return an error instead of crashing. The difference is subtle, granted. src/cairo-gl-surface.c | 8 ++++++++ src/cairo-glx-context.c | 7 ++++--- 2 files changed, 12 insertions(+), 3 deletions(-) commit e74cb7fa3eb9a9bcdeb298d969b70fd99ba50f3d Author: Adrian Johnson <ajohnson@redneon.com> Date: Tue Sep 22 22:44:34 2009 +0930 win32-printing: correct the use of the CAIRO_HAS_WIN32_FONT #ifdef it was wrapping code required for non windows fonts. src/cairo-win32-printing-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 3dd2cc8f6d84253f93affecd3022c9fe4a25d60f Author: Adrian Johnson <ajohnson@redneon.com> Date: Tue Sep 22 22:40:53 2009 +0930 win32-printing: fix meta surface pattern regression caused by bed2701e src/cairo-win32-printing-surface.c | 2 ++ 1 file changed, 2 insertions(+) commit 7658635c654b694b516e5dd44a43611d94e91abd Author: Adrian Johnson <ajohnson@redneon.com> Date: Tue Sep 22 22:36:20 2009 +0930 win32-printing: Allow GDI CTM with scale < 1 to be set To better support creating EMF files with the win32-printing surface, allow a GDI CTM with scale < 1 to be used. http://lists.cairographics.org/archives/cairo/2009-September/018110.html src/cairo-win32-printing-surface.c | 90 ++++++++++++++++++++++++++++++-------- src/cairo-win32-private.h | 2 + 2 files changed, 73 insertions(+), 19 deletions(-) commit 19c091670302fd23a676e680308df49a4ba06687 Author: Adrian Johnson <ajohnson@redneon.com> Date: Tue Sep 22 22:35:33 2009 +0930 Make win32 fonts work in EMF files created by win32-printing src/cairo-win32-printing-surface.c | 37 +++++++++++++++++++++++++++++++++++++ 1 file changed, 37 insertions(+) commit f01a54266d57763635f984982bec001a393d8aa2 Author: Adrian Johnson <ajohnson@redneon.com> Date: Tue Sep 22 22:34:56 2009 +0930 Fix compiler warnings in win32-surface src/cairo-win32-surface.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) commit 23663e2078e34659af43bd0e62bb1fb7a94cc71e Author: Adrian Johnson <ajohnson@redneon.com> Date: Tue Sep 22 22:34:21 2009 +0930 Fix compiler error on cygwin src/cairo-win32-surface.c | 1 + 1 file changed, 1 insertion(+) commit 7ecd413fab7c55918c6f3c4c272b5e8b5aa89aa7 Author: Adrian Johnson <ajohnson@redneon.com> Date: Tue Sep 22 22:33:14 2009 +0930 Add WINVER defines to cairo-mutex-impl-privaye.h The WINVER defines should always be defined before including windows.h src/cairo-mutex-impl-private.h | 9 +++++++++ 1 file changed, 9 insertions(+) commit 605ec22ab5103e16052591517f86fe558133fb8e Author: Adrian Johnson <ajohnson@redneon.com> Date: Tue Sep 22 22:17:27 2009 +0930 Use _WIN32 to check for win32 mutex Commit 535bcaa1 moved pthread after win32 to fix mingw build problems. However this breaks cygwin builds. On cygwin HAVE_WINDOWS_H is defined but _WIN32 is not. This was causing windows.h to be included which defines _WIN32. As a result the win32 code in cairo-misc.c was being compiled but the win32 declaration in cairint.h was not included. Fix this by using _WIN32 to enable the win32 mutex functions since _WIN32 is defined by mingw, visual C++, and winegcc, but not cygwin. On cygwin, posix functions are preferred as it is a unix emulation environment. src/cairo-mutex-impl-private.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit e00d0627494a4b15ed3b74a704695ca8b81a350e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Sep 21 13:50:00 2009 +0100 [xlib] Fix recent bug in unbounded trapezoids Gah! I had believed that the dst extents and the clip were correct to enable unbounded fixup for the unbounded trapezoids. I was wrong, so I need to requery the trapezoid extents. As this information is already known, I should update the interface to pass along all relevant information. src/cairo-surface.c | 23 ++++++++++------------- src/cairo-xlib-surface.c | 32 +++++++++++++++++++------------- 2 files changed, 29 insertions(+), 26 deletions(-) commit 378b1e73d9f27e9b54ea01b10e588b361848d0cd Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Sep 21 03:10:53 2009 +0100 [fallback] Special case single composite rectangle Avoid the overhead of region-from-traps extraction for the very frequent case of using a single (possibly clipped) rectangle with a pattern source. src/cairo-surface-fallback.c | 42 ++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 42 insertions(+) commit 35daf95db8aa4f0c254891f180aa4996dd464a60 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Sep 21 03:31:22 2009 +0100 [xlib] Trim a few redundant steps when uploading glyphs Cleanup the glyph coercion routines to avoid the temporary cairo_t (as we can just cairo_surface_paint() directly). src/cairo-xlib-surface.c | 23 ++++++----------------- 1 file changed, 6 insertions(+), 17 deletions(-) commit 9b7c5d95055c4ea5da46c53263050d6ec1331c08 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Sep 20 18:46:19 2009 +0100 [xlib] Cleanse creation of similar surfaces. Reduce two nearly identical routines to one. src/cairo-xlib-surface.c | 197 ++++++++++++++++++++++------------------------- 1 file changed, 91 insertions(+), 106 deletions(-) commit 878cef62b525e9f13ab2a8ab562ba8b6628aaa1f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Sep 20 18:45:09 2009 +0100 [xlib] Interim CAIRO_DEBUG variable to specify xrender level Use CAIRO_DEBUG=xrender-version=x.y to override what the server reports. Useful to test cairo's behaviour against older servers. src/cairo-xlib-display.c | 23 +++++++++++++++++++---- 1 file changed, 19 insertions(+), 4 deletions(-) commit 531e8045980c966b8e51df933721f9b86c6b1539 Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Fri Sep 18 12:17:46 2009 -0400 Add a description of how we compute the spline_error_squared. src/cairo-spline.c | 17 +++++++++++------ 1 file changed, 11 insertions(+), 6 deletions(-) commit f4336352405ee7c184d45a73cdd6c1a0526843db Author: Behdad Esfahbod <behdad@behdad.org> Date: Wed Sep 16 15:46:14 2009 -0400 Set LC_ALL=C instead of LANG=C doc/public/check-doc-coverage.sh | 4 +++- doc/public/check-doc-syntax.sh | 3 ++- src/check-def.sh | 3 ++- src/check-doc-syntax.sh | 3 ++- src/check-headers.sh | 3 ++- src/check-plt.sh | 3 ++- src/check-preprocessor-syntax.sh | 3 ++- 7 files changed, 15 insertions(+), 7 deletions(-) commit 56c9b2de7a2b93b2e0c59cf98326d8c0d4d508ba Author: Sebastian Dröge <sebastian.droege@collabora.co.uk> Date: Tue Sep 15 17:26:59 2009 +0200 Use __uint128_t and __int128_t if available GCC uses a peculiar name for a real 128-bit integer on x86-64. Speedups, xlib on a gm45 ======================== poppler 41246.56 -> 35102.82: 1.18x speedup swfdec-youtube 12623.01 -> 11936.79: 1.06x speedup Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> build/configure.ac.system | 2 +- src/cairo-wideint-private.h | 1 + src/cairo-wideint-type-private.h | 6 +++ src/cairo-wideint.c | 91 +++++++++++++++++++--------------------- 4 files changed, 50 insertions(+), 50 deletions(-) commit ccf84a8883ebb8d2ee8f55eb53da4dfc96887de0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Sep 15 14:22:57 2009 +0100 [perf] Reorganise cairo-perf In preparation to creating a new hub to control performance measuring and reporting, move the current cairo-perf out of the way. perf/.gitignore | 3 ++ perf/Makefile.am | 84 +++++++++++++++---------------- perf/cairo-perf-diff | 2 +- perf/{cairo-perf.c => cairo-perf-micro.c} | 0 perf/{ => micro}/box-outline.c | 0 perf/{ => micro}/cairo-perf-cover.c | 0 perf/{ => micro}/composite-checker.c | 0 perf/{ => micro}/dragon.c | 0 perf/{ => micro}/fill.c | 0 perf/{ => micro}/glyphs.c | 0 perf/{ => micro}/intersections.c | 0 perf/{ => micro}/long-dashed-lines.c | 0 perf/{ => micro}/long-lines.c | 0 perf/{ => micro}/mask.c | 0 perf/{ => micro}/mosaic.c | 0 perf/{ => micro}/mosaic.h | 0 perf/{ => micro}/paint-with-alpha.c | 0 perf/{ => micro}/paint.c | 0 perf/{ => micro}/pattern_create_radial.c | 0 perf/{ => micro}/pythagoras-tree.c | 0 perf/{ => micro}/rectangles.c | 0 perf/{ => micro}/rounded-rectangles.c | 0 perf/{ => micro}/spiral.c | 0 perf/{ => micro}/stroke.c | 0 perf/{ => micro}/subimage_copy.c | 0 perf/{ => micro}/tessellate.c | 0 perf/{ => micro}/text.c | 0 perf/{ => micro}/twin.c | 0 perf/{ => micro}/unaligned-clip.c | 0 perf/{ => micro}/world-map.c | 0 perf/{ => micro}/world-map.h | 0 perf/{ => micro}/zrusin-another.h | 0 perf/{ => micro}/zrusin.c | 0 33 files changed, 46 insertions(+), 43 deletions(-) commit 4152cd90e25ffa9d2a3683c790bd2891a58c52de Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Mon Sep 14 14:29:01 2009 +0300 [autoconf] Comment on the SIZEOF_VOID__ -> SIZEOF_VOID_P workaround. Explain where the weird SIZEOF_VOID__ comes from. src/cairo-atomic-private.h | 2 ++ util/cairo-trace/trace.c | 2 ++ 2 files changed, 4 insertions(+) commit cd94bf8bf30de9aa8dfec1a06a7ef3386944c974 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Mon Sep 14 03:41:42 2009 -0600 [trace] Look harder for the cairo-trace.so library. On OpenBSD the convention is to not use symlinks from unversioned library names to versioned library names. This breaks cairo-trace because it looks for cairo-trace.so explicitly, but on OpenBSD only cairo-trace.so.0.0 is installed. The right thing to do is probably to source the cairo-trace.la file and look for a file name there somehow. Instead this commit just looks for a likely looking file or symlink in the install directory. util/cairo-trace/cairo-trace.in | 17 ++++++++++++++++- 1 file changed, 16 insertions(+), 1 deletion(-) commit a343c8c42d3adc8f52afd09bdfbbcfd5299df49e Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Mon Sep 14 02:33:26 2009 -0600 [trace] Don't use pthread_key_delete. On OpenBSD libc doesn't provide pthread stubs like glibc on Linux or newer libcs from FreeBSD. However libX11 does provide a stubs for a subset of the pthread functions (formerly in libXThrStub, now moved into libX11 proper), but pthread_key_delete() is not one of the stubbed ones. So, on OpenBSD cairo's non-linking of libpthread accidentally works as long as the xlib-surface is enabled, which is nearly always the case. This patch makes trace.c stand at the same precipice as cairo itself by reverting to only a subset of the pthreads functions stubbed by libX11. util/cairo-trace/trace.c | 1 - 1 file changed, 1 deletion(-) commit cf0fed2191c2b5ebce8855def1dd302ccf8e039a Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Sun Sep 13 17:35:29 2009 -0600 [build] Check for dlsym in both libdl and libc. The BSDs have dlsym() in libc rather than libdl. boilerplate/cairo-boilerplate.c | 4 ++-- configure.ac | 12 +++++++----- util/cairo-trace/Makefile.am | 6 +++++- 3 files changed, 14 insertions(+), 8 deletions(-) commit 1c7ac6c47addd6825c4f49f514fe3ffb708e3813 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Sun Sep 13 17:56:56 2009 -0600 [build] Work around autoconf void* -> VOID__ name conversion bug. The autoconf shipping with OpenBSD 4.5 converts sizeof(void*) to SIZEOF_VOID__ rather than SIZEOF_VOID_P. Work around that by defining SIZEOF_VOID_P if required. src/cairo-atomic-private.h | 4 ++++ util/cairo-trace/trace.c | 4 ++++ 2 files changed, 8 insertions(+) commit 12470fd7010ea0748c8dcc9014d1cd0f52f51a3b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Sep 14 10:36:08 2009 +0100 [configure] Setup GL CFLAGS before checking for GLX configure.ac | 3 +++ 1 file changed, 3 insertions(+) commit 9c2790ccc48392bfe039626ebb383070aa18fe56 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Sep 14 10:08:57 2009 +0100 [build] Add a default message for when GLEW isn't built And it appears I touched the build system again and autoregenerated the feature headers. boilerplate/Makefile.win32.features | 8 -------- build/Makefile.win32.features-h | 1 - configure.ac | 5 +++-- src/Makefile.win32.features | 8 -------- util/cairo.modules | 1 + 5 files changed, 4 insertions(+), 19 deletions(-) commit 6d4f3c40eff7d070b1c9759038444928c6754efe Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Sep 14 09:19:10 2009 +0100 Add a private copy of GLEW This appears to be the simplest mechanism to build libglew at the moment - should a system copy be unavailable. Fortunately libglew is now distributed under a permissive licence. If you want to pass 'make -C src check' you have to use the system copy, or spend quite a bit of time cairo-fying libglew. boilerplate/Makefile.win32.features | 8 + build/Makefile.win32.features-h | 1 + build/configure.ac.features | 1 + configure.ac | 19 +- src/Makefile.am | 2 + src/Makefile.sources | 3 + src/Makefile.win32.features | 8 + src/glew/GL/glew.h | 12262 ++++++++++++++++++++++++++++++++++ src/glew/GL/glxew.h | 1397 ++++ src/glew/LICENSE.txt | 73 + src/glew/glew.c | 12180 +++++++++++++++++++++++++++++++++ util/cairo.modules | 10 +- 12 files changed, 25955 insertions(+), 9 deletions(-) commit 9ad26f74b0059645d865a20c387fa4bd460cabc8 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Sep 13 23:16:57 2009 +0100 [configure] Fallback to detect OpenGL headers When not using mesa, we can not rely on a pkgconfig file, so just rely on the presence of the OpenGL headers instead. configure.ac | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) commit 71c3b2888cc81e6d55782388d14bb8d806e77d07 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Sun Sep 13 20:24:29 2009 +0300 [win32] Sync Makefile.win32.features. Ooops.. wasn't sure if I should commit this or not. build/Makefile.win32.features | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit ba21f213560fa239a02ae28af4eb1d3a6b7254e5 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Sun Sep 13 18:21:38 2009 +0100 [trace] Don't trace internal cairo calls on Solaris. On various configurations of Solaris cairo-trace likes to trace internal calls to the API from within cairo and cairo-trace itself. On Linux the slim_hidden_proto stuff avoids this by name mangling and symbol alias magic, but on Solaris that doesn't work so we're left with either no hidden symbols at all or using normal ELF hidden symbol attributes, but neither of those prevent internal calls to cairo from being traced of course. This commit provides a per-thread entry/exit trace counter for use when we can't use name mangling to hide internal API calls. As a side-effect it may hide actual client API calls from callback functions called by cairo, but such use cases ought to be rare. util/cairo-trace/trace.c | 465 +++++++++++++++++++++++++++++++++-------------- 1 file changed, 327 insertions(+), 138 deletions(-) commit f5bcb2f36e7dba9cfba5d697e1401dcd00603812 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Sun Sep 13 12:47:19 2009 +0100 [build] Enable building cairo-trace on Solaris. Solaris supports LD_PRELOAD too! configure.ac | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit ce8c842a70d3f23527dcbd889b022103483341f2 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Sun Sep 13 12:53:07 2009 +0100 [trace] Make cairo-trace and its symbol-lookup automatically configured. The build shouldn't fail if we can't make our optional performance tools. configure.ac | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 2fb59a69f4509c2116f16bd6d376f35e3e2eb709 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Sun Sep 13 14:27:03 2009 +0100 [trace] Don't rely on the constructor attribute to initialise the tracer. Use pthread_once() at cairo API entry points to initialise the library if required. This side steps the issues with the __constructor__ attribute vs _init()/_fini() on Solaris and different tracer/tracee runtimes which result in the constructors not being run in cairo-trace.so. util/cairo-trace/trace.c | 176 ++++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 174 insertions(+), 2 deletions(-) commit 18a441984b649f558c6478f24a7987941acadbe1 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Sun Sep 13 13:53:27 2009 +0100 [trace] Don't try and propagate a void result. The pattern return <function returning void>( ... ); is a gccism not supported by Sun Studio. util/cairo-trace/trace.c | 136 ++++++++++++++++++++++++----------------------- 1 file changed, 70 insertions(+), 66 deletions(-) commit fee5c58c6caecdbdb387fe39bd6ed94faf7f6ae9 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Sun Sep 13 13:09:47 2009 +0100 [trace] Avoid warnings from assigning a void pointer to a function pointer. The Sun Studio compiler complains a *lot* when assigning the result of dlsym to a function pointer. Cast the result to the proper type first.:w util/cairo-trace/trace.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 19881012cb31ec2aebf556e2c25b53e8387dc689 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Sep 13 16:36:12 2009 +0100 [image] Add pixel masks for BGRA formats pixman introduced the BGRA pixel type not too long ago, and so we need to update our pixman_format_code_t to cairo_format_masks_t translator. src/cairo-image-surface.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) commit f8a8f297f298d81899803a1b2bbca007418408b5 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Sep 13 10:58:59 2009 +0100 [trace] Stray trailing ';' in bswap macros util/cairo-trace/trace.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit f851009ec76ee2290ffbc3f55edbf3cc16aff488 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Sep 13 10:57:42 2009 +0100 [trace] Compile without lookup-symbol And check compilation with --disable-symbol-lookup util/cairo-trace/trace.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) commit 6cb3ff9c1943c58f5b6892af3f18f6aaa24494b6 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Sep 13 10:52:14 2009 +0100 [configure] --enable-symbol-lookup There you go Joonas, I don't always ignore your suggestions! This is simple patch to allow the user to disable symbol loops in case the auto-detection fails on some obscure (perhaps OpenBSD) platform. Or in case the user really wants to trim a few bytes from a library only used during tracing! boilerplate/Makefile.win32.features | 7 +++++++ build/Makefile.win32.features | 1 + build/Makefile.win32.features-h | 3 +++ build/configure.ac.features | 1 + configure.ac | 6 ++++++ src/Makefile.win32.features | 7 +++++++ util/cairo-trace/Makefile.am | 14 +++++++++----- 7 files changed, 34 insertions(+), 5 deletions(-) commit 556d064f4eb3625022f0d99461929ff3fafc9c71 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Sep 11 15:35:27 2009 +0100 [test] Typos in README from previous commit test/README | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 43ecde2d22c5a8c591ac83f9a82c2bf83af484fd Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Sep 11 15:15:22 2009 +0100 [test] Add Debian packages for fonts Mention which .deb provides the required fonts. test/README | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) commit 33be73dfb76c26e3bb0ab59b2f570d00d9c7be62 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Sep 11 14:59:58 2009 +0100 [xlib] Fix big-line. Project lines that exceed the 16.16 limits onto the XFixedLine, as we know that top/bottom must fit within the confines of the surface and so will be less than 16 bits. Sadly this is a run-on patch that also does: 1. Make FillTiled the default for new GCs. 2. Stores extend mode as opposed to repeat, and thereby cleaning up some inconsistent code. 3. Remove the special casing for unbounded trapezoids, as it is redundant with the polygon clipping. 4. Tidy the logic for deciding when to use the core protocol (_categorize_composite_operation) src/cairo-xlib-screen.c | 6 +- src/cairo-xlib-surface-private.h | 2 +- src/cairo-xlib-surface.c | 402 ++++++++++++++++----------------------- 3 files changed, 164 insertions(+), 246 deletions(-) commit 6960162c5eae30e2d48992023be35e3dbf502a03 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Sep 11 14:56:17 2009 +0100 [xlib] Discard clip if larger than glyph extents Implement cheap calculation of glyph extents to see whether we can discard the clip region. This is effective around 50% of the time for firefox (and makes the xtrace so much neater). src/cairo-scaled-font.c | 36 ++++++++++++++++++++++++++++++++++++ src/cairo-xlib-surface.c | 23 +++++++++++++++++++++++ src/cairoint.h | 6 ++++++ 3 files changed, 65 insertions(+) commit 6e78409417117ad4de1aec65bb4753ff32e1e7a6 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Sep 11 14:19:38 2009 +0100 [test] Update big-line and remove XFAIL Now that we use polygon clipping, output geometry should no longer exceed the 2^16 limits imposed by pixman. For the image backend, we now use spans for stroking and for the xlib backend we have to double check the range on the output trapezoids. In short, cairo should pass this test. test/Makefile.am | 8 ++++---- test/big-line.c | 16 ++++++++-------- test/big-line.ps.argb32.ref.png | Bin 977 -> 0 bytes test/big-line.ps.rgb24.ref.png | Bin 709 -> 0 bytes test/big-line.quartz.ref.png | Bin 1075 -> 0 bytes test/big-line.quartz.rgb24.ref.png | Bin 941 -> 0 bytes test/big-line.ref.png | Bin 0 -> 977 bytes test/big-line.xlib-fallback.ref.png | Bin 0 -> 723 bytes test/big-line.xlib.ref.png | Bin 0 -> 933 bytes 9 files changed, 12 insertions(+), 12 deletions(-) commit b495e7eb2ee2521a400d4b42e04242023bdb1e44 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Sep 10 19:25:14 2009 +0100 [surface] Early return for (region IN white) Fill a region using IN and a white source is a frequent reduction of a complex clip (viz a path that covers the whole operation extents will be reduced to a single rectangle), and we can simply discard the fill (as it always has bounded semantics). src/cairo-surface.c | 7 +++++++ 1 file changed, 7 insertions(+) commit 94aeae9b2ec563201a0e65705504d1d70c33a619 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Sep 10 19:11:36 2009 +0100 [clip] Simplify applying a box clip If the clip path is a simple box, then for many operations it is entirely representation by the alteration of extents and so we can skip a few steps during reconstruction. src/cairo-clip-private.h | 3 +- src/cairo-clip.c | 84 ++++++++++++++++++++++++++---------------------- 2 files changed, 48 insertions(+), 39 deletions(-) commit 496e79b63762c70fd022b2d9e2d6107bcce1807a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Sep 10 17:36:46 2009 +0100 [xlib] Fix extraction of GC during XCloseDisplay() Jeff Muizelaar reported a bug from his firefox builds that was causing a double free during XCloseDisplay, and suggested it was related to c0e01d9cd. Reviewing the cleanup, suggested that the cause may just be a pair of missing parenthesis. src/cairo-xlib-screen.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit f34affaa1e0b52e255d882b84bf064f9a9d4a95d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Sep 9 18:03:20 2009 +0100 [util] And reuse the same name repository name. Completely confused by local caching of a broken cairo.modules by the jhbuild client. ARGH! util/cairo.modules | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit e1774185d6bb824a53465a037539b53fdc7af5d8 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Sep 9 17:52:10 2009 +0100 [util] And back to a unique repo name for cairo.modules Gah, further experiments into the unknown and the unclear. util/cairo.modules | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 2905ebc99e4cf8b8e94e16fcd3bc1566f51f4b57 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Sep 9 17:24:43 2009 +0100 [util] And restore the repository tag. jbhuild requires a local definition. util/cairo.modules | 2 ++ 1 file changed, 2 insertions(+) commit 4d7eb79d25bcfdc47abcae1fd51c7573043909fd Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Sep 9 17:12:14 2009 +0100 [util] Inherit the repository from xorg.modules Experiment with inheriting the repository from xorg.module, in order to avoid having two entries for the same server. util/cairo.modules | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) commit f5ee785144a8ddb6056b21f6c28bc4a9f4870b32 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Sep 9 13:59:50 2009 +0100 [test] Disable the antialias testing for the vector surfaces. Only the raster bckends obey the font anti-alias options. test/large-twin-antialias-mixed.c | 2 +- test/twin-antialias-gray.c | 2 +- test/twin-antialias-mixed.c | 2 +- test/twin-antialias-none.c | 2 +- test/twin-antialias-subpixel.c | 2 +- 5 files changed, 5 insertions(+), 5 deletions(-) commit abaef9bfbe11302beb6481547632b44be9d5be8d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Sep 9 13:57:06 2009 +0100 [test] Compare a failure against the image output If a backend fails in exactly the same way as the image, then we can safely assume that the failure is systematic and not an error in the backend, so change the result to XFAIL. test/cairo-test.c | 46 ++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 46 insertions(+) commit 41fbea053a8feb201a4f11aaf4dff86fea0bcb01 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Sep 9 12:19:51 2009 +0100 [test] Typo in ft-show-glyphs-positioning Note to self: remember to sleep more often. test/ft-show-glyphs-positioning.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit c61480a319bad55b0b4cf3f2be609bde9f2c5948 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Sep 9 11:08:17 2009 +0100 [test] An oversized twin test case. A larger variant of the twin mixed antialiasing test. test/Makefile.am | 2 + test/Makefile.sources | 1 + test/large-twin-antialias-mixed.c | 97 +++++++++++++++++++++++++++ test/large-twin-antialias-mixed.ref.png | Bin 0 -> 16668 bytes test/large-twin-antialias-mixed.xlib.ref.png | Bin 0 -> 14655 bytes 5 files changed, 100 insertions(+) commit 610b311a970877f3954ba829cf43caecce0ab72c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Sep 9 09:48:24 2009 +0100 [test] Variation of twin that intermixes antialiasing Check that we are substituting the right glyph depth and changing masks appropriately. test/Makefile.am | 2 + test/Makefile.sources | 1 + test/twin-antialias-mixed.c | 97 +++++++++++++++++++++++++++++++++ test/twin-antialias-mixed.ref.png | Bin 0 -> 2338 bytes test/twin-antialias-mixed.xlib.ref.png | Bin 0 -> 1760 bytes 5 files changed, 100 insertions(+) commit c218ff8cbec2cc7aa3efc4c7f7d8690fbdec5e9a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Sep 9 09:52:18 2009 +0100 [test] SUBPIXEL_ANTIALIAS varation of twin Force argb32 glyphs. test/Makefile.am | 2 + test/Makefile.sources | 1 + test/twin-antialias-subpixel.c | 71 ++++++++++++++++++++++++++++++ test/twin-antialias-subpixel.ref.png | Bin 0 -> 3979 bytes test/twin-antialias-subpixel.xlib.ref.png | Bin 0 -> 3863 bytes 5 files changed, 74 insertions(+) commit 456484514f4dacfe95952e8c0b73eb6a9a1a9d27 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Sep 9 09:38:18 2009 +0100 [test] Variation of twin with ANTIALIAS_GRAY Force generation of a8 glyphs. test/Makefile.am | 2 + test/Makefile.sources | 1 + test/twin-antialias-gray.c | 71 ++++++++++++++++++++++++++++++++++ test/twin-antialias-gray.ref.png | Bin 0 -> 3979 bytes test/twin-antialias-gray.xlib.ref.png | Bin 0 -> 3863 bytes 5 files changed, 74 insertions(+) commit 8555984b8e72ab5846783ed54afa1a857d70f241 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Sep 9 09:29:42 2009 +0100 [test] Add ANTIALIAS_NONE variation of twin I'm off hunting bugs. By using twin with ANTIALIAS_NONE we can construct a1 glyph images independently of the native font system. test/Makefile.am | 2 + test/Makefile.sources | 1 + test/twin-antialias-none.c | 71 ++++++++++++++++++++++++++++++++++ test/twin-antialias-none.ref.png | Bin 0 -> 718 bytes test/twin-antialias-none.xlib.ref.png | Bin 0 -> 745 bytes 5 files changed, 74 insertions(+) commit 1820cb7fd648283a5b477fbc6bd654200ed190e6 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Sep 9 09:12:54 2009 +0100 [test] Check for errors during ft-show-glyphs-positioning Do not blindly assume that we managed to construct a valid scaled-font before attempting to dereference the FT_Face. Consider a machine with no fonts which is substituting twin... test/ft-show-glyphs-positioning.c | 46 +++++++++++++++++++++++++++++++++------ 1 file changed, 39 insertions(+), 7 deletions(-) commit 7fb0d5e2091a09ef7b01574f4ecf87236ef43fec Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Sep 9 01:45:16 2009 +0100 [twin] Initialise all properties Left a couple of uninitialised properties along the non-toy font construction path. src/cairo-font-face-twin.c | 2 ++ 1 file changed, 2 insertions(+) commit 34778a883af806f77d0c89c85cebb315b43d7ef2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Sep 9 01:16:08 2009 +0100 [configure] Typo in test The have_dl clause was meant to have been removed from the requirements test for cairo-script-interpreter. Instead we left the test broken. configure.ac | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 4b221bd49a0c2c980a63779d905e8eb9b9cfa690 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Sep 8 19:02:04 2009 +0100 [spans] Correct offsets for trapezoids Pass on the destination offset from composite_trapezoids() to composite_polygon(). src/cairo-spans.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 9389cb78a8ea5b6579d091ab1030ab9f1b13efc2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Sep 8 17:51:33 2009 +0100 [build] Improve handling of missing test apparatus Improve detection, reporting and disabling of test backends when we lack the required libraries and utilities. boilerplate/Makefile.win32.features | 7 +++++++ boilerplate/cairo-boilerplate-pdf.c | 2 ++ boilerplate/cairo-boilerplate-ps.c | 2 ++ boilerplate/cairo-boilerplate-svg.c | 2 ++ build/Makefile.win32.features | 1 + build/Makefile.win32.features-h | 3 +++ build/configure.ac.features | 3 ++- configure.ac | 20 +++++++++++++------- perf/cairo-perf-compare-backends.c | 2 +- src/Makefile.win32.features | 7 +++++++ test/Makefile.am | 2 ++ test/any2ppm.c | 8 ++++++++ util/Makefile.am | 4 ++-- 13 files changed, 52 insertions(+), 11 deletions(-) commit 45759e5a0fb20a632f66a8dd206833b55a95f840 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Sep 8 09:19:53 2009 +0100 [gl] Supply extents for acquire source image I have no idea how we survived for so long without supplying the source extents... src/cairo-gl-surface.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) commit 67d40e5c7300c4082484dbda5c81808737bb2ac5 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Sep 6 10:17:40 2009 +0100 [xlib] Protect ourselves from liars that claim to have a 64k window Found using webkit, who attempt to paint an width X page height window. Please, please clip large windows to the visible area. Thanks. src/cairo-xlib-surface.c | 27 +++++++++++++++++++++++++-- 1 file changed, 25 insertions(+), 2 deletions(-) commit 310cebf194919cf3a7c37e724e64962ae47343cc Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Sep 5 15:21:50 2009 +0100 [configure] Add option to disable trace. Some environments may be broken beyond our capabilities to detect, or maybe the user is just insane and doesn't want to build my nice shiny cairo-trace. Whatever, give them the option to choose: $ ./configure --disable-trace boilerplate/Makefile.win32.features | 7 +++++++ build/Makefile.win32.features | 1 + build/Makefile.win32.features-h | 3 +++ build/configure.ac.features | 3 +++ configure.ac | 8 +++++--- src/Makefile.win32.features | 7 +++++++ util/Makefile.am | 2 +- 7 files changed, 27 insertions(+), 4 deletions(-) commit 9b932d7cd750b9bc61d108c80c4530352d952cfa Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Sep 5 08:46:39 2009 +0100 [perf] Move the calibration to its own function Improve code clarity and whatnot. perf/cairo-perf.c | 71 ++++++++++++++++++++++++++++++++----------------------- 1 file changed, 41 insertions(+), 30 deletions(-) commit 270b2f1be0e600621647f2e2486618c134217941 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Sat Sep 5 05:39:35 2009 +0300 [perf] Add a fast-and-sloppy mode to cairo-perf. A new -f option to cairo-perf reverts to a fast run mode for quick performance overviews. The number of milliseconds each iteration of a test is run for can be overriden using the new CAIRO_PERF_ITERATION_MS environment variable. The default remains 2000 ms/iter. perf/cairo-perf.c | 59 ++++++++++++++++++++++++++++++++++++------------------- perf/cairo-perf.h | 3 +++ 2 files changed, 42 insertions(+), 20 deletions(-) commit 7f91c4953ccfbb96d76ecd73715fa2e524be933e Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Sat Sep 5 05:07:25 2009 +0300 [boilerplate] Support wildcard ? in CAIRO_TEST_TARGET{,_EXCLUDE}. Useful for running tests only for a given content type. boilerplate/cairo-boilerplate.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) commit b3bafbc5f9201662702cc0c0d1c96411e31bb7d3 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Sat Sep 5 05:02:00 2009 +0300 [boilerplate] Support giving content in CAIRO_TEST_TARGET{,_EXCLUDE}. Sometimes it's convenient to run the regression or performance tests against a given target with a given content. Now we accept an optional content specifier as a suffix .<content> on a target name, where <content> is rgb or rgba. boilerplate/cairo-boilerplate.c | 61 ++++++++++++++++++++++++++++++++++++++--- 1 file changed, 57 insertions(+), 4 deletions(-) commit 29432d3d32bc84ec4a2e1815a84e4ac2089138fe Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Sat Sep 5 02:25:01 2009 +0300 [test] Reorder dash-infinite-loop to not hit a runaway allocation. This test is annoying enough as it is what with it wedging the test suite and all. There's no reason why it should DOS the running box as well by sitting in a loop allocating like mad. test/dash-infinite-loop.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 20b0e7f95448e496b609212b9608f75e03f223f8 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Fri Sep 4 23:54:21 2009 +0300 [test] Add large-source-roi.ref.png to test/Makefile.am Oops.. forgot to add it to the build files too test/Makefile.am | 1 + 1 file changed, 1 insertion(+) commit 402a0539489cc01094b464109ae28ae78b85a294 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Sep 4 09:47:10 2009 +0100 [gl] compile fix. Compiled fine during testing -- only I forgot I hadn't enable the gl surface. src/cairo-gl-surface.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit b065303740cc18fbcabfeef34d09c7d8b617647e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Sep 4 09:35:22 2009 +0100 [xlib] Enable pad_reflect by default If the XServer claims to support the required Render extension then send it the operations. However for the cases where we know it does not work, i.e. the current and previous generations (<=1.6) of Xorg servers, enable the buggy_pad_reflect fallbacks. src/cairo-xlib-display.c | 19 ++++++++++++------- 1 file changed, 12 insertions(+), 7 deletions(-) commit 634fdb4afd15e2467979609b1af7c1c198fc83d4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Sep 4 09:14:12 2009 +0100 [test] Add unbounded variants of clip-{fill,stroke} Add a couple of tests to exercise a bug that Joonas spotted that I had introduced with the clip-reduction scheme - namely that I had incorrectly removed the clip on unbounded operations. test/Makefile.am | 14 +++ test/Makefile.sources | 2 + test/clip-fill-unbounded.argb32.ref.png | Bin 0 -> 1615 bytes test/clip-fill-unbounded.c | 105 +++++++++++++++++++++ test/clip-fill-unbounded.rgb24.ref.png | Bin 0 -> 1312 bytes test/clip-fill-unbounded.svg12.argb32.xfail.png | Bin 0 -> 1601 bytes test/clip-fill-unbounded.svg12.rgb24.xfail.png | Bin 0 -> 1601 bytes ...clip-fill-unbounded.xlib-fallback.rgb24.ref.png | Bin 0 -> 1429 bytes test/clip-fill-unbounded.xlib.argb32.ref.png | Bin 0 -> 1758 bytes test/clip-fill-unbounded.xlib.rgb24.ref.png | Bin 0 -> 1426 bytes test/clip-stroke-unbounded.argb32.ref.png | Bin 0 -> 1703 bytes test/clip-stroke-unbounded.c | 102 ++++++++++++++++++++ test/clip-stroke-unbounded.rgb24.ref.png | Bin 0 -> 1383 bytes test/clip-stroke-unbounded.svg12.argb32.xfail.png | Bin 0 -> 1697 bytes test/clip-stroke-unbounded.svg12.rgb24.xfail.png | Bin 0 -> 1697 bytes ...ip-stroke-unbounded.xlib-fallback.rgb24.ref.png | Bin 0 -> 1429 bytes test/clip-stroke-unbounded.xlib.argb32.ref.png | Bin 0 -> 1792 bytes test/clip-stroke-unbounded.xlib.rgb24.ref.png | Bin 0 -> 1426 bytes 18 files changed, 223 insertions(+) commit de99f841886de5cc6e5bc91f9d3e82076d838dee Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Sep 4 08:55:19 2009 +0100 [fallback] Only eliminate the clip if the operation is bounded For unbounded operations we still need to pass along the clip in order to correctly limit the extents of the operation. src/cairo-surface-fallback.c | 57 ++++++++++++++++++++++++++++++++------------ 1 file changed, 42 insertions(+), 15 deletions(-) commit 0ffca06d7feec6fdc56c47a378fcc37c1951a00a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Sep 4 08:32:31 2009 +0100 [perf] Use milliseconds by default in cairo-perf-diff-files Necessary when comparing performance reports from different machines (or even at different CPU states). perf/cairo-perf-diff-files.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) commit 00ca0780b8f06ec160d8f969e316c21fbbde04bc Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Sep 3 23:25:53 2009 +0100 [win32] Remove unused clone_similar() The win32 backend handles surface sources directly and never calls _cairo_pattern_acquire_surface() which is the only other possible user of clone_similar(). src/cairo-win32-printing-surface.c | 2 +- src/cairo-win32-surface.c | 57 +------------------------------------- 2 files changed, 2 insertions(+), 57 deletions(-) commit 535bcaa1a1dac28fbe3106c6605949171bc36cb2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Sep 3 20:10:55 2009 +0100 [mutex] Hook into pthread last Check for native mutex types before hooking into pthread, as this workarounds broken builds on mingw that confuse us by including the pthread header file. src/cairo-mutex-impl-private.h | 47 +++++++++++++++++++++--------------------- 1 file changed, 24 insertions(+), 23 deletions(-) commit 769f4a4f47ab4636ec5cca3e92ce2c0ff19d2f75 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Sep 3 17:47:34 2009 +0100 [gl] Allocate small number of rectangles on the stack FillRectangle is most frequently used to fill an entire imagee with the background colour, i.e. with just a single, or few, rectangle. Avoid heap allocation for this common case by allocating enough space for 4 rectangles (vertices+colors) on the stack. src/cairo-gl-surface.c | 43 +++++++++++++++++++++++++++++-------------- 1 file changed, 29 insertions(+), 14 deletions(-) commit 6ce200da9d01a85a1de576229e75732db65a8b70 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Sep 3 17:38:00 2009 +0100 [gl] Assert that the error is impossible. As we created the image, it should not need coercing into a suitable format and so we should be able to upload it without failure. src/cairo-gl-surface.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit d3aeafb406da7bf69e9bf24e18a5975780a7a987 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Sep 3 17:32:28 2009 +0100 [gl] Remove reference to depth_stencil_tex We no longer use a depth-stencil, so remove the vestigial reference. src/cairo-gl-private.h | 1 - src/cairo-gl-surface.c | 2 -- 2 files changed, 3 deletions(-) commit 316c1683ceb39eb652608adc360cb4da9c22256a Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Thu Sep 3 21:38:31 2009 +0300 [test] Fix the order of random points random-intersections-curves*. The calls to uniform_random() to get the curve points were in the function arguments, but argument order evaluation is compiler implementation dependent. test/random-intersections-curves-eo.c | 17 ++++++++++------- test/random-intersections-curves-nz.c | 17 ++++++++++------- 2 files changed, 20 insertions(+), 14 deletions(-) commit 9e45673e197d0f43e296483cc6b5ca6df94e7f02 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Thu Sep 3 19:18:07 2009 +0300 [image] Check for out of bounds image surface sizes in constructors. The image surface code doesn't reliably work on images larger than 32767 in width or height. This patch makes the image surface constructors fail by returning a surface in the CAIRO_STATUS_INVALID_SIZE state when given negative or too large dimensions so that client code gets a prompt and correct error rather than flaky rendering on large images. src/cairo-image-surface.c | 27 +++++++++++++++++++++++++-- test/large-source-roi.c | 4 ++-- test/large-source-roi.ref.png | Bin 0 -> 112 bytes test/large-source.c | 4 ++-- 4 files changed, 29 insertions(+), 6 deletions(-) commit 12d0613210547b8a50dd7b21a12eb1485ee496b9 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Thu Sep 3 17:54:49 2009 +0300 [test] Allow tests to XFAIL by putting the cairo_t into an error state. The test runner was extra strict about never letting a test put the cairo_t into an error state, and never would it check for the expectedness status of the failure. This patch moves the check for a test being an XFAIL above the check on the cairo_t's final status. test/cairo-test.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) commit a76c36f2edd16d8e2b8f49e6409134aceab38a42 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Sep 3 14:29:20 2009 +0100 [xlib] solid pictures should only be 1x1 Creating an widthxheight solid picture for using with RenderCompositeTrapezoids defeats the optimization with the xserver that checks for a solid alpha pattern. The checks it performs are for CONTENT_ALPHA, Repeat, 1x1 and value == 0xff. src/cairo-xlib-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit df067be863bf714f22db0392ab29926edfa8e917 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Sep 3 08:45:40 2009 +0100 [polygon] Compute the limit of the limits slightly more efficiently Initialize the bbox to the first limit, as frequently there will only be a single (or at least a small number) limit. src/cairo-polygon.c | 24 ++++++++++++------------ 1 file changed, 12 insertions(+), 12 deletions(-) commit a1bac73f24069990842fa8f31e776b4d2c72546e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Sep 3 09:20:46 2009 +0100 [boilerplate] Handle errors whilst creating GL surface boilerplate/cairo-boilerplate-gl.c | 11 ++++++----- src/cairo-glx-context.c | 16 ++++++++-------- 2 files changed, 14 insertions(+), 13 deletions(-) commit f1d284f9976d38f636c6791f11479ae75d7bd199 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Sep 3 01:00:59 2009 +0100 [polygon] Fix discard with non-banded disjoint clip boxes The early discard checked if the line was below the last clip-box, or if above the first. However, the clip-boxes are only sorted on by the bottom (not the strict XY-banded sort of the regions) and so this was erroneously discarding lines. src/cairo-polygon.c | 29 ++++++++++++++++++++++++----- src/cairo-types-private.h | 1 + 2 files changed, 25 insertions(+), 5 deletions(-) commit a6dfdeec82ec34d88276fd0bb0ddcc94405d89f3 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Sep 3 00:59:21 2009 +0100 [test] Add clip-disjoint Soeren found another bug (thanks Soeren!) in the clipping code - as reproduced by this test case. test/Makefile.am | 2 + test/Makefile.sources | 1 + test/clip-disjoint.c | 90 ++++++++++++++++++++++++++++++++++++++++ test/clip-disjoint.ref.png | Bin 0 -> 5520 bytes test/clip-disjoint.xlib.ref.png | Bin 0 -> 5395 bytes 5 files changed, 93 insertions(+) commit ff6bc1e132836c68d4f538799c2cd26a2d9cace4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Sep 2 21:22:33 2009 +0100 [script] Always create a fallback font In order to get a baseline for win32 performance testing, always create a font so that the trace can be replayed. Not ideal, but I feel this the pragmatic solution for judging the performance differentials before I can work out a better solution for loading typ42 fonts. util/cairo-script/cairo-script-operators.c | 46 ++++++++++++++++++------------ 1 file changed, 27 insertions(+), 19 deletions(-) commit fcbb288e7cf2f5914c0d2b66ee6d81b694de6de6 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Sep 2 21:02:18 2009 +0100 [script] Attempt to select a font if ! CAIRO_HAS_FT_FONT In order to enable replay of traces on machines that do not use FreeType as the native font system, we need to convert a type42 font into something similar. Currently the fallback is just to select a font with the same name - this ignores weight and slant, and many other details. util/cairo-script/cairo-script-operators.c | 106 +++++++++++++++++++++++++++++ 1 file changed, 106 insertions(+) commit 2c03dfa0b406565fca798214a842bac69b042017 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Sep 2 10:09:50 2009 +0100 [boilerplate/test] Use numerical equality not string equality test == != -eq boilerplate/make-cairo-boilerplate-constructors.sh | 2 +- test/make-cairo-test-constructors.sh | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) commit f5df38ca5efcbc0cd3cc18d9fb67df49ec4859f8 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Wed Sep 2 04:38:52 2009 +0100 [build] Fix typos in prototype warnings. The -Wstrict-prototypes and -Wmissing-prototypes warnings weren't actually in use due to typos. build/configure.ac.warnings | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit c086b40a93057a6fd47d23c85c5026d6fe2f524a Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Wed Sep 2 04:29:18 2009 +0100 [build] Hush the Solaris compiler about enum abuse. Sun Studio 12 doesn't like it when we mix our enum values and types. We do that a lot on purpose so the warnings from compiles were very verbose. build/aclocal.cairo.m4 | 23 +++++++++++++++++++++-- build/configure.ac.warnings | 6 ++++++ 2 files changed, 27 insertions(+), 2 deletions(-) commit f081a5ff554267eebecea4652bb483eea11d1484 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Wed Sep 2 04:10:39 2009 +0100 [build] Refine the -Wno-attribute test to check our use cases. We don't actually check that -Wno-attribute does what we think it does. On clang it doesn't since it happily seems to recognize but ignore the attribute. This patch factors out a silent version of CAIRO_CC_TRY_FLAG which accepts an optional program argument and actually tests that the compiler doesn't produce any warning messages. It is then used to check that -Wno-attribute doesn't complain when the __warn_unused_result__ attribute is applied to void functions or variables. build/aclocal.cairo.m4 | 42 ++++++++++++++++++++++++++++++++++-------- build/configure.ac.analysis | 2 +- build/configure.ac.warnings | 10 ++++++---- 3 files changed, 41 insertions(+), 13 deletions(-) commit c87b366bfec4eeda2646b33cb8a33822a301456c Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Wed Sep 2 04:09:46 2009 +0100 [constructors] Guard against being called without any input files. The make-cairo-(test|boilerplate)-constructors scripts ought never to be called without arguments lest we are left constructorless. boilerplate/make-cairo-boilerplate-constructors.sh | 7 ++++++- test/make-cairo-test-constructors.sh | 7 ++++++- 2 files changed, 12 insertions(+), 2 deletions(-) commit b509b548b1e3ac5a9e3de2f9652cd1973d295fa3 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Fri Jul 24 09:12:10 2009 +0300 [trace] Check for __builtin_return_address explicitly. Some other compilers such as clang and icc support the __builtin_return_address() intrinsic as well, so we don't need to check for __GNUC__ >= 3 only. build/configure.ac.system | 12 ++++++++++++ util/cairo-trace/trace.c | 2 +- 2 files changed, 13 insertions(+), 1 deletion(-) commit 2b0e070f6a6bee415b1036fd149f0c41bcf87abb Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Fri Jul 24 07:48:57 2009 +0300 [trace] Replace an open coded test for matrix identity. The code has a _matrix_is_identity() function we can use instead of open coding the same test. util/cairo-trace/trace.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) commit 70ea9e3ce1b0037999340e484e62d3a1484da41a Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Fri Jul 24 07:46:20 2009 +0300 [trace] Don't crash on unknown enums values. If the tracer encounters an unknown enum value it ought not to crash. Theis patch replaces the idiom of looking up a name for an enumerated value directly from a table by a switch statement. As a bonus we get warnings from the compiler when the enums are updated in cairo. util/cairo-trace/trace.c | 313 ++++++++++++++++++++++++++--------------------- 1 file changed, 172 insertions(+), 141 deletions(-) commit 174c2620c82a47d067ec7b5062a4f513155a0b1f Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Fri Jul 24 07:42:57 2009 +0300 [trace] Check for object stack underflow. If the tracer's object stack underflows we want to know about is as soon as possible. This patch adds checks against the stack overflowing and aborts the program with an object stack dump if it does. util/cairo-trace/trace.c | 26 ++++++++++++++++++++++++-- 1 file changed, 24 insertions(+), 2 deletions(-) commit bb480d235882d8e7d5748a3837f9b274a8b8ac86 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Tue Jul 21 21:18:36 2009 +0300 [NEWS] Thank the AuroraUX team for facilitating Solaris testing. NEWS | 5 +++++ 1 file changed, 5 insertions(+) commit c64f6f8a15306cdcf5dd02480049b0f9112fd6ac Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Fri Jun 26 16:54:11 2009 +0300 [trace] Get the tracee program name from the environment. Support non-Linux systems which don't have a /proc/self/cmdline by transferring the application name given to cairo-trace via an environment variable CAIRO_TRACE_PROG_NAME. util/cairo-trace/cairo-trace.in | 3 ++- util/cairo-trace/trace.c | 31 +++++++++++++++++++------------ 2 files changed, 21 insertions(+), 13 deletions(-) commit 17cdffafda04514f08e0f0f11d0a85b938282c38 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Sep 2 00:41:18 2009 +0100 [xlib] Remove stray code Grr. Should have spotted this before pushing - remove the addition of the superfluous code. src/cairo-xlib-surface.c | 13 ++----------- 1 file changed, 2 insertions(+), 11 deletions(-) commit 59c4fe93ee30c8182ae1a29267b9c08602e2f6c5 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Sep 2 00:34:37 2009 +0100 [xlib] Eliminate GC clipping Eradicate the use of clipping with GC. By never using clipping, we never have to worry about retrieving a dirty clip from the GC cache. src/cairo-xlib-private.h | 12 +-- src/cairo-xlib-screen.c | 50 ++++------ src/cairo-xlib-surface-private.h | 2 - src/cairo-xlib-surface.c | 197 ++++++++++++++++++--------------------- 4 files changed, 116 insertions(+), 145 deletions(-) commit 075fc5666abe006c434bf7ec28d2d4f0226a1d3a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Sep 2 00:33:22 2009 +0100 [test] Add clip-image Exercise the XCopyArea() paths under clipping - whilst modifying that code I noticed that it was not being exercised by the test suite. test/Makefile.am | 1 + test/Makefile.sources | 1 + test/clip-image.c | 94 ++++++++++++++++++++++++++++++++++++++++++++++++ test/clip-image.ref.png | Bin 0 -> 2679 bytes 4 files changed, 96 insertions(+) commit 7d1eb259f93d3f2f2e754b2b8b90cb88359b477d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Sep 1 23:12:43 2009 +0100 [xlib] Make xlib_display_t private and rename xlib_screen_info_t The issue Joonas was trying to solve was the unwanted inclusion of the inlines via cairo-freelist-private.h. Unwittingly he included cairoint.h from cairo-xlib-private.h instead, a far more heinous crime as that causes the boilerplate to try to use the hidden, private symbols. Instead we resolve this issue by making the cairo_xlib_display_t structure private to cairo-xlib-display.c and provide functions to manipulate the abstract data type. Whilst in the vicinity, we rename cairo_xlib_screen_info_t to cairo_xlib_screen_t for consistency and cleanliness. src/cairo-xlib-display.c | 126 ++++++++++++++++++++++++++-- src/cairo-xlib-private.h | 88 ++++++++++++-------- src/cairo-xlib-screen.c | 155 ++++++++++++++++------------------- src/cairo-xlib-surface-private.h | 3 +- src/cairo-xlib-surface.c | 173 ++++++++++++++++++++++----------------- 5 files changed, 339 insertions(+), 206 deletions(-) commit b8ddd66cf6e0d16383580c3c3398343f577b89fd Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Sep 1 21:49:01 2009 +0100 Revert "[freelist] Make _cairo_freepool_alloc_from_new_pool static inline." This reverts commit 5a3fa29b370816acb3a08d60e4031ed82c1e4c73 as it breaks the boilerplate when linking with gcc. src/cairo-freelist-private.h | 29 ++--------------------------- src/cairo-freelist.c | 29 +++++++++++++++++++++++++++++ 2 files changed, 31 insertions(+), 27 deletions(-) commit 5a3fa29b370816acb3a08d60e4031ed82c1e4c73 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Tue Sep 1 20:57:01 2009 +0100 [freelist] Make _cairo_freepool_alloc_from_new_pool static inline. The xlib boilerplate includes a cairo private header to be able to disable usage of the render extension. This indirectly includes cairo-freelist-private.h which contains a bunch of static inline functions which use the private _cairo_freepool_alloc_from_new_pool function, but since that is not inline also, it causes an undefined extern which cannot be resolved. The binutils linker doesn't care since the freelist function aren't actually used in the boilerplate but the Solaris linker does. By making the .._alloc_from_new_pool function inline no dangling references are created and linking succeeds. src/cairo-freelist-private.h | 29 +++++++++++++++++++++++++++-- src/cairo-freelist.c | 29 ----------------------------- 2 files changed, 27 insertions(+), 31 deletions(-) commit 68c8eb955d6d16872c2fa488f0960b974611865d Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Mon Aug 31 23:57:20 2009 +0100 [wrapper] Avoid a void return gccism. Returning void using the pattern "return func_returning_void(...)" is a gccism not supported by Sun Studio 12. src/cairo-surface-wrapper.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 6ddab64025eacd5000fa295c86c0fddc469bdc5f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Sep 1 20:25:53 2009 +0100 [ps] Clip meta-surface to desired extents. Ensure that the meta surface does not extend beyond the operation by forcing a clip to the extents of the operation. Fixes test/device-offset and test/device-offset-positive src/cairo-ps-surface.c | 46 +++++++++++++++++++++++++++++++++++++++++++--- 1 file changed, 43 insertions(+), 3 deletions(-) commit 665f5829541703b29e33f80bebf579a17f47a892 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Sep 1 18:24:42 2009 +0100 [ps/pdf] Trim patterns to operation extents If we have to rasterise a pattern for use by PS/PDF (for example, to satisfy CAIRO_EXTENT_PAD) then only generate an image large enough to cover the operation extents. We ensure tight coverage by computing the extents afresh - we could do this lazily in the future, but we can not rely on the bounds as computed by the analysis surface as for native operations they may not be tight. src/cairo-pdf-surface.c | 67 +++++++++++++++++--- src/cairo-ps-surface.c | 76 +++++++++++++++-------- src/cairo-surface.c | 162 ++++++++++++++++++++++++++++++++++++++++++++++++ src/cairoint.h | 49 +++++++++++++++ 4 files changed, 319 insertions(+), 35 deletions(-) commit c4c7db96750337076b953a4938afc7b5f16815fc Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Sep 1 14:59:36 2009 +0100 [boilerplate] Use xlib as fallback reference for xcb And exercise RGB code-paths. boilerplate/cairo-boilerplate-xcb.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-) commit e7a118d3f82f2cbefb4a1529cc90f76ed7371ecc Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Sep 1 14:31:38 2009 +0100 [path] Standalone header-compilation Hmm, TRUE/FALSE apparently weren't defined before use. Odd. src/cairo-path-fixed-private.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 4f880deeabc36842b3b4b0d7ad16383db708cd42 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Sep 1 14:24:59 2009 +0100 Compiler warnings Add an impossible default condition to avoid a compiler warning. And tweak the code to avoid mismatching signed/unsigned comparisons. src/cairo-pdf-surface.c | 1 + src/cairo-xml-surface.c | 8 ++++---- 2 files changed, 5 insertions(+), 4 deletions(-) commit 1bcc3a3fa00445667adc47d4852237271c7eec0f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Sep 1 14:24:06 2009 +0100 [tee] Rename 'append' to 'add' and add symmetric 'remove' src/cairo-tee-surface.c | 51 ++++++++++++++++++++++++++++++++++++++++++++----- src/cairo.h | 8 ++++++-- util/cairo-fdr/fdr.c | 2 +- util/cairo-sphinx/fdr.c | 2 +- 4 files changed, 54 insertions(+), 9 deletions(-) commit af82670dd3bfdb661de5a01a6856ec01bdf80040 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Sep 1 13:31:52 2009 +0100 [pattern] Remove unused hidden symbol for cairo_pattern_status() src/cairo-pattern.c | 9 ++++----- src/cairoint.h | 1 - 2 files changed, 4 insertions(+), 6 deletions(-) commit aca1dff25967940c36caee26977512484f1720ac Author: Adrian Johnson <ajohnson@redneon.com> Date: Mon Aug 31 22:28:26 2009 +0930 PDF: Ensure operator is selected before painting surface src/cairo-pdf-surface.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) commit a402bdbd32d3ffe4b564318e6dd83831328257b1 Author: Adrian Johnson <ajohnson@redneon.com> Date: Mon Aug 31 22:15:57 2009 +0930 PDF: Flush operators before writing to the PDF stream src/cairo-pdf-surface.c | 4 ++++ 1 file changed, 4 insertions(+) commit 119355b2a4d81c903720484adaeb7dea3929ca1f Author: Adrian Johnson <ajohnson@redneon.com> Date: Mon Aug 31 22:14:15 2009 +0930 PDF: Reset alpha to 1.0 before painting a surface Fixes a test suite regression. src/cairo-pdf-surface.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) commit 20bd7d0edbb01a52bcc88bddcba4cc5237f3a673 Author: Adrian Johnson <ajohnson@redneon.com> Date: Mon Aug 31 22:03:34 2009 +0930 Update ref images test/meta-surface-pattern.pdf.argb32.ref.png | Bin 3835 -> 3908 bytes test/meta-surface-pattern.pdf.rgb24.ref.png | Bin 3682 -> 3760 bytes test/rotate-image-surface-paint.pdf.xfail.png | Bin 207 -> 217 bytes test/scale-offset-image.pdf.argb32.ref.png | Bin 0 -> 9981 bytes test/scale-offset-image.pdf.ref.png | Bin 10099 -> 0 bytes test/scale-offset-image.pdf.rgb24.ref.png | Bin 0 -> 9981 bytes test/scale-offset-similar.pdf.argb32.ref.png | Bin 0 -> 11027 bytes test/scale-offset-similar.pdf.ref.png | Bin 10890 -> 0 bytes test/scale-offset-similar.pdf.rgb24.ref.png | Bin 0 -> 11027 bytes test/surface-pattern-scale-down.pdf.ref.png | Bin 1589 -> 1603 bytes test/surface-pattern-scale-up.pdf.ref.png | Bin 2712 -> 2793 bytes 11 files changed, 0 insertions(+), 0 deletions(-) commit 135912721f5791ad8eac773fd042c751122c8760 Author: Adrian Johnson <ajohnson@redneon.com> Date: Sun Aug 30 23:58:53 2009 +0930 PDF: Avoid using patterns when filling a surface with EXTEND_NONE The fill equivilant of the previous optimization. src/cairo-pdf-surface.c | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+) commit 6512faeb9487dcc2d577b5fb3628cf7f13a25c0a Author: Adrian Johnson <ajohnson@redneon.com> Date: Sun Aug 30 23:38:17 2009 +0930 PDF: Avoid using patterns when painting surfaces with EXTEND_NONE This is an optimization the PS surface has been using to improve printing speed and prevent printers from choking on large images. Applying this optimzation to PDF prevents the same problem occuring when the PDF is converted to PS. src/cairo-pdf-surface.c | 58 +++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 58 insertions(+) commit 26d568a9ada523793b2e5855678922894e438ef4 Author: Adrian Johnson <ajohnson@redneon.com> Date: Sun Aug 30 23:50:53 2009 +0930 PDF: Fix bug when when embedding surfaces src/cairo-pdf-surface.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit feaf38d0cdc5a79966d3a412b644705b0f38e75a Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Mon Aug 31 16:25:08 2009 +0300 [test] Use HAVE_FLOCKFILE instead of _POSIX_C_SOURCE. The _POSIX_C_SOURCE 2001.. #define requires C99 mode and clang on Solaris is strict about such things. Use configure tests for flockfile() instead. test/cairo-test.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) commit 9bb469c51cf580e3ae18b21b3847488e8c4be051 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Fri Jul 24 06:15:46 2009 +0300 [trace] Use HAVE_FLOCKFILE/FUNLOCKFILE since configure test for them. The trace source was using the flockfile() and funlockfile() functions directly. util/cairo-trace/trace.c | 4 ++++ 1 file changed, 4 insertions(+) commit b5cf24a4fb0e48a0067c4b5a706890d4c1f3f1ae Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Fri Jun 26 20:06:40 2009 +0300 [build] Remove duplicate invocation of AC_CHECK_HEADERS. The fenv.h stuff was checked twice. build/configure.ac.system | 3 --- 1 file changed, 3 deletions(-) commit 3acccf0ea5ca1fde9cf6b91677588680a2644ee6 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Aug 30 12:33:40 2009 +0100 [perf] Trim outliers from chart Use "mild outliers" method to remove exceptional speed-ups and slow-downs from the graph, so that the majority of information is not lost by the scaling. Add the timing labels to the bars so that the true factor is always presented. perf/cairo-perf-chart.c | 211 +++++++++++++++++++++++++++++++++++++++--------- 1 file changed, 172 insertions(+), 39 deletions(-) commit 687462be89fd0ddf6b5412f2cb7b4f7b1a80135c Author: Carlos Garcia Campos <carlosgc@gnome.org> Date: Sun Aug 30 12:07:28 2009 +0200 [build] Fix a typo in configure.ac configure.ac | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit a6bcb6260ec8982493ac3411436ec007b40e4501 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Aug 30 10:09:51 2009 +0100 [bo-rectangular] Fix assertion failure with insertion sort We remember the location of the last insert as the next edge is likely to be nearby. However, we need to be careful when the pointer rests upon the HEAD and ensure that we begin the search from the appropriate end. src/cairo-bentley-ottmann-rectangular.c | 4 ++++ 1 file changed, 4 insertions(+) commit 19ebf83b6717e1f02e7be14218007858edf14ef0 Merge: 40aefac5d a77f1933a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Aug 29 17:45:48 2009 +0100 Merge branch 'stroke-with-spans' This branch brings self-intersection removal with virtually no performance regression. (Compare with the initial implementation that incurred a 5-10x slowdown due to having to tessellate whole strokes at a time.) The importance of self-intersection removal is the improved visual quality it brings - gone are those annoying sparkles on the outside of rounded-rectangles for instance. Most of the performance overhead associated with the self-intersection removal is avoided by switching from trapezoids to spans for strokes. Obviously we are not able to do so for the xlib backend as we do not yet have a polygon image type, and so the tessellators are overhauled instead, along with more special casing for frequent operations to avoid the increased complexity along the general paths. Speedups ======== xlib-rgba swfdec-youtube-0 11371.13 (11406.01 0.28%) -> 10450.00 (10461.84 0.66%): 1.09x speedup ▏ image-rgba firefox-talos-svg-0 73696.53 (73828.28 3.42%) -> 68324.30 (70269.79 1.36%): 1.08x speedup ▏ image-rgba swfdec-youtube-0 7843.08 (7873.89 2.57%) -> 7393.96 (7399.68 0.18%): 1.06x speedup xvfb-rgba swfdec-youtube-0 9627.25 (9634.43 0.16%) -> 9020.55 (9040.97 0.27%): 1.07x speedup ▏ Slowdowns ========= xvfb-rgba gnome-terminal-vim-0 7695.12 (7697.87 0.44%) -> 8569.45 (8588.29 0.19%): 1.11x slowdown ▏ xvfb-rgba swfdec-giant-steps-0 3811.77 (3815.06 0.23%) -> 4246.67 (4569.17 3.52%): 1.11x slowdown ▏ image-rgba gvim-0 7150.90 (7181.96 29.36%) -> 14641.04 (14651.36 0.11%): 2.05x slowdown █ One method for overcoming these regressions is to reduce the complexity of the polygons being fed into the tessellator (both in the number of edges and intersections). This should be feasible by coupling into Jeff Muizelaar's stroke-to-path work, which early indications suggest will bring a significant performance improvement. On top of this, our span implementation for the image backend is not as efficient as we would hope for - and Joonas promises a much faster implementation soon. commit a77f1933afebe28e0651906c96fd098cd8267a9f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Aug 29 14:23:41 2009 +0100 Use the more generic is_box when doing simple extent checks is_rectangle() is far stricter than is_box(), and is only required for a very limited set of operations (essentially were the rectangle must conform to the motion as described by cairo_rectangle). For the general case where we just want to know whether we have a single rectangular path that covers a certain area, is_box() is sufficient. src/cairo-pdf-surface.c | 2 +- src/cairo-ps-surface.c | 2 +- src/cairo-script-surface.c | 4 ++-- src/cairo-svg-surface.c | 2 +- 4 files changed, 5 insertions(+), 5 deletions(-) commit 0a548d08b5eae73a95fb41f41298c9c43379163e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Aug 29 16:27:27 2009 +0100 [clip] Correctly compute a geometric mask for a rectilinear + arbitrary Fix up the geometric clipper to handle intersecting a rectilinear path with an arbitrary path and inspecting the result to see if it becomes a a region. src/cairo-clip.c | 47 ++++++++++++++++++++++++++++++++++++++--------- 1 file changed, 38 insertions(+), 9 deletions(-) commit 8a323d7c8998f308fc222d65badb1289e3f7fb54 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Aug 29 14:15:07 2009 +0100 [clip] Apply surface offset when combining with clip mask In order to correctly combine the clip mask with the compositing mask the clip path must be offset so that it is relative to the destination surface. src/cairo-clip-private.h | 2 +- src/cairo-clip.c | 85 ++++++++++++++++++++++-------------------- src/cairo-region.c | 8 +++- src/cairo-surface-fallback.c | 8 +--- src/cairo-traps.c | 8 +--- src/cairo.h | 2 +- test/clip-stroke.xlib.ref.png | Bin 1563 -> 1490 bytes 7 files changed, 56 insertions(+), 57 deletions(-) commit 21225a7163bc93d34d3e395c840faaba24046bb6 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Aug 29 12:08:12 2009 +0100 [clip] Pass in destination offset for combining with clip-mask When combining a clip-mask with a subsurface, as when used to combine with the composite mask, we need to pass the destination surface offset to the clip so that the paths can be corrected for the new surface. src/cairo-clip-private.h | 4 +++- src/cairo-clip.c | 45 ++++++++++++++++++++++---------------------- src/cairo-surface-fallback.c | 14 +++++++++----- 3 files changed, 34 insertions(+), 29 deletions(-) commit ac6c6fe1d39effd5b6b382f0f1199af824868ef4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Aug 29 12:07:20 2009 +0100 [test] Add rotated clip. Exercise a bug found in not offsetting the clip mask when combining with the composite mask. test/Makefile.am | 3 ++ test/Makefile.sources | 1 + test/rotated-clip.c | 110 +++++++++++++++++++++++++++++++++++++++++ test/rotated-clip.ps.ref.png | Bin 0 -> 3378 bytes test/rotated-clip.ref.png | Bin 0 -> 3761 bytes test/rotated-clip.xlib.ref.png | Bin 0 -> 3922 bytes 6 files changed, 114 insertions(+) commit c60280782dfac7de1abe3230a500c7c98735dc02 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Aug 29 09:59:36 2009 +0100 [script] Implement invert Flesh out matrix inversion. util/cairo-script/cairo-script-operators.c | 26 +++++++++++++++++++++++++- 1 file changed, 25 insertions(+), 1 deletion(-) commit 06ca0b1475caf709fdf32b10a891dfb3b47cc6b1 Author: Damian Frank <damianf@ubuntu-vm.(none)> Date: Fri Aug 14 11:35:55 2009 -0500 Fix build on systems with older Xrender headers. This patch revises xlib so that it doesn't depend on having recent Xrender headers to build. In particular, some definitions were added to the private xrender header file, and an ifdef render version check CAIRO_SURFACE_RENDER_SUPPORTS_OPERATOR was changed to a run-time check using CAIRO_SURFACE_RENDER_HAS_PDF_OPERATORS. src/cairo-xlib-display.c | 7 ++++ src/cairo-xlib-surface.c | 30 ++------------ src/cairo-xlib-xrender-private.h | 90 ++++++++++++++++++++++++++++++++++++---- 3 files changed, 93 insertions(+), 34 deletions(-) commit 17ef949b6aaa812dd566f1db016055f8e1913320 Author: Damian Frank <damianf@ubuntu-vm.(none)> Date: Fri Aug 14 11:38:52 2009 -0500 Rename cairo-script static func to avoid MinGW conflict cairo-script-operators.c's _dup function was colliding with one defined in io.h by MinGW (gcc 4.3.0 package). I renamed it to _duplicate. util/cairo-script/cairo-script-operators.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 3e7e0eacef650f1035cfbe15e306116ff711f99c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Aug 29 17:07:41 2009 +0100 Update reference images Refresh the test reference images to match the current output where acceptable. test/Makefile.am | 14 ++++++++++++-- test/clip-fill.xlib-fallback.ref.png | Bin 1039 -> 1063 bytes test/clip-operator.pdf.argb32.ref.png | Bin 9340 -> 9266 bytes test/clip-operator.pdf.rgb24.ref.png | Bin 5146 -> 5166 bytes test/clip-operator.ref.png | Bin 8191 -> 8126 bytes test/clip-operator.rgb24.ref.png | Bin 3279 -> 3244 bytes test/clip-operator.svg12.argb32.xfail.png | Bin 8404 -> 8378 bytes test/clip-operator.svg12.rgb24.xfail.png | Bin 4523 -> 4566 bytes test/clip-operator.xlib-fallback.ref.png | Bin 3226 -> 3228 bytes test/clip-push-group.svg.ref.png | Bin 0 -> 175 bytes test/clip-stroke.ref.png | Bin 1528 -> 1451 bytes test/clip-stroke.xlib-fallback.ref.png | Bin 1566 -> 1490 bytes test/clip-stroke.xlib.ref.png | Bin 1563 -> 1563 bytes test/clip-text.svg.ref.png | Bin 933 -> 946 bytes test/clipped-group.ref.png | Bin 316 -> 289 bytes test/clipped-group.svg.ref.png | Bin 0 -> 250 bytes test/device-offset-fractional.pdf.xfail.png | Bin 275 -> 270 bytes test/extended-blend-alpha.svg12.argb32.xfail.png | Bin 6857 -> 6658 bytes test/filter-nearest-offset.pdf.xfail.png | Bin 1926 -> 2152 bytes test/filter-nearest-transformed.pdf.xfail.png | Bin 488 -> 532 bytes test/ft-text-vertical-layout-type1.svg.ref.png | Bin 3609 -> 3607 bytes test/ft-text-vertical-layout-type1.xlib.ref.png | Bin 3640 -> 3639 bytes test/ft-text-vertical-layout-type3.svg.ref.png | Bin 3626 -> 3626 bytes test/ft-text-vertical-layout-type3.xlib.ref.png | Bin 3605 -> 3605 bytes test/group-unaligned.svg.argb32.xfail.png | Bin 522 -> 520 bytes test/leaky-dashed-rectangle.pdf.ref.png | Bin 347 -> 380 bytes test/mask-glyphs.svg.ref.png | Bin 1211158 -> 1211144 bytes test/mask.pdf.argb32.ref.png | Bin 7837 -> 7866 bytes test/mask.pdf.rgb24.ref.png | Bin 7301 -> 7336 bytes test/mask.svg.argb32.xfail.png | Bin 8613 -> 8641 bytes test/mask.svg.rgb24.xfail.png | Bin 7189 -> 7199 bytes test/operator-source.svg12.argb32.xfail.png | Bin 2951 -> 2967 bytes test/operator-source.svg12.rgb24.xfail.png | Bin 2963 -> 2976 bytes test/overlapping-glyphs.svg.argb32.ref.png | Bin 0 -> 2338 bytes test/overlapping-glyphs.svg.rgb24.ref.png | Bin 0 -> 2338 bytes test/radial-gradient.pdf.ref.png | Bin 277162 -> 280944 bytes test/radial-gradient.svg.xfail.png | Bin 0 -> 264425 bytes test/random-intersections-curves-eo.ref.png | Bin 246996 -> 245762 bytes ...m-intersections-curves-eo.xlib-fallback.ref.png | Bin 246243 -> 244255 bytes test/random-intersections-curves-eo.xlib.ref.png | Bin 246243 -> 244255 bytes test/random-intersections-curves-nz.ref.png | Bin 267589 -> 264545 bytes ...m-intersections-curves-nz.xlib-fallback.ref.png | Bin 266364 -> 264288 bytes test/random-intersections-curves-nz.xlib.ref.png | Bin 266360 -> 264288 bytes ...df.xfail.png => scale-offset-image.pdf.ref.png} | Bin 10099 -> 10099 bytes test/scale-offset-image.script.xfail.png | Bin 0 -> 9963 bytes test/scale-offset-similar.pdf.ref.png | Bin 0 -> 10890 bytes test/scale-offset-similar.pdf.xfail.png | Bin 11024 -> 0 bytes test/scale-offset-similar.script.xfail.png | Bin 0 -> 9963 bytes test/self-intersecting.ref.png | Bin 301 -> 213 bytes test/smask-text.script.ref.png | Bin 0 -> 1663 bytes test/smask.pdf.xfail.png | Bin 4393 -> 2848 bytes test/smask.script.ref.png | Bin 0 -> 3396 bytes test/text-pattern.pdf.argb32.ref.png | Bin 1831 -> 1810 bytes test/text-pattern.pdf.rgb24.ref.png | Bin 1598 -> 1559 bytes test/unbounded-operator.svg12.rgb24.xfail.png | Bin 1719 -> 1731 bytes test/user-font-proxy.svg.ref.png | Bin 16817 -> 16814 bytes test/user-font.pdf.ref.png | Bin 0 -> 6241 bytes 57 files changed, 12 insertions(+), 2 deletions(-) commit 155e10e632cb647df5c3d54a75975bc16591287d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Aug 29 08:04:33 2009 +0100 [script] Fix use of freed list A typo, I missed converting the user over to the freshly sorted list, leaving it iterating over original but checking the sorted for termination conditions. src/cairo-script-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 5393aa6d6c4676f20d316f3cd0a18bb497574e50 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Aug 29 08:02:52 2009 +0100 [path] Return the canonical box. When returning the single box that represents a path, always return it consistently wound. src/cairo-path-fill.c | 32 -------------------------------- src/cairo-path-fixed.c | 28 ++++++++++++++++++++++++---- 2 files changed, 24 insertions(+), 36 deletions(-) commit afea5eb79d2159fe9a5dc1a1a7b9445e40fbb474 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Aug 28 17:10:35 2009 +0100 [scaled-font] Fix implementation-face refleak If we found the font via the holdover cache, or if we returned due to an error, we would leak a reference count on the implementaton face. src/cairo-scaled-font.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) commit e5d44937f34d35606ad7f07415be7331cf146567 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Aug 28 17:09:54 2009 +0100 [ft] Improve error path handling. Specifically check for an error during resolving the font and do not cache the error object. src/cairo-ft-font.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) commit e76856e6ee563affc0a273ed6eef865866802dec Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Aug 28 15:51:33 2009 +0100 [scaled-font] Refleak on error path. Perform the destroy of the local font before returning along the error path. src/cairo-scaled-font.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) commit d7d6f75ed26bc3ea040c618442ba71c89633ac7b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Aug 28 15:50:13 2009 +0100 [clip] Fix refleak of previous clipping surfaces. When combining previous clip masks, we leaked a referenced due to not destroying the returned reference. src/cairo-clip.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) commit 52e5ce3a36ab1cd7da25056df7f12ac0b23e410d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Aug 27 17:25:08 2009 +0100 [test] Add clip-stroke Soeren was the first to report a clipping regression in the xlib backend with strokes, and provided a test case to exercise the bug. This is an extension of his test to provide coverage of different clipping and stroking methods. test/Makefile.am | 3 + test/Makefile.sources | 1 + test/clip-stroke.c | 121 +++++++++++++++++++++++++++++++++ test/clip-stroke.ref.png | Bin 0 -> 1528 bytes test/clip-stroke.xlib-fallback.ref.png | Bin 0 -> 1566 bytes test/clip-stroke.xlib.ref.png | Bin 0 -> 1563 bytes 6 files changed, 125 insertions(+) commit 20cdb99ae8ae0cc02193468e811b1b74b3f6d3b7 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Aug 27 16:21:30 2009 +0100 [path-fixed] Distinguish cw and ccw boxes To correctly handle retessellating trapezods constructed from alternately wound boxes, then we need to pass that information from the path to the tessellator. We do this by switching the direction of the box if the first edge is horizontal as opposed to vertical. src/cairo-path-fixed.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 219d46a9ebc47f99fa8fe9c6e3f0aa440309f032 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Aug 27 14:45:24 2009 +0100 [scaled-font] Eliminate intermediate path when tracing glyphs Currently the tracing code for glyphs constructs an temporary path in order to replay and append to the output. This temporary allocation is extremely wasteful as we can just directly append the glyph path to the output path. src/cairo-scaled-font.c | 51 +++++++++++++++++++++---------------------------- 1 file changed, 22 insertions(+), 29 deletions(-) commit 2e05922737d63289a3f124699359b8d385315cbd Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Aug 27 13:58:43 2009 +0100 [stroke] Handle degenerate stroke extents If the stroke is degenerate, i.e. the path consists only of a single move-to and no edges, then the stroke may be visible due to end-capping (as opposed to fills which are empty). So we also need to pad out the extents around the current point for the degenerate case. src/cairo-path-bounds.c | 31 +++++++++++++------------------ 1 file changed, 13 insertions(+), 18 deletions(-) commit cfd78393f357bc69233d4d00d0fb3a2ff736f1a7 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Aug 26 23:32:34 2009 +0100 [path] Handle the implicit close for path_fixed_is_box() _cairo_path_fixed_is_box() is only called for filled paths and so must handle the implicit close (which was already being correctly handled by _cairo_path_fixed_iter_is_box). src/cairo-path-fixed.c | 33 ++++++++++++++++++--------------- 1 file changed, 18 insertions(+), 15 deletions(-) commit f22045bb4b9e700ce223c259ad41403dc7efe81f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Aug 26 23:30:02 2009 +0100 [fallback] Include implicit closes in the check for rectilinear paths Fixes test/implicit-close By forgetting the implicit-close when checking for rectilinear paths, we tried to feed the triangle (and other diagclose) into the specialised rectilinear tesselators which completely mishandled that final edge. src/cairo-path-fixed-private.h | 14 ++++++++++++++ src/cairo-surface-fallback.c | 6 ++++-- 2 files changed, 18 insertions(+), 2 deletions(-) commit 92f6f275fcb5407baf908485ffd08b6787b2caf9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Aug 26 23:27:46 2009 +0100 [test] Implicit close This is a simple test that broke with the determination of rectilinearity during path construction. I forgot the implicit close on fill and so the ignored the final diagonal edge and failed to draw the triangle. test/Makefile.am | 1 + test/Makefile.sources | 1 + test/implicit-close.c | 54 ++++++++++++++++++++++++++++++++++++++++++++ test/implicit-close.ref.png | Bin 0 -> 251 bytes 4 files changed, 56 insertions(+) commit 8078cd194e95a10cf653c970d1ddd39049a511f2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Aug 23 15:25:42 2009 +0100 [boilerplate] Runtime library check For the purposes of benchmarking it is useful to run cairo-perf against a different library from the one it was compiled against. In order to do so, we need to check that the runtime library contains the required entry points for our targets - which we can check by using dlsym. boilerplate/Makefile.am | 3 +++ boilerplate/cairo-boilerplate-directfb.c | 2 ++ boilerplate/cairo-boilerplate-drm.c | 2 ++ boilerplate/cairo-boilerplate-gl.c | 2 ++ boilerplate/cairo-boilerplate-glitz-agl.c | 2 ++ boilerplate/cairo-boilerplate-glitz-glx.c | 2 ++ boilerplate/cairo-boilerplate-glitz-wgl.c | 2 ++ boilerplate/cairo-boilerplate-pdf.c | 2 ++ boilerplate/cairo-boilerplate-ps.c | 4 ++++ boilerplate/cairo-boilerplate-quartz.c | 2 ++ boilerplate/cairo-boilerplate-script.c | 1 + boilerplate/cairo-boilerplate-skia.c | 2 ++ boilerplate/cairo-boilerplate-svg.c | 4 ++++ boilerplate/cairo-boilerplate-test-surfaces.c | 8 +++++++ boilerplate/cairo-boilerplate-vg.c | 4 ++++ boilerplate/cairo-boilerplate-win32-printing.c | 2 ++ boilerplate/cairo-boilerplate-win32.c | 2 ++ boilerplate/cairo-boilerplate-xcb.c | 1 + boilerplate/cairo-boilerplate-xlib.c | 4 ++++ boilerplate/cairo-boilerplate.c | 29 +++++++++++++++++++++++--- boilerplate/cairo-boilerplate.h | 1 + configure.ac | 7 +++++++ 22 files changed, 85 insertions(+), 3 deletions(-) commit 7447915381fc64bd0c66f7110c1dd0b8a10d73f5 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Aug 22 22:20:35 2009 +0100 [script] Conditionalise the build Allow the interpreter to be disabled if we can not satisfy its dependencies. configure.ac | 2 ++ util/Makefile.am | 11 +++++++++-- util/cairo-script/Makefile.am | 2 +- 3 files changed, 12 insertions(+), 3 deletions(-) commit fcda9fc2f29c2ee7d0bb45c17a2261badfdf430c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Aug 22 18:48:33 2009 +0100 [scaled-font] Remove assert from cairo_scled_font_create() The assert() is only correct for the normal paths, but failed on the error path. It has been run for long enough for me to be confident that the code is self-consistent, so I think I can now safely remove it. src/cairo-scaled-font.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) commit 93cfa7376fb505e676d113e8ef431bab9b497f56 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Aug 22 18:45:54 2009 +0100 [win32] Use the system scaled_font_done Pointless as both functions are empty, but lets be pedantically correct nevertheless. src/cairo-win32-font.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) commit 8654a4b35c060ab06860c124a643be912060c4a1 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Aug 22 18:16:20 2009 +0100 [win32] Initialize clip_region Eek, I attempted to destroy an uninitialised region on the first use of a clip. src/cairo-win32-surface.c | 3 +++ 1 file changed, 3 insertions(+) commit 9b33a2e1c794996cbe4cb0a5c773d50566d46bd4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Aug 22 17:40:12 2009 +0100 [win32] Compiler warnings Innocuous warnings about the use of mismatching explicit casts (I'm really not convinced by the merits of this particular compiler warning, but it does cleanse the code slightly.) src/cairo-win32-printing-surface.c | 24 ++++++++++++------------ 1 file changed, 12 insertions(+), 12 deletions(-) commit d1740d8782d90edb5b5e20dc5bcadb7a9eeeb4cc Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Aug 20 21:11:11 2009 +0100 [pattern] Ensure that no repeated pattern is clipped Previously the pattern_acquire_surface routine only had to worry about handling extend modes NONE or REPEAT and so the test for ! REPEAT sufficed when what was actually intended was a test for NONE. src/cairo-pattern.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 958f7ab1238b8242975d3ea774c26d23e090e70c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Aug 19 12:35:00 2009 +0100 [script] Support running on cairo-1.8 Pre 1.9 the application had to pass a resolved font to cairo -- so do so in the interpreter if the cairo version is less than 1.9 util/cairo-script/cairo-script-operators.c | 18 +++++++++++++++--- 1 file changed, 15 insertions(+), 3 deletions(-) commit 87175334a574fa0f69679b1a0baeeb881eaa439b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Aug 21 15:51:12 2009 +0100 [gl] Use spans for trapezois. Always use spans, even for unaligned boxes. In the future (given a new interface) we may want to emit the common unaligned box code more efficient than a per-scanline computation -- but for now simply avoid the requirements to write a temporary CPU buffer. src/cairo-gl-surface.c | 22 ++++++++++++++++++---- src/cairo-image-surface.c | 4 +--- src/cairo-surface-fallback.c | 10 ++-------- src/cairo-surface.c | 27 ++++++++++----------------- src/cairo-win32-surface.c | 4 +--- src/cairoint.h | 18 ++++++++++++++---- 6 files changed, 46 insertions(+), 39 deletions(-) commit e65dfacab5d123913a4b2f59967f27c0b0af0692 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Aug 21 15:50:24 2009 +0100 [gl] Simplify acquire_dest_image() Remove redundant code: the intersection of surface extents and the copy to a temporary buffer. src/cairo-gl-surface.c | 53 ++++++++++++-------------------------------------- 1 file changed, 12 insertions(+), 41 deletions(-) commit 5a13396373180ceae31287441ef9c838c771849d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Aug 21 15:48:38 2009 +0100 [gl] Use common ARRAY_LENGTH macro Forgo the local ARRAY_SIZE macro where the common one will suffice. src/cairo-gl-surface.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) commit efdb53425ee71b3bce6c92ce212f5baf3e2a43d7 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Aug 21 11:58:31 2009 +0100 [qt] Discard impossible status return from path construction As we never return an error status during the path construction, we can use the return value for the QPainterPath instead, greatly simplifying the callers. src/cairo-qt-surface.cpp | 105 +++++++++++++++++++---------------------------- 1 file changed, 42 insertions(+), 63 deletions(-) commit f2cde41cabaa0c190ef965f634cf2c73b6ba2334 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Aug 21 10:29:39 2009 +0100 [cairoint.h] Add missing cairo_private to debug prototypes. src/cairoint.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit b6d96bba8a4e9efe1033dd7fc875aba9bd8d0dc8 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Aug 20 20:31:49 2009 +0100 [win32] Trust the clipping code to trim roi to our surface The higher level code ensures that the region of interest is trimmed to our declared surface extents, so performing the intersection again is redundant. Furthermore with the change in the clipping code, the fallback region is no longer clipped, especially as the clip that is currently set upon the DC is likely to be stale and incorrect for the fallback. Hopefully this resolves the assertion failure reported by Damian Frank, http://lists.cairographics.org/archives/cairo/2009-August/018015.html CC: Damian Frank <damian.frank@gmail.com> src/cairo-win32-surface.c | 53 ++++++++--------------------------------------- 1 file changed, 9 insertions(+), 44 deletions(-) commit d7faec024ac9f702c9bb305599ef8960a972491a Author: Vladimir Vukicevic <vladimir@pobox.com> Date: Mon Jun 15 11:15:22 2009 +0100 Add skia backend Originally written by Vladimir Vukicevic to investigate using Skia for Mozilla, it provides a nice integration with a rather interesting code base. By hooking Skia underneath Cairo it allows us to directly compare code paths... which is interesting. [updated by Chris Wilson] boilerplate/Makefile.sources | 1 + boilerplate/Makefile.win32.features | 10 + boilerplate/cairo-boilerplate-skia.c | 50 ++ build/Makefile.win32.features | 1 + build/Makefile.win32.features-h | 3 + build/configure.ac.features | 1 + configure.ac | 13 + perf/cairo-perf-trace.c | 3 + perf/cairo-perf.c | 3 + src/Makefile.sources | 3 + src/Makefile.win32.features | 14 + src/cairo-skia-surface.cpp | 1174 ++++++++++++++++++++++++++++++++++ src/cairo-skia.h | 84 +++ src/cairo.h | 4 +- 14 files changed, 1363 insertions(+), 1 deletion(-) commit af6df4af51ad75a956e3b73542647206ab534bd2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Aug 20 12:54:12 2009 +0100 [configure] Choose a better name for the 'temporary' egl iterator The problem is that it remains referenced by egl_NONPKGCONFIG_LIBS and so must not be subsequently overwritten, so give it an egl_ prefix. configure.ac | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) commit 83c2f44dd954cb491d72834d0362d6d801ef7651 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Aug 20 12:16:55 2009 +0100 [test] Update REFERENCE_IMAGES test/Makefile.am | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) commit f99e1841448e98be6161be57551153a6d9e28a13 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Aug 20 12:11:51 2009 +0100 [build] Remove -Wcast-align Use -Wcast-align simply generates too much noise due to false-positive reports when casting pixels to uint32_t. build/configure.ac.warnings | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) commit 5fdf5b311e06e49843e924523e9bc91ba14da3bb Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Aug 19 08:35:01 2009 +0100 [fallback] Reduce paint + clipmask to fill Under simple, yet common, conditions using a bounded operator and painting with a single complex clip we can reduce the strength of that operation to a fill. In effect this removes the need for a temporary mask for some backends (GL, drm, xlib). src/cairo-surface-fallback.c | 26 +++++++++++++++++--------- 1 file changed, 17 insertions(+), 9 deletions(-) commit 425b0e35e2d7d44d99e22169b98fc3ca05763650 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Aug 18 14:25:25 2009 +0100 Add xml surface A very simple surface that produces a hierarchical DAG in a simple XML format. It is intended to be used whilst debugging, for example with the automatic regression finding tools of cairo-sphinx, and with test suites that just want to verify that their code made a particular Cairo call. boilerplate/Makefile.win32.features | 10 + build/Makefile.win32.features | 1 + build/Makefile.win32.features-h | 3 + build/configure.ac.features | 1 + configure.ac | 4 + src/Makefile.sources | 18 +- src/Makefile.win32.features | 14 + src/cairo-base64-stream.c | 143 ++++ src/cairo-base85-stream.c | 3 - src/cairo-output-stream-private.h | 4 + src/cairo-ps-surface.c | 4 + src/cairo-script-surface.c | 18 +- src/cairo-xml-surface.c | 1153 ++++++++++++++++++++++++++++++ src/cairo-xml.h | 72 ++ src/cairo.h | 4 +- util/Makefile.am | 11 +- util/cairo-script/cairo-script-scanner.c | 113 +++ util/trace-to-xml.c | 77 ++ util/xml-to-trace.c | 263 +++++++ 19 files changed, 1902 insertions(+), 14 deletions(-) commit 6e0b3be903a6c3ad948de95f490cff92430429a6 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Aug 18 10:55:54 2009 +0100 Add cairo-sphinx utility for regression analysis sphinx is an alternate version of the current cairo-test-trace. It's purpose is to replay a live application (which may just be a replay of a trace) against a backend and its reference. The improvement over the original cairo-test-trace is that the reference backend may be from an older version of cairo. configure.ac | 6 + util/Makefile.am | 6 + util/cairo-sphinx/.gitignore | 1 + util/cairo-sphinx/Makefile.am | 40 ++ util/cairo-sphinx/fdr.c | 260 +++++++ util/cairo-sphinx/sphinx.c | 1525 +++++++++++++++++++++++++++++++++++++++++ 6 files changed, 1838 insertions(+) commit c980affce05590f5f52273ba340463f00773c776 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Aug 18 10:54:26 2009 +0100 [script] Apply device offset when replaying meta surface As we set the size of the surface to fit the ink extents of the meta surface, we also need to ensure that the origin of the script lies at the origin of the ink extents. src/cairo-script-surface.c | 2 ++ 1 file changed, 2 insertions(+) commit eb33f842dc9a2555ba8f7948f49a8335db951959 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Aug 17 11:38:52 2009 +0100 [meta] Missing status check We need to check the status on the created image before use. src/cairo-meta-surface.c | 2 ++ 1 file changed, 2 insertions(+) commit ce6a2cc5d2cb8a299759d764de2e7d2b6b655cb4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Aug 17 10:31:47 2009 +0100 [wrapper] Always copy clip We always need to make a local copy of the clip as the backends are free to modify it as they process the operation. src/cairo-surface-wrapper.c | 60 +++++++++++++++++++++++++++++++++------------ 1 file changed, 44 insertions(+), 16 deletions(-) commit 6ff711b6305a9cf65e584d92258a6fa4e78c31ef Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Aug 17 10:11:32 2009 +0100 [matrix] Improve bbox finding for translation matrix If the matrix is a pure translation matrix than we can skip determination of the extents and just translate the input bbox. src/cairo-matrix.c | 56 ++++++++++++++++++++++++++++-------------------------- 1 file changed, 29 insertions(+), 27 deletions(-) commit cd7b27ff5c01a533c2c065c4b455ad19df2be3bb Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Aug 16 21:08:22 2009 +0100 Add 'flight-data-recorder' utility. This is a simple variation on cairo-trace that wraps records the last 16 contexts by wrapping the target surface inside a tee surface, along with a meta/recording surface. Then on receipt of a SIGUSR1, those last 16 contexts are played via a script-surface into /tmp/fdr.trace. Mostly proof-of-concept, it seems to be causing a number of rendering glitches whilst testing with firefox -- otherwise, it seems to works. configure.ac | 1 + src/cairo-scaled-font-private.h | 3 + src/cairo-scaled-font.c | 16 +++ src/cairo-script-surface.c | 58 +++++++- src/cairo-script.h | 4 + src/cairo.h | 4 + util/Makefile.am | 3 + util/cairo-fdr/Makefile.am | 13 ++ util/cairo-fdr/fdr.c | 304 ++++++++++++++++++++++++++++++++++++++++ 9 files changed, 400 insertions(+), 6 deletions(-) commit 658cdc7c9aac23f82f3ea5db8df10844aeb3ac75 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Aug 16 18:04:54 2009 +0100 Introduce cairo_tee_surface_t Add a new surface type that multiplies it input onto several output surfaces. The only limitation is that it requires a master surface that is used whenever we need to query surface options, such as font options and extents. boilerplate/Makefile.win32.features | 8 + build/Makefile.win32.features-h | 1 + build/configure.ac.features | 1 + configure.ac | 1 + src/Makefile.sources | 6 +- src/Makefile.win32.features | 8 + src/cairo-surface-wrapper-private.h | 4 + src/cairo-surface-wrapper.c | 7 + src/cairo-surface.c | 11 + src/cairo-tee-surface-private.h | 47 +++ src/cairo-tee-surface.c | 558 ++++++++++++++++++++++++++++++++++++ src/cairo.h | 13 +- 12 files changed, 662 insertions(+), 3 deletions(-) commit 4ae7186719b25f052b875549cc5377e16a557512 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Aug 16 17:07:00 2009 +0100 [script] Correct emission of get_target() for a popped context If the context is no longer on the operand stack, then we need to recall the surface via a definition. src/cairo-script-surface.c | 53 ++++++++++++++++++++++++++++------------------ 1 file changed, 32 insertions(+), 21 deletions(-) commit 8d1bf830c0137eac837091bda92a636c0fcb0456 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Aug 16 14:20:35 2009 +0100 Fix errors found by clang Shadowed variables, unused writes and some dead code. src/cairo-pdf-surface.c | 2 -- src/cairo-xlib-surface.c | 5 ++--- util/cairo-script/cairo-script-file.c | 10 +++------- util/cairo-script/cairo-script-operators.c | 23 ++++++++++++++++------- util/cairo-trace/trace.c | 14 ++++++++------ 5 files changed, 29 insertions(+), 25 deletions(-) commit d8dbce021a4493330864154e67ca6e4a1f2f50b2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Aug 16 13:26:59 2009 +0100 [script] recursive active (type3 glyphs) The assumption that an active surface could not recurse into another operation was invalid - due to the complexity of handling type3 glyphs. src/cairo-script-surface.c | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) commit eba6b5126a55c84706e677b3fc88743f64cc28d7 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Aug 16 12:47:14 2009 +0100 [build] Add options for warning about bad casts -Wbad-function-cast in particular. Triggers quite a few warnings where we have explicitly cast to an integer. build/configure.ac.warnings | 9 +- src/cairo-arc.c | 2 +- src/cairo-pdf-surface.c | 8 +- src/cairo-ps-surface.c | 20 ++-- src/cairo-svg-surface.c | 4 +- util/cairo-script/cairo-script-objects.c | 4 +- util/cairo-script/cairo-script-operators.c | 172 +++++++++++++++++++---------- util/cairo-script/cairo-script-private.h | 6 +- 8 files changed, 143 insertions(+), 82 deletions(-) commit 33ef32af4e7adce41f035ee378279c19577469b0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Aug 16 12:46:20 2009 +0100 [clip] Use the rectangular tessellator to extract boxes src/cairo-clip.c | 17 ++++++++++------- src/cairo-path-fill.c | 29 ++++++++++++++++++++++------- src/cairo-surface-fallback.c | 8 ++++---- 3 files changed, 36 insertions(+), 18 deletions(-) commit ab035ab2c7bec254fc94d6391398905b5039e777 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Aug 28 10:06:04 2009 +0100 [tessellate] Rectangular special case Add an even simpler sweep-line tessellator for rectangular trapezoids (as produced by the rectilinear stoker and box filler). This is so simple it even outperforms pixman's region validation code for the purposes of path-to-region conversion. src/Makefile.sources | 1 + src/cairo-bentley-ottmann-rectangular.c | 733 ++++++++++++++++++++++++++++++++ src/cairo-clip.c | 26 +- src/cairo-path-fill.c | 40 +- src/cairo-path-stroke.c | 3 +- src/cairo-surface-fallback.c | 95 ++++- src/cairo-traps.c | 28 +- src/cairoint.h | 5 + 8 files changed, 883 insertions(+), 48 deletions(-) commit d7b0c3b784faba756b10b66b9757e6e4c3fce38c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Aug 13 22:03:06 2009 +0100 [script] Track scaled-font Instead of emitting the (font-face, matrix and options) elements when setting up the desired font on the matrix, simply restore the scaled-font. src/cairo-script-surface.c | 39 ++++++++++++++++++++++----------------- 1 file changed, 22 insertions(+), 17 deletions(-) commit 7306305cc898c9f27957c6c6717028cbdcff6a5a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Aug 13 21:59:12 2009 +0100 [script] Emit surface content Include the desired content with the creation info. boilerplate/cairo-boilerplate-script.c | 2 +- src/cairo-script-surface.c | 34 +++++++++++++++++++++------------- src/cairo-script.h | 1 + util/cairo-trace/trace.c | 5 ++++- 4 files changed, 27 insertions(+), 15 deletions(-) commit 052211b072788f0977dccebdcf681d4874f2487a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Aug 13 20:16:51 2009 +0100 [script] Garbage collect contexts on context switch Previously the contexts were permanently associated with the surface and only destroyed along with the final reference to the surface. This meant that we kept a large number of unwanted contexts in memory. Most applications only have a few contexts active at any time, so remove inactive contexts from the operand stack every time we perform an operation on a different context. src/cairo-script-surface.c | 262 +++++++++++++++++++++++++++++++-------------- 1 file changed, 184 insertions(+), 78 deletions(-) commit 8f8b91d9049433c9210e0b3aad574cf659862ae0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Aug 13 18:37:01 2009 +0100 [script] Wrap snapshot. Use the snapshot of our target surface if available. src/cairo-script-surface.c | 20 ++++++++++++++++---- src/cairo-surface-wrapper-private.h | 3 +++ src/cairo-surface-wrapper.c | 6 ++++++ 3 files changed, 25 insertions(+), 4 deletions(-) commit fee647c98506eedad0fea667a9442786cb0804b7 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Aug 13 18:35:48 2009 +0100 [script] Read from a FILE stream For ease of implementing the interpreter inside a pipeline, add a method to execute a FILE *. util/cairo-script/cairo-script-interpreter.c | 22 ++++++++++++++++++++++ util/cairo-script/cairo-script-interpreter.h | 4 ++++ 2 files changed, 26 insertions(+) commit a9d997fecd54cea7dcd71487a24dbae14d0073a8 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Aug 13 17:21:05 2009 +0100 [script] Introduce cairo_script_context_t cairo_script_context_t is an encapsulation object for interfacing with the output - multiple surfaces can share the same context, meaning that they write to the same destination file/stream. boilerplate/cairo-boilerplate-script.c | 5 +- boilerplate/cairo-boilerplate.c | 26 +--- src/cairo-script-surface.c | 225 ++++++++++++++------------------- src/cairo-script.h | 47 ++++--- test/cairo-test-trace.c | 5 +- util/cairo-script/csi-trace.c | 13 +- util/cairo-trace/trace.c | 20 +-- 7 files changed, 145 insertions(+), 196 deletions(-) commit 28887ac272c8a36a41da4d6d58044164b94da6f3 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Aug 13 16:38:08 2009 +0100 [trace] Fix emission of similar (track operands) util/cairo-trace/trace.c | 13 ++++++------- 1 file changed, 6 insertions(+), 7 deletions(-) commit dbd9438f5d01666ea118254427c0dc66a519ec3e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Aug 13 15:57:30 2009 +0100 [stroke] Only mark traps as having intersection if non-empty. We were hitting an assertion attempting to eliminate intersections inside the rectilinear tessellator for empty strokes. We can avoid this assertion, by only marking the traps as having potential intersections iff it is non-empty. src/cairo-path-stroke.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) commit 60d73da9f2e148e982254c78773f0b925be184ff Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Aug 24 08:04:12 2009 +0100 [clip] Cache intermediate clip masks. As we now superimpose a per-operation clip, this defeats the current top-level caching mechanism. Instead we need to cache the mask for each path. This still seems quite wasteful, and an avenue would be to avoid caching if the path is rectilinear and reduce the number of required composite operations. (However, first find test case...) src/cairo-clip.c | 165 ++++++++++++++++++++++++++++++++++--------------------- src/cairoint.h | 3 + 2 files changed, 104 insertions(+), 64 deletions(-) commit 50c7d637f3a97031c5f4e2bc8d41d17e5397db45 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Aug 13 15:49:11 2009 +0100 [trace] Emit an 'index' for an undefined surface Larry Ewing hit a bug in cairo-trace whereby it tried to create a similar surface referencing an undefined object. This fix checks whether the object has yet to be defined, and if not issues an index in order to copy the appropriate operand from the stack. util/cairo-trace/trace.c | 21 ++++++++++----------- 1 file changed, 10 insertions(+), 11 deletions(-) commit 3a483c2896c28142a90bd0e282af3862e066adfe Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Aug 13 00:19:03 2009 +0100 [gstate] Convert simple mask() into a paint() As using mask() prevents various optimisations in the backends (for example the use of geometric clips) and for some may trigger fallbacks, perform the simplifications usually done (too late) by the pattern layer in the generic gstate layer. This allows us on the odd occasion to transform a mask() into a paint() but perhaps more importantly removes the need for identical transformations in each backend. src/cairo-gstate.c | 101 ++++++++++++++++++++++++++++++++++++++++++++++++----- src/cairoint.h | 9 ++--- 2 files changed, 98 insertions(+), 12 deletions(-) commit e2c31183e96f84e7d40a9e5e2b6b8802d47628d0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Aug 11 23:45:30 2009 +0100 [script] Handle zero-length strings Not sure where the zero length string is coming from, but we should nevertheless handle it. util/cairo-script/cairo-script-scanner.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit 4c215162d24453788b3a461aa47ccf2e638db35e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Aug 11 23:40:54 2009 +0100 [script] Off-by-one in allocation of string length. We need pass in the real number of bytes in the string, excluding the NUL terminator which is already accounted for. util/cairo-script/cairo-script-scanner.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 55eddb7ef4b6b561d6692a6f074834d5d5fb94fa Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Aug 11 23:40:04 2009 +0100 [script] A little utility to convert a trace into a trace Replay the trace using the interpreter onto a script surface - useful for checking idempotency. util/cairo-script/Makefile.am | 6 ++++++ util/cairo-script/csi-trace.c | 40 ++++++++++++++++++++++++++++++++++++++++ 2 files changed, 46 insertions(+) commit 4f129863df6392f3deaf6e76fd15adeba98e41b8 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Aug 11 23:38:42 2009 +0100 [script] Store the current stroke matrix We can skip re-emitting stroke parameters if the values are unchanged and the scaling matrix is unaltered. src/cairo-script-surface.c | 46 +++++++++++++++++++++++++++++++++++++--------- 1 file changed, 37 insertions(+), 9 deletions(-) commit 858211f3944507362b2a18d56a65e9a478ccd305 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Aug 11 22:43:31 2009 +0100 [script] Suppress resetting stroke-style elements after matrix switch If the user is just using the default values, there is no point re-emitting them. src/cairo-script-surface.c | 42 ++++++++++++++++++++++++++---------------- 1 file changed, 26 insertions(+), 16 deletions(-) commit b6db3053dcde99e26471fdeaedcadd4a6f93b5ef Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Aug 11 22:30:58 2009 +0100 [script] Hide the implicit CLEAR for similar surfaces Do emit the clear that is performed by the surface layer on similar surfaces. src/cairo-script-surface.c | 43 +++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 43 insertions(+) commit 005b195f062d896d76a87d608ce313b169f99201 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Aug 11 22:07:12 2009 +0100 [pattern] Ignore matrix/filter/extend when comparing solids Solid patterns do not use their matrices, filter or extend properties so ignore them for the purposes of comparing and hashing. src/cairo-pattern.c | 25 ++++++++++++++++--------- 1 file changed, 16 insertions(+), 9 deletions(-) commit bb919584c0054bb3e3c547f65c91cfe48302ac86 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Aug 11 21:40:37 2009 +0100 [script] Use a compact representation for horizontal offsets between glyphs Kerning is quite frequent, that is to apply a horizontal but no vertical offset to a glyph. For instance by discarding the vertical coordinate where it remains the same and only encoding the horizontal offset we reduce the file size by ~12.5% when tracing poppler. src/cairo-script-surface.c | 56 +++++++++++++++++++++--------- util/cairo-script/cairo-script-operators.c | 22 ++++++++---- 2 files changed, 55 insertions(+), 23 deletions(-) commit cbee97f0e3e784b4482429790fc3f42c81908557 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Aug 11 21:26:11 2009 +0100 [script] Preserve '{}' whilst translating util/cairo-script/cairo-script-private.h | 1 + util/cairo-script/cairo-script-scanner.c | 67 +++++++++++++++++--------------- 2 files changed, 37 insertions(+), 31 deletions(-) commit 24b23200025166e8343425f77a53bedd08790367 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Aug 11 21:12:59 2009 +0100 [script] Fix list handling during font destruction Use cairo_list to unhook the font correctly during the fini callback. src/cairo-script-surface.c | 33 +++++++++++++-------------------- 1 file changed, 13 insertions(+), 20 deletions(-) commit 4032438625819cfa5d1928d653f404364529a2e1 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Aug 11 17:02:41 2009 +0100 [path] Eliminate redundant line-to before a close As the close implicitly issues a line-to to the initial point, remove an identical line-to if present. src/cairo-path-fixed.c | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+) commit 111f2be71b1e51fac551fd9214d13899a8ec7909 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Aug 11 17:01:50 2009 +0100 [path] Discard redundant line-to Eliminate repeated line-to to the current point. src/cairo-path-fixed.c | 49 ++++++++++++++++++++++++++++--------------------- 1 file changed, 28 insertions(+), 21 deletions(-) commit a2d5f59e2158651ac85dcc8a2b8f49cd2861044e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Aug 11 17:01:07 2009 +0100 [debug] Path printer src/cairo-debug.c | 67 +++++++++++++++++++++++++++++++++++++++++++++++++++++++ src/cairoint.h | 3 +++ 2 files changed, 70 insertions(+) commit 4bf96bad9697cbe67907df69d40f46d8d7f24325 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Aug 11 15:37:35 2009 +0100 [fill] Use trivial rectilinear_to_traps Avoid a small amount of unnecessary overhead by performing a simple conversion of the path to traps when it consists solely of simple boxes. src/cairo-path-fill.c | 74 ++++++++++++++++++++++++++++++++++++++++++++ src/cairo-surface-fallback.c | 14 +++++++++ src/cairoint.h | 5 +++ 3 files changed, 93 insertions(+) commit 30e5fa0ce06b2572f09f3d47ee10b692f18ba0ae Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Aug 11 15:36:11 2009 +0100 [polygon] Return status from path ops This tidies the common case which was to call, for example, _cairo_polygon_line_to(); _cairo_polygon_status(); src/cairo-path-fill.c | 26 ++++---------------------- src/cairo-polygon.c | 18 ++++++++++++------ src/cairoint.h | 6 +++--- 3 files changed, 19 insertions(+), 31 deletions(-) commit 3fcac1ef21de9526bc1abca902db5755abe463ae Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Aug 11 13:43:32 2009 +0100 [slope] Inline _cairo_slope_init() Move the definition to a separate header file and allow callers to inline the simple function. src/Makefile.sources | 1 + src/cairo-hull.c | 2 ++ src/cairo-path-fixed.c | 1 + src/cairo-path-stroke.c | 2 ++ src/cairo-pen.c | 2 ++ src/cairo-polygon.c | 2 ++ src/cairo-slope-private.h | 65 +++++++++++++++++++++++++++++++++++++++++++++++ src/cairo-slope.c | 9 +------ src/cairo-spline.c | 2 ++ src/cairo-traps.c | 1 + src/cairoint.h | 17 ------------- 11 files changed, 79 insertions(+), 25 deletions(-) commit a1e0c4b30980d624bb3e015b7dcf39b4a2ef8c56 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Aug 11 13:34:29 2009 +0100 [clip] Combine directly onto target Where it is unlikely that we will reuse the temporary clip surface, combine the clip directly with the mask. src/cairo-clip-private.h | 3 ++ src/cairo-clip.c | 118 ++++++++++++++++++++++++++++++++++++++++--- src/cairo-surface-fallback.c | 26 +--------- 3 files changed, 115 insertions(+), 32 deletions(-) commit 3f12d9ec5db1ac372742c3c03408bdaeaffdc1e4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Aug 28 10:05:52 2009 +0100 [clip] Use geometric clipping for unaligned clips For the simple cases where the clip is an unaligned box (or boxes), apply the clip directly to the geometry and avoid having to use an intermediate clip-mask. src/cairo-bentley-ottmann.c | 285 +++---------------------------------- src/cairo-clip-private.h | 5 + src/cairo-clip.c | 296 ++++++++++++++++++++++++++++++++++++-- src/cairo-gstate.c | 15 +- src/cairo-path-fill.c | 8 +- src/cairo-path-stroke.c | 22 +-- src/cairo-polygon.c | 329 ++++++++++++++++++++++++------------------- src/cairo-rectangle.c | 23 +++ src/cairo-surface-fallback.c | 267 +++++++++++++++++++---------------- src/cairo-traps.c | 210 +++++++++++++++------------ src/cairo-types-private.h | 4 +- src/cairoint.h | 28 +++- 12 files changed, 830 insertions(+), 662 deletions(-) commit 2457c4bedef0447f7bff9b54dba96126010917ac Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Aug 10 14:48:03 2009 +0100 traps-as-spans Add an interface to spans that accepts trapezoids. This allows backends that have an efficient span-line interface but lack efficient handling of boxes (partly due to the current poor compositor interface) to redirect composite_trapezoids() to composite_polygon() and the span-renderer. src/cairo-spans-private.h | 19 ++++++++++ src/cairo-spans.c | 82 ++++++++++++++++++++++++++++++++++++++++++ src/cairo-tor-scan-converter.c | 27 +++++++++++++- 3 files changed, 127 insertions(+), 1 deletion(-) commit 3023330706b1237b0fbd697d015cad9a23c250b7 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Aug 10 10:57:41 2009 +0100 [fill] Early check for empty path/polygon src/cairo-path-fill.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) commit 9ba37a85d2f6c033d68eb547be6c63382164519e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Aug 10 10:57:01 2009 +0100 [gstate] Discard trivial all-clipped regions Avoid assertion failures later that we have a valid region. src/cairo-gstate.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) commit 85094c4eee4e50ec724bf1bb54ecff6f7c1014bf Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Aug 24 07:06:32 2009 +0100 [clip] Eliminate redundant clips First perform a simple geometric clip to catch the majority of cases where an unaligned clip has been set outside the operation extents that can be discarded without having to use an image surface. This causes a dramatic increase of over 13x for the poppler-bug-12266 trace and little impact elsewhere for more sensible clippers. src/cairo-clip-private.h | 8 +- src/cairo-clip.c | 187 ++++++++++++++++++++++++++++++++++--------- src/cairo-gstate.c | 106 +++++++++++++++--------- src/cairo-path-fill.c | 3 + src/cairo-surface-fallback.c | 111 ++++++++++++++++++------- 5 files changed, 309 insertions(+), 106 deletions(-) commit ff0ca6d02a2e8901e9cfca31326c3fdc16e77e2f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Aug 5 17:54:48 2009 +0100 [test] Stress the intersection handling by feeding spans/traps random curves Another stress test for the fill/stroke and intersection finders. test/Makefile.am | 10 +++ test/Makefile.sources | 2 + test/random-intersections-curves-eo.c | 81 ++++++++++++++++++++ test/random-intersections-curves-eo.pdf.ref.png | Bin 0 -> 246507 bytes test/random-intersections-curves-eo.ps.ref.png | Bin 0 -> 151676 bytes test/random-intersections-curves-eo.ref.png | Bin 0 -> 246996 bytes ...m-intersections-curves-eo.xlib-fallback.ref.png | Bin 0 -> 246243 bytes test/random-intersections-curves-eo.xlib.ref.png | Bin 0 -> 246243 bytes test/random-intersections-curves-nz.c | 82 +++++++++++++++++++++ test/random-intersections-curves-nz.pdf.ref.png | Bin 0 -> 266977 bytes test/random-intersections-curves-nz.ps.ref.png | Bin 0 -> 134732 bytes test/random-intersections-curves-nz.ref.png | Bin 0 -> 267589 bytes ...m-intersections-curves-nz.xlib-fallback.ref.png | Bin 0 -> 266364 bytes test/random-intersections-curves-nz.xlib.ref.png | Bin 0 -> 266360 bytes 14 files changed, 175 insertions(+) commit 6dfe050d6360409161418d853df3d7d7dfb216af Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Aug 4 15:32:25 2009 +0100 [polygon] Amalgamate collinear edges Combine sequential collinear edges into a single edge, this benefits immensely by feeding fewer edges into either the tessellator or spans. src/cairo-polygon.c | 58 +++++++++++++++++++++++++++++++++++++++++------ src/cairo-types-private.h | 3 +++ 2 files changed, 54 insertions(+), 7 deletions(-) commit dc886450ac7c04252cff77729e0653c6ea2768db Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Aug 4 13:14:38 2009 +0100 [util] Show total edge length in show-edges util/show-edges.c | 54 +++++++++++++++++++++++++++++++++++++++++++----------- 1 file changed, 43 insertions(+), 11 deletions(-) commit 0f8af054841c2d6dfe4bfeb3d13a7bab0cfbe2f3 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Aug 3 08:27:01 2009 +0100 [fallback] Avoid tessellating empty polygons I added an assert inside the tessellator to ensure that empty polygon were not being propagated that far... src/cairo-surface-fallback.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) commit 09377a716334df3683912747067cd396768cfab6 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Aug 2 13:09:30 2009 +0100 [freelist] Lazy initialisation of pools src/cairo-freelist-private.h | 31 ++++++++++++++++-- src/cairo-freelist.c | 75 +++++++++++++++++--------------------------- 2 files changed, 56 insertions(+), 50 deletions(-) commit 6f0340e2e5079eba597c0a3a7d39da21cf2b5e7a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jul 29 21:47:11 2009 +0100 [clip] Use the rectilinear tessellator We can ensure that we always produce a clip region when possible by using the rectilinear tessellator to convert complex, device-aligned polygons to regions. Prior to using the tessellator, we relied on pixman's region code which could only handle a union of rectangles. src/cairo-clip.c | 9 ++---- src/cairo-path-fill.c | 79 +++++++++++++++++++++++++++++++++++++++++---------- 2 files changed, 66 insertions(+), 22 deletions(-) commit e3820bef20fdd77ab804b9832f47dc286e4887c4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Aug 2 15:17:51 2009 +0100 [fill] Short-circuit extents on an empty path. If the path is empty, avoid redundant polygonisation and tessellation by simply returning the empty extents. src/cairo-gstate.c | 12 ++++++++++++ 1 file changed, 12 insertions(+) commit 4051ed328b618e28cf1df276899eefa225225c76 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Aug 25 20:51:06 2009 +0100 [tessellator] Special case rectilinear tessellation For the frequent cases where we know in advance that we are dealing with a rectilinear path, but can not use the simple region code, implement a variant of the Bentley-Ottmann tessellator. The advantages here are that edge comparison is very simple (we only have vertical edges) and there are no intersection, though possible overlaps. The idea is the same, maintain a y-x sorted queue of start/stop events that demarcate traps and sweep through the active edges at each event, looking for completed traps. The motivation for this was noticing a performance regression in box-fill-outline with the self-intersection work: 1.9.2 to HEAD^: 3.66x slowdown HEAD^ to HEAD: 5.38x speedup 1.9.2 to HEAD: 1.57x speedup The cause of which was choosing to use spans instead of the region handling code, as the complex polygon was no longer being tessellated. src/Makefile.sources | 1 + src/cairo-bentley-ottmann-rectilinear.c | 582 +++++++++++++++++++++++++++++ src/cairo-bentley-ottmann.c | 5 +- src/cairo-combsort-private.h | 4 +- src/cairo-path-fill.c | 100 +---- src/cairo-path-stroke.c | 1 + src/cairo-surface-fallback.c | 34 +- src/cairo-traps.c | 2 + src/cairoint.h | 18 +- test/ft-text-vertical-layout-type1.ref.png | Bin 3647 -> 3644 bytes test/ft-text-vertical-layout-type3.ref.png | Bin 3607 -> 3608 bytes 11 files changed, 631 insertions(+), 116 deletions(-) commit 82ccb4c70cbf28167c280e590017b221a406b5c3 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jul 29 18:47:29 2009 +0100 [clip] Use special-purpose fill_to_region() Avoid the creation of temporary traps when generating a region, by calling the to_region() directly. src/cairo-clip.c | 26 +++++++++----------------- 1 file changed, 9 insertions(+), 17 deletions(-) commit 41adeac9880d011278f083c620a42b849471a92f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jul 25 21:25:07 2009 +0100 [fallback] Avoid going through traps for trivial regions. src/cairo-path-fill.c | 129 ++++++++++++++++++++++++++++++ src/cairo-surface-fallback.c | 184 ++++++++++++++++++++++++++++++++----------- src/cairoint.h | 5 ++ 3 files changed, 272 insertions(+), 46 deletions(-) commit 55bd590561880136c54da0db1f7f095a426d96a9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jul 24 20:45:56 2009 +0100 [tessellator] Use a priority queue for the events The skip list was suffering from severe overhead, so though the search was quick, the extra copies during insertion and deletion were slow. src/Makefile.am | 5 +- src/Makefile.sources | 2 - src/cairo-bentley-ottmann.c | 341 +++++++++++++++++++++--------------- src/cairo-freelist.c | 8 +- src/cairo-skiplist-private.h | 118 ------------- src/cairo-skiplist.c | 399 ------------------------------------------- 6 files changed, 212 insertions(+), 661 deletions(-) commit ebfcc2ce8fb6fcaf28d1c59cf7a5b13168cbeb70 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jul 24 16:50:33 2009 +0100 [tessellator] Remove the skiplist for the active edges The active edge list is typically short, and the skiplist adds significant overhead that far outweigh the benefit of the O(n lg n) sort. Instead we track the position of the last insertion edge, knowing that the start events are lexicographically sorted, and begin a linear search from there. src/cairo-bentley-ottmann.c | 230 +++++++++++++++----------------------------- 1 file changed, 75 insertions(+), 155 deletions(-) commit 36480fe531f19d9c692ee1f8cf09accd4b2c0ad8 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jul 25 07:38:04 2009 +0100 [traps] Increase exponential expansion factor. Grow the traps more rapidly, as the allocations are very short-lived so the over-allocation is less of an issue. src/cairo-traps.c | 8 +++----- src/cairoint.h | 8 +++----- 2 files changed, 6 insertions(+), 10 deletions(-) commit 9d51c03bad5f10257e248f43375062902482c0c4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jul 24 13:51:23 2009 +0100 [traps] Compute extents on demand. src/cairo-region-private.h | 5 - src/cairo-region.c | 47 ++--- src/cairo-surface-fallback.c | 10 - src/cairo-traps.c | 440 +++++++++++++------------------------------ src/cairoint.h | 9 - 5 files changed, 144 insertions(+), 367 deletions(-) commit f8bb3617c3a7ec598c42eff1f8562e3ccc95127f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Dec 17 09:32:16 2008 +0000 Eliminate self-intersecting strokes. We refactor the surface fallbacks to convert full strokes and fills to the intermediate polygon representation (as opposed to before where we returned the trapezoidal representation). This allow greater flexibility to choose how then to rasterize the polygon. Where possible we use the local spans rasteriser for its increased performance, but still have the option to use the tessellator instead (for example, with the current Render protocol which does not yet have a polygon image). In order to accommodate this, the spans interface is tweaked to accept whole polygons instead of a path and the tessellator is tweaked for speed. Performance Impact ================== ... Still measuring, expecting some severe regressions. ... NEWS | 18 + src/Makefile.am | 2 + src/cairo-bentley-ottmann.c | 1833 ++++++++++++-------- src/cairo-clip.c | 14 +- src/cairo-fixed-private.h | 52 +- src/cairo-fixed-type-private.h | 5 + src/cairo-gstate.c | 4 +- src/cairo-hash.c | 6 - src/cairo-matrix.c | 27 +- src/cairo-path-fill.c | 109 +- src/cairo-path-stroke.c | 1025 ++++++----- src/cairo-pen.c | 191 -- src/cairo-polygon.c | 269 ++- src/cairo-skiplist-private.h | 4 +- src/cairo-skiplist.c | 8 +- src/cairo-slope.c | 4 +- src/cairo-spans-private.h | 45 +- src/cairo-spans.c | 179 +- src/cairo-surface-fallback.c | 208 ++- src/cairo-tor-scan-converter.c | 492 +++--- src/cairo-toy-font-face.c | 3 +- src/cairo-traps.c | 72 +- src/cairo-types-private.h | 15 +- src/cairo-wideint-private.h | 47 +- src/cairo-wideint.c | 34 - src/cairoint.h | 102 +- test/Makefile.am | 90 +- test/Makefile.sources | 3 +- test/caps-joins-alpha.ref.png | Bin 2463 -> 2570 bytes test/caps-joins-alpha.xlib.ref.png | Bin 0 -> 2278 bytes test/caps-joins-curve.ref.png | Bin 5132 -> 5725 bytes test/caps-joins-curve.xlib.ref.png | Bin 0 -> 5126 bytes test/clear.pdf.argb32.ref.png | Bin 0 -> 790 bytes test/clear.ps.argb32.ref.png | Bin 0 -> 790 bytes test/clip-operator.svg12.argb32.xfail.png | Bin 8406 -> 8404 bytes test/clip-operator.svg12.rgb24.xfail.png | Bin 4540 -> 4523 bytes test/clipped-group.pdf.ref.png | Bin 310 -> 334 bytes test/clipped-trapezoids-ref.png | Bin 0 -> 1055 bytes test/clipped-trapezoids.c | 95 + test/clipped-trapezoids.ref.png | Bin 0 -> 963 bytes test/close-path-current-point.ref.png | Bin 1916 -> 1893 bytes test/dash-caps-joins.ref.png | Bin 4718 -> 4520 bytes test/dash-curve.ref.png | Bin 39696 -> 41284 bytes test/dash-curve.xlib.ref.png | Bin 0 -> 40109 bytes test/dash-scale.ref.png | Bin 8058 -> 7616 bytes test/degenerate-arc.ref.png | Bin 544 -> 646 bytes test/degenerate-dash.ref.png | Bin 2005 -> 2086 bytes test/degenerate-dash.xlib.ref.png | Bin 0 -> 1898 bytes test/degenerate-path.argb32.ref.png | Bin 0 -> 240 bytes test/degenerate-path.ref.png | Bin 257 -> 0 bytes test/degenerate-path.rgb24.ref.png | Bin 224 -> 210 bytes test/degenerate-pen.ref.png | Bin 1139 -> 997 bytes test/degenerate-pen.xlib.ref.png | Bin 0 -> 997 bytes test/device-offset-fractional.pdf.xfail.png | Bin 275 -> 275 bytes test/fill-and-stroke-alpha-add.ref.png | Bin 591 -> 562 bytes test/fill-and-stroke-alpha-add.svg12.xfail.png | Bin 614 -> 631 bytes test/fill-and-stroke-alpha.ref.png | Bin 519 -> 515 bytes test/fill-and-stroke.argb32.ref.png | Bin 0 -> 313 bytes test/fill-and-stroke.ref.png | Bin 279 -> 0 bytes test/fill-and-stroke.rgb24.ref.png | Bin 257 -> 287 bytes test/fill-and-stroke.xlib.argb32.ref.png | Bin 0 -> 322 bytes test/fill-and-stroke.xlib.rgb24.ref.png | Bin 0 -> 291 bytes test/filter-nearest-offset.pdf.xfail.png | Bin 1867 -> 1926 bytes test/filter-nearest-transformed.pdf.xfail.png | Bin 482 -> 488 bytes test/ft-text-vertical-layout-type1.ps.ref.png | Bin 0 -> 1634 bytes test/ft-text-vertical-layout-type1.ps2.ref.png | Bin 1846 -> 0 bytes test/ft-text-vertical-layout-type1.ps3.ref.png | Bin 1846 -> 0 bytes test/ft-text-vertical-layout-type1.ref.png | Bin 3643 -> 3647 bytes test/ft-text-vertical-layout-type1.svg.ref.png | Bin 3607 -> 3609 bytes test/ft-text-vertical-layout-type1.xlib.ref.png | Bin 3980 -> 3640 bytes test/ft-text-vertical-layout-type3.ps.ref.png | Bin 0 -> 1636 bytes test/ft-text-vertical-layout-type3.ps2.ref.png | Bin 1804 -> 0 bytes test/ft-text-vertical-layout-type3.ps3.ref.png | Bin 1804 -> 0 bytes test/ft-text-vertical-layout-type3.ref.png | Bin 3608 -> 3607 bytes test/ft-text-vertical-layout-type3.svg.ref.png | Bin 3626 -> 3626 bytes test/ft-text-vertical-layout-type3.xlib.ref.png | Bin 3934 -> 3605 bytes test/group-unaligned.svg.argb32.xfail.png | Bin 520 -> 522 bytes test/joins.ref.png | Bin 5879 -> 5708 bytes test/leaky-dashed-rectangle.pdf.ref.png | Bin 347 -> 347 bytes test/leaky-dashed-rectangle.xlib.ref.png | Bin 0 -> 347 bytes test/leaky-dashed-stroke.ref.png | Bin 9214 -> 9387 bytes test/leaky-dashed-stroke.xlib.ref.png | Bin 0 -> 9305 bytes test/line-width-scale.ref.png | Bin 6445 -> 5716 bytes test/long-dashed-lines.ref.png | Bin 2116 -> 2070 bytes test/mask-glyphs.svg.ref.png | Bin 1211147 -> 1211158 bytes test/meta-surface-pattern.pdf.argb32.ref.png | Bin 3919 -> 3835 bytes test/meta-surface-pattern.pdf.rgb24.ref.png | Bin 3867 -> 3682 bytes test/meta-surface-pattern.svg.argb32.ref.png | Bin 3930 -> 3778 bytes test/meta-surface-pattern.svg.rgb24.ref.png | Bin 3910 -> 3760 bytes test/new-sub-path.argb32.ref.png | Bin 0 -> 421 bytes test/new-sub-path.pdf.argb32.ref.png | Bin 0 -> 512 bytes test/new-sub-path.ref.png | Bin 386 -> 0 bytes test/new-sub-path.rgb24.ref.png | Bin 355 -> 381 bytes test/operator-source.svg12.argb32.xfail.png | Bin 2967 -> 2951 bytes test/operator-source.svg12.rgb24.xfail.png | Bin 2976 -> 2963 bytes test/over-around-source.pdf.argb32.ref.png | Bin 585 -> 576 bytes ...m-intersections.c => random-intersections-eo.c} | 2 +- test/random-intersections-eo.ps.ref.png | Bin 0 -> 78806 bytes ....png => random-intersections-eo.quartz.ref.png} | Bin test/random-intersections-eo.ref.png | Bin 0 -> 136018 bytes test/random-intersections-eo.xlib.ref.png | Bin 0 -> 134655 bytes test/random-intersections-nonzero.c | 79 + test/random-intersections-nonzero.ps.ref.png | Bin 0 -> 77299 bytes test/random-intersections-nonzero.ref.png | Bin 0 -> 141154 bytes test/random-intersections-nonzero.xlib.ref.png | Bin 0 -> 141548 bytes test/random-intersections.ps2.ref.png | Bin 89253 -> 0 bytes test/random-intersections.ps3.ref.png | Bin 89253 -> 0 bytes test/random-intersections.ref.png | Bin 133462 -> 0 bytes test/random-intersections.xlib.ref.png | Bin 148722 -> 0 bytes test/reflected-stroke.ref.png | Bin 5704 -> 4978 bytes test/rel-path.ref.png | Bin 212 -> 198 bytes test/rel-path.rgb24.ref.png | Bin 228 -> 214 bytes test/scale-offset-image.meta.xfail.png | Bin 8599 -> 0 bytes test/scale-offset-image.pdf.xfail.png | Bin 8694 -> 10099 bytes test/scale-offset-image.xfail.png | Bin 8599 -> 9960 bytes test/scale-offset-image.xlib-fallback.xfail.png | Bin 8580 -> 8706 bytes test/scale-offset-image.xlib.xfail.png | Bin 9100 -> 9132 bytes test/scale-offset-similar.meta.xfail.png | Bin 9095 -> 10752 bytes test/scale-offset-similar.pdf.xfail.png | Bin 9368 -> 11024 bytes test/scale-offset-similar.xfail.png | Bin 8599 -> 9960 bytes test/scale-offset-similar.xlib-fallback.xfail.png | Bin 8580 -> 8706 bytes test/scale-offset-similar.xlib.xfail.png | Bin 8570 -> 8696 bytes test/self-intersecting.argb32.xfail.png | Bin 235 -> 0 bytes test/self-intersecting.c | 3 + test/self-intersecting.pdf.argb32.xfail.png | Bin 285 -> 0 bytes test/self-intersecting.pdf.rgb24.xfail.png | Bin 269 -> 0 bytes ...gb32.xfail.png => self-intersecting.ps.ref.png} | Bin test/self-intersecting.ps.rgb24.xfail.png | Bin 186 -> 0 bytes test/self-intersecting.ref.png | Bin 242 -> 301 bytes test/self-intersecting.rgb24.ref.png | Bin 304 -> 0 bytes test/self-intersecting.rgb24.xfail.png | Bin 240 -> 0 bytes test/self-intersecting.xlib.argb32.xfail.png | Bin 235 -> 0 bytes test/self-intersecting.xlib.ref.png | Bin 0 -> 252 bytes test/self-intersecting.xlib.rgb24.xfail.png | Bin 240 -> 0 bytes test/smask-stroke.ref.png | Bin 1845 -> 1714 bytes test/smask-stroke.xlib.ref.png | Bin 0 -> 1707 bytes test/smask.ps.ref.png | Bin 0 -> 3430 bytes test/smask.ps2.ref.png | Bin 3669 -> 0 bytes test/smask.ps3.ref.png | Bin 3669 -> 0 bytes test/smask.ref.png | Bin 3417 -> 3423 bytes test/smask.svg.ref.png | Bin 3458 -> 3457 bytes test/smask.xlib.ref.png | Bin 3393 -> 3386 bytes test/spline-decomposition.pdf.ref.png | Bin 19156 -> 19540 bytes test/spline-decomposition.ps.ref.png | Bin 9090 -> 9028 bytes test/spline-decomposition.ref.png | Bin 19107 -> 19569 bytes test/spline-decomposition.svg.ref.png | Bin 19156 -> 19540 bytes test/spline-decomposition.xlib.ref.png | Bin 0 -> 19158 bytes test/stroke-ctm-caps.ref.png | Bin 1057 -> 873 bytes test/stroke-image.pdf.ref.png | Bin 1554 -> 1475 bytes test/stroke-image.ps.ref.png | Bin 0 -> 1641 bytes test/stroke-image.ps2.ref.png | Bin 2121 -> 0 bytes test/stroke-image.ps3.ref.png | Bin 2121 -> 0 bytes test/stroke-image.ref.png | Bin 1866 -> 1344 bytes test/stroke-image.xlib.ref.png | Bin 0 -> 1440 bytes test/surface-pattern.pdf.xfail.png | Bin 10528 -> 10532 bytes test/text-rotate.pdf.ref.png | Bin 16639 -> 16757 bytes test/text-rotate.ref.png | Bin 17775 -> 16738 bytes test/text-rotate.svg.ref.png | Bin 17024 -> 17118 bytes test/text-rotate.xlib.ref.png | Bin 0 -> 16646 bytes test/twin.ps.ref.png | Bin 2197 -> 2142 bytes test/twin.ref.png | Bin 4038 -> 3979 bytes test/twin.xlib.ref.png | Bin 0 -> 4435 bytes test/unantialiased-shapes.ref.png | Bin 4459 -> 3969 bytes test/unbounded-operator.svg12.argb32.ref.png | Bin 0 -> 2767 bytes test/unbounded-operator.svg12.argb32.xfail.png | Bin 2771 -> 0 bytes test/unbounded-operator.svg12.rgb24.xfail.png | Bin 1731 -> 1719 bytes test/user-font-proxy.svg.ref.png | Bin 16814 -> 16817 bytes test/user-font.ps.ref.png | Bin 0 -> 4605 bytes test/user-font.ps2.ref.png | Bin 5043 -> 0 bytes test/user-font.ps3.ref.png | Bin 5043 -> 0 bytes test/user-font.ref.png | Bin 6082 -> 6270 bytes test/user-font.xlib.ref.png | Bin 6183 -> 5884 bytes util/.gitignore | 3 + util/Makefile.am | 23 +- util/show-edges.c | 1189 +++++++++++++ util/show-events.c | 845 +++++++++ util/show-traps.c | 1239 +++++++++++++ 177 files changed, 6226 insertions(+), 2113 deletions(-) commit 7c499db8afe8a7cf8c512ec166fe7dbf11a25c02 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Aug 25 20:37:55 2009 +0100 [test] Minor tweak of ft-text-vertical-layout Tidily destroy the font immediately after use. test/ft-text-vertical-layout-type1.c | 6 ++---- test/ft-text-vertical-layout-type3.c | 5 ++--- 2 files changed, 4 insertions(+), 7 deletions(-) commit bdd3c5ba6987280b455229dd12b20c22159ce61c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Aug 22 18:59:01 2009 +0100 [perf] Match directory names In order to handle 'cairo-perf-trace benchmark', we need to perform the can_run? test on the directory name as opposed to the individual trace names. Make it so. perf/box-outline.c | 2 +- perf/cairo-perf-trace.c | 37 ++++++++++++++++++++++++++++--------- perf/cairo-perf.c | 14 +++++++++++--- perf/cairo-perf.h | 3 ++- perf/composite-checker.c | 2 +- perf/dragon.c | 3 ++- perf/fill.c | 2 +- perf/glyphs.c | 2 +- perf/intersections.c | 2 +- perf/long-dashed-lines.c | 2 +- perf/long-lines.c | 2 +- perf/mask.c | 2 +- perf/mosaic.c | 2 +- perf/paint-with-alpha.c | 2 +- perf/paint.c | 2 +- perf/pattern_create_radial.c | 2 +- perf/pythagoras-tree.c | 4 ++-- perf/rectangles.c | 2 +- perf/rounded-rectangles.c | 2 +- perf/spiral.c | 2 +- perf/stroke.c | 2 +- perf/subimage_copy.c | 2 +- perf/tessellate.c | 2 +- perf/text.c | 2 +- perf/twin.c | 2 +- perf/unaligned-clip.c | 2 +- perf/world-map.c | 2 +- perf/zrusin.c | 2 +- 28 files changed, 68 insertions(+), 39 deletions(-) commit 77c11096169bc8af6aa08241a800a51a2292a27a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Aug 22 12:23:03 2009 +0100 [perf] Reset global caches after every trace I'd disabled this to look at cairo-qt performance, then forgot about it. Be clean, cleanup globals -- this should fix the huge performance loss when running in series multiple backends that need separate font caches. perf/cairo-perf-trace.c | 5 +++++ 1 file changed, 5 insertions(+) commit 9c8e60f7718b29051268541d2cfec58e51d8e7b8 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Aug 22 12:09:25 2009 +0100 [perf] Remove cpuset warning for cairo-perf-trace These traces run for much longer than the original synthetic benchmarks and seek to replicate 'real-world' applications, so the warning that the xserver and cairo-perf are not bound to any cpu is false. perf/cairo-perf-trace.c | 47 ----------------------------------------------- 1 file changed, 47 deletions(-) commit d07ed5d990634dc09d0d1b6ce45aa8ad1994b75a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Aug 19 12:36:56 2009 +0100 [perf] Add charting utility cairo-perf-chart takes multiple runs (currently it is limited to prefiltered data sets) and pretty-prints a chart showing performace improvements/regressions (in either ASCII or HTML) along with a cairo-perf-chart.png perf/Makefile.am | 4 + perf/cairo-perf-chart.c | 751 ++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 755 insertions(+) commit 015eccadcb8b22f302bdaa315b88981a38294686 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Aug 16 09:57:07 2009 +0100 [perf] Fix unaligned-clip We failed to reset the scale after each loop, eventually generating a vast clip-mask that exceeded memory capacity. perf/unaligned-clip.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) commit 6f8fae21e6fee07a8f6561f77bcb0bfaf64f5882 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Aug 15 10:16:41 2009 +0100 [perf] Add a box mode to spiral Add a very similar 'spiral' path (really just concentric boxes) that hit the rectangular optimisations so that we can compare how effective they are. perf/spiral.c | 73 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 73 insertions(+) commit e79a9a87c311fd0499c96cecbe560839a17d244e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Aug 15 10:33:48 2009 +0100 [perf] Support objdir!=srcdir in cairo-perf-diff Crude support to detect when we have cairo-perf-diff-files in the local directory which is not srcdir. perf/cairo-perf-diff | 26 ++++++++++++++++---------- 1 file changed, 16 insertions(+), 10 deletions(-) commit 6bfb77a33c3614cafdcd6206de5c71321c5fc1e5 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Aug 14 19:53:17 2009 +0100 [perf] Add spiral strokes Use cairo_stroke() to perform the equivalent of spiral-rect-(pix|non)align-evenodd-fill. A useful comparison of stroking versus filling, as we can assume the composition costs are similar. perf/spiral.c | 73 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 73 insertions(+) commit 54b2935c3842e77888faef06bbf247f99bd9de8c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Aug 13 23:23:41 2009 +0100 [perf] Reset the path after each spiral Oops we were accumulating paths during each spiral iteration and so the tests were getting slower and slower and slower... [And fix a couple of other instances of path accumulation.] perf/intersections.c | 2 ++ perf/spiral.c | 1 + perf/world-map.c | 2 ++ 3 files changed, 5 insertions(+) commit fddfe4973f5b49f280734ee7141ab72451cf94d2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Aug 13 15:52:06 2009 +0100 [perf] Add clipped modes to dragon Investigate the performance hit for unclipped/aligned/unaligned/masked modes with a reasonably complex geometry. perf/dragon.c | 35 +++++++++++++++++++++++++++++++++++ 1 file changed, 35 insertions(+) commit 8b3be26c7be0747378792e3a454e500d290c3982 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Aug 12 00:02:04 2009 +0100 [test] Remove misleading code/comments from a1-image-sample test/a1-image-sample.c | 4 ---- 1 file changed, 4 deletions(-) commit e702df59b05246a52a645792a65dc9c12d5b3feb Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Aug 22 17:53:59 2009 +0100 [test] Attempt to automatically detect running under gdb A common mistake is to forget to pass the foreground mode to cairo-test-suite when launching it under the debugger, resulting in the debugger not attaching to the children and missing the error you were trying to capture. Under linux, we can inspect the path to our parent's executable and if that looks like gdb, we assume it is and disable forking of traces. test/cairo-test-runner.c | 32 ++++++++++++++++++++++++++++++++ 1 file changed, 32 insertions(+) commit bf4977b645270bb88225501e73848f7814cccead Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Aug 25 07:16:16 2009 +0100 [test] Exit on first failure '-x' Add a command line option to the test suite to cause it to exit after the first failure. The purpose of this is for integration into 'git bisect run', where the failing test is unknown and we are looking for any failure. For example, for use in a regression script to find commits in the midst of as series that need a refresh of a reference image (or fixing!). test/cairo-test-runner.c | 15 +++++++++++++-- 1 file changed, 13 insertions(+), 2 deletions(-) commit 40aefac5d714bf7536632ed38c7a8ee05049f30b Author: Benjamin Otte <otte@gnome.org> Date: Wed Aug 26 21:22:07 2009 +0200 [xlib] DO_XCOPYAREA and DO_XTILE optimizations break with Window source Cairo should include the contents of subwindows when using a Window as a source but will clip to subwindows when using a Window as a destination. This can be set using the GC's subwindow_mode. XCopyArea and XFillRectangle can however only use one GC for both source and destination. Cairo's mode is set to (the default) ClipByChildren. This means that copying from a Window is broken, so we only allow the optimization when we know that the source is a Pixmap. The performance impact of this change has not been tested. It should be small, as the code will use XRender otherwise. If it turns out to be a bigger impact, the optimizations could be improved by doing a two-step copy process: 1) Copy to an intermediate Pixmap with IncludeInferiors 2) Copy to the destination with ClipByChildren (potentially omitting one one of the steps if source or destination are known to be Pixmaps). references: commit 0c5d28a4e5ce5e4dd72c0f416ce5e960e92b808b https://bugs.freedesktop.org/show_bug.cgi?id=12996 src/cairo-xlib-surface.c | 2 ++ 1 file changed, 2 insertions(+) commit 90536ef2dd534d9385d38125445faeadd0511850 Author: Behdad Esfahbod <behdad@behdad.org> Date: Tue Aug 25 00:28:08 2009 -0400 [build] Remove bogus warning that we require at least one native font backend That limitation was lifted long ago after twin face was introduced. build/configure.ac.features | 4 ---- 1 file changed, 4 deletions(-) commit 20e227fc397d064af0b4b8bbf4028430695a218d Author: Behdad Esfahbod <behdad@behdad.org> Date: Mon Aug 24 23:57:21 2009 -0400 [build] Fix typo build/aclocal.enable.m4 | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 87fb9a9ded0fb0355f91cad188393d26b9e47e0b Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk> Date: Mon Aug 17 00:59:46 2009 -0400 [test] Update reference image for the extended-blend-mode test. A recent pixman change caused one-bit differences in some pixels. test/extended-blend-alpha.argb32.ref.png | Bin 9777 -> 9406 bytes 1 file changed, 0 insertions(+), 0 deletions(-) commit d1b8e260d43c7dee71244f26d50d8b2e49e9b6e0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Aug 10 21:29:13 2009 +0100 [qt] Fix compilation Enabling 'FAST CLIP' appears to trigger an infinite loop so disable. Enabling 'FAST FILL' has limited effect on performance, so disable whilst the basic QT surface is improved. perf/cairo-perf-trace.c | 2 +- src/cairo-qt-surface.cpp | 63 ++++++++++++++++++++++++++++++++---------------- 2 files changed, 43 insertions(+), 22 deletions(-) commit dff0a91742908098a1ee110fadf11d4b802b1587 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Aug 10 17:57:16 2009 +0100 [xlib] Rewrite UNSUPPORTED() avoiding gcc-ism Behdad pointed out that fprintf() returns a value so that we could simply use the comma operator to return the correct value instead of the expression-block gcc-ism. src/cairo-xlib-surface.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) commit e63d725ad4fb92a0085224f627cbced3e3205735 Author: Behdad Esfahbod <behdad@behdad.org> Date: Sat Jun 6 01:08:22 2009 -0400 [utils] Remove unused code util/malloc-stats.c | 30 ------------------------------ 1 file changed, 30 deletions(-) commit d4d0fcb4a8055f1fdd24fc5b7b64bb55d55e1485 Author: Andrea Canciani <ranma42@gmail.com> Date: Mon Aug 10 17:43:31 2009 +0100 [quartz] Compile fix for show_glyphs Correct the prototype for _cairo_quartz_surface_show_glyphs(). src/cairo-quartz-surface.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 2a2a19457bb934f9585bfdb97730fab99c8e386c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Aug 10 17:42:05 2009 +0100 [surface] Check for a NULL snapshot The backend is permitted to return a NULL surface when snapshotting to indicate that it cannot (or choses not to) implement the method. src/cairo-surface.c | 28 +++++++++++++++------------- 1 file changed, 15 insertions(+), 13 deletions(-) commit 2e8ce34454a4d93986f590b1d2b024d1a1eb6bde Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Aug 10 16:51:07 2009 +0100 [quartz] Use the implementation font-face not the associated As the associated is now explicitly the font-face used to create the font by the user, whereas what we require is the current implementation (quartz) font. src/cairo-quartz-font.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) commit a4dc372bab8ebe75a9f7d17a9a2e6fc8d9d2d747 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Aug 10 16:50:00 2009 +0100 [scaled-font] Update API documentation Now that the toy-font-face is exposed to the user, it is expected to be returned when the user queries the font face associated with a scaled font. src/cairo-scaled-font.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) commit c4828666b9b4dad39c3f17fdfd7d1dd3cbb83cbc Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Aug 10 16:45:41 2009 +0100 [script] Handle translation of compressed strings. After introducing the new syntax to mark a deflated string, we also need to add the equivalent binary tokens. util/cairo-script/cairo-script-file.c | 10 ++-- util/cairo-script/cairo-script-scanner.c | 81 ++++++++++++++++++++------------ 2 files changed, 56 insertions(+), 35 deletions(-) commit 34870491951f5a712833b5fe7ed84c935bccfc5a Author: Andrea Canciani <ranma42@gmail.com> Date: Sun Aug 9 22:33:28 2009 +0200 [quartz] Fix one more compile error src/cairo-quartz-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit c2e75b9993ab1d30002877fc11b52ea7b9acd40b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Aug 9 21:12:36 2009 +0100 [quartz] Attempt to fix compile errors ranma42 spotted that the quartz-backend no longer compiled after bed270, so once again attempt to blindly fix those errors... src/cairo-quartz-surface.c | 18 +++++++++--------- 1 file changed, 9 insertions(+), 9 deletions(-) commit 81c4594712843abca188a33c1edfb70fa6c5a0d9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Aug 6 23:41:11 2009 +0100 [script] Ensure strings are nul terminated Infrequently, but, for example, handling glyph strings, we require the string to be nul terminated. (Otherwise an error occurs, which was previously compounded by a drastic leak.) util/cairo-script/cairo-script-scanner.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit 86d6a489527d125e739aa5a6d17893bb44977010 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Aug 6 23:39:21 2009 +0100 [script] Check for failure to store the glyph cache If we fail to add the glyph cache (presumably because the font is in error) do not leak the allocation. As this occurs for every single glyph string, the leak can grow very quickly and mask the original bug. util/cairo-script/cairo-script-operators.c | 14 ++++++++++---- 1 file changed, 10 insertions(+), 4 deletions(-) commit af392f7e45edcbf101ca7c91fc1ee372d283e8b9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Aug 6 23:17:06 2009 +0100 [script] Extend csi-replay to show how to use a single surface Currently the replay creates a fresh surface for every new surface. Whilst using it to view replays (such as with --xlib) this is often not what is desired so add a mode (compile-time only for now) to use similar surfaces and blits to the front buffer instead. util/cairo-script/csi-replay.c | 73 +++++++++++++++++++++++++++++++++++++++++- 1 file changed, 72 insertions(+), 1 deletion(-) commit b7639a5f79eaecfbaf1710fc679c7be375638efd Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Aug 6 10:55:18 2009 +0100 [trace] Close the dictionary before font font expects the dictionary to be constructed on the stack for its use, so close it. (I missed the closing '>>' when switching between dictionary constructors.) util/cairo-trace/trace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 19f54d872fcfa9801abee9c57e30c6b3fc7eef40 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Aug 6 10:40:32 2009 +0100 [gl] Move hidden cairo_gl_surface_create() proto to header Make check-plt.sh happy src/cairo-gl-private.h | 2 ++ src/cairo-gl-surface.c | 1 - 2 files changed, 2 insertions(+), 1 deletion(-) commit ee69f8aa5e0697aaeee9c00839fb7b30fe7e3184 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Aug 6 10:30:52 2009 +0100 [trace] Whitespace A few gratuitous whitespace and increase the buffer size used whilst reading in data files. util/cairo-trace/trace.c | 12 ++++-------- 1 file changed, 4 insertions(+), 8 deletions(-) commit 6d30021d65aa4126cd5eef47b0e6057dfc17d7ac Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Aug 6 10:29:49 2009 +0100 [trace] Only write the 4 byte length to the stream We were attempting to write out the whole length of the block when we were just trying to write out the length value. util/cairo-trace/trace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit a439bd99d0439c007bc74584c1eb56700c520b52 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Aug 5 17:34:29 2009 +0100 [perf] Compare drawing random curves Extend the intersection tests with straight lines, with random curves as well. perf/intersections.c | 58 +++++++++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 57 insertions(+), 1 deletion(-) commit 1327df2cf2f2dd9f98533d824503dfb081803b05 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Aug 4 20:15:13 2009 +0100 [pattern] Invalid access beyond end of freed-pool Oops, we were attempting to return a pointer from beyond the end of our array. src/cairo-pattern.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) commit 920f59a755ce142b068f21b9db07f76f35449f16 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Aug 4 20:13:43 2009 +0100 [build] Configure switch to disable atomics Workaround for my arm toolchain which succeeds in linking the configure program, only to complain when linking a program (such as cairo-perf) against libcairo.so. Annoying. build/configure.ac.system | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) commit cb30c1f367967e3395296a5d01a3eb9050635e3b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Aug 4 13:59:03 2009 +0100 [path] Extend identical lines. If a subsequent PATH_OP is just a continuation of the previous line, i.e. it has the same gradient, then just replace the end-point of the previous line with the new point rather than adding a new operation. Surprisingly this occurs in the wild, but the main motivation is a future optimisation to reduce the number of intersections during stroke-to-path. src/cairo-path-fixed.c | 31 ++++++++++++++++++++++++++++--- src/cairoint.h | 7 +++++++ 2 files changed, 35 insertions(+), 3 deletions(-) commit 85b688a3f6271427befca699de3a7a15162fc59e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Aug 3 22:27:22 2009 +0100 [matrix] Use hypot() Use hypot() instead of open-coding sqrt(x*x + y*y). In theory, the compiler could emit highly efficient code. In practice it's slower, but more likely to be more accurate -- but the difference over a bare sqrt() is likely not to be perceptible. src/cairo-matrix.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 0db9e010fa70c65451d324cc9c0ade989f77fedd Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Aug 3 22:23:19 2009 +0100 [perf] Calibrate tests to run for at least 2 seconds By ensuring that tests take longer than a couple of seconds we eliminate systematic errors in our measurements. However, we also effectively eliminate the synchronisation overhead. To compensate, we attempt to estimate the overhead by reporting the difference between a single instance and the minimum averaged instance. perf/box-outline.c | 14 +++-- perf/cairo-perf-report.c | 4 +- perf/cairo-perf.c | 39 +++++++++--- perf/cairo-perf.h | 2 +- perf/composite-checker.c | 22 ++++--- perf/dragon.c | 142 ++++++++++++++++++++++--------------------- perf/fill.c | 21 +++++-- perf/glyphs.c | 32 +++++----- perf/intersections.c | 27 ++++---- perf/long-dashed-lines.c | 9 +-- perf/long-lines.c | 68 +++++++++++---------- perf/mask.c | 45 ++++++++------ perf/mosaic.c | 38 ++++++------ perf/paint-with-alpha.c | 5 +- perf/paint.c | 5 +- perf/pattern_create_radial.c | 24 ++++---- perf/pythagoras-tree.c | 20 +++--- perf/rectangles.c | 38 ++++++------ perf/rounded-rectangles.c | 47 +++++++++++--- perf/spiral.c | 51 ++++++++-------- perf/stroke.c | 19 ++++-- perf/subimage_copy.c | 7 ++- perf/tessellate.c | 17 +++--- perf/text.c | 27 ++++---- perf/twin.c | 29 +++++---- perf/unaligned-clip.c | 37 ++++++----- perf/world-map.c | 78 ++++++++++++------------ perf/zrusin.c | 10 +-- 28 files changed, 496 insertions(+), 381 deletions(-) commit 60c574ad062f3d8526056942bb7c9e71610a2773 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Aug 5 15:11:00 2009 +0100 [surface] Protect against the user setting a fallback resolution of 0. Bug 23067 -- using clear drawing operator crashes printing http://bugs.freedesktop.org/show_bug.cgi?id=23067 Here we were hitting an assert within the paginated surface after creating a zero sized fallback image, due to the paginated surface being created with an x fallback resolution of 0 dpi (by _gtk_printer_create_cairo_surface(), gtk/gtkprinter.c:924). Avoid the bug by guarding against bad input to cairo_surface_set_fallback_resolution() which also allows us to identity the invalid caller. src/cairo-surface.c | 8 ++++++++ 1 file changed, 8 insertions(+) commit a3f393cf644d75065fe524e2fbb28f55844f43bf Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Aug 6 09:52:34 2009 +0100 [configure] Also check for libiberty.h cairo-trace also uses the libiberty.h without checking for its presence -- causing a build failure if bfd was installed but not that header. configure.ac | 1 + 1 file changed, 1 insertion(+) commit 51bd27afa147f78c8f4f3778cee725b6444e7eb0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Aug 6 09:41:15 2009 +0100 [boilerplate/test] Convert make-*-constructors to shell Remove the intermediate C program that was a nuisance whilst cross-compiling and replace it with a simple shell script that is just a combination of cat + sed. boilerplate/Makefile.am | 17 +-- boilerplate/make-cairo-boilerplate-constructors.c | 163 --------------------- boilerplate/make-cairo-boilerplate-constructors.sh | 24 +++ build/configure.ac.tools | 8 - test/Makefile.am | 11 +- test/make-cairo-test-constructors.c | 162 -------------------- test/make-cairo-test-constructors.sh | 24 +++ 7 files changed, 56 insertions(+), 353 deletions(-) commit 5dd29d7e2da8684ea46c8f1baba42e6dc64f1351 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Aug 6 09:55:43 2009 +0100 [directfb] Conditionally use DSPF_BGR555 Appears that this token is more recent than some of my headers, so check for its presence before use. src/cairo-directfb-surface.c | 4 ++++ 1 file changed, 4 insertions(+) commit 1d2abe7d1061e580e189604b25465c8025cedfb7 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jul 31 19:38:23 2009 +0100 [gl] Refresh clip-operator ref. A minor renderering difference, acceptable. test/Makefile.am | 1 + test/clip-operator.gl.argb32.ref.png | Bin 0 -> 8296 bytes 2 files changed, 1 insertion(+) commit 640a0be2e2ef0124424aeca32915a2f644b03d27 Merge: 067de9cf6 5c2239be4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jul 31 18:03:35 2009 +0100 Merge branch 'gl-glyph-cache' Comparing the glyph-cache on an i915: gnome-terminal-20090728: 1.51x speedup firefox-20090601: No change commit 5c2239be4b11437b1ab1731ef2dfac6cfb8095b1 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jul 31 16:29:49 2009 +0100 [gl] Use scissors rather than allocating a temporary depth buffer. src/cairo-gl-glyphs.c | 9 ---- src/cairo-gl-private.h | 9 ++++ src/cairo-gl-surface.c | 114 ++++++++++--------------------------------------- 3 files changed, 32 insertions(+), 100 deletions(-) commit 067de9cf62223ee7a36aab10e3d8a4e0dcda4741 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jul 31 17:49:39 2009 +0100 [test] Add clip-fill. Test filling through a complex clip. test/Makefile.am | 4 ++ test/Makefile.sources | 1 + test/clip-fill.c | 78 +++++++++++++++++++++++++++++++++++ test/clip-fill.ps.xfail.png | Bin 0 -> 522 bytes test/clip-fill.ref.png | Bin 0 -> 1035 bytes test/clip-fill.xlib-fallback.ref.png | Bin 0 -> 1039 bytes test/clip-fill.xlib.ref.png | Bin 0 -> 1049 bytes 7 files changed, 83 insertions(+) commit 7db1fb6fa97e0c44e66c0ae533b94fb3f26b273d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jul 31 16:57:14 2009 +0100 [test] Add clip-text Test applying a complex clip to text. test/Makefile.am | 4 ++ test/Makefile.sources | 1 + test/clip-text.c | 88 ++++++++++++++++++++++++++++++++++++++++++++ test/clip-text.ps.xfail.png | Bin 0 -> 434 bytes test/clip-text.ref.png | Bin 0 -> 928 bytes test/clip-text.svg.ref.png | Bin 0 -> 933 bytes test/clip-text.xlib.ref.png | Bin 0 -> 900 bytes 7 files changed, 93 insertions(+) commit 728d3f6383bba218de03f2693f289ddad6a80d0c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jul 31 16:24:37 2009 +0100 Revert "[gl] Fixup unbounded regions after spans" This reverts commit f7121ca88844fa70ecb1a35f3f6c9b653324718d. src/cairo-gl-surface.c | 68 ++++++-------------------------------------------- 1 file changed, 7 insertions(+), 61 deletions(-) commit 74bdea1249bc1fcd223e9bab83d9b9e22fa98b99 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jul 31 16:20:54 2009 +0100 [spans] Emit empty rows. A problem that does not present itself whilst using spans to intermediate masks is that the tor-span-convertor did not emit the empty rows. When compositing directly using spans with an unbounded operator this was causing rendering artefacts, for overlapping-glyphs and the gl backend. src/cairo-tor-scan-converter.c | 15 +++++++++++++++ 1 file changed, 15 insertions(+) commit 1e4b9ab04bfe18f50b9b7c2cdb75d5828e33ef69 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jul 31 15:27:20 2009 +0100 [gl] Return status from _render_glyphs() src/cairo-gl-glyphs.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) commit 097484da6fd70867ed2128e78247eab5c43217fb Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jul 31 14:48:32 2009 +0100 [boilerplate] Conditionally tweak the xlib surface Prior to introduction of the buggy members to the surface, we obviously cannot set them. However, the boilerplate code is meant to compile against older revisions of the library so we need to check for the existence prior to use. boilerplate/cairo-boilerplate-xlib.c | 10 ++++++++-- src/cairo-xlib-surface-private.h | 3 +++ 2 files changed, 11 insertions(+), 2 deletions(-) commit 397538a251ae19bb1a34650763eb9a80108a96de Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jul 31 12:17:28 2009 +0100 [boilerplate] Update fallbacks after disabling RENDER The xlib surface caches flags based on RENDER version, so after disabling render to force fallbacks, we need to ensure that the flags are reset. boilerplate/cairo-boilerplate-xlib.c | 5 +++++ 1 file changed, 5 insertions(+) commit 786d1f962a7e7ee7036ea85d4ea3aec7580e25db Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jul 31 13:17:24 2009 +0100 [xlib] Use server-side gradients. We can offload creation of gradients to server that support RENDER 0.10 and later. This greatly reduces the amount of traffic we need to send over our display connection as the gradient patterns are much smaller than the full image. Even if the server fallbacks to using pixman, performance should be improved by the reduced transport overhead. Furthermore this is a requisite to enable hardware accelerated gradients with the xlib backend. Running cairo-perf-trace on tiny, Celeron/i915: before: firefox-20090601 211.585 after: firefox-20090601 270.939 and on tiger, CoreDuo/nvidia: before: firefox-20090601 70.143 after: firefox-20090601 87.326 where linear gradients are used extensively throughout the GTK+ theme. Not quite the result I was expecting! In particular, looking at tiny: xlib-rgba paint-with-alpha_linear-rgba_over-512 47.11 (47.16 0.05%) -> 123.42 (123.72 0.13%): 2.62x slowdown █▋ xlib-rgba paint-with-alpha_linear3-rgba_over-512 47.27 (47.32 0.04%) -> 123.78 (124.04 0.13%): 2.62x slowdown █▋ xlib-rgba paint-with-alpha_linear-rgb_over-512 47.19 (47.21 0.02%) -> 123.37 (123.70 0.13%): 2.61x slowdown █▋ xlib-rgba paint-with-alpha_linear3-rgb_over-512 47.30 (47.31 0.04%) -> 123.52 (123.62 0.09%): 2.61x slowdown █▋ xlib-rgba paint_linear3-rgb_over-512 47.29 (47.32 0.05%) -> 118.95 (119.60 0.29%): 2.52x slowdown █▌ xlib-rgba paint_linear-rgba_over-512 47.14 (47.17 0.06%) -> 116.76 (117.06 0.16%): 2.48x slowdown █▌ xlib-rgba paint_linear3-rgba_over-512 47.32 (47.34 0.04%) -> 116.85 (116.98 0.05%): 2.47x slowdown █▌ xlib-rgba paint_linear-rgb_over-512 47.15 (47.19 0.03%) -> 114.08 (114.55 0.20%): 2.42x slowdown █▍ xlib-rgba paint-with-alpha_radial-rgb_over-512 117.25 (119.43 1.21%) -> 194.36 (194.73 0.09%): 1.66x slowdown ▋ xlib-rgba paint-with-alpha_radial-rgba_over-512 117.22 (117.26 0.02%) -> 193.81 (194.17 0.11%): 1.65x slowdown ▋ xlib-rgba paint_radial-rgba_over-512 117.23 (117.26 0.02%) -> 186.35 (186.41 0.03%): 1.59x slowdown ▋ xlib-rgba paint_radial-rgb_over-512 117.23 (117.27 0.02%) -> 184.14 (184.62 1.51%): 1.57x slowdown ▋ Before 1.10, we may choose to disable server-side gradients for the current crop of Xorg servers, similar to the extended repeat modes. [Updated by Chris Wilson. All bugs are his.] NEWS | 28 ++++ src/cairo-xlib-display.c | 9 +- src/cairo-xlib-private.h | 3 +- src/cairo-xlib-surface-private.h | 5 +- src/cairo-xlib-surface.c | 306 +++++++++++++++++++++++++++++++++++---- 5 files changed, 318 insertions(+), 33 deletions(-) commit 53cbbad8fbbd2b9423a5eb179d0cec3fc7678791 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jul 31 11:35:45 2009 +0100 [xlib] Disable extended repeat modes if RENDER < 0.10 The extended repeat modes were only introduced in RENDER 0.10, so disable them if the server reports an earlier version. This is in addition to disabling the repeat modes if we know (guess!) the server to have a buggy implementation. src/cairo-xlib-surface.c | 4 ++++ 1 file changed, 4 insertions(+) commit e283e53563915daa11566ba006e1bf1e12b0d2fc Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jul 31 10:50:13 2009 +0100 [perf] Print min_ticks in report As the change and ranking is based on the min_ticks, and as this can sometimes deviate wildly from median_ticks, include min_ticks in the output. In particular it helps to explain cases like: xlib-rgba rectangles_similar-rgba-mag_source-512 10.13 88.41% -> 5.77 0.19%: 1.50x slowdown which becomes xlib-rgba rectangles_similar-rgba-mag_source-512 3.83 (10.13 88.41%) -> 5.75 (5.77 0.19%): 1.50x slowdown (Considering the poor standard deviation on the initial measurement, this is more likely a sampling error than a true regression.) perf/cairo-perf-diff-files.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) commit ce8a8424aa4331f119b115123faf7b773b5f25d0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jul 31 09:36:10 2009 +0100 [perf] Change the order of slowdowns. More the large slowdowns to the end. This has two pleasing effects: 1. There is symmetry between large speedups at the top, and large slowdowns at the bottom, with long bars -> short bars -> long bars. 2. After a cairo-perf-diff run the largest slowdowns are immediately visible on the console. What better way to flag performance regressions? perf/cairo-perf-diff-files.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) commit 877ed645f95ff048f7c5d2ce0fb92389918711a6 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jul 30 22:09:55 2009 +0100 [perf] Add a more complex fill, a set of rings In investigating performance regressions with the wip/tessellator, I'm keen to inspect how the tessellator scales with polygon complexity. perf/fill.c | 37 +++++++++++++++++++++++++++++++++++++ 1 file changed, 37 insertions(+) commit 3dbc4170f0e89f02455b82c6496d6322e56e0491 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jul 30 10:35:11 2009 +0100 [perf] Add a single shot long-lines Compare the performance of self-intersection removal by comparing the cost of individually stroking each line versus stroking them all en-mass. perf/long-lines.c | 45 ++++++++++++++++++++++++++++++++++----------- 1 file changed, 34 insertions(+), 11 deletions(-) commit 2f033af5ab15dc56675c42434bb1c18655012b03 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jul 30 09:42:31 2009 +0100 [perf] Fix the asymmetry in long lines We missed the final long diagonal to the bottom-right. perf/long-lines.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 944206e55ba997a4a55334be3c7ebf8141016be7 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jul 30 09:36:28 2009 +0100 [perf] Correct typo in long-lines.c Computed may_y using height not width, fortunately the test cases have always been square. perf/long-lines.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 229887e980f03d90248add8af9cd0a9be6cb0f9a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jul 29 19:48:00 2009 +0100 [perf] Check output Add a CAIRO_PERF_OUTPUT environment variable to cause cairo-perf to first generate an output image in order to manually check that the test is functioning correctly. This needs to be automated, so that we have absolute confidence that the performance tests are not broken - but isn't that the role of the test suite? If we were ever to publish cairo-perf results, I would want some means of verification that the test-suite had first been passed. perf/cairo-perf.c | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+) commit 8c6ecfe6488dff93090d5d0a2d814466804bc7de Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jul 29 16:17:36 2009 +0100 [perf] Remove the warning about failing to open a directory The warning is repeated in the error message if we fail to find any traces, and now that we search a path it is likely that some elements do not exist. Thus we annoy the user with irrelevant, non-fatal warnings. Still looking for suggestions for the most appropriate home for the system wide cairo-traces dir... perf/cairo-perf-trace.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) commit 8c1aed17ca5dafb00175ac413d56760a3ef012f9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jul 29 13:24:19 2009 +0100 [perf] Add another stroking micro-benchmark The original stroke only contains a single subpath. Self-intersection removal particularly affects strokes with multiple curved segments, so add a path that encompasses both straight edges and rounded corners. perf/stroke.c | 36 ++++++++++++++++++++++++++++++++++++ 1 file changed, 36 insertions(+) commit 71f5649846aa8e9e2178e7caf69ab47554f86c4d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jul 29 15:36:25 2009 +0100 [path] Fix iter to handle circular list of buffers When switching the path over to use the circularly linked list, 73f801, I missed updating the path iterator. src/cairo-path-fixed-private.h | 1 + src/cairo-path-fixed.c | 9 +++++++-- 2 files changed, 8 insertions(+), 2 deletions(-) commit acfcf4a31b3370ca7bbdd522fad9ddc483df3472 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jul 29 16:10:04 2009 +0100 [path] Remove the broken rel-curve-to as line-to As pointed out by Andrea, and now tested by test/degenerate-rel-curve-to, this attempt at removing degenerate curve-to was broken. src/cairo-path-fixed.c | 6 ------ 1 file changed, 6 deletions(-) commit 81d3e15e6ca7b7c1a07fdfb6d03c43484ce9f502 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jul 29 16:07:17 2009 +0100 [test] Yet another bug in curve-to as line-to. Andrea Canciani (ranma42) found another instance of my broken 'degenerate' curve-to as line-to optimisation. All I can say is when I do something wrong, at least I'm consistent! This test case highlights the bug in the rel-curve-to path. test/Makefile.am | 2 + test/Makefile.sources | 1 + test/degenerate-rel-curve-to.c | 99 ++++++++++++++++++++++++++++++ test/degenerate-rel-curve-to.ps.xfail.png | Bin 0 -> 202 bytes test/degenerate-rel-curve-to.ref.png | Bin 0 -> 270 bytes 5 files changed, 102 insertions(+) commit f7121ca88844fa70ecb1a35f3f6c9b653324718d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jul 28 20:48:26 2009 +0100 [gl] Fixup unbounded regions after spans src/cairo-gl-surface.c | 68 ++++++++++++++++++++++++++++++++++++++++++++------ 1 file changed, 61 insertions(+), 7 deletions(-) commit 5631f01075a030255c5eb47a8ee600b062705d0f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jul 28 18:56:12 2009 +0100 [gl] Cache texture snapshots. src/cairo-gl-glyphs.c | 32 ++++-- src/cairo-gl-private.h | 2 +- src/cairo-gl-surface.c | 297 +++++++++++++++++++++++++++++-------------------- 3 files changed, 204 insertions(+), 127 deletions(-) commit 3b0bb2cfd2ad8579e88f1c0ab46f5ccf762e0974 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jul 28 18:04:59 2009 +0100 [gl] Replace DrawPixels with TexSubImage src/cairo-gl-surface.c | 60 +++++++++++++++++++++++--------------------------- 1 file changed, 27 insertions(+), 33 deletions(-) commit 4b7a1a2da7a7f55db788cbfb3164f9f948bbb26b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jul 28 17:32:30 2009 +0100 [gl] Disable the EXTEND_REFLECT convertor As we can handle EXTEND_REFLECT via the samplers, we do not need to convert such patterns to EXTEND_REPEAT during acquisition. src/cairo-gl-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 780a95b024c5880353ffc256a634eb3bcc641601 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jul 28 16:49:04 2009 +0100 [gl] Trim glyphs to surface/clip extents. src/cairo-gl-glyphs.c | 21 ++++++++++++++++++--- 1 file changed, 18 insertions(+), 3 deletions(-) commit b1227dea29d3e357cc69c49eed0ae10d5578f24a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jul 28 16:39:52 2009 +0100 [gl] Render glyphs to an intermediate mask as opposed to falling back. There are many common scenarios, mostly involving overlapping glyphs, for which to guarantee correct rendering we have to composite the glyphs via an explicit mask. That is instead of just blending the glyphs on to the destination, we have to add the glyphs to a mask, and then composite that mask+source with the destination. src/cairo-gl-glyphs.c | 306 ++++++++++++++++++++++++++++++++------------------ 1 file changed, 194 insertions(+), 112 deletions(-) commit cd259489e2f41caaa7441233d87b46fbd43991ae Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jul 28 15:58:54 2009 +0100 [gl] Fallback for overlapping glyphs. src/cairo-analysis-surface.c | 6 ++++-- src/cairo-gl-glyphs.c | 35 ++++++++++++++++++++--------------- src/cairo-gl-private.h | 13 +++++++++++++ src/cairo-scaled-font.c | 23 +++++++++++++++++++++-- src/cairo-surface-fallback.c | 3 ++- src/cairo-xcb-surface.c | 3 ++- src/cairo-xlib-surface.c | 3 ++- src/cairoint.h | 3 ++- 8 files changed, 66 insertions(+), 23 deletions(-) commit 3d8f1d3dc83b9a86f2f104f0e2afa192a34d18c8 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jul 28 15:38:45 2009 +0100 [gl] Only unlock the glyph cache when full. src/cairo-gl-glyphs.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) commit 922b1fe8e380ceba0630935460ce758ef660df75 Author: Adrian Johnson <ajohnson@redneon.com> Date: Tue Jul 28 22:34:42 2009 +0930 Ensure Type 3 fonts do not use the name .notdef in the PS encoding src/cairo-scaled-font-subsets.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit a251e34dbbcb652906f7fddd2a72cbb5992b58ae Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jul 28 13:44:19 2009 +0100 [gl] Move check for vertex buffer support to individual operations. As we can fallback instead of using vertex buffers, do so rather than preventing any use of GL acceleration. src/cairo-gl-glyphs.c | 3 +++ src/cairo-gl-surface.c | 17 +++++++++-------- 2 files changed, 12 insertions(+), 8 deletions(-) commit 90657bacc306e65e07e9eb16561a805150aef778 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jul 28 13:30:31 2009 +0100 [gl] Separate out supported operator check. By performing the check on whether the operator is supported prior to acquiring the context, we do not need to handle the error part way through the context setup. This makes the code much cleaner, and save some work for the unsupported cases. src/cairo-gl-glyphs.c | 7 ++++--- src/cairo-gl-private.h | 5 ++++- src/cairo-gl-surface.c | 46 ++++++++++++++++++---------------------------- 3 files changed, 26 insertions(+), 32 deletions(-) commit 21d50e2163038c8acc0eef738acf61f851835f7d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jul 28 13:24:43 2009 +0100 [gl] Unthaw font along contended path. If we bail after freezing the font, make sure we thaw that it is thawed before returning. src/cairo-gl-glyphs.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 5440055fc0329c726d324984cd2fdf07badc7e0f Author: Adrian Johnson <ajohnson@redneon.com> Date: Tue Jul 28 20:59:25 2009 +0930 Ensure Type 3 fonts do not add a .notdef glyph src/cairo-scaled-font-subsets.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 1ae5a41951b8a8cb415835559f91d1636a885216 Author: Adrian Johnson <ajohnson@redneon.com> Date: Tue Jul 28 20:54:58 2009 +0930 PDF: Only Type 3 fonts should include glyph 0 in /ToUnicode 4c498098 made all 8-bit fonts include glyph 0 which incorrectly included Type 1 fallback fonts. src/cairo-pdf-surface.c | 20 +++++++++++++------- src/cairo-scaled-font-subsets.c | 1 + src/cairoint.h | 1 + 3 files changed, 15 insertions(+), 7 deletions(-) commit c11f369057c5ebb958bec58ef41f8ad4b43bdbee Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jul 28 13:02:10 2009 +0100 [perf] Avoid NULL derefs on broken result files. Instead of testing for a NULL return we checked that the pointer didn't point to NIL. Oops. perf/cairo-perf-report.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) commit 19f44982348d247057df4481010c1e3ddc80bac2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jul 28 12:58:07 2009 +0100 [test] Wrap yet another call to get_image_surface() inside a timeout. I missed this call to get_image_surface() that is now being hit having restored the reference image for dash-infinite-loop. test/cairo-test.c | 3 +++ 1 file changed, 3 insertions(+) commit e2f912dc5bbfad3a6f4b5e1dbfbe51be148b3232 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jul 28 09:32:36 2009 +0100 [script] Add cvi, cvr, mod Add implementations of convert-to-integer, convert-to-real and modulus supplied by Zack Weinberg <zweinberg@mozilla.com>. util/cairo-script/cairo-script-operators.c | 95 +++++++++++++++++++++++++++++- util/cairo-script/cairo-script-private.h | 3 + util/cairo-script/cairo-script-scanner.c | 6 +- 3 files changed, 100 insertions(+), 4 deletions(-) commit 24b5ac6377f0e4b76b73460a497b9861fc01a889 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jul 28 08:05:22 2009 +0100 [test] Add timeouts around finishing and converting surfaces to images. In order to catch infinite loops whilst replaying and converting vector surfaces to images (via external renderers) we need to also install alarms around the calls to finish() and get_image(). test/cairo-test.c | 14 ++++++++++---- 1 file changed, 10 insertions(+), 4 deletions(-) commit dacc380dd9ea23fe060d0397104ecd4b2a6d0eae Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Sat Jul 11 02:59:12 2009 +0300 [test] Test extremely small dash lengths. The stroker code is liable to wedge when passed dash patterns which don't advance the dash offset due to limited precision arithmetic. This test attempts to hit all the places in the stroker where that can happen. Reported on the cairo mailing list by Hans Breuer: http://lists.cairographics.org/archives/cairo/2009-June/017506.html test/Makefile.am | 1 + test/Makefile.sources | 1 + test/dash-infinite-loop.c | 83 ++++++++++++++++++++++++++++++++++++++++ test/dash-infinite-loop.ref.png | Bin 0 -> 642 bytes 4 files changed, 85 insertions(+) commit 128d6fb2daf4e57def813511cc89167217041bf4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jul 28 08:37:54 2009 +0100 [path] Remove the erroneous conversion of a 'zero-length' curve-to As pointed out by Andrea, and now tested by test/degenerate-curve-to, a curve-to that begins and ends on the same point may extend further due to its control points. It can not be simply replaced with a degenerate line-to. In order to do so we will need more extensive degeneracy checking, ala _cairo_spline_init(). src/cairo-path-fixed.c | 3 --- 1 file changed, 3 deletions(-) commit 660c389d11bda29498807f41d3b761fa540ae9b9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jul 28 08:18:17 2009 +0100 [test] Add degenerate-curve-to Andrea Canciani (ranma42) pointed out a second bug in the curve-to as line-to optimisation, that is a curve starting and finishing on the same point is not necessarily degenerate. This test case exercises 5 different curves that start and end on the same point. test/Makefile.am | 2 + test/Makefile.sources | 1 + test/degenerate-curve-to.c | 103 ++++++++++++++++++++++++++++++++++ test/degenerate-curve-to.ps.xfail.png | Bin 0 -> 202 bytes test/degenerate-curve-to.ref.png | Bin 0 -> 276 bytes 5 files changed, 106 insertions(+) commit d31de83e01468eaf9e1a906aef536b63e77cb752 Author: Carl Worth <cworth@cworth.org> Date: Mon Jul 27 15:58:27 2009 -0700 Mark curve-to-as-line-to as XFAIL for PS backend. Looks like ghostscript has a similar buggy optimization like we just fixed in cairo. I'm just waiting on a new bugzilla account from bugs.ghostscript.com after which I plan to report the bug there. test/curve-to-as-line-to.ps.xfail.png | Bin 0 -> 112 bytes 1 file changed, 0 insertions(+), 0 deletions(-) commit 45919a4f0c94a247b1c6941dbc4a57f6c9399396 Author: Carl Worth <cworth@cworth.org> Date: Mon Jul 27 15:45:55 2009 -0700 Revert "[path] Convert straight curve-to to line-to" This reverts commit c72ca2f2296b5fbc5859059b98221e5ffe087dae. This commit was broken as verified by the curve-to-as-line-to test case. src/cairo-path-fixed.c | 12 ------------ 1 file changed, 12 deletions(-) commit 5bed405b278a6b934369873f55117d497f69bc3a Author: Carl Worth <cworth@cworth.org> Date: Mon Jul 27 15:45:15 2009 -0700 Revert "[spline] Treat a straight spline as degenerate." This reverts commit f3d265559a2f97152ce8f307ea3ce83463083088. This commit was broken as verified by the curve-to-as-line-to test case. src/cairo-spline.c | 9 --------- 1 file changed, 9 deletions(-) commit 01acad1659caf766c57849f85dc3536fe2167608 Merge: 3cc9a6050 cf15aed0c Author: Carl Worth <cworth@cworth.org> Date: Mon Jul 27 15:26:19 2009 -0700 Merge branch 'master' of git.cairographics.org:/git/cairo commit 3cc9a6050d9704976d8efe373b913e5309bde6d2 Author: Carl Worth <cworth@cworth.org> Date: Mon Jul 27 15:24:55 2009 -0700 Add new test: curve-to-as-line-to This test anticipates a future optimization, (already pushed upstream but not pulled yet), with a buggy implementation of replacing curve_to with line_to. test/Makefile.sources | 1 + test/curve-to-as-line-to.c | 95 +++++++++++++++++++++++++++++++++++++++ test/curve-to-as-line-to.ref.png | Bin 0 -> 95 bytes 3 files changed, 96 insertions(+) commit e19b2866ed095cf66e9344efa442ae01dc7913ac Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jul 27 19:51:36 2009 +0100 [gl] Actually remove the glyph from the rtree on scaled_font_fini Mark the node as available when the scaled_glyph is finished, and then attempt to collapse the parent node. src/cairo-gl-glyphs.c | 5 +++-- src/cairo-rtree-private.h | 3 +++ src/cairo-rtree.c | 13 +++++++++++++ 3 files changed, 19 insertions(+), 2 deletions(-) commit e0b7979a30fff2a22fef5c170fd72c37c197ee0e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jul 27 17:44:10 2009 +0100 [gl] Switch to using the common rtree implementation. src/cairo-gl-glyphs.c | 440 ++++++++----------------------------------------- src/cairo-gl-private.h | 52 ++---- src/cairo-gl-surface.c | 5 + 3 files changed, 89 insertions(+), 408 deletions(-) commit 37bf06d66efa238c4ff9c74c86a0c9d037b300e6 Author: Eric Anholt <eric@anholt.net> Date: Mon Jul 27 09:16:46 2009 -0700 [gl] Fix handling of clip for glyphs (clip-operator test). src/cairo-gl-glyphs.c | 43 ++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 42 insertions(+), 1 deletion(-) commit 74f33513194ec3d8e69a21aa71aa7fc854415e83 Author: Eric Anholt <eric@anholt.net> Date: Fri Jul 24 14:08:55 2009 -0700 [gl] Switch the glyph cache to using VBOs. This is a major performance improvement for GL even on non-TNL hardware (i915), as we get to emit one giant DrawArrays and make GL drivers love us. Now we're actually faster than not having the glyph cache. Before glyph cache (HP 945GM): [ 0] gl firefox-20090601 238.103 238.195 0.35% 5/6 After: [ 0] gl firefox-20090601 68.181 76.735 5.46% 6/6 src/cairo-gl-glyphs.c | 177 ++++++++++++++++++++++++++++++++++---------------- 1 file changed, 120 insertions(+), 57 deletions(-) commit ee4138e587a67c3fc4d0e89b899109af89f16e72 Author: Eric Anholt <eric@anholt.net> Date: Thu Jul 23 21:12:59 2009 -0700 [gl] Start adding glyph cache support. This is pretty much cut-and-paste from ickle's cairo-drm-i915-surface, and uses fixed function and doesn't use VBOs. src/Makefile.sources | 2 +- src/cairo-gl-glyphs.c | 829 +++++++++++++++++++++++++++++++++++++++++++++++++ src/cairo-gl-private.h | 115 +++++++ src/cairo-gl-surface.c | 69 ++-- 4 files changed, 965 insertions(+), 50 deletions(-) commit cf15aed0c4e843e7297c5a1979ac0318f6df4947 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jul 27 19:18:21 2009 +0100 [xlib] Add a timely reminder to ensure that bugs are properly filed. In a discussion on IRC, attention was drawn to a dubious comment in _cairo_xlib_show_glyphs() - the precise details of which have passed out of the collective memory. src/cairo-xlib-surface.c | 6 ++++++ 1 file changed, 6 insertions(+) commit 2da01ed552d48808cdf3aa7798ddfb959d016f0f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jul 27 18:20:22 2009 +0100 [rtree] Merge the common unpin_and_evict_unused() routine Having written the same method to prune glyphs from the rtree three times, I thought wise to add the common method to the core routines. src/cairo-rtree-private.h | 1 + src/cairo-rtree.c | 50 +++++++++++++++++++++++++++++++++++++++++++++-- 2 files changed, 49 insertions(+), 2 deletions(-) commit 9f6a0f5668601c74886378d6cdb9588621d30e6d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jul 27 14:01:32 2009 +0100 [paginated] Propagate error return during finish. In bed2701, I removed the explicit finish of the paginated's target surface, since as a wrapper it did not explicitly own the surface and so should not be calling finish(). However, we do need to propagate errors from the backing surface, such as PDF, which will only be detected during the implicit finish in the final destroy. So check to see it we hold the last reference to the target (and so our destroy will trigger the implicit finish) and call the finish explicitly and check the error status afterwards. src/cairo-paginated-surface.c | 10 ++++++++++ 1 file changed, 10 insertions(+) commit 6ca3720ba66f443e70c961f7d4fdac604c831e97 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jul 27 10:06:34 2009 +0100 [test] Overlapping glyphs Add a test case to explicitly check handling of overlapping glyphs. test/Makefile.am | 4 + test/Makefile.sources | 1 + test/overlapping-glyphs.argb32.ref.png | Bin 0 -> 2696 bytes test/overlapping-glyphs.c | 123 +++++++++++++++++++++++++++ test/overlapping-glyphs.pdf.argb32.xfail.png | Bin 0 -> 2199 bytes test/overlapping-glyphs.pdf.rgb24.xfail.png | Bin 0 -> 1692 bytes test/overlapping-glyphs.rgb24.ref.png | Bin 0 -> 1650 bytes 7 files changed, 128 insertions(+) commit 09a2b2ed3189fe12483cbd673c24ceddc2c20f3f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jul 27 10:14:24 2009 +0100 [xlib] Fast-path the likely case of retrieving a known xrender_format If we know the CPU can read pointers atomically, then we can simply peek into the cached_xrender_formats to see if we already have a match, before taking the mutex. (Acquiring the mutex here is a minor nuisance that appears on the callgrind profiles.) src/cairo-xlib-display.c | 6 ++++++ 1 file changed, 6 insertions(+) commit efb3445ee1553c91e62f6b0fe50ab72d65f5c81c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jul 27 10:18:25 2009 +0100 Add a simple rectangle tree (rtree) implementation In order to efficient store small images, we need to pack them into a large texture. The rtree handles allocation of small rectangles out of a much larger whole. As well as tracking free rectangles, it can also be used to note which parts of the texture are 'pinned' -- that is have operations currently pending and so can not be modified until that batch of operations have been flushed. When the rtree is full, i.e. there is no single free rectangle to accommodate the allocation request, it will randomly evict an unpinned block large enough to fit the request. The block may comprise just a single glyph, or a subtree of many glyphs. This may not be the best strategy, but it is an effective start. src/Makefile.sources | 2 + src/cairo-rtree-private.h | 124 ++++++++++++++++ src/cairo-rtree.c | 349 ++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 475 insertions(+) commit c72ca2f2296b5fbc5859059b98221e5ffe087dae Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Jul 26 09:07:48 2009 +0100 [path] Convert straight curve-to to line-to Avoid the high cost associated with curves if we can convert the curve to a straight line. src/cairo-path-fixed.c | 12 ++++++++++++ 1 file changed, 12 insertions(+) commit f3d265559a2f97152ce8f307ea3ce83463083088 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Jul 26 09:03:36 2009 +0100 [spline] Treat a straight spline as degenerate. The fallback for degenerate splines is to treat them as a line-to, so if the spline is straight, we can just replace it with a simple line-to by treating as degenerate. src/cairo-spline.c | 9 +++++++++ 1 file changed, 9 insertions(+) commit 25858d524d158531a71a2e772551b1e497d9d423 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Jul 26 08:13:28 2009 +0100 [script] Fix reference to font source after transferring to mmap After diverting the pointers to accommodate lazy decompressing of the source, the bytecode pointer was left pointing to the original location that had already been freed - thus passing an invalid block to FreeType and unsurprisingly then, blowing up. util/cairo-script/cairo-script-operators.c | 6 ++++++ 1 file changed, 6 insertions(+) commit ede0f7e7df4e17ddc1ffe98dcdfe0115a58e7b14 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jul 25 07:19:22 2009 +0100 [pdf] Eliminate redundant whole-page clip. We do not to emit a clip if it covers the whole page. src/cairo-pdf-surface.c | 16 ++++++++++++++++ test/meta-surface-pattern.pdf.argb32.ref.png | Bin 3868 -> 3919 bytes test/meta-surface-pattern.pdf.rgb24.ref.png | Bin 3806 -> 3867 bytes test/scale-offset-image.pdf.xfail.png | Bin 8692 -> 8694 bytes test/scale-offset-similar.pdf.xfail.png | Bin 9369 -> 9368 bytes 5 files changed, 16 insertions(+) commit 4c498098c0dc66dc8c3e8cd60f79e5e5eda206b9 Author: Adrian Johnson <ajohnson@redneon.com> Date: Fri Jul 24 21:36:47 2009 +0930 PDF: Fix glyph 0 in toUnicode stream when using user fonts _cairo_pdf_surface_emit_to_unicode_stream() was reserving glyph 0 for the .notdef glyph (as required by TrueType/CFF/Type1 fallback fonts). However Type 3 fonts do not reserve glyph 0 for .notdef and need glyph 0 to be included in the toUnicode stream. http://lists.cairographics.org/archives/cairo/2009-July/017731.html src/cairo-pdf-surface.c | 71 +++++++++++++++++++++++++++++++------------------ 1 file changed, 45 insertions(+), 26 deletions(-) commit e577096dff714266101ba86ac275864b854096a4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jul 20 23:17:07 2009 +0100 [boilerplate] Fix compilation against old revisions (as old as 1.9.2!) boilerplate/Makefile.am | 7 +++++- boilerplate/cairo-boilerplate-pdf.c | 4 ++++ boilerplate/cairo-boilerplate-ps.c | 4 ++++ boilerplate/cairo-boilerplate-svg.c | 4 ++++ boilerplate/cairo-boilerplate-test-surfaces.c | 31 ++++++++++++++++----------- boilerplate/cairo-boilerplate.c | 10 ++++++--- src/cairo-types-private.h | 5 ++++- util/cairo-script/cairo-script-operators.c | 2 ++ 8 files changed, 50 insertions(+), 17 deletions(-) commit 0466053099b7f27065267d49c84e9e24bff6c7a1 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jul 23 23:45:01 2009 +0100 [image] Discard redundant clears On slow machines the call to pixman_fill_sse2() on similar surfaces that we know are already zeroed takes a significant amount of time [12.77% of the profile for a firefox trace, cf to just 3% of the profile is spent inside memset]. Rather than solve why the pixman_fill_sse2() is so slow, simply skip the redundant clears. src/cairo-image-surface.c | 35 ++++++++++++++++++++++++++++++++++- src/cairoint.h | 1 + 2 files changed, 35 insertions(+), 1 deletion(-) commit 8b8e03503d7a4fdacfc48e0b35c9b02d7a08c1e2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jul 23 22:00:05 2009 +0100 [script] mark scan_read() inline Small function that is critical to performance whilst scanning scripts. util/cairo-script/cairo-script-scanner.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) commit f02ba09475b751fba411addb96718dbcb1de5132 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jul 23 20:23:51 2009 +0100 [xlib] Check workqueue before taking the display mutex Optimistically check to see if there is any outstanding work before checking under the mutex. We don't care if we occasionally do not run the queue this time due to contention, since we will most likely check again very shortly or clean up with the display. src/cairo-xlib-display.c | 6 ++++++ 1 file changed, 6 insertions(+) commit 4dc62bbedc5d8669c0508963497c568d0202f579 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jul 23 19:34:10 2009 +0100 [test] Extend radial gradients to check r2 < r1 Include additional gradients to cover the cases where r1 > r2, as a pixman bug was exposed by: [Bug 22908] Invalid output of radial gradient http://bugs.freedesktop.org/show_bug.cgi?id=22908 test/radial-gradient.c | 36 +++++++++++++++++++++++------------- test/radial-gradient.pdf.ref.png | Bin 79612 -> 277162 bytes test/radial-gradient.ref.png | Bin 91159 -> 280971 bytes 3 files changed, 23 insertions(+), 13 deletions(-) commit e98e2690cb4ac475819394334cb5e61a8e3d5557 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jul 22 18:32:47 2009 +0100 [test] Fix use of uninitialized widths. user-font-rescale copied unitialized values from the widths array into the desired array. Although these corresponded to unused glyphs and so were never used during the rendering, the values may have been illegal causing FPE as they were copied. test/user-font-rescale.c | 13 +++++-------- 1 file changed, 5 insertions(+), 8 deletions(-) commit 596dec95610ef02d86ede94eca5b83221b0b533a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jul 22 18:15:38 2009 +0100 [test] Trap SIGFPE Note the crash if we hit a floating-point exception. test/cairo-test.c | 7 +++++++ 1 file changed, 7 insertions(+) commit 0561539880b480273767a9559da89a25970802e2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jul 22 18:31:37 2009 +0100 [scaled-font] cairo_scaled_font_glyph_extents() initialise extents Ensure that the extents are zeroed on error. src/cairo-scaled-font.c | 23 ++++++++++++++++------- 1 file changed, 16 insertions(+), 7 deletions(-) commit 5c843b5f8eca4b6f9756644143bc1be3bc3cbc84 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jul 21 20:43:28 2009 +0100 [scaled-font] Skip empty glyphs Skip the composition of empty glyphs - prevents redundant calls through the stack. src/cairo-scaled-font.c | 57 +++++++++++++++++++++++++++---------------------- 1 file changed, 31 insertions(+), 26 deletions(-) commit 0c4692a1d0667f5850f59c65754a3ffbaf688afe Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jul 20 23:34:55 2009 +0100 [perf] Specify html file on cmdline instead of redirecting all output Redirecting all output was causing the build messages to be entangled with the Performance Change html. perf/Makefile.am | 4 ++-- perf/cairo-perf-diff | 6 +++--- 2 files changed, 5 insertions(+), 5 deletions(-) commit 1ec1d6148e85d4a9b9da825ea57adb0b8ceba1c0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jul 20 22:42:44 2009 +0100 [perf] Compare performance against most recent tag. perf/Makefile.am | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) commit 9d7fd626c39c37368ea05f91a358899e5b5b898c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jul 23 17:05:28 2009 +0100 [script] Decompress font sources Need to decompress patterns and embedded bytecode before passing to fontconfig/FreeType. util/cairo-script/cairo-script-operators.c | 119 ++++++++++++++++++++++++----- 1 file changed, 99 insertions(+), 20 deletions(-) commit 5d8f366581c0ca31fac76c991cb681d036af275f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jul 23 16:36:02 2009 +0100 [test] clip representation is no longer surface dependent So we can test the api just once in the preamble and not per-target. test/get-clip.c | 200 +++++++++++++++++++++----------------------------------- 1 file changed, 74 insertions(+), 126 deletions(-) commit c3f2db4f73e93a3dafe0f52a5f9ca09ca78cc906 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jun 11 15:18:55 2009 +0100 [drm] Add an accelerated image surface. Use the DRM interface to h/w accelerate composition on image surfaces. The purpose of the backend is simply to explore what such a hardware interface might look like and what benefits we might expect. The use case that might justify writing such custom backends are embedded devices running a drm compositor like wayland - which would, for example, allow one to write applications that seamlessly integrated accelerated, dynamic, high quality 2D graphics using Cairo with advanced interaction (e.g. smooth animations in the UI) driven by a clutter framework... In this first step we introduce the fundamental wrapping of GEM for intel and radeon chipsets, and, for comparison, gallium. No acceleration, all we do is use buffer objects (that is use the kernel memory manager) to allocate images and simply use the fallback mechanism. This provides a suitable base to start writing chip specific drivers. boilerplate/Makefile.sources | 2 + boilerplate/Makefile.win32.features | 20 + boilerplate/cairo-boilerplate-drm.c | 91 ++++ build/Makefile.win32.features | 2 + build/Makefile.win32.features-h | 6 + build/configure.ac.features | 1 + configure.ac | 25 + perf/cairo-perf-trace.c | 3 + perf/cairo-perf.c | 5 +- src/Makefile.sources | 13 + src/Makefile.win32.features | 28 ++ src/cairo-debug.c | 4 + src/cairo-drm.h | 135 ++++++ src/cairo-freelist-private.h | 59 ++- src/cairo-freelist.c | 73 +++ src/cairo-image-surface.c | 1 + src/cairo-mutex-list-private.h | 3 + src/cairo.h | 2 + src/cairoint.h | 8 + src/drm/cairo-drm-bo.c | 120 +++++ src/drm/cairo-drm-gallium-surface.c | 696 +++++++++++++++++++++++++++ src/drm/cairo-drm-intel-private.h | 182 +++++++ src/drm/cairo-drm-intel-surface.c | 475 ++++++++++++++++++ src/drm/cairo-drm-intel.c | 933 ++++++++++++++++++++++++++++++++++++ src/drm/cairo-drm-ioctl-private.h | 12 + src/drm/cairo-drm-private.h | 257 ++++++++++ src/drm/cairo-drm-radeon-private.h | 110 +++++ src/drm/cairo-drm-radeon-surface.c | 437 +++++++++++++++++ src/drm/cairo-drm-radeon.c | 447 +++++++++++++++++ src/drm/cairo-drm-surface.c | 517 ++++++++++++++++++++ src/drm/cairo-drm.c | 362 ++++++++++++++ test/cairo-test.c | 2 +- test/get-clip.c | 1 + 33 files changed, 5025 insertions(+), 7 deletions(-) commit bed2701e1c89095878d549cbca8f22d84f3dda3c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jul 23 15:32:13 2009 +0100 Remove clip handling from generic surface layer. Handling clip as part of the surface state, as opposed to being part of the operation state, is cumbersome and a hindrance to providing true proxy surface support. For example, the clip must be copied from the surface onto the fallback image, but this was forgotten causing undue hassle in each backend. Another example is the contortion the meta surface endures to ensure the clip is correctly recorded. By contrast passing the clip along with the operation is quite simple and enables us to write generic handlers for providing surface wrappers. (And in the future, we should be able to write more esoteric wrappers, e.g. automatic 2x FSAA, trivially.) In brief, instead of the surface automatically applying the clip before calling the backend, the backend can call into a generic helper to apply clipping. For raster surfaces, clip regions are handled automatically as part of the composite interface. For vector surfaces, a clip helper is introduced to replay and callback into an intersect_clip_path() function as necessary. Whilst this is not primarily a performance related change (the change should just move the computation of the clip from the moment it is applied by the user to the moment it is required by the backend), it is important to track any potential regression: ppc: Speedups ======== image-rgba evolution-20090607-0 1026085.22 0.18% -> 672972.07 0.77%: 1.52x speedup ▌ image-rgba evolution-20090618-0 680579.98 0.12% -> 573237.66 0.16%: 1.19x speedup ▎ image-rgba swfdec-fill-rate-4xaa-0 460296.92 0.36% -> 407464.63 0.42%: 1.13x speedup ▏ image-rgba swfdec-fill-rate-2xaa-0 128431.95 0.47% -> 115051.86 0.42%: 1.12x speedup ▏ Slowdowns ========= image-rgba firefox-periodic-table-0 56837.61 0.78% -> 66055.17 3.20%: 1.09x slowdown ▏ NEWS | 6 + boilerplate/cairo-boilerplate-glitz-glx.c | 4 - boilerplate/cairo-boilerplate-pdf.c | 1 - boilerplate/cairo-boilerplate-ps.c | 8 +- boilerplate/cairo-boilerplate-qt.cpp | 4 +- boilerplate/cairo-boilerplate-svg.c | 1 - boilerplate/cairo-boilerplate-test-surfaces.c | 113 +- boilerplate/cairo-boilerplate.c | 100 +- boilerplate/cairo-boilerplate.h | 2 + perf/unaligned-clip.c | 3 + src/Makefile.sources | 7 + src/cairo-analysis-surface-private.h | 6 +- src/cairo-analysis-surface.c | 539 ++++---- src/cairo-beos-surface.cpp | 409 +++---- src/cairo-clip-private.h | 87 +- src/cairo-clip.c | 1290 ++++++++++---------- src/cairo-debug.c | 2 + src/cairo-directfb-surface.c | 491 ++++---- src/cairo-ft-font.c | 53 +- src/cairo-gl-private.h | 1 + src/cairo-gl-surface.c | 216 +++- src/cairo-glitz-surface.c | 213 ++-- src/cairo-gstate.c | 233 ++-- src/cairo-image-surface.c | 173 ++- src/cairo-meta-surface-private.h | 43 +- src/cairo-meta-surface.c | 550 ++++----- src/cairo-os2-surface.c | 14 +- src/cairo-paginated-private.h | 2 - src/cairo-paginated-surface-private.h | 9 - src/cairo-paginated-surface.c | 234 ++-- src/cairo-path-bounds.c | 65 +- src/cairo-path-fill.c | 11 +- src/cairo-path-fixed-private.h | 36 +- src/cairo-path-fixed.c | 201 ++- src/cairo-path-in-fill.c | 27 +- src/cairo-path-stroke.c | 8 +- src/cairo-pattern.c | 127 +- src/cairo-pdf-surface-private.h | 3 + src/cairo-pdf-surface.c | 220 ++-- src/cairo-ps-surface-private.h | 4 + src/cairo-ps-surface.c | 367 +++--- src/cairo-qt-surface.cpp | 373 +++--- src/cairo-quartz-image-surface.c | 14 +- src/cairo-quartz-private.h | 2 + src/cairo-quartz-surface.c | 136 ++- src/cairo-region-private.h | 73 ++ src/cairo-region.c | 170 ++- src/cairo-scaled-font.c | 132 +- src/cairo-script-surface.c | 611 +++++++-- src/cairo-script.h | 10 + src/cairo-spans-private.h | 19 +- src/cairo-spans.c | 20 +- src/cairo-stroke-style.c | 8 +- src/cairo-surface-clipper-private.h | 72 ++ src/cairo-surface-clipper.c | 138 +++ src/cairo-surface-fallback-private.h | 21 +- src/cairo-surface-fallback.c | 995 ++++++++------- src/cairo-surface-private.h | 19 +- src/cairo-surface-wrapper-private.h | 156 +++ src/cairo-surface-wrapper.c | 449 +++++++ src/cairo-surface.c | 857 ++++--------- src/cairo-svg-surface-private.h | 4 +- src/cairo-svg-surface.c | 527 ++++---- src/cairo-traps.c | 26 +- src/cairo-type3-glyph-surface-private.h | 9 +- src/cairo-type3-glyph-surface.c | 94 +- src/cairo-types-private.h | 9 +- src/cairo-user-font.c | 30 +- src/cairo-vg-surface.c | 122 +- src/cairo-win32-font.c | 16 +- src/cairo-win32-printing-surface.c | 194 ++- src/cairo-win32-private.h | 15 +- src/cairo-win32-surface.c | 279 ++--- src/cairo-xcb-surface.c | 248 ++-- src/cairo-xlib-surface-private.h | 2 +- src/cairo-xlib-surface.c | 492 ++++---- src/cairo.c | 83 +- src/cairo.h | 30 +- src/cairoint.h | 325 ++--- src/test-fallback-surface.c | 4 +- src/test-fallback16-surface.c | 4 +- src/test-null-surface.c | 35 +- src/test-paginated-surface.c | 100 +- src/test-paginated-surface.h | 6 +- src/test-wrapping-surface.c | 272 +++++ src/test-wrapping-surface.h | 51 + test/Makefile.am | 75 +- test/Makefile.sources | 2 + test/buffer-diff.c | 12 +- test/cairo-test.c | 18 +- test/clip-operator.pdf.argb32.ref.png | Bin 0 -> 9340 bytes test/clip-operator.pdf.argb32.xfail.png | Bin 9340 -> 0 bytes test/clip-operator.pdf.rgb24.ref.png | Bin 0 -> 5146 bytes test/clip-operator.pdf.rgb24.xfail.png | Bin 5155 -> 0 bytes test/clip-operator.svg12.argb32.xfail.png | Bin 8404 -> 8406 bytes test/clip-operator.svg12.rgb24.xfail.png | Bin 4523 -> 4540 bytes test/clip-operator.xlib-fallback.ref.png | Bin 0 -> 3226 bytes test/clip-operator.xlib-fallback.rgb24.ref.png | Bin 3254 -> 0 bytes test/clip-push-group.pdf.ref.png | Bin 0 -> 164 bytes test/clip-push-group.ref.png | Bin 199 -> 164 bytes test/clip-push-group.xlib.ref.png | Bin 0 -> 155 bytes test/clip-unbounded.c | 80 ++ test/clip-unbounded.ref.png | Bin 0 -> 100 bytes test/clip-unbounded.rgb24.ref.png | Bin 0 -> 97 bytes test/clip-unbounded.svg12.rgb24.xfail.png | Bin 0 -> 100 bytes test/clipped-group.pdf.argb32.ref.png | Bin 298 -> 0 bytes test/clipped-group.pdf.ref.png | Bin 0 -> 310 bytes test/clipped-group.pdf.rgb24.ref.png | Bin 298 -> 0 bytes test/clipped-group.ref.png | Bin 350 -> 316 bytes ...integer-translate-source.svg12.argb32.xfail.png | Bin 0 -> 16392 bytes ...-integer-translate-source.svg12.rgb24.xfail.png | Bin 0 -> 16392 bytes test/device-offset-fractional.pdf.argb32.ref.png | Bin 275 -> 0 bytes test/device-offset-fractional.pdf.ref.png | Bin 277 -> 0 bytes test/device-offset-fractional.pdf.rgb24.ref.png | Bin 275 -> 0 bytes test/device-offset-fractional.pdf.xfail.png | Bin 0 -> 275 bytes test/extended-blend-alpha.svg12.argb32.xfail.png | Bin 0 -> 6857 bytes test/extended-blend-alpha.svg12.rgb24.xfail.png | Bin 0 -> 5014 bytes test/extended-blend.svg12.argb32.xfail.png | Bin 0 -> 2273 bytes test/extended-blend.svg12.rgb24.xfail.png | Bin 0 -> 1856 bytes test/fill-empty.argb32.ref.png | Bin 0 -> 99 bytes test/fill-empty.c | 62 + test/fill-empty.rgb24.ref.png | Bin 0 -> 97 bytes test/fill-empty.svg12.rgb24.xfail.png | Bin 0 -> 99 bytes test/filter-nearest-offset.pdf.xfail.png | Bin 1926 -> 1867 bytes test/filter-nearest-transformed.pdf.xfail.png | Bin 488 -> 482 bytes ...t-text-vertical-layout-type1.pdf.argb32.ref.png | Bin 3632 -> 0 bytes test/ft-text-vertical-layout-type1.pdf.ref.png | Bin 3968 -> 3635 bytes ...ft-text-vertical-layout-type1.pdf.rgb24.ref.png | Bin 3632 -> 0 bytes ...t-text-vertical-layout-type3.pdf.argb32.ref.png | Bin 3642 -> 0 bytes test/ft-text-vertical-layout-type3.pdf.ref.png | Bin 3634 -> 3639 bytes ...ft-text-vertical-layout-type3.pdf.rgb24.ref.png | Bin 3642 -> 0 bytes test/ft-text-vertical-layout-type3.ref.png | Bin 3609 -> 3608 bytes test/group-unaligned.svg.argb32.xfail.png | Bin 522 -> 520 bytes test/group-unaligned.xlib-fallback.ref.png | Bin 496 -> 474 bytes test/image-surface-source.svg12.argb32.xfail.png | Bin 0 -> 278 bytes test/image-surface-source.svg12.rgb24.xfail.png | Bin 0 -> 278 bytes test/in-fill-trapezoid.c | 19 +- test/mask-glyphs.svg.ref.png | Bin 1211170 -> 1211147 bytes test/mask.pdf.argb32.ref.png | Bin 0 -> 7837 bytes test/mask.pdf.argb32.xfail.png | Bin 7944 -> 0 bytes test/mask.pdf.rgb24.ref.png | Bin 0 -> 7301 bytes test/mask.pdf.rgb24.xfail.png | Bin 7388 -> 0 bytes test/mask.xlib-fallback.rgb24.ref.png | Bin 7216 -> 0 bytes test/meta-surface-pattern.pdf.argb32.ref.png | Bin 4009 -> 3868 bytes test/meta-surface-pattern.pdf.rgb24.ref.png | Bin 3905 -> 3806 bytes test/operator-clear.xlib.argb32.ref.png | Bin 0 -> 1071 bytes ...b.ref.png => operator-clear.xlib.rgb24.ref.png} | Bin ...xfail.png => operator-source.pdf.rgb24.ref.png} | Bin test/operator-source.svg12.argb32.xfail.png | Bin 2951 -> 2967 bytes test/operator-source.svg12.rgb24.xfail.png | Bin 2963 -> 2976 bytes test/operator-source.xlib-fallback.ref.png | Bin 0 -> 3177 bytes test/operator-source.xlib-fallback.rgb24.ref.png | Bin 3193 -> 0 bytes test/path-append.xlib-fallback.ref.png | Bin 6357 -> 6320 bytes test/pdf-surface-source.svg12.argb32.xfail.png | Bin 0 -> 278 bytes test/pdf-surface-source.svg12.rgb24.xfail.png | Bin 0 -> 278 bytes test/ps-surface-source.svg12.argb32.xfail.png | Bin 0 -> 278 bytes test/ps-surface-source.svg12.rgb24.xfail.png | Bin 0 -> 278 bytes test/push-group.xlib-fallback.rgb24.ref.png | Bin 2912 -> 0 bytes test/scale-offset-image.meta.xfail.png | Bin 0 -> 8599 bytes test/scale-offset-image.pdf.rgb24.ref.png | Bin 8691 -> 0 bytes ...32.ref.png => scale-offset-image.pdf.xfail.png} | Bin test/scale-offset-image.xlib-fallback.xfail.png | Bin 0 -> 8580 bytes test/scale-offset-similar.meta.xfail.png | Bin 0 -> 9095 bytes test/scale-offset-similar.pdf.argb32.ref.png | Bin 9278 -> 0 bytes test/scale-offset-similar.pdf.rgb24.ref.png | Bin 9278 -> 0 bytes test/scale-offset-similar.pdf.xfail.png | Bin 0 -> 9369 bytes test/scale-offset-similar.xlib-fallback.xfail.png | Bin 0 -> 8580 bytes test/self-intersecting.argb32.xfail.png | Bin 287 -> 235 bytes test/self-intersecting.pdf.argb32.xfail.png | Bin 295 -> 285 bytes test/self-intersecting.pdf.rgb24.xfail.png | Bin 272 -> 269 bytes test/self-intersecting.rgb24.xfail.png | Bin 282 -> 240 bytes test/smask-fill.xlib-fallback.ref.png | Bin 1156 -> 0 bytes test/smask-text.xlib.ref.png | Bin 0 -> 1672 bytes test/smask.xlib.ref.png | Bin 0 -> 3393 bytes test/surface-pattern-big-scale-down.ps.xfail.png | Bin 0 -> 238 bytes test/svg-surface-source.svg12.argb32.xfail.png | Bin 0 -> 278 bytes test/svg-surface-source.svg12.rgb24.xfail.png | Bin 0 -> 278 bytes ...allback16-surface-source.svg12.argb32.xfail.png | Bin 0 -> 278 bytes ...fallback16-surface-source.svg12.rgb24.xfail.png | Bin 0 -> 278 bytes test/text-pattern.pdf.argb32.ref.png | Bin 0 -> 1831 bytes test/text-pattern.pdf.argb32.xfail.png | Bin 1845 -> 0 bytes test/text-pattern.pdf.rgb24.ref.png | Bin 0 -> 1598 bytes test/text-pattern.pdf.rgb24.xfail.png | Bin 1599 -> 0 bytes test/unbounded-operator.svg12.argb32.xfail.png | Bin 2767 -> 2771 bytes test/unbounded-operator.svg12.rgb24.xfail.png | Bin 1719 -> 1731 bytes test/user-font-proxy.svg.ref.png | Bin 16817 -> 16814 bytes test/xlib-surface-source.svg12.argb32.xfail.png | Bin 0 -> 278 bytes test/xlib-surface-source.svg12.rgb24.xfail.png | Bin 0 -> 278 bytes util/cairo-script/cairo-script-file.c | 39 +- util/cairo-script/cairo-script-objects.c | 22 + util/cairo-script/cairo-script-private.h | 9 +- util/cairo-script/cairo-script-scanner.c | 52 +- util/cairo-trace/trace.c | 119 +- 193 files changed, 7741 insertions(+), 5903 deletions(-) commit f5a1cdf283a6aa1f4409ccbf3c2274fb587724fe Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jul 23 14:46:16 2009 +0100 [gl] Simplify the common CLEAR of a surface Almost every surface is at sometime cleared, so catch the operation in paint(), and emit a simple glClear() instead. src/cairo-gl-surface.c | 25 ++++++++++++++++++++++++- 1 file changed, 24 insertions(+), 1 deletion(-) commit eed17527f62eab826af65258b60cd4545a71a22d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jul 23 14:30:46 2009 +0100 [gl] Do not use unchecked GL v1.5 features The span renderer uses ARB_vertex_buffer_object which was included into the core as part of OpenGL 1.5. We failed to check for the required version during initialisation, and to my surprise the i915 can only support OpenGL 1.4 as it lacks ARB_occlusion_query. So just use the ARB extension instead which is present on i915. src/cairo-gl-surface.c | 24 +++++++++++++----------- 1 file changed, 13 insertions(+), 11 deletions(-) commit 6c28c7a1e268fad0ad34cac0d8a9a3951d8f3c7a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jul 23 08:45:58 2009 +0100 [perf] Add a simple report printer. After a run, it can be useful to reprint the results, so add cairo-perf-print to perform that task. For the future, I'd like to move the performance suite over to the git/perf style of single, multi-function binary. The sequence of operations that I typically do are: ./cairo-perf-trace -r -v -i 6 > `git describe`.`hostname`.perf ./cairo-perf-diff-files REVA REVB ./cairo-perf-print REVA ./cairo-perf-compare-backends REVA which misses the caching available with cairo-perf-diff. 'make html' is almost what I want, but still too prescriptive. However, that does need to be addressed for continuous performance monitoring. Along the perf lines, those sequence of operations become: ./cairo-perf record -i 6 ./cairo-perf report ./cairo-perf report REVA REVB ./cairo-perf report --backends="image,xlib,gl" REVA REVB ./cairo-perf report --html REVA REVB Also we want to think about installing the cairo-perf binary. So we want to differentiate when run inside a git checkout. perf/Makefile.am | 8 ++++-- perf/cairo-perf-print.c | 72 +++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 78 insertions(+), 2 deletions(-) commit 92ba74d760f20cd257b3916a51c8efef1c0e021e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jul 22 19:04:07 2009 +0100 [vg] Compile fixes for EGL boilerplate boilerplate/cairo-boilerplate-vg.c | 19 +++++++++---------- 1 file changed, 9 insertions(+), 10 deletions(-) commit bfbe875ded0b0666a8048ef5fb598179a81c610a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jul 22 07:52:30 2009 +0100 [gl] Enable GL backend for cairo-perf-trace perf/cairo-perf-trace.c | 3 +++ perf/cairo-perf.c | 4 +++- 2 files changed, 6 insertions(+), 1 deletion(-) commit fed000620335a5fc2bda2ffeca5cbdd0ca5a158c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jul 22 07:49:05 2009 +0100 [NEWS] Add stub for OpenGL. NEWS | 8 ++++++++ 1 file changed, 8 insertions(+) commit bff0b11634ffe52a5c8b2bbc9b6aee5bc3d2772c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jul 22 07:45:21 2009 +0100 [gl] Update reference images test/Makefile.am | 14 ++++++++++++++ test/a1-image-sample.gl.xfail.png | Bin 0 -> 120 bytes test/alpha-similar.gl.argb32.xfail.png | Bin 0 -> 99 bytes test/alpha-similar.gl.rgb24.xfail.png | Bin 0 -> 93 bytes test/device-offset-fractional.gl.xfail.png | Bin 0 -> 311 bytes test/filter-nearest-offset.gl.xfail.png | Bin 0 -> 260 bytes test/filter-nearest-transformed.gl.xfail.png | Bin 0 -> 514 bytes test/finer-grained-fallbacks.gl.argb32.ref.png | Bin 0 -> 1075 bytes test/mask-glyphs.gl.ref.png | Bin 0 -> 1188669 bytes test/meta-surface-pattern.gl.argb32.ref.png | Bin 0 -> 3130 bytes test/scale-offset-image.gl.ref.png | Bin 0 -> 9197 bytes test/scale-offset-similar.gl.ref.png | Bin 0 -> 9197 bytes test/source-clip-scale.gl.ref.png | Bin 0 -> 164 bytes test/unbounded-operator.gl.argb32.xfail.png | Bin 0 -> 2805 bytes test/unbounded-operator.gl.rgb24.xfail.png | Bin 0 -> 1355 bytes 15 files changed, 14 insertions(+) commit 653351cd2e9fbe42333cd5ce57b7b49c6da3d122 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jul 22 07:38:24 2009 +0100 [gl] Copy font options from image backend. Enable hint metrics similar to the default image backend. However, consider using the display stored in the GLX context to query the screen default. src/cairo-gl-surface.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) commit a2ac67274d4ef89079c47cc415144a6ac0303f70 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jul 22 00:05:07 2009 +0100 [gl] Add warning about missing hook for XCloseDisplay(). src/cairo-glx-context.c | 2 ++ 1 file changed, 2 insertions(+) commit f2f42ec70a626a69f4b29f55f9d1818942bb0b19 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jul 22 00:01:22 2009 +0100 [gl] Apply pixel sample bias for spans src/cairo-gl-surface.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) commit 9e420f5d0669ec12eaaf7faf4e55642c10b95fb3 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jul 21 23:45:00 2009 +0100 [gl] Another whitespace pass. src/cairo-gl-surface.c | 69 +++++++++++++++++++++++++------------------------- 1 file changed, 35 insertions(+), 34 deletions(-) commit 222fd8b49b25c56f38ad375059240983a8f576ce Merge: 5471603b5 7ee0fee90 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jul 21 23:42:34 2009 +0100 Merge commit 'anholt/gl-span-renderer' commit 5471603b527ecf43aaa72665d47929a1038aa642 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jul 21 23:38:05 2009 +0100 [gl] Use QUADS for drawing fill_rectangles() not TRIFAN As each rectangle is independent, we need to draw the array using QUADS. src/cairo-gl-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit c3ef0c8d2011a23d73ff919dbb947ca68e1d8c1b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jul 21 23:31:30 2009 +0100 [gl] Whitespace Ensure we leave a space between a function call and its parameters. src/cairo-gl-surface.c | 90 +++++++++++++++++++++++++------------------------- 1 file changed, 45 insertions(+), 45 deletions(-) commit b71c27bf4ebd19f74f22392ddb2612d843c74a38 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jul 21 23:26:15 2009 +0100 [gl] Remove dead code. The loop between texture_setup() and clone_similar() should be impossible, since every compositing backend should know how to clone an image surface. cairo-gl is no longer an exception and so this code can safely be removed. src/cairo-gl-surface.c | 25 +------------------------ 1 file changed, 1 insertion(+), 24 deletions(-) commit c8cc6a205a68a7c87239424aa93a29b5b166561d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jul 21 23:08:39 2009 +0100 [gl] Avoid temporary allocation of pixman trapezoids When creating the trapezoid mask, avoid having to allocate a temporary array to hold the converted pixman trapezoids by instead rasterizing each trapezoid separately into the mask. src/cairo-gl-surface.c | 82 ++++++++++++++++++++++---------------------------- 1 file changed, 36 insertions(+), 46 deletions(-) commit 5deb8fa213ce86759053be391966618ee0959f50 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jun 9 16:20:32 2009 +0100 [gl] Construct a dummy window for the initial context In order to make the initial context current we need a Drawable that matches the context. In general, the RootWindow may not match the desired context so we need to query the context and construct an appropriate Drawable. src/cairo-glx-context.c | 70 ++++++++++++++++++++++++++++++++++++++++++++----- 1 file changed, 63 insertions(+), 7 deletions(-) commit 911482e5ae88ba26a1b3290731968f8b8a80cd70 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri May 22 23:44:27 2009 +0100 [gl] Constrain image sources to max texture size src/cairo-gl-private.h | 2 ++ src/cairo-gl-surface.c | 19 +++++++++++++++++++ 2 files changed, 21 insertions(+) commit 1ae5942a3aa9f73aa71438dc40221836b0dff7e2 Merge: fd9df4978 6a5957475 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jul 21 22:28:44 2009 +0100 Merge commit 'anholt/gl' Conflicts: boilerplate/Makefile.sources boilerplate/cairo-boilerplate.c build/configure.ac.features src/cairo.h util/cairo-script/Makefile.am commit fd9df4978d56bbb85ecd795afce72c0310dbcf3f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jul 21 20:51:46 2009 +0100 [perf] Don't override CAIRO_TRACE_DIR in cairo-perf-diff perf/cairo-perf-diff | 1 - 1 file changed, 1 deletion(-) commit 2dd97ae59c482f09e42d719a941e189560f00b8f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jul 20 22:19:21 2009 +0100 [perf] Include trace comparison in html output build/Makefile.am.analysis | 2 ++ perf/Makefile.am | 2 +- perf/cairo-perf-diff | 17 +++++++---------- perf/make-html.py | 3 +-- 4 files changed, 11 insertions(+), 13 deletions(-) commit f30cee70f5fadd1f2af60310a148f61ed4fc7069 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jul 20 22:17:30 2009 +0100 [perf] Search multiple directories for traces In view of sharing traces between multiple builder, add some system wide directories to the search path. This should be refined to a single canonical location before release. perf/cairo-perf-trace.c | 23 +++++++++++++++++++++-- 1 file changed, 21 insertions(+), 2 deletions(-) commit ebac096648d638d6405152d37d7b58a0f6ad3242 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jul 20 19:54:12 2009 +0100 [build] And export CC_FOR_BUILD so that we can actually use it! build/configure.ac.tools | 1 + 1 file changed, 1 insertion(+) commit ed7853d4991f17a5e6df694808d33603eecf9186 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jul 20 19:37:27 2009 +0100 [build] Use ${CC} as the default value for CC_FOR_BUILD build/configure.ac.tools | 7 +++++++ 1 file changed, 7 insertions(+) commit 809f77a5cb7c17582a0ed90285a3c48c0696ed3e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jul 20 13:44:48 2009 +0100 [test] Summarise tests that fail during the preamble. Some tests only run and check during the preamble phase, and those failures were being ignored during the summary. test/cairo-test-runner.c | 109 ++++++++++++++++++++++++++++++++++++----------- 1 file changed, 85 insertions(+), 24 deletions(-) commit dbaa08e80b6d53f905974f3d2012f9425d9b8603 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jul 20 11:53:17 2009 +0100 [test] Fallback to comparing the base image references After looking at backend specific images, check against the base image reference. This is useful to fallback surfaces like xlib-fallback, which should look closer to the image backend than the xlib backend. test/cairo-test.c | 231 ++++++++++++++++++++++++++++++++++++++---------------- 1 file changed, 163 insertions(+), 68 deletions(-) commit 164e0d2ea38baacd5888bffa5bebb5d64bfee01b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jul 20 11:02:04 2009 +0100 [test] Misidentification of XFAIL as NEW cut'n'paste error compared the image against the known failure instead of any recorded new failure, when checking for NEW fails. test/cairo-test.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit e2883177487ba294a0c97bafb00e2f56e5fe84d9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jul 20 18:55:04 2009 +0100 Add a jhbuild moduleset. util/cairo.modules | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+) commit 0c1e86d661e145981050129b163eae0fb8f4a963 Author: Pierre Tardy <tardyp@gmail.com> Date: Sat Jul 18 19:16:20 2009 +0100 Cross-compilation fix for boilerplate/test boilerplate/Makefile.am | 2 +- test/Makefile.am | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) commit aac256b2692d15fc21abcdc9c1b9e5fd35f63f66 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jul 18 16:15:47 2009 +0100 Compile fixes for mingw32. Trivial typos found whilst using a cross-compiler. src/cairo-win32-surface.c | 1 + util/cairo-script/cairo-script-private.h | 4 ++-- 2 files changed, 3 insertions(+), 2 deletions(-) commit 922b01ad141305f909a4cfffc6c65a27c6d7a8a6 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jul 17 14:27:19 2009 +0100 [test] Add a simplified CLEAR test. Just performs a very basic masked clear. This does not exhibit the bug I was looking for, but is still useful for its simplicity. test/Makefile.am | 4 ++ test/Makefile.sources | 1 + test/clear.argb32.ref.png | Bin 0 -> 695 bytes test/clear.c | 86 ++++++++++++++++++++++++++++++++++++++ test/clear.rgb24.ref.png | Bin 0 -> 618 bytes test/clear.svg12.argb32.xfail.png | Bin 0 -> 170 bytes test/clear.svg12.rgb24.xfail.png | Bin 0 -> 170 bytes 7 files changed, 91 insertions(+) commit 52fa8760aeef38abbab0484a6978adaf4f100f90 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jul 15 16:37:25 2009 +0100 Add OpenVG backend. Based on the work by Øyvind Kolås and Pierre Tardy -- many thanks to Pierre for pushing this backend for inclusion as well as testing and reviewing my initial patch. And many more thanks to pippin for writing the backend in the first place! Hacked and chopped by myself into a suitable basis for a backend. Quite a few issues remain open, but would seem to be ready for testing on suitable hardware. AUTHORS | 3 +- NEWS | 3 + boilerplate/Makefile.sources | 1 + boilerplate/Makefile.win32.features | 30 + boilerplate/cairo-boilerplate-vg.c | 350 +++++++ build/Makefile.win32.features | 3 + build/Makefile.win32.features-h | 9 + build/configure.ac.features | 3 + configure.ac | 43 + src/Makefile.sources | 3 + src/Makefile.win32.features | 42 + src/cairo-vg-surface.c | 1938 +++++++++++++++++++++++++++++++++++ src/cairo-vg.h | 103 ++ src/cairo.h | 4 +- test/cairo-test.c | 3 +- test/get-clip.c | 1 + util/cairo-trace/trace.c | 70 ++ 17 files changed, 2606 insertions(+), 3 deletions(-) commit 34d5b862bce3e9273391eb5d2c384394cca94d2d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jul 16 14:18:13 2009 +0100 [xlib] Enable Render support for the extended PDF operators With the latest 0.11 release of renderproto we now have the ability to use the extended PDF operators via Render. src/cairo-image-surface.c | 10 ++++++---- src/cairo-xlib-surface.c | 48 ++++++++++++++++++++++++++++++++++++++++++++--- 2 files changed, 51 insertions(+), 7 deletions(-) commit 01eec61f4ffae4f492d667b6accb5ed3f6960d2c Author: Tobias Hunger <tobias.hunger@gmail.com> Date: Thu Jul 16 00:16:59 2009 -0400 [qt] Remove use of unnecessary QBrush pointer QBrush is a implicitly shared datastructure. There is on need to layer "explicit pointer based sharing" atop of it. Mozilla Bug #498871 src/cairo-qt-surface.cpp | 31 +++++++++++++++---------------- 1 file changed, 15 insertions(+), 16 deletions(-) commit 222c12cc4e48c2c9178a7390ea8edd73d7cb56bb Author: Tobias Hunger <tobias.hunger@gmail.com> Date: Thu Jul 16 00:16:59 2009 -0400 [qt] Remove use of unnecessary QPen pointer QPen is a implicitly shared class, so there is no need to do explicit sharing via pointers. Mozilla Bug #498873 src/cairo-qt-surface.cpp | 22 ++++++++++------------ 1 file changed, 10 insertions(+), 12 deletions(-) commit 0797b2fcaad0c17c35e67ec5d01a0556a76ee39d Author: Tobias Hunger <tobias.hunger@gmail.com> Date: Thu Jul 16 00:16:59 2009 -0400 [qt] Cleanup some whitespace Mozilla bug #498843 src/cairo-qt-surface.cpp | 212 +++++++++++++++++++++++------------------------ 1 file changed, 106 insertions(+), 106 deletions(-) commit d782ba04b76ae18c562840242dc535e9db48aa13 Author: Tobias Hunger <tobias.hunger@gmail.com> Date: Thu Jul 16 00:16:58 2009 -0400 [qt] Add method to flush qpainter This is required for non-X11 rendering backends. Mozilla bug #498910 src/cairo-qt-surface.cpp | 14 +++++++++++++- 1 file changed, 13 insertions(+), 1 deletion(-) commit e713857b85ef45d923eae94d8e4c05c7b2049ca5 Author: Tobias Hunger <tobias.hunger@gmail.com> Date: Thu Jul 16 00:16:58 2009 -0400 [qt] Move _opstr() inside the #if 0 to avoid warnings Mozilla bug #498917 src/cairo-qt-surface.cpp | 50 ++++++++++++++++++++++++------------------------ 1 file changed, 25 insertions(+), 25 deletions(-) commit 4f9551e5e533deb66401b601766bf4263192b6de Author: Tobias Hunger <tobias.hunger@gmail.com> Date: Thu Jul 16 00:16:57 2009 -0400 [qt] Reduce padding in cairo_qt_surface_t struct Removes 4 bytes of padding (on a 32bit system) from the datastructure by reordering the members. Mozilla bug #498880 src/cairo-qt-surface.cpp | 27 ++++++++++++++------------- 1 file changed, 14 insertions(+), 13 deletions(-) commit 1caa34103925c53b180cd614f5c6b425b5a3f271 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jul 15 13:26:34 2009 +0100 Fix build under distcheck Need to play a little dance to get the sources included and paths correct whilst building source files with objdir != srcdir under distcheck. boilerplate/Makefile.am | 10 +-- boilerplate/make-cairo-boilerplate-constructors.c | 19 ++++-- doc/public/tmpl/cairo-surface.sgml | 2 + doc/public/tmpl/cairo.sgml | 15 +++++ test/Makefile.am | 9 ++- test/make-cairo-test-constructors.c | 13 +++- util/cairo-script/csi-bind.c | 76 +++++++++++++++++++++++ 7 files changed, 125 insertions(+), 19 deletions(-) commit 818ad38a0e323892231469520a4b077f4dae31d4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jul 14 19:07:42 2009 +0100 [boilerplate] Add make-...-constructors to .gitignore boilerplate/.gitignore | 1 + 1 file changed, 1 insertion(+) commit 9a65b4a116a3ae451cbba2826467ad3375235a5c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jul 14 19:06:53 2009 +0100 [perf] Add cairo-traces to .gitignore perf/.gitignore | 1 + 1 file changed, 1 insertion(+) commit 5796a9240ea02dbdbef0d51a4ee71398fea858ad Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jul 14 19:04:56 2009 +0100 [test] Reconstruct REFERENCE_IMAGES Rebuild the list from scratch using the contents of my git working tree. test/Makefile.am | 449 ++++++++++++++++++++++--------------------------------- 1 file changed, 178 insertions(+), 271 deletions(-) commit 8c55ca9ebce487c5ca8b1712f8358a6361032d43 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jul 14 16:54:51 2009 +0100 [directfb] Silence compiler warnings. Add extended blend mode to operator switch. src/cairo-directfb-surface.c | 15 +++++++++++++++ 1 file changed, 15 insertions(+) commit 8ad3fca2109f18b5125e7087b3059eb4225eec3e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jul 14 16:51:31 2009 +0100 [glitz] Extended blend mode support Ensure that we fallback given an extended operator. src/cairo-glitz-surface.c | 80 +++++++++++++++++++++++++++++++++++------------ 1 file changed, 60 insertions(+), 20 deletions(-) commit 333d37a60fa80f325612ea94b19ec9892dff8cfb Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jul 14 16:41:00 2009 +0100 [pdf] Silence copmiler. Fill in missing case values to silence the compiler, and remind us should we every add more operators in future. src/cairo-pdf-surface.c | 95 +++++++++++++++++++++++++------------------------ 1 file changed, 48 insertions(+), 47 deletions(-) commit 710308c64e298ae35988fb0881a33cbe92af48ab Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jul 14 16:31:41 2009 +0100 [script] Extended blend mode support. src/cairo-script-surface.c | 18 +++++++++++++++++- util/cairo-script/cairo-script-operators.c | 15 +++++++++++++++ util/cairo-trace/trace.c | 18 +++++++++++++++++- 3 files changed, 49 insertions(+), 2 deletions(-) commit 100f0d94eebcec05372263b2e68c27b91b52a1c8 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jul 14 16:30:36 2009 +0100 [cairo] Remove trailing spaces in comments. vim highlights these errors making my screen go red and upsets me. So they must die. src/cairo.h | 22 +++++++++++----------- 1 file changed, 11 insertions(+), 11 deletions(-) commit 75736603d3b976dab8ac1ef473164c618084ee60 Author: Emmanuel Pacaud <emmanuel.pacaud@free.fr> Date: Wed Aug 29 15:11:23 2007 +0200 [SVG] Add extended blend modes. src/cairo-svg-surface.c | 6 ++++++ 1 file changed, 6 insertions(+) commit 47af6cf2803737ab7248d1ce4d76f038d1f3188e Author: Adrian Johnson <ajohnson@redneon.com> Date: Tue Jul 14 15:34:13 2009 +0200 [pdf] Add support for the extra blend modes HSL modes support added by Benjamin Otte <otte@gnome.org> src/cairo-pdf-surface-private.h | 4 + src/cairo-pdf-surface.c | 162 +++++++++++++++++++++++++++++++++++++++- 2 files changed, 164 insertions(+), 2 deletions(-) commit cf186d60b0d96cde859869237fa859e28a74a037 Author: Benjamin Otte <otte@gnome.org> Date: Thu Aug 30 10:16:00 2007 +0200 [xlib] Add support for new operators This is of course just making sure fallbacks get used as Render does not support the new operators yet. src/cairo-xlib-surface.c | 28 +++++++++++++++++++++++++++- 1 file changed, 27 insertions(+), 1 deletion(-) commit 0fd944d4bfbc2fff9960378eafd18a7d8fb9f296 Author: Benjamin Otte <otte@gnome.org> Date: Tue Jul 14 15:28:10 2009 +0200 Use new pixman extended blend operators It also adds extended-blend tests. Based on a previous patch by Emmanuel Pacaud <emmanuel.pacaud@free.fr> src/cairo-image-surface.c | 31 ++++++++ src/cairo-misc.c | 30 ++++++++ src/cairo.h | 53 +++++++++++++- test/Makefile.am | 4 ++ test/Makefile.sources | 2 + test/extended-blend-alpha.argb32.ref.png | Bin 0 -> 9777 bytes test/extended-blend-alpha.c | 115 ++++++++++++++++++++++++++++++ test/extended-blend-alpha.rgb24.ref.png | Bin 0 -> 5743 bytes test/extended-blend.argb32.ref.png | Bin 0 -> 4071 bytes test/extended-blend.c | 117 +++++++++++++++++++++++++++++++ test/extended-blend.rgb24.ref.png | Bin 0 -> 3772 bytes 11 files changed, 351 insertions(+), 1 deletion(-) commit 16387f0a7dd3b474bcaf637d3e290029b79afec1 Author: Carlos Garcia Campos <carlosgc@gnome.org> Date: Tue Jul 14 11:16:24 2009 +0200 [configure] Bump pixman dependency Version 0.15.16 contains the new PDF blend mode operators. configure.ac | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 655a4dbc36d95ce4a82cbc13aa9e2002b41fa4de Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jul 13 14:31:43 2009 +0100 [test] Track XFAIL using expected results stored as xfail.png Instead of tagging the sources, which is insensitive to changes, track the known failure modes by recording the current fail as an xfail.png reference. (We also introduce a new.png to track a fresh error, so that they are not lost in the noise of the old XFAILs and hopefully do not cause everyone to fret). As we have removed the XFAIL tagging we find, surprise surprise, that some tests are now working -- so review all the reference images (as also some .ref.png now should be .xfail.png). Note: I've only checked image,pdf,ps,svg. The test surfaces report some failures that probably need to addressed in source. I've not correct the changes for win32 and quartz. Nor fixed up the experimental backends. test/alpha-similar.c | 2 +- test/alpha-similar.pdf.argb32.xfail.png | Bin 0 -> 95 bytes test/alpha-similar.pdf.rgb24.xfail.png | Bin 0 -> 93 bytes test/alpha-similar.ps.argb32.xfail.png | Bin 0 -> 95 bytes test/alpha-similar.ps.rgb24.xfail.png | Bin 0 -> 93 bytes test/alpha-similar.svg.argb32.xfail.png | Bin 0 -> 99 bytes test/alpha-similar.svg.rgb24.xfail.png | Bin 0 -> 95 bytes test/big-line.c | 2 +- test/big-line.ps.argb32.ref.png | Bin 0 -> 977 bytes test/big-line.ps.rgb24.ref.png | Bin 0 -> 709 bytes test/big-line.ps2.ref.png | Bin 1148 -> 0 bytes test/big-line.ps2.rgb24.ref.png | Bin 830 -> 0 bytes test/big-line.ps3.ref.png | Bin 1148 -> 0 bytes test/big-line.ps3.rgb24.ref.png | Bin 830 -> 0 bytes test/big-line.ref.png | Bin 1486 -> 0 bytes test/big-line.rgb24.ref.png | Bin 1308 -> 0 bytes test/big-trap.c | 7 +- test/cairo-test-runner.c | 161 +++----- test/cairo-test.c | 405 ++++++++++++++------- test/cairo-test.h | 56 ++- test/caps-joins-alpha.svg11.ref.png | Bin 2454 -> 0 bytes test/caps-joins-alpha.svg12.ref.png | Bin 2454 -> 0 bytes test/clip-fill-rule.test-fallback.rgb24.ref.png | Bin 361 -> 0 bytes test/clip-nesting.test-fallback.rgb24.ref.png | Bin 936 -> 0 bytes test/clip-operator.c | 2 +- test/clip-operator.pdf.argb32.ref.png | Bin 11603 -> 0 bytes test/clip-operator.pdf.argb32.xfail.png | Bin 0 -> 9340 bytes test/clip-operator.pdf.rgb24.ref.png | Bin 6882 -> 0 bytes test/clip-operator.pdf.rgb24.xfail.png | Bin 0 -> 5155 bytes test/clip-operator.svg12.argb32.xfail.png | Bin 0 -> 8404 bytes test/clip-operator.svg12.rgb24.xfail.png | Bin 0 -> 4523 bytes test/clip-operator.test-fallback.argb32.ref.png | Bin 8252 -> 0 bytes test/clip-operator.test-fallback.rgb24.ref.png | Bin 3241 -> 0 bytes test/clip-twice.test-fallback.argb32.ref.png | Bin 1343 -> 0 bytes test/clip-twice.test-fallback.rgb24.ref.png | Bin 1179 -> 0 bytes test/composite-integer-translate-source.c | 2 +- test/degenerate-arc.test-fallback.argb32.ref.png | Bin 547 -> 0 bytes test/degenerate-arc.test-fallback.rgb24.ref.png | Bin 547 -> 0 bytes test/degenerate-dash.c | 2 +- test/degenerate-dash.ps.xfail.png | Bin 0 -> 1835 bytes test/degenerate-path.c | 2 +- test/degenerate-path.ps.argb32.xfail.png | Bin 0 -> 236 bytes test/degenerate-path.ps.rgb24.xfail.png | Bin 0 -> 186 bytes test/degenerate-path.ps2.argb32.ref.png | Bin 390 -> 0 bytes test/degenerate-path.ps2.rgb24.ref.png | Bin 358 -> 0 bytes test/degenerate-path.ps3.argb32.ref.png | Bin 390 -> 0 bytes test/degenerate-path.ps3.rgb24.ref.png | Bin 358 -> 0 bytes test/device-offset-scale.c | 2 +- test/device-offset-scale.svg.xfail.png | Bin 0 -> 128 bytes test/extend-pad-border.c | 2 +- test/extend-pad-border.ps.ref.png | Bin 0 -> 649 bytes test/extend-pad-border.svg.xfail.png | Bin 0 -> 1063 bytes test/extend-pad-similar.svg.xfail.png | Bin 0 -> 270 bytes test/extend-pad.svg.xfail.png | Bin 0 -> 270 bytes test/fallback-resolution.c | 14 +- test/fill-alpha-pattern.pdf.argb32.ref.png | Bin 3750 -> 0 bytes test/fill-alpha-pattern.pdf.rgb24.ref.png | Bin 3758 -> 0 bytes ...fill-alpha-pattern.test-fallback.argb32.ref.png | Bin 3379 -> 0 bytes .../fill-alpha-pattern.test-fallback.rgb24.ref.png | Bin 3379 -> 0 bytes test/fill-alpha.test-fallback.argb32.ref.png | Bin 2824 -> 0 bytes test/fill-alpha.test-fallback.rgb24.ref.png | Bin 2824 -> 0 bytes test/fill-and-stroke-alpha-add.c | 2 +- test/fill-and-stroke-alpha-add.svg12.xfail.png | Bin 0 -> 614 bytes test/fill-degenerate-sort-order.c | 2 +- .../fill-degenerate-sort-order.ps.argb32.xfail.png | Bin 0 -> 1901 bytes test/fill-degenerate-sort-order.ps.rgb24.xfail.png | Bin 0 -> 1645 bytes ...enerate-sort-order.test-fallback.argb32.ref.png | Bin 2378 -> 0 bytes ...generate-sort-order.test-fallback.rgb24.ref.png | Bin 2041 -> 0 bytes test/fill-image.xlib.ref.png | Bin 0 -> 1460 bytes test/fill-rule.test-fallback.argb32.ref.png | Bin 1979 -> 0 bytes test/fill-rule.test-fallback.rgb24.ref.png | Bin 1703 -> 0 bytes test/filter-bilinear-extents.pdf.ref.png | Bin 798 -> 0 bytes test/filter-bilinear-extents.pdf.xfail.png | Bin 0 -> 401 bytes test/filter-nearest-offset.pdf.ref.png | Bin 997 -> 0 bytes test/filter-nearest-offset.pdf.xfail.png | Bin 0 -> 1926 bytes test/filter-nearest-offset.svg.xfail.png | Bin 0 -> 4419 bytes test/filter-nearest-offset.svg11.ref.png | Bin 4739 -> 0 bytes test/filter-nearest-offset.svg12.ref.png | Bin 4739 -> 0 bytes test/filter-nearest-transformed.pdf.ref.png | Bin 552 -> 0 bytes test/filter-nearest-transformed.pdf.xfail.png | Bin 0 -> 488 bytes test/filter-nearest-transformed.svg.xfail.png | Bin 0 -> 546 bytes test/filter-nearest-transformed.svg11.ref.png | Bin 559 -> 0 bytes test/filter-nearest-transformed.svg12.ref.png | Bin 559 -> 0 bytes test/finer-grained-fallbacks.c | 2 +- test/finer-grained-fallbacks.svg12.argb32.ref.png | Bin 0 -> 190 bytes test/finer-grained-fallbacks.svg12.rgb24.ref.png | Bin 0 -> 191 bytes ...-grained-fallbacks.test-fallback.argb32.ref.png | Bin 1111 -> 0 bytes ...r-grained-fallbacks.test-fallback.rgb24.ref.png | Bin 854 -> 0 bytes test/font-matrix-translation.svg.ref.png | Bin 0 -> 870 bytes test/font-matrix-translation.svg11.argb32.ref.png | Bin 857 -> 0 bytes test/font-matrix-translation.svg11.ref.png | Bin 970 -> 0 bytes test/font-matrix-translation.svg11.rgb24.ref.png | Bin 857 -> 0 bytes test/font-matrix-translation.svg12.argb32.ref.png | Bin 857 -> 0 bytes test/font-matrix-translation.svg12.ref.png | Bin 970 -> 0 bytes test/font-matrix-translation.svg12.rgb24.ref.png | Bin 857 -> 0 bytes test/ft-show-glyphs-positioning.svg.ref.png | Bin 0 -> 6018 bytes test/ft-show-glyphs-positioning.svg11.ref.png | Bin 6590 -> 0 bytes test/ft-show-glyphs-positioning.svg12.ref.png | Bin 6590 -> 0 bytes test/ft-show-glyphs-table.svg.ref.png | Bin 0 -> 10005 bytes test/ft-show-glyphs-table.svg11.argb32.ref.png | Bin 9953 -> 0 bytes test/ft-show-glyphs-table.svg11.rgb24.ref.png | Bin 9953 -> 0 bytes test/ft-show-glyphs-table.svg12.argb32.ref.png | Bin 9953 -> 0 bytes test/ft-show-glyphs-table.svg12.rgb24.ref.png | Bin 9953 -> 0 bytes test/ft-text-vertical-layout-type1.svg.ref.png | Bin 0 -> 3607 bytes ...text-vertical-layout-type1.svg11.argb32.ref.png | Bin 3614 -> 0 bytes test/ft-text-vertical-layout-type1.svg11.ref.png | Bin 3973 -> 0 bytes ...-text-vertical-layout-type1.svg11.rgb24.ref.png | Bin 3614 -> 0 bytes ...text-vertical-layout-type1.svg12.argb32.ref.png | Bin 3614 -> 0 bytes test/ft-text-vertical-layout-type1.svg12.ref.png | Bin 3973 -> 0 bytes ...-text-vertical-layout-type1.svg12.rgb24.ref.png | Bin 3614 -> 0 bytes ...tical-layout-type1.test-fallback.argb32.ref.png | Bin 3639 -> 0 bytes ...rtical-layout-type1.test-fallback.rgb24.ref.png | Bin 3639 -> 0 bytes test/ft-text-vertical-layout-type3.svg.ref.png | Bin 0 -> 3626 bytes ...text-vertical-layout-type3.svg11.argb32.ref.png | Bin 3640 -> 0 bytes test/ft-text-vertical-layout-type3.svg11.ref.png | Bin 3957 -> 0 bytes ...-text-vertical-layout-type3.svg11.rgb24.ref.png | Bin 3640 -> 0 bytes ...text-vertical-layout-type3.svg12.argb32.ref.png | Bin 3640 -> 0 bytes test/ft-text-vertical-layout-type3.svg12.ref.png | Bin 3957 -> 0 bytes ...-text-vertical-layout-type3.svg12.rgb24.ref.png | Bin 3640 -> 0 bytes ...tical-layout-type3.test-fallback.argb32.ref.png | Bin 3605 -> 0 bytes ...rtical-layout-type3.test-fallback.rgb24.ref.png | Bin 3605 -> 0 bytes test/gradient-alpha.pdf.argb32.ref.png | Bin 126 -> 0 bytes test/gradient-alpha.pdf.rgb24.ref.png | Bin 121 -> 0 bytes test/gradient-constant-alpha.pdf.argb32.ref.png | Bin 116 -> 0 bytes test/gradient-constant-alpha.pdf.rgb24.ref.png | Bin 115 -> 0 bytes test/group-unaligned.pdf.new.png | Bin 0 -> 427 bytes test/group-unaligned.pdf.ref.png | Bin 414 -> 0 bytes test/group-unaligned.svg.argb32.ref.png | Bin 524 -> 0 bytes test/group-unaligned.svg.argb32.xfail.png | Bin 0 -> 522 bytes test/group-unaligned.svg.rgb24.ref.png | Bin 434 -> 0 bytes test/group-unaligned.svg.rgb24.xfail.png | Bin 0 -> 425 bytes test/huge-linear.c | 2 +- test/huge-linear.pdf.ref.png | Bin 0 -> 1577 bytes test/huge-radial.c | 2 +- test/huge-radial.pdf.argb32.ref.png | Bin 0 -> 177382 bytes test/huge-radial.pdf.rgb24.ref.png | Bin 0 -> 225358 bytes test/large-source-roi.c | 2 +- test/large-source-roi.ref.png | Bin 102 -> 0 bytes test/linear-gradient.pdf.ref.png | Bin 1125 -> 0 bytes test/linear-gradient.svg11.argb32.ref.png | Bin 988 -> 0 bytes test/linear-gradient.svg11.ref.png | Bin 1000 -> 0 bytes test/linear-gradient.svg11.rgb24.ref.png | Bin 988 -> 0 bytes test/linear-gradient.svg12.argb32.ref.png | Bin 988 -> 0 bytes test/linear-gradient.svg12.ref.png | Bin 1000 -> 0 bytes test/linear-gradient.svg12.rgb24.ref.png | Bin 988 -> 0 bytes test/linear-gradient.test-fallback.argb32.ref.png | Bin 923 -> 0 bytes test/linear-gradient.test-fallback.rgb24.ref.png | Bin 923 -> 0 bytes test/long-lines.c | 2 +- test/long-lines.ref.png | Bin 247 -> 0 bytes test/make-html.pl | 13 +- test/mask-alpha.svg.rgb24.xfail.png | Bin 0 -> 585 bytes test/mask-alpha.svg11.argb32.ref.png | Bin 642 -> 0 bytes test/mask-alpha.svg11.rgb24.ref.png | Bin 592 -> 0 bytes test/mask-alpha.svg12.argb32.ref.png | Bin 642 -> 0 bytes test/mask-alpha.svg12.rgb24.ref.png | Bin 592 -> 0 bytes test/mask-alpha.test-fallback.argb32.ref.png | Bin 627 -> 0 bytes test/mask-ctm.svg11.argb32.ref.png | Bin 108 -> 0 bytes test/mask-ctm.svg12.argb32.ref.png | Bin 108 -> 0 bytes test/mask-glyphs.c | 2 +- test/mask-glyphs.pdf.ref.png | Bin 0 -> 1187713 bytes test/mask-glyphs.svg.ref.png | Bin 0 -> 1211170 bytes test/mask-surface-ctm.svg11.argb32.ref.png | Bin 108 -> 0 bytes test/mask-surface-ctm.svg12.argb32.ref.png | Bin 108 -> 0 bytes test/mask-transformed-similar.svg.ref.png | Bin 0 -> 3365 bytes test/mask-transformed-similar.svg11.ref.png | Bin 4051 -> 0 bytes test/mask-transformed-similar.svg12.ref.png | Bin 4051 -> 0 bytes test/mask.pdf.argb32.ref.png | Bin 9946 -> 0 bytes test/mask.pdf.argb32.xfail.png | Bin 0 -> 7944 bytes test/mask.pdf.rgb24.ref.png | Bin 8745 -> 0 bytes test/mask.pdf.rgb24.xfail.png | Bin 0 -> 7388 bytes test/mask.svg.argb32.xfail.png | Bin 0 -> 8613 bytes test/mask.svg.rgb24.xfail.png | Bin 0 -> 7189 bytes test/mask.svg11.argb32.ref.png | Bin 8682 -> 0 bytes test/mask.svg11.rgb24.ref.png | Bin 7203 -> 0 bytes test/mask.svg12.argb32.ref.png | Bin 8682 -> 0 bytes test/mask.svg12.rgb24.ref.png | Bin 7203 -> 0 bytes test/mask.test-fallback.argb32.ref.png | Bin 8457 -> 0 bytes test/mask.test-fallback.rgb24.ref.png | Bin 7058 -> 0 bytes test/meta-surface-pattern.pdf.argb32.ref.png | Bin 0 -> 4009 bytes test/meta-surface-pattern.pdf.ref.png | Bin 4111 -> 0 bytes test/meta-surface-pattern.pdf.rgb24.ref.png | Bin 0 -> 3905 bytes test/meta-surface-pattern.svg.argb32.ref.png | Bin 0 -> 3930 bytes test/meta-surface-pattern.svg.rgb24.ref.png | Bin 0 -> 3910 bytes test/meta-surface-pattern.svg11.argb32.ref.png | Bin 3924 -> 0 bytes test/meta-surface-pattern.svg11.rgb24.ref.png | Bin 3914 -> 0 bytes test/meta-surface-pattern.svg12.argb32.ref.png | Bin 3924 -> 0 bytes test/meta-surface-pattern.svg12.rgb24.ref.png | Bin 3914 -> 0 bytes test/operator-alpha.c | 2 +- test/operator-alpha.svg12.argb32.xfail.png | Bin 0 -> 274 bytes test/operator-alpha.svg12.rgb24.xfail.png | Bin 0 -> 248 bytes test/operator-clear.c | 2 +- test/operator-clear.pdf.argb32.ref.png | Bin 1607 -> 0 bytes test/operator-clear.pdf.rgb24.ref.png | Bin 1402 -> 0 bytes test/operator-clear.svg12.argb32.xfail.png | Bin 0 -> 405 bytes test/operator-clear.svg12.rgb24.xfail.png | Bin 0 -> 535 bytes test/operator-clear.xlib.ref.png | Bin 0 -> 951 bytes test/operator-source.c | 2 +- test/operator-source.pdf.argb32.ref.png | Bin 5112 -> 0 bytes test/operator-source.pdf.rgb24.ref.png | Bin 4186 -> 0 bytes test/operator-source.pdf.rgb24.xfail.png | Bin 0 -> 3645 bytes test/operator-source.svg12.argb32.xfail.png | Bin 0 -> 2951 bytes test/operator-source.svg12.rgb24.xfail.png | Bin 0 -> 2963 bytes test/operator-source.test-fallback.argb32.ref.png | Bin 4401 -> 0 bytes test/operator-source.test-fallback.rgb24.ref.png | Bin 3200 -> 0 bytes test/operator.c | 2 +- test/operator.svg12.argb32.xfail.png | Bin 0 -> 238 bytes test/operator.svg12.rgb24.xfail.png | Bin 0 -> 242 bytes test/over-above-source.c | 2 +- test/over-above-source.svg12.rgb24.xfail.png | Bin 0 -> 563 bytes .../over-above-source.test-fallback.argb32.ref.png | Bin 533 -> 0 bytes test/over-above-source.test-fallback.rgb24.ref.png | Bin 450 -> 0 bytes test/over-around-source.c | 2 +- test/over-around-source.svg12.argb32.xfail.png | Bin 0 -> 559 bytes test/over-around-source.svg12.rgb24.xfail.png | Bin 0 -> 559 bytes ...over-around-source.test-fallback.argb32.ref.png | Bin 610 -> 0 bytes test/over-below-source.c | 2 +- test/over-below-source.svg12.argb32.xfail.png | Bin 0 -> 224 bytes test/over-below-source.svg12.rgb24.xfail.png | Bin 0 -> 224 bytes test/over-between-source.c | 2 +- test/over-between-source.svg12.argb32.xfail.png | Bin 0 -> 224 bytes test/over-between-source.svg12.rgb24.xfail.png | Bin 0 -> 224 bytes ...ver-between-source.test-fallback.argb32.ref.png | Bin 578 -> 0 bytes test/paint-source-alpha.c | 2 +- test/paint-source-alpha.svg.ref.png | Bin 0 -> 693 bytes test/paint-source-alpha.svg11.ref.png | Bin 756 -> 0 bytes test/paint-source-alpha.svg12.ref.png | Bin 505 -> 0 bytes test/paint-with-alpha.svg.ref.png | Bin 0 -> 483 bytes test/paint-with-alpha.svg11.ref.png | Bin 516 -> 0 bytes test/paint-with-alpha.svg12.ref.png | Bin 516 -> 0 bytes test/path-precision.c | 9 +- test/push-group.svg11.argb32.ref.png | Bin 3034 -> 0 bytes test/push-group.svg12.argb32.ref.png | Bin 3034 -> 0 bytes test/push-group.test-fallback.argb32.ref.png | Bin 3107 -> 0 bytes test/push-group.test-fallback.rgb24.ref.png | Bin 2942 -> 0 bytes test/radial-gradient.svg11.ref.png | Bin 91039 -> 0 bytes test/radial-gradient.svg12.ref.png | Bin 91039 -> 0 bytes ...ndom-intersections.test-fallback.argb32.ref.png | Bin 132312 -> 0 bytes ...andom-intersections.test-fallback.rgb24.ref.png | Bin 132312 -> 0 bytes test/rotate-image-surface-paint.pdf.ref.png | Bin 196 -> 0 bytes test/rotate-image-surface-paint.pdf.xfail.png | Bin 0 -> 207 bytes test/rotate-image-surface-paint.svg.ref.png | Bin 0 -> 209 bytes test/rotate-image-surface-paint.svg11.ref.png | Bin 248 -> 0 bytes test/rotate-image-surface-paint.svg12.ref.png | Bin 248 -> 0 bytes test/scale-offset-image.c | 2 +- test/scale-offset-image.pdf.argb32.ref.png | Bin 0 -> 8692 bytes test/scale-offset-image.pdf.rgb24.ref.png | Bin 0 -> 8691 bytes test/scale-offset-image.xfail.png | Bin 0 -> 8599 bytes test/scale-offset-image.xlib.xfail.png | Bin 0 -> 9100 bytes test/scale-offset-similar.c | 2 +- test/scale-offset-similar.pdf.argb32.ref.png | Bin 0 -> 9278 bytes test/scale-offset-similar.pdf.rgb24.ref.png | Bin 0 -> 9278 bytes test/scale-offset-similar.xfail.png | Bin 0 -> 8599 bytes test/scale-offset-similar.xlib.xfail.png | Bin 0 -> 8570 bytes ...cale-source-surface-paint.svg.argb32.xfail.png} | Bin .../scale-source-surface-paint.svg.rgb24.xfail.png | Bin 0 -> 222 bytes .../scale-source-surface-paint.svg11.rgb24.ref.png | Bin 243 -> 0 bytes ...scale-source-surface-paint.svg12.argb32.ref.png | Bin 229 -> 0 bytes .../scale-source-surface-paint.svg12.rgb24.ref.png | Bin 243 -> 0 bytes test/self-copy-overlap.c | 2 +- test/self-copy-overlap.ref.png | Bin 1140 -> 0 bytes test/self-copy-overlap.rgb24.ref.png | Bin 608 -> 0 bytes test/self-intersecting.argb32.xfail.png | Bin 0 -> 287 bytes test/self-intersecting.c | 2 +- test/self-intersecting.pdf.argb32.xfail.png | Bin 0 -> 295 bytes test/self-intersecting.pdf.rgb24.xfail.png | Bin 0 -> 272 bytes test/self-intersecting.ps.argb32.xfail.png | Bin 0 -> 186 bytes test/self-intersecting.ps.rgb24.xfail.png | Bin 0 -> 186 bytes test/self-intersecting.rgb24.xfail.png | Bin 0 -> 282 bytes test/self-intersecting.xlib.argb32.xfail.png | Bin 0 -> 235 bytes test/self-intersecting.xlib.rgb24.xfail.png | Bin 0 -> 240 bytes test/set-source.svg11.argb32.ref.png | Bin 104 -> 0 bytes test/set-source.svg12.argb32.ref.png | Bin 104 -> 0 bytes test/smask-fill.c | 2 +- test/smask-fill.svg.ref.png | Bin 0 -> 1150 bytes test/smask-fill.svg11.argb32.ref.png | Bin 1128 -> 0 bytes test/smask-fill.svg11.ref.png | Bin 1166 -> 0 bytes test/smask-fill.svg11.rgb24.ref.png | Bin 1128 -> 0 bytes test/smask-fill.svg12.argb32.ref.png | Bin 1128 -> 0 bytes test/smask-fill.svg12.ref.png | Bin 1166 -> 0 bytes test/smask-fill.svg12.rgb24.ref.png | Bin 1128 -> 0 bytes test/smask-fill.test-fallback.argb32.ref.png | Bin 1148 -> 0 bytes test/smask-fill.test-fallback.rgb24.ref.png | Bin 1148 -> 0 bytes test/smask-image-mask.c | 2 +- test/smask-mask.c | 2 +- test/smask-mask.pdf.xfail.png | Bin 0 -> 4401 bytes test/smask-mask.svg.ref.png | Bin 0 -> 2376 bytes test/smask-mask.svg11.ref.png | Bin 2529 -> 0 bytes test/smask-mask.svg12.ref.png | Bin 2529 -> 0 bytes test/smask-paint.c | 2 +- test/smask-paint.pdf.xfail.png | Bin 0 -> 4513 bytes test/smask-paint.svg.ref.png | Bin 0 -> 2453 bytes test/smask-paint.svg11.ref.png | Bin 2612 -> 0 bytes test/smask-paint.svg12.ref.png | Bin 2612 -> 0 bytes test/smask-stroke.c | 2 +- test/smask-stroke.pdf.xfail.png | Bin 0 -> 392 bytes test/smask-text.c | 2 +- test/smask-text.pdf.ref.png | Bin 0 -> 1874 bytes test/smask-text.ref.png | Bin 1777 -> 1874 bytes test/smask-text.svg.ref.png | Bin 0 -> 1794 bytes test/smask-text.svg11.argb32.ref.png | Bin 1791 -> 0 bytes test/smask-text.svg11.ref.png | Bin 1886 -> 0 bytes test/smask-text.svg11.rgb24.ref.png | Bin 1791 -> 0 bytes test/smask-text.svg12.argb32.ref.png | Bin 1791 -> 0 bytes test/smask-text.svg12.ref.png | Bin 1886 -> 0 bytes test/smask-text.svg12.rgb24.ref.png | Bin 1791 -> 0 bytes test/smask.c | 2 +- test/smask.pdf.xfail.png | Bin 0 -> 4393 bytes test/smask.ref.png | Bin 3587 -> 3417 bytes test/smask.svg.ref.png | Bin 0 -> 3458 bytes test/smask.svg11.ref.png | Bin 3634 -> 0 bytes test/smask.svg12.ref.png | Bin 3634 -> 0 bytes test/source-clip-scale.svg.ref.png | Bin 0 -> 116 bytes test/source-clip-scale.svg11.ref.png | Bin 119 -> 0 bytes test/source-clip-scale.svg12.ref.png | Bin 119 -> 0 bytes test/surface-pattern-big-scale-down.c | 2 +- test/surface-pattern-big-scale-down.ref.png | Bin 226 -> 191 bytes test/surface-pattern-scale-down.pdf.argb32.ref.png | Bin 1532 -> 0 bytes test/surface-pattern-scale-down.pdf.ref.png | Bin 0 -> 1589 bytes test/surface-pattern-scale-down.pdf.rgb24.ref.png | Bin 1532 -> 0 bytes test/surface-pattern-scale-up.pdf.argb32.ref.png | Bin 3834 -> 0 bytes test/surface-pattern-scale-up.pdf.ref.png | Bin 0 -> 2712 bytes test/surface-pattern-scale-up.pdf.rgb24.ref.png | Bin 3834 -> 0 bytes test/surface-pattern.pdf.ref.png | Bin 14800 -> 0 bytes test/surface-pattern.pdf.xfail.png | Bin 0 -> 10528 bytes test/surface-pattern.ps.xfail.png | Bin 0 -> 2188 bytes test/surface-pattern.ps2.ref.png | Bin 2364 -> 0 bytes test/surface-pattern.ps3.ref.png | Bin 2364 -> 0 bytes ...n.svg.ref.png => surface-pattern.svg.xfail.png} | Bin test/text-pattern.pdf.argb32.ref.png | Bin 2334 -> 0 bytes test/text-pattern.pdf.argb32.xfail.png | Bin 0 -> 1845 bytes test/text-pattern.pdf.rgb24.ref.png | Bin 1915 -> 0 bytes test/text-pattern.pdf.rgb24.xfail.png | Bin 0 -> 1599 bytes test/text-pattern.svg.argb32.ref.png | Bin 0 -> 1745 bytes test/text-pattern.svg.rgb24.ref.png | Bin 0 -> 1453 bytes test/text-pattern.svg11.argb32.ref.png | Bin 1743 -> 0 bytes test/text-pattern.svg11.rgb24.ref.png | Bin 1563 -> 0 bytes test/text-pattern.svg12.argb32.ref.png | Bin 1743 -> 0 bytes test/text-pattern.svg12.rgb24.ref.png | Bin 1563 -> 0 bytes test/text-rotate.svg.ref.png | Bin 0 -> 17024 bytes test/text-rotate.svg11.argb32.ref.png | Bin 16942 -> 0 bytes test/text-rotate.svg11.ref.png | Bin 18301 -> 0 bytes test/text-rotate.svg11.rgb24.ref.png | Bin 16942 -> 0 bytes test/text-rotate.svg12.argb32.ref.png | Bin 16942 -> 0 bytes test/text-rotate.svg12.ref.png | Bin 18301 -> 0 bytes test/text-rotate.svg12.rgb24.ref.png | Bin 16942 -> 0 bytes test/text-transform.svg.ref.png | Bin 0 -> 5682 bytes test/text-transform.svg11.argb32.ref.png | Bin 5677 -> 0 bytes test/text-transform.svg11.rgb24.ref.png | Bin 5677 -> 0 bytes test/text-transform.svg12.argb32.ref.png | Bin 5677 -> 0 bytes test/text-transform.svg12.rgb24.ref.png | Bin 5677 -> 0 bytes test/trap-clip.pdf.argb32.ref.png | Bin 6746 -> 0 bytes test/trap-clip.pdf.rgb24.ref.png | Bin 6616 -> 0 bytes test/trap-clip.test-fallback.argb32.ref.png | Bin 5753 -> 0 bytes test/trap-clip.test-fallback.rgb24.ref.png | Bin 5379 -> 0 bytes test/twin.svg.ref.png | Bin 3027 -> 3040 bytes test/unbounded-operator.c | 2 +- test/unbounded-operator.svg12.argb32.xfail.png | Bin 0 -> 2767 bytes test/unbounded-operator.svg12.rgb24.xfail.png | Bin 0 -> 1719 bytes .../unbounded-operator.test-fallback.rgb24.ref.png | Bin 1306 -> 0 bytes test/user-font-mask.svg.ref.png | Bin 0 -> 2030 bytes test/user-font-mask.svg11.ref.png | Bin 2224 -> 0 bytes test/user-font-proxy.svg.ref.png | Bin 0 -> 16817 bytes test/user-font-proxy.svg11.argb32.ref.png | Bin 16836 -> 0 bytes test/user-font-proxy.svg11.ref.png | Bin 17902 -> 0 bytes test/user-font-proxy.svg11.rgb24.ref.png | Bin 16836 -> 0 bytes test/user-font-proxy.svg12.argb32.ref.png | Bin 16836 -> 0 bytes test/user-font-proxy.svg12.ref.png | Bin 17902 -> 0 bytes test/user-font-proxy.svg12.rgb24.ref.png | Bin 16836 -> 0 bytes test/user-font-proxy.test-fallback.argb32.ref.png | Bin 16835 -> 0 bytes test/user-font-proxy.test-fallback.rgb24.ref.png | Bin 16835 -> 0 bytes test/user-font-rescale.svg.ref.png | Bin 0 -> 14873 bytes test/user-font-rescale.svg11.ref.png | Bin 15741 -> 0 bytes test/user-font-rescale.svg12.ref.png | Bin 15741 -> 0 bytes test/user-font.svg.ref.png | Bin 0 -> 6379 bytes test/user-font.svg11.argb32.ref.png | Bin 6411 -> 0 bytes test/user-font.svg11.ref.png | Bin 6134 -> 0 bytes test/user-font.svg11.rgb24.ref.png | Bin 6411 -> 0 bytes test/user-font.svg12.argb32.ref.png | Bin 6411 -> 0 bytes test/user-font.svg12.ref.png | Bin 6134 -> 0 bytes test/user-font.svg12.rgb24.ref.png | Bin 6411 -> 0 bytes test/user-font.test-fallback.argb32.ref.png | Bin 5601 -> 0 bytes test/user-font.test-fallback.rgb24.ref.png | Bin 5601 -> 0 bytes 382 files changed, 438 insertions(+), 303 deletions(-) commit a81cf14bd6de545e9108f5c24c59e55026df86de Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jul 13 10:58:44 2009 +0100 [test] Sanitise REFERENCE_IMAGES Add and delete files to satisfy 'make check-ref-missing'. test/Makefile.am | 6 ------ test/fill-image.test-fallback.ref.png | Bin 1462 -> 0 bytes test/fill-image.xlib.ref.png | Bin 1462 -> 0 bytes test/group-unaligned.ref.png | Bin 0 -> 434 bytes test/meta-surface-pattern.pdf.argb32.ref.png | Bin 4011 -> 0 bytes test/meta-surface-pattern.pdf.rgb24.ref.png | Bin 3910 -> 0 bytes test/operator-clear.test-fallback.ref.png | Bin 1071 -> 0 bytes test/operator-clear.xlib.ref.png | Bin 1071 -> 0 bytes 8 files changed, 6 deletions(-) commit 7f3eda4e9f9e6cb6f3a33871ed42099c822d0ea2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jul 13 10:43:34 2009 +0100 [test] Update reference images for gs 8.64 test/Makefile.am | 7 +++++-- test/README | 2 +- test/close-path-current-point.ps.ref.png | Bin 0 -> 1490 bytes test/extend-pad.ps.ref.png | Bin 0 -> 314 bytes test/fill-image.ps.ref.png | Bin 0 -> 1799 bytes test/test-fallback16-surface-source.ps.ref.png | Bin 0 -> 292 bytes test/user-font-proxy.ps.ref.png | Bin 0 -> 7766 bytes test/user-font-proxy.ps2.ref.png | Bin 8171 -> 0 bytes test/user-font-proxy.ps3.ref.png | Bin 8171 -> 0 bytes 9 files changed, 6 insertions(+), 3 deletions(-) commit 0d44416b9dd8149032039cf44ea14bf4867f3c54 Author: Adrian Johnson <ajohnson@redneon.com> Date: Sun Jul 12 19:21:28 2009 +0930 PDF: Don't round pattern y location to an integer Using cairo_surface_get_extents() to get page height does not work when the page height is not an integer. http://lists.cairographics.org/archives/cairo/2009-July/017570.html src/cairo-pdf-surface.c | 10 +--------- 1 file changed, 1 insertion(+), 9 deletions(-) commit 9686331cb07a8f9018f28dc7d6d6e3726e289fda Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Sat May 2 18:05:20 2009 +0300 [spans] Round rasterization grid pixel areas instead of truncating. When converting a grid pixel area into the range [0,255] the GRID_AREA_TO_ALPHA() macro would truncate extra bits off the result rather than rounding. This could cause seams between abutting collinear edges of separately rendered polygons even when the coordinates of the abutting edges were the same. Reported by Soeren Sandmann on the cairo mailing list: http://lists.cairographics.org/archives/cairo/2009-May/017043.html src/cairo-tor-scan-converter.c | 4 ++-- test/clip-operator.ref.png | Bin 8210 -> 8191 bytes test/clip-operator.test-paginated.argb32.ref.png | Bin 8247 -> 8189 bytes test/fill-degenerate-sort-order.ref.png | Bin 2406 -> 2399 bytes test/fill-rule.ref.png | Bin 2061 -> 2065 bytes test/finer-grained-fallbacks.ref.png | Bin 1069 -> 1070 bytes test/linear-gradient.pdf.argb32.ref.png | Bin 1112 -> 0 bytes test/linear-gradient.pdf.ref.png | Bin 935 -> 1125 bytes test/linear-gradient.pdf.rgb24.ref.png | Bin 1112 -> 0 bytes test/mask-alpha.ref.png | Bin 643 -> 632 bytes test/mask.pdf.argb32.ref.png | Bin 9918 -> 9946 bytes test/mask.ref.png | Bin 8581 -> 8603 bytes test/operator-clear.ref.png | Bin 1084 -> 1070 bytes test/operator-clear.test-fallback.ref.png | Bin 0 -> 1071 bytes test/operator-clear.xlib.ref.png | Bin 0 -> 1071 bytes test/operator-source.ref.png | Bin 4425 -> 4395 bytes test/radial-gradient.pdf.argb32.ref.png | Bin 79601 -> 0 bytes test/radial-gradient.pdf.ref.png | Bin 79490 -> 79612 bytes test/radial-gradient.pdf.rgb24.ref.png | Bin 79601 -> 0 bytes test/trap-clip.pdf.argb32.ref.png | Bin 6745 -> 6746 bytes test/trap-clip.pdf.rgb24.ref.png | Bin 6621 -> 6616 bytes test/trap-clip.ref.png | Bin 5829 -> 5820 bytes test/trap-clip.test-paginated.argb32.ref.png | Bin 5865 -> 5898 bytes 23 files changed, 2 insertions(+), 2 deletions(-) commit fbd88db938bf3bb03934795da9b3d184c4f43608 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Jul 5 14:43:50 2009 +0100 [path] Don't include current-point in hash As we don't strictly use the current-point in comparing paths, exclude it from the hash. Similarly use the path content flags as a cheap means to differentiate contents. src/cairo-path-fixed.c | 17 +++++++++-------- 1 file changed, 9 insertions(+), 8 deletions(-) commit 73f8019fd2afd0ad41c2515bef9877348355d246 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Jul 5 14:37:02 2009 +0100 [path] Use cairo_list_t instead of open-coding its own Use the cairo_list_t and its style of iterators to improve the readability of the cairo_path_buf_t management. Note the complications that arise from the embedding of the initial buf -- however the macros do help make the unusual manipulations more identifiable. src/cairo-path-fixed-private.h | 14 +-- src/cairo-path-fixed.c | 226 +++++++++++++++++++---------------------- src/cairo.c | 2 +- 3 files changed, 114 insertions(+), 128 deletions(-) commit a04e3726648c62a1385b67cfc16a785a468e1d13 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Jul 5 09:21:47 2009 +0100 [path] Add path watch debugging Simple debug macro to print the path to stderr during construction. src/cairo-path-fixed-private.h | 9 +++++++++ src/cairo-path-fixed.c | 26 ++++++++++++++++++++++++++ 2 files changed, 35 insertions(+) commit 1645352bfb69cc6d43fcbf3c77b4da5693af322a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Jul 5 09:09:42 2009 +0100 [path] Evaluate is_box && is_region during construction Whilst constructing the path, if the operations continue to be axis-aligned lines, allow the is_box and is_region flags to persist. These are set to false as soon as a curve-to is added, a diagonal or in the case of is_region a non-integer point. src/cairo-path-fixed-private.h | 8 ++ src/cairo-path-fixed.c | 286 +++++++++++++++-------------------------- src/cairo.c | 2 + src/cairoint.h | 3 - 4 files changed, 115 insertions(+), 184 deletions(-) commit 4bb157672ef8cb7d486d6ccdaaa3890e3f45e17a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jul 4 23:44:09 2009 +0100 [test] Rename some structures within cairo-test-trace Simply rename a few structures so that their names are less confusing with the mix of process and threads used to executes traces. test/cairo-test-trace.c | 224 ++++++++++++++++++++++++------------------------ 1 file changed, 112 insertions(+), 112 deletions(-) commit 53a54827f7d8ca8500f3f9674edbd7e768e8a50d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Jul 5 09:22:05 2009 +0100 [boilerplate] Add missing header file. Deleted so many, and forgot to add just this little one. Sigh. boilerplate/cairo-boilerplate-private.h | 49 +++++++++++++++++++++++++++++++++ 1 file changed, 49 insertions(+) commit ee8c50f2d6f1dba41f829c715b7bc59b5eeda50d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jul 4 22:00:09 2009 +0100 Remove the defunct test-meta surface Replace the internal test-meta surface will the freshly exported, real meta-surface. boilerplate/cairo-boilerplate-test-surfaces.c | 38 --- boilerplate/cairo-boilerplate.c | 35 +++ src/Makefile.sources | 2 - src/test-meta-surface.c | 358 -------------------------- src/test-meta-surface.h | 50 ---- util/cairo-trace/trace.c | 49 ++-- 6 files changed, 69 insertions(+), 463 deletions(-) commit 88cb69b10c66751f687c3745c8e9861b105de3a2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jul 4 21:43:27 2009 +0100 [boilerpate] Move target definition to backends. By moving the backend target definition out of the massive amlagamated block in cairo-boilerplate.c and into each of the cairo-boilerplate-backend.c, we make it much easier to add new targets as the information need only be entered in a single file and not scattered across three. However, updating the target interface means trawling across all the files -- except given that I found it difficult maintaining the single massive array I do not see this as an increase in the maintenance burden. boilerplate/.gitignore | 2 + boilerplate/Makefile.am | 12 + boilerplate/Makefile.sources | 27 +- boilerplate/cairo-boilerplate-beos-private.h | 40 -- boilerplate/cairo-boilerplate-beos.cpp | 48 +- boilerplate/cairo-boilerplate-directfb-private.h | 26 - boilerplate/cairo-boilerplate-directfb.c | 29 +- boilerplate/cairo-boilerplate-glitz-agl.c | 31 +- boilerplate/cairo-boilerplate-glitz-glx.c | 31 +- boilerplate/cairo-boilerplate-glitz-private.h | 88 --- boilerplate/cairo-boilerplate-glitz-wgl.c | 32 +- boilerplate/cairo-boilerplate-pdf-private.h | 60 -- boilerplate/cairo-boilerplate-pdf.c | 48 +- boilerplate/cairo-boilerplate-ps-private.h | 71 --- boilerplate/cairo-boilerplate-ps.c | 73 ++- boilerplate/cairo-boilerplate-qt-private.h | 60 -- boilerplate/cairo-boilerplate-qt.cpp | 29 +- boilerplate/cairo-boilerplate-quartz-private.h | 44 -- boilerplate/cairo-boilerplate-quartz.c | 29 +- boilerplate/cairo-boilerplate-script-private.h | 57 -- boilerplate/cairo-boilerplate-script.c | 28 +- boilerplate/cairo-boilerplate-svg-private.h | 72 --- boilerplate/cairo-boilerplate-svg.c | 70 ++- .../cairo-boilerplate-test-surfaces-private.h | 99 ---- boilerplate/cairo-boilerplate-test-surfaces.c | 117 +++- boilerplate/cairo-boilerplate-win32-printing.c | 43 +- boilerplate/cairo-boilerplate-win32-private.h | 65 --- boilerplate/cairo-boilerplate-win32.c | 28 +- boilerplate/cairo-boilerplate-xcb-private.h | 47 -- boilerplate/cairo-boilerplate-xcb.c | 28 +- boilerplate/cairo-boilerplate-xlib-private.h | 71 --- boilerplate/cairo-boilerplate-xlib.c | 207 ++++--- boilerplate/cairo-boilerplate.c | 639 +++------------------ boilerplate/cairo-boilerplate.h | 3 +- boilerplate/make-cairo-boilerplate-constructors.c | 156 +++++ test/make-cairo-test-constructors.c | 28 +- 36 files changed, 946 insertions(+), 1592 deletions(-) commit 8a10ab1c04298d6c22ae8aabec5d762141a8e98f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jul 4 18:46:21 2009 +0100 [script] Tidy dict construction for glyphs Replace dict with '<< ... >>' similar to surface definitions as it is easier to read in scripts. src/cairo-script-surface.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) commit fef3649953a3dcc3a17d9d5858f56690a8782501 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jul 4 18:32:57 2009 +0100 [script] Fix the trivial errors running the test suite A few typos crept in breaking the script output and various refreshes of the expected results. src/cairo-script-surface.c | 33 +++++++++++++++++++++---------- test/mime-data.script.ref.png | Bin 1982 -> 2041 bytes test/scale-offset-image.c | 2 +- test/scale-offset-similar.c | 2 +- util/cairo-script/cairo-script-scanner.c | 6 ++++-- 5 files changed, 29 insertions(+), 14 deletions(-) commit 2395bba1e66a2623cb288e13effc206f86e5e62e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jul 4 18:31:49 2009 +0100 [any2ppm] Choose a more appropriate format for the replay content Actually select a format that matches the request content when replaying the script to an image. test/any2ppm.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-) commit 9942a89870f7754e28ff334010c4432553cf7d91 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jul 4 18:30:24 2009 +0100 [boilerplate] Read whole RGB lines at a time Rather than read+unpack each pixel, read in a whole RGB packed line and unpack the whole line in a single pass. boilerplate/cairo-boilerplate.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) commit 23648e2fdfefba4df08bc854d725758209998e1f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jul 3 00:40:32 2009 +0100 [script] Prototypical binary translator Hook into the scanner to write out binary version of the tokenized objects -- note we bind executable names (i.e. check to see if is an operator and substitute the name with an operator -- this breaks overloading of operators by scripts). By converting scripts to a binary form, they are more compact and execute faster: firefox-world-map.trace 526850146 bytes bound.trace 275187755 bytes [ # ] backend test min(s) median(s) stddev. count [ 0] null bound 34.481 34.741 0.68% 3/3 [ 1] null firefox-world-map 89.635 89.716 0.19% 3/3 [ 0] drm bound 79.304 79.350 0.61% 3/3 [ 1] drm firefox-world-map 135.380 135.475 0.58% 3/3 [ 0] image bound 95.819 96.258 2.85% 3/3 [ 1] image firefox-world-map 156.889 156.935 1.36% 3/3 [ 0] xlib bound 539.130 550.220 1.40% 3/3 [ 1] xlib firefox-world-map 596.244 613.487 1.74% 3/3 This trace has a lot of complex paths and the use of binary floating point reduces the file size by about 50%, with a commensurate reduction in scan time and significant reduction in operator lookup overhead. Note that this test is still IO/CPU bound on my i915 with its pitifully slow flash... util/cairo-script/Makefile.am | 5 +- util/cairo-script/cairo-script-file.c | 44 +- util/cairo-script/cairo-script-interpreter.c | 85 ++- util/cairo-script/cairo-script-interpreter.h | 9 +- util/cairo-script/cairo-script-objects.c | 8 +- util/cairo-script/cairo-script-operators.c | 2 +- util/cairo-script/cairo-script-private.h | 27 +- util/cairo-script/cairo-script-scanner.c | 1059 ++++++++++++++++++-------- 8 files changed, 906 insertions(+), 333 deletions(-) commit 07c0c8c5c0820e93403623eb5c1fa2df8ca89c60 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jul 4 14:07:16 2009 +0100 [surface] Refactor meta-surface clone Whilst still obnoxious, at least moving it to a separate function is more readable. src/cairo-surface.c | 140 +++++++++++++++++++++++++++++----------------------- 1 file changed, 79 insertions(+), 61 deletions(-) commit 7903c80ee81777bab6eec408c10b2b59330b10f7 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jul 4 13:50:15 2009 +0100 [meta] Cache replays when used as a pattern source Use the cow-snapshotting mechanism to store the meta surface replay (either to an image inside acquire_source_image() or to a similar surface during clone_similar()). Fixes Bug 17971 -- Extreme slowdown for manual convolutions in most vector backends. https://bugs.freedesktop.org/show_bug.cgi?id=17971 src/cairo-meta-surface.c | 18 ++++++++-- src/cairo-surface.c | 92 ++++++++++++++++++++++++++++++++++++------------ src/cairoint.h | 3 +- 3 files changed, 88 insertions(+), 25 deletions(-) commit f353ba46a3b8c260f96588fd6bb8d07c2a09d01a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jul 4 13:49:35 2009 +0100 [meta] Copy extents on snapshotting. Need to copy across the extents as well when snapshotting. src/cairo-meta-surface.c | 1 + 1 file changed, 1 insertion(+) commit f4019be7bdff5c67d679f4de1070c25aa99a993d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jul 4 13:16:42 2009 +0100 [trace] Remove a few transient pattern def/undef It is easier on the eye to use '1 index set-source exch pop' rather than 'dup /p0 exch def p0 set-source /p0 undef' (as patterns are expected to be temporary so we strive to avoid naming them). util/cairo-trace/trace.c | 42 +++++++++++++++++++++++++++++++----------- 1 file changed, 31 insertions(+), 11 deletions(-) commit dcb7f1d034438c5c890490df93a86b40fea3036c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jul 4 08:08:25 2009 +0100 [script] Add missing include <string.h> As reported by Magnus Boman <captain.magnus@gmail.com>. Fixes Bug 22610 -- Patch to add missing include in util/cairo-script/csi-replay.c http://bugs.freedesktop.org/show_bug.cgi?id=22610 util/cairo-script/csi-replay.c | 1 + 1 file changed, 1 insertion(+) commit 97bd2228a6635bbb0eac4ce95801754957dbb7ba Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jul 3 19:41:51 2009 +0100 [script] Insert spaces around '<<' and '>>' The dictionary constructors are just ordinary tokens so need space for delimiters. src/cairo-script-surface.c | 62 ++++++++++++++++------------------------------ 1 file changed, 21 insertions(+), 41 deletions(-) commit 6003ab77e1ebefadb97338de0e7da4a76d973b1a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jul 3 18:26:50 2009 +0100 Export meta-surface The meta-surface is a vital tool to record a trace of drawing commands in-memory. As such it is used throughout cairo. The value of such a surface is immediately obvious and should be applicable for many applications. The first such case is by cairo-test-trace which wants to record the entire graph of drawing commands that affect a surface in the event of a failure. NEWS | 27 +++++ boilerplate/Makefile.win32.features | 8 ++ boilerplate/cairo-boilerplate.c | 12 +-- build/Makefile.win32.features-h | 1 + build/configure.ac.features | 1 + configure.ac | 4 + perf/cairo-perf.c | 6 +- src/Makefile.win32.features | 8 ++ src/cairo-meta-surface-private.h | 14 +-- src/cairo-meta-surface.c | 202 +++++++++++++++++++++++++++--------- src/cairo-paginated-surface.c | 16 +-- src/cairo-pdf-surface.c | 2 +- src/cairo-ps-surface.c | 2 +- src/cairo-script-surface.c | 10 +- src/cairo-surface.c | 2 +- src/cairo-svg-surface.c | 8 +- src/cairo-type3-glyph-surface.c | 8 +- src/cairo-types-private.h | 3 +- src/cairo-user-font.c | 7 +- src/cairo-win32-printing-surface.c | 2 +- src/cairo.h | 22 +++- src/test-meta-surface.c | 16 +-- src/test-meta-surface.h | 8 +- test/cairo-test-trace.c | 58 ++++------- test/get-clip.c | 5 + 25 files changed, 296 insertions(+), 156 deletions(-) commit 2a9903dbbfeb3fe843f0d618d15674b37a29f3a9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jul 3 12:48:40 2009 +0100 [script] Trim some excess verbage Remove some redundant defining of surfaces and contexts and of setting defaults. In order to reduce the number of defines, we need to operate on the operand stack more frequently - though in practice those operations are quite rare. src/Makefile.sources | 1 + src/cairo-list-private.h | 189 +++++++++++++++ src/cairo-script-surface.c | 561 ++++++++++++++++++++++++++++++--------------- 3 files changed, 572 insertions(+), 179 deletions(-) commit 28773fe79f883c32ee4e55e49f5dc3fced00c4f6 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jul 3 13:17:47 2009 +0100 [test-meta] Initialize image Oops, stupid mistake - missed off the initialization of the image to NULL from the last commit. src/test-meta-surface.c | 1 + 1 file changed, 1 insertion(+) commit ac6c3dbbaed377273b73da6e9e03bb8380a78ec4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jul 3 12:56:20 2009 +0100 [test-meta] Lazily allocate image Only allocate the image on first use, as we may never actually need to construct the image output. src/test-meta-surface.c | 42 ++++++++++++++++++++++++------------------ 1 file changed, 24 insertions(+), 18 deletions(-) commit fe73a9dd1413a851beeca7a0fc4ab7cb351ea9ea Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jul 3 11:24:42 2009 +0100 [test] Record trace to an in-memory meta-surface Requires hooking into test-meta-surface currently. Export meta-surface! The idea is that on detection of an error, we can reconstruct a minimal trace from the meta-surface. The first step is to simply dump the trace for the failing meta-surface. Later, we should automatically minimise this further. src/test-meta-surface.c | 101 +++++++------ src/test-meta-surface.h | 4 + test/Makefile.am | 3 + test/cairo-test-trace.c | 371 ++++++++++++++++++++++++++++++++++++++++++------ 4 files changed, 397 insertions(+), 82 deletions(-) commit 633efe81873168506ec3594eae1f6ac38d71026c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jul 2 09:17:43 2009 +0100 [surface] Allocate temporary region on stack Avoid a very short-lived temporary heap allocation of a region. src/cairo-surface.c | 13 +++++-------- 1 file changed, 5 insertions(+), 8 deletions(-) commit b2e2a2f31af33ff01d78c6c8786249a623017ffa Author: Carlos Garcia Campos <carlosgc@gnome.org> Date: Tue Jun 30 15:47:51 2009 +0200 [test] Update argb32 pdf reference image for trap-clip test/trap-clip.pdf.argb32.ref.png | Bin 6720 -> 6745 bytes 1 file changed, 0 insertions(+), 0 deletions(-) commit 910afc7a8e5c40f4fe682ec0b0f3836a07ea721d Author: Carlos Garcia Campos <carlosgc@gnome.org> Date: Tue Jun 30 15:40:24 2009 +0200 [test] Mark all smask tests as XFAIL for pdf Those tests use linear gradients which are not correctly handled by poppler yet. See https://bugs.freedesktop.org/show_bug.cgi?id=10942 test/README | 11 +++++++++++ test/smask-fill.c | 2 +- test/smask-image-mask.c | 2 +- test/smask-mask.c | 2 +- test/smask-paint.c | 2 +- test/smask-stroke.c | 2 +- test/smask-text.c | 2 +- test/smask.c | 2 +- 8 files changed, 18 insertions(+), 7 deletions(-) commit 5becfc2bd59956090daa8afd68db2f139dcb8392 Author: Carlos Garcia Campos <carlosgc@gnome.org> Date: Tue Jun 30 14:58:40 2009 +0200 [test] Add pdf reference image for rotate-image-surface-paint Remove also argb32 and rgb24 pdf reference images. test/Makefile.am | 3 +-- test/rotate-image-surface-paint.pdf.argb32.ref.png | Bin 215 -> 0 bytes test/rotate-image-surface-paint.pdf.ref.png | Bin 0 -> 196 bytes test/rotate-image-surface-paint.pdf.rgb24.ref.png | Bin 215 -> 0 bytes 4 files changed, 1 insertion(+), 2 deletions(-) commit f81203cde864af1a1fea81a438e4f4effe928808 Author: Carlos Garcia Campos <carlosgc@gnome.org> Date: Tue Jun 30 14:44:19 2009 +0200 [test] Add pdf reference image for mask-transformed-image test/Makefile.am | 1 + test/mask-transformed-image.pdf.ref.png | Bin 0 -> 3528 bytes 2 files changed, 1 insertion(+) commit 9d538ef337e9ca1600f842cd2164ff5e6629090a Author: Carlos Garcia Campos <carlosgc@gnome.org> Date: Tue Jun 30 14:42:32 2009 +0200 [test] Update argb32 and rgb24 pdf reference image for mask test/mask.pdf.argb32.ref.png | Bin 9903 -> 9918 bytes test/mask.pdf.rgb24.ref.png | Bin 8735 -> 8745 bytes 2 files changed, 0 insertions(+), 0 deletions(-) commit 97def877bb1c290735875061118bd795b9b83215 Author: Carlos Garcia Campos <carlosgc@gnome.org> Date: Tue Jun 30 14:18:30 2009 +0200 [test] Update pdf reference image for surface-pattern Remove also argb32 and rgb24 pdf reference images. test/Makefile.am | 2 -- test/surface-pattern.pdf.argb32.ref.png | Bin 14808 -> 0 bytes test/surface-pattern.pdf.ref.png | Bin 14791 -> 14800 bytes test/surface-pattern.pdf.rgb24.ref.png | Bin 14808 -> 0 bytes 4 files changed, 2 deletions(-) commit 8a6bdc63c1491dc77e8d9e192b9e44d9ff2e3c9b Author: Carlos Garcia Campos <carlosgc@gnome.org> Date: Tue Jun 30 13:33:07 2009 +0200 [test] Update argb32 and rgb24 pdf reference image for text-pattern test/text-pattern.pdf.argb32.ref.png | Bin 2151 -> 2334 bytes test/text-pattern.pdf.rgb24.ref.png | Bin 508 -> 1915 bytes 2 files changed, 0 insertions(+), 0 deletions(-) commit b46ba7aa33ba604628f1c94449019895b1d7ef82 Author: Carlos Garcia Campos <carlosgc@gnome.org> Date: Tue Jun 30 13:25:39 2009 +0200 [test] Update pdf reference image for text-transform test/text-transform.pdf.ref.png | Bin 3796 -> 5586 bytes 1 file changed, 0 insertions(+), 0 deletions(-) commit 9bc7ab980d6e834dacaf4f48f199e8c95cfba94e Author: Carlos Garcia Campos <carlosgc@gnome.org> Date: Tue Jun 30 12:09:50 2009 +0200 [test] Update pdf reference image for filter-nearest-transformed Remove also argb32 and rgb24 pdf reference images. test/Makefile.am | 2 -- test/filter-nearest-transformed.pdf.argb32.ref.png | Bin 488 -> 0 bytes test/filter-nearest-transformed.pdf.ref.png | Bin 548 -> 552 bytes test/filter-nearest-transformed.pdf.rgb24.ref.png | Bin 488 -> 0 bytes 4 files changed, 2 deletions(-) commit 27938f7e3ff357e5fe8a05e547cf83dbad7551ef Author: Carlos Garcia Campos <carlosgc@gnome.org> Date: Tue Jun 30 12:04:06 2009 +0200 [test] Update pdf reference image for filter-nearest-offset Remove also argb32 and rgb24 pdf reference images. test/Makefile.am | 2 -- test/filter-nearest-offset.pdf.argb32.ref.png | Bin 4295 -> 0 bytes test/filter-nearest-offset.pdf.ref.png | Bin 4537 -> 997 bytes test/filter-nearest-offset.pdf.rgb24.ref.png | Bin 4295 -> 0 bytes 4 files changed, 2 deletions(-) commit e6576e8912196607fdd3613c6924e26c2ce7a951 Author: Carlos Garcia Campos <carlosgc@gnome.org> Date: Tue Jun 30 11:56:18 2009 +0200 [test] Update argb32 pdf reference image for clip-operator test/clip-operator.pdf.argb32.ref.png | Bin 11604 -> 11603 bytes 1 file changed, 0 insertions(+), 0 deletions(-) commit 13d9374ec13c62233312beb0feb9a47531aec2e3 Author: Carlos Garcia Campos <carlosgc@gnome.org> Date: Mon Jun 29 17:36:47 2009 +0200 [test] Remove PDF reference images for scale-source-surface-paint It works with current poppler now test/Makefile.am | 2 -- test/scale-source-surface-paint.pdf.argb32.ref.png | Bin 182 -> 0 bytes test/scale-source-surface-paint.pdf.rgb24.ref.png | Bin 198 -> 0 bytes 3 files changed, 2 deletions(-) commit 3710c4184c2e87368dde5ccb25f3648b861b6059 Author: Carlos Garcia Campos <carlosgc@gnome.org> Date: Mon Jun 29 17:32:54 2009 +0200 [test] Remove PDF reference images for paint-with-alpha It works with current poppler now test/Makefile.am | 1 - test/paint-with-alpha.pdf.ref.png | Bin 466 -> 0 bytes 2 files changed, 1 deletion(-) commit 8162acd81a898247f19054a136af95062e3b0e91 Author: Carlos Garcia Campos <carlosgc@gnome.org> Date: Mon Jun 29 12:42:52 2009 +0200 [test] Remove PDF reference images for paint-source-alpha It works with current poppler now test/Makefile.am | 2 -- test/paint-source-alpha.pdf.argb32.ref.png | Bin 471 -> 0 bytes test/paint-source-alpha.pdf.ref.png | Bin 473 -> 0 bytes 3 files changed, 2 deletions(-) commit 1e1f4fd0920cb2b0882137b4cc500dcbae42d2dc Author: Carlos Garcia Campos <carlosgc@gnome.org> Date: Tue Jun 30 11:05:50 2009 +0200 [test] Use poppler_page_render with a transparent surface Instead of using an opaque surface filled in white, use a transparent surface and fill it in white by using CAIRO_OPERATOR_DEST_OVER after rendering. test/any2ppm.c | 9 +++++---- test/pdf2png.c | 9 +++++---- 2 files changed, 10 insertions(+), 8 deletions(-) commit 700a555d194a20c00df69b963cdc5b6153e162a4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jul 1 19:41:42 2009 +0100 [surface] Expose a SURFACE_TYPE_MISMATCH nil object Gah, yet more error object bloat. Must finish cairo-object. src/cairo-surface.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) commit 8f4611a94bda407b5944520b95599441a4ff5327 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jul 1 19:39:57 2009 +0100 [perf] Fix use-after-free when retrieving error line number We find out the status on destroying the script and then attempt to query the defunct script for more info about the error. Wrong. perf/cairo-perf-trace.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) commit 7c160dea5456b783d9b04b5c11bbacb9a89fa031 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jul 1 10:12:50 2009 +0100 [perf] Unbreak cairo-perf-diff builder perf/cairo-perf-diff | 19 ++++++++++--------- 1 file changed, 10 insertions(+), 9 deletions(-) commit 3d88f99668308315db759af3dac789973caa8eae Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jul 1 09:57:56 2009 +0100 [perf] Enable cairo-perf-trace in cairo-perf-diff Only as a --trace option for now. perf/cairo-perf-diff | 51 ++++++++++++++++++++++++++++++++++++--------------- 1 file changed, 36 insertions(+), 15 deletions(-) commit 1564d67d63618e12c58dc1d65c7e68505950e4d4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jun 30 22:22:39 2009 +0100 [perf] Exclude the xlib-reference target from cairo-perf perf/cairo-perf-trace.c | 3 ++- perf/cairo-perf.c | 3 ++- 2 files changed, 4 insertions(+), 2 deletions(-) commit e5b4f78bff20903fef5f55a0fcd026590ea444a4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jun 30 22:12:50 2009 +0100 [boilerplate] Only include null-test-surface >= cairo-1.9.3 boilerplate/cairo-boilerplate-test-surfaces.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) commit e0bdbb437ee699d6f75df34494a6e290dc55aa39 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jun 30 22:04:01 2009 +0100 [xlib] Remove redundant checks for src_picture==None Before attempting to even set the attributes on the source Picture, we ensure that it exists. So remove the redundant safe-guards to do nothing if it doesn't exist. src/cairo-xlib-surface.c | 12 +----------- 1 file changed, 1 insertion(+), 11 deletions(-) commit b7f68d3518e25fb02c9d36d14de9dc0c905ffe8f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jun 30 20:46:12 2009 +0100 [perf] Enable trace directory recursion. I have an idea to categorise traces within their own subdirectories and so for convenience added path walking to cairo-perf-trace. Principally this should allow for forests of symlinks of all types. perf/cairo-perf-trace.c | 104 +++++++++++++++++++++++++++++++----------------- perf/cairo-perf.h | 1 + 2 files changed, 68 insertions(+), 37 deletions(-) commit 6102dd0c1c26e289ad08d1cf0cc366e6298f5711 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jun 29 10:50:44 2009 +0100 [ft] Correct reference counting for implementation font face We are expected to return a new reference from get_implementation(), and failing to do so was causing use-after-frees (detected by the reference counting checks). src/cairo-ft-font.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) commit 4f745fbe3329a4fd5164c7bd7fe61c0395e8d1ee Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jun 29 10:49:51 2009 +0100 [test] Only run fallback-resolution for enabled targets A secondary check that both activates CAIRO_TEST_TARGET and prevents a NULL-deref with the null test surface. test/fallback-resolution.c | 3 +++ 1 file changed, 3 insertions(+) commit d80f5580ef371a1d4b1746b4ad54d97274d2d6cd Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Jun 28 21:11:41 2009 +0100 [pattern] Mark _cairo_pattern_nil as static Fixes check-def.sh failure. src/cairo-pattern.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 758a18b354cefca841536c7d91f8f5758456f72f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Jun 28 20:56:33 2009 +0100 [test] Timeout support for tests Enforce that each test must render within 60 seconds or be considered to have hit an infinite loop and be reported as a CRASH. The timeout value is adjustable via CAIRO_TEST_TIMEOUT -- a value of 0 will disable. build/configure.ac.system | 3 +++ test/cairo-test.c | 18 ++++++++++++++++++ test/cairo-test.h | 2 ++ 3 files changed, 23 insertions(+) commit f7021d8f3e59d1f7fa77135366bbbd6845dff684 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Jun 28 20:41:04 2009 +0100 [script] Enable error handling for recursive scanners It's conceivable that a script could execute another file and so we should only setjmp on the first invocation. util/cairo-script/cairo-script-private.h | 1 + util/cairo-script/cairo-script-scanner.c | 15 ++++++++++++--- 2 files changed, 13 insertions(+), 3 deletions(-) commit 963664727b4754f92a876c36557922821706f591 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Jun 28 10:49:37 2009 +0100 [trace] Compile fix for --enable-script Silly typo as pointed out by Hans Breuer <hans@breuer.org>. util/cairo-trace/trace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 2ecafb025c8b15a86372b655e763d742a0f6ef85 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Jun 28 10:09:01 2009 +0100 [script] Reuse glyph advance cache over multiple invocations Store the metric cache as user-data on the scaled-font so that we can retrieve it on future calls to show-glyphs and friends. util/cairo-script/cairo-script-operators.c | 413 +++++++++-------------------- 1 file changed, 131 insertions(+), 282 deletions(-) commit 663da31bbe7cf6021c8d59a4d4ddd2702cfe860f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Jun 28 01:17:13 2009 +0100 [boilerplate] Propagate original error from create_from_png() During cairo_boilerplate_get_image_surface_from_png() the original status returned by cairo_image_surface_create_from_png() could have been masked by inappropriate context construction when flattening or extracting. Simply check after creating the surface from the png file, and return immediately if in error. boilerplate/cairo-boilerplate.c | 2 ++ 1 file changed, 2 insertions(+) commit 2bbb69420289aa372512602db312611369e70f54 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Jun 28 00:57:10 2009 +0100 [test] Disable fallback16 testing Only use the RGB16_565 surface as a source during testing. We presume that pixman is itself rigorously tested and the output is equivalent to the image surface. Enabling this test surface would require too large of a burden to maintain the reference images for the 16-bit backend. boilerplate/cairo-boilerplate.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 34ac8673b2d51edeb0b29a4949d7d16c3d07a790 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Jun 28 00:48:05 2009 +0100 [script] Use setjmp exception handling during scanning Save ~5% of overhead by using longjmp on error to terminate scanning instead of propagating the error, incurring lots of redundant error checks. util/cairo-script/cairo-script-interpreter.c | 14 +- util/cairo-script/cairo-script-objects.c | 46 +--- util/cairo-script/cairo-script-private.h | 51 ++-- util/cairo-script/cairo-script-scanner.c | 390 +++++++++++++-------------- 4 files changed, 221 insertions(+), 280 deletions(-) commit 2b1ec7a4d87399824dacb9803627c9322b12d894 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 27 22:48:19 2009 +0100 [cairo] Report true error via pattern from cairo_pop_group() Report the correct error via the returned pattern->status rather than simply returning the NO_MEMORY nil object. src/cairo-pattern.c | 4 ++-- src/cairo.c | 18 +++++++++++------- src/cairoint.h | 4 +++- 3 files changed, 16 insertions(+), 10 deletions(-) commit 5c3be3ece27b7cf6e4b1262ea18ac51b45c62389 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 27 22:40:15 2009 +0100 [cairo] Mark cr->status guard as unlikely For correct operation, the context should never be in error, so mark the error context as being unlikely to optimize the common path. src/cairo.c | 199 +++++++++++++++++++++++++++++------------------------------- 1 file changed, 97 insertions(+), 102 deletions(-) commit f3880283331f3bda374e24fd0e1de677d8d368a9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 27 20:18:18 2009 +0100 [gstate] Avoid temporary allocation for transformed patterns Observe that patterns are not altered during an operation and so we are safe to use the data from the original pattern without copying. (This is enforced through the declaration that the backends operate on constant patterns which are not allowed to be referenced or destroyed.) src/cairo-gstate.c | 184 ++++++++++++++++------------------------------------ src/cairo-pattern.c | 31 +++++++++ src/cairoint.h | 4 ++ 3 files changed, 92 insertions(+), 127 deletions(-) commit 3759c7b34b662ccdc0486c68798aca0a6438ab7e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 27 19:38:30 2009 +0100 [perf] Enable the null-backend for trace replays. perf/cairo-perf-trace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 8d4c704129132effa10fb9242239952c2f4b7721 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 27 19:34:20 2009 +0100 [ft] Cache the resolved font face Keep the resolved font face alive whilst the default FcConfig remains identical i.e. resolve the pattern again if the system configuration changes (e.g. a new font is added or the rules are updated). This should retain the benefits of lazily resolving font patterns whilst improving performance by reducing the number of calls to FcFontMatch(). src/cairo-ft-font.c | 39 ++++++++++++++++++++++++++++++++++----- 1 file changed, 34 insertions(+), 5 deletions(-) commit ddadd41a6c3d6a7cdaa07373b24325f59d3436e3 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 27 18:07:36 2009 +0100 [test] Remove broken PDF reference images for push-group More XFAIL reference images that are now invalidated by changes in poppler. test/Makefile.am | 3 --- test/push-group.pdf.argb32.ref.png | Bin 2722 -> 0 bytes test/push-group.pdf.ref.png | Bin 2807 -> 0 bytes test/push-group.pdf.rgb24.ref.png | Bin 2740 -> 0 bytes 4 files changed, 3 deletions(-) commit 4d233fa4a6a16d400ba959f4c1c5d84fdd26d8fc Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 27 18:04:37 2009 +0100 [test] Remove invalid smask PDF reference images These need to be recorded using an XFAIL reference. In the meantime, as the poppler output has changed, just remove all the invalid PDF reference images. test/Makefile.am | 19 ------------------- test/smask-fill.pdf.argb32.ref.png | Bin 1909 -> 0 bytes test/smask-fill.pdf.ref.png | Bin 1075 -> 0 bytes test/smask-fill.pdf.rgb24.ref.png | Bin 1909 -> 0 bytes test/smask-image-mask.pdf.argb32.ref.png | Bin 1651 -> 0 bytes test/smask-image-mask.pdf.ref.png | Bin 421 -> 0 bytes test/smask-image-mask.pdf.rgb24.ref.png | Bin 1651 -> 0 bytes test/smask-mask.pdf.argb32.ref.png | Bin 4398 -> 0 bytes test/smask-mask.pdf.ref.png | Bin 3731 -> 0 bytes test/smask-mask.pdf.rgb24.ref.png | Bin 4398 -> 0 bytes test/smask-paint.pdf.argb32.ref.png | Bin 4496 -> 0 bytes test/smask-paint.pdf.ref.png | Bin 3800 -> 0 bytes test/smask-paint.pdf.rgb24.ref.png | Bin 4496 -> 0 bytes test/smask-stroke.pdf.argb32.ref.png | Bin 1417 -> 0 bytes test/smask-stroke.pdf.ref.png | Bin 449 -> 0 bytes test/smask-stroke.pdf.rgb24.ref.png | Bin 1417 -> 0 bytes test/smask-text.pdf.ref.png | Bin 1806 -> 0 bytes test/smask.pdf.argb32.ref.png | Bin 4496 -> 0 bytes test/smask.pdf.ref.png | Bin 3800 -> 0 bytes test/smask.pdf.rgb24.ref.png | Bin 4496 -> 0 bytes 20 files changed, 19 deletions(-) commit 86624627e4b196a78cfd9bd642299441e838e0f2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 27 17:53:18 2009 +0100 [test] Add group-unaligned Test case for: Bug 22441 -- Unexpected shift with push_group and pop_group https://bugs.freedesktop.org/show_bug.cgi?id=22441 This is a test that demonstrates the error in the pdf backend when using groups on surfaces with non-integer sizes. In order to create such a surface, we need to update the boilerplate to use doubles instead of integers when specifying the surface size. boilerplate/cairo-boilerplate-beos-private.h | 16 +++--- boilerplate/cairo-boilerplate-beos.cpp | 8 +-- boilerplate/cairo-boilerplate-directfb-private.h | 8 +-- boilerplate/cairo-boilerplate-directfb.c | 8 +-- boilerplate/cairo-boilerplate-glitz-agl.c | 8 +-- boilerplate/cairo-boilerplate-glitz-glx.c | 10 ++-- boilerplate/cairo-boilerplate-glitz-private.h | 24 ++++----- boilerplate/cairo-boilerplate-glitz-wgl.c | 8 +-- boilerplate/cairo-boilerplate-pdf-private.h | 8 +-- boilerplate/cairo-boilerplate-pdf.c | 14 ++--- boilerplate/cairo-boilerplate-ps-private.h | 16 +++--- boilerplate/cairo-boilerplate-ps.c | 30 +++++------ boilerplate/cairo-boilerplate-qt-private.h | 10 ++-- boilerplate/cairo-boilerplate-qt.cpp | 8 +-- boilerplate/cairo-boilerplate-quartz-private.h | 8 +-- boilerplate/cairo-boilerplate-quartz.c | 8 +-- boilerplate/cairo-boilerplate-script-private.h | 8 +-- boilerplate/cairo-boilerplate-script.c | 12 ++--- boilerplate/cairo-boilerplate-svg-private.h | 16 +++--- boilerplate/cairo-boilerplate-svg.c | 30 +++++------ .../cairo-boilerplate-test-surfaces-private.h | 40 +++++++------- boilerplate/cairo-boilerplate-test-surfaces.c | 40 +++++++------- boilerplate/cairo-boilerplate-win32-printing.c | 8 +-- boilerplate/cairo-boilerplate-win32-private.h | 16 +++--- boilerplate/cairo-boilerplate-win32.c | 8 +-- boilerplate/cairo-boilerplate-xcb-private.h | 8 +-- boilerplate/cairo-boilerplate-xcb.c | 8 +-- boilerplate/cairo-boilerplate-xlib-private.h | 24 ++++----- boilerplate/cairo-boilerplate-xlib.c | 45 +++++++++------- boilerplate/cairo-boilerplate.c | 10 ++-- boilerplate/cairo-boilerplate.h | 8 +-- test/Makefile.am | 7 +++ test/Makefile.sources | 1 + test/cairo-test.c | 2 +- test/cairo-test.h | 4 +- test/group-unaligned.c | 60 +++++++++++++++++++++ test/group-unaligned.pdf.ref.png | Bin 0 -> 414 bytes test/group-unaligned.ps.ref.png | Bin 0 -> 332 bytes test/group-unaligned.svg.argb32.ref.png | Bin 0 -> 524 bytes test/group-unaligned.svg.rgb24.ref.png | Bin 0 -> 434 bytes test/group-unaligned.xlib-fallback.ref.png | Bin 0 -> 496 bytes test/group-unaligned.xlib.ref.png | Bin 0 -> 466 bytes 42 files changed, 312 insertions(+), 235 deletions(-) commit c506ddb58612c69b80b08be3a6d14c5bcc1ac21f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 27 16:54:37 2009 +0100 [pdf] Restore original surface size after emitting group mask A trivial fix to reset the original surface size after emitting the group. The annoying aspect is that this should have been caught by the test suite. Alas, no. A gentle reminder that simple line coverage is insufficient. :( src/cairo-pdf-surface.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) commit f8a2665840f12ae48798b0ffca49681e276823ba Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 26 07:22:49 2009 +0100 [cairo] Trivial code tidy src/cairo.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit c7cf5317f58b382c47ad2d1b4d8db3133b0c4d4b Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Wed Jun 24 23:33:44 2009 +0100 [test] Fix Makefile.am for Solaris make. Sun's make doesn't like conditional assignments to variables in dependencies in the form of "target: VAR+=VAL". This construct was used to force the test suite run in the foreground in the check-valgrind target. It is replaced by a more explicit set of the CAIRO_TEST_MODE environment variable. test/Makefile.am | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) commit b82e7740c102afa7b9184b2556b887cf7a0211b0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Jun 21 22:25:08 2009 +0100 [atomic] Silence compiler warnings by defining an intptr_t Evaulate the integer sizes during configure to find one of the exact same size as a void* to use in the conversion of the atomic ptr cmpxchg to an atomic int cmpxchg. build/configure.ac.system | 3 +++ src/cairo-atomic-private.h | 19 ++++++++++++------- 2 files changed, 15 insertions(+), 7 deletions(-) commit a2d4fb50092c0cbb31a5df1bca9111957238139b Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Sun Jun 21 14:09:10 2009 +0100 [test] Fix a typo in the _POSIX_SOURCE version number. I hope POSIX isn't around in 20000 AD! test/cairo-test.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 80990c7f729a666fc646182a36ffd311d7396a1a Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Sun Jun 21 14:02:46 2009 +0100 [atomic] Use an integer __sync_val_compare_and_swap() for pointer CAS. Fix an implicit pointer/integer cast in _cairo_atomic_ptr_cmpxchg() when building with LLVM/clang. The Intel synchronization primitives __sync_val_compare_and_swap() are only defined by Intel for types int, long, long long and their unsigned variants. This patch uses one of those for _cairo_atomic_ptr_cmpxchg() instead of relying on a gcc extension of __sync_val_compare_and_swap() to pointer types. src/cairo-atomic-private.h | 9 ++++++++- src/cairo-atomic.c | 6 +++++- 2 files changed, 13 insertions(+), 2 deletions(-) commit d0c1c9282164bffb76f7419873e93608dc6d9876 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Jun 21 10:08:21 2009 +0100 [boilerplate] Implement a reference xlib surface Create an xlib target that uses a separate display, specified by CAIRO_REFERENCE_DISPLAY, that can be used as a reference implementation for the normal display. So this can be used by cairo-trace-test to create a reference surface that mimics exactly the true surface, but if it targets an Xvfb display will be entirely rendered in software. If CAIRO_REFERENCE_DISPLAY is not set, the reference surface degrades into an xlib-fallback surface which is a close approximation (but still has behavioural differences). boilerplate/cairo-boilerplate-xlib-private.h | 11 ++++++ boilerplate/cairo-boilerplate-xlib.c | 51 ++++++++++++++++++++++++++++ boilerplate/cairo-boilerplate.c | 14 ++++++-- 3 files changed, 74 insertions(+), 2 deletions(-) commit d2244053dcca687a878b1008e3a8219f11d7bd0d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Jun 21 08:58:08 2009 +0100 [perf] Force rebuild if make fails The build system has a singular failure whereby if a backend disappears between on compile and the next, automake will fail to reconstruct the Makefiles - resulting in a broken build. Attempt to fix this by removing the build dir and recloning, which should work for any corrupt caches but obviously will fail again at a true build failure. perf/cairo-perf-diff | 63 ++++++++++++++++++++++++++++++---------------------- 1 file changed, 37 insertions(+), 26 deletions(-) commit ef0d307d549b9cc7a98cdce2ee4bae40d0fd0bd1 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Jun 21 08:43:55 2009 +0100 [perf] Need to version surface types As cairo-perf-diff will execute the current cairo-perf against historical revisions, any introduced api must be protect in order to compile on old versions. perf/cairo-perf-trace.c | 8 ++++++++ perf/cairo-perf.c | 6 ++++++ 2 files changed, 14 insertions(+) commit 44b72ba6d7bdcc5f4e431af178bc7c22f13e88e0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 20 13:09:46 2009 +0100 [perf] Add a non-build mode to graph Add --show to cairo-perf-graph just to graph perf files that currently exist and not build the missing cases. perf/cairo-perf-graph | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) commit d4dd6fcf3a4e0d7426657592ebb3d65a2ae0bc4a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 20 13:08:50 2009 +0100 [perf] Show speedup/slowdownn labels on graph Add labels show that it is clear which direction is faster on the graph. perf/cairo-perf-graph-widget.c | 166 +++++++++++++++++++++++++++++++++++++---- 1 file changed, 153 insertions(+), 13 deletions(-) commit b71b6ecb59bf83605bb2bb10e3f8acd7dd950242 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 19 21:21:02 2009 +0100 [test] Fix the image compare Oh, it's a bad sign when I can't even correctly compare a bunch of pixels. test/cairo-test-trace.c | 101 +++++++++++++++++++++++++++++++++++++++++------- 1 file changed, 86 insertions(+), 15 deletions(-) commit d544828934e479071fe5246fc0bdbbad33c097e8 Author: M Joonas Pihlaja <rowan@sal.math.ualberta.ca> Date: Sat Jun 20 06:16:51 2009 -0600 [build] Look for libpng.pc as well when configuring. The OpenBSD png package installs only libpng.pc and not any of libpng{13,12,10}.pc. configure.ac | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 5733d8d6c35e1a77513a49267a47f6caae68e829 Author: M Joonas Pihlaja <rowan@sal.math.ualberta.ca> Date: Sat Jun 20 06:11:44 2009 -0600 [build] Remove a level of quoting from CAIRO_PARSE_VERSION. The m4 that comes with OpenBSD gets confused by the extra quoting applied to the version number argument when building the bodies of the cairo_version_{major,minor,micro} macros using m4_bpatsubst(). Since we're making sure that the argument is always a number, we can safely use the unquoted \2 body rather than \[\2\]. acinclude.m4 | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 3bad3efaa2fa00275c1d27c95ccbf1fcdfdfef3e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 19 19:09:03 2009 +0100 [test] Add cairo-test-trace to .gitignore test/.gitignore | 1 + 1 file changed, 1 insertion(+) commit 62dc0ae46c26652da9a76b851bab594020020115 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 19 19:03:17 2009 +0100 [cairo] Offset the current path when pushing/popping groups. We need to translate the path in order to compensate for the device offset applied to the group surface when pushing and popping. (The path is transformed to device space on creation, and so needs recomputing for the new device.) src/cairo.c | 10 ++++++++++ 1 file changed, 10 insertions(+) commit 851c8839622619779c9169299e3efccd108d0395 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 19 19:00:39 2009 +0100 [test] Add group-clip Larry Ewing found an ancient and nasty bug whereby a path was not corrected whilst pushing and popping groups. AUTHORS | 1 + test/Makefile.am | 1 + test/Makefile.sources | 1 + test/group-clip.c | 57 ++++++++++++++++++++++++++++++++++++++++++++++++ test/group-clip.ref.png | Bin 0 -> 191 bytes 5 files changed, 60 insertions(+) commit 0a34abace781976b92287dd07daa2b41e168c4f6 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 19 18:40:43 2009 +0100 [test] Experiment with reference targets Specify another boilerplate target to use as the reference for this target. We then use this in cairo-test-trace in preference to using the image surface. Still not perfect, though the framework is improving. boilerplate/cairo-boilerplate.c | 254 ++++++++++++++++++++++------------------ boilerplate/cairo-boilerplate.h | 5 + test/Makefile.am | 6 +- test/cairo-test-trace.c | 229 ++++++++++++++++++++++++++++++------ 4 files changed, 341 insertions(+), 153 deletions(-) commit cec8579348fb39ca879285bebfb1ee65cbae502d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 19 16:09:26 2009 +0100 [xlib] Remove redundant code to query a xrender_format from a visual We always query an xrender_format for a Visual upon surface creation, so checking again in create_similar() is redundant. (It also interferes with disabling XRender...) src/cairo-xlib-surface.c | 9 ++------- 1 file changed, 2 insertions(+), 7 deletions(-) commit dd842a2bed1568000c9727fc35d4683db349e506 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 19 14:59:59 2009 +0100 [boilerplate] Missing static on pdf user data key. boilerplate/cairo-boilerplate-pdf.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 2da906f4764d5fe7cbb08ec238b01e2a03696899 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Fri Jun 19 18:44:23 2009 +0300 [script] Include unistd.h if we're twiddling file descriptors. Including sys/mman.h doesn't pull in unistd.h on Solaris 9 so we have to include it explicitly. util/cairo-script/cairo-script-operators.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) commit 1b6dbcbbebf18d6239fcc618f3389e9c0fc981f3 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Fri Jun 19 18:39:53 2009 +0300 [trace] Compile without fontconfig. Use CAIRO_HAS_FC_FONT to decide whether to export cairo_ft_font_face_create_for_pattern() or not. util/cairo-trace/trace.c | 2 ++ 1 file changed, 2 insertions(+) commit d9b5a98a78b712a2ad64408fa47ba6aa34e29f1f Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Fri Jun 19 18:34:34 2009 +0300 [build] Require fontconfig >= 2.2.95 if we're using it at all. We require at least fontconfig 2.2.95 when we're using fontconfig because we depend on the FcResultOutOfMemory error code introduced in that version. configure.ac | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit 84e587bbfad6586e41fac86c4d19005418115ac5 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 19 14:41:39 2009 +0100 [trace] Prevent a child process from writing to the same file After opening a specific file or fd for ourselves, reset the CAIRO_TRACE_FD to point to an invalid fd in order to prevent any child processes (who inherit our environment) from attempting to trace cairo calls. If we allow them to continue, then the two traces will intermix and be unreplayable. util/cairo-trace/trace.c | 7 +++++++ 1 file changed, 7 insertions(+) commit 18edea36ba6604e4cbdbda1ed56e6117b5768d94 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 19 14:14:32 2009 +0100 [perf] Report line of error during trace Query the number of new lines processed so far and report that on hitting an error. perf/cairo-perf-trace.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) commit 55721d380d5a75a9448f522f9ad48cd18a6c2f65 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 19 14:13:34 2009 +0100 [script] Report line number Count the number of newlines processed in the trace and provide an API for the user to query. util/cairo-script/cairo-script-interpreter.c | 6 ++++++ util/cairo-script/cairo-script-interpreter.h | 3 +++ util/cairo-script/cairo-script-private.h | 2 ++ util/cairo-script/cairo-script-scanner.c | 12 +++++++++--- 4 files changed, 20 insertions(+), 3 deletions(-) commit 030ef4ca0023166fcb82f0a0b1512eb31beb3385 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 19 11:56:14 2009 +0100 [trace] Change the threshold at which pixel data is immediately copied Embed the pixels for images less than 32*32 as this catches most icons which are frequently uploaded, but is still an unlikely size for a destination image surface. util/cairo-trace/trace.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) commit d3330d7beba0e0d2543e40a0e0e4d81ea9a78b45 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 19 11:15:30 2009 +0100 [ft] Substitute twin if there are no fonts Fixes bug 22356 -- Spurious "out of memory" error on system without fonts https://bugs.freedesktop.org/show_bug.cgi?id=22356 If FcFontMatch() fails, then it means that there are no fonts available on the system (or it may have been a malloc error, we have no way of telling). Instead of report NO_MEMORY and disabling all drawing, one of the rationales for including a builtin font was so that we could continue even in the face of this error and show *something* to the user. (This being a last resort (and especially important for demos!) and hopefully easier to diagnose than no output at all.) src/cairo-font-face-twin.c | 75 +++++++++++++++++++++++++++++++++++----------- src/cairo-ft-font.c | 6 +++- src/cairoint.h | 3 ++ 3 files changed, 66 insertions(+), 18 deletions(-) commit 6d8b353658da1ad973e3bda9e89799a7467d8182 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Fri Jun 19 02:37:19 2009 +0300 [spans] Squash a compiler warning when creating error objects. Handle the new CAIRO_STATUS_USER_FONT_NOT_IMPLEMENTED status code in the switch/case of the ..._create_in_error() functions for creating span renderers or scan converters. src/cairo-spans.c | 2 ++ 1 file changed, 2 insertions(+) commit 7651227d3dfe4ebf40e95a72106957a9ec51dd3e Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Fri Jun 19 02:03:39 2009 +0300 Revert "[test] Make the xlib-fallback use the image refs." This reverts commit 903b39c30448d62e2cbf9d075c5256a333bd5d8f. The xlib-fallback actually renders mostly via traps at the moment, so staying with xlib refs is the right thing to do. boilerplate/cairo-boilerplate.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 729457b174cb19f4e0750486caed9582c5708930 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Sat May 2 23:30:02 2009 +0300 [test] Add trap rendered ref images for fill-image. The fill code path for the xlib and fallback surfaces use trapezoids rather than spans, so there's lots of differences due to antialising differences at the edge of the fill. This patch adds a ref image for the xlib and test-fallback surfaces. test/fill-image.test-fallback.ref.png | Bin 0 -> 1462 bytes test/fill-image.xlib.ref.png | Bin 0 -> 1462 bytes 2 files changed, 0 insertions(+), 0 deletions(-) commit 8e7f667abb94262b63b4fd589f5ee06cb8974aaf Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Sat May 2 19:09:27 2009 +0300 [test] Mark scale-offset tests as XFAIL for image, xlib-fallback. The scale-offset tests trigger a loss-of-precision bug when converting double matrices from cairo to pixman format. http://lists.cairographics.org/archives/cairo/2008-November/015671.html test/Makefile.am | 2 ++ test/scale-offset-image.c | 2 +- test/scale-offset-similar.c | 2 +- 3 files changed, 4 insertions(+), 2 deletions(-) commit c12533b1358c9e5c1c3440efca4dc44ba2de9f6d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jun 18 17:46:14 2009 +0100 [pattern] Add _cairo_pattern_fini_snapshot Symmetric operation to _cairo_pattern_init_snapshot() this exists simply to break the circular reference between the meta-surface and snapshot-cow. src/cairo-meta-surface.c | 24 ++++++++++++------------ src/cairo-pattern.c | 17 +++++++++++++++++ src/cairoint.h | 3 +++ 3 files changed, 32 insertions(+), 12 deletions(-) commit 2213c447373b600c1583b2edd24775c5c031231c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jun 18 17:14:53 2009 +0100 [svg] Break circular snapshot reference by explicit finish The svg backend snapshots the meta surface which because of snapshot-cow creates a circular reference back to the creator. So in order to break the circular reference when we have finished with the snapshot, we need to call cairo_surface_finish() in addition to simply destroying the surface. src/cairo-svg-surface.c | 4 ++++ 1 file changed, 4 insertions(+) commit cd9eef1aad87558d3f2d47cfcfa2f5b9b254692c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jun 18 16:42:56 2009 +0100 [scaled-font] Check the MRU font before resolving the implementor This should hide most of the overhead of resolving a frequently used toy font. src/cairo-scaled-font.c | 23 +++++++++++++++-------- 1 file changed, 15 insertions(+), 8 deletions(-) commit 5ad64dfda26f97f243f13e9b766567a30fda67a9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jun 18 16:39:02 2009 +0100 [test] Add the sha1sum of the required fonts (Note, I think these are currently out-of-date... But it will serve as a useful reminder to verify and update them in future.) test/README | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) commit 3063f0f44c61869bc6bb07d46bf881f1558ac015 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jun 18 16:33:12 2009 +0100 [scaled-font-subset] Check for matching font-face implementors Catch toy and user fonts which may have different font-faces for their implementation than the original ones they were created with. src/cairo-scaled-font-subsets.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) commit 6e78c94615e407d72d4778d5558dc218e0402cbc Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jun 18 16:30:41 2009 +0100 [test] Move calls to the getters from out of the innermost loops. Tidy the code slightly by removing excess and unsightly calls to cairo_test_get_context() and cairo_get_scaled_font(). test/show-glyphs-many.c | 19 ++++++++++++------- 1 file changed, 12 insertions(+), 7 deletions(-) commit 3da32e35afa57533379c3de35459f23aef7c3d04 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jun 18 15:31:27 2009 +0100 [analysis] Restore nops for the analysis null surface Joonas reported that adding the extra routines to the null-surface as used by the analysis surface broke user-fonts. So create a separate null backend to be exported via the test-null surface. src/cairo-analysis-surface.c | 57 ++-------------- src/test-null-surface.c | 157 ++++++++++++++++++++++++++++++++++++++++++- 2 files changed, 161 insertions(+), 53 deletions(-) commit c0e01d9cd71bd958e1b31a03cea4c08a1bdf4926 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jun 18 14:32:53 2009 +0100 [xlib] Improve GC caching efficacy Shrink the overall size of the per-screen GC cache, but allow multiple GCs per depth, as it quite common to need up to two temporary GCs along some drawing paths. Decrease the number of GCs we obtain in total by returning clean (i.e. a GC without a clip set) back to the screen pool after use. Compensate for the increased number of put/get by performing the query using atomic operations where available. So overall we see a dramatic reduction on the numbers of XCreateGC and XFreeGC, of even greater benefit for RENDER-less servers. src/cairo-xlib-private.h | 11 +-- src/cairo-xlib-screen.c | 200 +++++++++++++++++++++++++++++++++++------------ src/cairo-xlib-surface.c | 58 +++++++++----- 3 files changed, 197 insertions(+), 72 deletions(-) commit 7f238f542441cc5912d14e5c6f9c49ffd0b83fad Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Thu Jun 18 17:13:13 2009 -0400 Use the scanline rasterizer in the win32 backend. Brings http://people.mozilla.com/~jmuizelaar/world-map.html from 8fps to 13fps in Vista on a Mac Mini src/cairo-win32-surface.c | 146 +++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 144 insertions(+), 2 deletions(-) commit d066154e62d6b4ecad15e0eef431aa2f4a56fecb Author: Jeff Muizelaar <jeff@infidigm.net> Date: Tue Jun 16 15:55:53 2009 -0400 Factor out _cairo_image_surface_span_render_row() This allows other backends use the same function for rendering rows. More specifically, I intend to use this with the win32-backend. src/cairo-image-surface.c | 36 +++++++++++++++++++++++------------- src/cairoint.h | 6 ++++++ 2 files changed, 29 insertions(+), 13 deletions(-) commit fd142d0a6e21c492624ca8bc5019d89b0d7c5457 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Thu Jun 18 16:14:50 2009 +0100 [shave] Robuster usage of the test shell built-in. Solaris 9 confuses shave, resulting in an empty variable. This patch papers over the resulting build failure from the test shell built-in being called with an empty argument which was unquoted. build/shave.in | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 9ecee571c12e4d1645d9db2929280c9b35b1aad4 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Thu Jun 18 16:09:31 2009 +0100 [test/build] Don't test Freetype font stuff without Fontconfig. Most Freetype hitting tests require Fontconfig to find them some fonts. test/Makefile.am | 2 ++ 1 file changed, 2 insertions(+) commit 742b1ad4b069653a99fc6bd4510f2b6abb6abf95 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Sun May 24 23:41:28 2009 +0100 [test] Fix checking of the fenv.h include. This adds a configure check for fenv.h and makes invalid-matrix.c check for it with HAVE_FENV_H instead of HAVE_FEDISABLEEXCEPT -- turns out Solaris doesn't have fedisableexcept(), but it does have feclearexcept(). The same issue appears on OSX and was fixed in ab86662ab499e1f29c0f8c4248771e730c281e3f. This patch adds some configure magic. build/configure.ac.system | 2 +- test/invalid-matrix.c | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) commit 75f1d11d93236b19d06515b8ed2baf612b0767ab Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Sun May 24 21:28:04 2009 +0300 [test] Fallback to HUGE_VAL in place of INFINITY in invalid-matrix.c. GCC 3.4.3 on OpenSolaris does not recognise INFINITY. Use HUGE_VAL instead for every compiler instead of just MSVC. test/invalid-matrix.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit d1994d1ac292cbc896cffd24094ddfa5f2b1e8a9 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Sun May 24 23:46:13 2009 +0300 [test/any2ppm] Daemonize without BSD's daemon(). Solaris libc doesn't provide daemon() so implement any2ppm daemon's detaching without it. test/Makefile.am | 2 +- test/any2ppm.c | 62 +++++++++++++++++++++++++++++++++++++++++++++++++++----- 2 files changed, 58 insertions(+), 6 deletions(-) commit aafff0b9528952fbbe9d04a70bf8c76ee701743f Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Sun May 24 23:50:03 2009 +0300 [boilerplate] Check for connect() in libsocket. Solaris requires -lsocket to be able to talk to the any2ppm daemon over unix domain sockets. boilerplate/Makefile.am | 2 ++ build/configure.ac.system | 4 ++++ 2 files changed, 6 insertions(+) commit c4c06931690226e12622a957bbb6d652134f78b0 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Sun May 24 16:14:33 2009 +0300 [test] Make test constructors without perl. Remove the dependency on perl when building the test suite by replacing the script which makes cairo-test-constructors.c with a C program. test/.gitignore | 1 + test/Makefile.am | 9 ++- test/Makefile.win32 | 7 +- test/make-cairo-test-constructors.c | 153 +++++++++++++++++++++++++++++++++++ test/make-cairo-test-constructors.pl | 18 ----- 5 files changed, 165 insertions(+), 23 deletions(-) commit cdd27a8d4cc3486867cff4ce81bb4e064eb74d58 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Sun May 24 16:19:31 2009 +0300 [test] Never use gcc __attribute__ magic for constructors. The configure test for __attribute__((constructor)) succeeds when compiling with Sun Studio 12 on OpenSolaris but the attribute is just ignored and has no actual effect. This causes the test suite to not run any tests at all. With this patch we revert to always using make-cairo-test-constructors.pl. build/configure.ac.system | 10 ---------- test/Makefile.am | 7 ++----- test/cairo-test-private.h | 2 -- test/cairo-test-runner.c | 2 -- test/cairo-test.h | 14 -------------- 5 files changed, 2 insertions(+), 33 deletions(-) commit 46eb56a99f72b364e3eabe46f9b2995ed0696bd3 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Thu Jun 18 16:42:53 2009 +0300 [cairo-surface] Handle the all clipped out case for regions. Explicitly handle a region clip which represents that the entire surface is clipped out by passing in a temporary empty region to the backend set_clip_region() method. Previously the passed in region may have been NULL even when clip->all_clipped = TRUE. Fixes a bug tickled by the clip-all test case which was brought to light by 394e139213e8f6692115c4c24818bfeb5e6d456a. src/cairo-surface.c | 34 +++++++++++++++++++++++++++++++--- 1 file changed, 31 insertions(+), 3 deletions(-) commit 002264d572841f06930869a6c87fa2c987346dd6 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jun 17 17:15:26 2009 +0100 [clip] Missing {} (coding style) Carl spotted that I'd slipped in a little CODING_STYLE violation. src/cairo-clip.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit bc87074185592eae367b6c091241957b94202fda Author: Nis Martensen <nis.martensen@web.de> Date: Tue Jun 16 23:27:20 2009 +0200 [path] Fix missing implicit move-to When cairo_curve_to happens to start a new subpath (e.g., after a call to cairo_new_sub_path()), it also needs to update the last_move_point. Otherwise the new current point after a close_path() will be at an unexpected position. Therefore, call _cairo_path_fixed_move_to() explicitly. AUTHORS | 1 + src/cairo-path-fixed.c | 6 +++--- 2 files changed, 4 insertions(+), 3 deletions(-) commit 3c5e57e694d1a8a0a48a66e0d61d8257c23a780e Author: Nis Martensen <nis.martensen@web.de> Date: Tue Jun 16 23:38:50 2009 +0200 [test] Verify current point position after subpath closure Make sure cairo_close_path always takes the current point where it should be. test/Makefile.am | 1 + test/Makefile.sources | 1 + test/close-path-current-point.c | 95 ++++++++++++++++++++++++++++++++++ test/close-path-current-point.ref.png | Bin 0 -> 1916 bytes 4 files changed, 97 insertions(+) commit 394e139213e8f6692115c4c24818bfeb5e6d456a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jun 17 14:46:53 2009 +0100 [clip] During _clip() limit the extracted traps to the current clip extents By applying a tight _cairo_traps_limit() we can reduce the amount of work we need to do when tessellating the path and extracting the trapezoids. src/cairo-clip.c | 39 +++++++++++++++++++++++++++++++++++---- 1 file changed, 35 insertions(+), 4 deletions(-) commit 650b85ec7721fb1302284e3ca4b7f4b72358abed Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jun 17 14:19:19 2009 +0100 [clip] Avoid copying region within _cairo_clip_intersect_region() Within _cairo_clip_intersect_region() we can simply assign the freshly extracted traps-region if we previously had no region set. src/cairo-clip.c | 12 +++--------- 1 file changed, 3 insertions(+), 9 deletions(-) commit 4bc89e224f74f24272c97dec93d70f0c4dd1963c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jun 16 15:13:46 2009 +0100 [NEWS] Add release notes from 1.8.8 NEWS | 69 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 68 insertions(+), 1 deletion(-) commit 116c93cb7fe8ad55457c53b4652e8d3117581158 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jun 16 11:36:36 2009 +0100 [cairo] Remove instance of 'return free()' Gah! This is a GCC-ism, so therefore it has to go. src/cairo.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit 22587f57bd5d1b4440d936cd4655a7e8fcebdf36 Author: Vladimir Vukicevic <vladimir@pobox.com> Date: Sun Jun 14 20:43:05 2009 +0100 Import Qt backend by Mozilla Written by Vladimir Vukicevic to enable integration with Qt embedded devices, this backend allows cairo code to target QPainter, and use it as a source for other cairo backends. This imports the sources from mozilla-central: http://mxr.mozilla.org/mozilla-central/find?text=&kind=text&string=cairo-qpainter renames them from cairo-qpainter to cairo-qt, and integrates the patch by Oleg Romashin: https://bugs.freedesktop.org/attachment.cgi?id=18953 And then attempts to restore 'make check' to full functionality. However: - C++ does not play well with the PLT symbol hiding, and leaks into the global namespace. 'make check' fails at check-plt.sh - Qt embeds a GUI into QApplication which it requires to construct any QPainter drawable, i.e. used by the boilerplate to create a cairo-qt surface, and this leaks fonts (cairo-ft-fonts no less) causing assertion failures that all cairo objects are accounted for upon destruction. [Updated by Chris Wilson] Acked-by: Jeff Muizelaar <jeff@infidigm.net> Acked-by: Carl Worth <cworth@cworth.org> boilerplate/Makefile.sources | 3 + boilerplate/Makefile.win32.features | 10 + boilerplate/cairo-boilerplate-qt-private.h | 60 + boilerplate/cairo-boilerplate-qt.cpp | 90 ++ boilerplate/cairo-boilerplate.c | 23 + boilerplate/cairo-boilerplate.h | 3 + build/Makefile.win32.features | 1 + build/Makefile.win32.features-h | 3 + build/configure.ac.features | 1 + configure.ac | 11 + perf/cairo-perf-trace.c | 1 + perf/cairo-perf.c | 3 +- src/Makefile.am | 2 +- src/Makefile.sources | 3 + src/Makefile.win32.features | 14 + src/cairo-debug.c | 42 + src/cairo-qt-surface.cpp | 1876 ++++++++++++++++++++++++++++ src/cairo-qt.h | 89 ++ src/cairo-xlib-surface.c | 6 + src/cairo.h | 4 +- src/check-def.sh | 7 +- src/check-preprocessor-syntax.sh | 3 +- 22 files changed, 2247 insertions(+), 8 deletions(-) commit 7d3881114add18d5934073d0b04755d343ea38c6 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jun 15 15:13:43 2009 +0100 [perf] Report errors during replays perf/cairo-perf-trace.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) commit 061e0477e9a894241806eeacbde2f3770b751f34 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jun 15 15:10:50 2009 +0100 [perf] Report iteration count Show number of discard results by showing the number of statistically valid samples out of total population count. perf/cairo-perf-trace.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) commit bab8af62e772844f08a244d99f705e318af3f58c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jun 15 15:01:34 2009 +0100 [region] Add PLT entry for cairo_region_create_rectangles() src/cairo-region.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) commit 6da9c410d4696ebba0fb563a858f1625bf3b1f9e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jun 15 12:14:51 2009 +0100 Missing header file for cairo-test-null-surface. src/test-null-surface.h | 46 ++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 46 insertions(+) commit e5727e20f52eb0308a8ad1c44a6eb4c7c65d7ff0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jun 15 11:58:58 2009 +0100 Expose _cairo_null_surface_create() via a test surface Using a null surface is a convenient method to measure the overhead of the performance testing framework, so export it although as a test-surface so that it will only be available in development builds and not pollute distributed libraries. .../cairo-boilerplate-test-surfaces-private.h | 10 ++++ boilerplate/cairo-boilerplate-test-surfaces.c | 16 +++++++ boilerplate/cairo-boilerplate.c | 9 ++++ perf/cairo-perf-trace.c | 6 +-- src/Makefile.sources | 2 + src/cairo-analysis-surface.c | 56 ++++++++++++++++++++-- src/test-null-surface.c | 48 +++++++++++++++++++ test/cairo-test.c | 14 ++++-- util/cairo-trace/trace.c | 26 ++++++++++ 9 files changed, 175 insertions(+), 12 deletions(-) commit f2057061c67236a79c8bb69b84e44e151109629a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Jun 14 09:10:08 2009 +0100 [trace] Add sticky bit to (system-wide) trace output dir Need to allow user programs to dump their traces into the common output directory, when using /etc/ld.so.preload to capture traces for the entire desktop. util/cairo-trace/Makefile.am | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit 3fe50a77ea21bcddb701a8640b5ecad9f6f5ab01 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 13 22:27:17 2009 +0100 [configure] Rephrase warning message for disabling core features. "strictly recommended" is an odd phrase and though the stern warning is appropriate as replacing a system library with a more limited version is likely to cause mayhem, we are but mere servants and should be reserved in our parlance. build/configure.ac.features | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) commit 5d57aeaa23566739ba01a0fac1c3b90ce27b28cd Author: Søren Sandmann Pedersen <sandmann@redhat.com> Date: Tue May 26 17:53:37 2009 -0400 Reinstate cairo_region_create_rectangles() cairo_region_union_rectangle() is linear in the number of rectangles in the region. There is no way to make it significantly faster without losing the ability to return errors synchronously, so a cairo_region_create_rectangles() is needed to avoid a large performance regression. src/cairo-region.c | 44 ++++++++++++++++++++++++++++++++++++++++++++ src/cairo-traps.c | 47 ++++++++++++++++++++++++++++------------------- src/cairo.h | 4 ++++ src/cairoint.h | 1 + 4 files changed, 77 insertions(+), 19 deletions(-) commit 216a2e24c7375e48677ce30f9fe8fb3d8bbdd10d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 13 21:40:05 2009 +0100 [test] Minor tweak to cairo-test-trace Cleanse the code of a couple of redundant pointer manipulations. test/cairo-test-trace.c | 16 +++++++++------- 1 file changed, 9 insertions(+), 7 deletions(-) commit 2ab4245c866bcefb755a33f7eaa6ef1277f82df1 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 13 21:33:09 2009 +0100 [test] Need SOURCE when copying image data As cairo-test-trace does not clear the image data before reuse, using the default OVER operator will cause differing results for each process when inadvertently alpha blending into the shared memory region. As we essentially want to just copy the source pixels, be explicit and set the SOURCE operator. test/cairo-test-trace.c | 1 + 1 file changed, 1 insertion(+) commit e003d665fe45905606a792e4a783f65fc7e84550 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 13 19:18:11 2009 +0100 [test] Disable compilation of cairo-test-trace without shm_open() Remember to only add cairo-test-trace to the list of EXTRA_PROGRAM so long as we can actually compile it. test/Makefile.am | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit c092136fe9e5fee71dfe02f945fe5676f43b2d05 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 13 18:59:17 2009 +0100 [configure] Check for shm_open() As needed for cairo-test-trace. build/configure.ac.system | 8 ++++++++ test/Makefile.am | 4 +++- 2 files changed, 11 insertions(+), 1 deletion(-) commit f298dd92ca0b789a3d2b591753c992e1673b2d87 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 13 18:49:16 2009 +0100 [test] Remove dlmalloc cairo-test-trace's shared memory allocation pattern is much simpler than anticipated as it allocates a bunch of images and then frees them all, and so only needs a simple linear allocator. test/Makefile.am | 1 - test/cairo-test-trace.c | 62 +- test/dlmalloc.c | 5099 ----------------------------------------------- 3 files changed, 11 insertions(+), 5151 deletions(-) commit 37f32f29518bdc1428c7ec137f212313e377486c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 13 18:31:35 2009 +0100 [util] Disable default build of font-view It's only a demonstration utility after a all and we do not need to incur the cost of relinking it everytime we modify the main library. util/Makefile.am | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 0eca1b17b14057ef88ae2df500679e62ffb50118 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 13 17:16:20 2009 +0100 [surface] Give ownership of the snapshot to the snapshotter. Previously the reference to the newly created snapshot was owned by the containing pattern. The consequence of this was that when the pattern was destroyed the snapshot was cleaned up which prevent reuse of the snapshot across multiple pages. Transferring ownership upon attachment of the snapshot to the target means that the snapshot stays in existence until the target itself is destroyed or modified *and* the containing pattern is consumed. Obvious in hindsight. src/cairo-surface.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-) commit a1b3392e60b6b4ff4b650f4d48d56568c1851245 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 13 16:37:43 2009 +0100 [surface] Mark cairo_surface_show_page() with begin-modification cairo_surface_show_page() clears the surface after presentation, so we need to discard the snapshots here. src/cairo-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 1798bdd322e0d25147195198bf0c5e9506c61a15 Author: Adrian Johnson <ajohnson@redneon.com> Date: Sat Jun 13 22:47:41 2009 +0930 Fix PDF pattern embedding The PDF snapshot cow patch was reusing a previously emitted surface pattern if the surface unique id matched the current surface. This resulted in incorrect output as the new pattern may have a different pattern matrix. This patch fixes the PDF backend to always emit a new pattern but re-use previously emitted image or metasurface XObjects. src/cairo-pdf-surface-private.h | 21 +- src/cairo-pdf-surface.c | 532 +++++++++++++++++++++++++++------------- 2 files changed, 372 insertions(+), 181 deletions(-) commit f115c5479ead934774fea99d74d780e2bb1681c4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 13 15:49:15 2009 +0100 [trace] Strip absolute path off name when profiling Carl spotted this last night, but I misinterpreted it as an old problem caused by the application changing its working directory before its first cairo call - thus causing cairo-trace to attempt to open a file in the new directory. Instead the problem was attempting to trace an executable with an absolute path, where we just tagged it with a .lzma extentsion and attempted to pipe the output there. Obviously this fails for the user profiling system binaries. So use basename to strip the leading path. util/cairo-trace/cairo-trace.in | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) commit c52eed6680dd13f6bc65dc4a26b6c8eab5a94a54 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 13 15:42:43 2009 +0100 [trace] Handle wrapping python python lazily loads libcairo.so and so it is not available via RTLD_NEXT, and we need to dlopen cairo ourselves. Similarly the linker is not able to resolve any naked function references and so we need to ensure that all of our own calls into the library are wrapped with DLCALL. util/cairo-trace/trace.c | 145 +++++++++++++++++++++++++++-------------------- 1 file changed, 83 insertions(+), 62 deletions(-) commit 606790a86083d31fa50ed5e0cf621eca56c16e33 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 13 14:07:22 2009 +0100 [script] Disable compilation of examples Now that cairo-perf-trace exists and is a far superior replay tool, use EXTRA_PROGRAM for the original replay tools. util/cairo-script/Makefile.am | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 31b7652e1fa2ab2bd9e9df7f5517b7d678faf6aa Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 13 12:19:51 2009 +0100 [test] Never, ever, blame a crashing test on external factors. test/cairo-test-runner.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 3dde883b779b81b95f420039c02b51b029311f78 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 13 10:13:20 2009 +0100 [test] Code review after sleep Review cairo-test-trace.c and rewrite parts to ease understanding and fix various bugs - such as failure to notice the slaves crashing and not releasing our shared memory after an interrupt. test/cairo-test-trace.c | 400 ++++++++++++++++++++++++++++-------------------- 1 file changed, 236 insertions(+), 164 deletions(-) commit 1f542965f0310aed493651b8ddc1e98a4192b977 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 13 08:24:22 2009 +0100 [perf] Mention cairo-perf-compare-backends in the README perf/README | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) commit 3b6e4018c69f88135f8d937724acf90adb003bad Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 12 22:19:56 2009 +0100 [test] Add cairo-test-trace The basic premise is that we feed the trace to multiple backends in parallel and compare the output at the end of each context (based on the premise that contexts demarcate expose events, or their logical equivalents) with that of the image[1] backend. Each backend is executed in a separate process, for robustness, with the image data residing in shared memory and synchronising over a socket. [1] Should be reference implementation, currently the image backend is considered to be the reference for all other backends. test/Makefile.am | 16 +- test/cairo-test-trace.c | 1225 ++++++++++++ test/dlmalloc.c | 5099 +++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 6339 insertions(+), 1 deletion(-) commit a41e1275d28317c88e29b7f142cfc835673c01f5 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 12 12:32:51 2009 +0100 [boilerplate] Make array of targets const. Protect the boilerplate targets from unexpected modifications. boilerplate/cairo-boilerplate.c | 8 ++++---- boilerplate/cairo-boilerplate.h | 4 ++-- perf/cairo-perf-trace.c | 6 +++--- perf/cairo-perf.c | 4 ++-- perf/cairo-perf.h | 4 ++-- test/cairo-test.h | 2 +- 6 files changed, 14 insertions(+), 14 deletions(-) commit 6aebde6598c629d667b06e20111106832d139c4d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 12 12:25:22 2009 +0100 [boilerplate] Export cairo_boilerplate_get_image_target() A routine to scan the compiled list of targets and return the entry corresponding to the image backend. boilerplate/cairo-boilerplate.c | 17 +++++++++++++++++ boilerplate/cairo-boilerplate.h | 3 +++ 2 files changed, 20 insertions(+) commit d008167b9bcd6a24199740183bbfcabcfdccdd63 Author: Carl Worth <cworth@cworth.org> Date: Fri Jun 12 12:08:22 2009 -0700 Makefile: Fix two typos preventing doc-publish from working It seems that everything up to this point in release-publish worked fine, so with these fixes, hopefully the next run of make release-publish will work without any snags. build/Makefile.am.releasing | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 98eb84cfbbf2cfdb0153c8e924099cef016a412a Author: Carl Worth <cworth@cworth.org> Date: Fri Jun 12 12:07:59 2009 -0700 Increment version to 1.9.3 after the 1.9.2 snapshot. cairo-version.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit e9b9d2a7c17ca4b2bc2991fdc4893aed850578db Author: Carl Worth <cworth@cworth.org> Date: Thu Jun 11 15:28:59 2009 -0700 Update version to 1.9.2 In preparation for the (long overdue) 1.9.2 snapshot today. cairo-version.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 8ac8c5b428d11b11ba9fb55264cd9091368f6acb Author: Carl Worth <cworth@cworth.org> Date: Fri Jun 12 11:49:52 2009 -0700 NEWS: Fix date and name of 1.9.2 snapshot It's a snapshot, not a release, and the broken REFERENCE_IMAGE list derailed me from finishing yesterday. NEWS | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 3f474555dd00b8585ae3db848cf46984d77bd16e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 12 11:19:50 2009 +0100 [test] Add jp2.jp2 to EXTRA_DIST Required by test/mime-data and so triggering failures during distcheck as it was not packaged. test/Makefile.am | 1 + 1 file changed, 1 insertion(+) commit 7f114b781f5c530d57530e5f76402e41cdabac6b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 12 11:15:16 2009 +0100 [doc] Workaround read-only tree during make distcheck This looks to be an ugly necessity to work-around the nasty issue that we currently gtkdoc expect to be run inside the source tree. I'm sure Behdad will be able to resolve this much more elegantly than this quick and fragile attempt. build/Makefile.am.gtk-doc | 11 ++++++----- build/configure.ac.features | 2 +- doc/public/Makefile.am | 5 +++-- 3 files changed, 10 insertions(+), 8 deletions(-) commit 1498c9c333d7579231ac44a91f00f0a49360702d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 12 07:38:00 2009 +0100 [script] Also mmap backing storage for the pattern descriptions These can be reasonably large and persist for long times due to the font holdover caches, so manually swap them out to save space on tiny swapless machines. util/cairo-script/cairo-script-operators.c | 14 +++++++++++++- 1 file changed, 13 insertions(+), 1 deletion(-) commit 80ba06803b50cc0f76039939f25372cbbe2e8d81 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 12 08:03:12 2009 +0100 [Makefile] Update list of reference images. test/Makefile.am | 1432 +++++++++++++++++++++++++++++++----------------------- 1 file changed, 818 insertions(+), 614 deletions(-) commit fb21a0da8730823ff092f8e298af3d5b4e8c90e4 Author: Carl Worth <cworth@cworth.org> Date: Thu Jun 11 15:28:02 2009 -0700 Update CLEANFILES for 'make distcheck' This is a standard part of the release ritual. test/Makefile.am | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) commit a4f023ef2eb041fecba5f3fd2d5a1594975ca211 Author: Carl Worth <cworth@cworth.org> Date: Thu Jun 11 14:51:14 2009 -0700 NEWS: Update date for snapshot planned for today. NEWS | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 82139b42919be1acf6bade733bb2ff489e50812d Author: Carl Worth <cworth@cworth.org> Date: Thu Jun 11 14:51:10 2009 -0700 NEWS: Explain bovine polaroids a bit more. It's a fun joke, but the COW snapshots feature is important enough that we should actually tell users what it is. NEWS | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) commit 4d5f66d5b1f243e58d43dc89568be9240262fd5c Author: Carl Worth <cworth@cworth.org> Date: Thu Jun 11 14:42:11 2009 -0700 NEWS: Cleanup in preparation for 1.9.2 snapshot Mostly just wrapping to a reasonable line length, and some minor wording changes. NEWS | 71 +++++++++++++++++++++++++++++++++++++++++++------------------------- 1 file changed, 45 insertions(+), 26 deletions(-) commit 19630f4b73a4fdfb9855ececd6614493e68d3b95 Author: Carl Worth <cworth@cworth.org> Date: Wed Jun 10 17:39:48 2009 -0700 Make the warning about the need to --enable-gtk-doc more explicit. I must have read it three times and kept thinking, "but I *did* just install gtk-doc", before I finally understood what it was trying to tell me. build/Makefile.am.gtk-doc | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 3aaea453b0f0880d6c8825be21e67821732d1635 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jun 11 22:36:01 2009 +0100 [script] Store fonts inside mmaped files. As an aide to tiny swapless systems write the rarely used bytes that define type42 fonts into a deleted file and mmap them back into our address space. build/configure.ac.system | 3 + util/cairo-script/cairo-script-operators.c | 91 ++++++++++++++++++++++++------ 2 files changed, 78 insertions(+), 16 deletions(-) commit 6167f27adfd530bdf091a0e7a3e0141d5db66f00 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jun 11 22:33:00 2009 +0100 [script] Fix memleak of duplicated strings from files csi_string_new() duplicated the bytes which was not what was desired, so implement a csi_string_new_for_bytes() to take ownership and prevent the leak that was occuring, for example, every time we create a new font face. util/cairo-script/cairo-script-file.c | 5 +++-- util/cairo-script/cairo-script-objects.c | 27 +++++++++++++++++++++++++++ util/cairo-script/cairo-script-private.h | 6 ++++++ 3 files changed, 36 insertions(+), 2 deletions(-) commit 13b56c42bfdb9ad36f9b1bd22f83d7ef3a074ddc Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Thu Jun 11 17:32:23 2009 -0400 Fix compilation on OS X when compiling for 64 bit OS X doesn't support ATSUI in 64 bit so don't include the cairo ATSUI API in that case. src/cairo-quartz-font.c | 3 ++- src/cairo-quartz.h | 2 ++ 2 files changed, 4 insertions(+), 1 deletion(-) commit 9e3019ef987f8d3ad424c13ffae5a15b08beb327 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jun 11 22:27:18 2009 +0100 [perf] Add libcairoperf.la to DEPENDENCIES It seems adding the explicit dependencies to encourage it to rebuild components from other parts of the source tree removed the automagic dependency of libcairoperf.la. So add it to the list. Maybe this is not the correct solution, but it works again for now. perf/Makefile.am | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) commit 55cbf525479cde8251cc9199f9db89e510d6a9c1 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jun 11 15:40:27 2009 +0100 [perf] Fix -i getopt parsing. Gah, I presumed that the ':' separated options that required arguments from stand-alone options. I was wrong. The ':' indicates that the preceding option takes an argument. So add it back to -i. perf/cairo-perf-trace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 614bfdc0846856653a742b035e70d009b0a716cc Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jun 11 15:35:05 2009 +0100 [xcb] Trivial compile fix. Update show_glyphs() interface. src/cairo-xcb-surface.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) commit f7dbdf1fd1d7cc14491a9419c316b078247e145e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jun 8 09:39:20 2009 +0100 [script] Hide pixman-ops when copying pixels Seeing unexpected time inside pixman composite is quite disturbing when trying to track down the apparent slowness in some benchmarks. Remove one source of this artefact by simply memcpy'ing pixel data when trivial. util/cairo-script/cairo-script-operators.c | 70 +++++++++++++++++++++++------- 1 file changed, 55 insertions(+), 15 deletions(-) commit ecee5663bf900ab97d2bb7ec41ade838f167e1cd Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jun 8 01:05:28 2009 +0100 [perf] Fixup compare-backends to work with just a single test. perf/cairo-perf-compare-backends.c | 17 ++++++----------- 1 file changed, 6 insertions(+), 11 deletions(-) commit 42c0aee1d9ff5e0182c41ed505a2b0f56e564cbb Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jun 11 12:58:05 2009 +0100 [perf] Add a -x to exclude traces from a benchmark. Read names of traces to exclude from a file specified using -x on the commandline, i.e. $ ./cairo-perf-trace -x cairo-traces/tiny.exclude This is a convenient method for me to exclude certain traces for particular machines. For example tiny cannot run firefox-36-20090609.trace as that has a greater working set than the available RAM on tiny. perf/cairo-perf-trace.c | 135 +++++++++++++++++++++++++++++++++++++++++++++--- perf/cairo-perf.h | 2 + 2 files changed, 130 insertions(+), 7 deletions(-) commit b40ce402ada8d57c5093715be530df08459d1654 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jun 11 09:16:29 2009 +0100 [NEWS] Add more notes NEWS | 50 ++++++++++++++++++++++++++++++++++++++++++++------ 1 file changed, 44 insertions(+), 6 deletions(-) commit 610da573e91810c53305b6bfe02eb7b714e3d08e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jun 10 12:59:36 2009 +0100 [win32-font] Non-fatal error from GetGlyphOutlineW(GGO_METRICS) If GetGlyphOutlineW(GGO_METRICS) fails to retreive the metrics for the specified glyph it returns GDI_ERROR. Like ft, do not interpret this as a fatal error but just mark the glyph as empty. Fixes http://bugs.freedesktop.org/show_bug.cgi?id=20255 Bug 20255 -- cairo_scaled_font_glyph_extents breaks with invalid glyph id src/cairo-win32-font.c | 10 ++-------- 1 file changed, 2 insertions(+), 8 deletions(-) commit 81b5dc42b0e754d602506a8ccd231df9afd71593 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jun 10 08:49:39 2009 +0100 [perf] Expand the section on cairo-perf-trace in the README Promote the information on how to use cairo-perf-trace and include it immediately after the details on cairo-perf. This should make it much clearer on how to replay the traces, and the difference between the two benchmarks. perf/README | 37 ++++++++++++++++++++++++++++++++++--- 1 file changed, 34 insertions(+), 3 deletions(-) commit ec92e633edd377747155b60aa225b266c38bc498 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jun 10 08:34:57 2009 +0100 [perf] Couple cairo-perf-traces into make perf Rather than complicating cairo-perf to extend it to perform both micro- and macro-benchmarks, simply run the two binaries in succession during make perf. For bonus points, consider whether we should hook cairo-perf-trace into cairo-perf-diff. perf/Makefile.am | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) commit 0db946cdd9fe601a060b62c5df7df3022ea5e2d6 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jun 9 16:19:36 2009 +0100 [perf] Add explicit dependences for the local libraries Update the build rules and add the dependences for the libraries built in other parts of cairo. perf/Makefile.am | 11 +++++++++++ 1 file changed, 11 insertions(+) commit 844c809698635cf44658fcbfb5da6976890185b9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jun 9 12:45:53 2009 +0100 [boilerplate] Print out known targets on error. Improve the 'Cannot find target ...' error message for an incorrect CAIRO_TEST_TARGET by actually listing the targets that have been compiled into the test suite. boilerplate/cairo-boilerplate.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-) commit ac563715506b911f142dd0e4460a6effee45119d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jun 8 23:58:56 2009 +0100 [test] Exercise overlapping glyphs Quite an expensive test that converts an image into a distorted array of glyphs, using a perspective transformation taking the intensity of the pixel as depth. This generates a pretty picture and many overlapping glyphs. test/Makefile.am | 1 + test/Makefile.sources | 1 + test/mask-glyphs.c | 175 +++++++++++++++++++++++++++++++++++++++++++++++ test/mask-glyphs.ref.png | Bin 0 -> 1189352 bytes 4 files changed, 177 insertions(+) commit 015df191ba947e714285145c3a4ead198ba0d07e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jun 8 17:40:16 2009 +0100 [trace] Prevent overflowing the operand stack when recreating objects Objects like cairo_scaled_font_t may return a reference to a previously defined scaled-font instead of creating a new token each time. This caused cairo-trace to overflow its operand stack by pushing a new instance of the old token every time. Modify the tracer such that a font token can only appear once on the stack -- for font-faces we remove the old operand and for scaled-fonts we simply pop, chosen to reflect expected usage. util/cairo-trace/trace.c | 89 ++++++++++++++++++++++++++++++++++-------------- 1 file changed, 63 insertions(+), 26 deletions(-) commit 09492288b33c36093e50d39e4e7e632ab659a0e2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jun 8 17:38:14 2009 +0100 [trace] Option to re-enable mark-dirty Applications such as swfdec have a strictly correct use of mark-dirty and so we need an option to re-enable mark-dirty tracing in conjunction with --profile. util/cairo-trace/cairo-trace.in | 5 +++++ 1 file changed, 5 insertions(+) commit 39bac6edddb8913d07fb25f14f088967ca846a78 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jun 8 11:52:56 2009 +0100 [perf] Mention cairo-perf-trace early on in the README perf/README | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 34ccaeae2d3cf1b5bdd8b97581d6e70324494356 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jun 8 00:12:28 2009 +0100 [perf] Continuously update the summary for traces perf/cairo-perf-trace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 055e9b0eb650bec6ae8297e5a36f1cac22c43d56 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Jun 7 23:35:57 2009 +0100 [test] Cache resolved scaled-font The intention is to stress test the solid pattern caches, so we can cheat and avoid repeatedly resolving the toy font. test/solid-pattern-cache-stress.c | 6 ++++++ 1 file changed, 6 insertions(+) commit 25c37509396a89deb644cbd6ada99137ba9d3095 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Jun 7 19:23:04 2009 +0100 [scaled-font] Report the original-font-face When queried with cairo_scaled_font_get_font_face() return the original font-face which matches the one supplied by the user, rather than the implementation derived face. Fixes test/font-face-get-type. src/cairo-scaled-font.c | 3 +++ 1 file changed, 3 insertions(+) commit 6d693f6bd7c629372cfb6c284ca98dc04619bfe9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Jun 7 18:59:36 2009 +0100 [ft] Support font_face->get_implementation The lazy resolution of patterns was defeating the scaled_font cache as ft-fonts that resolved to the same unscaled font were being given different font-faces upon creation. We can keep the lazy resolution by simply asking the ft backend to create a fully resolved ft-font-face when we need to create a scaled-font. This font is then keyed by the resolved font-face and so will match all future lazily resolved identical patterns. src/cairo-ft-font.c | 117 ++++++++++++++++++++-------------------------- src/cairo-scaled-font.c | 20 +++++--- src/cairo-toy-font-face.c | 41 +++++++++------- src/cairoint.h | 9 ++-- 4 files changed, 96 insertions(+), 91 deletions(-) commit a29426f4bffc0a3f5bb349a2ba44203394339ae8 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Jun 7 17:24:40 2009 +0100 [ft] Assert that we create an unscaled font for the resolved pattern src/cairo-ft-font.c | 2 ++ 1 file changed, 2 insertions(+) commit 7d8a0a1cdd62112ccdae2a80bb55ec9cec99b978 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Jun 7 17:01:57 2009 +0100 [cairo] Keep a small stash of cairo contexts A cairo context is meant to be extremely cheap to create such that it can be used in transient expose events. Thus these are allocated reasonably frequently and show up malloc profiles. src/cairo.c | 51 +++++++++++++++++++++++++++++++++++++++++++++++++-- 1 file changed, 49 insertions(+), 2 deletions(-) commit f08fe5b550d53c93b73c6295f8a85a8dcd4a1f4a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Jun 7 15:55:20 2009 +0100 [trace] Redirect stdout to /dev/null when compressing Otherwise we mix in stdout in with the trace and prevent replay. util/cairo-trace/cairo-trace.in | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 750c1b5b48dcd33ba4a4d5290c50f564bf45bc58 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 6 21:45:18 2009 +0100 [configure] Check for FcInit() configure.ac | 2 +- test/cairo-test.c | 2 ++ 2 files changed, 3 insertions(+), 1 deletion(-) commit f7a562a590493904cf10b4fb9af3b42fb3f8a3a2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 6 14:58:46 2009 +0100 [configure] Don't attempt to build GTK+ utilities on system without GTK+ configure.ac | 3 ++- util/Makefile.am | 5 ++++- 2 files changed, 6 insertions(+), 2 deletions(-) commit 6a5957475139a7c80a6bb55fb63927e5ec8866f6 Author: Eric Anholt <eric@anholt.net> Date: Sun Jun 7 10:43:43 2009 +0000 [gl] consolidate the pixman format to GL format/type/internalformat code. This means that draw_image gets all the love that the pattern texture code does. Not that I'm sure it still needs to exist. src/cairo-gl-surface.c | 321 ++++++++++++++++++++++++------------------------- 1 file changed, 158 insertions(+), 163 deletions(-) commit cb1e5674cd7a61a2d96bd4ae8895b26686571e78 Author: Eric Anholt <eric@anholt.net> Date: Sun Jun 7 10:09:28 2009 +0000 [gl] Add support for more 16-bit image formats. Fixes test-fallback16-surface-source src/cairo-gl-surface.c | 54 ++++++++++++++++++++++++++++++++++++++++++++------ 1 file changed, 48 insertions(+), 6 deletions(-) commit 7a95b2745ae619f6ca4631f3cdb03e12203126f3 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 6 21:36:08 2009 +0100 [boilerplate] Minor typo in win32-printing boilerplate/cairo-boilerplate.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 16eb160c7794ad34c17c5b12ec586d956baf9831 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 6 20:09:15 2009 +0100 [script] Enable compilation without fontconfig Some insane people actually try compiling cairo with freetype and without fontconfig... util/cairo-script/cairo-script-operators.c | 6 ++++++ 1 file changed, 6 insertions(+) commit d753ba96aba4dbbcbd0da1823be8824ba233f079 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 5 17:59:38 2009 +0100 [script] Manage used entries within hash tables Apply the patch from Karl Tomlinson, https://bugs.freedesktop.org/attachment.cgi?id=19802, to repack the hash table if the number of free slots falls too low. util/cairo-script/cairo-script-hash.c | 109 ++++++++++++++++++++----------- util/cairo-script/cairo-script-private.h | 1 + 2 files changed, 73 insertions(+), 37 deletions(-) commit 4ccfd474a36f482adcab49a8d38742121817b47e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 6 13:32:21 2009 +0100 [perf] Switch to using clock_gettime() Try using clock_gettime() for a high resolution stable time source in preference to the potentially unstable TSC. build/configure.ac.system | 9 ++++-- perf/cairo-perf-posix.c | 72 ++++++++++++++++++++++++++++++++++------------- 2 files changed, 60 insertions(+), 21 deletions(-) commit 867c88ae90e4e69a556c032fa67d94a292f5d817 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 6 12:59:39 2009 +0100 [script] Add a finish method to the interpreter When using fonts circular references are established between the holdover font caches and the interpreter which need manual intervention via cairo_script_interpreter_finish() to break. perf/cairo-perf-trace.c | 3 +- util/cairo-script/cairo-script-interpreter.c | 51 ++++++++++++++++++++++------ util/cairo-script/cairo-script-interpreter.h | 3 ++ util/cairo-script/cairo-script-objects.c | 35 ++++++++++++------- util/cairo-script/cairo-script-private.h | 3 ++ 5 files changed, 70 insertions(+), 25 deletions(-) commit 098822d7ee7f14096dd269c1bc96e39da126622c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jun 6 10:35:19 2009 +0100 [script] Fix erroneous error return when rendering type3 glyphs Only return a USER_FONT_ERROR if an error occurs whilst rendering the glyph! util/cairo-script/cairo-script-operators.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 24e51dd2ee04884e1f6f9f527b6675ad5f883ee2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 5 18:34:29 2009 +0100 [atomic] Hide compiler warnings Add uint variants of _cairo_atomic_int_*() to hide the compiler warnings about sign conversions. src/cairo-atomic-private.h | 3 +++ src/cairo-surface.c | 6 +++--- 2 files changed, 6 insertions(+), 3 deletions(-) commit 4ae5e2d44555a7219eb76ee187a8f588606c112f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 5 18:29:40 2009 +0100 [atomic] Provide mutex-based ptr cmpxchg To handle those CPUs where we do not have an atomic cmpxchg. src/cairo-atomic-private.h | 3 +++ src/cairo-atomic.c | 13 +++++++++++++ 2 files changed, 16 insertions(+) commit a1d0a06b6275cac3974be84919993e187394fe43 Author: Thomas Jaeger <ThJaeger@gmail.com> Date: Thu Feb 19 12:02:41 2009 -0500 xlib, xcb: Hand off EXTEND_PAD/EXTEND_REFLECT to Render Most drivers and the X server used to have incorrect RepeatPad/RepeatReflect implementations, forcing cairo to fall back to client-side software rendering, which is painfully slow due to pixmaps being transfered over the wire. These issues are mostly fixed in the drivers (with the exception of radeonhd, whose developers didn't respond) and the RepeatPad software fallback is implemented correctly as of pixman-0.15.0, so this patch will hand off composite operations with EXTEND_PAD/EXTEND_REFLECT source patterns to XRender. There is no way to detect whether the X server or the drivers use a broken Render implementation, we make a guess based on the server version: It's probably safe to assume that 1.7 X servers will use fixed drivers and a recent enough version of pixman. AUTHORS | 1 + src/cairo-xcb-surface.c | 15 +++++++++++++-- src/cairo-xlib-display.c | 15 +++++++++++---- src/cairo-xlib-private.h | 1 + src/cairo-xlib-surface-private.h | 1 + src/cairo-xlib-surface.c | 21 ++++++++++++++++++--- 6 files changed, 45 insertions(+), 9 deletions(-) commit 60aefd0d634c531353c92b77b36424b142efeb93 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jun 5 07:30:13 2009 +0100 [configure] Bump pixman dependency. Reflect was only fixed in the 0.15 series, so require at least that version of pixman so we can that advantage of that and drop our workaround. configure.ac | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 9d7a7e76b8bccc2517950976bb8557cd7b955a11 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jun 4 22:19:31 2009 +0100 [scaled-font] Close race from 16d128 Whilst waiting for the fontmap lock on destruction another thread may not only have resurrected the font but also destroyed it acquired the lock first and inserted into the holdovers before the first thread resumes. So check that the font is not already in the holdovers array before inserting. src/cairo-scaled-font-private.h | 6 +++--- src/cairo-scaled-font.c | 33 +++++++++++++++++++++++---------- 2 files changed, 26 insertions(+), 13 deletions(-) commit f4d9a4482fa1bfa37783ca139e3716c8d8649f32 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jun 4 14:13:06 2009 +0100 [pattern] Generalise the freed pattern pool Also cache recently freed surface and gradient patterns. With thanks to Jonathan Morton for the inspiration and initial pointer pool code for pixman. src/cairo-atomic-private.h | 1 + src/cairo-mutex-list-private.h | 1 - src/cairo-pattern.c | 170 +++++++++++++++++++++++++++-------------- 3 files changed, 112 insertions(+), 60 deletions(-) commit 132f44dce1884c9beed8c5e6588140ee8fa9cfbe Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jun 4 14:12:20 2009 +0100 valgrindify init/fini routines Annotate object init/fini routines to detect use-after-free for on-stack/embedded objects. src/cairo-debug.c | 2 -- src/cairo-freelist.c | 7 ------- src/cairo-gstate.c | 7 +++++++ src/cairo-path-fixed.c | 10 +++------- src/cairo-pattern.c | 49 +++++++++++++++++++++++++++++++++++++++++++++++- src/cairo-pen.c | 8 ++++++-- src/cairo-polygon.c | 4 ++++ src/cairo-region.c | 5 +++++ src/cairo-stroke-style.c | 6 ++++++ src/cairo-traps.c | 4 ++++ src/cairo.c | 1 + src/cairoint.h | 4 ++++ 12 files changed, 88 insertions(+), 19 deletions(-) commit e89828c5331579d0d4019b6339df09ba1e2aa32e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jun 3 21:52:02 2009 +0100 [freelist] valgrindify freed blocks Mark freed blocks as unaccessible upon free() and then as undefined on alloc(). src/cairo-freelist.c | 26 ++++++++++++++++++++++---- 1 file changed, 22 insertions(+), 4 deletions(-) commit 4e11d56f8e5f409ea2476315c9d67116b213b738 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jun 4 11:16:48 2009 +0100 [perf] Update trace summary after each iteration Running the macro-benchmarks takes so long, feedback is important. perf/cairo-perf-trace.c | 29 ++++++++++++++++++++++++++++- perf/cairo-perf.h | 1 + 2 files changed, 29 insertions(+), 1 deletion(-) commit 5c2e73e74cb9aef92fdd90a91df44b417a036ebd Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jun 4 10:03:45 2009 +0100 [perf] Enable traces to be interrupted Waiting for a long running benchmark can be very annoying, especially if you just want a rough-and-ready result. So hook into SIGINT and stop the current benchmark (after the end of the iteration) on the first ^C. A second ^C within the same iteration will kill the program as before. perf/cairo-perf-trace.c | 20 +++++++++++++++++++- 1 file changed, 19 insertions(+), 1 deletion(-) commit 2eaced24475622e73fbf6e3307ab46e3fe37eaef Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jun 3 21:49:17 2009 +0100 [perf] Reconstruct interpreter for each run Destroy the interpreter after each run of a trace so that we collect any left-over garbage. perf/cairo-perf-trace.c | 21 ++++++++++----------- 1 file changed, 10 insertions(+), 11 deletions(-) commit 1b92ea250fdaef9add1a68bb8565847e6d56b023 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jun 3 20:28:49 2009 +0100 [ft] Don't call FT_Done_Face() on a face we do not own _font_map_release_face_lock_held() was being called unconditionally during _cairo_ft_font_reset_static_data(). This presents two problems. The first is that we call FT_Done_Face() on an object not owned by cairo, and the second is that the bookkeeping is then incorrect which will trigger an assert later. src/cairo-ft-font.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) commit 8e5295979fe087fadeff0f56c036f9d002086996 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jun 3 20:17:59 2009 +0100 [pattern] Typo in document Remove extraneous markup from program-listing. https://bugs.freedesktop.org/show_bug.cgi?id=20441 src/cairo-pattern.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit fa5d6c7afdb9b742ad653c65e8804e84f2964599 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jun 3 19:29:19 2009 +0100 [trace] Add a --profile mode To save typing when creating macro-benchmarks --profile disables mark-dirty and caller-info and compresses the trace using LZMA. Not for computers short on memory! perf/README | 11 +++++++++-- util/cairo-trace/cairo-trace.in | 18 ++++++++++++++++++ 2 files changed, 27 insertions(+), 2 deletions(-) commit 9451f0b983665d13ad777bc8fd88507fb7a317b4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jun 3 13:26:52 2009 +0100 [perf] Drop ticks from trace benchmarks Don't report the tick values for the macro-benchmarks as these are so large as to be meaningless. perf/cairo-perf-trace.c | 37 ++++++++++++++++++++----------------- 1 file changed, 20 insertions(+), 17 deletions(-) commit c988dc3719e345f6e1db6b0d656077a6b253b26d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jun 3 10:51:23 2009 +0100 [cairo] Doc typo. Fix conflicting sentence as pointed out by https://bugs.freedesktop.org/show_bug.cgi?id=20183 src/cairo.c | 4 ---- 1 file changed, 4 deletions(-) commit 35d0dd00c9f21cc4fd392d4627756f8d5b53a11b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jun 3 10:49:16 2009 +0100 [image] Doc typo. Correct parameter reference as pointed out by https://bugs.freedesktop.org/show_bug.cgi?id=20182 src/cairo-image-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 58459ac6435a2889b90580ddf00e69107361f431 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jun 3 10:45:51 2009 +0100 [font-options] Doc typo Remove duplicate 'with' as pointed out by https://bugs.freedesktop.org/show_bug.cgi?id=20180 src/cairo-font-options.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 8d39dee8e2aea2200fec460e8540838ec29cae5f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jun 3 10:40:38 2009 +0100 [cairo] Update documentation for bug 20154 Several typographical errors were pointed out in https://bugs.freedesktop.org/show_bug.cgi?id=20154 src/cairo.c | 17 ++++++++--------- 1 file changed, 8 insertions(+), 9 deletions(-) commit 73469e693c02e6a99a32f12ac333737b47408e5e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jun 3 09:51:52 2009 +0100 [cairo] Tweak error strings. Rewrite a few error strings so that they more closer match the documentation. Where they differ, I believe I have chosen the more informative combination of the two texts. src/cairo-misc.c | 12 ++++++------ src/cairo.h | 4 ++-- 2 files changed, 8 insertions(+), 8 deletions(-) commit ab86662ab499e1f29c0f8c4248771e730c281e3f Author: Andrew Lavin <aj.lavin@gmail.com> Date: Fri May 22 16:36:40 2009 -0400 [test] Compile fix for invalid-matrix Fixed test compile error on OS X that caused fenv.h not to be included. test/invalid-matrix.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit eec547648439884a464eef832067a363cf1547e2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jun 3 08:57:15 2009 +0100 [perf] s/git/anongit/ Third time lucky? perf/cairo-perf-trace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit ce7e6339428703069f798dffcf23ec3e98cb554f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jun 2 21:54:44 2009 +0100 [perf] Warn if no traces were found perf/cairo-perf-trace.c | 28 ++++++++++++++++++++-------- 1 file changed, 20 insertions(+), 8 deletions(-) commit f0366f0f3263397181699bf6bf2ed3968ea66403 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jun 2 21:48:26 2009 +0100 [perf] Update to point to new toplevel cairo-traces repo perf/cairo-perf-trace.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) commit c9cd2f9ca101c25d04f52578d54be36ae646049e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jun 2 18:47:07 2009 +0100 [cairo] PLT symbols for cairo_set_source_rgb Added a private use in user-font, but forgot to add a hidden symbol. src/cairo.c | 1 + src/cairoint.h | 1 + 2 files changed, 2 insertions(+) commit 21550a753059b4f880f03ea96148befaa34a6426 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jun 2 18:08:45 2009 +0100 [user-font] Set the initial colour to white An issue occured when using subpixel antialiasing with user-fonts and XRender - the glyphs were transparent, as demonstrated by the font-view example. The problem lies in that enabling subpixel antialiasing triggers use of an ARGB32 image surface for rendering the glyph, but the default colour is black (so the only information is in the alpha-channel). Given an ARGB32 glyph XRender treats it as a per-channel mask, but since the R,G,B channels were uniformly zero, the glyph is rendered as transparent. Fix this by setting the initial colour to white before rendering the image surface for a user-font glyph, which generates the appropiate gray-level mask by default. src/cairo-user-font.c | 1 + 1 file changed, 1 insertion(+) commit b5799e073e4e404aea277b7640eeee2120318823 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jun 2 18:07:31 2009 +0100 [xlib] Use bswap_32 Byteswap the ARGB32 glyphs using bswap_32 instead of open-coding. src/cairo-xlib-surface.c | 30 +++++++++++------------------- 1 file changed, 11 insertions(+), 19 deletions(-) commit 56e9765f82a64940b36a64688267fbe5d1c8919e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 2 14:25:04 2009 +0000 [util] Add font-view A C variant of Behdad's python example font viewer. util/Makefile.am | 5 ++ util/font-view.c | 184 +++++++++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 189 insertions(+) commit d331dda48017667c15f95bf7d510bce2f38111de Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jun 2 15:41:19 2009 +0100 [perf] Missing ')' in output perf/cairo-perf-compare-backends.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit c9ddaf8e7f0006f308aab0b87cf0425b2690d914 Author: Adrian Johnson <ajohnson@redneon.com> Date: Sun Sep 7 13:09:09 2008 +0930 PDF: Don't embed the same pattern more than once The PDF surface now keeps track of all the patterns it is embedding in a hash table keyed by the unique_id returned by the _cairo_surface_get_unique_id(). src/cairo-pdf-surface-private.h | 13 +++- src/cairo-pdf-surface.c | 134 +++++++++++++++++++++++++++++++--------- 2 files changed, 115 insertions(+), 32 deletions(-) commit f7613eeecea1218d353a183078fb77bc5cf4e72e Author: Paolo Bonzini <bonzini@gnu.org> Date: Tue Nov 25 20:15:15 2008 +0100 Add more surface snapshots providers This patch adds more implementation of the snapshot method. For surface types where acquire_source_image is already making a copy of the bits, doing another one as is the case for the fallback implementation is a waste. src/cairo-glitz-surface.c | 16 +++++++++++++++- src/cairo-quartz-surface.c | 19 ++++++++++++++++++- src/cairo-xcb-surface.c | 17 ++++++++++++++++- src/cairo-xlib-surface.c | 20 +++++++++++++++++++- 4 files changed, 68 insertions(+), 4 deletions(-) commit cffb398f5a484000be458d04ef8f8bce3f6c7e3d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue May 26 21:07:07 2009 +0100 Add a generic cow-snapshotting framework Provide a mechanism for backends to attach and remove snapshots. This can be used by backends to provide a cache for _cairo_surface_clone_similar(), or by the meta-surfaces to only emit a single pattern for each unique snapshot. In order to prevent stale data being returned upon a snapshot operation, if the surface is modified (via the 5 high level operations, and on notification of external modification) we break the association with any current snapshot of the surface and thus preserve the current data for their use. src/cairo-meta-surface.c | 1 - src/cairo-surface-fallback.c | 2 +- src/cairo-surface-private.h | 7 +- src/cairo-surface.c | 272 +++++++++++++++++++++++++++++++++---------- src/cairoint.h | 12 ++ test/mask.c | 12 +- test/trap-clip.c | 12 +- 7 files changed, 249 insertions(+), 69 deletions(-) commit 7ed050fd435f17d25c7b757b02cfe200f8779fc2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jun 2 13:08:25 2009 +0100 [perf] Benchmark traces Add a variant of cairo-perf that measures the time to replay traces. boilerplate/cairo-boilerplate-system.c | 15 + boilerplate/cairo-boilerplate-system.h | 4 + perf/.gitignore | 1 + perf/Makefile.am | 17 ++ perf/README | 31 ++ perf/cairo-perf-compare-backends.c | 10 +- perf/cairo-perf-trace.c | 530 +++++++++++++++++++++++++++++++++ 7 files changed, 605 insertions(+), 3 deletions(-) commit 403f780b292762aa45056c2fb5a48bb806521173 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jun 2 09:16:16 2009 +0100 [perf] Add a verbose flag for summary output with raw Use 'cairo-perf -v -r' to have both the summary output along with the raw values. This gives a progress report whilst benchmarking, very reassuring with long running tests. perf/cairo-perf.c | 82 +++++++++++++++++++++++++++++++++++-------------------- perf/cairo-perf.h | 3 ++ 2 files changed, 55 insertions(+), 30 deletions(-) commit 70fd2bbce38fca65db6725c632ee591352dfe42c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jun 2 09:00:21 2009 +0100 [perf] Exclude similar testing by default There are synchronisation issues with similar surfaces (as only the original target surface is synced) which interferes with making performance comparisons. (There still maybe some value should you be aware of the limitations...) perf/cairo-perf.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) commit bc49cb377c70c4c6926ca52e34094e99a4591f14 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat May 30 12:28:23 2009 +0100 [perf] Benchmark mixing different masks and sources perf/Makefile.am | 1 + perf/cairo-perf.c | 1 + perf/cairo-perf.h | 1 + perf/mask.c | 290 ++++++++++++++++++++++++++++++++++++++++++++++++++++++ 4 files changed, 293 insertions(+) commit 4aca84ddb22bc178cbc9b132b9ce06af3f4b300b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu May 28 18:05:26 2009 +0100 [perf] Add a pure glyphs performance metric Use the new API Behdad exposed in 1.8 to precompute a glyph string using Cairo and then benchmark cairo_show_glyphs(). This is then equivalent to the text benchmark but without the extra step of converting to glyphs on every call to cairo_show_text() i.e. it shows the underlying glyph rendering performance. perf/Makefile.am | 1 + perf/cairo-perf.c | 1 + perf/cairo-perf.h | 1 + perf/glyphs.c | 97 +++++++++++++++++++++++++++++++++++++++++++++++++++++++ 4 files changed, 100 insertions(+) commit 55f4e0e4e8c7df59bfc9e6ffea8daa065276e42f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat May 30 16:50:17 2009 +0100 [perf] Change seperators from '-' to '.' This allows the perf tests to use '-' in the name which is easier to read and differentiates with using '_' to separate source and operators. perf/cairo-perf-compare-backends.c | 22 ++++++++++++---------- perf/cairo-perf-cover.c | 32 ++++++++++++++++---------------- perf/cairo-perf-report.c | 16 +++++++++++++--- perf/cairo-perf.c | 8 ++++---- perf/mosaic.c | 8 ++++---- perf/pattern_create_radial.c | 4 ++-- perf/pythagoras-tree.c | 2 +- perf/subimage_copy.c | 4 ++-- perf/unaligned-clip.c | 2 +- perf/world-map.c | 4 ++-- perf/zrusin.c | 4 ++-- 11 files changed, 59 insertions(+), 47 deletions(-) commit b7c03d4e400245fc2b2d748fe808a9724e61f22f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed May 27 16:31:51 2009 +0100 [ps] Trivial warning fix. src/cairo-ps-surface.c | 1 + 1 file changed, 1 insertion(+) commit 99482b17a50a8ae52b7627bf624e358ed3623540 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed May 27 18:34:54 2009 +0100 [script] Improve caching of glyph advance The glyph advance cache was only enabled for glyph indices < 256, causing a large number of misses for non-ASCII text. Improve this by simply applying the modulus of the index to select the cache slot - which may cause some glyph advances to be overwritten and re-queried, but improves the hit rate. util/cairo-script/cairo-script-operators.c | 39 +++++++++++------------------- 1 file changed, 14 insertions(+), 25 deletions(-) commit 51193f1610c0f6893768e43e90a3395ec96c28fb Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu May 28 14:16:58 2009 +0100 [pattern] Pass flags to _cairo_pattern_acquire_surface() Allow the caller to choose whether or not various conversions take place. The first flag is used to disable the expansion of reflected patterns into a repeating surface. src/cairo-directfb-surface.c | 1 + src/cairo-glitz-surface.c | 1 + src/cairo-image-surface.c | 3 +++ src/cairo-pattern.c | 13 +++++++++++-- src/cairo-xcb-surface.c | 4 ++++ src/cairo-xlib-surface.c | 5 +++++ src/cairoint.h | 6 ++++++ 7 files changed, 31 insertions(+), 2 deletions(-) commit 4ec451a2fa4050a60a81d01473d23ff2973f47e6 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed May 27 17:35:54 2009 +0100 [scaled-font] Cache repeated glyphs during probing device_extents Maintain a local on-stack cache of recently queried glyphs to avoid relatively expensive hash table queries. src/cairo-scaled-font.c | 35 +++++++++++++++++++++++------------ 1 file changed, 23 insertions(+), 12 deletions(-) commit ec22ae5b1d7e8ec364813f3b1c73f538018ecb57 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed May 27 11:11:06 2009 +0100 [scaled-font] Compare most recently used font directly Avoid the relatively expensive _cairo_scaled_font_init_key() operation when checking against the most recently used font. src/cairo-scaled-font.c | 28 +++++++++++++++++++++++----- 1 file changed, 23 insertions(+), 5 deletions(-) commit 16d128c15edf36a6e285fe4fbd6a828b64fd7f87 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed May 27 21:10:04 2009 +0100 [scaled-font] Lazily acquire the font_map_lock on font destruction. We can defer taking the cairo_scaled_font_map_lock until we drop the last reference to the scaled font so long as we double check the reference count after waiting for the lock and not making assumptions about unreferenced fonts during construction. This is significant as even acquiring the uncontended cairo_scaled_font_map_lock during cairo_scaled_font_destroy() was showing up as a couple of percent on text heavy profiles (e.g. gnome-terminal). src/cairo-scaled-font.c | 32 +++++++++++++++++--------------- 1 file changed, 17 insertions(+), 15 deletions(-) commit dc083ab30a5b781e205354c525ee054982364abd Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed May 27 14:54:34 2009 +0100 [cairo] Track the MRU scaled font When observing applications two patterns emerge. The first is due to Pango, which wraps each glyph run within a context save/restore. This causes the scaled font to be evicted after every run and reloaded on the next. This is caught by the MRU slot on the cairo_scaled_font_map and prevents a relatively costly traversal of the hash table and holdovers. The second pattern is by applications that directly manage the rendering of their own glyphs. The prime example of this is gnome-terminal/vte. Here the application frequently alternates between a few scaled fonts - which requires a hash table retrieval every time. By introducing a MRU slot on the gstate we are able to directly recover the scaled font around 90% of the time. Of 110,000 set-scaled-fonts: 4,000 were setting the current font 96,000 were setting to the previous font 2,500 were recovered from the MRU on the cairo_scaled_font_map 7,500 needed a hash retrieval which compares to ~106,000 hash lookups without the additional MRU slot on the gstate. This translates to an elapsed time saving of ~5% when replaying a gnome-terminal trace using the drm backend. src/cairo-gstate-private.h | 1 + src/cairo-gstate.c | 18 ++++++++++++++---- src/cairo.c | 6 ++++++ 3 files changed, 21 insertions(+), 4 deletions(-) commit fda89c56ff484a8cd33cd780e8b9396d3538284d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed May 27 15:59:16 2009 +0100 Markup a few more functions as const/pure src/cairoint.h | 59 ++++++++++++++++++++++++++++++---------------------------- 1 file changed, 31 insertions(+), 28 deletions(-) commit f59f44c140e5f60d336423e0585d2bb8a6c0ea01 Merge: 2da78fd46 4232719af Author: Eric Anholt <eric@anholt.net> Date: Tue Jun 2 00:37:19 2009 -0700 Merge commit 'origin/master' into gl Felt like pulling the latest stuff, since I branched back in February. Conflicts: build/configure.ac.features src/cairo.h util/cairo-script/csi-replay.c commit 2da78fd4666faa27d037ae3625ca83353a6e7629 Author: Eric Anholt <eric@anholt.net> Date: Mon Jun 1 14:04:21 2009 -0700 [gl] Drop use of packed datatypes where it's disallowed. I had naively assumed that GL image specification let you do useful things and describe the most common datatypes in graphics, since we do things that way inside of the DRI drivers. Silly me. GL_BGR and GL_RGB can't do GL_UNSIGNED_INT_8_8_8_8{,_REV}, so no specifying 24-depth 32-bpp data with implicit alpha. GL_BGR can't even do r5g6b5! This fixes 20 regressions in the test suite. src/cairo-gl-surface.c | 25 +++++++++++++------------ 1 file changed, 13 insertions(+), 12 deletions(-) commit 69a150b7d8115babf2c60d7c5f7ca340162dc098 Author: Eric Anholt <eric@anholt.net> Date: Mon Jun 1 13:36:32 2009 -0700 [gl] Initialize has_alpha in the image surface composite fastpath. Fixes bilevel-image, set-source, and zero-alpha. src/cairo-gl-surface.c | 4 ++++ 1 file changed, 4 insertions(+) commit 7ee0fee900bf30a7ca0497ab64199c9013f81cdd Merge: 882f1d5a6 d3a478b6c Author: Eric Anholt <eric@anholt.net> Date: Mon Jun 1 10:45:36 2009 -0700 Merge branch 'gl' into gl-span-renderer Conflicts: src/cairo-gl-surface.c commit d3a478b6c0dca3884a68c5014185cad0a166801c Author: Eric Anholt <eric@anholt.net> Date: Mon Jun 1 10:19:24 2009 -0700 [gl] Fix check-plt.sh for the GL surface. src/cairo-gl-surface.c | 7 +++++++ 1 file changed, 7 insertions(+) commit 1f7c393a6d63282a632a62264b99e67e26ecf38e Author: Eric Anholt <eric@anholt.net> Date: Mon Jun 1 10:10:00 2009 -0700 [gl] Fix check-doc-syntax.sh for GL surface. src/cairo-gl-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 19bc0f2f1d75b1514b22d0bb737bad67ff8d1ee6 Author: Eric Anholt <eric@anholt.net> Date: Mon Jun 1 10:09:16 2009 -0700 [gl] Fix check-def.sh for GL surface. src/cairo-gl-surface.c | 2 +- src/cairo-gl.h | 6 ------ 2 files changed, 1 insertion(+), 7 deletions(-) commit e05097c604c607d27b341cd33426001bd2d1f690 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed May 27 16:45:20 2009 +0100 [surface] Assign a unique id to the surface. Allocate an ever-increasing, non-zero, unique identifier to each surface. True for the first 4-billion... src/cairo-pattern.c | 11 ++++++++--- src/cairo-surface-private.h | 2 ++ src/cairo-surface.c | 25 +++++++++++++++++++++++++ 3 files changed, 35 insertions(+), 3 deletions(-) commit 45835f623f5ddda6e3258361b9d6ab27860a9198 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue May 26 12:46:46 2009 +0100 [cache] Expose init/fini methods so that caches can be embedded. The structure is already exposed, so just expose the constructors/destructors in order to enable caches to be embedded and remove a superfluous malloc. src/cairo-cache-private.h | 27 +++++++++--- src/cairo-cache.c | 104 ++++++++++++++++------------------------------ src/cairo-scaled-font.c | 35 ++++++++-------- src/cairo-types-private.h | 14 ------- 4 files changed, 74 insertions(+), 106 deletions(-) commit c3aac9cf49362b726a54a33a46bd8511a10f644f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue May 26 10:37:30 2009 +0100 [image] Eliminate trapezoid array allocation By simply iterating over the array cairo_trapezoid_t, converting each one separately to a pixman_trapezoid_t and rasterizing each one individually we can avoid the common heap allocation. pixman performs exactly the same iteration internally so there is no efficiency loss. src/cairo-image-surface.c | 92 +++++++++++++++++++++++------------------------ 1 file changed, 45 insertions(+), 47 deletions(-) commit 7b2bc441387abcf3967587f571dc5741d0a53938 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue May 26 10:41:53 2009 +0100 [surface-fallback] Tidy pattern handling. Make the treatment of replacing the NULL source pattern with WHITE consistent. As it is a solid pattern, we can skip _cairo_pattern_fini() and so make the code more readable, and consistent along the error paths. src/cairo-surface-fallback.c | 83 ++++++++++++++++++-------------------------- 1 file changed, 33 insertions(+), 50 deletions(-) commit e83e113eae9e7cb3e09719bfc0ad68450faf3ecd Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon May 25 23:25:38 2009 +0100 [surface] Speed up cairo_surface_get_mime_data(). The number of mime-types attached to a surface is usually small, typically zero. Therefore it is quicker to do a strcmp() against each key in the private mime-data array than it is to intern the string (i.e. compute a hash, search the hash table, and do a final strcmp). src/cairo-array.c | 8 -------- src/cairo-surface.c | 33 ++++++++++++++++++--------------- src/cairoint.h | 6 ++++++ 3 files changed, 24 insertions(+), 23 deletions(-) commit 247e76b923d54f639e2ca50c8da945c15b691158 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue May 26 08:53:45 2009 +0100 [script] Speed up floating-point scanner Hard-code frequent exponents to reduce number of calls to pow(). util/cairo-script/cairo-script-scanner.c | 45 ++++++++++++++++++++++---------- 1 file changed, 31 insertions(+), 14 deletions(-) commit c5e85835b29c7654e1c28a90b2e587afba7b6f0f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue May 26 15:52:30 2009 +0100 [debug] Relax ASSERT_NOT_REACHED Need to actually handle random pixman image formats... For the time being, ignore them. src/cairo-debug.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 4232719af968ed05636fe34f2ffe2520dc02d737 Author: Behdad Esfahbod <behdad@behdad.org> Date: Sat May 30 23:03:55 2009 -0400 [ft] Fix vertical advance metrics of bitmap fonts (#21985) src/cairo-ft-font.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 4314cae874dec2747df6be88859d7b6174eece31 Author: Adrian Johnson <ajohnson@redneon.com> Date: Wed May 27 17:56:05 2009 +0930 Fix typo in comment src/cairo-truetype-subset.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 7dbc2fe80a4fe0dcee4a293e47ab6edcefc24e18 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon May 25 20:36:34 2009 +0100 [script] Simplify _dictionary_name_equal Just a simple return TRUE since all necessary checking is performed by _csi_hash_table_lookup(). util/cairo-script/cairo-script-objects.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) commit 0b5e92e66be94ce7cc9c31e911f23c4cb7ec77af Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon May 25 19:56:57 2009 +0100 [script] Check hash value between comparing keys util/cairo-script/cairo-script-hash.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) commit a364f71194aa2ea92071662f156c9b2a4d9211b4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon May 25 21:25:27 2009 +0100 [trace] Set output location using pwd Record the current working directory and pass that along to cairo-trace so that the trace output is local to the user and not the application. This is vital if the application is called via a script that changes directory. util/cairo-trace/cairo-trace.in | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit a76e09ea656faa63fbfa159e8f52c9c9ec7d35c6 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon May 25 21:22:43 2009 +0100 [trace] Missing newlines in error messages. util/cairo-trace/trace.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit b7f199fde25c960bf87302d5e868a7c2dffa4f5d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon May 25 12:40:35 2009 +0100 [pattern] Trim REPEAT source size when applicable. Some backends are quite constrained with surface sizes and so trigger fallbacks when asked to clone large images. To avoid this we attempt to trim ROIs (as these are often limited to the destination image, and so can be accommodated by the hardware). This patch allows trimming REPEAT sources both horizontally and vertically independently. src/cairo-pattern.c | 21 ++++++++++++++++++--- 1 file changed, 18 insertions(+), 3 deletions(-) commit e4efc80b8e89b05afc22d74f984f4ec9012bc39b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon May 25 21:15:22 2009 +0100 [build] Make valgrind support user-configurable As the number of places where we add valgrind client requests grows, it becomes imperative that we should be able to disable them with a simple configure time option. build/configure.ac.analysis | 37 ++++++++++++++++++++++--------------- 1 file changed, 22 insertions(+), 15 deletions(-) commit d840deb57b51236820dc8c320ecd7540973de873 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon May 25 22:11:22 2009 +0100 [in-fill] Fix typo in on-vertex test. Eeek! Comparing point->y against in_fill->x is a bad idea. src/cairo-path-in-fill.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) commit cfd484cd01a77b1f91e27daccfc5f240cf7c692d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon May 25 22:10:20 2009 +0100 [test] Change test semantics to match new in-fill definition The in-fill definition has changed to include queries on the edges and vertices, so update the test case to match. test/in-fill-trapezoid.c | 77 +++++++++++++++++++++++++++++++++++------------- 1 file changed, 57 insertions(+), 20 deletions(-) commit b71b019fe50a9188ddbecd1945606da8ba3bad53 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Feb 27 16:32:21 2009 +0000 [in-fill] Treat on-edge queries as inside. Jeff Muizelaar noted that the treatment of edges differed with firefox's canvas definition, which considers a point on any edge as inside. The current implementation has a similar definition to that of flash, for which the top and right edges are outside. Arguably, firefox has the more intuitive definition here... src/cairo-path-in-fill.c | 21 ++++++++++++++++++++- 1 file changed, 20 insertions(+), 1 deletion(-) commit efd0f0b2922d0801e4e0e8e75ddf9b9892a2f2e6 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed May 20 18:49:28 2009 +0100 [xlib] Handle too-large images gracefully. Trigger a fallback to an image surface for massive patterns. src/cairo-xlib-surface.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 3c6838532a62c294e7f18b13b0b7f8da9a1a1e1e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed May 20 18:46:35 2009 +0100 [surface] Early return if the backend cannot clone an image If the backend cannot create a similar surface to hold the image, then report back the failure before embarking upon an infinite recursion. The UNSUPPORTED return should percolate up through the call stack and trigger a fallback. src/cairo-surface.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) commit bf2c7356d4c955fb770863874a0ea111e9ba02a4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri May 22 12:52:43 2009 +0100 [surface] Provide nil-surface for INVALID_SIZE So that we can faithfully report back failure if the user tries to create a surface larger than the backend can support. src/cairo-surface.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) commit 6717f0d748000416172057d0aab2471377456e27 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed May 20 17:35:19 2009 +0100 [win32] Wrap win32-font usage with CAIRO_HAS_WIN32_FONT Bug 19251: --enable-win32=yes and --enable-win32-font=no causes compilation failure [https://bugs.freedesktop.org/show_bug.cgi?id=19251] Wrap use of the win32 font backend within win32-printing-surface within CAIRO_HAS_WIN32_FONT ifdefs to prevent compilation failure if the user explicitly disables the win32 font backend. src/cairo-win32-printing-surface.c | 4 ++++ 1 file changed, 4 insertions(+) commit c7d2b0de830f4c653fbb5016c4efdc5f0e5af0eb Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed May 20 17:34:56 2009 +0100 [win32] Typo Correct function name in error string. src/cairo-win32-printing-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 87b764908a38cbb4159ac76d8b7d1d08a24b838e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue May 19 15:53:55 2009 +0100 [gl] Add EGL target Split the GLX context from the GL surface to enable use of an alternative EGL interface. boilerplate/Makefile.win32.features | 20 ++ boilerplate/cairo-boilerplate-gl.c | 2 +- build/Makefile.win32.features | 2 + build/Makefile.win32.features-h | 6 + build/configure.ac.features | 3 + configure.ac | 30 ++- src/Makefile.sources | 4 +- src/Makefile.win32.features | 28 +++ src/cairo-egl-context.c | 181 ++++++++++++++++++ src/cairo-gl-private.h | 88 +++++++++ src/cairo-gl-surface.c | 191 ++++++++----------- src/cairo-gl.h | 41 +++-- src/cairo-glx-context.c | 136 ++++++++++++++ test/.gitignore | 2 + test/Makefile.am | 10 + test/egl-flowers.c | 358 ++++++++++++++++++++++++++++++++++++ test/glx-flowers.c | 256 ++++++++++++++++++++++++++ util/cairo-script/.gitignore | 2 + util/cairo-script/Makefile.am | 15 ++ util/cairo-script/csi-egl.c | 263 ++++++++++++++++++++++++++ util/cairo-script/csi-glx.c | 150 +++++++++++++++ util/cairo-script/csi-replay.c | 61 ++++++ 22 files changed, 1723 insertions(+), 126 deletions(-) commit 6e87e354e42d1a758a252533407e6e1829638544 Author: Adrian Johnson <ajohnson@redneon.com> Date: Sun May 17 18:14:28 2009 +0930 Correct the comment for the index_to_ucs4 font backend function src/cairoint.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit 2a34992cccfd77c2acf30fe851311f16137ba32f Author: Adrian Johnson <ajohnson@redneon.com> Date: Sun May 17 18:12:39 2009 +0930 Ensure win32 font index_to_ucs4() sets ucs4 to -1 if lookup fails src/cairo-win32-font.c | 1 + 1 file changed, 1 insertion(+) commit 4c8e5485a10478c19e8a6cb48595473f5bb8ee8a Author: Adrian Johnson <ajohnson@redneon.com> Date: Sun May 17 18:05:42 2009 +0930 Fix win32 build breakage that was caused by c25992479aca481d326f72665ebdcf0904273eac src/cairo-win32-private.h | 1 + 1 file changed, 1 insertion(+) commit 31596cf2b298054bbd3e340ae77a3388eadc0d8b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri May 15 17:40:26 2009 +0100 [debug] Check image contents using memcheck. As an aide to tracking down the source of uninitialised reads, run VALGRIND_CHECK_MEM_IS_DEFINED() over the contents of image surfaces at the boundary between backends, e.g. upon setting a glyph image or acquiring a source image. src/cairo-debug.c | 42 ++++++++++++++++++++++++++++++++++++++++++ src/cairo-ft-font.c | 2 ++ src/cairo-pattern.c | 2 ++ src/cairo-png.c | 2 ++ src/cairo-scaled-font.c | 3 +++ src/cairo-surface.c | 32 +++++++++++++++++++++++--------- src/cairoint.h | 11 +++++++++++ 7 files changed, 85 insertions(+), 9 deletions(-) commit 791a6fa399894e735d522781b15b7f18c3ecd4d1 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri May 15 16:57:49 2009 +0100 [memfault] Update macros to avoid namescape collision with memcheck Basing the macro names of the memfault skin lead to a namespace collision with memcheck. After updating the headers, update cairo's usage to match. src/cairo-malloc-private.h | 2 +- test/cairo-test.c | 62 +++++++++++++++++++++++----------------------- 2 files changed, 32 insertions(+), 32 deletions(-) commit c897f36a6b09f3591d858142e3236f256f0bd6a2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri May 15 20:37:52 2009 +0100 [test] Stress the ft from-face cache Create a set of font faces using the same FT_Face to stress test the handling of from-face fonts within the backend cache. test/ft-font-create-for-ft-face.c | 109 +++++++++++++++++++++++++++++++++++--- 1 file changed, 102 insertions(+), 7 deletions(-) commit 477df1f5504a507d0c5960aa7e21375284a6f99c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri May 15 10:48:56 2009 +0100 [test] Clear expected floating point exceptions test/invalid-matrix deliberately feeds garbage into the API to test our error detection. This causes FPE to be raised during the course of the test - so they are deliberately disable for the duration. However, the exceptions were not being cleared and so the FPE could be triggered on the next floating point operation. This was being masked during make check, by the fact that each test is run in its own forked process and was only observed when multiple tests were run in foreground mode. build/configure.ac.system | 2 +- test/invalid-matrix.c | 4 ++++ 2 files changed, 5 insertions(+), 1 deletion(-) commit f47a93ef30249b531e588335045d052c3dcc71bd Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat May 9 09:55:49 2009 +0100 [test] Check text->path with user-fonts (twin) Check behaviour of user-fonts, twin in particular, when using the text as a path. test/twin.c | 11 ++++++++++- test/twin.ps.ref.png | Bin 1114 -> 2197 bytes test/twin.ref.png | Bin 1492 -> 4038 bytes test/twin.svg.ref.png | Bin 1487 -> 3027 bytes 4 files changed, 10 insertions(+), 1 deletion(-) commit 669e7ae5c98812360bf20b798c20024ae2e9810b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri May 8 17:36:01 2009 +0100 [test] Relax APPROX_EQUALS condition The bounding box of the text is rounded to the nearest pixel boundaries, so therefore the test must accept a similar level of imprecision. test/get-path-extents.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) commit fe7410c6cdc19bb366bf1e93481bf8287a14d52e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat May 9 19:01:06 2009 +0100 [test] Add a note to "Getting the elusive zero failures" The test suite depends upon many external factors and in order to achieve a pass, your system must match that upon which the reference images were generated. Add a note to read test/README in case of failures so that the casual user is not unduly alarmed by cairo reportedly failing. test/cairo-test-runner.c | 12 ++++++++++++ 1 file changed, 12 insertions(+) commit 2f962799a77faef0e6ad62c79bd7eebc9b8c5a4b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri May 15 18:49:03 2009 +0100 [test] Fix typos that excluded backend test sources test/Makefile.am | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) commit c25992479aca481d326f72665ebdcf0904273eac Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat May 9 10:10:14 2009 +0100 [xlib] Use minimal depth for similar clones. Damian Frank noted [http://lists.cairographics.org/archives/cairo/2009-May/017095.html] a performance problem with an older XServer with an unaccelerated composite - similar problems will be seen with non-XRender servers which will trigger extraneous fallbacks. The problem he found was that painting an ARGB32 image onto an RGB24 destination window (using SOURCE) was going via the RENDER protocol and not core. He was able to demonstrate that this could be worked around by declaring the pixel data as an RGB24 image. The issue is that the image is uploaded into a temporary pixmap of matching depth (i.e. 32 bit for ARGB32 and 24 bit for RGB23 data), however the core protocol can only blit between Drawables of matching depth - so without the work-around the Drawables are mismatched and we either need to use RENDER or fallback. This patch adds a content mask to _cairo_surface_clone_similar() to provide the extra bit of information to the backends for when it is possible for them to drop channels from the clone. This is used by the xlib backend to only create a 24 bit source when blitting to a Window. src/cairo-clip.c | 1 + src/cairo-directfb-surface.c | 2 ++ src/cairo-glitz-surface.c | 3 +++ src/cairo-image-surface.c | 4 ++++ src/cairo-pattern.c | 18 +++++++++++++--- src/cairo-quartz-surface.c | 1 + src/cairo-surface-fallback-private.h | 1 + src/cairo-surface-fallback.c | 3 ++- src/cairo-surface.c | 7 +++++- src/cairo-win32-surface.c | 3 ++- src/cairo-xcb-surface.c | 5 +++++ src/cairo-xlib-surface.c | 42 +++++++++++++++++++++++++++--------- src/cairoint.h | 4 ++++ src/test-fallback-surface.c | 1 + src/test-fallback16-surface.c | 2 ++ test/surface-source.c | 1 + 16 files changed, 82 insertions(+), 16 deletions(-) commit 0238fe2cafea2e1ed19bb222117bd73ee6898d4d Author: Karl Tomlinson <karlt+@karlt.net> Date: Thu May 14 11:46:29 2009 +0100 [ft] Resolve mutual referencing problems with zombie faces Bug 21706 -- zombie ft_font_face / ft_unscaled_font mutual referencing problems [http://bugs.freedesktop.org/show_bug.cgi?id=21706] There can be more than one zombie font_face belonging to an unscaled_font, but only the first is destroyed. This leaks the client's FT_Face (and associated font data) as release of the FT_Face depends on release of the font_face. (The reason why Firefox ends up with two different font_faces for one unscaled_font is that load_flags for faces with artificial oblique have FT_LOAD_NO_BITMAP set. https://bugzilla.mozilla.org/show_bug.cgi?id=486974) Also it's possible for _cairo_ft_font_face_create to pull out a zombie font_face from the unscaled_font, which would crash _cairo_ft_font_face_scaled_font_create, as that expects non-null font_face->unscaled (if !font-face->pattern). AUTHORS | 2 +- src/cairo-ft-font.c | 38 ++++++++++++++++++++++++++++++-------- 2 files changed, 31 insertions(+), 9 deletions(-) commit d6f6ec9082c86b9fd9e2389b9627f08a91c2cdd3 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri May 8 16:43:43 2009 +0100 [ft] Restore the ability to lazily resolve patterns. I broke the ability for the ft font backend to resolve patterns whilst fixing the font creation to propagate the error status from fontconfig (be27e8). By adjusting the sequence of error checks we do not confuse the absence of a match with a fatal error and thereby restoring the lazy pattern resolution whilst ensuring error propagation. src/cairo-ft-font.c | 32 +++++++++++++++++--------------- 1 file changed, 17 insertions(+), 15 deletions(-) commit a352fd46020e18f9d9f839f0c3f3a63c1d8c0ae1 Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Fri May 15 15:34:05 2009 -0400 Only include <strings.h> for ffs() when we HAVE_FFS This fixes breakage caused by 3d3d10a31e04498ef91a288d89b3dec291bfc323 src/cairo-skiplist.c | 2 ++ 1 file changed, 2 insertions(+) commit 882f1d5a60abfbf83c65fe00b31bb4c8e5c82f62 Author: Eric Anholt <eric@anholt.net> Date: Sun May 10 16:56:49 2009 -0700 [gl] Use spans directly as geometry instead of rasterizing to a temporary. I was hoping for a bigger win than this, but cairogears is now significantly faster than it was with just traps. One potential problem is that adding src texture coordinates blew up vertex size from 12 bytes to 20, and we're emitting quite a few vertices. There's plenty of hope for the future, though: With shaders we could remove the 2 new texcoords again by stuffing a position to source coordinate transformation matrix in constants. Then with EXT_geometry_shader4 plus EXT_gpu_shader4 we should be able to get the 24 bytes for 2 vertices down to 16 or less. src/cairo-gl-surface.c | 316 ++++++++++++++++++++++++++----------------------- 1 file changed, 171 insertions(+), 145 deletions(-) commit 7b4f2d5e6dd37232d4fa764b69065c7d96856dcc Author: Eric Anholt <eric@anholt.net> Date: Sun May 10 14:50:22 2009 -0700 [gl] Use Enable/DisableClientState for arrays for consistency. src/cairo-gl-surface.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) commit 705ee3da1372e8533f25e0b4bde072e4bf76872b Author: Eric Anholt <eric@anholt.net> Date: Sat May 9 12:17:12 2009 -0700 [gl] Add HW spans implementation using VBOs and GL_LINES. This is a minor performance regression over software, but is the basis for not rasterizing the geometry into a temporary mask, which should be a significant win. src/cairo-gl-surface.c | 145 ++++++++++++++++++++++++++++++++++++++++++------- 1 file changed, 124 insertions(+), 21 deletions(-) commit 934af24ccce9c06fc38ab7735d643d230fb3b8ad Author: Eric Anholt <eric@anholt.net> Date: Sat May 9 10:03:06 2009 -0700 [gl] Use the span renderer with an image surface temporary. This appears to be no performance difference in cairogears versus the traps fallback before, but is a step towards handing the information off to the hardware. src/cairo-gl-surface.c | 210 ++++++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 208 insertions(+), 2 deletions(-) commit 17b41fe7e3b8b493341be6384d816105aadf0cb6 Author: Eric Anholt <eric@anholt.net> Date: Sun May 10 10:17:27 2009 -0700 [gl] Use filtering of GL_NEAREST to hint to TexImage not to alloc for mipmaps. Gl defaults to NEAREST_MIPMAP_LINEAR on new texture objects. This should save space or bandwidth with DRI drivers, as they'll use the parameter as a hint, and then not have to either fix up later or waste space for unused mipmap levels. src/cairo-gl-surface.c | 7 +++++++ 1 file changed, 7 insertions(+) commit a438acd37c5aaedc7e7a05e6a9d8758315931b44 Author: Eric Anholt <eric@anholt.net> Date: Sat May 9 11:33:53 2009 -0700 [gl] Handle PIXMAN_b8g8r8a and PIXMAN_b8g8r8x8 in the image tex shortcut. src/cairo-gl-surface.c | 6 ++++++ 1 file changed, 6 insertions(+) commit 29768c2193b09d3bef54a178c8e3ba6f7e7cebd0 Author: Eric Anholt <eric@anholt.net> Date: Sat May 9 11:30:11 2009 -0700 [gl] Don't forget that we require ARB_texture_non_power_of_two currently. src/cairo-gl-surface.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit 088d2a6cc70b01f55d178d235542247180b3e5ee Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu May 7 11:35:17 2009 +0100 [doc] Update identifiers doc/public/tmpl/cairo-status.sgml | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit d8fb6a03d3034cf776687bcc1f0625a8a9f2ae07 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu May 7 14:57:13 2009 +0100 [path] Define _BSD_SOURCE to enable hypot() hypot() is only declared for BSD/SVID/XOPEN/C99 sources. Choose BSD as we've used that elsewhere. src/cairo-path-stroke.c | 1 + 1 file changed, 1 insertion(+) commit e38f85c5de51016b8858e755e2752816c9995ab2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu May 7 14:42:15 2009 +0100 [script] Pedantic violations Fixup a few trivial errors with -pedantic. util/cairo-script/cairo-script-operators.c | 2 +- util/cairo-script/cairo-script-private.h | 8 ++++---- 2 files changed, 5 insertions(+), 5 deletions(-) commit 0c00556990a842e0ee0056134c6e196c2fb372b7 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu May 7 14:11:12 2009 +0100 [script] Missing include for INT32_MAX Avoid depending upon stdint.h, or rather cut'n'pasting Cairo's portable defines, by using INT_MAX and limits.h instead. util/cairo-script/cairo-script-file.c | 3 ++- util/cairo-script/cairo-script-objects.c | 3 ++- util/cairo-script/cairo-script-operators.c | 13 +++++++------ util/cairo-script/cairo-script-scanner.c | 5 +++-- util/cairo-script/cairo-script-stack.c | 7 ++++--- 5 files changed, 18 insertions(+), 13 deletions(-) commit 3d3d10a31e04498ef91a288d89b3dec291bfc323 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu May 7 14:10:47 2009 +0100 [skiplist] Missing include for ffs() src/cairo-skiplist.c | 2 ++ 1 file changed, 2 insertions(+) commit e5752865ab0ebefec313ab9a6f90451ef2d7af95 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu May 7 14:58:45 2009 +0100 [cairo] Protect typeof against -ansi s/typeof/__typeof__/ to allow the [gcc-specific] macro to continue to work despite -ansi. src/cairoint.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 5f4f2ab01ab3fa6b07ac7521dfeaf90a787a5dee Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu May 7 14:34:47 2009 +0100 [cairo] Remove stray semi-colon Pedantic finds another victim. src/cairoint.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit d86ad461cf7a9d857bd07a6b775c84acbf35647b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu May 7 14:16:05 2009 +0100 [cairo] Convert C99 comments to /* ... */ First victim of -pedantic... src/cairoint.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 695f648607dd5a78cf29dcc33ed19aa223d0416a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu May 7 11:16:10 2009 +0100 [cairo] Fix errors reported by check-doc Identity a few new instances of CAIRO_STATUS_USER_FONT_NOT_IMPLEMENTED in comments with %. src/cairo.h | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) commit 849bdee199b5ebba467603a69d30bb519c10bc75 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu May 7 11:36:38 2009 +0100 [check] Quieten output during checking headers-standalone Remove the duplication of the compile line and enable shave to replace it entirely with a "CHECK $header". This should make errors more obvious. build/shave.in | 3 +++ src/Makefile.am.analysis | 2 +- 2 files changed, 4 insertions(+), 1 deletion(-) commit 84fad4b11bc26cfd0847660e3309eb902d783713 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu May 7 11:31:22 2009 +0100 [build] Fix gtk-doc interoperation with shave Add --tag=CC to the libtool invocations. build/Makefile.am.gtk-doc | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit fd72c56af833a738ddc8f188cabe03d9f8b45475 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu May 7 11:27:43 2009 +0100 [build] Report status of gtk-doc in summary build/configure.ac.features | 1 + 1 file changed, 1 insertion(+) commit 428fb58a4c96391f96f48026be49a4533cd12834 Author: Behdad Esfahbod <behdad@behdad.org> Date: Tue Apr 28 15:03:53 2009 -0400 [user-font] Add CAIRO_STATUS_USER_FONT_NOT_IMPLEMENTED This is useful for language bindings to signal that a method is not implemented. src/cairo-misc.c | 2 ++ src/cairo-surface.c | 1 + src/cairo-user-font.c | 14 +++++++++++--- src/cairo.h | 23 ++++++++++++++--------- 4 files changed, 28 insertions(+), 12 deletions(-) commit f9573d03dde2c38674d36b783044715ab5104b0a Author: Behdad Esfahbod <behdad@behdad.org> Date: Tue Mar 17 19:48:12 2009 -0400 Err, make gtk-doc happy again doc/public/tmpl/cairo-status.sgml | 2 +- src/cairo.h | 2 -- 2 files changed, 1 insertion(+), 3 deletions(-) commit 30735f790aa3cef822f132932f43a4738cd8cd95 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue May 5 18:37:25 2009 +0100 [xlib] Use a similar content surface for cloning images Simply request a surface with a similar content to the source image when uploading pixel data. Failing to do so prevents using a 16-bit (or otherwise non-standard pixman image format) window as a source - in fact it will trigger an infinite recursion. src/cairo-xlib-surface.c | 16 +++++++++------- 1 file changed, 9 insertions(+), 7 deletions(-) commit 60e38d0530443aa9c78e74e47ba5574887ae220c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue May 5 17:05:39 2009 +0100 [surface] Convert FORMAT_INVALID during snapshots Currently the surface snapshotting attempts to clone the source using a new surface of identical format. This will raise an error if the source is an unusual xserver, for example one operating at 16bpp. The solution to this is to create the surface using the content type rather than the format (as elsewhere within the code base). However, we also wish to preserve FORMAT_A1 (which is lost if we only choose the format based on _cairo_format_from_content) as the various backends may be able to trivially special case such bitmaps. src/cairo-surface-fallback.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) commit 31f5a2e94d669b9d2785de944d4aee584fd1d76e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue May 5 15:52:55 2009 +0100 [png] Coerce FORMAT_INVALID to a known image format It is possible for cairo_surface_write_to_png() to acquire a non-standard image surface when, for example, we try to dump a low bit-depth XServer. Handle this scenario by coercing the unknown image format to a standard type via pixman. src/cairo-png.c | 44 ++++++++++++++++++++++++++++++-------------- 1 file changed, 30 insertions(+), 14 deletions(-) commit c488b336449a1a7ca4d3f90785afeec9e21784c3 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue May 5 13:59:07 2009 +0100 [test] Create a new fallback surface to exercise 16-bit paths. Add a variation of test-fallback-surface that forces the use of a 16-bit pixman format code instead of the standard 32-bit types. This creates an image surface akin to the fallbacks used with low bit-depth xservers. .../cairo-boilerplate-test-surfaces-private.h | 11 + boilerplate/cairo-boilerplate-test-surfaces.c | 16 ++ boilerplate/cairo-boilerplate.c | 18 ++ src/Makefile.sources | 2 + src/test-fallback16-surface.c | 234 +++++++++++++++++++++ src/test-fallback16-surface.h | 52 +++++ test/Makefile.am | 4 + test/Makefile.sources | 3 + test/test-fallback16-surface-source.c | 43 ++++ test/test-fallback16-surface-source.ref.png | Bin 0 -> 268 bytes 10 files changed, 383 insertions(+) commit 1d609d672273da494fd596606b59ab1c0010ae6d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue May 5 17:11:55 2009 +0100 [image] Treat A1 image surfaces as BILEVEL_ALPHA Categorise the transparency of FORMAT_A1 image surfaces as BILEVEL_ALPHA. src/cairo-image-surface.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) commit aac132a76a2af3719088678295169f1962a555e6 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Apr 24 10:14:36 2009 +0100 [image] Make _cairo_image_analayze_transparency() more format agnostic Use the content in preference to the format to determine CAIRO_IMAGE_IS_OPAQUE/CAIRO_IMAGE_HAS_ALPHA. src/cairo-image-surface.c | 17 +++++++---------- 1 file changed, 7 insertions(+), 10 deletions(-) commit 4be6de9fc6192d59c8d61e8edafed941e868a756 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Apr 24 10:03:20 2009 +0100 [image] Return FORMAT_INVALID for an error surface. The default error value should be CAIRO_FORMAT_INVALID [-1] not 0 (which corresponds to CAIRO_FORMAT_ARGB32). src/cairo-image-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 6675cf558719f81afe2a4331bc6adb3cda637a26 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri May 1 09:31:53 2009 +0100 [build] Add lcov-1.7 to known list build/configure.ac.analysis | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 526fcdb7e6cc8b522508762b1a68a5585fddf823 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Apr 20 10:56:06 2009 +0100 [build] Enable shave support shave transforms the messy output of autotools into a pretty (quiet!) Kbuild-like one. Lets see how controversial a simple change can be... build/.gitignore | 2 ++ build/aclocal.dolt.m4 | 1 + build/aclocal.shave.m4 | 77 ++++++++++++++++++++++++++++++++++++++++++++++++ build/shave-libtool.in | 69 +++++++++++++++++++++++++++++++++++++++++++ build/shave.in | 79 ++++++++++++++++++++++++++++++++++++++++++++++++++ configure.ac | 4 +++ 6 files changed, 232 insertions(+) commit 1ae2ddc1dd4c90d50b8c57c4de677f8ab96b1fa2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Apr 16 17:19:12 2009 +0100 [memfault] Manually inject faults when using stack allocations Ensure that no assumptions are made that a small allocation will succeed by manually injecting faults when we may be simply allocating from an embedded memory pool. The main advantage in manual fault injection is improved code coverage - from within the test suite most allocations are handled by the embedded memory pools. src/cairo-array.c | 3 +++ src/cairo-bentley-ottmann.c | 3 +++ src/cairo-gstate.c | 3 +++ src/cairo-hull.c | 3 +++ src/cairo-image-surface.c | 6 ++++++ src/cairo-malloc-private.h | 7 +++++++ src/cairo-misc.c | 3 +++ src/cairo-path-stroke.c | 2 ++ src/cairo-pattern.c | 9 +++++++++ src/cairo-pen.c | 9 +++++++++ src/cairo-polygon.c | 5 +++++ src/cairo-scaled-font.c | 3 +++ src/cairo-stroke-style.c | 3 +++ src/cairo-traps.c | 5 +++++ 14 files changed, 64 insertions(+) commit 817589e1967ebdd5e5bda1781eb76010fd8c37dc Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Apr 16 13:01:52 2009 +0100 [test] Call FcInit() manually. Pre-initialise fontconfig whilst memfault is disabled to avoid a lot of expensive, redundant testing of FcInit() throughout the test suite. test/cairo-test.c | 6 ++++++ 1 file changed, 6 insertions(+) commit be27e844c83c0f5cf25dee1c62768dbf70897a06 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Apr 16 17:13:52 2009 +0100 [ft] Propagate status from font creation Return the true error status whel _cairo_ft_unscaled_font_create_internal(). This ensures that the original error is not masked and we are able to report the error during fontconfig pattern resolution. src/cairo-ft-font.c | 90 ++++++++++++++++++++++++++++++----------------------- 1 file changed, 51 insertions(+), 39 deletions(-) commit d46c56f18c8a93d3d50be9333292c7c9b0ac0a78 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Apr 16 13:00:59 2009 +0100 [ft] Check for pattern duplication failure. Check the return of FcPatternDuplicate() for failure, and propagate the error. src/cairo-ft-font.c | 2 ++ 1 file changed, 2 insertions(+) commit 50302f156dfba4dcb3639843d7eda819f2c6797c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Apr 16 12:59:36 2009 +0100 [ft] Check (correctly!) for pattern duplication failure. Check the return from FcPatternDuplicate() for allocation failure (and not the original pattern)! src/cairo-ft-font.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 8bf109bd2a9c9bf755671185733def898a4f96f2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Apr 16 17:17:06 2009 +0100 [pattern] Silence compiler with impossible case. Assert that the pattern is one of the four known types, and return an error so that the compiler knows that the local variable can not be used uninitialised. src/cairo-pattern.c | 3 +++ 1 file changed, 3 insertions(+) commit 5a76c233bffda65140c6ada9b5f2ed7d3bcc999e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Apr 14 11:02:40 2009 +0100 [type3] Propagate error from font creation. Perform an error check after creating a scaled-font to prevent operations on a nil-object. src/cairo-type3-glyph-surface.c | 2 ++ 1 file changed, 2 insertions(+) commit d46bd67c8b3f63d6286845e814f8f1e040f89737 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Apr 10 14:55:24 2009 +0100 [type3] Add status checks. Propagate error from the stream when creating the surface and add status guards before writing to the potential nil object. src/cairo-type3-glyph-surface.c | 26 +++++++++++++++++++++++--- 1 file changed, 23 insertions(+), 3 deletions(-) commit a6ffb69a54f15659dac7c6b2e1c98bd7893ccdf1 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Apr 3 17:28:05 2009 +0100 [type3] Check error status before emit_image(). Be careful not to pass an error object down to emit_image() hook, propagate the error instead. This relieves the burden of the error check from the callee -- which was causing an assertion failure in the ps backend. src/cairo-type3-glyph-surface.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) commit 21d1138da9bcf9c006e0cf3f3d1d50e1c7cec1de Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Apr 10 14:53:55 2009 +0100 [scaled-font-subsets] Check for malloc failure. Check that the utf8 string is successfully allocated before writing to it, otherwise propagate the error status back to the callers. src/cairo-scaled-font-subsets.c | 63 +++++++++++++++++++++++------------------ 1 file changed, 35 insertions(+), 28 deletions(-) commit 0f0e2d738437176e72317b37ef66fbc56c60ba93 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Apr 10 14:52:48 2009 +0100 [pdf] Propagate failure from type3 glyph surface creation. Add an early check that we successfully created the glyph surface. src/cairo-pdf-surface.c | 11 +++++++++++ 1 file changed, 11 insertions(+) commit 1496c5cf51a2304dfd4e9d359a0645ff44fadd3e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Apr 10 14:50:26 2009 +0100 [analysis] Use _cairo_region_init() Avoid secondary allocations of the thin region wrappers during surface creation by embedding them into the parent structure. This has the satisfactory side-effect of not requiring status checks which current code lacks. src/cairo-analysis-surface.c | 26 +++++++++++++------------- 1 file changed, 13 insertions(+), 13 deletions(-) commit ad83765a1441cb369c3e7f3de77c243297406c7a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Apr 3 09:23:58 2009 +0100 [xlib] Allocate bounded region on stack. Eliminate the extremely short-lived and oft unnecessary heap allocation of the region by first checking to see whether the clip exceeds the surface bounds and only then intersect the clip with a local stack-allocated region. src/cairo-region.c | 23 +++++++++++++++++++ src/cairo-xlib-surface.c | 58 +++++++++++++++++++++++++++++++----------------- src/cairoint.h | 10 +++++++++ 3 files changed, 71 insertions(+), 20 deletions(-) commit 75b06b8bdb5efaf030e063c1f66583d8fde347dc Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Apr 1 11:43:16 2009 +0100 [pattern] Do not modify outparam on error path. The pdf backend was crashing as a failed pattern copy stored an invalid pointer with the resource group. src/cairo-pattern.c | 21 +++++++++++---------- 1 file changed, 11 insertions(+), 10 deletions(-) commit 82cccb26723697492504f395fed33afba28d84ba Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Apr 14 11:01:39 2009 +0100 [test] Handle memfaults during surface-finish-twice Check and report memory allocation failures. test/surface-finish-twice.c | 17 +++++++++++------ 1 file changed, 11 insertions(+), 6 deletions(-) commit f12d52bfcae0792729b33a1fd1b2d23bd3fc7cf3 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Apr 4 19:02:15 2009 +0100 [test] Use xmalloc() to evade memfault. Do not use the simple malloc() as memfault will inject allocation failures (unlike xmalloc() for which faults are excluded) - as this is unnecessary inside the test harness and thus does not exercise any cairo code paths. test/show-glyphs-many.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit ca501d99bbec432252cbb134d863801d4a792a44 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Apr 1 11:41:59 2009 +0100 [test] Disable signal handling under valgrind Capturing and handling the signals prevents valgrind from providing backtraces for fatal signals -- which is often more useful. test/cairo-test.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit f8539726368141f8b42c7817cfd59b4abfce33f3 Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Tue Apr 7 14:50:25 2009 -0400 [win32] Fix building with libpng and zlib Include zlib and libpng dependencies using something like '$(cairo_dir)/../zlib/zdll.lib' instead of just 'zdll.lib'. Also, do similarly for the headers. build/Makefile.win32.common | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) commit 1798fc6607a89811929d1f21cc71c751438f5699 Author: Adrian Johnson <ajohnson@redneon.com> Date: Sun Apr 5 11:17:02 2009 +0930 Fix the TrueType subsetting fix in 9b496af5c Oops. src/cairo-truetype-subset.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit fe97e815e81f4429889fd52bb525c260f9967356 Author: Adrian Johnson <ajohnson@redneon.com> Date: Sun Apr 5 01:04:12 2009 +1030 Fix TrueType subsetting bug Fix incorrect counting of arguments in composite glyphs src/cairo-truetype-subset-private.h | 2 +- src/cairo-truetype-subset.c | 12 +++++++----- 2 files changed, 8 insertions(+), 6 deletions(-) commit 9b496af5c214283c297a52c90a04743f9da9f6a3 Author: Adrian Johnson <ajohnson@redneon.com> Date: Sun Apr 5 00:44:02 2009 +1030 Fix buffer size check in TrueType subsetting Andrew Cowie reported a problem with the Charis SIL font being embedded as a fallback font. The buffer size check for composite glyphs was incorrect causing the subsetting to fail for some fonts. src/cairo-truetype-subset.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit ba2ec8e1d9144c4a27917e59cb15039552eb008c Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Wed Apr 1 13:29:12 2009 -0400 Remove trailing comma in cairo_region_overlap_t enum. src/cairo.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 2d6336624c33e9c674c192b68eb501373a96dc87 Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Wed Apr 1 11:32:58 2009 -0400 [test-win32] Fix static linking of the test suite. Previously, the test suite needed to be dynamically linked to work. boilerplate/Makefile.win32 | 2 +- test/Makefile.win32 | 9 ++++++--- test/pdiff/Makefile.win32 | 2 +- 3 files changed, 8 insertions(+), 5 deletions(-) commit 0f21893b0510a188439c2d8cd448bcfcf17ef8c6 Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk> Date: Wed Apr 1 08:54:30 2009 -0400 [region] pixman_region32_contains_point() does not allow NULL for box src/cairo-region.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) commit e46c1d7fa34b4ba89fc3e0fe6f3042a6fa8c0398 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Mar 31 10:39:46 2009 +0100 [scaled-font-subset] Cleanup after failure to convert to utf16. Avoid leaking the local hashtable and strings after failing to convert the string to utf16. src/cairo-scaled-font-subsets.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 180b964aac4e058e6783778d33772f08e13b3669 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Mar 31 10:37:44 2009 +0100 [scaled-font-subset] Propagate error from scaled-font to collection. Don't attempt to collect the sub_font if it is in error. src/cairo-scaled-font-subsets.c | 4 ++++ 1 file changed, 4 insertions(+) commit 8362c6f726979abc43ad9f7303bd45fcb03f83b3 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Mar 31 10:36:08 2009 +0100 [type1] Fixup error path during write_charstrings() On the common error path we attempted to unlock a mutex that was not always held, so reorder the error paths appropriately. src/cairo-type1-fallback.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) commit 88956cd42e9cc03cb8b4ec730062993eaaf3938d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Mar 31 10:35:24 2009 +0100 [ps] Check for error during stroking. Add a missing error status check that caused errors during stroke to be masked. src/cairo-ps-surface.c | 3 +++ 1 file changed, 3 insertions(+) commit 043352aa8d1e3aeacf3b877f45e1bc451a676e15 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Mar 31 10:34:41 2009 +0100 [pdf] Prevent leak of pad_image on error path. Ensure that the local pad_image is destroyed after an error. src/cairo-pdf-surface.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit b0689f56118dd8ccda6f29901d41cf8f80983aa0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Mar 31 10:27:15 2009 +0100 Check for errors during get_mime_data() cairo_surface_get_mime_data() may raise an error on the surface, so we need to check lest it goes unnoticed and we generate a corrupt file. src/cairo-pdf-surface.c | 2 ++ src/cairo-ps-surface.c | 2 ++ src/cairo-svg-surface.c | 2 ++ 3 files changed, 6 insertions(+) commit 9f63cbb870892253f363ddb7aac908263672c8dc Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Mar 30 19:09:13 2009 +0100 [pdf] Free compressed stream after error. Ensure that the compressed stream is destroyed after encountering an error. src/cairo-pdf-surface.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) commit ba1a0fa601a817ff489bc5373af57977e41eb99f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Mar 30 19:08:41 2009 +0100 [truetype] Free local names on failure. Ensure that all local allocations are freed on the error path. src/cairo-truetype-subset.c | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-) commit edce97a750acf4368bd7249ec6b9a195f8584cdf Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Mar 30 16:50:51 2009 +0100 [pdf] Remove false assertion. The stream itself may be in an error state, so an error could be raised. src/cairo-pdf-surface.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) commit b580a4a8d6c056ba8b47be32ea8c5b9a1d90d01e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Mar 30 16:50:10 2009 +0100 [test] Improve fault injection coverage In order to exercise the meta-surfaces, we need to inject faults into cairo_surface_finish(). boilerplate/cairo-boilerplate.c | 16 +++++++++++++--- test/cairo-test.c | 35 +++++++++++++++++++++++++++++++++++ 2 files changed, 48 insertions(+), 3 deletions(-) commit 80d5b53b474c94219c51ae16dcc19098a2d86165 Author: Søren Sandmann Pedersen <sandmann@redhat.com> Date: Tue Mar 31 05:47:44 2009 -0400 [region] Change name of cairo_region_empty() to cairo_region_is_empty() src/cairo-clip.c | 2 +- src/cairo-region.c | 6 +++--- src/cairo-surface-fallback.c | 2 +- src/cairo.h | 2 +- src/cairoint.h | 2 +- 5 files changed, 7 insertions(+), 7 deletions(-) commit 77ee65fd03d06064be023f022d565c5038fe26df Author: Jonathan Kew <jfkthame@gmail.com> Date: Mon Mar 30 14:45:48 2009 -0400 [win32] Fix horizontal glyph positioning bug The _cairo_win32_scaled_font_backend version of show_glyphs collects glyph runs to hand to ExtTextOutW until the y-offset changes, then flushes the glyphs buffered so far. As each glyph is buffered, it also calculates and buffers the dx value for the preceding glyph. However, when it sees a change in dy and decides to flush, it should *not* append an entry to the dx buffer, as this would be the "dx" of the previous glyph, and instead the new start_x value will be used for the new glyph run that's being collected. This bug means that after any vertically-offset glyph, the remaining glyphs in the run will get incorrect dx values (horizontal escapement). Mozilla bug #475092 src/cairo-win32-font.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) commit aee71e2063b1d6d23cd8dcef7789c9cf106af32a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Mar 30 13:41:00 2009 +0100 [test] Memfault checks. Don't assume an error means the test failed, check for injected allocation errors. test/create-from-png.c | 164 ++++++++++++++++++++++++++++++++++--------------- 1 file changed, 113 insertions(+), 51 deletions(-) commit ea6197c2f5f04d5e8e8035a330c5199b37beb702 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Mar 30 10:54:26 2009 +0100 [surface] Propagate region allocation failure. Propagate the error status from failing to allocate the region. src/cairo-surface.c | 21 ++++++++++----------- 1 file changed, 10 insertions(+), 11 deletions(-) commit c35d226f7d3654b312e18068b8ccb47a51002a39 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Mar 30 10:46:37 2009 +0100 [traps] Propagate allocation failure. Report failure to allocation region. src/cairo-traps.c | 23 ++++++++++------------- 1 file changed, 10 insertions(+), 13 deletions(-) commit e238d10b30e18084bf07eb1643fdf48ef35467ab Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Mar 30 10:22:18 2009 +0100 [region] Use const cairo_rectangle_int_t consistently. Add the const declaration to a couple of functions. src/cairo-region.c | 4 ++-- src/cairo.h | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) commit de1612bdd767ca37e01938f8e41d9699531a49d9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Mar 30 10:17:49 2009 +0100 [region] Use _cairo_status_is_error Replace the open-coded version with the more readable macro. src/cairo-region.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit f027405429d0133b2840c4b82bc553355fa5f3d2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Mar 30 10:12:41 2009 +0100 [region] Add leading underscore to private _cairo_region_set_error() Differentiate the private _cairo_region_set_error() function by using a leading underscore. src/cairo-region.c | 24 ++++++++++++------------ 1 file changed, 12 insertions(+), 12 deletions(-) commit ed7188a471f73abcc4ca3e2a92685088134391c7 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Mar 30 10:07:48 2009 +0100 [region] Add slim_hidden_def. Fixes for check-plt.sh (and a few adjacent whitespace). src/cairo-region.c | 244 +++++++++++++++++++++++++++++------------------------ 1 file changed, 133 insertions(+), 111 deletions(-) commit f31c6548f818e1d4e257d94d623705284bcc4274 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Mar 28 19:26:03 2009 +0000 Silence compiler warnings for CAIRO_STATUS_LAST_STATUS Add ASSERT_NOT_REACHED (or similar) cases to the error handling switches to silence the compiler. src/cairo-misc.c | 5 +++-- src/cairo-spans.c | 2 ++ src/cairo-surface.c | 1 + 3 files changed, 6 insertions(+), 2 deletions(-) commit 979337dd4da40abb2ea49968a2c01709d9046aab Author: Eric Anholt <eric@anholt.net> Date: Sat Mar 28 21:53:04 2009 -0700 [gl] Wire blend factors using dst alpha to constants for CAIRO_CONTENT_COLOR. This fixes the operator and operator-alpha tests to rgb24 destinations. While we request an RGB texture, the returned texture has alpha bits, so when we blend against it as a renderbuffer, we get the junk alpha values. Whether or not this is a driver bug, we'll have this problem when we get visuals with alpha bits for windows despite not requestiong alpha, so we have to handle it in cairo. src/cairo-gl-surface.c | 22 ++++++++++++++++++---- 1 file changed, 18 insertions(+), 4 deletions(-) commit 809f7dc4c83009f0a67bc8af0608b480bf7fb00e Author: Eric Anholt <eric@anholt.net> Date: Sat Mar 28 21:30:13 2009 -0700 [gl] Make sure test visuals for CAIRO_CONTENT_COLOR_ALPHA have alpha bits. Since we're just making FBOs, the visual chosen probably shouldn't matter, and it doesn't on Intel. But it does seem like the right thing to do. boilerplate/cairo-boilerplate-gl.c | 25 ++++++++++++++++++------- 1 file changed, 18 insertions(+), 7 deletions(-) commit 93c437d4b9261bc1ce3daf9cbd02d9cc4d2584cc Author: Eric Anholt <eric@anholt.net> Date: Sat Mar 28 20:59:01 2009 -0700 [gl] Use GLEW to detect required extension presence. configure.ac | 9 +++++++++ src/cairo-gl-surface.c | 13 +++++++++++++ 2 files changed, 22 insertions(+) commit 62d84847c3b857b8fd5aeea3f4b5c0f2f4c91625 Author: Søren Sandmann Pedersen <sandmann@redhat.com> Date: Sat Mar 28 03:58:51 2009 -0400 [region] Add documentation for all the new region methods. src/cairo-region.c | 260 ++++++++++++++++++++++++++++++++++++++++++++++------- src/cairo.h | 6 +- 2 files changed, 232 insertions(+), 34 deletions(-) commit 980e9f0dbd4f21c0ccf4cc33b0b0f2f370667290 Author: Søren Sandmann Pedersen <sandmann@redhat.com> Date: Sat Mar 28 03:04:28 2009 -0400 [region] Fix status propagation for regions When an operation fails, store the status code in the destination region rather than leaving it unchanged. src/cairo-region.c | 66 ++++++++++++++++++++++++++++++++++++++++++------------ 1 file changed, 52 insertions(+), 14 deletions(-) commit 64e490a464472cfe06d779503601d41972ed4518 Author: Søren Sandmann Pedersen <sandmann@redhat.com> Date: Sat Mar 28 01:10:24 2009 -0400 [region] Use signed ints for width and height in cairo_rectangle_int_t src/cairo.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 26999e5aa8b5d353596e31d4b3823cafe007125c Author: Søren Sandmann <sandmann@redhat.com> Date: Fri Mar 27 08:48:42 2009 -0400 [region] Add cairo_region_intersect_rectangle() and _subtract_rectangle() Intersecting with and subtracting rectangles is quite common, and we already have cairo_union_rectangle(). src/cairo-clip.c | 27 ++++----------------------- src/cairo-region.c | 45 ++++++++++++++++++++++++++++++++++++++++++++- src/cairo-surface.c | 9 +-------- src/cairo.h | 11 ++++++++++- src/cairoint.h | 2 ++ 5 files changed, 61 insertions(+), 33 deletions(-) commit 4b3245481cfbf96388c140421a071fd1f79ad601 Author: Søren Sandmann Pedersen <sandmann@redhat.com> Date: Sat Feb 21 08:13:25 2009 -0500 [region] Expand rect to rectangle in a couple of names Specifically, cairo_region_union_rect -> cairo_region_union_rectangle cairo_region_create_rect -> cairo_region_create_rectangle Also delete cairo_region_clear() which is not that useful. src/cairo-analysis-surface.c | 4 ++-- src/cairo-clip.c | 4 ++-- src/cairo-region.c | 16 +++------------- src/cairo-surface-fallback.c | 2 +- src/cairo-surface.c | 4 ++-- src/cairo-traps.c | 2 +- src/cairo-xlib-surface.c | 2 +- src/cairo.h | 7 ++----- src/cairoint.h | 5 ++--- 9 files changed, 16 insertions(+), 30 deletions(-) commit bf6d9bc1758ac1971485b6565d29934d6b06bef2 Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk> Date: Wed Feb 18 08:01:19 2009 -0500 [region] Delete cairo_region_create_rectangles() It was only used in _cairo_traps_extract_region() which could be simplified significantly by calling cairo_region_union_rect() repeatedly instead. src/cairo-region.c | 44 --------------------------------------- src/cairo-traps.c | 60 +++++++++++++++++++++--------------------------------- src/cairo.h | 4 ---- src/cairoint.h | 1 - 4 files changed, 23 insertions(+), 86 deletions(-) commit e6d82afcab01cc91b88185abc8d1bfb9df5c2c19 Author: Søren Sandmann <sandmann@redhat.com> Date: Wed Feb 18 00:20:46 2009 -0500 [region] Add slim_hidden_proto for region functions src/cairoint.h | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+) commit e29103c08189cab9de34f577951219a69bd7c8e2 Author: Søren Sandmann <sandmann@redhat.com> Date: Wed Feb 18 00:11:35 2009 -0500 [region] Move region function prototypes to cairo.h along with helper types. Move struct _cairo_region to cairoint.h and delete cairo-region-private.h. Delete cairo_private from the function definitions that had it. src/Makefile.sources | 1 - src/cairo-analysis-surface.c | 1 - src/cairo-clip-private.h | 1 - src/cairo-region-private.h | 128 ------------------------------------------- src/cairo-region.c | 6 +- src/cairo-types-private.h | 6 -- src/cairo.h | 75 +++++++++++++++++++++++++ src/cairoint.h | 6 +- 8 files changed, 83 insertions(+), 141 deletions(-) commit 1cca5a1348d071a3c8e2ba9d3ba0d9670b9564f4 Author: Søren Sandmann <sandmann@redhat.com> Date: Tue Feb 17 23:52:18 2009 -0500 [region] Remove underscores from _cairo_region_* src/cairo-analysis-surface.c | 18 ++++++------- src/cairo-clip.c | 50 +++++++++++++++++------------------ src/cairo-directfb-surface.c | 4 +-- src/cairo-paginated-surface.c | 4 +-- src/cairo-region-private.h | 36 ++++++++++++------------- src/cairo-region.c | 61 +++++++++++++++++++++---------------------- src/cairo-surface-fallback.c | 20 +++++++------- src/cairo-surface.c | 14 +++++----- src/cairo-traps.c | 10 +++---- src/cairo-win32-surface.c | 8 +++--- src/cairo-xcb-surface.c | 4 +-- src/cairo-xlib-surface.c | 14 +++++----- 12 files changed, 120 insertions(+), 123 deletions(-) commit fcdca966941f4f7cc908bd46df1ec9a6b0f757c1 Author: Søren Sandmann <sandmann@redhat.com> Date: Tue Feb 17 22:52:50 2009 -0500 [region] Always define cairo_rectangle_int_t with ints Use ints in cairo_rectangle_int_t, no matter the fixed type. src/cairo-types-private.h | 38 +++++++++----------------------------- 1 file changed, 9 insertions(+), 29 deletions(-) commit cfa7a384b730289233bef556fe1919f781eac684 Author: Søren Sandmann <sandmann@redhat.com> Date: Tue Feb 17 23:40:02 2009 -0500 [region] Add _cairo_region_contains_point() src/cairo-region-private.h | 5 ++++- src/cairo-region.c | 15 +++++++++++++-- src/cairo.h | 1 + 3 files changed, 18 insertions(+), 3 deletions(-) commit 15564d1949778124e553d792f85cae2b031aad45 Author: Søren Sandmann Pedersen <sandmann@redhat.com> Date: Tue Feb 17 05:58:34 2009 -0500 [region] Add _cairo_region_union src/cairo-region-private.h | 4 ++++ src/cairo-region.c | 16 ++++++++++++++++ 2 files changed, 20 insertions(+) commit df883aa937d2f3ecf52048b60caff48b1c9edac9 Author: Søren Sandmann Pedersen <sandmann@redhat.com> Date: Tue Feb 17 06:06:40 2009 -0500 [region] Add a cairo_region_overlap_t type src/cairo-analysis-surface.c | 4 ++-- src/cairo-region-private.h | 8 +++++++- src/cairo-region.c | 33 ++++++++++++++++++++++++--------- 3 files changed, 33 insertions(+), 12 deletions(-) commit ebd0e685ae2d76c3af96d748a33a7194c70749f3 Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk> Date: Mon Feb 16 05:41:31 2009 -0500 [region] Consistently use rectangles in the API of regions Usually, rectangles are more useful than boxes, so regions should only expose rectangles in their public API. Specifically, _cairo_region_num_boxes becomes _cairo_region_num_rectangles _cairo_region_get_box becomes _cairo_region_get_rectangle Remove the cairo_box_int_t type src/cairo-clip.c | 22 ++++++++------------ src/cairo-directfb-surface.c | 26 +++++++++++------------ src/cairo-paginated-surface.c | 34 +++++++++++++++--------------- src/cairo-region-private.h | 14 ++++++------- src/cairo-region.c | 28 ++++++++++++------------- src/cairo-surface-fallback.c | 2 +- src/cairo-surface.c | 25 ++++++++-------------- src/cairo-traps.c | 31 ++++++++++++++-------------- src/cairo-types-private.h | 6 ------ src/cairo-win32-surface.c | 48 +++++++++++++++++++++---------------------- src/cairo-xcb-surface.c | 24 +++++++++++----------- src/cairo-xlib-surface.c | 29 +++++++++++++------------- 12 files changed, 134 insertions(+), 155 deletions(-) commit c29aeee2d3569cd43e88adc05662bb6306987b8b Author: Søren Sandmann <sandmann@redhat.com> Date: Thu Dec 11 18:37:40 2008 -0500 [region] Change sense of _cairo_region_not_empty() to _cairo_region_empty() Having "not" in the name causes double negatives. src/cairo-clip.c | 2 +- src/cairo-region-private.h | 2 +- src/cairo-region.c | 6 +++--- src/cairo-surface-fallback.c | 2 +- 4 files changed, 6 insertions(+), 6 deletions(-) commit c88321d0a2b9d200b2044674d02fa0683feca74c Author: Søren Sandmann <sandmann@redhat.com> Date: Thu Dec 11 16:20:23 2008 -0500 [region] Miscellaneous bug fixes. - Initialize region->status in _cairo_region_create_boxes() - Make _cairo_region_copy() actually return a region. - Fix a bug where a NULL region could be dereferenced Also add an assertion that the result of cairo_region_copy() is never NULL. src/cairo-clip.c | 2 ++ src/cairo-image-surface.c | 2 +- src/cairo-region.c | 4 +++- 3 files changed, 6 insertions(+), 2 deletions(-) commit 93b285dc2c9c9ec486e0096933c3611f73374fcb Author: Søren Sandmann <sandmann@redhat.com> Date: Thu Dec 11 15:50:15 2008 -0500 [region] Get rid of clip->has_region src/cairo-clip-private.h | 1 - src/cairo-clip.c | 36 +++++++++++++++--------------------- src/cairo-surface.c | 2 +- 3 files changed, 16 insertions(+), 23 deletions(-) commit e3e1b35eb9eb3cf90d882a7452580f9efe00b89a Author: Søren Sandmann <sandmann@redhat.com> Date: Thu Dec 11 15:29:23 2008 -0500 [region] Make cairo_region_t a malloced object. src/cairo-analysis-surface.c | 26 ++++--- src/cairo-clip-private.h | 2 +- src/cairo-clip.c | 99 +++++++++++++------------ src/cairo-region-private.h | 42 +++++------ src/cairo-region.c | 168 ++++++++++++++++++++++++++++++++++--------- src/cairo-surface-fallback.c | 65 ++++++++--------- src/cairo-surface.c | 29 ++++---- src/cairo-traps.c | 14 ++-- src/cairo-xlib-surface.c | 16 ++--- src/cairoint.h | 2 +- 10 files changed, 280 insertions(+), 183 deletions(-) commit c2c637cf1d89fc8fdcced4467d602a55ef0b14e0 Author: Søren Sandmann Pedersen <sandmann@redhat.com> Date: Sat Mar 28 04:26:10 2009 -0400 Simplify _cairo_clip_intersect_region() The intermediate intersection region is not necessary because if the operation fails, clip->region will end up as a pixman error region. src/cairo-clip.c | 19 +++++-------------- 1 file changed, 5 insertions(+), 14 deletions(-) commit fe10cd6467f2ec58532dc7c713bb6d08dab08442 Author: Bertram Felgenhauer <int-e@gmx.de> Date: Fri Mar 27 07:10:21 2009 +0100 [spline] fix wrong sign in _cairo_spline_bound. src/cairo-spline.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 53107de63a954105f8820e5b96e2a1956cb7af8c Author: Bertram Felgenhauer <int-e@gmx.de> Date: Thu Mar 26 04:56:27 2009 +0100 [in_fill] Correctly track current point in curve_to. When discarding a bezier path segment, we still need to update the current point. src/cairo-path-in-fill.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) commit 578b06a978d51dc3d1d844ee7eea9ddd8329cc8d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Mar 23 10:56:43 2009 +0000 [perf] Use CAIRO_LDFLAGS Use CAIRO_LDFLAGS in order to pull in additional link options, such as --coverage. perf/Makefile.am | 2 ++ 1 file changed, 2 insertions(+) commit ac30ced6135c5cf0fb34a67fe8f863030c1fbeb9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Mar 22 19:02:27 2009 +0000 [test] Check for surface create failure. If the image surface creation fails, the data pointer will be NULL leading to a segfault -- so check! test/user-font-mask.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) commit b5a4a2c4b1b1bab25e2ff0842e3c27205ec73d51 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Mar 22 19:00:04 2009 +0000 [test] Early test for memfault. Check the error status within after each iteration in solid-pattern-cache-stress. test/solid-pattern-cache-stress.c | 41 ++++++++++++++++++++++++++++----------- 1 file changed, 30 insertions(+), 11 deletions(-) commit 1a7b94f934f8c9a25e60d9466651b0b7fb919656 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Mar 20 11:06:29 2009 +0000 [test] Check for errors during get-path-extents. Need to check that an error was not raised on the context before checking whether the extents match expectations. test/get-path-extents.c | 9 +++++++++ 1 file changed, 9 insertions(+) commit 3b545330887209c68128fca0d1bb40bf4927c2a6 Author: Vladimir Vukicevic <vladimir@pobox.com> Date: Sat Mar 21 11:56:18 2009 -0700 [win32] Use _hypot instead of hypot The hypot symbol is deprecated on Win32, and nonexistent on Windows CE src/cairo-compiler-private.h | 1 + 1 file changed, 1 insertion(+) commit 895fce1632069ddfb8c581d2b23b1b3552e35104 Author: Stefan Klug <Klug.Stefan@gmx.de> Date: Fri Mar 20 17:20:22 2009 -0400 [wince] Use the official _WIN32_WCE define instead of WINCE src/cairo-misc.c | 4 ++-- src/cairoint.h | 2 +- 2 files changed, 3 insertions(+), 3 deletions(-) commit 86a935de42318c4be92a657ce8e8c452693d4414 Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Fri Mar 20 17:14:08 2009 -0400 [msvc] Fix definition of CAIRO_ENSURE_UNIQUE for non-x86 The current definition of CAIRO_ENSURE_UNIQUE uses x86 assembly so make sure we only compile it when targeting x86. src/cairo-compiler-private.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit c932a809d6484503d7ee267d934bbc87c8d44092 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Mar 17 12:01:13 2009 +0000 [test] Propagate allocation failure. Check for memfaults during rel-path test. test/rel-path.c | 19 ++++++++++++++++--- 1 file changed, 16 insertions(+), 3 deletions(-) commit 7db55b37d4aef188e04771b45076f6735507c209 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Mar 17 11:58:16 2009 +0000 [test] Destroy pattern on error paths. Destroy the pattern after encountering an error in pattern-getters test. test/pattern-getters.c | 29 +++++++++++++++++++++++++---- 1 file changed, 25 insertions(+), 4 deletions(-) commit dc176d88ac03ae71fc32abb27329a35650801d99 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Mar 17 10:42:37 2009 +0000 [test] Trivial leak on error in show-glyphs-many. Free the allocated glyph array after failure. test/show-glyphs-many.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) commit a4b44ca89eb9975e6af2913e50ec3c3eb566cfdd Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Mar 17 14:56:42 2009 +0000 [pattern] Report the true error status from getters. As the getters actually return an error status, use it to report any pre-existing error status on the pattern. src/cairo-pattern.c | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+) commit 496bbcf58233ceecfffc52ae5b1a3a1f07516c67 Author: Behdad Esfahbod <behdad@behdad.org> Date: Tue Mar 17 19:46:25 2009 -0400 Make CAIRO_STATUS_LAST_STATUS public doc/public/tmpl/cairo-status.sgml | 1 + src/cairo.h | 12 ++++++++++-- src/cairoint.h | 6 ------ 3 files changed, 11 insertions(+), 8 deletions(-) commit 922c108365b940dbabf23358189bbaa8602446c9 Author: Behdad Esfahbod <behdad@behdad.org> Date: Tue Mar 17 19:22:31 2009 -0400 [twin] Reorganize matching code to better reflect the code in Pango Makes it easier to update later. src/cairo-font-face-twin.c | 179 +++++++++++++++++++++++++++++++++------------ src/cairoint.h | 4 + 2 files changed, 138 insertions(+), 45 deletions(-) commit 2b4044a36f8b156ca0e58b72614659324a9b022e Author: Behdad Esfahbod <behdad@behdad.org> Date: Tue Mar 17 18:52:16 2009 -0400 [twin] Update parsed weights and stretches from Pango src/cairo-font-face-twin.c | 64 ++++++++++++++++++++++++++++------------------ 1 file changed, 39 insertions(+), 25 deletions(-) commit 3ec94f9b5917b8f71304c982bd153e5a810f3b12 Author: Behdad Esfahbod <behdad@behdad.org> Date: Tue Mar 17 00:53:30 2009 -0400 Support compiling without fontconfig Adds a new, fake, fontconfig font backend. Fontconfig can be disabled using --disable-fc, in which case the toy text API wont find fonts and the internal font will always be used. Also defines the feature macro CAIRO_HAS_FC_FONT. The two fontconfig-specific functions in cairo-ft.h depend on that macro now. boilerplate/Makefile.win32.features | 10 +++++++++ build/Makefile.win32.features | 1 + build/Makefile.win32.features-h | 3 +++ build/configure.ac.features | 1 + configure.ac | 26 ++++++++++++++-------- doc/public/cairo-sections.txt | 1 + doc/public/tmpl/cairo-ft.sgml | 9 ++++++++ doc/public/tmpl/cairo-status.sgml | 2 +- src/Makefile.win32.features | 14 ++++++++++++ src/cairo-ft-font.c | 43 +++++++++++++++++++++++++++++++------ src/cairo-ft.h | 23 +++++++++++++------- src/cairo-os2-surface.c | 6 ++++-- src/cairo-path-fixed.c | 2 +- src/cairo-scaled-font.c | 4 ++-- src/cairo-xlib-display.c | 2 -- src/cairo-xlib-screen.c | 23 +++++++++----------- 16 files changed, 125 insertions(+), 45 deletions(-) commit 0e7bcb5e3793771d50f7d391d7b1e538ef220a80 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Mar 16 21:32:44 2009 +0000 [scaled-font] Improve comments. Grr. I thought I had merged this in with the previous commit... src/cairo-scaled-font.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) commit 9c80392ac415e7f07c71261d280ac4376d3c8471 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Mar 16 19:31:38 2009 +0000 [scaled-font] Lean and mean global glyph cache. Jeff Muizelaar pointed out that the severe overallocation implicit in the current version of the glyph cache is obnoxious and prevents him from accepting the trunk into Mozilla. Jeff captured a trace of scaled font and glyph usage during a tp run and presented his analysis in http://lists.cairographics.org/archives/cairo/2009-March/016706.html Using that data, the design was changed to allocate pages of glyphs from a capped global pool but with per-font hash tables. This should allow the glyph cache to have tight memory bounds with fair allocation according to usage. Note that both the old design and the 1.8 glyph cache had essentially unbounded memory constraints, since each scaled font could cache up to 256 glyphs (1.8) or had a reserved page (old), with no limit on the number of active fonts. Currently the eviction policy is a simple random strategy, this gives a 'fair' allotment of the cache, but a LRU variant might perform better. On a sample run of firefox-3.0.7 perusing BBC news in 32 languages: 1.8: cache allocation 8190x, ~1.2 MiB; elapsed 88.2s old: cache allocation 771x, ~13.8 MiB; elapsed 81.7s lean: cache allocation 433x, ~1.8 MiB; elapsed 82.4s src/cairo-cache-private.h | 5 +- src/cairo-cache.c | 39 ++-- src/cairo-ft-font.c | 4 +- src/cairo-hash-private.h | 6 +- src/cairo-hash.c | 55 ------ src/cairo-scaled-font-private.h | 6 +- src/cairo-scaled-font.c | 396 ++++++++++++++++++++-------------------- src/cairo-types-private.h | 3 + src/cairoint.h | 7 +- 9 files changed, 232 insertions(+), 289 deletions(-) commit 121d4bb656755b7ca89065bf87e3e4e47c49c89d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Mar 16 19:42:47 2009 +0000 [scaled-font] Initialise original_font_face Eek, attempting to use+free an uninitialised pointer! src/cairo-scaled-font.c | 1 + 1 file changed, 1 insertion(+) commit 2df498ba452b188f2f40413c91fec5e5ed9cf967 Author: Eric Anholt <eric@anholt.net> Date: Thu Mar 5 23:26:44 2009 -0800 [gl] Use triangle fan instead of GL_QUADS for the single quad we draw. This saves the driver from converting the quad into tris on its own. We'd rather be able to specify a rect as 2-3 points and use the rectangular rendering feature that most of our hardware has, if it was possible, but it isn't exposed in GL. src/cairo-gl-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 9099c7e7307a39bc630919faa65bba089fd15104 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Mar 4 09:26:17 2009 +0000 [trace] Disable mark dirty tracing. Applications like firefox have a very conservative approach and mark surfaces dirty before every render. As we record the image data every time, firefox traces can grow very large with redundant data - so allow the user to disable mark dirty tracing. util/cairo-trace/cairo-trace.in | 19 +++++++++++++++---- util/cairo-trace/trace.c | 27 ++++++++++++++++++++------- 2 files changed, 35 insertions(+), 11 deletions(-) commit addeb32c751ac080fe634ea6f83076d018944e4a Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Tue Mar 3 12:20:47 2009 -0500 [pdf] Intialize 'interpolate' Intialize 'interpolate' to prevent a gcc warning. Do this instead of adding a 'default' case to the switch statement so that we still get warnings if new filter types are added. src/cairo-pdf-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit a8158d443e6a4bafae28a46b883cbdcfd5789eec Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Mar 3 11:21:35 2009 +0000 [truetype] Move the sizeof asserts to compile time. Eliminate the need for a runtime test on the sizeof the private structures by performing the check at compile time. This was provoked by Ginn Chenn noting that the test was including a private header. src/cairo-truetype-subset.c | 15 +++++++++++ test/Makefile.sources | 1 - test/truetype-tables.c | 63 --------------------------------------------- 3 files changed, 15 insertions(+), 64 deletions(-) commit e4b1f871e9cbb04590df5dc6f6f7854642fa0340 Author: Ginn Chen <Ginn.Chen@Sun.COM> Date: Tue Mar 3 10:27:11 2009 +0000 [configure] Detect mkdir variant with non-gcc compilers. It doesn't work for non GCC compiler right now, as "-Werror -Wall" is an error to non GCC compiler. I swapped the sequence of build/configure.ac.system and build/ configure.ac.warnings, then WARN_CFLAGS can be used. build/configure.ac.system | 2 +- configure.ac | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) commit f178f55e47013a92a47a24ac04b1041963b03976 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Mar 3 10:25:05 2009 +0000 [test/any2ppm] Update the cairo-script-interpreter callback. Reflect the current prototype which now specifies the content to use when creating the surface. test/any2ppm.c | 1 + 1 file changed, 1 insertion(+) commit 694f2eea9feecfdc437e6964d1e758fab7315af9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Mar 3 10:24:25 2009 +0000 [test/any2ppm] Add feature checks for daemon() Add the feature checks for the presence of the daemon() function call. test/any2ppm.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) commit 9304984f4e20beec7b4de6a4141e2fd489130006 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Mar 1 10:10:24 2009 +0000 [path] Fix regression introduced with 005436 The order of the multiplication of the CTM and device_transform was reversed. src/cairo-path.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit b30de64a8ca3de7632696f45bdb580217bd9f8a1 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Mar 1 10:05:16 2009 +0000 [test] Add regression test for 005436 Jeff Muizelaar found a regression in commit 005436 and submitted this little test to exercise it. The essence of the bug appears to be wrt to the product of the CTM and device transform matrices. test/Makefile.sources | 1 + test/path-append.c | 81 +++++++++++++++++++++++++++++++++ test/path-append.ps.ref.png | Bin 0 -> 4516 bytes test/path-append.ref.png | Bin 0 -> 6165 bytes test/path-append.test-fallback.ref.png | Bin 0 -> 6461 bytes test/path-append.xlib-fallback.ref.png | Bin 0 -> 6357 bytes test/path-append.xlib.ref.png | Bin 0 -> 6461 bytes 7 files changed, 82 insertions(+) commit e20f1a0c76f2185760ca3a7466e25de6beedeaac Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Feb 27 16:30:05 2009 +0000 [in-fill] Close the path, cf fill() In order for in-fill treat the path equivalently to a fill, we need to close the path after interpretation. src/cairo-path-in-fill.c | 2 ++ 1 file changed, 2 insertions(+) commit 102ddd37bb329d41241f434ede0f9cba8424e71e Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Thu Feb 26 16:04:49 2009 -0500 [win32] Include $(OPT) flags when linking cairo.dll This will pass in -Zi which causes the /DEBUG flag to be passed to the linker keeping the debug information from being discarded during link. src/Makefile.win32 | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit eb2a73ba6d290497cabb0f917c0375eea919178e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Feb 24 20:27:36 2009 +0000 [ps] Check the status of _cairo_type3_glyph_surface_create() Doing so serves two purposes. The first is to do an early error check and the second is to clearly initialise the status variable. src/cairo-ps-surface.c | 3 +++ 1 file changed, 3 insertions(+) commit 6eb0a9d97ff7eaaee69ca10e4081cb950a543ce3 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Feb 15 21:27:29 2009 +0000 [scaled-font] Hold reference to original font face As noted by Carl during his LCA talk, caching of toy fonts was broken because we create the scaled font using the implementation font face and lose the reference to the containing font face that is cached by the toy font face create routines. So the toy fonts were not being preserved for the duration of the holdover scaled fonts and we recreated a new font face, new scaled font and new glyph caches every time we needed a font. src/cairo-ft-font.c | 2 +- src/cairo-scaled-font-private.h | 2 ++ src/cairo-scaled-font.c | 17 +++++++++++++---- src/cairo-toy-font-face.c | 1 + 4 files changed, 17 insertions(+), 5 deletions(-) commit fe7d5323f5bc734e76179b74d68fcba9b924ba94 Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Thu Feb 19 16:00:17 2009 -0500 Relanding: Add CAIRO_ENSURE_UNIQUE macro and use it in _cairo_error() When using MSVC, _cairo_error() can be folded into other identical functions. If that happens, _cairo_error isn't really useful anymore. Using the CAIRO_ENSURE_UNIQUE macro makes sure this doesn't happen. Use __asm to serve as a line delimiter. This allows us to use the __asm{} block in a macro. src/cairo-compiler-private.h | 24 ++++++++++++++++++++++++ src/cairo.c | 1 + 2 files changed, 25 insertions(+) commit 78de0e045ec0c10708052ed48df228e786839ddb Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Thu Feb 19 15:45:11 2009 -0500 Revert "Add CAIRO_ENSURE_UNIQUE macro and use it in _cairo_error()" This reverts commit 126824f5e6279f56dab0c040394b6c326b848cf7. It turns out MSVC doesn't handle line continuation characters in __asm{} blocks very well, so revert for now until I come up with something that works. src/cairo-compiler-private.h | 24 ------------------------ src/cairo.c | 1 - 2 files changed, 25 deletions(-) commit 126824f5e6279f56dab0c040394b6c326b848cf7 Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Thu Feb 19 13:31:55 2009 -0500 Add CAIRO_ENSURE_UNIQUE macro and use it in _cairo_error() When using MSVC, _cairo_error() can be folded into other identical functions. If that happens, _cairo_error isn't really useful anymore. Using the CAIRO_ENSURE_UNIQUE macro makes sure this doesn't happen. src/cairo-compiler-private.h | 24 ++++++++++++++++++++++++ src/cairo.c | 1 + 2 files changed, 25 insertions(+) commit d4227fc9126ffbb3a967aea1bc9795e7e64ee8e1 Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Thu Feb 19 11:45:14 2009 -0500 [test] Fix assert on default font family in toy-font-face toy-font-face was checking that cairo_toy_font_face_get_family returned "" which is CAIRO_FONT_FAMILY_DEFAULT when the freetype font backend is the default. However, when other font backends are the default the returned family is different. Therefore, instead of checking for "", we check for the appropriate string depending on the backend. test/toy-font-face.c | 19 +++++++++++++++---- 1 file changed, 15 insertions(+), 4 deletions(-) commit 06fa7a8386b267fdbd7635b952adc75a58f62584 Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Wed Feb 18 18:43:29 2009 -0500 Fix a const warning in _cairo_user_data_array_foreach() _cairo_user_data_array_foreach() was taking a function with a void *key parameter instead of a const void *key to match cairo_user_data_slot_t. src/cairo-array.c | 2 +- src/cairo-surface.c | 2 +- src/cairoint.h | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) commit f7b3c9df5885a1cd7981ee12b91962bdbfff47af Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Mon Feb 16 22:28:32 2009 -0500 Add test/Makefile.sources This lets test/Makefile.am and test/Makefile.win32 share a common list of sources. It also makes test/Makefile.win32 useful again. test/Makefile.am | 251 ++++---------------------------------------------- test/Makefile.sources | 251 ++++++++++++++++++++++++++++++++++++++++++++++++++ test/Makefile.win32 | 125 +++++-------------------- 3 files changed, 293 insertions(+), 334 deletions(-) commit e380beae5382df547a1b538de94e90c0e2339141 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Mon Feb 16 14:01:43 2009 +0200 [sdl] Remove new backend. The SDL backend makes invalid assumptions about SDL_Surface locking semantics and doesn't deal correctly with the unpremultiplied pixel format supported by SDL. Removed as per discussion on the mailing list. http://lists.cairographics.org/archives/cairo/2009-February/016595.html boilerplate/Makefile.sources | 3 - boilerplate/Makefile.win32.features | 10 - boilerplate/cairo-boilerplate-sdl-private.h | 56 ---- boilerplate/cairo-boilerplate-sdl.c | 69 ----- boilerplate/cairo-boilerplate.c | 15 - build/Makefile.win32.features | 1 - build/Makefile.win32.features-h | 3 - build/configure.ac.features | 1 - configure.ac | 8 - doc/public/tmpl/cairo-surface.sgml | 1 - perf/Makefile.am | 4 - perf/cairo-perf.c | 4 - src/Makefile.sources | 3 - src/Makefile.win32.features | 14 - src/cairo-sdl-surface.c | 418 ---------------------------- src/cairo-sdl.h | 54 ---- src/cairo.h | 2 - test/Makefile.am | 4 - test/cairo-test-runner.c | 4 - test/get-clip.c | 1 - 20 files changed, 675 deletions(-) commit 79a72e63585d7fce7898f2c8bd997c6d88f8895f Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk> Date: Mon Feb 16 05:44:15 2009 -0500 Delete _cairo_region_boxes_fini() src/cairo-region.c | 6 ------ 1 file changed, 6 deletions(-) commit f6daa664c1b2c894ba3baf2e7a72566bda1fd636 Author: Søren Sandmann <sandmann@daimi.au.dk> Date: Sun Feb 15 18:55:17 2009 -0500 Delete _cairo_region_get_boxes() in favor of _cairo_region_get_box() The _cairo_region_get_boxes() interface was difficult to use and often caused unnecessary memory allocation. With _cairo_region_get_box() it is possible to access the boxes of a region without allocating a big temporary array. src/cairo-clip.c | 21 ++++++++----------- src/cairo-directfb-surface.c | 21 ++++++++----------- src/cairo-paginated-surface.c | 19 ++++++++--------- src/cairo-region-private.h | 10 +++------ src/cairo-region.c | 39 +++++++++-------------------------- src/cairo-sdl-surface.c | 21 ++++++++----------- src/cairo-surface.c | 44 +++++++++++++++------------------------ src/cairo-win32-surface.c | 48 +++++++++++++++++++++---------------------- src/cairo-xcb-surface.c | 26 ++++++++++------------- src/cairo-xlib-surface.c | 31 ++++++++++++---------------- 10 files changed, 112 insertions(+), 168 deletions(-) commit 54228cb5fe0c41b10f0ec4a206b5fac28823cde6 Author: Eric Anholt <eric@anholt.net> Date: Sun Feb 15 14:29:55 2009 -0800 [gl] Use textures and not fbo-based surfaces for image surface patterns. This cuts the BO allocation overhead for traps in more than half, and reduces the load on the accelerator. src/cairo-gl-surface.c | 191 +++++++++++++++++++++++++++++++++++++++++++++---- 1 file changed, 176 insertions(+), 15 deletions(-) commit fcb13d262c5a09e56dc25c76103ef4ba623929ff Author: Søren Sandmann Pedersen <sandmann@daimi.au.dk> Date: Sun Feb 15 15:59:10 2009 -0500 Simplify region handling in xlib surface src/cairo-xlib-surface.c | 11 +++-------- 1 file changed, 3 insertions(+), 8 deletions(-) commit b43e7aee98a8d69677f7e6d2584fe01f550f896b Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Fri Feb 13 12:15:23 2009 -0500 [wince] We don't need cairo_win32_tmpfile on Windows CE _cairo_win32_tmpfile() uses _open_osfhandle() which is not available on Windows CE. However, Windows CE doesn't have the permisions problems that necessitated _cairo_win32_tmpfile() in the first place so we can just use tmpfile() on Windows CE. src/cairo-misc.c | 5 ++++- src/cairoint.h | 2 +- 2 files changed, 5 insertions(+), 2 deletions(-) commit fa66291c8862ed592fca469ceab0ac9b1d270835 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jan 28 14:22:58 2009 +0000 [perf] Test non-antialiased fill. Just because the i915 can special case such fills using a single-pass tessellation in the stencil buffer. perf/fill.c | 21 +++++++++++++++++++++ 1 file changed, 21 insertions(+) commit aab9ed3432f0ea7b8b24990de4aa134bb599a3e5 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 30 10:22:47 2009 +0000 [perf] Env variable to ignore similar targets whilst benchmarking. Use CAIRO_TEST_IGNORE_SIMILAR to skip similar targets. perf/cairo-perf.c | 3 +++ 1 file changed, 3 insertions(+) commit 655f6987334b991763a5ab7746dbfd73c6b05f3e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Feb 6 20:11:17 2009 +0000 [perf] Split can_run? into a separate precondition. Allow tests to skip unnecessary setup when pruning the list of perf cases. perf/box-outline.c | 3 +++ perf/cairo-perf.c | 24 ++++++++++++++++-------- perf/cairo-perf.h | 4 ++++ perf/composite-checker.c | 3 +++ perf/dragon.c | 3 +++ perf/fill.c | 3 +++ perf/intersections.c | 3 +++ perf/long-dashed-lines.c | 3 +++ perf/long-lines.c | 3 +++ perf/mosaic.c | 3 +++ perf/paint-with-alpha.c | 3 +++ perf/paint.c | 3 +++ perf/pattern_create_radial.c | 3 +++ perf/pythagoras-tree.c | 3 +++ perf/rectangles.c | 3 +++ perf/rounded-rectangles.c | 3 +++ perf/spiral.c | 3 +++ perf/stroke.c | 3 +++ perf/subimage_copy.c | 3 +++ perf/tessellate.c | 3 +++ perf/text.c | 3 +++ perf/twin.c | 3 +++ perf/unaligned-clip.c | 3 +++ perf/world-map.c | 3 +++ perf/zrusin.c | 3 +++ 25 files changed, 89 insertions(+), 8 deletions(-) commit 798581a1b5a8a56ce9d16c5b21eab82851061732 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 30 22:12:49 2009 +0000 [perf] Extend range of testing. Primarily to test longer glyph runs, but also test large upper bounds for strokes and fills. perf/cairo-perf.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) commit 1522fac5c71708fc9e98e03da9f51926c1e3769c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Feb 3 17:18:53 2009 +0000 [perf] Fix infinite loop in text The row would wrap-around with the character index, causing an infinite loop when trying to fill a window of size 512x512 and above. perf/text.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) commit 187e3473512e40fa1d046783e797ec3a198b09b2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Feb 13 13:09:32 2009 +0000 [test] Free ref_name in fallback-resolution. Trivial leak of a few thousand bytes. test/fallback-resolution.c | 2 ++ 1 file changed, 2 insertions(+) commit 005436758c5679f76cc462841678fb93d6c7e0ac Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Feb 3 10:20:26 2009 +0000 [path] Inline path ops during append_to_context() By inlining the operations, and most significantly, precomputing the combined user-to-backend matrix, we can achieve a speed up of over 50%, which is a noticeable performance boost in swfdec - where append-to-path accounts for over 35% [inclusive] of the time for a h/w accelerated backend. src/cairo-path.c | 76 ++++++++++++++++++++++++++++++++++++++++++-------------- 1 file changed, 58 insertions(+), 18 deletions(-) commit d295942b9d4da3be3318cd5fe2d3b0b1fe005d11 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jan 31 00:56:45 2009 +0000 Inline _cairo_restrict_value() This is one instance where the function call overhead dominated the function call in both time and size. src/cairo-misc.c | 9 --------- src/cairo-pattern.c | 32 ++++++++++++++++---------------- src/cairo.c | 14 ++++++++------ src/cairoint.h | 13 +++++++++++-- 4 files changed, 35 insertions(+), 33 deletions(-) commit cc8a09567ca034e7d95e2ef8a3ec833b12c9f87a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Feb 13 13:23:50 2009 +0000 [surface] Move the meta-data copy-on-snapshot to the generic layer. As pointed out by Paolo Bonzini, copying the meta data for a snapshot is common for all backends and so should be handled by the generic layer. src/cairo-surface-fallback.c | 11 ----------- src/cairo-surface.c | 30 +++++++++++++++++++++++++++--- 2 files changed, 27 insertions(+), 14 deletions(-) commit adaf70a93f4449e85997bcde531b76c9044758ea Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Feb 13 12:56:46 2009 +0000 [surface] Separate the mime-data from the user-data. Move the mime-data into its own array so that it cannot be confused with user-data and we do not need to hard-code the copy list during snapshotting. The copy-on-snapshotting code becomes far simpler and will accommodate all future mime-types. Keeping mime-data separate from user-data is important due to the principle of least surprise - the API is different and so it would be surprising if you queried for user-data and were returned an opaque mime-data pointer, and vice versa. (Note this should have been prevented by using interned strings, but conceptually it is cleaner to make the separation.) Also it aides in trimming the user data arrays which are linearly searched. Based on the original patch by Adrian Johnson: http://cgit.freedesktop.org/~ajohnson/cairo/commit/?h=metadata&id=37e607cc777523ad12a2d214708d79ecbca5b380 src/cairo-array.c | 36 ++++++++++++++++++++++++++++++++++++ src/cairo-surface-fallback.c | 27 +++++++-------------------- src/cairo-surface-private.h | 1 + src/cairo-surface.c | 42 +++++++++++++++++++++--------------------- src/cairoint.h | 14 ++++++++++++-- 5 files changed, 77 insertions(+), 43 deletions(-) commit 2280de9d0282a599309ec12dc576bae54857f873 Merge: 17ce8584e de7270cb1 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Feb 13 10:08:21 2009 +0000 Merge branch '1.8' commit de7270cb1e9510cb230e80045c812fa68fcfa585 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Feb 13 10:03:22 2009 +0000 [cairo] Describe the restrictions upon cairo_set_tolerance() Truc Troung reported that the behaviour of cairo_set_tolerance()/cairo_get_tolerance() was inconsistent with the documentation, i.e. we failed to mention that the tolerance would be restricted to the smalled fixed-point value. Add a sentence to the documentation that describes the restriction without mentioning what that is... Hopefully that is sufficient detail to accommodate the reporter, without exposing internal implementation details. Fixes https://bugs.freedesktop.org/show_bug.cgi?id=20095 Bug 20095 - The cairo_set_tolerance() function behavior is inconsistency with the spec src/cairo.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) commit 17ce8584e7142d13bd7a777c9570e5548a06a90c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Feb 12 18:26:57 2009 +0000 [NEWS] Add API changes. Scan the public headers for obvious additions. NEWS | 20 +++++++++++++++++++- 1 file changed, 19 insertions(+), 1 deletion(-) commit 2df611a3810eb64c8ed22dfae5f3d3157eef7e6a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Feb 10 10:28:28 2009 +0000 [script] Expose a normal xlib window for replay Provide a visible surface for replaying scripts against. util/cairo-script/csi-replay.c | 122 ++++++++++++++++++++++------------------- 1 file changed, 66 insertions(+), 56 deletions(-) commit 8e1f103540392b17c51b4fb4b3c3480430cbc212 Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Thu Feb 12 12:02:17 2009 -0500 Divert pclose to _pclose for MSVC Commit bf62798b1284533e28b78717dac8070ca6d29e54 added a diversion for popen but not for pclose. This is needed for linking the boilerplate with the test suite. src/cairo-compiler-private.h | 1 + 1 file changed, 1 insertion(+) commit 4c6e7a16032fb223f4224d4355c2ea046cd8fc4b Author: Eric Anholt <eric@anholt.net> Date: Wed Feb 11 05:09:23 2009 -0800 [gl] Add support for constant color directly, not through a texture. src/cairo-gl-surface.c | 337 ++++++++++++++++++++++++++++++++++++------------- 1 file changed, 248 insertions(+), 89 deletions(-) commit 31fb7382d22c9293c093b60716ee4e393fa0d5cf Author: Eric Anholt <eric@anholt.net> Date: Wed Feb 11 05:09:51 2009 -0800 [gl] Fix some memory leaks on boilerplate-based runs. boilerplate/cairo-boilerplate-gl.c | 4 ++++ 1 file changed, 4 insertions(+) commit df96fc1f92b68ef3e76750d0891377980e039b09 Author: Eric Anholt <eric@anholt.net> Date: Mon Feb 9 15:42:09 2009 -0800 [gl] Avoid re-binding the same draw buffer. The driver could be smarter and avoid a lot of the cost, but this is easy and avoids needing much driver smarts. src/cairo-gl-surface.c | 30 ++++++++++++++++++++---------- 1 file changed, 20 insertions(+), 10 deletions(-) commit fb6937565be558e7ff8ba3dd815a427721783cba Author: Eric Anholt <eric@anholt.net> Date: Mon Feb 9 15:30:21 2009 -0800 [gl] Add interface for creating surfaces from window backbuffers. src/cairo-gl-surface.c | 85 +++++++++++++++++++++++++++++++++++++++++++++++--- src/cairo-gl.h | 9 +++--- 2 files changed, 85 insertions(+), 9 deletions(-) commit 6cc3d8dcc0fe418c56a519180163bd49fbdb9730 Author: Eric Anholt <eric@anholt.net> Date: Mon Feb 9 12:30:20 2009 -0800 [gl] Fix unantialiased-shapes tests. src/cairo-gl-surface.c | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-) commit 23a8c58dc13bd9ff44f235bee8e4e7240831728a Author: Eric Anholt <eric@anholt.net> Date: Thu Feb 5 11:00:59 2009 -0800 [gl] Fix infinite recursion on compositing A1 images. We could do this fast, but let's do it correct first. src/cairo-gl-surface.c | 28 ++++++++++++++++++++++++++++ 1 file changed, 28 insertions(+) commit b6dde667fa1c72c6da1970244c4f28c48a38f493 Author: Eric Anholt <eric@anholt.net> Date: Mon Feb 2 23:22:49 2009 -0800 [gl] Add support for composite and composite_trapezoids. This is a simple implementation that uses GL_ARB_texture_env_combine after converting everything to GL surfaces (FBOs). src/cairo-gl-surface.c | 375 ++++++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 373 insertions(+), 2 deletions(-) commit 38a1061bbcf3e69acc05c35f2084d462b43f5477 Merge: b3e2433f1 e7d4bc3d8 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Feb 12 10:46:29 2009 +0000 Merge branch '1.8' commit e7d4bc3d864b1b42bb1cae031036ddf6a4445d3c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Feb 12 10:42:24 2009 +0000 [png] Correct documentation to avoid reference to NULL As pointed out by Truc Truong, cairo_image_surface_create_from_png_stream() cannot return NULL and so the documentation was incorrect. Fixes http://bugs.freedesktop.org/show_bug.cgi?id=20075 Bug 20075 There is a misprint in the spec for cairo_image_surface_create_from_png_stream() function src/cairo-png.c | 17 +++++++++++++++-- 1 file changed, 15 insertions(+), 2 deletions(-) commit b3e2433f1e78e8799a9f57bfb6da108016687fc9 Author: Søren Sandmann <sandmann@redhat.com> Date: Wed Feb 11 17:55:15 2009 -0500 Correct the reference image for the rotate-image-surface-paint test test/rotate-image-surface-paint.ref.png | Bin 190 -> 191 bytes 1 file changed, 0 insertions(+), 0 deletions(-) commit 41cbd935f9dba276db716e2c71ac21dc60505be9 Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Wed Feb 11 15:24:20 2009 -0500 [test] Add cairo_test_NaN and use it in place of strtod strtod("NaN") returns 0.0 with the MSVC runtime so we need to generate NaN some other way. test/cairo-test.h | 15 +++++++++++++++ test/invalid-matrix.c | 2 +- test/user-font-rescale.c | 2 +- 3 files changed, 17 insertions(+), 2 deletions(-) commit 70297f257d4dc0accb5183b806d43a033887acb7 Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Tue Feb 10 17:58:28 2009 -0500 [test] Add crtdbg.h include crtdbg.h is required for _CrtSetReportMode() and _CrtSetReportFile(). test/cairo-test-runner.c | 4 ++++ 1 file changed, 4 insertions(+) commit b9f0c4b25223830ce73f7e3adef85a0e97a31c0e Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Tue Feb 10 17:57:00 2009 -0500 Use _cairo_round() instead of round() MSVC doesn't have round() and the code probably wants to round in the same direction regardless of whether the values are negative or positive. src/cairo-font-face-twin.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) commit dffdbd85157395bceb27d30d6426aa47173f6a18 Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Tue Feb 10 17:53:39 2009 -0500 Add _cairo_round() _cairo_round() has the same behaviour as _cairo_lround() except it returns a double instead of an integer. src/cairoint.h | 9 +++++++++ 1 file changed, 9 insertions(+) commit 41e46c7754fff3a99927979925300a7588b9962f Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Tue Feb 10 11:44:20 2009 -0500 [test] Don't embed preprocessor directives inside macros MSVC can't handle this. GCC will warn with -pedantic, but I'm not sure we want to enable that. test/user-font-mask.c | 11 ++++++----- test/user-font-proxy.c | 10 +++++----- test/user-font.c | 10 +++++----- 3 files changed, 16 insertions(+), 15 deletions(-) commit fe40d3bdb0d0cb93169b935e44d14b15e014fa6d Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Mon Feb 9 23:43:04 2009 -0500 [test] Allocate glyph array with malloc Keeping it on the stack causes a stack overflow of Window's default 1mb stack. test/show-glyphs-many.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) commit 50f8c2e4f3407ef237afd94c7317b1e185ccc9dc Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Mon Feb 9 23:39:38 2009 -0500 [test] define isnan() on MSVC Needed for user-font-rescale.c test/cairo-test.h | 4 ++++ 1 file changed, 4 insertions(+) commit dd9fc47418b86610d7896fcf585037c96beaa66f Author: Jeff Muizelaar <jeff@infidigm.net> Date: Mon Feb 9 16:31:08 2009 -0500 Remove zero size data array for compilation with MSVC MSVC doesn't support zero sized arrays very well. For example, zero sized arrays in arrays. src/cairo-tor-scan-converter.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) commit a6c8f18a391407044b33340cf13c49368a2e810e Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Mon Feb 9 15:59:01 2009 -0500 [test] Avoid C99 designated initializers MSVC doesn't support these so we shouldn't use them. test/path-precision.c | 20 ++++++++++---------- 1 file changed, 10 insertions(+), 10 deletions(-) commit 3d2fcec948b93f75ba49b898633a84f60d419779 Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Mon Feb 9 15:12:41 2009 -0500 [test] Define INFINITY on MSVC 1./0. produces a compiler error on MSVC so we'll use HUGEVAL instead. test/cairo-test.h | 3 --- test/invalid-matrix.c | 4 ++-- 2 files changed, 2 insertions(+), 5 deletions(-) commit 6fb4c6200c840d438e1c9cdd2c9e7b0f140863d8 Author: Eric Anholt <eric@anholt.net> Date: Thu Feb 5 10:29:01 2009 -0800 [gl] Clamp surface size to a minimum of (1,1) in create_similar. Fixes clip-zero test. src/cairo-gl-surface.c | 5 +++++ 1 file changed, 5 insertions(+) commit a8a40810275ff0a59f5767034372523d34fbbd1b Author: Eric Anholt <eric@anholt.net> Date: Thu Feb 5 09:43:38 2009 -0800 [gl] clear surfaces to transparent on creation src/cairo-gl-surface.c | 7 +++++++ 1 file changed, 7 insertions(+) commit e7eb6dab12b9d622b80e4202d753324f7161b269 Author: Eric Anholt <eric@anholt.net> Date: Thu Jan 29 22:31:56 2009 -0800 [gl] Hook up fill_rectangles. First actual "acceleration" for cairogears. Actually slows things down since we're now pingponging between cpu and gpu. src/cairo-gl-surface.c | 119 ++++++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 118 insertions(+), 1 deletion(-) commit 2ad5c3fc8fdb5941bdedcda3c972b3bae426fb32 Author: Eric Anholt <eric@anholt.net> Date: Wed Feb 4 13:10:47 2009 -0800 [gl] Add a shortcut from image to gl in clone_similar. src/cairo-gl-surface.c | 27 +++++++++++++++++++++++++++ 1 file changed, 27 insertions(+) commit d9d56bb2ed552dd40415af53105a98b9d6ba39ab Author: Eric Anholt <eric@anholt.net> Date: Wed Feb 4 13:06:31 2009 -0800 [gl] Make create_similar actually create similar instead of image. src/cairo-gl-surface.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) commit 90b30219b945d6cce941a1b702c9b789a60d8b27 Author: Eric Anholt <eric@anholt.net> Date: Thu Jan 29 22:27:39 2009 -0800 [gl] Replace all-software implementation with all-GL-read/drawpixels. src/cairo-gl-surface.c | 315 +++++++++++++++++++++++++++++++++++++++++++------ 1 file changed, 282 insertions(+), 33 deletions(-) commit c22ca79863a54b4293e607c8e4d515868c216842 Author: Behdad Esfahbod <behdad@behdad.org> Date: Fri Feb 6 12:56:48 2009 -0500 Don't distribute cairo-features.h (#19992) Second time I fix this bug. Mabe we should add a test for it! src/Makefile.am | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) commit a55655a82de8dee0f171efb2f7dfb1da341ecdd6 Author: Behdad Esfahbod <behdad@behdad.org> Date: Fri Feb 6 12:47:38 2009 -0500 [test] Fix "make dist" doc/public/tmpl/cairo-status.sgml | 1 + test/Makefile.am | 1 - 2 files changed, 1 insertion(+), 1 deletion(-) commit af5ca7249f7628a8b0758f51934e3c1f6d9f4f36 Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Thu Feb 5 16:46:50 2009 -0500 [test] Fix the definition of INFINITY In my defence, even with INFINITY as 0, I was seeing new asserts on win32. Caught by Jeremy Lea. test/invalid-matrix.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 235de8d7a492ccd132fb83f78d33b408bd76aec1 Author: Eric Anholt <eric@anholt.net> Date: Thu Jan 29 09:19:01 2009 -0800 [gl] Add basics for GL surface backend using test-fallback as base. boilerplate/Makefile.sources | 3 + boilerplate/Makefile.win32.features | 10 + boilerplate/cairo-boilerplate-gl-private.h | 59 ++++++ boilerplate/cairo-boilerplate-gl.c | 124 +++++++++++++ boilerplate/cairo-boilerplate.c | 25 +++ build/Makefile.win32.features | 1 + build/Makefile.win32.features-h | 3 + build/configure.ac.features | 1 + configure.ac | 7 + perf/cairo-perf.c | 1 + src/Makefile.sources | 4 + src/Makefile.win32.features | 14 ++ src/cairo-gl-surface.c | 287 +++++++++++++++++++++++++++++ src/cairo-gl.h | 89 +++++++++ src/cairo-mutex-list-private.h | 4 + src/cairo.h | 4 +- 16 files changed, 635 insertions(+), 1 deletion(-) commit 83ec4b16b7b9a27439fee2d84c50e30a1ec2d68c Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Wed Feb 4 14:04:05 2009 -0500 Elaborate the meaning of arithmetic rounding as done by _cairo_lround() src/cairo-misc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 2676b3bdd6fbeef62d34c100718f76fe5bbe647d Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Tue Feb 3 14:18:44 2009 -0500 [test] Define INFINITY if it hasn't been defined. This lets us use more of invalid-matrix on win32 test/cairo-test.h | 3 +++ test/invalid-matrix.c | 22 ++-------------------- 2 files changed, 5 insertions(+), 20 deletions(-) commit 6b0c2c8ba61b6fdaad4c77e1a7c02c13c70e03c4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jan 31 11:58:19 2009 +0000 [directfb] Don't access the scaled_font->glyphs cache directly Not only does it no longer exist, but doing so bypassed the locking mechanisms and sanity checks! Spotted by Lance Hepler. src/cairo-directfb-surface.c | 20 ++++++++++---------- 1 file changed, 10 insertions(+), 10 deletions(-) commit d63c1ab3ffcb64220a05c80e674324f524f29dc2 Author: Behdad Esfahbod <behdad@behdad.org> Date: Sat Jan 31 22:10:57 2009 -0500 [util/malloc-stats] Do a single backtrace_symbols() call I was hoping that this may speed things up, but it didn't. :( util/malloc-stats.c | 58 +++++++++++++++++++++++++++++++++++++++++++++-------- 1 file changed, 50 insertions(+), 8 deletions(-) commit 64d1c7587041f765b393e1802a10cce02b807ad1 Author: Behdad Esfahbod <behdad@behdad.org> Date: Sat Jan 31 21:34:21 2009 -0500 [util] Fix memset bug in malloc-stats util/malloc-stats.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) commit b02aeb367dc4bada5412798fa93ce74f3d861273 Author: Behdad Esfahbod <behdad@behdad.org> Date: Sat Jan 31 21:33:42 2009 -0500 [util] Don't link backtrace-symbols into malloc-stats On my rawhide machine, seems like the glibc backgrace-symbols is now as good as our private backtrace-symbols. util/Makefile.am | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 7375f4c76bb169cc24d5d714520c337172d3cf69 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 30 21:54:56 2009 +0000 Trivial warning fixes. Cleanup a few compiler warnings about unused variables and mismatching pointer types. src/cairo-bentley-ottmann.c | 1 - src/cairo-ps-surface.c | 4 ++-- 2 files changed, 2 insertions(+), 3 deletions(-) commit f10eaadf89fa94172a81c05a59febab3f7e32396 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 30 21:50:24 2009 +0000 Revert "[png] Complete the idempotent read_png() -> write_png()" This reverts commit 564d64a1323c5cbcde2dd9365ac790fe8aa1c5a6. In hindsight, and with further discussion with Jeff Muizelaar, this behaviour of using the stored contents from the mime-data is completely the opposite of the users' expectations. When the user calls cairo_surface_write_to_png(), usually in the course of debugging their rendering code, they expect the precise contents of the surface to be saved. src/cairo-png.c | 14 -------------- 1 file changed, 14 deletions(-) commit dd11d905a54a123ddf619e5f0194fb1800ba643d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 30 10:09:29 2009 +0000 [util] Use a hash-table for malloc-stats. At Behdad's request, convert the array of allocators into a simple hash table (large static number of buckets + chaining) in order to speed up malloc profiling. util/malloc-stats.c | 150 ++++++++++++++++++++++++++++++++++++---------------- 1 file changed, 103 insertions(+), 47 deletions(-) commit 322fb00066cc4655122fcf7d738a0cbbe46fcdd1 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jan 29 22:05:06 2009 +0000 [test] Compile again without memfault. Hide the valgrind macro when not using memfault. test/cairo-test.c | 26 ++++++++++++++++---------- 1 file changed, 16 insertions(+), 10 deletions(-) commit 5e6d25e204b681c5d5fba90abfe4d7401f23460f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Dec 17 20:47:09 2008 +0000 [skiplist] Provide an initial stack allocated pool. Since we only need to allocate elts for intersection events and edges, the number of elts in flight at any one time is actually quite small and can usually be accommodated from an embedded pool. src/cairo-bentley-ottmann.c | 33 ++++++++++----------------------- src/cairo-skiplist-private.h | 9 +++++---- src/cairo-skiplist.c | 13 ++++++------- 3 files changed, 21 insertions(+), 34 deletions(-) commit dd4276c6618aa250637e4499bc7cb0a35b24448c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Nov 30 13:49:24 2008 +0000 [stroker] Rectilinear dashing. Extend the rectilinear stroker to handle dashes, so that for pixel-aligned dashed strokes we completely avoid tessellation overhead. src/cairo-matrix.c | 3 +- src/cairo-path-stroke.c | 492 +++++++++++++++++---- src/cairo-pen.c | 11 +- src/cairo-stroke-style.c | 18 +- src/cairoint.h | 11 +- test/Makefile.am | 4 +- test/leaky-dashed-rectangle.pdf.ref.png | Bin 0 -> 347 bytes ...2.ref.png => leaky-dashed-rectangle.ps.ref.png} | Bin test/leaky-dashed-rectangle.ps3.ref.png | Bin 444 -> 0 bytes test/leaky-dashed-rectangle.ref.png | Bin 366 -> 347 bytes 10 files changed, 437 insertions(+), 102 deletions(-) commit 7f95288c03a400bf770165d427ef623d924b3b47 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jan 29 09:47:01 2009 +0000 [ft] Improve error status propagation. Propagate the error status from deep within the bowels, in order to reduce the number of duplicate _cairo_error() and generally clean up the return values. src/cairo-ft-font.c | 65 ++++++++++++++++++++++++++--------------------------- 1 file changed, 32 insertions(+), 33 deletions(-) commit 53bd2ae2ce27f9b954f34bc9921d798c9a074125 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jan 26 21:36:27 2009 +0000 [ft] Distinguish fatal backend errors whilst constructing scaled fonts. We now have the ability to distinguish an error case where the backend is left in an inconsistent state from a transitory error. For the former we need to report the error condition via the return value, which will be propagated to the font-face. For the latter we just construct an in-error scaled font nil-object which is passed back to the user. src/cairo-ft-font.c | 38 ++++++++++++++++++++++++++------------ 1 file changed, 26 insertions(+), 12 deletions(-) commit f17aeedab31753974cce027f92571107425b1bcd Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jan 26 21:11:41 2009 +0000 [scaled-font] Differentiate fatal error when creating fonts We only want to set the error state on the backend when it implies that the font-face is in an inconsistent state. For example, this may be due to a locking error in the backend or that we have detected a corrupt font. In contrast, if we merely fail to allocated the scaled font then we just wish to return that error to the user, without making the font-face itself inert. src/cairo-scaled-font.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) commit 58cab06c4c2c49bbecb11efaae6b41d30c06eff0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jan 26 20:52:29 2009 +0000 [scaled-font] Guard against invalid matrices when creating the scaled font. Check the user input for validity before passing the values on to the backend. Currently the error is detected by the backend and the error is propagated onto the font-face. src/cairo-scaled-font.c | 22 ++++++++++++++++++++-- 1 file changed, 20 insertions(+), 2 deletions(-) commit 97c88f2af04b6bc5161fa2b567b5e922d7fd326a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jan 26 20:51:02 2009 +0000 [surface] Fix memleak of along set_mime_data() error path Free the mime_data holder if we fail to attach it to the surface. src/cairo-surface.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) commit d20e5fc2d95c61ab04e085bf3a99d2cb958421a5 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jan 26 15:08:58 2009 +0000 [ps] Free images after emitting patterns. Ensure that the temporary images are freed after we finish with the pattern. Note that we are using 3 members of the surface for temporary storage whilst emitting patterns, this should be reviewed. src/cairo-ps-surface-private.h | 3 +++ src/cairo-ps-surface.c | 18 ++++++++++++++++-- 2 files changed, 19 insertions(+), 2 deletions(-) commit e6102dbe028ca93db936b2f4cd6368e2ba0a2209 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jan 26 14:00:13 2009 +0000 [png] Avoid a double free of the memory stream after error. _cairo_memory_stream_destroy() finalizes the stream even if the stream was in error and that error is reported back to the caller - so ensure we don't try to free the stream again. src/cairo-png.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 0f3e366f8bbbaa80b518eb1b0297a6122901ce66 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jan 26 10:54:45 2009 +0000 [font-face] Close a race when resurrecting fonts. Paul Messmer provided a thorough analysis of a race between destroying the final reference on a font and a concurrent recreation of the font - demonstrating how it is possible for the create() to return the font that was in the process of being freed. To stop the race, we need to recheck the reference count upon taking the mutex guarding the hash table. src/cairo-ft-font.c | 6 ++++++ src/cairo-toy-font-face.c | 6 ++++++ 2 files changed, 12 insertions(+) commit 312b5680a5754c8e7ee1332206b81449cf9bf8a3 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 23 21:13:15 2009 +0000 [cff-subset] Free ps_name. Ensure ps_name is freed along error paths and by the normal destructor. src/cairo-cff-subset.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) commit 5176507fcb61ae1ec1143aa0b6b098bc92575c48 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 23 13:37:54 2009 +0000 [truetype] Free font name. Remember to free the font name on destruction. src/cairo-truetype-subset.c | 2 ++ 1 file changed, 2 insertions(+) commit ab0ac1b8a84b0d259602f0029a3b5552466f35a6 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jan 29 09:45:11 2009 +0000 [truetype] Initialise font_name Ensure the font_name is initialized to NULL. src/cairo-cff-subset.c | 2 ++ src/cairo-truetype-subset.c | 1 + 2 files changed, 3 insertions(+) commit 8388af137841679b7c510980daf3cec1427b6e6b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jan 29 09:50:38 2009 +0000 [test] Trivial fixes for error paths. Kill a few leaks along error paths in the test code. test/create-from-png-stream.c | 36 ++++++++++++++++++++++++++++-------- test/mime-data.c | 4 ++++ 2 files changed, 32 insertions(+), 8 deletions(-) commit 3752f690b467432ab5b1058d450cb79d719a794a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jan 29 09:49:41 2009 +0000 [test] Suppress suppressed memfault report Check to see if there are any *unsuppressed* memfaults before declaring unreported faults. test/cairo-test.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) commit 8dc4c0da9b13b16c593e874d59c13a89a77a2481 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 23 13:17:24 2009 +0000 [toy-font] Fix unwind behaviour following error during construction. We failed to cleanup the font face correctly after an allocation failure during _cairo_toy_font_face_init() leading to memleaks and live entries being left in the font-face hash tables. src/cairo-toy-font-face.c | 43 +++++++++++++++++++++---------------------- 1 file changed, 21 insertions(+), 22 deletions(-) commit 1d52fbc8f4f70e9e2419a6ed66cd907552d1d13b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 23 12:51:52 2009 +0000 [tessellator] Memleak on error path. Add a missing _cairo_skip_list_fini() after failure to allocate the events. src/cairo-bentley-ottmann.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) commit 6b5d2bf1a742b34a58d65f188fe15ffbf2f83118 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 23 10:42:42 2009 +0000 [trace] Comment out the redundant wrapping of FT_Open_Face() Remove the left-over debugging spew, but leave a comment to hopefully clarify the situation with wrapping FT_Open_Face(). util/cairo-trace/trace.c | 9 +++++++++ 1 file changed, 9 insertions(+) commit faa004033cde21cd81890c7f82abae8eb766bb4b Author: Carl Worth <cworth@cworth.org> Date: Thu Jan 22 11:52:54 2009 +1100 cairo-trace: Print name of trace file. It's just a lot easier to use cairo-trace if it tells you what file it just created. util/cairo-trace/trace.c | 3 +++ 1 file changed, 3 insertions(+) commit d108b2777fcd6ef4fa45aeeef457dc58522e325b Author: Daniel Holbert <dholbert@mozilla.com> Date: Fri Jan 23 10:18:48 2009 +0000 Spelling corrections: s/it's/its/ As a fun itch to scratch, I've been fixing incorrect uses of the contraction "it's" in comments within the mozilla source tree (tracked in https://bugzilla.mozilla.org/show_bug.cgi?id=458167 ), and I ran across 6 instances of this typo in mozilla's snapshot of cairo. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> src/cairo-ft-font.c | 2 +- src/cairo-matrix.c | 2 +- src/cairo-mutex-impl-private.h | 2 +- src/cairo-pdf-operators.c | 2 +- src/cairo-surface-fallback.c | 2 +- 5 files changed, 5 insertions(+), 5 deletions(-) commit 6394ec3048f31b867d9588853fa400c6c630c6f1 Author: Paolo Bonzini <bonzini@gnu.org> Date: Mon Dec 15 09:32:43 2008 +0100 [surface] add CAIRO_STATUS_INVALID_SIZE Adds an error code replacing CAIRO_STATUS_NO_MEMORY in one case where it is not really appropriate. CAIRO_STATUS_INVALID_SIZE is used by several backends that do not support image sizes beyond 2^15 pixels on each side. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> src/cairo-misc.c | 4 +++- src/cairo-os2-surface.c | 9 +++++---- src/cairo-quartz-image-surface.c | 5 +++-- src/cairo-quartz-surface.c | 4 ++-- src/cairo-spans.c | 2 ++ src/cairo-surface.c | 1 + src/cairo-xlib-surface.c | 7 +++---- src/cairo.h | 6 ++++-- src/cairoint.h | 2 +- 9 files changed, 24 insertions(+), 16 deletions(-) commit 46acfd2e85dd6f7a73e1172d363d509c769376f2 Author: Paolo Bonzini <bonzini@gnu.org> Date: Tue Dec 23 08:31:30 2008 +0100 [glitz] use image fallback if the cairo_content_t is unsupported The agreement on the mailing list was that returning NULL is the right thing to do, and indeed the callers of _cairo_glitz_surface_create_similar are prepared to receive NULL and return CAIRO_STATUS_INT_UNSUPPORTED in that case. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> src/cairo-glitz-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 01d20b79daf0abe0f69ccec4ecd5122c5bfe9a4e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 23 10:09:37 2009 +0000 [scaled-font] Fix up syntax in doc comments The old NULL vs %NULL conflict. src/cairo-scaled-font-subsets-private.h | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) commit aaec63d48386ec825cd4d6e67b6adf7c5fd3b167 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Nov 21 15:04:47 2008 +0000 [scaled-font] Global glyph cache Currently glyphs are cached independently in each font i.e. each font maintains a cache of up to 256 glyphs, and there can be as many scaled fonts in use as the application needs and references (we maintain a holdover cache of 512 scaled fonts as well). Alternatively, as in this patch, we can maintain a global pool of glyphs split between all open fonts. This allows a heavily used individual font to cache more glyphs than we could allow if we used per-font glyph caches, but at the same time maintains fairness across all fonts (by using random replacement) and provides a cap on the maximum number of global glyphs. The glyphs are allocated in pages, which are cached in the global pool. Using pages means we can exploit spatial locality within the font (nearby indices are typically used in clusters) to reduce frequency of small allocations and allow the scaled font to reserve a single MRU page of glyphs. This caching dramatically reduces the cairo overhead during the cairo-perf benchmarks, and drastically reduces the number of allocations made by the application (for example browsing multi-lingual site with firefox). boilerplate/cairo-boilerplate.c | 5 +- src/cairo-cache-private.h | 8 ++ src/cairo-cache.c | 36 +++-- src/cairo-hash-private.h | 4 + src/cairo-hash.c | 117 ++++++++++----- src/cairo-mutex-list-private.h | 1 + src/cairo-scaled-font-private.h | 6 +- src/cairo-scaled-font.c | 306 ++++++++++++++++++++++++++++------------ src/cairoint.h | 9 +- 9 files changed, 351 insertions(+), 141 deletions(-) commit 54f6a49ebb18cf396823d0d70b95e4e264142171 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jan 19 15:36:29 2009 +0000 [bounds] Skip spline evaluation based on bounding bbox of control points. The bounding polygon of the control points, defines the extents of the spline. Therefore if the control points are entirely contained within the current path extents, so is the spline and we do not need to evaluate its tight bounds. src/cairo-path-bounds.c | 21 +++++++++++++++++++-- 1 file changed, 19 insertions(+), 2 deletions(-) commit e217c4da7bc5c4817e0d829ff61dd2bd5b3145a6 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jan 19 15:28:05 2009 +0000 [in-stroke] Check point against extents before computing path. We can avoid tessellating the path entirely by first checking whether the query point is inside the path extents. src/cairo-gstate.c | 15 +++++++++++++++ 1 file changed, 15 insertions(+) commit 48f9a0e6da0dd24ea9c809876ef3c745dcfd0d52 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jan 15 00:26:03 2009 +0000 [spline] Correct the definition of a cubic Bezier curve. Add the missing coefficients for p1 and p2 so the derivation of the derivative and the solution for its inflection points stands correct. src/cairo-spline.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) commit ee7ac5681fa6a74b68beeae667d96d1421050fc9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jan 21 12:04:06 2009 +0000 [path] A degenerate curve_to becomes a line_to. Be consistent. src/cairo-path-fixed.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 778ced4879b09f7482bd41c398bf2d984754ed0b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jan 14 18:55:32 2009 +0000 [path] Rename _cairo_path_fixed_approximate_extents() Rename approximate_extents() to approximate_clip_extents() so that it is consistent with the fill and stroke variants and clearer under what circumstances you may wish to use it. src/cairo-analysis-surface.c | 2 +- src/cairo-clip.c | 4 ++-- src/cairo-path-bounds.c | 4 ++-- src/cairo-surface-fallback.c | 2 +- src/cairoint.h | 4 ++-- 5 files changed, 8 insertions(+), 8 deletions(-) commit 75f7c420b624049c1f6c51795679f8029cd2231d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jan 21 13:09:41 2009 +0000 [perf] Remove a redundant clear during source init. After a short wild goose chase to see why cairo_image_surface_fill_rectangles() was appearing in the profile, tweak init_and_set_source_surface() to remove the redundant clear and to propagate any errors in the auxiliary context. perf/cairo-perf-cover.c | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) commit 706f6de68da65911f434d2065dcb143649fa793e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Jan 18 16:47:59 2009 +0000 [perf] Add another variation on the many-rectangles case This variation aims to show the difference between calling fill once per-rectangle, or once for all rectangles. perf/rectangles.c | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+) commit ff5d37a8ad063e84e88f453a403715bc85f8a3ec Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jan 29 10:09:11 2009 +0000 [mutex] Civilise the comment. Note bene that Behdad does not like people shouting. src/cairo-mutex-impl-private.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit f4ff6128d7fb6e4b5ba361a597fc6c0c88d5ec00 Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Wed Jan 28 17:16:32 2009 -0500 Avoid "empty body in an if-statement" warning Use '(void)expr;' instead of 'if (expr) ;' to avoid getting the warning. 'if (expr) {}' is an option, however '(void)expr;' seems like a more common idiom and getting warnings for __attribute__((warn_unsed_result)) functions is probably prefered. src/cairo-mutex-impl-private.h | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) commit 41feeedcc14bf8caef3c039de49f4f28143712c7 Author: Adrian Johnson <ajohnson@redneon.com> Date: Wed Dec 3 23:58:22 2008 +1030 Use PS font name in PS TrueType fonts src/cairo-ps-surface.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) commit b7a9e1d4ac3972bc3d215070124b6a9eda68d3e3 Author: Adrian Johnson <ajohnson@redneon.com> Date: Wed Dec 3 23:58:05 2008 +1030 Embed full font name in PDF TrueType and CFF fonts if the full font name was available in the font. src/cairo-pdf-surface.c | 32 ++++++++++++++++++++++++-------- 1 file changed, 24 insertions(+), 8 deletions(-) commit 6f2db9a4b07cde2c4932ea481228abc248e90145 Author: Adrian Johnson <ajohnson@redneon.com> Date: Wed Dec 3 23:57:38 2008 +1030 Use PS font name in CFF and TrueType PDF font subsets James Cloos found that the font name in embedded fonts should be the PostScript font name (nameID=6 in the name table). http://lists.cairographics.org/archives/cairo/2008-December/015919.html src/cairo-cff-subset.c | 70 +++++++++++++++--------- src/cairo-pdf-surface.c | 16 +++--- src/cairo-scaled-font-subsets-private.h | 19 +++++-- src/cairo-truetype-subset.c | 97 +++++++++++++++++++++++---------- 4 files changed, 136 insertions(+), 66 deletions(-) commit 2ed08f7801a2af27e35afcf57f00f4bf5d48384a Author: Adrian Johnson <ajohnson@redneon.com> Date: Wed Dec 3 23:56:01 2008 +1030 Factor out duplicate code in truetype and cff subsetting The code for reading the font name from the name table has been moved to a new function: _cairo_truetype_read_font_name(). src/cairo-cff-subset.c | 60 ++----------- src/cairo-scaled-font-subsets-private.h | 18 ++++ src/cairo-truetype-subset.c | 149 +++++++++++++++++++------------- 3 files changed, 114 insertions(+), 113 deletions(-) commit 1deb1e451022b9dd5aa6ecb0b580b006047b630e Author: Adrian Johnson <ajohnson@redneon.com> Date: Tue Nov 25 23:11:01 2008 +1030 PDF: Include subset tag in font name PDF requires font names of subsetted fonts to be preprended with "XXXXXX+" where XXXXXX is a sequence of 6 uppercase letters unique the font and the set of glyphs in the subset. src/cairo-pdf-surface.c | 112 ++++++++++++++++++++++++++++++++++++++++++++---- 1 file changed, 104 insertions(+), 8 deletions(-) commit c4e54629bb444ed3e850ca8deec175936b90c4e4 Author: Carl Worth <cworth@cworth.org> Date: Thu Jan 22 12:26:55 2009 +1100 Add details to test/COPYING about license of bundled fonts. Just the public-doamin 6x13.pcf file for now, but Adrian will be bundling more soon. test/COPYING | 8 ++++++++ 1 file changed, 8 insertions(+) commit 6801f28f6dfeb21eec44052e75156e9d2b82422e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jan 14 13:59:28 2009 +0000 [perf] Add a utility to compare backends. A minor variation on cairo-perf-diff-files that compares tests with the same name for multiple backends. perf/.gitignore | 1 + perf/Makefile.am | 8 +- perf/cairo-perf-compare-backends.c | 393 +++++++++++++++++++++++++++++++++++++ perf/cairo-perf-diff-files.c | 2 +- perf/cairo-perf-graph-files.c | 2 +- perf/cairo-perf-report.c | 43 +++- perf/cairo-perf.h | 9 +- 7 files changed, 448 insertions(+), 10 deletions(-) commit 4c79cd480db1cf10b6018bce3ea966587efc6081 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Jan 13 12:52:52 2009 +0000 [perf] Tweak dragon to hit fill_rectangles(). Add a second dragon path that is pixel-aligned and uses a solid pattern, so that it can be drawn using fill-rectangles. perf/dragon.c | 42 +++++++++++++++++++++++++++++++++++++++++- 1 file changed, 41 insertions(+), 1 deletion(-) commit de9e6b5a3f4e4752e0f99e3ae20ac263a5aae4bb Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jan 12 01:00:30 2009 +0000 [perf] Cover linear gradient with 3 stops. The i915 is able to special case gradients with just 2 color stops to avoid creating temporary gradient textures, so add a 3 stop linear gradient to compare the speed difference. perf/cairo-perf-cover.c | 36 ++++++++++++++++++++++++++++++++++++ 1 file changed, 36 insertions(+) commit 7cbc55f21624159dfa58a9a50ec004af9368ab3d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jan 10 00:12:06 2009 +0000 [perf] Add scaled similar surface sources. Cover the similar source with min/mag scale factors as well, so we can compare the performance impact with scaled image sources. This is useful to distinguish between transport overhead and transform cost. perf/cairo-perf-cover.c | 44 ++++++++++++++++++++++++++++++++++++++++---- 1 file changed, 40 insertions(+), 4 deletions(-) commit afce1cfe987eeec6516aed1eb8fd97c2d3b9b07b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 9 14:22:36 2009 +0000 [scaled-font] Avoid repeated lookup of the same unicode during text->glyphs Performing the unicode to index is quite expensive, the FcFreeTypeCharIndex() taking over 12% in the cairo-perf text benchmarks. By adding a simple cache of translated unicode indices, we save around 25% of the lookups during benchmarks, with a relative reduction in runtime. src/cairo-scaled-font.c | 63 ++++++++++++++++++++++++++++++++----------------- 1 file changed, 41 insertions(+), 22 deletions(-) commit 49eca78a4265432e285af58435219e8b804b38bb Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Jan 12 12:42:46 2009 +0000 [test] Add a huge-radial test case. Also test the handling of radial gradients with large radii. test/Makefile.am | 11 ++-- test/{huge-pattern.c => huge-linear.c} | 6 +- ...pattern.ps3.ref.png => huge-linear.ps3.ref.png} | Bin test/{huge-pattern.ref.png => huge-linear.ref.png} | Bin test/huge-pattern.pdf.argb32.ref.png | Bin 2430 -> 0 bytes test/huge-pattern.pdf.ref.png | Bin 2716 -> 0 bytes test/huge-pattern.pdf.rgb24.ref.png | Bin 2817 -> 0 bytes test/huge-radial.c | 69 +++++++++++++++++++++ test/huge-radial.ps3.ref.png | Bin 0 -> 18449 bytes test/huge-radial.ref.png | Bin 0 -> 41702 bytes 10 files changed, 79 insertions(+), 7 deletions(-) commit 7709d1d9d43a42dd8f9293f18141c57c76ca0970 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Jan 7 21:12:21 2009 +0000 [test] Add fill-image. A filled equivalent of stroke-image, that checks that the pattern matrices are applied correctly during fills - useful with the segregation between fills and strokes introduced by spans. test/Makefile.am | 2 ++ test/fill-image.c | 83 ++++++++++++++++++++++++++++++++++++++++++++++++ test/fill-image.ref.png | Bin 0 -> 1317 bytes 3 files changed, 85 insertions(+) commit 5605e4bfcd0b6e3e34eed3785bc8ae51b24a7385 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Jan 4 11:26:45 2009 +0000 [test] Propagate failure from painting large-source-roi Use cairo_get_target() to propagate any failure when creating the large-source. test/large-source-roi.c | 28 ++++++++++++++++++++-------- 1 file changed, 20 insertions(+), 8 deletions(-) commit 59a14f622389ae3f34f93c78b709df2ad6d1d624 Author: Behdad Esfahbod <behdad@behdad.org> Date: Sat Jan 10 13:09:06 2009 -0500 Comment win32 maintainer-clean files again Makefile.am | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) commit cc5119173918623ce13363f4b25cd06076fc3fbc Author: Søren Sandmann Pedersen <sandmann@redhat.com> Date: Sat Jan 10 12:13:40 2009 -0500 Uncomment win32 maintainer-clean files to make Makefile.am syntactically correct Makefile.am | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 06deaa98b67f9e674a6933a8f54dae6a45faf6ff Author: Behdad Esfahbod <behdad@behdad.org> Date: Fri Jan 9 15:55:24 2009 -0500 [build] Include all generated win32 build files in the repo So a git clone can be built on win32. The files only change after adding new backends. Makefile.am | 4 +- boilerplate/.gitignore | 2 +- boilerplate/Makefile.am | 2 +- boilerplate/Makefile.win32.features | 233 +++++++++++++++++++++++++++ build/.gitignore | 4 +- build/Makefile.win32.features | 22 +++ build/Makefile.win32.features-h | 70 ++++++++ src/.gitignore | 3 +- src/Makefile.am | 2 +- src/Makefile.win32.features | 311 ++++++++++++++++++++++++++++++++++++ 10 files changed, 644 insertions(+), 9 deletions(-) commit a5c1cdf2b030959309e1645bd115f058f1367296 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Thu Nov 13 12:11:12 2008 +0200 [clip] Fix uninitialised status return in _cairo_clip_intersect_mask() for empty extents. This fixes the clip-all test case crashing for me. src/cairo-clip.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 8157bc8a2dc0d555606d19ad52ae7a603471edd6 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Sun Jan 4 04:14:28 2009 +0200 [test] Stress test using large source images. This test attempts to trigger failures in those backend clone_similar methods that have size restrictions on the resulting image. It also triggers errors in scaling down large image surfaces as the image backend also fails this test. test/Makefile.am | 2 ++ test/large-source-roi.c | 64 ++++++++++++++++++++++++++++++++++++++++++ test/large-source-roi.ref.png | Bin 0 -> 102 bytes 3 files changed, 66 insertions(+) commit 388ae177e4100698289819429fa1f8e6958d1c60 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jan 3 22:51:28 2009 +0000 [boilerplate] Remove CAIRO_BOILERPLATE_LOG() The variadic macro is only used within boilerplate/ so replace it with a simple, and portable, call to fprintf. boilerplate/cairo-boilerplate-beos.cpp | 19 +++---------------- boilerplate/cairo-boilerplate-glitz-agl.c | 10 +++++----- boilerplate/cairo-boilerplate-glitz-glx.c | 6 +++--- boilerplate/cairo-boilerplate-glitz-wgl.c | 10 +++++----- boilerplate/cairo-boilerplate-system.c | 16 ++++++++-------- boilerplate/cairo-boilerplate-xcb.c | 4 ++-- boilerplate/cairo-boilerplate.h | 4 ---- test/cairo-test.h | 1 - 8 files changed, 26 insertions(+), 44 deletions(-) commit 5f816ccd25d1cd303fc1e9e44e80c1207b2a424a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jan 3 22:30:55 2009 +0000 [boilerplate] Redefine DEBUG() for portability concerns Behdad warned that using an empty variadic macro was non-portable. boilerplate/cairo-boilerplate-xlib.c | 18 +++++++++--------- boilerplate/cairo-boilerplate.h | 2 +- 2 files changed, 10 insertions(+), 10 deletions(-) commit acb2717372f3862ddbde8cfdc814132808b71f86 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jan 3 21:50:55 2009 +0000 [quartz] Define RTLD_DEFAULT RTLD_DEFAULT is a gnu-ism (at least according to the manpage on my linux system) so declare _GNU_SOURCE before including dlfcn.h and failing that provide our own definition. src/cairo-quartz-surface.c | 5 +++++ 1 file changed, 5 insertions(+) commit ff1f5de5511ba0b7842b53223c26986e4bcdc38a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Jan 3 11:44:58 2009 +0000 [boilerplate] Suppress xlib warnings on stderr If we cannot test the xlib backend simply because there is no Display, just report UNTESTED and do not clutter the output with superfluous warnings [see the output from the buildbots for an example]. However, keep the warnings around so that a developer can re-enable them and so simply move them to a new "lower priority" macro. boilerplate/cairo-boilerplate-xlib.c | 18 +++++++++--------- boilerplate/cairo-boilerplate.h | 4 ++++ 2 files changed, 13 insertions(+), 9 deletions(-) commit 75538962c8af11b1ec669caca6259b7769b5cc1d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 2 17:45:26 2009 +0000 [boilerplate] Check the return of pclose() pclose() returns the child exit status, so we can use that to detect errors in the convertor process. boilerplate/cairo-boilerplate.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) commit 333158ec85cf3c610cc8965fc3f99d72b534cc2e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 2 15:36:39 2009 +0000 [configure] Replace awk comparator with an aclocal version As reported in https://bugs.freedesktop.org/show_bug.cgi?id=19283, the fallback freetype version compare is broken inside the configure script as the $1-$3 arguments are interpreted as the script is constructed. To avoid making that awk comparison any more complicated, we import a version compare from the autoconf archives - such that we have a reusable macro for the furture. build/aclocal.compare.m4 | 162 +++++++++++++++++++++++++++++++++++++++++++++++ configure.ac | 18 ++---- 2 files changed, 168 insertions(+), 12 deletions(-) commit dc33ae24619f4602c23716e9e407f8dd4f1b4a1d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 2 14:29:53 2009 +0000 [boilerplate] Use pclose() after popen Joonas pointed out that we should be using pclose() on a stream returned by popen(). boilerplate/cairo-boilerplate.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit f230ce7658910c7f4f8feb722b77a2141824f963 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 2 14:03:35 2009 +0000 [path] Fix typo in bounds for empty path. We set the width to be zero, twice, and the height not even once! src/cairo-path-bounds.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) commit 163c326c82a45c1f3ee84bbfaee2cc2e6dc1fafc Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 2 13:46:16 2009 +0000 [test] Free test name Adding the missing free for the converted test name. test/cairo-test.c | 3 +++ 1 file changed, 3 insertions(+) commit dd65be740c475daf75c602fc79ff25977674d9cf Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 2 13:42:53 2009 +0000 [test/pdf-mime-data] Free data on error paths. Cleanup the allocated buffers on error. test/pdf-mime-data.c | 4 ++++ 1 file changed, 4 insertions(+) commit 65f9760d661a0eb2edf9e53fb1b74666ce0ba3b9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 2 13:43:33 2009 +0000 [toy-font-face] Return defaults for error objects. Similar to the behaviour of the other objects, we return the default conditions if the object is in any error (and not just a nil object). src/cairo-toy-font-face.c | 28 ++++++++++++++++++++++++---- 1 file changed, 24 insertions(+), 4 deletions(-) commit d478d5ed5ce7c19a02e11435bcf69e867c6705d9 Author: Behdad Esfahbod <behdad@behdad.org> Date: Fri Jan 2 06:48:54 2009 -0500 [doc] Give a small hint about Twin font Though, the details are not documented yet. I'm not sure how much of it I do want to document. src/cairo-toy-font-face.c | 2 +- src/cairo.c | 7 +++++++ 2 files changed, 8 insertions(+), 1 deletion(-) commit a023104400f7f08775e9e52c304f5df2bc96382d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Jan 2 10:32:39 2009 +0000 [trace] Adopt _cairo_dtostr In order to have locale-independent output of decimal values, we need to manually transform such numbers into strings. As this is a solved problem for cairo, we adopt _cairo_output_stream_printf() and in particular the _cairo_dtostr() routine for our own printf processing. util/cairo-trace/trace.c | 920 ++++++++++++++++++++++++++++------------------- 1 file changed, 546 insertions(+), 374 deletions(-) commit c601f308430f4d33929109fb9170b9322edca3c6 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Jan 1 16:55:08 2009 +0000 [cairo] Early return if we attempt to set the same colour. Profiling a silly video renderer that called set-source; rectangle; fill; for each pixel, we can shave 5% off the cairo overhead by introducing an early return if we attempt to reset the current colour. src/cairo.c | 30 ++++++++++++++++++++++++++++++ 1 file changed, 30 insertions(+) commit fb3522f33a565576002595bf03e07f6c8b16a471 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Dec 20 19:19:54 2008 +0000 [os2] Fix memory leak of surface on error path Of we fail to create the mutex, free the surface before returning the failure. Reported: http://bugs.freedesktop.org/show_bug.cgi?id=19208. src/cairo-os2-surface.c | 1 + 1 file changed, 1 insertion(+) commit 8d23c3a6c2ab0ae168afb695e2b8c5f121ed2be3 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Dec 20 19:17:28 2008 +0000 [quartz] Delay allocation of string until after guard. Fixes a memory leak should we bail due to the version of Quartz being insufficient. Reported: http://bugs.freedesktop.org/show_bug.cgi?id=19209. src/cairo-quartz-font.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) commit 9c9ed8f0b5ce8bc3036dcb0bf841ffaa24797cfa Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Dec 19 13:04:31 2008 +0000 [scaled-font] Post-process hash value. Mix the bits within the hash value to reduce clustering. src/cairo-scaled-font.c | 15 +++++++++++++++ 1 file changed, 15 insertions(+) commit 555dd6b97ec432787d83242727164d25b459cf54 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Dec 19 13:02:00 2008 +0000 [scaled-font] Switch to a constant loop for hashing. As we only use the FNV hash for hashing matrices, expose the constant size to the compiler so that it can perform its magic. src/cairo-scaled-font.c | 18 ++++++++---------- 1 file changed, 8 insertions(+), 10 deletions(-) commit f5274f5847519208865159fa9bb254d76ba8ddac Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Dec 19 16:31:28 2008 +0000 Iterate over hash table using foreach() in destructors. Don't use the remarkably inefficient _cairo_hash_table_random_entry() to remove all entries from the hash table! src/cairo-cff-subset.c | 21 +++++++++++---------- src/cairo-ft-font.c | 47 ++++++++++++++++++++++------------------------- 2 files changed, 33 insertions(+), 35 deletions(-) commit 1c4ea84b24f3f7e480f9c50b5bd77ac703c36ef0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Dec 18 20:32:44 2008 +0000 [cairo] Early return if we attempt to set the same scaled_font If the application calls cairo_set_scaled_font() with the current scaled font, we can return early as it is a no-op. src/cairo.c | 3 +++ 1 file changed, 3 insertions(+) commit b661f3d27ba77cdf470a86f6320b1bb31d92a9f0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Dec 18 18:34:16 2008 +0000 [cairo] Embed a second gstate. Experiment with embedding a second gstate into the initial context to reduce allocations. src/cairo-private.h | 2 +- src/cairo.c | 12 ++++++------ 2 files changed, 7 insertions(+), 7 deletions(-) commit 8abd21bd3afc143501f66fdcdc0c6c4139df490b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Dec 23 15:09:15 2008 +0000 [script] Call the context creation hook Hook in the creation hook so applications can record or even modify the contexts used in replay. util/cairo-script/cairo-script-operators.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) commit 2c9323d3942cbf93b100f711000f48a3eb869271 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Dec 19 14:11:37 2008 +0000 [script] more targets util/cairo-script/csi-replay.c | 101 +++++++++++++++++++++++++++++++++++------ 1 file changed, 88 insertions(+), 13 deletions(-) commit fd96cea4fefeeee8bcccde663faa559151e2606e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Dec 19 12:54:53 2008 +0000 [script] Improve array construction. Limit the memory allocation to the initial array size and perform a direct copy from the operand stack to the array. util/cairo-script/cairo-script-objects.c | 27 +++++++++++++----- util/cairo-script/cairo-script-operators.c | 45 ++++++++++++------------------ util/cairo-script/cairo-script-private.h | 1 + util/cairo-script/cairo-script-scanner.c | 2 +- util/cairo-script/cairo-script-stack.c | 2 +- 5 files changed, 41 insertions(+), 36 deletions(-) commit ecb8dce27c769158fe4a92432a90b24a96f8afa9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Dec 18 18:32:44 2008 +0000 [script] Recreate surface content. Use the content recorded in the trace, defaulting to COLOR_ALPHA, to determine the replayed surface type. util/cairo-script/cairo-script-interpreter.h | 1 + util/cairo-script/cairo-script-operators.c | 8 ++- util/cairo-script/csi-exec.c | 1 + util/cairo-script/csi-replay.c | 80 ++++++++++++++++++++-------- util/cairo-trace/trace.c | 8 ++- 5 files changed, 74 insertions(+), 24 deletions(-) commit d21529b9ef31228182d2f61bb4f7beb6319b10c4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Dec 20 15:38:13 2008 +0000 [trace] Correct escaped characters in string literal. Laziness caused the control character to be used instead of its replacement in the escaped string! util/cairo-trace/trace.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-) commit ca39c4bf7f90646b70b2bbe9e7318fcfae99e35e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Dec 20 15:35:40 2008 +0000 [trace] Correct a couple of typos. A couple of minor typos in the escaped characters strings. util/cairo-trace/trace.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 1f2fec7388092c8fea87651f9094ca2e0be2104b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Dec 19 18:04:03 2008 +0000 [trace] Fix OBO in emit_glyphs() The largest index that will fit within 8 bits is 255 not 256! util/cairo-trace/trace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 7ebe9e68371e4e77139d9abeb7feb390203b1c8e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Dec 18 21:59:16 2008 +0000 [trace] Minor tidy. Improve scoping of variables. util/cairo-trace/trace.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) commit 931556005a1946c5e730afc7977551a2bc5be42c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Dec 18 21:57:19 2008 +0000 [trace] Remove the forced switching to C locale This interferes with the application being traced. It is not clear from printf(3) whether "%.f" is locale dependent or not - but until we have a failure do not break applications unnecessarily! util/cairo-trace/cairo-trace.in | 4 ---- 1 file changed, 4 deletions(-) commit d0f41b92bf9a9ee313dde6f451b26fffb65a0906 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Dec 18 17:26:43 2008 +0000 [trace] Do not force the output filename. Handle the case of tracing an application that spawns it own graphical children but using the autonaming facility within cairo-trace. Currently the traced process tree would all attempt to write to the same file, creating a broken trace. This means sacrificing the display of the output name, but allows use for a wider range of applications. util/cairo-trace/cairo-trace.in | 25 ++++++------------------- 1 file changed, 6 insertions(+), 19 deletions(-) commit 19e8c5c7f4a561f2454380c8f4098a832d716d28 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Dec 20 19:27:33 2008 +0000 [test] Fix mismatched free. A couple of strings allocated using glib functions but freed with system malloc. Reported: http://bugs.freedesktop.org/show_bug.cgi?id=19210. test/pdf2png.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 6662eede2af2da50bff1902e694204c69e84f036 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Dec 20 19:14:23 2008 +0000 [perf] Fix errors reported by cppcheck Trivial mistakes, identified in bugs http://bugs.freedesktop.org/show_bug.cgi?id=19206 and http://bugs.freedesktop.org/show_bug.cgi?id=19207. perf/cairo-perf-graph-files.c | 2 +- perf/cairo-perf-report.c | 7 ++++--- 2 files changed, 5 insertions(+), 4 deletions(-) commit 8419c4f124c6ffa75b44901dccab33df8bb6287a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Dec 20 15:34:41 2008 +0000 [perf] Correct another reference to '<cairo>' I think that's the last use of the old internal name for twin! perf/twin.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 14d94d9354170003ec15847873a273c1752cf9f2 Author: Behdad Esfahbod <behdad@behdad.org> Date: Fri Jan 2 02:56:38 2009 -0500 Fix various README typos HACKING | 2 +- src/README | 4 ++-- util/README | 6 ++++++ 3 files changed, 9 insertions(+), 3 deletions(-) commit 28a72648ba7abe02ebd4df7234424e333b85dc9c Author: Behdad Esfahbod <behdad@behdad.org> Date: Tue Dec 30 13:48:47 2008 -0500 [gstate] Change the glyph dropping safety margin from 2em to 10em The small margin caused bugs with math fonts. See: https://bugzilla.mozilla.org/show_bug.cgi?id=460023 src/cairo-gstate.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) commit fa63c43532de4a38d59d2746c34c815d18121062 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Dec 29 16:11:29 2008 +0000 [spline] Be pedantic and propagate errors. We know that the current users will always return SUCCESS, but propagate the status return for future users. src/cairo-path-bounds.c | 6 ++---- src/cairo-spline.c | 15 +++++++++++---- src/cairo-types-private.h | 2 +- src/cairoint.h | 2 +- 4 files changed, 15 insertions(+), 10 deletions(-) commit 7df082dd8aaa9d31479a8bb1f6f1dfe07f52019a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Dec 29 16:06:36 2008 +0000 [test] Update twin reference images Subsequent to recent tweaks, update the reference images for twin. test/Makefile.am | 6 ++---- test/twin.c | 2 +- test/twin.pdf.ref.png | Bin 1669 -> 0 bytes test/twin.ps.ref.png | Bin 0 -> 1114 bytes test/twin.ps2.ref.png | Bin 1095 -> 0 bytes test/twin.ps3.ref.png | Bin 1095 -> 0 bytes test/twin.ref.png | Bin 1712 -> 1492 bytes test/twin.svg.ref.png | Bin 0 -> 1487 bytes test/twin.svg11.argb32.ref.png | Bin 1797 -> 0 bytes test/twin.svg11.ref.png | Bin 1662 -> 0 bytes test/twin.svg11.rgb24.ref.png | Bin 1797 -> 0 bytes test/twin.svg12.argb32.ref.png | Bin 1797 -> 0 bytes test/twin.svg12.ref.png | Bin 1662 -> 0 bytes test/twin.svg12.rgb24.ref.png | Bin 1797 -> 0 bytes 14 files changed, 3 insertions(+), 5 deletions(-) commit e10af38799eb77f1a7b5f75e76c733353c7b8622 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Dec 29 12:54:13 2008 +0000 make "make check" happy again Add the missing scoping that caused check-plt to complain. src/cairo-font-face-twin.c | 6 ++++-- src/cairo-scaled-font.c | 2 ++ src/cairo.c | 4 ++++ src/cairoint.h | 8 +++++++- 4 files changed, 17 insertions(+), 3 deletions(-) commit 010085622674bd02098742f401409da8e7c1b1dc Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Dec 29 12:45:13 2008 +0000 [path] Remove tolerance from path bounders With Behdad's analytical analysis of the spline bbox, tolerance is now redundant for the path extents and the approximate bounds, so remove it from the functions parameters. src/cairo-analysis-surface.c | 3 +-- src/cairo-gstate.c | 3 +-- src/cairo-path-bounds.c | 18 ++++++------------ src/cairo-surface-fallback.c | 1 - src/cairoint.h | 5 +---- 5 files changed, 9 insertions(+), 21 deletions(-) commit 84b81388bef4a2f300580081415fe09947edb96e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Dec 29 12:16:53 2008 +0000 [test] Draw spline bbox Add the stroked extents to the spline-decomposition test. test/Makefile.am | 6 +-- test/spline-decomposition.c | 63 ++++++++++++++++++++++++++++++++ test/spline-decomposition.pdf.ref.png | Bin 20404 -> 19156 bytes test/spline-decomposition.ps.ref.png | Bin 0 -> 9090 bytes test/spline-decomposition.ps2.ref.png | Bin 8957 -> 0 bytes test/spline-decomposition.ps3.ref.png | Bin 8957 -> 0 bytes test/spline-decomposition.ref.png | Bin 20402 -> 19107 bytes test/spline-decomposition.svg.ref.png | Bin 0 -> 19156 bytes test/spline-decomposition.svg11.ref.png | Bin 20404 -> 0 bytes test/spline-decomposition.svg12.ref.png | Bin 20404 -> 0 bytes 10 files changed, 65 insertions(+), 4 deletions(-) commit efb17160904eda8fdb21fa1fd10a09081cf5cb3e Author: Behdad Esfahbod <behdad@behdad.org> Date: Sun Dec 28 16:06:27 2008 -0500 [_cairo_spline_bound] Protect against b == 0 src/cairo-spline.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) commit 0b59e29004bb19eb9bd458bbe6a48a7c367f72d0 Author: Behdad Esfahbod <behdad@behdad.org> Date: Sun Dec 28 02:49:39 2008 -0500 [_cairo_spline_bound] Simplify condition src/cairo-spline.c | 14 +++++++++----- 1 file changed, 9 insertions(+), 5 deletions(-) commit 3bf1b7d574620b1ab3c1fa50977a23d36ab04a40 Author: Behdad Esfahbod <behdad@behdad.org> Date: Sun Dec 28 02:41:39 2008 -0500 [_cairo_spline_bound] Fix the check for feasible solutions Also make it more strict. The only times we call sqrt now is when a solution in (0,1) exists. src/cairo-spline.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) commit 7f840d156c1212eeb0fddad4b4a6844a9c070bd5 Author: Behdad Esfahbod <behdad@behdad.org> Date: Sun Dec 28 02:02:30 2008 -0500 [spline] Save a couple more muls src/cairo-spline.c | 26 +++++++++++++------------- 1 file changed, 13 insertions(+), 13 deletions(-) commit 3292f9906b3637564c37863dde5d214b0fee2885 Author: Behdad Esfahbod <behdad@behdad.org> Date: Sun Dec 28 01:59:12 2008 -0500 [spline] Do some checks to avoid calling sqrt() if no feasible solution exists src/cairo-spline.c | 27 ++++++++++++++++++++------- 1 file changed, 20 insertions(+), 7 deletions(-) commit efe4d2ce99e65f9841951d3ff3e0536973bfb12c Author: Behdad Esfahbod <behdad@behdad.org> Date: Sun Dec 28 01:22:40 2008 -0500 [spline] Simplify code src/cairo-spline.c | 19 +++++++++---------- 1 file changed, 9 insertions(+), 10 deletions(-) commit 8672178bf6c7f3a38e11e224f1d484b1f0c7fd7b Author: Behdad Esfahbod <behdad@behdad.org> Date: Sun Dec 28 01:20:37 2008 -0500 [spline] Remove duplicated code by using a macro src/cairo-spline.c | 50 +++++++++++++++++++++----------------------------- 1 file changed, 21 insertions(+), 29 deletions(-) commit ef0f6c3ca311c41c9062e1298b020eae1212984e Author: Behdad Esfahbod <behdad@behdad.org> Date: Sat Dec 27 23:13:45 2008 -0500 [spline] Add an analytical bounder for splines The way this works is very simple: Once for X, and once for Y, it takes the derivative of the bezier equation, equals it to zero and solves to find the extreme points, and if the extreme points are interesting, adds them to the bounder. Not the fastest algorithm out there, but my estimate is that if _de_casteljau() ends up breaking a stroke in at least 10 pieces, then the new bounder is faster. Would be good to see some real perf data. src/cairo-path-bounds.c | 18 ++----- src/cairo-spline.c | 131 ++++++++++++++++++++++++++++++++++++++++++++++++ src/cairoint.h | 6 +++ 3 files changed, 141 insertions(+), 14 deletions(-) commit f2f62c7c4402a28a65c8171238d163af6b2fb95a Author: Behdad Esfahbod <behdad@behdad.org> Date: Fri Dec 26 20:28:54 2008 -0500 [twin] Micro-optimize src/cairo-font-face-twin.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 50bc2bc0170be2a9c84ae3064525b18190e22b48 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Dec 27 11:46:24 2008 +0000 [path] Simply track the current point for bounds. The idea is to track always update the current point, but not add it during a move-to. src/cairo-path-bounds.c | 25 +++++++++++-------------- 1 file changed, 11 insertions(+), 14 deletions(-) commit 078ebb01ba0b99ab4973ba479fe438f0674aa102 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Dec 27 11:29:15 2008 +0000 [path] Initialise spline from current point Joonas spotted that the breakage with the curve bounds was the result of initialising the spline using the original move to point and not the current point. Fixes: Bug 19256 Gnome Foot in gnome-games rendered incorrectly (https://bugs.freedesktop.org/show_bug.cgi?id=19256) src/cairo-path-bounds.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) commit 4320ea68875cc015dfecdf5ed40195e276efca07 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Dec 26 00:01:01 2008 +0000 [trace] Fix up positional arguments James Cloos pointed out that the correct form to use is "$@" when executing the command line. util/cairo-trace/cairo-trace.in | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) commit e76a676c8f3d6dc9199cf0bc0856af0c659f4a16 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Dec 25 09:51:51 2008 +0000 [twin] Trivial spelling correction. This ain't no decadent descendent of the Hershey font... src/cairo-font-face-twin.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 9332c0a8eaf266e99555df9ad7769f8b5804cbfd Author: Behdad Esfahbod <behdad@behdad.org> Date: Fri Dec 26 16:38:04 2008 -0500 [util] Add waterfall util/Makefile.am | 3 +- util/waterfall | 97 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 99 insertions(+), 1 deletion(-) commit ab1febbf22a641feb2fda4b21156b2bd45df039b Author: Behdad Esfahbod <behdad@behdad.org> Date: Fri Dec 26 16:35:33 2008 -0500 [twin] Resnap margin under monospace src/cairo-font-face-twin.c | 14 +++++++++++++- 1 file changed, 13 insertions(+), 1 deletion(-) commit 3c91d9f5a2a290653a56242eed6acae4f16e99e3 Author: Behdad Esfahbod <behdad@behdad.org> Date: Fri Dec 26 16:30:01 2008 -0500 [twin] Adjust margins src/cairo-font-face-twin.c | 114 ++++++++++++++++++++++++++++----------------- 1 file changed, 70 insertions(+), 44 deletions(-) commit 0b5c60bae1fb0924bbd43daab38932052e10b3be Author: Behdad Esfahbod <behdad@behdad.org> Date: Fri Dec 26 15:48:46 2008 -0500 [twin] Clean up hinting src/cairo-font-face-twin.c | 89 ++++++++++++++++++++++------------------------ 1 file changed, 42 insertions(+), 47 deletions(-) commit 6e7a2c4ce32ddfb0e58a74fcba8c4e46e46d32a4 Author: Behdad Esfahbod <behdad@behdad.org> Date: Fri Dec 26 15:34:50 2008 -0500 [twin] Fix monospace for narrow glyphs src/cairo-font-face-twin.c | 23 ++++++++++++++--------- 1 file changed, 14 insertions(+), 9 deletions(-) commit e8e6ae1294f11addcbc8a95db471bb48a50b0b1a Author: Behdad Esfahbod <behdad@behdad.org> Date: Fri Dec 26 15:15:15 2008 -0500 [twin] Cache pen and other properties on the scaled font src/cairo-font-face-twin.c | 170 +++++++++++++++++++++++++++------------------ 1 file changed, 101 insertions(+), 69 deletions(-) commit e133cc14469e8ebb8311bf8db206147f6b07786d Author: Behdad Esfahbod <behdad@behdad.org> Date: Fri Dec 26 14:56:32 2008 -0500 [twin] Don't stretch pen src/cairo-font-face-twin.c | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) commit dec4d791f4cc626d1fa52d68518f787210240812 Author: Behdad Esfahbod <behdad@behdad.org> Date: Fri Dec 26 14:24:08 2008 -0500 [twin] Minor shuffling src/cairo-font-face-twin.c | 142 ++++++++++++++++++++++++++------------------- 1 file changed, 81 insertions(+), 61 deletions(-) commit c3de08ee2fcfb0e092f0299e249b0f5fe56b87bf Author: Behdad Esfahbod <behdad@behdad.org> Date: Fri Dec 26 14:11:52 2008 -0500 [twin] Further reduce weight src/cairo-font-face-twin.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 5c201ef5cfdfdd4e19db7e0995b85fbb3f2e6090 Author: Behdad Esfahbod <behdad@behdad.org> Date: Fri Dec 26 00:05:19 2008 -0500 [scaled-font] Make cairo_debug_reset_static_data() work again I broke it in a5983929f8f07f384f78e0b604e22037cf5ed716 src/cairo-debug.c | 2 ++ src/cairo-scaled-font.c | 4 +--- src/cairoint.h | 3 +++ 3 files changed, 6 insertions(+), 3 deletions(-) commit 730ed68ec1ac10dcb9c7c69759f72fcaa3ea82e5 Author: Behdad Esfahbod <behdad@behdad.org> Date: Thu Dec 25 23:29:02 2008 -0500 [pattern] Fix comment typo src/cairo-pattern.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit a5f4a97aa4122d8c4b8f31053d421a52c356e270 Author: Behdad Esfahbod <behdad@behdad.org> Date: Thu Dec 25 02:48:26 2008 -0500 [twin] Don't hint pen if hint-style is SLIGHT src/cairo-font-face-twin.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 0c7eaf6e0a2c2f0db634f8120fcbc03f7e8fd751 Author: Behdad Esfahbod <behdad@behdad.org> Date: Thu Dec 25 02:44:19 2008 -0500 [twin] Improve glyph hints src/cairo-font-face-twin-data.c | 202 +++++++++++++++++++++++----------------- 1 file changed, 116 insertions(+), 86 deletions(-) commit e6205ea5a38c9de4c08b313a1beb70d4a38b0676 Author: Behdad Esfahbod <behdad@behdad.org> Date: Thu Dec 25 02:10:16 2008 -0500 [twin] Improve hinting src/cairo-font-face-twin.c | 90 +++++++++++++++++++--------------------------- 1 file changed, 36 insertions(+), 54 deletions(-) commit da8658b67510daaefd542b80e6a997bbdfd82d44 Author: Behdad Esfahbod <behdad@behdad.org> Date: Thu Dec 25 01:35:36 2008 -0500 [twin] Hint dots src/cairo-font-face-twin-data.c | 24 ++++++++++++------------ 1 file changed, 12 insertions(+), 12 deletions(-) commit f4c81e18f85c6a68d682301abfd75e7c208c1e1a Author: Behdad Esfahbod <behdad@behdad.org> Date: Thu Dec 25 01:29:41 2008 -0500 [twin] Disable pen hinting if hinting is off src/cairo-font-face-twin.c | 27 +++++++++++++++++++-------- 1 file changed, 19 insertions(+), 8 deletions(-) commit af91fc1974ce2dfbaab187e22769cdfb9e6279d9 Author: Behdad Esfahbod <behdad@behdad.org> Date: Thu Dec 25 01:24:46 2008 -0500 [twin] Optimize hinting src/cairo-font-face-twin.c | 7 +------ 1 file changed, 1 insertion(+), 6 deletions(-) commit 439959d1af1f6c4ce174a3fee524eacfc18a00bd Author: Behdad Esfahbod <behdad@behdad.org> Date: Thu Dec 25 01:06:47 2008 -0500 [twin] Implement hinting src/cairo-font-face-twin.c | 148 +++++++++++++++++++++++++++++++++++++++------ 1 file changed, 129 insertions(+), 19 deletions(-) commit 9f9f5317dca6a9988a0a62aaa60393570157af63 Author: Behdad Esfahbod <behdad@behdad.org> Date: Wed Dec 24 22:29:59 2008 -0500 [util] Fix pangram util/cairo-view | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 8296bc74c8687a2f52e06174e9dc70d32d1181c8 Author: Behdad Esfahbod <behdad@behdad.org> Date: Wed Dec 24 18:00:04 2008 -0500 [twin] Reduce the weight just a bit, such that bold doesn't look as bad src/cairo-font-face-twin.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit b2f89625453e2a15da175ea5b4ca5af6c2201d96 Author: Behdad Esfahbod <behdad@behdad.org> Date: Wed Dec 24 17:29:21 2008 -0500 [twin] Adjust baseline src/cairo-font-face-twin.c | 11 ++++------- 1 file changed, 4 insertions(+), 7 deletions(-) commit f980d017d2360634f391eb1129317446bfe42cc9 Author: Behdad Esfahbod <behdad@behdad.org> Date: Wed Dec 24 17:00:43 2008 -0500 [twin] Remove serif setting I don't think I'm going to implement serif-drawing soon, so, remove the infrastructure. Can always be added back later. src/cairo-font-face-twin.c | 21 ++------------------- 1 file changed, 2 insertions(+), 19 deletions(-) commit 1116febb40cb5e5f7eafe97999d1143347819b3f Author: Behdad Esfahbod <behdad@behdad.org> Date: Wed Dec 24 16:32:25 2008 -0500 [twin] Implement small-caps src/cairo-font-face-twin.c | 27 ++++++++++++++++----------- 1 file changed, 16 insertions(+), 11 deletions(-) commit 6767673961401c7ab7b92eb8dfbb345efb560741 Author: Behdad Esfahbod <behdad@behdad.org> Date: Wed Dec 24 16:13:12 2008 -0500 [twin] Implement stretch src/cairo-font-face-twin.c | 27 +++++++++++++++++++++------ 1 file changed, 21 insertions(+), 6 deletions(-) commit 0656e947f11356877014df93bd083123c7313dc3 Author: Behdad Esfahbod <behdad@behdad.org> Date: Wed Dec 24 15:43:17 2008 -0500 [twin] Implement slant src/cairo-font-face-twin.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) commit 9d493fcd16fb770f21fa732b08834f3a435fe1f2 Author: Behdad Esfahbod <behdad@behdad.org> Date: Wed Dec 24 15:36:57 2008 -0500 [twin] Disable the serif mode src/cairo-font-face-twin.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) commit 6c1d21bf8947f5b11702626ddfd9bac18073188d Author: Behdad Esfahbod <behdad@behdad.org> Date: Wed Dec 24 03:47:21 2008 -0500 [twin] Implement monospace src/cairo-font-face-twin.c | 30 +++++++++++++++++++++++------- 1 file changed, 23 insertions(+), 7 deletions(-) commit 5def8c587b2b897bfebc6d77a6a3e13355cae2f2 Author: Behdad Esfahbod <behdad@behdad.org> Date: Wed Dec 24 03:45:06 2008 -0500 [twin] Fix dots src/cairo-font-face-twin-data.c | 54 +++++++++++++++++++++++++---------------- 1 file changed, 33 insertions(+), 21 deletions(-) commit a303bbaea332df20b81f068366d701d66913c043 Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Tue Dec 23 16:12:56 2008 -0500 [font-face-get-type] Fix test to cleanup the surface and context properly Destroy the context and the surface even when the test fails. I ran into this because cairo_debug_reset_static_data() is called between the tests on the quartz backend (it doesn't fork) and isn't on other backends? This is perhaps a mistake... test/font-face-get-type.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) commit 52cc31a7fa6570e538762631359a92e7b717710e Author: Behdad Esfahbod <behdad@behdad.org> Date: Tue Dec 23 13:48:52 2008 -0500 [util] Add cairo-view util/Makefile.am | 3 +- util/cairo-view | 113 +++++++++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 115 insertions(+), 1 deletion(-) commit 418c7ef133840c10eb0be7e75c7dd11533fbe897 Author: Behdad Esfahbod <behdad@behdad.org> Date: Tue Dec 23 02:39:17 2008 -0500 [twin] Adjust font extents src/cairo-font-face-twin.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) commit 96e41c28e980f313c51a0eed81681e7e48752797 Author: Behdad Esfahbod <behdad@behdad.org> Date: Sat Dec 20 23:30:59 2008 -0500 [twin] Minor cleanup src/cairo-font-face-twin.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) commit 556f6ce364d81f9309162d21deb86a3d5b8b9a6e Author: Behdad Esfahbod <behdad@behdad.org> Date: Sat Dec 20 23:16:53 2008 -0500 [twin] close_path the Q src/cairo-font-face-twin-data.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 6929ed800d19f359af7436636e8adc6203083a10 Author: Behdad Esfahbod <behdad@behdad.org> Date: Sat Dec 20 20:25:01 2008 -0500 [twin] Fix serif option parsing. Also make serif and weight do something src/cairo-font-face-twin.c | 20 ++++++++++++++++---- 1 file changed, 16 insertions(+), 4 deletions(-) commit d423339fcef3c41225196c4371fb2c133ca61144 Author: Behdad Esfahbod <behdad@behdad.org> Date: Sat Dec 20 19:39:58 2008 -0500 [twin] Add face properties Just parse them now. We don't use them yet. src/cairo-font-face-twin.c | 191 +++++++++++++++++++++++++++++++++++++++++++-- src/cairo-user-font.c | 4 +- src/cairoint.h | 6 +- 3 files changed, 189 insertions(+), 12 deletions(-) commit 93672d842f2b2c840bb9e73901faa4ce5df9c792 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Dec 23 15:00:47 2008 +0000 [test] Fix reference handling in user-font-rescale user-font-rescale stored the current font on the context in order to create a rescaling proxy font. As we failed to take a reference to the font, it caught us by surprise when the font disappeared as we modified the context before creating our proxy. Ho hum. test/user-font-rescale.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) commit 5ad65dc4be5964265946de59a951f2304d64159e Author: Adrian Johnson <ajohnson@redneon.com> Date: Tue Dec 23 14:56:48 2008 +1030 Document PDF restrict_to_version API doc/public/cairo-sections.txt | 4 ++++ doc/public/tmpl/cairo-pdf.sgml | 35 +++++++++++++++++++++++++++++++++++ src/cairo-pdf.h | 2 ++ 3 files changed, 41 insertions(+) commit 1869e0240bcc93c93e8abd1c8f930cf08ca8ee24 Author: Adrian Johnson <ajohnson@redneon.com> Date: Tue Dec 23 14:37:30 2008 +1030 Document cairo_surface_(set|get)_mime_data() and mime types doc/public/cairo-sections.txt | 5 ++++ doc/public/tmpl/cairo-surface.sgml | 50 ++++++++++++++++++++++++++++++++++++-- 2 files changed, 53 insertions(+), 2 deletions(-) commit 1f894033f077731485e1228f7e071e75c37a9947 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Tue Dec 23 03:14:38 2008 +0200 [spans] Close open subpaths when filling with a scan converter. As reported by Christian Persch, open subpaths weren't being closed when rendering to an image surface: http://bugs.freedesktop.org/show_bug.cgi?id=19240 src/cairo-spans.c | 53 +++++++++++++++++++++++++++++++---------------------- 1 file changed, 31 insertions(+), 22 deletions(-) commit 0aa34c6435eaa260292cf10d270ebbf3314c7924 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Tue Dec 23 02:18:14 2008 +0200 [svg] Fix build when building only the svg vector surface. The SVG vector surface pulls in font subsetting and that in turns uses the PDF operators in cairo-type3-glyph-surface.c. src/cairoint.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 38ec6e302cdd703447f169d95121d434c7495501 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Tue Dec 23 02:05:32 2008 +0200 [test] Fix any2ppm build when building without the full complement of surfaces. It was complaining about g_init_type () being used without the proper includes which would have been pulled in by the svg or poppler includes. test/any2ppm.c | 2 ++ 1 file changed, 2 insertions(+) commit 6ce97907edfa8a7b09abfc090a9c03d235de2ada Author: Adrian Johnson <ajohnson@redneon.com> Date: Mon Dec 22 11:09:27 2008 +1030 win32-font: truncate instead of fail if toy font name too long http://lists.cairographics.org/archives/cairo/2008-December/016109.html src/cairo-win32-font.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) commit d99583bae680fea852e957afdf674e2031aa7928 Author: Adrian Johnson <ajohnson@redneon.com> Date: Sun Dec 21 17:03:17 2008 +1030 Fix win32 font breakage as a result of the toy font face changes in f7ab65e9b0c0f08be8d294e90131baa2decf1f1d src/cairo-win32-font.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) commit a5983929f8f07f384f78e0b604e22037cf5ed716 Author: Behdad Esfahbod <behdad@behdad.org> Date: Sat Dec 20 01:18:12 2008 -0500 [toy-font-face] Move toy font face code in a file of its own src/Makefile.sources | 1 + src/cairo-debug.c | 2 +- src/cairo-font-face.c | 449 ------------------------------------- src/cairo-mutex-list-private.h | 2 +- src/cairo-os2-surface.c | 9 +- src/cairo-scaled-font.c | 4 +- src/cairo-toy-font-face.c | 489 +++++++++++++++++++++++++++++++++++++++++ src/cairoint.h | 5 +- 8 files changed, 497 insertions(+), 464 deletions(-) commit 8bf49b27aeea8179b5862a424fb3ef1934be6e0e Author: Behdad Esfahbod <behdad@behdad.org> Date: Fri Dec 19 20:46:36 2008 -0500 [font-face] Use cairo_font_face_t instead of toy, for nil objects And with a NULL backend. We are not supposed to read the nil objects afterall. If this causes crashes, the crash site should be fixed. src/cairo-font-face.c | 55 ++++++++++----------------------------------------- src/cairoint.h | 2 +- 2 files changed, 11 insertions(+), 46 deletions(-) commit e3675f8ac0acb285b2d669c0a44322623dac5fe9 Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Fri Dec 19 08:50:26 2008 -0500 Fix definition of _cairo_quartz_font_face_backend Removes the NULL so that _cairo_quartz_font_face_scaled_font_create is in the correct location. src/cairo-quartz-font.c | 1 - 1 file changed, 1 deletion(-) commit fa7d96e1236cdfa0016f96e5c7b6bd209c405cba Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Fri Dec 19 08:47:07 2008 -0500 Fix compilation of quartz surface. Declares _cairo_quartz_scaled_font_backend ahead of time and makes it static. Also, removes the 'static' from the _cairo_quartz_font_backend definition. src/cairo-quartz-font.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) commit f8886ad1b30ec9c29ec5ac3e9cd5519dbc9e7b65 Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Thu Dec 18 20:50:20 2008 -0500 [scaled-font-zero-matrix] Destrory scaled font and font options Oops. test/scaled-font-zero-matrix.c | 2 ++ 1 file changed, 2 insertions(+) commit 4567692326daf592ab432670c6d33b49f940dd35 Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Thu Dec 18 18:03:40 2008 -0500 Add test creating a scaled font with a zero ctm matrix test/Makefile.am | 1 + test/scaled-font-zero-matrix.c | 59 ++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 60 insertions(+) commit 18054ef00c69f62804e08734fd2c3286373b451f Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Thu Dec 18 18:01:52 2008 -0500 [test] Quartz doesn't like being forked When the cairo-test-suite forks CoreFoundation complains with: "The process has forked and you cannot use this CoreFoundation functionality safely. You MUST exec(). When forked so avoid doing it in the test suite for now. In the future we should investigate the possibility of a work around. test/cairo-test-runner.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) commit b4e0d489bb0317fdd935ad2d29830e48b5ce4f6c Merge: 813cbf13d f60da9a37 Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Thu Dec 18 17:25:17 2008 -0500 Merge branch '1.8' Conflicts: NEWS build/Makefile.am.changelog cairo-version.h src/cairo-gstate.c commit f60da9a379890a3a4cd74cdad48e73c22cb74160 Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Thu Dec 18 17:07:25 2008 -0500 Fix _compute_transform to check for nearly degenerate matrices If a matrix was something like [0 .000001 0, .000001 0 0] the old code would assume that xx and yy were greater than 0 and turn the nearly degenerate matrix into an actual degenerate one. This caused things to blow up later on. Now we check that our nearly rectangular matrices are not nearly degenerate, and let the nearly degenerate ones fall through to the non-rectangular path. Note: I'm not sure why NEARLY_ZERO(d) is fabs(d) < 1/65536 instead of some other value. Hopefully, it's a useful definition. This problem was found by a test case attached to: https://bugzilla.mozilla.org/show_bug.cgi?id=467423 src/cairo-win32-font.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) commit 813cbf13ddbd3d4b708b3b362dd6c108966f44d5 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Dec 18 14:52:03 2008 +0000 [path] Separate the approx. bounds into 3 distinct functions Based on feedback from Jeff Muizelaar, there is a case for a very quick and dirty extents approximation based solely on the curve control points (for example when computing the clip intersect rectangle of a path) and by moving the stroke extension into a core function we can clean up the interface for all users, and centralise the logic of approximating the stroke extents. src/cairo-analysis-surface.c | 32 ++++++-------- src/cairo-clip.c | 14 +++--- src/cairo-path-bounds.c | 101 ++++++++++++++++++++++++++++++++++++++++--- src/cairo-rectangle.c | 10 ----- src/cairo-surface-fallback.c | 10 ++--- src/cairoint.h | 19 +++++--- 6 files changed, 134 insertions(+), 52 deletions(-) commit dea40e61babe608315b6d365094cf518814b134e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Dec 18 11:50:00 2008 +0000 [path] Return the fixed-point bounds of the path When analysing the stroke extents, we need the original fixed-point extents so that we do not incur an OBO when we round-to-integer a second time. We also need a more accurate estimate than simply using the control points of the curve, so pass in tolerance and decompose until someone discovers a cheaper algorithm to determine the precise aligned bounding box of a bezier curve. src/cairo-analysis-surface.c | 30 +++++++------- src/cairo-clip.c | 25 +++++++----- src/cairo-path-bounds.c | 95 ++++++++++++++++++++++---------------------- src/cairo-rectangle.c | 10 +++++ src/cairo-surface-fallback.c | 9 +++-- src/cairoint.h | 9 ++++- 6 files changed, 98 insertions(+), 80 deletions(-) commit 3a53e0261b1b5af21bf37e2a211eefd501bd5358 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Dec 18 11:10:17 2008 +0000 [test] Update twin reference images. Closing the 'o' in twin has removed a slight artefact - update the reference images to match. test/twin.pdf.ref.png | Bin 1673 -> 1669 bytes test/twin.ref.png | Bin 1718 -> 1712 bytes 2 files changed, 0 insertions(+), 0 deletions(-) commit c2478de26cc1781b8f8604191a9f75ad29eac746 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Dec 18 10:50:45 2008 +0000 [twin] Switch internal users to the new name. s/<cairo>/@cairo:/ doc/tutorial/src/twin.c | 2 +- test/twin.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) commit 3424b5f8c8cb1f2ef16b9e8429e64f720106f4d8 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Dec 18 10:47:44 2008 +0000 [scaled-font] Make check-doc happy Replace a structure marker '#' with a macro marker '%'. src/cairo-scaled-font.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 449bf0bb7bff38da1c5d0286cb4f22c858bcc3a9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Dec 18 10:45:22 2008 +0000 [script] Silence incorrect compiler warnings Bah, the compiler failed to do its IPA correctly and wrongly complained about potential use of uninitialised out-params guarded by the return value. Silence them! util/cairo-script/cairo-script-operators.c | 38 +++++++++++++++--------------- 1 file changed, 19 insertions(+), 19 deletions(-) commit ce0b136a441382b25d0932cd66a906a30a693365 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Dec 17 20:34:45 2008 +0000 Query the backend to see if we can repaint the solid pattern. If we are dithering on the Xlib backend we can not simply repaint the surface used for a solid pattern and must recreate it from scratch. However, for ordinary XRender usage we do not want to have to pay that price - so query the backend to see if we can reuse the surface. src/cairo-analysis-surface.c | 2 ++ src/cairo-meta-surface.c | 1 + src/cairo-paginated-surface.c | 1 + src/cairo-pdf-surface.c | 1 + src/cairo-script-surface.c | 1 + src/cairo-surface.c | 41 ++++++++++++++++++++++++----------------- src/cairo-xlib-surface.c | 12 +++++++++++- src/cairoint.h | 5 +++++ src/test-meta-surface.c | 1 + src/test-paginated-surface.c | 1 + 10 files changed, 48 insertions(+), 18 deletions(-) commit 0df43251d4b4641d16c0a2e2c49d18dd2a9c832e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Dec 17 23:45:39 2008 +0000 [replay] Take advantage of unresolved fonts As Behdad has changed the underlying ft-font to accept unresolved patterns, take advantage of that to simplify the replay code. util/cairo-script/cairo-script-operators.c | 16 ++-------------- 1 file changed, 2 insertions(+), 14 deletions(-) commit 97a00bdd4eefade8d0808b5572bcdbba26c9c8f3 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Dec 17 17:32:07 2008 +0000 [cairoint.h] Wrap macro in parenthesis Ensure that the stride calculation macro is not influenced by surrounding precedence issues by enclosing it in a pair of parenthesis. src/cairoint.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit aa65e756d496f4c3a8b43d2100e17dc39c2c26b7 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Dec 5 11:21:11 2008 +0000 [script] Avoid creating 0x0 windows XCreateWindow dies if asked to create a 0x0 window, so don't and use a 1x1 pixel window for a zero-sized surface. util/cairo-script/csi-replay.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) commit 6458903c95c2bba3f1b9ceaaafb6979d180ab039 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Dec 17 10:07:32 2008 +0000 [image] Eliminate the short-lived context used for coercing. When coercing from one image format to another we performed a paint operation using a temporary context - this is overkill as we can just call _cairo_surface_paint() directly. src/cairo-image-surface.c | 42 +++++++++++++++++++++++------------------ src/cairo-scaled-font.c | 40 ++++++++++++++++++++------------------- src/cairo-svg-surface.c | 18 ++++++++---------- src/cairo-type3-glyph-surface.c | 20 ++++---------------- src/cairoint.h | 4 ++-- 5 files changed, 59 insertions(+), 65 deletions(-) commit 9b46d13b6cd58c711010df89a41b216c5cc8c881 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Dec 17 23:52:32 2008 +0000 [trace] Runtime version check. Check that the application is using cairo >= 1.9 before attempting to query the mime-data. util/cairo-trace/trace.c | 47 +++++++++++++++++++++++++---------------------- 1 file changed, 25 insertions(+), 22 deletions(-) commit 2d790daa957471670f4ae0d3b22da89e4ee7111f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Dec 18 00:47:08 2008 +0000 [pattern] Use a solid pattern for a uniform gradient. If each color stop in a gradient is identical, replace the gradient surface with a simple solid surface. As seen in the wild. src/cairo-pattern.c | 54 ++++++++++++++++++++++++++++++++++++++--------------- 1 file changed, 39 insertions(+), 15 deletions(-) commit 25a4677200ea6689be0dca9d533842dc7da54837 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Dec 17 09:29:35 2008 +0000 [analysis] Use approximate extents. Use the approximate path based extents to avoid tessellation. src/cairo-analysis-surface.c | 54 ++++++++++++++------------------------------ 1 file changed, 17 insertions(+), 37 deletions(-) commit 18bca91411e9ba6e1e58dbc945a0567d2661568f Author: Behdad Esfahbod <behdad@behdad.org> Date: Thu Dec 18 02:12:14 2008 -0500 [twin] close_path the 'o' src/cairo-font-face-twin-data.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 5ee6aad471d460f960a5e7a5a8e35aca0f0af60e Author: Behdad Esfahbod <behdad@behdad.org> Date: Wed Dec 17 18:05:51 2008 -0500 [toy] Use twin font if font backend returns UNSUPPORTED src/cairo-font-face.c | 19 ++++++++----------- 1 file changed, 8 insertions(+), 11 deletions(-) commit eb069094ea4471b3e63881fe0f8b958ad8924c7d Author: Behdad Esfahbod <behdad@behdad.org> Date: Wed Dec 17 17:30:34 2008 -0500 Treat any toy family starting with "@cairo:" as request for twin src/cairo-font-face.c | 3 ++- src/cairoint.h | 2 +- 2 files changed, 3 insertions(+), 2 deletions(-) commit 32c66109059398e88f50335bef75fa8c0e51c4c8 Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Wed Dec 17 17:43:37 2008 -0500 Call _cairo_error when propagating error status from the font_face. src/cairo-gstate.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 1d72e53c3e99f93aa59967c76f1b1b4c035b9130 Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Wed Dec 17 17:42:18 2008 -0500 Add a missing _cairo_error() to a bunch of status returns. src/cairo-misc.c | 4 ++-- src/cairo-scaled-font.c | 14 +++++++------- src/cairo.c | 2 +- 3 files changed, 10 insertions(+), 10 deletions(-) commit af3a892c3ef6a7c1d28d523944bbb67e59edc1de Author: Behdad Esfahbod <behdad@behdad.org> Date: Wed Dec 17 16:45:28 2008 -0500 [ft] Remove stale comment src/cairo-ft-font.c | 4 ---- 1 file changed, 4 deletions(-) commit 0137b9bd320783264d865a397392b0ee14fd69b3 Author: Karl Tomlinson <karlt+@karlt.net> Date: Wed Dec 17 16:40:12 2008 -0500 [ft] Don't call FT_Done_Face() on faces we did not create src/cairo-ft-font.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit e4d7c87b5e449a743a68dd256faaa3242f191a44 Author: Karl Tomlinson <karlt+@karlt.net> Date: Wed Dec 17 16:40:12 2008 -0500 [ft] Don't call FT_Done_Face() on faces we did not create src/cairo-ft-font.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 6778a5f67ae2e35e809cf4464530e2bb05870378 Author: Behdad Esfahbod <behdad@behdad.org> Date: Wed Dec 17 16:19:45 2008 -0500 [.gitignore] Update build/.gitignore | 2 ++ src/.gitignore | 1 + 2 files changed, 3 insertions(+) commit 26f471999973c538e45db0ae9f0227fbe5dedf24 Author: Behdad Esfahbod <behdad@behdad.org> Date: Wed Dec 17 16:08:32 2008 -0500 [scaled-font] Improve docs src/cairo-scaled-font.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) commit f7ab65e9b0c0f08be8d294e90131baa2decf1f1d Author: Behdad Esfahbod <behdad@behdad.org> Date: Wed Dec 17 06:25:50 2008 -0500 Clean up toy font face handling This commit moves the toy-to-real mapping from the scaled font creation time to font face creation. A toy font face will keep an internal ref to an implementation face. Then cairo_scaled_font_create() will simply substitute the implementation face before creating anything. This also modifies the cairo-ft toy creation in that we now create a non-resolved pattern and store it in a cairo-ft font-face. We then do the resolving and unscaled font creation at scaled-font creation time. This also means that cairo_ft_font_face_create_for_pattern() now accepts non-resolved patterns too, and does the right thing about them. As much as that can be called right. Some testing of toy font creation performance is in order, as is testing win32 and quartz font backends. src/cairo-font-face.c | 143 +++++++++---------- src/cairo-ft-font.c | 371 +++++++++++++++++++++++++++++------------------- src/cairo-quartz-font.c | 208 ++++++++++----------------- src/cairo-scaled-font.c | 19 ++- src/cairo-user-font.c | 62 +------- src/cairo-win32-font.c | 20 +-- src/cairoint.h | 40 +++--- 7 files changed, 412 insertions(+), 451 deletions(-) commit 43edb4dd7b8d0614a6c2be15dfa72f980dd55f1d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Dec 12 18:10:58 2008 +0000 Fix compilation with gcov We need to add --coverage to LDFLAGS, so create CAIRO_LDFLAGS and use that to populate AM_LDFLAGS. boilerplate/Makefile.am | 1 + build/configure.ac.analysis | 6 +++++- build/configure.ac.features | 1 + src/Makefile.am | 3 ++- test/Makefile.am | 17 +++++++++-------- 5 files changed, 18 insertions(+), 10 deletions(-) commit bcea3151d66495f03d67ce71274203d0515caf4c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Dec 12 14:08:11 2008 +0000 [test] Add missing joins.ref.png Missed this file when adding the reference images for joins. test/joins.ref.png | Bin 0 -> 5879 bytes 1 file changed, 0 insertions(+), 0 deletions(-) commit 8cec548854d86dac8f0c99e99461421c8ad653b1 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Dec 14 14:06:07 2008 +0000 [NEWS] Add notes for 1.8.6 Add the historical blurb for 1.8.6. NEWS | 49 +++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 49 insertions(+) commit 0a1d194ad85398c8711268fd9ba679f8609206a1 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Dec 14 14:53:27 2008 +0000 [perf] Fix for git-1.6 git-1.6 moved all the subcommands out of the PATH so update our usage. perf/cairo-perf-graph | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 1c2eba92c16207a48f9c8f52065fbf9a65d16fcd Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Dec 14 13:05:01 2008 +0000 [RELEASING] Update Makefile for git-1.6 Git moved all the subcommands out of the PATH, so we need to update our usage. build/Makefile.am.changelog | 4 ++-- build/Makefile.am.releasing | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) commit 2fa6b6353740b7941e90e50bf6b0eb5f06aa11b5 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Dec 14 11:40:12 2008 +0000 [RELEASING] Correct a few minor typos. As I was reading through the instructions, take the opportunity to fix a few spelling mistakes. RELEASING | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) commit 1950abd14c28d58913c085d3829eb99a078cf2d1 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Dec 14 11:26:56 2008 +0000 [doc] Fix erroneous doc comment. A normal comment was marked as a gtk-doc entry and thus causing gtk-doc to complain. src/cairo-path-fixed.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 9485cd947f982ecd1e4f8d948786af3265d6eb74 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Dec 14 16:35:14 2008 +0000 Increment version to 1.8.7 after the 1.8.6 release. cairo-version.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit e2a2eddcfb8fc73f3decdb91c00e8e6d5282e09c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Dec 14 14:07:11 2008 +0000 Increment version to 1.8.6 In preparation for the next stable release. cairo-version.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit ef7c42a7d390064b2a5b08e14d5f65b8ea4628e3 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Dec 14 14:06:07 2008 +0000 [NEWS] Add notes for 1.8.6 Add the historical blurb for 1.8.6. NEWS | 50 ++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 50 insertions(+) commit ce91e6ac360e1dbe57a4e729b68d84499603e9ae Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Dec 14 14:53:27 2008 +0000 [perf] Fix for git-1.6 git-1.6 moved all the subcommands out of the PATH so update our usage. perf/cairo-perf-graph | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit f8372509aff662e9cbabeb99560b01019395c15e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Dec 14 13:05:01 2008 +0000 [RELEASING] Update Makefile for git-1.6 Git moved all the subcommands out of the PATH, so we need to update our usage. build/Makefile.am.changelog | 4 ++-- build/Makefile.am.releasing | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) commit 9faa9a214994fe27a0cb27a298881f23c2050bbe Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Dec 14 11:40:12 2008 +0000 [RELEASING] Correct a few minor typos. As I was reading through the instructions, take the opportunity to fix a few spelling mistakes. RELEASING | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) commit b1be4196513263bdbf58944ea6ac87708a399c5f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Dec 14 11:26:56 2008 +0000 [doc] Fix erroneous doc comment. A normal comment was marked as a gtk-doc entry and thus causing gtk-doc to complain. src/cairo-path-fixed.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 79bd32a6ccb1a882ddd2b1585ab155f568bf5ec7 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Dec 6 13:32:37 2008 +0200 [perf-diff] Fix cairo-perf-diff for git 1.6 Since git 1.6 the plumbing commands aren't installed in the user's path by default. This patch fixes cairo-perf-diff to find the git-sh-setup command from git's lib dir. (cherry picked from commit 0c0f4862c52d68776024f125b003ade455044b27) perf/cairo-perf-diff | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 97b5240beeb9206f4dbda6ffce33b51aa16eec2f Author: Julien Danjou <julien@danjou.info> Date: Tue Nov 18 10:01:49 2008 +0100 [xcb] check for render extension presence Otherwise this may leads to an invalid memory access to r. Fixes: Bug 18588 - XCB backend fails with missing render. https://bugs.freedesktop.org/show_bug.cgi?id=18588 Signed-off-by: Julien Danjou <julien@danjou.info> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> (cherry picked from commit 834f1d7b7097dcc3a32f6c65d21e87fd272d924a) AUTHORS | 1 + src/cairo-xcb-surface.c | 8 ++++++-- 2 files changed, 7 insertions(+), 2 deletions(-) commit 6fed98c0d3b0708ad5171654179a482fa7d88805 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Dec 5 21:14:45 2008 +0000 [gstate] Remove culled glyphs from clusters. Sascha Steinbiss reported a bug where the PDF backend was reading beyond the end of the glyph array: http://lists.cairographics.org/archives/cairo/2008-December/015976.html. It transpires that in the early glyph culling in the gstate we were not updating the clusters to skip culled glyphs. src/cairo-gstate.c | 223 ++++++++++++++++++++++++++++++++++++++++++----------- 1 file changed, 176 insertions(+), 47 deletions(-) commit 2b7c6f361a3cfe309ff0bcb606b808acbf36aa0f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Nov 26 12:33:12 2008 +0000 [skiplist] Allocate elements in chunks. Use a pool allocator to preallocate a chunk from which to allocate the skiplist elements (if we failed to reallocate from the freelists). src/cairo-bentley-ottmann.c | 33 ++++++++++++----- src/cairo-skiplist-private.h | 3 +- src/cairo-skiplist.c | 85 +++++++++++++++++++++++++++++++++++--------- 3 files changed, 94 insertions(+), 27 deletions(-) commit 903b39c30448d62e2cbf9d075c5256a333bd5d8f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Dec 4 15:22:01 2008 +0000 [test] Make the xlib-fallback use the image refs. The xlib-fallback is closer to the image than the xlib backend, so prefer not to use the xlib.ref.png. boilerplate/cairo-boilerplate.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 792057539bf814cc00447a0a53978e0af3efe270 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Dec 3 10:52:59 2008 +0000 [test] Only delete output images beneath output. We were using an overly-liberal find that also deleted copied output for use in CAIRO_REF_DIR if that directory was below test/. So only delete files below output/ (which should only be used by cairo-test). test/Makefile.am | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) commit 913cbad25e08a07c05b8c2e6ddd3c343ca2462b2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Dec 2 13:06:50 2008 +0000 [test] Add a simple joins test case Exercise joins between short (<LINE_WIDTH) lines - used in debugging stroke-to-path. test/Makefile.am | 3 ++ test/joins.c | 109 ++++++++++++++++++++++++++++++++++++++++++++++++++ test/joins.ps.ref.png | Bin 0 -> 5496 bytes 3 files changed, 112 insertions(+) commit 9f4f41de7d20ee46ee8ca06716cbc18e55cfcf86 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Dec 2 13:05:30 2008 +0000 [tutorial] Correct twin font name We changed the name for the builtin font from "cairo" to "<cairo>" to reduce possible naming conflicts - update the tutorial to match. doc/tutorial/src/twin.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 4ff884fe4df64234a7da4635d78ffafabe00f18c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Nov 30 19:18:15 2008 +0000 [test] Check idempotency of append_path() -> copy_path() The API should preserve the precision across the public interface so that the user is able to retrieve the co-ordinates that he used to construct the path. However since we transform the path to a 24.8 fixed-point internal represent we currently incur a precision-loss - the affects of which can be seen in the miter-precision test case for example. It is planned to move to keeping the path as doubles until the backend explicitly requests the fixed-point coodinates (and some backends, e.g. pdf, might only ever use the doubles). Then, barring rounding errors during path transformations, we should be able to return the exact path the user set (under an identity CTM, of course ;-). test/Makefile.am | 1 + test/path-precision.c | 108 ++++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 109 insertions(+) commit fe4af195a7880336894a5fbae86740ef55c14cbf Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Nov 30 13:48:11 2008 +0000 [test] Add a rectilinear-dash test case. Exercise dashing on pixel-aligned boundaries to test extending the rectilinear stroker to handle dashes. test/Makefile.am | 2 + test/rectilinear-dash.c | 176 ++++++++++++++++++++++++++++++++++++++++++ test/rectilinear-dash.ref.png | Bin 0 -> 291 bytes 3 files changed, 178 insertions(+) commit 1d68ee73f8d406671d25a1ab6c3cfb096a7a6ce7 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Nov 23 08:59:56 2008 +0000 [pattern] Compute the combined color content. When multiplying two colors together, the combined content is simply the or of the their contents. src/cairo-pattern.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit 30cef3116ed960a94323477154896a03c4fb30bb Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Dec 11 01:45:31 2008 +0000 [spans] Add a sentinel cell to the cell_list. By terminating the list with a cell for INT_MAX, we remove a few conditionals from the hot-paths. src/cairo-tor-scan-converter.c | 25 +++++++++++++++---------- 1 file changed, 15 insertions(+), 10 deletions(-) commit 3b2bb95249afdc3a705d0ce5205019e7d9d31020 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Dec 11 01:22:37 2008 +0000 [spans] Embed 64 edge buckets into polygon Frequently we only need a few y-buckets, so embed 64 into the parent structure. src/cairo-tor-scan-converter.c | 40 ++++++++++++---------------------------- 1 file changed, 12 insertions(+), 28 deletions(-) commit 70235df72e3f33482527090d7039d8cab7ef6ff9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Dec 11 00:32:13 2008 +0000 [spans] Pool capacity is in bytes not elements. Set the polygon capacity to a shade under 8k, not 200 bytes, reducing the number of malloc calls made by the scan convertor by a factor of 40! src/cairo-tor-scan-converter.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 6f284587a4ef829a1666fe730e590b59e4eedd5e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Dec 10 13:50:23 2008 +0000 [script] Add examples Add some simple examples to demonstrate CairoScript, courtesy of M Joonas Philaja. configure.ac | 1 + util/cairo-script/Makefile.am | 2 + util/cairo-script/examples/Makefile.am | 10 + util/cairo-script/examples/dragon.cs | 48 + util/cairo-script/examples/hilbert.cs | 51 + util/cairo-script/examples/infinichess.cs | 29 + util/cairo-script/examples/interference.cs | 46 + util/cairo-script/examples/pythagoras-tree.cs | 53 + util/cairo-script/examples/sierpinski.cs | 37 + .../examples/wedgeAnnulus_crop_ybRings.cs | 30 + util/cairo-script/examples/world-map.cs | 209 +++ util/cairo-script/examples/zrusin.cs | 1333 ++++++++++++++++++++ 12 files changed, 1849 insertions(+) commit 463dc2891312e3658875d6964a91e7e875de4b38 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Dec 10 21:09:14 2008 +0000 [script] Allocate the interned strings from a permanent pool Reduce memory fragmentation by allocating the permanent strings from a single pool. util/cairo-script/cairo-script-interpreter.c | 52 ++++++++++++++++++++++------ util/cairo-script/cairo-script-private.h | 4 +++ 2 files changed, 45 insertions(+), 11 deletions(-) commit adc9c90d79cf4bf5e3789ceadab6a49da746274a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Dec 10 21:00:11 2008 +0000 [script] Freed object cache. Cache the last freed object to reduce malloc pressure. util/cairo-script/cairo-script-hash.c | 3 - util/cairo-script/cairo-script-interpreter.c | 16 ++-- util/cairo-script/cairo-script-objects.c | 118 ++++++++++++++++++--------- util/cairo-script/cairo-script-private.h | 5 +- 4 files changed, 91 insertions(+), 51 deletions(-) commit 1042909796295234d8aadf674824d306d7bcc9b5 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Dec 10 17:56:19 2008 +0000 [script] Replay multiple files. Iterate over the argument vector and replay each file on the command line. util/cairo-script/csi-replay.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) commit c5c04528b51b9025f48fb0f3b1c4a8dc98acb214 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Dec 10 17:30:21 2008 +0000 [script] Slab allocator Allocate small objects from a pool and maintain a per-size free-list. util/cairo-script/cairo-script-interpreter.c | 65 +++++++++++++++++++++++++++- util/cairo-script/cairo-script-private.h | 22 ++++++++++ 2 files changed, 85 insertions(+), 2 deletions(-) commit 83f0e6cf6245c0b3fd7a9b572e5dcea9b633a644 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Dec 10 16:46:05 2008 +0000 [script] Inline the stack push Frequently to push an object onto the stack all we need is to simply perform the struct copy - so inline it and only call the out-of-line function if we need to enlarge the stack. util/cairo-script/cairo-script-interpreter.c | 20 ++++++------- util/cairo-script/cairo-script-operators.c | 43 +++++++--------------------- util/cairo-script/cairo-script-private.h | 14 ++++++++- util/cairo-script/cairo-script-stack.c | 13 ++++----- 4 files changed, 39 insertions(+), 51 deletions(-) commit 565b3d6ad64fe9d35ebddb4f6b46e492740ee59f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Dec 10 15:16:02 2008 +0000 [script] Switch to hyphens. Using hyphen in the operators is easier to read than underscores. src/cairo-script-surface.c | 60 ++++----- util/cairo-script/cairo-script-operators.c | 189 +++++++++++++++++++---------- util/cairo-trace/trace.c | 114 ++++++++--------- 3 files changed, 211 insertions(+), 152 deletions(-) commit 3abfd836a021a2950b7eea97f3b30011d58efcc8 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Dec 10 14:34:41 2008 +0000 [script] Null constructor Expose 'null', principally for comparison purposes. util/cairo-script/cairo-script-operators.c | 8 +++++++- util/cairo-script/cairo-script-private.h | 7 +++++++ 2 files changed, 14 insertions(+), 1 deletion(-) commit 7540ac7f23db47ee3f58a557ce91aaf1c0bc72d5 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Tue Dec 9 07:52:21 2008 +0200 [script] Reverse direction of bitshift. Positive shift counts should shift left according to PostScript. util/cairo-script/cairo-script-operators.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 4be479cc63a783da18652528f33bd08427a74fe7 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Dec 10 13:20:06 2008 +0000 [script] Add simple to integer converter Simple operator to coerce the type of the object on top of the stack to integer. util/cairo-script/cairo-script-operators.c | 23 +++++++++++++++++++++++ 1 file changed, 23 insertions(+) commit 6711b591b9efd8f116f2683f17d52e9a9ddfbaba Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Dec 10 12:34:30 2008 +0000 [script] Simple replay. Basic script executer. util/cairo-script/.gitignore | 1 + util/cairo-script/Makefile.am | 5 ++++- util/cairo-script/csi-exec.c | 29 +++++++++++++++++++++++++++++ 3 files changed, 34 insertions(+), 1 deletion(-) commit 177a68ef3ecd1717b81c93c53f8c7ae27838e2e5 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Dec 10 13:21:19 2008 +0000 [script] Add a few mathematical constants Just pi and the sqrt(2) for the moment, more to come on demand. util/cairo-script/cairo-script-interpreter.c | 26 ++++++++++++++++++++++++++ util/cairo-script/cairo-script-operators.c | 17 +++++++++++++++++ util/cairo-script/cairo-script-private.h | 8 ++++++++ 3 files changed, 51 insertions(+) commit 5dfaa22a578ff6561e8d15a8c08546c840476313 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Dec 10 12:26:09 2008 +0000 [script] Add repeat operator. Simple operator to repeat procedure n times. util/cairo-script/cairo-script-operators.c | 37 +++++++++++++++++++++++++++++- 1 file changed, 36 insertions(+), 1 deletion(-) commit 738cb327450821db2d3689e514376c0d9bc55e87 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Tue Dec 9 05:59:19 2008 +0200 [script] Fix \t escapes in string literals. Typo \r -> \t. util/cairo-script/cairo-script-scanner.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 265ebd372a4fd510bc29c749a46393ee7caace41 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Mon Dec 8 10:14:33 2008 +0200 [script] Flesh out the relational comparison operators to be more PostScript-like. The relational comparison operators can now compare strings vs names by content as well as performing automatic type promotions on the numeric types. For other types relational comparisons succeeed only if the values compare equal according to the eq operator, and put the interpreter into a type-error state otherwise. util/cairo-script/cairo-script-objects.c | 103 +++++++++++++++++++++ util/cairo-script/cairo-script-operators.c | 144 +++++------------------------ util/cairo-script/cairo-script-private.h | 5 + 3 files changed, 132 insertions(+), 120 deletions(-) commit 05afec7a8ac590d0d53a254895796f36a8b0eaf3 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Mon Dec 8 10:06:53 2008 +0200 [script] Implement the ne operator. util/cairo-script/cairo-script-operators.c | 19 ++++++++++++++++++- 1 file changed, 18 insertions(+), 1 deletion(-) commit 1aeb96dc287b516955db1909995e45fe26d4f87a Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Mon Dec 8 10:03:57 2008 +0200 [script] Implement PostScript-like semantics for the eq operator. The eq operator would only work for some types and put the interpreter in an error state if passed objects it didn't know how to compare. It would also not compare strings by value nor allow strings to be compared to names. This patch makes any two objects comparable. util/cairo-script/cairo-script-objects.c | 92 ++++++++++++++++++++++++++++++ util/cairo-script/cairo-script-operators.c | 64 +-------------------- util/cairo-script/cairo-script-private.h | 4 ++ 3 files changed, 97 insertions(+), 63 deletions(-) commit 620028fd19b091f525964b689ebd3a70e1636da2 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Mon Dec 8 09:23:37 2008 +0200 [script] Don't segfault when hashing empty strings. Check for the empty string. util/cairo-script/cairo-script-interpreter.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) commit 4ba77f776509caad030edf818a076ea9b071ad23 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Mon Dec 8 07:16:14 2008 +0200 [script] Fix dictionary construction to access the right stack slots. It bound values to themselves and ignored keys. util/cairo-script/cairo-script-operators.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit af53297a97eec505ac2a90768badda72056c6e3b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Dec 11 00:35:20 2008 +0000 [cairo] Use a stack buffer for text path. First try to allocate glyphs from the stack, similar to cairo_show_text(). src/cairo.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) commit 09589e20e14476df609581027d14628f52091e71 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Dec 10 23:45:44 2008 +0000 [clip] Intersect to region extents. When querying the intersection of a rectangle with the clip region, the result only depends upon the region extents so we do not need to perform an expensive region-region intersection computation. src/cairo-clip.c | 18 +++++------------- src/cairo-surface-fallback.c | 1 - 2 files changed, 5 insertions(+), 14 deletions(-) commit c5acb125e9509570c407a069f7e89acfbadda57b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Dec 10 23:06:22 2008 +0000 [surface] Avoid a short-lived context during fallback_clone_similar() We can paint to the surface directly without need of a temporary context, so do so. src/cairo-surface-fallback.c | 41 +++++++++++++++++++---------------------- 1 file changed, 19 insertions(+), 22 deletions(-) commit 888f62feaa73be326d41a485c0bcf81a3e466ee1 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Dec 10 17:42:16 2008 +0000 [test] Propagate error from path If the path is in error, just append it to the context so that the error is propagated correctly. test/copy-path.c | 5 +++++ 1 file changed, 5 insertions(+) commit f13f63e3d85d28a17a3cb60b7d1a7b552f2b6872 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Dec 10 08:54:03 2008 +0000 [test/clip-nesting] Propagate status from sub-context. Use cairo_path_append() to propagate an error status from a child context to it parent so that it is properly reported under memfault. test/clip-nesting.c | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) commit e6e33b036f9de5087a72cf1de9c0a4ce97c75812 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Dec 10 08:50:57 2008 +0000 [test/solid-pattern-cache-stress] Propagate status from sub-contexts. Use a cairo_append_path() to set an arbitrary error status on the parent context, in order to propagate errors from a child. test/solid-pattern-cache-stress.c | 32 +++++++++++++++++++++++++------- 1 file changed, 25 insertions(+), 7 deletions(-) commit bcb2724920aa1bca1a9a82b5017d3180f5f9523e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Dec 9 20:44:25 2008 +0000 [path] Mark points as const during _cairo_path_fixed_interpret() Use const to document the read-only nature of the arguments passed to the callbacks. src/cairo-path-bounds.c | 33 ++++-------- src/cairo-path-fill.c | 33 +++--------- src/cairo-path-fixed.c | 86 ++++++++++++++--------------- src/cairo-path-in-fill.c | 12 +++-- src/cairo-path-stroke.c | 108 +++++++++++++------------------------ src/cairo-path.c | 26 +++++---- src/cairo-pdf-operators.c | 12 +++-- src/cairo-pen.c | 4 +- src/cairo-quartz-surface.c | 11 ++-- src/cairo-rectangle.c | 6 +-- src/cairo-scaled-font.c | 12 +++-- src/cairo-script-surface.c | 12 +++-- src/cairo-spans.c | 18 +++---- src/cairo-svg-surface.c | 15 +++--- src/cairo-type1-fallback.c | 16 +++--- src/cairo-types-private.h | 3 +- src/cairo-win32-printing-surface.c | 12 +++-- src/cairoint.h | 12 ++--- 18 files changed, 186 insertions(+), 245 deletions(-) commit 68b29cafa597128e7cae86608e04ecae6070dad9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Dec 9 20:15:34 2008 +0000 [spline] Propagate errors during add point. Yikes! The callback could fail so we need to propagate the error status. src/cairo-path-fill.c | 11 +++-------- src/cairo-path-fixed.c | 7 ++----- src/cairo-path-in-fill.c | 7 ++----- src/cairo-path-stroke.c | 14 ++++++-------- src/cairo-pen.c | 23 ++++++++++++----------- src/cairo-spline.c | 40 ++++++++++++++++++++-------------------- src/cairo-types-private.h | 6 +++++- src/cairoint.h | 9 ++------- 8 files changed, 52 insertions(+), 65 deletions(-) commit 2f3905dec38a710234aba30e1983b80ea3066a50 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Dec 9 19:21:35 2008 +0000 [image] Set status on error path. We failed to set the status after cairo_image_surface_create() returned an error. src/cairo-image-surface.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit dfc49a67e62bea359b33413e9b882e92952431de Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Dec 9 14:44:51 2008 +0000 [test] Trim the number of fallback resolutions tested. Simplify the test matrix to only include testing the fallback-resolution of a uniform scale in one or both axes. test/Makefile.am | 20 --- test/fallback-resolution.c | 222 ++++++++++++++------------ test/fallback-resolution.ppi150x150.ref.png | Bin 8043 -> 8058 bytes test/fallback-resolution.ppi150x300.ref.png | Bin 6848 -> 0 bytes test/fallback-resolution.ppi150x37.5.ref.png | Bin 13176 -> 0 bytes test/fallback-resolution.ppi150x600.ref.png | Bin 6338 -> 0 bytes test/fallback-resolution.ppi150x72.ref.png | Bin 8370 -> 8303 bytes test/fallback-resolution.ppi150x75.ref.png | Bin 10445 -> 0 bytes test/fallback-resolution.ppi300x150.ref.png | Bin 6851 -> 0 bytes test/fallback-resolution.ppi300x300.ref.png | Bin 5637 -> 5639 bytes test/fallback-resolution.ppi300x37.5.ref.png | Bin 12870 -> 0 bytes test/fallback-resolution.ppi300x600.ref.png | Bin 4848 -> 0 bytes test/fallback-resolution.ppi300x72.ref.png | Bin 7053 -> 7014 bytes test/fallback-resolution.ppi300x75.ref.png | Bin 9617 -> 0 bytes test/fallback-resolution.ppi37.5x150.ref.png | Bin 12728 -> 0 bytes test/fallback-resolution.ppi37.5x300.ref.png | Bin 11981 -> 0 bytes test/fallback-resolution.ppi37.5x37.5.ref.png | Bin 14098 -> 14092 bytes test/fallback-resolution.ppi37.5x600.ref.png | Bin 11654 -> 0 bytes test/fallback-resolution.ppi37.5x72.ref.png | Bin 12990 -> 12906 bytes test/fallback-resolution.ppi37.5x75.ref.png | Bin 13787 -> 0 bytes test/fallback-resolution.ppi600x150.ref.png | Bin 6096 -> 0 bytes test/fallback-resolution.ppi600x300.ref.png | Bin 4812 -> 0 bytes test/fallback-resolution.ppi600x37.5.ref.png | Bin 12401 -> 0 bytes test/fallback-resolution.ppi600x600.ref.png | Bin 4104 -> 4090 bytes test/fallback-resolution.ppi600x72.ref.png | Bin 6398 -> 6326 bytes test/fallback-resolution.ppi600x75.ref.png | Bin 8974 -> 0 bytes test/fallback-resolution.ppi72x150.ref.png | Bin 8069 -> 8061 bytes test/fallback-resolution.ppi72x300.ref.png | Bin 6803 -> 6911 bytes test/fallback-resolution.ppi72x37.5.ref.png | Bin 13147 -> 13167 bytes test/fallback-resolution.ppi72x600.ref.png | Bin 6245 -> 6214 bytes test/fallback-resolution.ppi72x72.ref.png | Bin 8409 -> 8416 bytes test/fallback-resolution.ppi72x75.ref.png | Bin 10628 -> 10620 bytes test/fallback-resolution.ppi75x150.ref.png | Bin 9940 -> 0 bytes test/fallback-resolution.ppi75x300.ref.png | Bin 9003 -> 0 bytes test/fallback-resolution.ppi75x37.5.ref.png | Bin 14020 -> 0 bytes test/fallback-resolution.ppi75x600.ref.png | Bin 8455 -> 0 bytes test/fallback-resolution.ppi75x72.ref.png | Bin 10202 -> 10119 bytes test/fallback-resolution.ppi75x75.ref.png | Bin 12101 -> 12127 bytes 38 files changed, 121 insertions(+), 121 deletions(-) commit 0ebbcd671c362d98a60214872fa48aee1e42dde1 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Dec 9 14:30:58 2008 +0000 [test] Fix memleak from fallback-resolution. The test-name was being recreated for every pass, but was never freed. test/fallback-resolution.c | 2 ++ 1 file changed, 2 insertions(+) commit a9697e5fb04d045fe7aaf0cee80c41eb9d09e1c1 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Dec 12 11:08:13 2008 +0000 [test] Exercise glyph culling. Sascha Steinbiss reported an issue with glyph culling, http://lists.cairographics.org/archives/cairo/2008-December/015976.html, whereby we failed to update the text clusters upon culling the glyphs in the gstate and proceeded to read beyond the end of the glyph array in the PDF backend. This test case setups a similar condition as reported, by trying to write a wide string into a small box. test/Makefile.am | 3 ++ test/culled-glyphs.c | 62 ++++++++++++++++++++++++++++++++++++++++++ test/culled-glyphs.ps.ref.png | Bin 0 -> 372 bytes test/culled-glyphs.ref.png | Bin 0 -> 434 bytes 4 files changed, 65 insertions(+) commit 095a1fd7860bd1f6d4116766ade0ec9a382bb957 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Dec 5 21:14:45 2008 +0000 [gstate] Remove culled glyphs from clusters. Sascha Steinbiss reported a bug where the PDF backend was reading beyond the end of the glyph array: http://lists.cairographics.org/archives/cairo/2008-December/015976.html. It transpires that in the early glyph culling in the gstate we were not updating the clusters to skip culled glyphs. src/cairo-gstate.c | 223 ++++++++++++++++++++++++++++++++++++++++++----------- src/cairo.c | 9 ++- 2 files changed, 183 insertions(+), 49 deletions(-) commit 834f1d7b7097dcc3a32f6c65d21e87fd272d924a Author: Julien Danjou <julien@danjou.info> Date: Tue Nov 18 10:01:49 2008 +0100 [xcb] check for render extension presence Otherwise this may leads to an invalid memory access to r. Fixes: Bug 18588 - XCB backend fails with missing render. https://bugs.freedesktop.org/show_bug.cgi?id=18588 Signed-off-by: Julien Danjou <julien@danjou.info> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> AUTHORS | 1 + src/cairo-xcb-surface.c | 8 ++++++-- 2 files changed, 7 insertions(+), 2 deletions(-) commit e184bf0d3592c8703b1193d37553cdcdee313139 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Dec 12 10:12:43 2008 +0000 Fix compilation with --enable-png=no Adrian Johnson spotted that the build was broken if the PNG functions were disabled. Fix. src/cairoint.h | 1 - util/cairo-script/cairo-script-operators.c | 8 ++++++++ util/cairo-trace/trace.c | 4 ++++ 3 files changed, 12 insertions(+), 1 deletion(-) commit 391c6026402dee09bb34816637114cea990cd93a Merge: 2f1b581f5 d93bf10ed Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Thu Dec 11 14:41:01 2008 -0500 Merge branch '1.8' commit 2f1b581f54713c899f3b03af1e0ac8c38c36c385 Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Thu Dec 11 10:34:37 2008 -0500 [quartz] Create a copy instead of increasing the reference The pattern could be stack allocated so we can't take a reference to it. Some testing of quartz shows that it doesn't deal with malloc failure particularily well. In the best case CGFunctionCreate returns NULL, in the worst case it just crashes. Quartz does seem to be able to handle a NULL CGFunctionRef, so returning NULL if we fail to copy the pattern avoids complicating the code to deal with propagating the failure and shouldn't cause any additional crashes. Based on a patch by Paolo Bonzini. src/cairo-quartz-surface.c | 16 ++++++++++++++-- 1 file changed, 14 insertions(+), 2 deletions(-) commit f7b3f1b7d70a271575d3bb233b7ee51a21861c05 Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Wed Dec 10 17:54:59 2008 -0500 [quartz] Propagate const patterns through the casts As part of this avoid using cairo_pattern_get_matrix() because it requires a 'cairo_pattern_t *' instead of 'const cairo_pattern *' Also, make a copy of the pattern before pasing it in to cairo_set_source() src/cairo-quartz-surface.c | 39 +++++++++++++++++++++++---------------- 1 file changed, 23 insertions(+), 16 deletions(-) commit d93bf10edc7e432349524221a3d1b0f0b2ec8090 Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Wed Dec 10 17:02:55 2008 -0500 [quartz] Create a copy of the pattern so that the reference counts are balanced The pattern could be stack allocated so we can't take a reference to it; instead make a copy. Based on a patch by Paolo Bonzini. src/cairo-quartz-surface.c | 16 ++++++++++++++-- 1 file changed, 14 insertions(+), 2 deletions(-) commit db4c427e8460b0e326691bd6d3784178a713eb33 Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Tue Dec 9 12:03:11 2008 -0500 [quartz] Change some pattern casts to access to the base class Avoiding casts gives us some more type safety. src/cairo-quartz-surface.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) commit 5f60e2e00455f2894c7ce3805a029c806f21514a Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Tue Dec 9 11:41:10 2008 -0500 Reorganize the (set|get)_antialias prototypes Move _cairo_gstate_set_antialias() and _cairo_gstate_get_antialias() into the gstate section of cairoint.h src/cairoint.h | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) commit 673640a3b3931995897b01d49c5dd8d82b50dac2 Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Thu Dec 4 17:53:06 2008 -0500 [win32] Use MOD instead of the '%' operator Repeat should be handled using MOD instead of '%' so that negative numbers are handled as expected. E.g. -1 mod 600 = 599, not 495 as the '%' operator gives. This was causing https://bugzilla.mozilla.org/show_bug.cgi?id=466258 Patch from Robert O'Callahan src/cairo-win32-surface.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) commit 5e06085b483dcaaa7b1b29b13cd2813c7e51e02a Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Fri Aug 1 23:28:15 2008 +0300 [cairo-spans] Render clip mask surfaces with spans if we can. Generating surface masks for clipping can also benefit from span rendering sometimes. src/cairo-clip.c | 140 +++++++++++++++++++++ test/clip-fill-rule.pdf.argb32.ref.png | Bin 0 -> 509 bytes test/clip-fill-rule.rgb24.ref.png | Bin 380 -> 390 bytes test/clip-fill-rule.test-paginated.rgb24.ref.png | Bin 0 -> 361 bytes test/clip-fill-rule.xlib.rgb24.ref.png | Bin 0 -> 380 bytes test/clip-nesting.pdf.argb32.ref.png | Bin 0 -> 850 bytes test/clip-nesting.rgb24.ref.png | Bin 955 -> 963 bytes test/clip-nesting.test-fallback.rgb24.ref.png | Bin 0 -> 936 bytes test/clip-nesting.test-paginated.rgb24.ref.png | Bin 0 -> 936 bytes test/clip-nesting.xlib.rgb24.ref.png | Bin 0 -> 955 bytes test/clip-operator.pdf.argb32.ref.png | Bin 11600 -> 11604 bytes test/clip-operator.ps3.argb32.ref.png | Bin 7574 -> 7576 bytes test/clip-operator.ref.png | Bin 8247 -> 8210 bytes test/clip-operator.test-paginated.argb32.ref.png | Bin 0 -> 8247 bytes test/clip-twice.pdf.argb32.ref.png | Bin 1458 -> 1498 bytes test/clip-twice.ref.png | Bin 1362 -> 1342 bytes test/clip-twice.rgb24.ref.png | Bin 1198 -> 1203 bytes test/clip-twice.test-fallback.argb32.ref.png | Bin 0 -> 1343 bytes test/clip-twice.test-fallback.rgb24.ref.png | Bin 0 -> 1179 bytes test/clip-twice.test-paginated.argb32.ref.png | Bin 0 -> 1361 bytes test/clip-twice.test-paginated.rgb24.ref.png | Bin 0 -> 1199 bytes test/clip-twice.xlib.ref.png | Bin 0 -> 1362 bytes test/clip-twice.xlib.rgb24.ref.png | Bin 0 -> 1198 bytes test/device-offset-fractional.pdf.argb32.ref.png | Bin 0 -> 275 bytes test/device-offset-fractional.pdf.rgb24.ref.png | Bin 0 -> 275 bytes test/filter-nearest-offset.pdf.argb32.ref.png | Bin 0 -> 4295 bytes test/filter-nearest-offset.pdf.rgb24.ref.png | Bin 0 -> 4295 bytes test/filter-nearest-transformed.pdf.argb32.ref.png | Bin 0 -> 488 bytes test/filter-nearest-transformed.pdf.rgb24.ref.png | Bin 0 -> 488 bytes test/linear-gradient.pdf.argb32.ref.png | Bin 1118 -> 1112 bytes test/linear-gradient.pdf.rgb24.ref.png | Bin 1118 -> 1112 bytes test/mask.pdf.argb32.ref.png | Bin 9956 -> 9903 bytes test/mask.ref.png | Bin 8571 -> 8581 bytes test/mask.svg11.argb32.ref.png | Bin 8625 -> 8682 bytes test/mask.svg12.argb32.ref.png | Bin 8625 -> 8682 bytes test/meta-surface-pattern.pdf.argb32.ref.png | Bin 4017 -> 4011 bytes test/meta-surface-pattern.svg11.argb32.ref.png | Bin 3928 -> 3924 bytes test/meta-surface-pattern.svg12.argb32.ref.png | Bin 3928 -> 3924 bytes test/rotate-image-surface-paint.pdf.argb32.ref.png | Bin 209 -> 215 bytes test/rotate-image-surface-paint.pdf.rgb24.ref.png | Bin 207 -> 215 bytes test/surface-pattern-scale-down.pdf.argb32.ref.png | Bin 2386 -> 1532 bytes test/surface-pattern-scale-down.pdf.rgb24.ref.png | Bin 2386 -> 1532 bytes test/surface-pattern-scale-up.pdf.argb32.ref.png | Bin 4247 -> 3834 bytes test/surface-pattern-scale-up.pdf.rgb24.ref.png | Bin 4247 -> 3834 bytes test/surface-pattern.pdf.argb32.ref.png | Bin 0 -> 14808 bytes test/surface-pattern.pdf.rgb24.ref.png | Bin 0 -> 14808 bytes test/trap-clip.pdf.rgb24.ref.png | Bin 6642 -> 6621 bytes test/trap-clip.ref.png | Bin 5865 -> 5829 bytes test/trap-clip.test-paginated.argb32.ref.png | Bin 0 -> 5865 bytes 49 files changed, 140 insertions(+) commit 18634c37026a2d6147443cb6d991576f62b07e6d Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Thu Jul 24 20:47:14 2008 +0300 [cairo-spans] Hook up filling paths with spans to cairo-surface-fallback.c. This speeds up the mask generation step in cairo_fill() for the image surface by up to 10x in especially favourable cases. image-rgba twin-800 7757.80 0.20% -> 749.41 0.29%: 10.36x speedup image-rgba spiral-diag-pixalign-nonzero-fill-512 15.16 0.44% -> 3.45 8.80%: 5.54x speedup More typical simple non-rectilinear geometries are sped up by 30-50%. This patch does not affect any stroking operations or any fill operations of pixel aligned rectilinear geometries; those are still rendered using trapezoids. src/cairo-surface-fallback.c | 86 ++++++++++++++++++++- test/clip-fill-rule.test-fallback.rgb24.ref.png | Bin 0 -> 361 bytes test/clip-operator.pdf.argb32.ref.png | Bin 12125 -> 11600 bytes test/clip-operator.pdf.rgb24.ref.png | Bin 7367 -> 6882 bytes test/clip-operator.ps2.rgb24.ref.png | Bin 3624 -> 3736 bytes test/clip-operator.ps3.argb32.ref.png | Bin 0 -> 7574 bytes test/clip-operator.ps3.rgb24.ref.png | Bin 3624 -> 3736 bytes test/clip-operator.ref.png | Bin 8271 -> 8247 bytes test/clip-operator.rgb24.ref.png | Bin 3258 -> 3279 bytes test/clip-operator.test-fallback.argb32.ref.png | Bin 0 -> 8252 bytes test/clip-operator.test-fallback.rgb24.ref.png | Bin 0 -> 3241 bytes test/clip-operator.xlib-fallback.rgb24.ref.png | Bin 0 -> 3254 bytes test/clip-operator.xlib.ref.png | Bin 0 -> 8271 bytes test/clip-operator.xlib.rgb24.ref.png | Bin 0 -> 3258 bytes test/clip-twice.pdf.argb32.ref.png | Bin 0 -> 1458 bytes test/clipped-group.pdf.argb32.ref.png | Bin 0 -> 298 bytes test/clipped-group.pdf.rgb24.ref.png | Bin 0 -> 298 bytes test/degenerate-arc.ref.png | Bin 616 -> 544 bytes test/degenerate-arc.test-fallback.argb32.ref.png | Bin 0 -> 547 bytes test/degenerate-arc.test-fallback.rgb24.ref.png | Bin 0 -> 547 bytes test/degenerate-arc.xlib.ref.png | Bin 0 -> 616 bytes test/fill-alpha-pattern.pdf.argb32.ref.png | Bin 3887 -> 3750 bytes test/fill-alpha-pattern.pdf.rgb24.ref.png | Bin 3840 -> 3758 bytes test/fill-alpha-pattern.ps3.argb32.ref.png | Bin 0 -> 4070 bytes test/fill-alpha-pattern.ps3.rgb24.ref.png | Bin 0 -> 4473 bytes test/fill-alpha-pattern.ref.png | Bin 3653 -> 3374 bytes ...fill-alpha-pattern.test-fallback.argb32.ref.png | Bin 0 -> 3379 bytes .../fill-alpha-pattern.test-fallback.rgb24.ref.png | Bin 0 -> 3379 bytes test/fill-alpha-pattern.xlib.ref.png | Bin 0 -> 3653 bytes test/fill-alpha.ref.png | Bin 2989 -> 2728 bytes test/fill-alpha.test-fallback.argb32.ref.png | Bin 0 -> 2824 bytes test/fill-alpha.test-fallback.rgb24.ref.png | Bin 0 -> 2824 bytes test/fill-alpha.xlib.ref.png | Bin 0 -> 2989 bytes test/fill-degenerate-sort-order.ref.png | Bin 2397 -> 2406 bytes test/fill-degenerate-sort-order.rgb24.ref.png | Bin 2060 -> 2052 bytes ...enerate-sort-order.test-fallback.argb32.ref.png | Bin 0 -> 2378 bytes ...generate-sort-order.test-fallback.rgb24.ref.png | Bin 0 -> 2041 bytes test/fill-degenerate-sort-order.xlib.ref.png | Bin 0 -> 2397 bytes test/fill-degenerate-sort-order.xlib.rgb24.ref.png | Bin 0 -> 2060 bytes test/fill-missed-stop.pdf.argb32.ref.png | Bin 0 -> 452 bytes test/fill-rule.ref.png | Bin 1979 -> 2061 bytes test/fill-rule.rgb24.ref.png | Bin 1722 -> 1780 bytes test/fill-rule.test-fallback.argb32.ref.png | Bin 0 -> 1979 bytes test/fill-rule.test-fallback.rgb24.ref.png | Bin 0 -> 1703 bytes test/fill-rule.xlib.ref.png | Bin 0 -> 1979 bytes test/fill-rule.xlib.rgb24.ref.png | Bin 0 -> 1722 bytes test/finer-grained-fallbacks.ps2.argb32.ref.png | Bin 0 -> 1173 bytes test/finer-grained-fallbacks.ps2.rgb24.ref.png | Bin 1096 -> 1154 bytes test/finer-grained-fallbacks.ps3.argb32.ref.png | Bin 0 -> 1173 bytes test/finer-grained-fallbacks.ps3.rgb24.ref.png | Bin 1096 -> 1154 bytes test/finer-grained-fallbacks.ref.png | Bin 1111 -> 1069 bytes test/finer-grained-fallbacks.rgb24.ref.png | Bin 1114 -> 839 bytes ...-grained-fallbacks.test-fallback.argb32.ref.png | Bin 0 -> 1111 bytes ...r-grained-fallbacks.test-fallback.rgb24.ref.png | Bin 0 -> 854 bytes test/finer-grained-fallbacks.xlib.ref.png | Bin 0 -> 1111 bytes test/finer-grained-fallbacks.xlib.rgb24.ref.png | Bin 0 -> 1114 bytes test/font-matrix-translation.svg11.argb32.ref.png | Bin 0 -> 857 bytes test/font-matrix-translation.svg11.rgb24.ref.png | Bin 0 -> 857 bytes test/font-matrix-translation.svg12.argb32.ref.png | Bin 0 -> 857 bytes test/font-matrix-translation.svg12.rgb24.ref.png | Bin 0 -> 857 bytes test/ft-show-glyphs-table.svg11.argb32.ref.png | Bin 0 -> 9953 bytes test/ft-show-glyphs-table.svg11.rgb24.ref.png | Bin 0 -> 9953 bytes test/ft-show-glyphs-table.svg12.argb32.ref.png | Bin 0 -> 9953 bytes test/ft-show-glyphs-table.svg12.rgb24.ref.png | Bin 0 -> 9953 bytes ...t-text-vertical-layout-type1.pdf.argb32.ref.png | Bin 0 -> 3632 bytes ...ft-text-vertical-layout-type1.pdf.rgb24.ref.png | Bin 0 -> 3632 bytes test/ft-text-vertical-layout-type1.ref.png | Bin 3980 -> 3643 bytes ...text-vertical-layout-type1.svg11.argb32.ref.png | Bin 0 -> 3614 bytes ...-text-vertical-layout-type1.svg11.rgb24.ref.png | Bin 0 -> 3614 bytes ...text-vertical-layout-type1.svg12.argb32.ref.png | Bin 0 -> 3614 bytes ...-text-vertical-layout-type1.svg12.rgb24.ref.png | Bin 0 -> 3614 bytes ...tical-layout-type1.test-fallback.argb32.ref.png | Bin 0 -> 3639 bytes ...rtical-layout-type1.test-fallback.rgb24.ref.png | Bin 0 -> 3639 bytes test/ft-text-vertical-layout-type1.xlib.ref.png | Bin 0 -> 3980 bytes ...t-text-vertical-layout-type3.pdf.argb32.ref.png | Bin 0 -> 3642 bytes ...ft-text-vertical-layout-type3.pdf.rgb24.ref.png | Bin 0 -> 3642 bytes test/ft-text-vertical-layout-type3.ref.png | Bin 3934 -> 3609 bytes ...text-vertical-layout-type3.svg11.argb32.ref.png | Bin 0 -> 3640 bytes ...-text-vertical-layout-type3.svg11.rgb24.ref.png | Bin 0 -> 3640 bytes ...text-vertical-layout-type3.svg12.argb32.ref.png | Bin 0 -> 3640 bytes ...-text-vertical-layout-type3.svg12.rgb24.ref.png | Bin 0 -> 3640 bytes ...tical-layout-type3.test-fallback.argb32.ref.png | Bin 0 -> 3605 bytes ...rtical-layout-type3.test-fallback.rgb24.ref.png | Bin 0 -> 3605 bytes test/ft-text-vertical-layout-type3.xlib.ref.png | Bin 0 -> 3934 bytes test/huge-pattern.pdf.argb32.ref.png | Bin 0 -> 2430 bytes test/linear-gradient.pdf.argb32.ref.png | Bin 0 -> 1118 bytes test/linear-gradient.pdf.rgb24.ref.png | Bin 0 -> 1118 bytes test/linear-gradient.ref.png | Bin 1021 -> 983 bytes test/linear-gradient.svg11.argb32.ref.png | Bin 0 -> 988 bytes test/linear-gradient.svg11.rgb24.ref.png | Bin 0 -> 988 bytes test/linear-gradient.svg12.argb32.ref.png | Bin 0 -> 988 bytes test/linear-gradient.svg12.rgb24.ref.png | Bin 0 -> 988 bytes test/linear-gradient.test-fallback.argb32.ref.png | Bin 0 -> 923 bytes test/linear-gradient.test-fallback.rgb24.ref.png | Bin 0 -> 923 bytes test/linear-gradient.xlib.ref.png | Bin 0 -> 1021 bytes test/mask-alpha.ref.png | Bin 640 -> 643 bytes test/mask-alpha.svg11.argb32.ref.png | Bin 615 -> 642 bytes test/mask-alpha.svg11.rgb24.ref.png | Bin 0 -> 592 bytes test/mask-alpha.svg12.argb32.ref.png | Bin 615 -> 642 bytes test/mask-alpha.svg12.rgb24.ref.png | Bin 0 -> 592 bytes test/mask-alpha.test-fallback.argb32.ref.png | Bin 0 -> 627 bytes test/mask-alpha.xlib.ref.png | Bin 0 -> 640 bytes test/mask-alpha.xlib.rgb24.ref.png | Bin 0 -> 599 bytes test/mask.pdf.argb32.ref.png | Bin 8881 -> 9956 bytes test/mask.pdf.rgb24.ref.png | Bin 8267 -> 8735 bytes test/mask.ref.png | Bin 8476 -> 8571 bytes test/mask.rgb24.ref.png | Bin 7041 -> 7216 bytes test/mask.svg11.argb32.ref.png | Bin 8527 -> 8625 bytes test/mask.svg11.rgb24.ref.png | Bin 7935 -> 7203 bytes test/mask.svg12.argb32.ref.png | Bin 8527 -> 8625 bytes test/mask.svg12.rgb24.ref.png | Bin 7935 -> 7203 bytes test/mask.test-fallback.argb32.ref.png | Bin 0 -> 8457 bytes test/mask.test-fallback.rgb24.ref.png | Bin 0 -> 7058 bytes test/mask.xlib-fallback.rgb24.ref.png | Bin 0 -> 7216 bytes test/mask.xlib.ref.png | Bin 0 -> 8476 bytes test/mask.xlib.rgb24.ref.png | Bin 0 -> 7041 bytes test/meta-surface-pattern.pdf.argb32.ref.png | Bin 0 -> 4017 bytes test/meta-surface-pattern.pdf.rgb24.ref.png | Bin 4009 -> 3910 bytes test/meta-surface-pattern.svg11.argb32.ref.png | Bin 3924 -> 3928 bytes test/meta-surface-pattern.svg11.rgb24.ref.png | Bin 4593 -> 3914 bytes test/meta-surface-pattern.svg12.argb32.ref.png | Bin 3924 -> 3928 bytes test/meta-surface-pattern.svg12.rgb24.ref.png | Bin 4593 -> 3914 bytes test/operator-clear.pdf.argb32.ref.png | Bin 1614 -> 1607 bytes test/operator-clear.ps2.argb32.ref.png | Bin 0 -> 1156 bytes test/operator-clear.ps3.argb32.ref.png | Bin 0 -> 1156 bytes test/operator-source.pdf.argb32.ref.png | Bin 5149 -> 5112 bytes test/operator-source.pdf.rgb24.ref.png | Bin 4354 -> 4186 bytes test/operator-source.ref.png | Bin 4420 -> 4425 bytes test/operator-source.rgb24.ref.png | Bin 3201 -> 3231 bytes test/operator-source.test-fallback.argb32.ref.png | Bin 0 -> 4401 bytes test/operator-source.test-fallback.rgb24.ref.png | Bin 0 -> 3200 bytes test/operator-source.xlib-fallback.rgb24.ref.png | Bin 0 -> 3193 bytes test/operator-source.xlib.ref.png | Bin 0 -> 4420 bytes test/operator-source.xlib.rgb24.ref.png | Bin 0 -> 3201 bytes test/over-above-source.ps2.argb32.ref.png | Bin 636 -> 558 bytes test/over-above-source.ps3.argb32.ref.png | Bin 636 -> 558 bytes test/over-above-source.ref.png | Bin 538 -> 560 bytes test/over-above-source.rgb24.ref.png | Bin 461 -> 466 bytes .../over-above-source.test-fallback.argb32.ref.png | Bin 0 -> 533 bytes test/over-above-source.test-fallback.rgb24.ref.png | Bin 0 -> 450 bytes test/over-above-source.xlib.ref.png | Bin 0 -> 538 bytes test/over-above-source.xlib.rgb24.ref.png | Bin 0 -> 461 bytes test/over-around-source.pdf.argb32.ref.png | Bin 0 -> 585 bytes test/over-around-source.ps2.argb32.ref.png | Bin 632 -> 522 bytes test/over-around-source.ps3.argb32.ref.png | Bin 632 -> 522 bytes test/over-around-source.ref.png | Bin 614 -> 645 bytes ...over-around-source.test-fallback.argb32.ref.png | Bin 0 -> 610 bytes test/over-around-source.xlib.ref.png | Bin 0 -> 614 bytes test/over-around-source.xlib.rgb24.ref.png | Bin 0 -> 503 bytes test/over-below-source.pdf.argb32.ref.png | Bin 0 -> 464 bytes test/over-between-source.ps2.argb32.ref.png | Bin 678 -> 551 bytes test/over-between-source.ps3.argb32.ref.png | Bin 678 -> 551 bytes test/over-between-source.ref.png | Bin 575 -> 612 bytes ...ver-between-source.test-fallback.argb32.ref.png | Bin 0 -> 578 bytes test/over-between-source.xlib.ref.png | Bin 0 -> 575 bytes test/over-between-source.xlib.rgb24.ref.png | Bin 0 -> 473 bytes test/push-group.pdf.argb32.ref.png | Bin 0 -> 2722 bytes test/push-group.pdf.rgb24.ref.png | Bin 2714 -> 2740 bytes test/push-group.ref.png | Bin 3126 -> 3060 bytes test/push-group.rgb24.ref.png | Bin 2961 -> 2912 bytes test/push-group.svg11.argb32.ref.png | Bin 2935 -> 3034 bytes test/push-group.svg12.argb32.ref.png | Bin 2935 -> 3034 bytes test/push-group.test-fallback.argb32.ref.png | Bin 0 -> 3107 bytes test/push-group.test-fallback.rgb24.ref.png | Bin 0 -> 2942 bytes test/push-group.xlib-fallback.rgb24.ref.png | Bin 0 -> 2912 bytes test/push-group.xlib.ref.png | Bin 0 -> 3126 bytes test/push-group.xlib.rgb24.ref.png | Bin 0 -> 2961 bytes test/radial-gradient.pdf.argb32.ref.png | Bin 0 -> 79601 bytes test/radial-gradient.pdf.rgb24.ref.png | Bin 0 -> 79601 bytes test/random-intersections.ref.png | Bin 148722 -> 133462 bytes ...ndom-intersections.test-fallback.argb32.ref.png | Bin 0 -> 132312 bytes ...andom-intersections.test-fallback.rgb24.ref.png | Bin 0 -> 132312 bytes test/random-intersections.xlib.ref.png | Bin 0 -> 148722 bytes test/smask-fill.pdf.argb32.ref.png | Bin 0 -> 1909 bytes test/smask-fill.pdf.rgb24.ref.png | Bin 0 -> 1909 bytes test/smask-fill.ref.png | Bin 1223 -> 1156 bytes test/smask-fill.svg11.argb32.ref.png | Bin 0 -> 1128 bytes test/smask-fill.svg11.rgb24.ref.png | Bin 0 -> 1128 bytes test/smask-fill.svg12.argb32.ref.png | Bin 0 -> 1128 bytes test/smask-fill.svg12.rgb24.ref.png | Bin 0 -> 1128 bytes test/smask-fill.test-fallback.argb32.ref.png | Bin 0 -> 1148 bytes test/smask-fill.test-fallback.rgb24.ref.png | Bin 0 -> 1148 bytes test/smask-fill.xlib-fallback.ref.png | Bin 0 -> 1156 bytes test/smask-fill.xlib.ref.png | Bin 0 -> 1223 bytes test/smask-image-mask.pdf.argb32.ref.png | Bin 0 -> 1651 bytes test/smask-image-mask.pdf.rgb24.ref.png | Bin 0 -> 1651 bytes test/smask-mask.pdf.argb32.ref.png | Bin 0 -> 4398 bytes test/smask-mask.pdf.rgb24.ref.png | Bin 0 -> 4398 bytes test/smask-paint.pdf.argb32.ref.png | Bin 0 -> 4496 bytes test/smask-paint.pdf.rgb24.ref.png | Bin 0 -> 4496 bytes test/smask-stroke.pdf.argb32.ref.png | Bin 0 -> 1417 bytes test/smask-stroke.pdf.rgb24.ref.png | Bin 0 -> 1417 bytes test/smask-text.svg11.argb32.ref.png | Bin 0 -> 1791 bytes test/smask-text.svg11.rgb24.ref.png | Bin 0 -> 1791 bytes test/smask-text.svg12.argb32.ref.png | Bin 0 -> 1791 bytes test/smask-text.svg12.rgb24.ref.png | Bin 0 -> 1791 bytes test/smask.pdf.argb32.ref.png | Bin 0 -> 4496 bytes test/smask.pdf.rgb24.ref.png | Bin 0 -> 4496 bytes test/text-pattern.pdf.argb32.ref.png | Bin 1823 -> 2151 bytes test/text-pattern.svg11.argb32.ref.png | Bin 1733 -> 1743 bytes test/text-pattern.svg12.argb32.ref.png | Bin 1733 -> 1743 bytes test/text-rotate.svg11.argb32.ref.png | Bin 0 -> 16942 bytes test/text-rotate.svg11.rgb24.ref.png | Bin 0 -> 16942 bytes test/text-rotate.svg12.argb32.ref.png | Bin 0 -> 16942 bytes test/text-rotate.svg12.rgb24.ref.png | Bin 0 -> 16942 bytes test/text-transform.svg11.argb32.ref.png | Bin 0 -> 5677 bytes test/text-transform.svg11.rgb24.ref.png | Bin 0 -> 5677 bytes test/text-transform.svg12.argb32.ref.png | Bin 0 -> 5677 bytes test/text-transform.svg12.rgb24.ref.png | Bin 0 -> 5677 bytes test/trap-clip.pdf.argb32.ref.png | Bin 5809 -> 6720 bytes test/trap-clip.pdf.rgb24.ref.png | Bin 5768 -> 6642 bytes test/trap-clip.ps2.argb32.ref.png | Bin 5690 -> 4849 bytes test/trap-clip.ref.png | Bin 5772 -> 5865 bytes test/trap-clip.rgb24.ref.png | Bin 5365 -> 5457 bytes test/trap-clip.test-fallback.argb32.ref.png | Bin 0 -> 5753 bytes test/trap-clip.test-fallback.rgb24.ref.png | Bin 0 -> 5379 bytes test/trap-clip.xlib.ref.png | Bin 0 -> 5772 bytes test/trap-clip.xlib.rgb24.ref.png | Bin 0 -> 5365 bytes test/twin.svg11.argb32.ref.png | Bin 0 -> 1797 bytes test/twin.svg11.rgb24.ref.png | Bin 0 -> 1797 bytes test/twin.svg12.argb32.ref.png | Bin 0 -> 1797 bytes test/twin.svg12.rgb24.ref.png | Bin 0 -> 1797 bytes test/unbounded-operator.pdf.argb32.ref.png | Bin 0 -> 2713 bytes test/unbounded-operator.ps2.argb32.ref.png | Bin 0 -> 2713 bytes test/unbounded-operator.ps3.argb32.ref.png | Bin 0 -> 2713 bytes test/unbounded-operator.rgb24.ref.png | Bin 1315 -> 1341 bytes .../unbounded-operator.test-fallback.rgb24.ref.png | Bin 0 -> 1306 bytes test/unbounded-operator.xlib.rgb24.ref.png | Bin 0 -> 1315 bytes test/user-font-proxy.pdf.argb32.ref.png | Bin 0 -> 16937 bytes test/user-font-proxy.pdf.rgb24.ref.png | Bin 0 -> 16937 bytes test/user-font-proxy.ref.png | Bin 18121 -> 16937 bytes test/user-font-proxy.svg11.argb32.ref.png | Bin 0 -> 16836 bytes test/user-font-proxy.svg11.rgb24.ref.png | Bin 0 -> 16836 bytes test/user-font-proxy.svg12.argb32.ref.png | Bin 0 -> 16836 bytes test/user-font-proxy.svg12.rgb24.ref.png | Bin 0 -> 16836 bytes test/user-font-proxy.test-fallback.argb32.ref.png | Bin 0 -> 16835 bytes test/user-font-proxy.test-fallback.rgb24.ref.png | Bin 0 -> 16835 bytes test/user-font-proxy.xlib.ref.png | Bin 0 -> 18121 bytes test/user-font.ref.png | Bin 6183 -> 6082 bytes test/user-font.svg11.argb32.ref.png | Bin 0 -> 6411 bytes test/user-font.svg11.rgb24.ref.png | Bin 0 -> 6411 bytes test/user-font.svg12.argb32.ref.png | Bin 0 -> 6411 bytes test/user-font.svg12.rgb24.ref.png | Bin 0 -> 6411 bytes test/user-font.test-fallback.argb32.ref.png | Bin 0 -> 5601 bytes test/user-font.test-fallback.rgb24.ref.png | Bin 0 -> 5601 bytes test/user-font.xlib.ref.png | Bin 0 -> 6183 bytes 246 files changed, 85 insertions(+), 1 deletion(-) commit 85b81a3e59401e2fc68209634f7622694e7d30e1 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Fri Jul 11 13:10:20 2008 +0300 [cairo-spans] Implement a span renderer for cairo_image_surface_t. This implementation first produces an A8 alpha mask and then pixman_image_composites the result to the destination with the source. Clipping is handled by pixman when it is region clipping or by cairo-surface-fallback when it is something more complex. src/cairo-image-surface.c | 273 +++++++++++++++++++++++++++++++++++++++------- 1 file changed, 233 insertions(+), 40 deletions(-) commit 7994fc06ad66e31fcbc16f6e8cd9ad226022ec8c Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Thu Aug 28 22:16:07 2008 +0300 [cairo-spans] New cairo_tor_scan_converter_t. Imports a new polygon scan converter implementation from the repository at http://cgit.freedesktop.org/~joonas/glitter-paths/ Glitter paths is a stand alone polygon rasteriser derived from David Turner's reimplementation of Tor Anderssons's 15x17 supersampling rasteriser from the Apparition graphics library. The main new feature in this implementation is cheaply choosing per-scan line between doing fully analytical coverage computation for an entire row at a time vs. using a supersampling approach. src/Makefile.sources | 1 + src/cairo-spans-private.h | 10 + src/cairo-spans.c | 18 +- src/cairo-tor-scan-converter.c | 2003 ++++++++++++++++++++++++++++++++++++++++ 4 files changed, 2027 insertions(+), 5 deletions(-) commit 2078e5b20fdff76ada6e13b29b2775b7dcd35439 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Thu Jul 31 01:58:08 2008 +0300 [cairo-spans] New _cairo_path_fixed_fill_using_spans(). Adds a helper function for backends to use for filling a path using spans. src/cairo-spans-private.h | 10 ++++ src/cairo-spans.c | 145 ++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 155 insertions(+) commit 4a9b274eebe674bbc5c66dc3e33256723cdf9829 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Fri Jul 11 00:59:47 2008 +0300 [cairo-spans] Add a check/create_span_renderer backend methods. A surface will have the chance to use span rendering at cairo_fill() time by creating a renderer for a specific combination of pattern/dst/op before the path is scan converted. The protocol is to first call check_span_renderer() to see if the surface wants to render with spans and then later call create_span_renderer() to create the renderer for real once the extents of the path are known. No backends have an implementation yet. src/cairo-analysis-surface.c | 4 +++ src/cairo-beos-surface.cpp | 2 ++ src/cairo-directfb-surface.c | 2 ++ src/cairo-glitz-surface.c | 2 ++ src/cairo-image-surface.c | 2 ++ src/cairo-meta-surface.c | 2 ++ src/cairo-os2-surface.c | 2 ++ src/cairo-paginated-surface.c | 2 ++ src/cairo-pdf-surface.c | 2 ++ src/cairo-ps-surface.c | 2 ++ src/cairo-quartz-image-surface.c | 2 ++ src/cairo-quartz-surface.c | 2 ++ src/cairo-script-surface.c | 2 ++ src/cairo-sdl-surface.c | 2 ++ src/cairo-surface.c | 53 ++++++++++++++++++++++++++++++++++++++ src/cairo-svg-surface.c | 2 ++ src/cairo-type3-glyph-surface.c | 2 ++ src/cairo-win32-printing-surface.c | 2 ++ src/cairo-win32-surface.c | 2 ++ src/cairo-xcb-surface.c | 2 ++ src/cairo-xlib-surface.c | 2 ++ src/cairoint.h | 30 +++++++++++++++++++++ src/test-fallback-surface.c | 2 ++ src/test-meta-surface.c | 2 ++ src/test-paginated-surface.c | 2 ++ 25 files changed, 131 insertions(+) commit 948c3526dcdbc440395fff4ce9bf4b7553930d92 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Fri Oct 24 17:37:30 2008 +0300 [cairo-spans] New abstract types for scan converting polygons. A cairo_span_renderer_t implementation can be provided by a surface if it wants to render paths as horizontal spans of the alpha component of a mask. Its job is to composite a source pattern to the destination surface when given spans of alpha coverage for a row while taking care of backend specific clipping. A cairo_scan_converter_t takes edges of a flattened path and generates spans for a span renderer to render. src/Makefile.sources | 2 + src/cairo-spans-private.h | 124 ++++++++++++++++++++++++ src/cairo-spans.c | 242 ++++++++++++++++++++++++++++++++++++++++++++++ src/cairoint.h | 1 + 4 files changed, 369 insertions(+) commit 4b227143b3daab75148cd54c9e7580d509864e0d Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Thu Jul 31 01:54:53 2008 +0300 [cairo-spans] Introduce a type to track which pixels combine in a compositing op. A cairo_composite_rectangles_t contains the coordinates of rectangular windows into each of the source pattern, mask, clip and destination surface containing the pixels that will combine in a compositing operation. The idea is to have a uniform way to represent all the translations involved rather than overloading parameters like src_x/y, dst_x/y, etc., sometimes with different incompatible meanings across functions. src/cairo-rectangle.c | 21 +++++++++++++++++++++ src/cairo-types-private.h | 24 ++++++++++++++++++++++++ src/cairoint.h | 7 +++++++ 3 files changed, 52 insertions(+) commit a370d077bc697588b6dac2556afa0b95ff83a77d Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Fri Jul 25 16:35:14 2008 +0300 [path-fixed] New _cairo_path_fixed_is_region(). We want to hit the current fast paths for rendering axis aligned rectilinear paths rather than spans, and for that we need to be able to identify regional paths. src/cairo-path-fixed.c | 96 ++++++++++++++++++++++++++++++++++++++++++++++++++ src/cairoint.h | 3 ++ 2 files changed, 99 insertions(+) commit 6acb8223930081f70b422ef93a49ea645c2cc12d Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Fri Sep 12 19:32:12 2008 +0300 [path-fixed] Avoid extra indirection when iterating already flat paths. Perform a plain iteration rather than a flattening one if the path knows it doesn't have any curves. src/cairo-path-fixed.c | 9 +++++++++ 1 file changed, 9 insertions(+) commit 0c0f4862c52d68776024f125b003ade455044b27 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Dec 6 13:32:37 2008 +0200 [perf-diff] Fix cairo-perf-diff for git 1.6 Since git 1.6 the plumbing commands aren't installed in the user's path by default. This patch fixes cairo-perf-diff to find the git-sh-setup command from git's lib dir. perf/cairo-perf-diff | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit afba0c312958852586b846ec615cff1bd3e5cde7 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Fri Sep 12 17:41:45 2008 +0300 [perf] Add perf tests to hit rectilinear code paths. These tests look at the differences in code paths hit by filling paths that are rectilinear (or not) and pixel aligned (or not) with the even-odd and non-zero fill rules. The paths are not simple, so they don't hit the special case quad/triangle tessellator. perf/Makefile.am | 3 +- perf/cairo-perf.c | 1 + perf/cairo-perf.h | 1 + perf/spiral.c | 200 ++++++++++++++++++++++++++++++++++++++++++++++++++++++ 4 files changed, 204 insertions(+), 1 deletion(-) commit 8ec58113df417e1ec1d331ab65267c34e6d32096 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Fri Sep 12 17:32:40 2008 +0300 [perf] Explicitly test rendering a path with lots of intersections. We don't have one just for this purpose. The only other path with many intersections that gets actually rendered is zrusin-another, but that might be sped up in the future (say by identifying collinearities up front or something like that.) perf/Makefile.am | 3 +- perf/cairo-perf.c | 1 + perf/cairo-perf.h | 1 + perf/intersections.c | 97 ++++++++++++++++++++++++++++++++++++++++++++++++++++ 4 files changed, 101 insertions(+), 1 deletion(-) commit 08b486c9aa1d6ad01f1b17a16dcb4d695e8cbf7d Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Thu Dec 4 17:53:06 2008 -0500 [win32] Use MOD instead of the '%' operator Repeat should be handled using MOD instead of '%' so that negative numbers are handled as expected. E.g. -1 mod 600 = 599, not 495 as the '%' operator gives. This was causing https://bugzilla.mozilla.org/show_bug.cgi?id=466258 Patch from Robert O'Callahan src/cairo-win32-surface.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) commit 540de34453d16092acd2978b513831a02f01f59f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Oct 26 12:53:29 2008 +0000 [matrix] Optimise invert for simple scaling|translation matrices. Peter Hercek reported, and provided a very useful test case for, a bug that caused his applications to crash with Cairo detecting an non-invertible pattern matrix and thus asserting the impossible happened. Bisecting revealed that the bug first appeared with 3c18d95 and disappeared with 0d0c6a1. Since neither of these explain the crash, further investigation revealed a compiler bug (gcc 4.3.3 20081130, earlier versions have different bugs!) that caused the matrix inversion to be invalid iff _cairo_matrix_scalar_multiply() was inlined (i.e. -O0, or an explicit noinline atttribute on that function prevented the bug, as did -msse.) So we apply this workaround to hide the bug in the stable series... The matrix is quite often just a simple scale and translate (or even identity!). For this class of matrix, we can skip the full adjoint rearrangement and determinant calculation and just compute the inverse directly. (cherry picked from commit 0d0c6a199c5b631299c72dce80d66ac0f4936a64) src/cairo-matrix.c | 26 +++++++++++++++++++++++++- 1 file changed, 25 insertions(+), 1 deletion(-) commit aff32019a70600fec5040a5a73c4f0b688826064 Author: Behdad Esfahbod <behdad@behdad.org> Date: Sun Nov 30 23:28:59 2008 -0500 [Makefile.am] Use top_srcdir instead of srcdir Doesn't make much difference here, but makes our intention clear. build/Makefile.am.changelog | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 51a75ad932af4a3c3da96ef98de79328afa200c3 Author: Luo Jinghua <sunmoon1997@gmail.com> Date: Sun Nov 30 20:42:49 2008 +0800 glitz: Replace specified color with an opaque one if dst surface don't have an alpha channel. Otherwise if underlying glitz drawable has an alpha channel, glitz_set_rectangles will set its alpha channel to specified value instead of opaque one and effects following composite operations since glitz draws to attached drawable then copies its content to the dst surface. With this commit, three test cases such as operator, operator-alpha and unbounded-operator passes now. src/cairo-glitz-surface.c | 18 ++++++++++++------ 1 file changed, 12 insertions(+), 6 deletions(-) commit 7247017cf5e6b497a5836d9081ee153d27c6b15e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Nov 29 11:47:25 2008 +0000 Fix up a couple of likely(malloc==NULL) Adrian Johnson spotted that I marked a few malloc failures as likely, whoops. src/cairo-cff-subset.c | 4 ++-- src/cairo-clip.c | 2 +- 2 files changed, 3 insertions(+), 3 deletions(-) commit 1659db2c1f4cc72e51fb43d119822ec7aaa77ac0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Nov 29 10:37:58 2008 +0000 [test] Fix surface leak from device-offset-positive. A forgotten cairo_surface_destroy() caused the similar surface to be leaked during the test. test/device-offset-positive.c | 1 + 1 file changed, 1 insertion(+) commit 0d757a793d7fe2860c6a522124a123864fc97f0a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Nov 29 10:36:34 2008 +0000 [png] Fix leak of original png data on error path. The error path was missing a _cairo_output_stream_destroy() to cleanup a copy of the incoming PNG data. src/cairo-png.c | 15 +++++++++++---- 1 file changed, 11 insertions(+), 4 deletions(-) commit e6963a5bfebda69a1ef0a986cede84bcd955b6d4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Nov 18 17:26:55 2008 +0000 Mark allocation failures as unlikely. Use the gcc likelihood annotation to indicate that allocation failures are extremely unlikely. src/cairo-analysis-surface.c | 4 +-- src/cairo-array.c | 4 +-- src/cairo-base85-stream.c | 2 +- src/cairo-bentley-ottmann.c | 13 ++++---- src/cairo-cache.c | 6 ++-- src/cairo-cff-subset.c | 66 ++++++++++++++++++++--------------------- src/cairo-clip.c | 8 ++--- src/cairo-deflate-stream.c | 2 +- src/cairo-directfb-surface.c | 2 +- src/cairo-font-face.c | 6 ++-- src/cairo-ft-font.c | 25 ++++++++-------- src/cairo-gstate.c | 17 ++++++----- src/cairo-hash.c | 6 ++-- src/cairo-hull.c | 2 +- src/cairo-image-surface.c | 12 ++++---- src/cairo-lzw.c | 4 +-- src/cairo-meta-surface.c | 24 +++++++-------- src/cairo-misc.c | 24 ++++++++++----- src/cairo-output-stream.c | 14 ++++----- src/cairo-paginated-surface.c | 2 +- src/cairo-path-fixed.c | 4 +-- src/cairo-path-stroke.c | 4 +-- src/cairo-path.c | 8 ++--- src/cairo-pattern.c | 36 +++++++++++----------- src/cairo-pdf-operators.c | 6 ++-- src/cairo-pdf-surface.c | 34 ++++++++++----------- src/cairo-pen.c | 8 ++--- src/cairo-png.c | 14 ++++----- src/cairo-polygon.c | 2 +- src/cairo-ps-surface.c | 16 +++++----- src/cairo-region.c | 2 +- src/cairo-scaled-font-subsets.c | 28 ++++++++--------- src/cairo-scaled-font.c | 29 +++++++++--------- src/cairo-script-surface.c | 14 ++++----- src/cairo-sdl-surface.c | 2 +- src/cairo-skiplist.c | 2 +- src/cairo-stroke-style.c | 2 +- src/cairo-surface-fallback.c | 2 +- src/cairo-surface.c | 4 +-- src/cairo-svg-surface.c | 16 +++++----- src/cairo-traps.c | 4 +-- src/cairo-truetype-subset.c | 30 +++++++++---------- src/cairo-type1-fallback.c | 15 +++++----- src/cairo-type1-subset.c | 22 +++++++------- src/cairo-type3-glyph-surface.c | 2 +- src/cairo-user-font.c | 2 +- src/cairo-xlib-display.c | 4 +-- src/cairo-xlib-screen.c | 2 +- src/cairo-xlib-surface.c | 25 ++++++++-------- src/cairo-xlib-visual.c | 2 +- src/cairo.c | 4 +-- src/test-fallback-surface.c | 2 +- src/test-meta-surface.c | 2 +- src/test-paginated-surface.c | 2 +- 54 files changed, 300 insertions(+), 294 deletions(-) commit d1801c23fae3777c7c59e084894a3410f7a1f932 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Nov 18 15:38:37 2008 +0000 Mark if(status) as being unlikely. The error paths should be hit very rarely during normal operation, so mark them as being unlikely so gcc may emit better code. src/cairo-analysis-surface.c | 24 +-- src/cairo-array.c | 6 +- src/cairo-cache.c | 4 +- src/cairo-cff-subset.c | 208 ++++++++++++------------- src/cairo-clip.c | 23 ++- src/cairo-font-face.c | 14 +- src/cairo-ft-font.c | 44 +++--- src/cairo-gstate.c | 66 ++++---- src/cairo-image-surface.c | 16 +- src/cairo-lzw.c | 4 +- src/cairo-meta-surface.c | 42 ++--- src/cairo-misc.c | 6 +- src/cairo-output-stream.c | 2 +- src/cairo-paginated-surface.c | 38 ++--- src/cairo-path-fill.c | 6 +- src/cairo-path-fixed.c | 14 +- src/cairo-path-stroke.c | 56 +++---- src/cairo-path.c | 6 +- src/cairo-pattern.c | 34 ++-- src/cairo-pdf-operators.c | 60 ++++---- src/cairo-pdf-surface.c | 334 ++++++++++++++++++++-------------------- src/cairo-pen.c | 6 +- src/cairo-png.c | 14 +- src/cairo-ps-surface.c | 122 +++++++-------- src/cairo-scaled-font-subsets.c | 30 ++-- src/cairo-scaled-font.c | 72 ++++----- src/cairo-script-surface.c | 160 +++++++++---------- src/cairo-sdl-surface.c | 4 +- src/cairo-surface-fallback.c | 82 +++++----- src/cairo-surface.c | 60 ++++---- src/cairo-svg-surface.c | 84 +++++----- src/cairo-traps.c | 12 +- src/cairo-truetype-subset.c | 98 ++++++------ src/cairo-type1-fallback.c | 44 +++--- src/cairo-type1-subset.c | 46 +++--- src/cairo-type3-glyph-surface.c | 10 +- src/cairo-user-font.c | 18 +-- src/cairo-xlib-screen.c | 4 +- src/cairo-xlib-surface.c | 92 +++++------ src/cairo.c | 138 ++++++++--------- 40 files changed, 1053 insertions(+), 1050 deletions(-) commit f0804d4856496a46d0b2270d5815856bf63b4cf8 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Nov 28 23:58:55 2008 +0000 [test] Exercise caps and joins under reflection. The nature of the joins depends critically upon whether the joint is clockwise or counter-clockwise, so extend the basic caps-joins test to exercise both conditions i.e. repeat the test under a reflection. test/Makefile.am | 6 ++- test/caps-joins-curve.c | 111 +++++++++++++++++++++++++++++++++++++++ test/caps-joins-curve.ps.ref.png | Bin 0 -> 3728 bytes test/caps-joins-curve.ref.png | Bin 0 -> 5132 bytes test/caps-joins.c | 40 +++++++++----- test/caps-joins.ps.ref.png | Bin 0 -> 2282 bytes test/caps-joins.ps2.ref.png | Bin 1459 -> 0 bytes test/caps-joins.ps3.ref.png | Bin 1459 -> 0 bytes test/caps-joins.ref.png | Bin 1488 -> 2380 bytes test/dash-curve.ref.png | Bin 39642 -> 39696 bytes 10 files changed, 143 insertions(+), 14 deletions(-) commit f39dd86e292e9af3438acb041a3a7330fb4a3b22 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Nov 27 00:23:11 2008 +0000 [test] Add simple cap test. Add a test case that only exercises capping, useful for developing new strokers. test/Makefile.am | 3 ++ test/caps.c | 87 +++++++++++++++++++++++++++++++++++++++++++++++++++ test/caps.ps.ref.png | Bin 0 -> 1466 bytes test/caps.ref.png | Bin 0 -> 1601 bytes 4 files changed, 90 insertions(+) commit f2982e103c27cad1edf648dadcfffe1a15763efd Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Nov 25 17:23:33 2008 +0000 [fill] Allow rectangles to be implicitly closed. Also scan for appendages of simple rectangles. src/cairo-path-fill.c | 2 +- src/cairo-path-fixed-private.h | 4 ++-- src/cairo-path-fixed.c | 14 +++++++++----- 3 files changed, 12 insertions(+), 8 deletions(-) commit d2bcf1d76defda861e1a7f2271046851083ad694 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Nov 28 17:12:03 2008 +0000 [path] Carefully check for no more points. As the empty path points to an embedded buf, we cannot rely on the buf pointer being NULL to mark end-of-path. src/cairo-path-fixed-private.h | 10 +++++----- src/cairo-path-fixed.c | 15 ++++++++++++--- 2 files changed, 17 insertions(+), 8 deletions(-) commit 424aba9be558e89b4d42554ca6a5042574e3df75 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Nov 26 20:38:24 2008 +0000 [check] Search for the word NOTE Add word boundary markers around the regexp so that we only match for the word NOTE and not on substrings like GL_NOTEQUAL. src/check-doc-syntax.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 36147140ce56d8d3492b971a9b6ceff97e212e01 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Nov 28 18:54:02 2008 +0000 [perf] Fix build If automake detects the use of cairo_perf_LDADD in an unused conditional that overrides the default - so we need to manually set cairo_perf_LDADD. perf/Makefile.am | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit ed6bd4766cb47e97b1a6ecb81eff5bf4673d549b Author: Luo Jinghua <sunmoon1997@gmail.com> Date: Fri Nov 28 07:54:41 2008 +0800 image: Use unsigned long to compute color masks. Use unsigned long in the first place to prevent compiler from expanding signed bit to all upper bits. e.g, a alpha mask 0xff0000 will expand to 0xffffffffff00000 on 64 bit platform which is not what we expected. src/cairo-image-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 41cb91d417e7c0dc1cc767ea0fb38a1f8b3f4e43 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Wed Nov 26 23:49:12 2008 +0200 [script] Use the in-tree libcairo for csi-replay. csi-replay was missing the in-tree libcairo.la so the build was failing while looking for new symbols from the system libcairo. util/cairo-script/Makefile.am | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit f5634818f1e5cc0adaa326662c099a75f4e615eb Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Nov 13 14:56:38 2008 +0000 [os2] Move include cairo.h before os2.h The defines need to come first, but we specify that cairo.h is the first header file to be included by files. src/cairo-os2.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 2f0f9a1a593db43dd04507c5989cd0af4b1486de Author: Behdad Esfahbod <behdad@behdad.org> Date: Thu Nov 13 10:50:41 2008 +0100 [os2] Fix includes Patch from Dave Yeo to make cairo-os2.h include os2.h directly so the header is standalone. src/cairo-os2-private.h | 11 ----------- src/cairo-os2-surface.c | 5 ----- src/cairo-os2.h | 8 ++++++++ 3 files changed, 8 insertions(+), 16 deletions(-) commit 5591cb58fd94483ec3dc7c169ae8cf0a6b7841bd Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Nov 26 17:43:22 2008 +0000 [script] Only use zlib if available. Conditionally link against zlib - replaying a script will fail if the trace uses a compressed format. configure.ac | 5 ++++- util/cairo-script/Makefile.am | 2 +- util/cairo-script/cairo-script-file.c | 5 ++++- util/cairo-script/cairo-script-operators.c | 2 ++ 4 files changed, 11 insertions(+), 3 deletions(-) commit 9432c0c9b89ed6b3d369b377cf884918cd6180db Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Nov 26 17:33:35 2008 +0000 [script] Link the interpreter against the built library. Ensure we correctly link against the version of cairo just built. util/cairo-script/Makefile.am | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 815e5b962ef8477c344da0f0b52c1f85c9854777 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Nov 26 17:30:29 2008 +0000 [script] Add a fallback image surface for replay If we do not have a native windowing surface compiled into the library, just replay the script to an image surface. util/cairo-script/csi-replay.c | 8 ++++++++ 1 file changed, 8 insertions(+) commit 5e376523628d5e2078e395ea7263a04bacd37c47 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Nov 26 17:26:57 2008 +0000 [skiplist] Check for ffs() Wine at least does not provide ffs(), so check at configure time and open-code a replacement. build/configure.ac.system | 2 +- src/cairo-skiplist.c | 11 +++++++++++ 2 files changed, 12 insertions(+), 1 deletion(-) commit 60282b866aab359840b4bcaa6aaccaca2eccd3d2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Nov 26 16:58:29 2008 +0000 [glitz] Fix clone_similar(). Clone similar open-coded various image surface functions and failed to clone a sub-region resulting in failures for mask-transformed-* and large-source. src/cairo-glitz-surface.c | 458 ++++++++++++++-------------------------------- 1 file changed, 137 insertions(+), 321 deletions(-) commit 47275c7ece06c0d02f2d7c9aa81d87a7144392b0 Author: Paolo Bonzini <bonzini@gnu.org> Date: Wed Nov 26 15:56:48 2008 +0100 [test] Fix glitz-surface-source test The recent changes to separate glitz/agl broke the compilation of the glitz-surface-source test. test/glitz-surface-source.c | 96 +++++++++++++++++++++++++++++++++- test/glitz-surface-source.ps2.ref.png | Bin 0 -> 376 bytes test/glitz-surface-source.ps3.ref.png | Bin 0 -> 376 bytes 3 files changed, 94 insertions(+), 2 deletions(-) commit a84ea7829065b7e1d33d6fed6bcf0c7c8abd383a Author: Paolo Bonzini <bonzini@gnu.org> Date: Wed Nov 26 15:49:00 2008 +0100 [test] Add quartz-surface-source test Add a simple test to exercise using a Quartz surface as a source. test/Makefile.am | 6 ++++- test/quartz-surface-source.c | 42 +++++++++++++++++++++++++++++++++ test/quartz-surface-source.ps2.ref.png | Bin 0 -> 376 bytes test/quartz-surface-source.ps3.ref.png | Bin 0 -> 376 bytes test/quartz-surface-source.ref.png | Bin 0 -> 332 bytes 5 files changed, 47 insertions(+), 1 deletion(-) commit 993941cfd701eb222b81cdddeabf6f7ff2daa150 Author: Paolo Bonzini <bonzini@gnu.org> Date: Wed Nov 26 13:32:11 2008 +0000 [perf] Fix SDL compilation for MacOS X The attached patch makes the SDL tests compile under Mac OS X. The problem is: 1) that <SDL_main.h> should be included in files that define the main function for SDL Mac OS X programs (this is not true with the upcoming SDL 1.3 release). 2) that -lSDLmain, because it is statically linked, needs the Cocoa framework in the LDADD of the main program. Again, 1.3 will not require this. perf/Makefile.am | 4 ++++ perf/cairo-perf.c | 4 ++++ test/Makefile.am | 4 ++++ test/cairo-test-runner.c | 4 ++++ 4 files changed, 16 insertions(+) commit 8a5b55ca6c69671a138f65ab15bcf93163f24a37 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Nov 26 13:26:40 2008 +0000 [matrix] Impose a maximum number of refinement iterations Ensure we do not loop forever trying to minimise the error between the pixman and cairo matrices - for instance when the FPU is not running at full precision. src/cairo-matrix.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) commit 4218699642c621eb3098a5251ef88d8c7d8a96d2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Nov 26 11:54:29 2008 +0000 [skiplist] Use ffs() Use ffs() [find-first-set-bit] instead of open-coding, which is about 25% faster on my test machine. src/cairo-skiplist.c | 11 +---------- 1 file changed, 1 insertion(+), 10 deletions(-) commit 039437c9cfaeed7a2cc1977623ab8bda9854b58a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Nov 26 10:31:09 2008 +0000 [png] Use FILE* instead of void*. Adrian Johnson reported that cygwin complained about the use of the void * within feof() as it was using a macro and attempted a to deference the void*... src/cairo-png.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) commit 8fa3cd95641e9138406d651b0734ee84f1da60d2 Author: Adrian Johnson <ajohnson@redneon.com> Date: Wed Nov 26 06:40:47 2008 +1030 Change uint to uint32_t src/cairo-image-info.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit e828f7330633cd839f094bf8409de0bcff17c95c Author: Luo Jinghua <sunmoon1997@gmail.com> Date: Wed Nov 26 00:09:11 2008 +0800 glitz: Apply fixup to unbounded operators to clear area outside of mask src/cairo-glitz-surface.c | 79 +++++++++++++++++++++++++++++++++++++++-------- 1 file changed, 66 insertions(+), 13 deletions(-) commit 449e7518bee5d0238138176b3b71d137c6fc01e7 Author: Luo Jinghua <sunmoon1997@gmail.com> Date: Tue Nov 25 22:55:24 2008 +0800 glitz: set correct parameters for linear and radial pattern. Glitz expects 16.16 fixed point, but we use 24.8 fixed point by default. src/cairo-glitz-surface.c | 20 ++++++++++---------- 1 file changed, 10 insertions(+), 10 deletions(-) commit 3bde440e00ba19e968b854c1505dc4d1ace83504 Author: Luo Jinghua <sunmoon1997@gmail.com> Date: Tue Nov 25 22:36:53 2008 +0800 glitz: fixup another stupid bug in fill_rectangles. Wny I doesn't notice them before. Without fixes it fills n_rect times of the first one in the rectangle list. src/cairo-glitz-surface.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) commit 4ac38f7c2bde67cab37805cab8a2effb2a8617e4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Nov 25 11:52:01 2008 +0000 [fill] Emit rectangles for GdkRegion Scan the path for a series of consistently wound rectangles. src/cairo-path-fill.c | 85 ++++++++++++++++++++++------ src/cairo-path-fixed-private.h | 17 ++++++ src/cairo-path-fixed.c | 124 ++++++++++++++++++++++++++++++++++++++--- 3 files changed, 201 insertions(+), 25 deletions(-) commit 23df74e5ffd6be876f3c19ee9d71683f5a0ed6f4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Nov 25 11:45:30 2008 +0000 [xlib] Cosmetic tweak. Tightly scope the local rects. src/cairo-xlib-surface.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) commit 06fabd6cbd0ad187f5f9f155d6b7962f76ec5dda Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Nov 25 12:05:26 2008 +0000 [path] Fix up extents. Forgot to round the box to the integer rectangle and missed why only testing on image. Very naughty. src/cairo-analysis-surface.c | 2 +- src/cairo-path-bounds.c | 20 +++++++++++--------- 2 files changed, 12 insertions(+), 10 deletions(-) commit 1f44fb97f3973aa90c4f27bcf9341149370c825d Author: Luo Jinghua <sunmoon1997@gmail.com> Date: Tue Nov 25 20:45:03 2008 +0800 cairo-script: add $(top_builddir)/src to CPPPATH. $(top_builddir)/src is needed to find cairo-features.h while building out of tree. util/cairo-script/Makefile.am | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 487c708b3b6727eb2f8d3714edeae356a7cbbd40 Author: Paolo Bonzini <bonzini@gnu.org> Date: Tue Nov 25 20:40:37 2008 +0800 [boilerplate] fix compilation for glitz-agl As suggested on the list, this splits the cairo-boilerplate-glitz.c file in three separate files, one for each backend. Furthermore, it fixes a few problems in compilation of the AGL backend test harness. boilerplate/Makefile.sources | 6 +- boilerplate/cairo-boilerplate-glitz-agl.c | 166 +++++++++ boilerplate/cairo-boilerplate-glitz-glx.c | 244 ++++++++++++ boilerplate/cairo-boilerplate-glitz-private.h | 6 + boilerplate/cairo-boilerplate-glitz-wgl.c | 164 ++++++++ boilerplate/cairo-boilerplate-glitz.c | 517 -------------------------- 6 files changed, 585 insertions(+), 518 deletions(-) commit ca80b8c652dde4449216da9d11691202eef97bbd Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Nov 25 10:25:24 2008 +0000 [pdf] Add a default case to silence the compiler. The foolish compiler was emitting a warning about a potential uninitialized variable even though the switch was fully populated. src/cairo-pdf-surface.c | 1 + 1 file changed, 1 insertion(+) commit b8991a1c69ae5d8fb630296a3c689aa8d1546671 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Nov 25 10:11:59 2008 +0000 [in-fill] Add the implicit close-path during move-to. When interpreting a fixed-path for a fill operation, any move-to implicitly closes the current path. src/cairo-gstate.c | 12 ++++++------ src/cairo-path-in-fill.c | 23 +++++++++++++---------- src/cairo.c | 11 ++++------- src/cairoint.h | 4 ++-- 4 files changed, 25 insertions(+), 25 deletions(-) commit 59de6fb89e80ee6aeeb2984b545ceb9bb9f0f7bb Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Nov 25 10:04:50 2008 +0000 [path] Compute approximate extents. When computing the bounds of the clip path, we care more for a fast result than absolute precision as the extents are only used as a guide to trim the future operations. So computing the extents of the path suffices. src/cairo-analysis-surface.c | 16 +++-------- src/cairo-clip.c | 25 +++--------------- src/cairo-gstate.c | 13 +++------ src/cairo-path-bounds.c | 63 ++++++++++++++++++++++++++++++++++++++++---- src/cairo-surface.c | 8 +++--- src/cairo.c | 10 +++---- src/cairoint.h | 8 ++++-- 7 files changed, 82 insertions(+), 61 deletions(-) commit b6bf047494fc308fff00d818b2920d8ba4aa7aed Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Nov 21 15:27:31 2008 +0000 [clip] Check for error surface The update to use a NULL backend with an error surface broke creating a context from an error surface. src/cairo-clip.c | 2 +- src/cairo.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) commit 721cad9b24c2ad049320950d231ed84046c5b8ab Author: Adrian Johnson <ajohnson@redneon.com> Date: Mon Nov 24 22:41:03 2008 +1030 Fix typo src/cairo-misc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 70e4c532722bbcad1eca50438e6ab2cdd0ea9b53 Author: Adrian Johnson <ajohnson@redneon.com> Date: Mon Nov 24 22:38:30 2008 +1030 Document location of image format specifications src/cairo-image-info.c | 15 +++++++++++++++ 1 file changed, 15 insertions(+) commit 4ca644af274bda4603e7d02dbeca8641bbb3ff79 Author: Adrian Johnson <ajohnson@redneon.com> Date: Mon Nov 24 00:29:54 2008 +1030 Win32-print: Add PNG embedding support src/cairo-win32-printing-surface.c | 106 +++++++++++++++++++++++++++---------- src/cairo-win32-private.h | 3 ++ 2 files changed, 82 insertions(+), 27 deletions(-) commit f59a3e03fc91dc4a1769b07bda77a364cc570bc8 Author: Adrian Johnson <ajohnson@redneon.com> Date: Sun Nov 23 22:27:49 2008 +1030 Add PNG get info function src/cairo-image-info-private.h | 5 +++++ src/cairo-image-info.c | 46 ++++++++++++++++++++++++++++++++++++------ 2 files changed, 45 insertions(+), 6 deletions(-) commit 1892907e24d9f9f31942c6962aaa6d2ff9553ce7 Author: Adrian Johnson <ajohnson@redneon.com> Date: Sun Nov 23 21:36:40 2008 +1030 PDF: Disable PDF 1.5 features when version 1.4 is selected JPEG 2000 image embedding and ActualText marked content are PDF 1.5 features. src/cairo-pdf-operators-private.h | 5 +++++ src/cairo-pdf-operators.c | 8 ++++++++ src/cairo-pdf-surface.c | 7 +++++++ 3 files changed, 20 insertions(+) commit c4a57385fa2e69020f43519bea21b98182bf5fd7 Author: Adrian Johnson <ajohnson@redneon.com> Date: Sun Nov 23 21:12:08 2008 +1030 PDF: Add cairo_pdf_surface_restrict_to_version API Now that we are using PDF 1.5 features, add an api to select between version 1.4 or 1.5. src/cairo-pdf-surface-private.h | 2 + src/cairo-pdf-surface.c | 116 +++++++++++++++++++++++++++++++++++++--- src/cairo-pdf.h | 24 +++++++++ 3 files changed, 136 insertions(+), 6 deletions(-) commit d85e836911d661275c29f2348a047e5d911f9004 Author: Adrian Johnson <ajohnson@redneon.com> Date: Sun Nov 23 21:03:22 2008 +1030 PDF: Add newline to end of binary streams PDF requires white space before the "endstream" src/cairo-pdf-surface.c | 2 ++ 1 file changed, 2 insertions(+) commit b87d81ef0bb52570385a3c9e331651cbeb87b2bb Author: Adrian Johnson <ajohnson@redneon.com> Date: Sun Nov 23 19:29:26 2008 +1030 Add image/jp2 to mime-data test test/jp2.jp2 | Bin 0 -> 2999 bytes test/mime-data.c | 7 ++++++- test/mime-data.pdf.ref.png | Bin 4466 -> 6482 bytes test/mime-data.ps.ref.png | Bin 4466 -> 4554 bytes test/mime-data.ref.png | Bin 155 -> 185 bytes test/mime-data.svg.ref.png | Bin 6153 -> 6264 bytes 6 files changed, 6 insertions(+), 1 deletion(-) commit 7fdb712cbaadc95c02f607a9f6c995f8beb01342 Author: Adrian Johnson <ajohnson@redneon.com> Date: Sun Nov 23 18:56:12 2008 +1030 PDF: Add JPEG2000 image embedding Requires increasing the PDF version to 1.5. src/cairo-pdf-surface.c | 52 ++++++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 51 insertions(+), 1 deletion(-) commit 0746efbf0718095920d9258942d32a7023d22131 Author: Adrian Johnson <ajohnson@redneon.com> Date: Sun Nov 23 18:47:13 2008 +1030 Add JPEG2000 mimetype and image info function src/cairo-image-info-private.h | 4 ++ src/cairo-image-info.c | 100 +++++++++++++++++++++++++++++++++++++++++ src/cairo-surface-fallback.c | 1 + src/cairo.h | 1 + 4 files changed, 106 insertions(+) commit 5de1e4de938d03406ce3364c6c1baa958f210410 Author: Adrian Johnson <ajohnson@redneon.com> Date: Sun Nov 23 18:33:58 2008 +1030 Move cairo-jpeg-info.c to cairo-image-info.c Other image formats will be added to the same file. src/Makefile.sources | 4 ++-- ...ro-jpeg-info-private.h => cairo-image-info-private.h} | 16 ++++++++-------- src/{cairo-jpeg-info.c => cairo-image-info.c} | 16 ++++++++-------- src/cairo-pdf-surface.c | 6 +++--- src/cairo-ps-surface.c | 6 +++--- src/cairo-svg-surface.c | 6 +++--- src/cairo-win32-printing-surface.c | 6 +++--- 7 files changed, 30 insertions(+), 30 deletions(-) commit 63180edf6f7cc81beda626190ca1055efa330fda Author: Luo Jinghua <sunmoon1997@gmail.com> Date: Sun Nov 23 11:42:26 2008 +0800 glitz: set clone_offset_{x, y} on success in clone_similar. With this fix, glitz backend passes 135 test cases and fails 66 test cases. src/cairo-glitz-surface.c | 2 ++ 1 file changed, 2 insertions(+) commit 36c1b3e5919f119a054e425ca03f7ce810d3e7ec Author: Luo Jinghua <sunmoon1997@gmail.com> Date: Sun Nov 23 10:27:49 2008 +0800 glitz: fixup a stupid bug in get_image. _pixman_format_from_masks returns a boolean instead of cairo status code. Without this fix, get_image bails out over and over again even operations was completed successfully. :-( src/cairo-glitz-surface.c | 14 +++++++++----- 1 file changed, 9 insertions(+), 5 deletions(-) commit c8b70aacc6b5523e2f73bebf4f61e876de1c3b3f Author: Nicolas Bruguier <gandalfn@club-internet.fr> Date: Sat Nov 22 18:16:17 2008 +0800 glitz: bring glitz backend to more or less working state. Thanks Nicolas Bruguier for writing the original patch and thanks Paolo Bonzini for pointing out. This patch converts 24.8 fixed point values into 16.16 ones before passing them down to glitz. And I fixed a use-after-free issue in _cairo_glitz_surface_set_clip_region. glitz_surface_set_clip_region takes the pointer and doesn't copy the clip boxes, the original code frees the clip boxes immediately after setting clip region. Keeping the box around with the life time of clip region fixes the bug. src/cairo-glitz-surface.c | 329 +++++++++++++++++++++++----------------------- 1 file changed, 163 insertions(+), 166 deletions(-) commit 77e60df32fa59328bd32095c77f8c360805b5db7 Author: Maarten Maathuis <madman2003@gmail.com> Date: Sat Nov 8 15:29:40 2008 +0100 cairo-xcb: avoid leaking memory src/cairo-xcb-surface.c | 3 +++ 1 file changed, 3 insertions(+) commit 361962b7fb4385f164673991670b6fce0839d32a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 31 08:42:30 2008 +0000 [directfb] Compile fix with debug enabled. Fix a trivial compile failure reported here: Bug 18322 - bug in _cairo_directfb_surface_release_source_image function (http://bugs.freedesktop.org/show_bug.cgi?id=18322) src/cairo-directfb-surface.c | 1 + 1 file changed, 1 insertion(+) commit f15b1f26becf28457e9ccf8903257a0dec25d547 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Nov 20 23:19:19 2008 +0000 [region] Use the caller supplied array for extracting boxes. Allow the user to pass in a pre-allocated array and use it if the number of boxes permits. This eliminates the frequent allocations during clipping by toolkits. src/cairo-directfb-surface.c | 6 ++++-- src/cairo-glitz-surface.c | 1 + src/cairo-paginated-surface.c | 3 +++ src/cairo-region.c | 11 ++++++----- src/cairo-sdl-surface.c | 1 + src/cairo-surface.c | 15 ++++++++------- src/cairo-win32-surface.c | 3 ++- src/cairo-xcb-surface.c | 1 + src/cairo-xlib-surface-private.h | 2 +- src/cairo-xlib-surface.c | 13 ++++++++----- 10 files changed, 35 insertions(+), 21 deletions(-) commit 9d2c55c661885c06eed44e810004c2ebe07038d3 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Nov 20 21:53:34 2008 +0000 [traps] Embed sufficient traps to accommodate a stroke rectangle Small numbers of traps are frequently generated from either a path of a single line, or that of a rectangle. Therefore should embed sufficient storage with cairo_traps_t to accommodate the stroke of a single rectangle (e.g. a pango unknown glyph hexbox) to avoid frequent allocations. src/cairoint.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit a11442154d958d5c9694eb944a715a6c689fda85 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Nov 20 18:00:40 2008 +0000 [cairo] Allocate glyphs on the stack for show_text(). First try to use a stack buffer for the glyphs and clusters if the user calls cairo_show_text() - for example, as pango does to draw the unknown hex box. src/cairo.c | 18 ++++++++++++++---- 1 file changed, 14 insertions(+), 4 deletions(-) commit 9556266ffcb1c34187730af90bb847950d6db66e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Nov 20 09:40:25 2008 +0000 [surface] Make the error surfaces have a NULL backend. By using a NULL backend for the error surfaces, instead of a pointer to the image surface backend end, we save a few lookup/redirections during dynamic linking. src/cairo-surface.c | 22 ++++++---------------- 1 file changed, 6 insertions(+), 16 deletions(-) commit cfd2c73826f5bf20624fbdf5b16fd08fbe18a914 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Nov 20 09:36:02 2008 +0000 [image] Trim image surface size by a few bytes Convert an infrequently accessed series of cairo_bool_t and short enums to a common bitfield. src/cairo-image-surface.c | 11 ++++++----- src/cairoint.h | 7 ++++--- 2 files changed, 10 insertions(+), 8 deletions(-) commit 1f48b36933b5ff082edf3e221563c15c3bf16b75 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Nov 20 12:41:18 2008 +0000 [script] Build fix for ! HAS_FT_FONT Correct the macro definition used when compiling without FreeType. util/cairo-script/cairo-script-operators.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) commit 0d5e533b6cc43c4ebc030140933247cf160d47cd Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Nov 19 17:03:13 2008 +0000 [trace] Correctly push font-face onto the operand stack. We tried to push a reference to an undefined font-face, now we just copy from the operand stack. util/cairo-trace/trace.c | 61 ++++++++++++++++++++++++++++-------------------- 1 file changed, 36 insertions(+), 25 deletions(-) commit b426346b482e2687081433213ae107cb7b2ebe58 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Nov 19 16:04:57 2008 +0000 [trace] Construct matrices directly Avoid building a temporary array by constructing the matrix directly. util/cairo-trace/trace.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) commit 727ff2bd7260f26c9390079258c43e6890ef2284 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Nov 19 14:25:20 2008 +0000 [xlib] Explicitly track whether the GC has clip_rects. Simplify the logic and remove some pessimism from the code by explicitly tracking when we set clip rectangles on the GC. src/cairo-xlib-surface-private.h | 1 + src/cairo-xlib-surface.c | 6 +++--- 2 files changed, 4 insertions(+), 3 deletions(-) commit 63a86a470fb88ee876414164e7e26789c1065b82 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Nov 19 14:04:01 2008 +0000 [surface] Replay meta surfaces to a similar surface. When cloning a meta-surface, first attempt to replay it to a similar surface rather than a full-sized image buffer. src/cairo-surface.c | 30 ++++++++++++++++++++++++++++-- 1 file changed, 28 insertions(+), 2 deletions(-) commit 7657bda0172f823fba61db4f66e0166c7619fbd2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Dec 18 11:23:33 2007 +0000 [perf/pythagoras_tree] Another fractal. Test lots of rectangles and recursion path construction. perf/Makefile.am | 3 +- perf/cairo-perf.c | 1 + perf/cairo-perf.h | 1 + perf/pythagoras-tree.c | 86 ++++++++++++++++++++++++++++++++++++++++++++++++++ 4 files changed, 90 insertions(+), 1 deletion(-) commit abd0a2627d198eddb628bbc1f2974435a74655a2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Dec 18 10:49:23 2007 +0000 [perf/dragon] Add a dragon curve perf case. Inspired by http://labs.trolltech.com/blogs/2007/08/31/rasterizing-dragons/ and http://en.wikipedia.org/wiki/Dragon_curve, add a performance test case to measure drawing this space-filling fractal curve. perf/Makefile.am | 3 +- perf/cairo-perf.c | 1 + perf/cairo-perf.h | 1 + perf/dragon.c | 166 ++++++++++++++++++++++++++++++++++++++++++++++++++++++ 4 files changed, 170 insertions(+), 1 deletion(-) commit 81ef772aa0bf512ec5ad8752da160117498bdb30 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Nov 19 11:00:35 2008 +0000 [xlib] Mark GC as dirty if we have an outstanding clip. Ginn Chen reported a regression with Firefox where "the whole area of web page is transparent until it redraws", and bisected it to the change to lazily clear the clip. The bug would appears to be when we have an inconsistent GC clip - i.e. the clip on the surface has been cleared, but we have not yet used and thus cleared the GC, so that we did not mark the GC as having a clip set when we freed it. src/cairo-xlib-surface.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) commit c41b99268dd2424d09ab12ca560d5db30b6b6faf Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Nov 19 11:49:04 2008 +0000 Conditionally include byteswap.h Fixup compilation by copying the checks from cairo-wideint-private.h to conditionally include byteswap.h and provide fallback implementations. util/cairo-script/cairo-script-interpreter.c | 1 - util/cairo-script/cairo-script-operators.c | 1 - util/cairo-script/cairo-script-private.h | 17 +++++++++++++++++ util/cairo-script/cairo-script-scanner.c | 1 - util/cairo-trace/trace.c | 17 ++++++++++++++++- 5 files changed, 33 insertions(+), 4 deletions(-) commit 8345fedbe4d4d003c1f26a78ac7c512c04d04173 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Nov 19 11:44:42 2008 +0000 [spline] Fix compile. Do not return the result of a void function. gcc chose to not warn about this when removing the return parameter... src/cairo-spline.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) commit b6c371a47f33ec10d4d6130cc15677761df2bdfd Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Nov 19 08:44:24 2008 +0000 [pattern] Cosmetic. surface is equal to pattern->surface at this point. src/cairo-pattern.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 7894abbe6d43b10ab2a92d99bdd6a08878e9022c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Nov 19 08:37:26 2008 +0000 [test] Support foreground only execution. Add an option to prevent forking - which makes it difficult to valgrind/gdb individual tests. test/Makefile.am | 1 + test/cairo-test-runner.c | 120 ++++++++++++++++++++++++++--------------------- 2 files changed, 67 insertions(+), 54 deletions(-) commit 645df0c6d2a3d9999bb18ddb8bb9995b3d597554 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Nov 18 16:37:59 2008 +0000 [scaled-font] Clean-up compiler warning. gcc warns that the status may be used uninitialized, so fix it. src/cairo-scaled-font-subsets.c | 20 ++++++++++---------- 1 file changed, 10 insertions(+), 10 deletions(-) commit 2fdee490745a6c3a75691907aadf8ae38c57234c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Nov 17 12:22:39 2008 +0000 [compiler] likelihood macros Behdad prefers these to be upper-case to be consistent with G_UNLIKELY and friends. However, as I intend to use these for nearly all instances of if(status), I suggest that we keep to the short and not so loud: if (unlikely (status)) return status; src/cairo-compiler-private.h | 8 ++++---- src/cairo-hash.c | 2 +- 2 files changed, 5 insertions(+), 5 deletions(-) commit 97edc680c189205ac2f4e150009f1f1cbe55ba1a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Nov 18 10:49:27 2008 +0000 [twin] Reduce tolerance. As the glyphs are rendered to cache, ensure that they are rendered at the highest quality settings. src/cairo-font-face-twin.c | 1 + src/cairo.c | 1 + src/cairoint.h | 1 + test/Makefile.am | 1 + test/twin.pdf.ref.png | Bin 0 -> 1673 bytes test/twin.ref.png | Bin 1673 -> 1718 bytes 6 files changed, 4 insertions(+) commit e50538863a2c063eba61b36cc08eff6eeb712956 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Nov 18 10:45:19 2008 +0000 [twin] Tweak line width. Slightly increase line width to eliminate internal holes in the characters where the strokes were not quite overlapping. src/cairo-font-face-twin.c | 2 +- test/twin.c | 2 +- test/twin.ps2.ref.png | Bin 1167 -> 1095 bytes test/twin.ps3.ref.png | Bin 1167 -> 1095 bytes test/twin.ref.png | Bin 1836 -> 1673 bytes test/twin.svg11.ref.png | Bin 1773 -> 1662 bytes test/twin.svg12.ref.png | Bin 1773 -> 1662 bytes 7 files changed, 2 insertions(+), 2 deletions(-) commit 91c17d33324b51a3876bc6ce778c684111139303 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Nov 18 00:32:54 2008 +0000 [trace] Simple unbounded cache for symbol lookups. Reparsing the dwarf info for every lookup is very slow, so cache the symbol lookups. This initial implementation is unbounded in the simple belief that the actual number of unique lookups during a program's lifetime should be fairly small. (Extending to a bounded MRU list is left as an exercise for the reader.) util/cairo-trace/lookup-symbol.c | 70 ++++++++++++++++++++++++++++++++++------ util/cairo-trace/trace.c | 9 +++--- 2 files changed, 65 insertions(+), 14 deletions(-) commit 2839a0e800d6cc12b28da44f30a9e278ceac65aa Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Nov 16 20:04:55 2008 +0000 [script] Add a simple replay. A very simple replay program for trace replay. util/cairo-script/.gitignore | 1 + util/cairo-script/Makefile.am | 7 +++- util/cairo-script/csi-replay.c | 85 ++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 92 insertions(+), 1 deletion(-) commit b091c5bfcad170fd563699a1c8b11e7d25d159a2 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Tue Nov 18 00:20:38 2008 +0200 BIBLIOGRAPHY: Cite Hars & Petruska's Pseudorandom Recursions paper. BIBLIOGRAPHY | 15 +++++++++++++++ 1 file changed, 15 insertions(+) commit 3a82f943988413e0d3fada9932bb385f4cbdb3c7 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Nov 16 18:21:52 2008 +0000 [test] Add util/cairo-script to include path. any2ppm needs the include path for cairo-script-interpreter so that it can be built without an existing install. test/Makefile.am | 3 ++- test/any2ppm.c | 3 --- 2 files changed, 2 insertions(+), 4 deletions(-) commit 997fa6a7729b72b25d9275929bf190e71eed1b9d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Nov 16 17:43:40 2008 +0000 [configure] Remove old line for CairoScript testing. CairoScript testing should now be unconditional so remove the surplus message. build/configure.ac.features | 1 - 1 file changed, 1 deletion(-) commit a2eff7c7deb9b41c387472bb91b9a7b9c8f319f0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Nov 16 16:19:25 2008 +0000 [test] Add build rule for libcairo-script-interpreter.la Add a rule to build the cairo-script-interpreter on behalf of any2ppm. test/Makefile.am | 3 +++ 1 file changed, 3 insertions(+) commit bf309aab6072ee1004073e71bccdf3c8040a8ad6 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Nov 15 11:36:41 2008 +0000 [configure] Delete CAN_TEST_SCRIPT A CairoScript interpreter is built under utils and so is always available. configure.ac | 3 --- test/any2ppm.c | 8 -------- 2 files changed, 11 deletions(-) commit d7873eecc598a558a2a862add8e9b056c4a23a4a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Nov 14 17:18:47 2008 +0000 [spline] Eliminate intermediate allocations during spline decomposition. The spline decomposition code allocates and stores points in a temporary buffer which is immediately consumed by the caller. If the caller supplies a callback that handles each point computed along the spline, then we can use the point immediately and avoid the allocation. src/cairo-bentley-ottmann.c | 12 +-- src/cairo-path-fill.c | 23 ++-- src/cairo-path-fixed.c | 27 ++--- src/cairo-path-in-fill.c | 21 ++-- src/cairo-path-stroke.c | 101 +++++++++--------- src/cairo-pen.c | 254 +++++++++++++++++++++++++++++--------------- src/cairo-polygon.c | 23 ++-- src/cairo-spline.c | 121 ++++----------------- src/cairo-types-private.h | 13 ++- src/cairoint.h | 65 ++++++++---- 10 files changed, 331 insertions(+), 329 deletions(-) commit 3bf8379408ce9b1e08d130bcd1076766e36f1bef Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Nov 14 09:50:29 2008 +0000 [type1] Propagate fatal NO_MEMORY erro from FreeType. If FreeType fails to load the glyph, check for a fatal error before falling back (and effectively masking the fatal condition). src/cairo-type1-subset.c | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 deletions(-) commit db9ed77d8aa4f4b4c8410e52e5bbd16ffa94ee24 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Nov 13 20:36:14 2008 +0000 [cairo] FLush surface on cairo_destroy() When discussing the implications of snapshot cow, one of the questions that we raised was what happens on cairo_destroy()? The lifetime of the context implicitly marks the extents of the drawing operations, typically the expose event (or perhaps one phase of it). Therefore at the end of the sequence we implicitly wish to flush the graphics events to the surface. src/cairo.c | 11 +++++++++++ 1 file changed, 11 insertions(+) commit 032be9849dfc32ec8802a4f086619755a3a397f5 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Nov 13 16:36:01 2008 +0000 [skiplist] Inline testing. Add a loop to create a skip list to allow manual checking. src/Makefile.am | 5 ++++- src/cairo-skiplist.c | 36 ++++++++++++++++++++++++++++++++++++ 2 files changed, 40 insertions(+), 1 deletion(-) commit e44c1f26e48cfb2f74834b6a8ef8532b0bd28982 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Sat Oct 18 01:15:44 2008 +0300 Use a smaller and faster random number generator. The previous LFSR generator used a large table to be speedy. The replacement generator is the single stage generator (1) from section 5.4 of [HP2007], requires no tables or extra state. I've run the generator through Rob Brown's extended Dieharder test suite and it passes all but one self-proclaimed buggy test (-d2) and the 2D sphere minimum distance test (-d12). [HP2007] Hars L. and Petruska G., ``Pseudorandom Recursions: Small and Fast Pseurodandom Number Generators for Embedded Applications'', Hindawi Publishing Corporation EURASIP Journal on Embedded Systems Volume 2007, Article ID 98417, 13 pages doi:10.1155/2007/98417 http://www.hindawi.com/getarticle.aspx?doi=10.1155/2007/98417&e=cta src/cairo-skiplist.c | 200 +++------------------------------------------------ 1 file changed, 11 insertions(+), 189 deletions(-) commit a4c5371b5e6d0df71efc75fee6f6a8fe8c6d3488 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Nov 13 14:56:38 2008 +0000 [os2] Move include cairo.h before os2.h The defines need to come first, but we specify that cairo.h is the first header file to be included by files. src/cairo-os2.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 456252a1c2b00685bee18c01bf27738d6125b4f8 Merge: 3b11997a2 ab61544b3 Author: Carl Worth <cworth@cworth.org> Date: Fri Nov 14 15:27:34 2008 +0100 Merge branch '1.8' Conflicts: NEWS cairo-version.h commit ab61544b30406a49abc1f559d81129edce6d41ec Author: Carl Worth <cworth@cworth.org> Date: Fri Nov 14 13:17:01 2008 +0100 Increment version to 1.8.5 after the 1.8.4 release. cairo-version.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 66e8f142e381501d114888c2d1fc1e7f6d6a9857 Author: Carl Worth <cworth@cworth.org> Date: Fri Nov 14 11:08:40 2008 +0100 Increment version to 1.8.4 cairo-version.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit b9c1344ad82d38bb86fa85fbb565b2de64aa0e35 Author: Carl Worth <cworth@cworth.org> Date: Fri Nov 14 11:07:10 2008 +0100 NEWS: Add notes for cairo 1.8.4 Just a few bug fixes here. NEWS | 52 ++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 52 insertions(+) commit 3b11997a23baa9e38e87b584ee29f4888a5d92f7 Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Thu Nov 13 18:01:47 2008 +0200 [cairo-script] Remove outdated #error to allow compilation to succeed. The interpreter doesn't depend on the script surface anymore. util/cairo-script/cairo-script-interpreter.h | 6 ------ 1 file changed, 6 deletions(-) commit d654d528de16769932131da80f4b925151d50103 Author: Carl Worth <cworth@cworth.org> Date: Thu Nov 13 16:45:15 2008 +0100 Blacklist "X.Org" < 7.0 with the buggy_repeat workaround. This is in response to a report that a 6.9 server crashes with cairo's extend-reflect test: https://bugs.freedesktop.org/show_bug.cgi?id=15628#c2 src/cairo-xlib-display.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) commit 35a1ba0ddc45f9cc785c6f8a37b6bb49c953e047 Author: Carl Worth <cworth@cworth.org> Date: Thu Nov 13 16:36:20 2008 +0100 Document the buggy_repeat workaround more carefully. It's a confusing condition, so let's be explicit about where the various numbers come from. src/cairo-xlib-display.c | 54 ++++++++++++++++++++++++++++++++++-------------- 1 file changed, 38 insertions(+), 16 deletions(-) commit cdfffc7420e005b2a7d1979feef8bd304183126c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Nov 13 11:07:45 2008 +0000 Add CairoScript interpreter Add a CairoScript interpreter library and use it to replay the test output for the CairoScript backend. The library is also used by the currently standalone Sphinx debugger [git://anongit.freedesktop.org/~ickle/sphinx]. The syntax/operator semantics are not yet finalized, but are expected to mature before the next stable release. configure.ac | 20 +- test/Makefile.am | 4 +- test/any2ppm.c | 41 +- util/Makefile.am | 2 +- util/cairo-script/COPYING | 17 + util/cairo-script/Makefile.am | 21 + util/cairo-script/cairo-script-file.c | 1018 +++++ util/cairo-script/cairo-script-hash.c | 448 ++ util/cairo-script/cairo-script-interpreter.c | 473 +++ util/cairo-script/cairo-script-interpreter.h | 104 + util/cairo-script/cairo-script-objects.c | 666 +++ util/cairo-script/cairo-script-operators.c | 5791 ++++++++++++++++++++++++++ util/cairo-script/cairo-script-private.h | 853 ++++ util/cairo-script/cairo-script-scanner.c | 1180 ++++++ util/cairo-script/cairo-script-stack.c | 196 + 15 files changed, 10801 insertions(+), 33 deletions(-) commit a856371bef496da0e84226f4fd2fc3cb72e955ac Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Nov 4 10:45:34 2008 +0000 Add CairoScript backend. A new meta-surface backend for serialising drawing operations to a CairoScript file. The principal use (as currently envisaged) is to provide a round-trip testing mechanism for CairoScript - i.e. we can generate script files for every test in the suite and check that we can replay them with perfect fidelity. (Obviously this does not provide complete coverage of CairoScript's syntax, but should give reasonable coverage over the operators.) boilerplate/Makefile.sources | 3 + boilerplate/cairo-boilerplate-script-private.h | 57 + boilerplate/cairo-boilerplate-script.c | 125 ++ boilerplate/cairo-boilerplate.c | 16 + build/configure.ac.features | 2 + configure.ac | 16 + doc/public/tmpl/cairo-surface.sgml | 1 + src/Makefile.sources | 3 + src/cairo-base85-stream.c | 1 + src/cairo-cache-private.h | 2 +- src/cairo-cache.c | 41 +- src/cairo-deflate-stream.c | 1 + src/cairo-ft-font.c | 12 + src/cairo-ft-private.h | 3 + src/cairo-gstate.c | 3 +- src/cairo-output-stream-private.h | 19 +- src/cairo-output-stream.c | 43 +- src/cairo-path-fixed-private.h | 11 + src/cairo-path-fixed.c | 164 ++ src/cairo-pattern.c | 255 +++ src/cairo-pdf-operators.c | 1 + src/cairo-ps-surface.c | 2 + src/cairo-scaled-font-private.h | 9 +- src/cairo-scaled-font.c | 5 + src/cairo-script-surface.c | 2598 ++++++++++++++++++++++++ src/cairo-script.h | 74 + src/cairo-types-private.h | 1 + src/cairo.h | 4 +- src/cairoint.h | 18 + test/Makefile.am | 4 +- test/any2ppm.c | 87 +- test/mime-data.script.ref.png | Bin 0 -> 1982 bytes 32 files changed, 3540 insertions(+), 41 deletions(-) commit 47a56e08501ec9375f75c15e35a68c77b313ada4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Sep 10 16:51:52 2008 +0100 [type1] Bind the RD procedure As the RD procedure may be called several hundred times for a full font, it is worth binding. src/cairo-type1-fallback.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 0769d39d0093ec3bff5b72ecbfcef1bae2f03be2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Nov 13 11:13:22 2008 +0000 Add predicate likelihood macros. Define the couple of standard macros that we can use to guide gcc optimisations of which code path will be most likely taken. src/cairo-compiler-private.h | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) commit b06c50cc54140a662adacde6336a3e7777f26ac3 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Nov 13 11:21:26 2008 +0000 [cairoint.h] Remove unused PLT entries. Trim the slim_hidden_proto/def to match the currently used functions. src/cairo-pattern.c | 1 - src/cairo.c | 3 --- src/cairoint.h | 16 ++++++---------- 3 files changed, 6 insertions(+), 14 deletions(-) commit 20fc2a2f92e27082a8806b9acb4b904dc84da672 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Nov 11 11:35:14 2008 +0000 [in-fill] Fixup boundary insideness testing Benjamin Otte noticed that the top-left vertex was now included within the fill extents. Having updated the test case to check the insideness conditions as defined by flash, fix the edge handling to match. src/cairo-path-in-fill.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 9afad537301529a5a4b4ddf8b7e4425002ea39c7 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Nov 11 10:22:17 2008 +0000 [test] Expand in-fill-trapezoid Test vertices and edges of a rectangle to probe 'corner-cases', using the insideness conditions from flash. test/in-fill-trapezoid.c | 34 ++++++++++++++++++++++++++++++++++ 1 file changed, 34 insertions(+) commit cebc84f367a81eedebf7ab0b6b082691923c3ef7 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Nov 8 00:06:38 2008 +0000 [hash] Separate out unique patterns of iterating over the table. Avoid unnecessary conditionals for the hotpaths by separating out the iteration over the elements into their distinct modes. src/cairo-font-face.c | 2 +- src/cairo-hash.c | 185 ++++++++++++++++++++++-------------------------- src/cairo-scaled-font.c | 11 +-- 3 files changed, 93 insertions(+), 105 deletions(-) commit 5f0aa274459fa182d1f82d393224c46ce2b12785 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Nov 13 10:45:15 2008 +0000 [trace] Add caller line info. Use lookup_symbol() on the return address to identify the caller and emit useful debug info to the trace. util/cairo-trace/cairo-trace.in | 27 +++++++- util/cairo-trace/lookup-symbol.c | 72 ++++++++++----------- util/cairo-trace/trace.c | 134 +++++++++++++++++++++++++++++++++++---- 3 files changed, 178 insertions(+), 55 deletions(-) commit 4cd6b0c0b9862f398877cb23c294399b480ac8d0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Nov 12 00:37:40 2008 +0000 [trace] Check for an undefined pattern and use index instead. util/cairo-trace/trace.c | 20 ++++++++++++++++++-- 1 file changed, 18 insertions(+), 2 deletions(-) commit 76f249a49399744d6d936e36ca4d68ab4a1f3d3b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Nov 11 09:12:50 2008 +0000 [trace] Detect and report broken traces. We cannot trace the cairo-script-interpreter as that uses Cairo's private symbols. util/cairo-trace/trace.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-) commit 01e2021b2b7b66e7cdb9c4b0326fd5793759848b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Nov 10 16:15:28 2008 +0000 [trace] Trace mark-dirty Capture the image data on mark dirty and record in the trace. util/cairo-trace/cairo-trace.in | 2 +- util/cairo-trace/trace.c | 121 ++++++++++++++++++++++++++++++++++------ 2 files changed, 106 insertions(+), 17 deletions(-) commit 8f3d0b55f50168f061a01eecfcb4ebecbff0e1c6 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Nov 11 09:25:24 2008 +0000 [trace] Add unstable warning. Add warning that CairoScript is unfinished and I may radically alter the format/syntax/language before release. util/cairo-trace/trace.c | 1 + 1 file changed, 1 insertion(+) commit 7ad2635ebe84ba7c050a7f00c5389cdc31ab1172 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Nov 11 09:46:38 2008 +0000 [trace] Disable stdout by default. Don't hog the terminal when we save to a trace file anyway. util/cairo-trace/cairo-trace.in | 48 +++++++++++++++++++++++++---------------- 1 file changed, 29 insertions(+), 19 deletions(-) commit efa9e1088cbf1b5331cc0bab3348520f3b91ae4b Author: M Joonas Pihlaja <jpihlaja@cc.helsinki.fi> Date: Thu Nov 13 12:11:12 2008 +0200 [clip] Fix uninitialised status return in _cairo_clip_intersect_mask() for empty extents. This fixes the clip-all test case crashing for me. src/cairo-clip.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 279a7cef349cf5aa87cdeacb9424413f51117fe8 Author: Behdad Esfahbod <behdad@behdad.org> Date: Thu Nov 13 10:50:41 2008 +0100 [os2] Fix includes Patch from Dave Yeo to make cairo-os2.h include os2.h directly so the header is standalone. src/cairo-os2-private.h | 11 ----------- src/cairo-os2-surface.c | 5 ----- src/cairo-os2.h | 8 ++++++++ 3 files changed, 8 insertions(+), 16 deletions(-) commit 0a682034a04fa7fc1407e7bc093ae01f4b385ad7 Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Tue Nov 11 13:26:09 2008 -0500 Make the clipping implementation comment clearer When reviewing the spans changes, I reread this comment and it wasn't as clear as it could be. src/cairo-clip.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) commit 4fdd856569a67636aaf29bc1a76eed44418db484 Author: Peter Weilbacher <mozilla@weilbacher.org> Date: Tue Nov 11 15:37:09 2008 +0100 README: adapt pixman version requirement to match configure.ac README | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 9c0042840c70c14e3e608ac3849edf5f1ef99816 Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Tue Nov 11 08:41:31 2008 -0500 [quartz] Add the proper propagation of extents This fixes the problems introduced by d682d275b90f7326df76d2764c513c8b6d5b551b src/cairo-quartz-surface.c | 24 +++++++++++++----------- 1 file changed, 13 insertions(+), 11 deletions(-) commit 57551a91168a68399add9e5c324fba03961b7747 Author: Jeff Muizelaar <jeff@infidigm.net> Date: Tue Nov 11 08:31:23 2008 -0500 [quartz] Fix fallout caused by const cairo_pattern_t * 794460186459981cd43ca38f3eba07e3f6bf8908 broke the quartz backend. Fix it by adding const to the appropriate parameters in the function declarations and copying the pattern when needed. src/cairo-quartz-surface.c | 41 ++++++++++++++++++++++++----------------- 1 file changed, 24 insertions(+), 17 deletions(-) commit 164069b339843837e04e2522a615f03a83b362c0 Author: Maarten Maathuis <madman2003@gmail.com> Date: Sat Nov 8 15:29:40 2008 +0100 cairo-xcb: avoid leaking memory src/cairo-xcb-surface.c | 3 +++ 1 file changed, 3 insertions(+) commit 5badcf6e370b9808374a1fa547383b236a9dcd21 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Nov 8 11:17:34 2008 +0000 [trace] Use a string literal for the type. In order that the string is interned and so is only allocated the once. util/cairo-trace/trace.c | 28 ++++++++++++++-------------- 1 file changed, 14 insertions(+), 14 deletions(-) commit 1d0faefbc8700833e9bde8a9c825b02b0f3cf910 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Nov 7 22:48:21 2008 +0000 [trace] Record the XID as a number Don't waste string constants (which get interned) for a simple number! util/cairo-trace/trace.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) commit 899a159b133e22f71e2a176785c91e78838c3fee Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Nov 8 11:16:03 2008 +0000 [trace] Remove stray /source An accidental /source was pushing pushed onto the operand stack and never used. util/cairo-trace/trace.c | 1 - 1 file changed, 1 deletion(-) commit f1cae5c52a5ee71b7933fca702bb525d98936838 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Nov 7 20:52:08 2008 +0000 [ps] Trivial compile warning clenaup. Add a missing const to silence the compiler. src/cairo-ps-surface.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 2b32c8b9e572c96ce8ba5c7d43b568f18f6da295 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Nov 7 20:30:33 2008 +0000 [hash] Return lookup entry. Use the return value to return the result from _cairo_hash_table_lookup() (as opposed to filling an output parameter on the stack) as this (a) results in cleaner code (no strict-alias breaking pointer casts), (b) produces a smaller binary and (c) is measurably faster. src/cairo-cache-private.h | 5 ++--- src/cairo-cache.c | 8 +++---- src/cairo-cff-subset.c | 24 +++++++++------------ src/cairo-font-face.c | 9 ++++---- src/cairo-ft-font.c | 6 +++--- src/cairo-hash-private.h | 7 +++--- src/cairo-hash.c | 24 +++++++-------------- src/cairo-misc.c | 7 +++--- src/cairo-scaled-font-subsets.c | 47 ++++++++++++++++++++--------------------- src/cairo-scaled-font.c | 18 +++++++--------- 10 files changed, 67 insertions(+), 88 deletions(-) commit d15fb9344bf86dd52cda0b43d3dfc49397fd84ec Author: Karl Tomlinson <karlt+@karlt.net> Date: Fri Nov 7 20:06:35 2008 +0000 [hash] Set is_unique when finding an available for inserts As we obey the rule in Cairo that we only insert if we know that there is no existing entry in the hash table, we can therefore perform a much quicker search knowing that the key is unique. src/cairo-hash.c | 16 +++++++--------- 1 file changed, 7 insertions(+), 9 deletions(-) commit cd2e18ddc65959a736fc7b7f6bbd3e76af0495a9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Nov 7 18:35:39 2008 +0000 [test] Fix-up rgb byte packing Another embarrassing, but thankfully, trivial bug. boilerplate/cairo-boilerplate.c | 2 ++ test/any2ppm.c | 14 +++++++++----- 2 files changed, 11 insertions(+), 5 deletions(-) commit 2554d1759835a174b89107808d81d044c3e2e098 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Nov 7 13:26:46 2008 +0000 [surface] Pass a separate closure for the mime-type destroy notifier. A limitation of the current API was that the destroy notifier was called on the mime-data block. This prevents the user from passing in a pointer to a managed block, for example a mime-data block belonging to a ref-counted object. We can overcome this by allowing the user to specify the closure to be used with the destroy notifier. src/cairo-png.c | 3 ++- src/cairo-surface.c | 9 ++++++--- src/cairo-types-private.h | 1 + src/cairo.h | 11 ++++++----- test/mime-data.c | 3 ++- test/pdf-mime-data.c | 4 +++- util/cairo-trace/trace.c | 6 ++++-- 7 files changed, 24 insertions(+), 13 deletions(-) commit ff0bd64e9436026f11e85eafcd74e9a0131c8b9f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Nov 6 01:22:39 2008 +0000 [NEWS] Add a few notes. Help Carl with a speedy snapshot by writing a few notes about what has been added so far to 1.9. NEWS | 42 ++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 42 insertions(+) commit 13627b46209f9239d10a155f2de7e53c0585e4c2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Nov 6 01:04:52 2008 +0000 [in-fill] Fix transposed arguments. Silly typo that escaped the test suite. src/cairo-path-in-fill.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 9dee7af41f4f5a4c1285e9d7951148e78659c064 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Nov 6 01:02:23 2008 +0000 [test] Add off-centre tests to in-fill-trapezoid. Reading through the previous commit spotted that the arguments to edge_compare_for_y_against_x were transposed, but the test-suite had failed to catch detect it. This is due that in order to actually solve the equation we need to have a diagonal edge passing near an off-centre point of interest, which was not among the test cases. So add some off-centre tests to fully exercise the code. test/in-fill-trapezoid.c | 66 ++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 66 insertions(+) commit f5965cb7d6559e051c2581fe446d0b9f40427eb2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Nov 5 23:48:52 2008 +0000 [in-fill] Avoid tessellation by counting number of edge crossing to -∞ Benjamin Otte reports that in one particular benchmark cairo_in_fill() is a hotspot in the profile. Currently we tessellate the entire path and then search for a containing trapezoid. This is very expensive compared to the simple method of counting the number of edge crossing between the point of interest and x=-∞. For example, this speeds tessellate-256 up by almost 3 orders of magnitude. src/Makefile.sources | 1 + src/cairo-gstate.c | 30 +----- src/cairo-path-in-fill.c | 264 +++++++++++++++++++++++++++++++++++++++++++++++ src/cairoint.h | 9 ++ 4 files changed, 279 insertions(+), 25 deletions(-) commit 0ac98461597420d3dfe52e405c6b3322d32f4854 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Nov 6 00:04:11 2008 +0000 [test] Add WINDING variants to in-fill test Check cairo_in_fill() with some WINDING tests as well as the current EVEN_ODD. test/in-fill-trapezoid.c | 56 ++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 56 insertions(+) commit 476d5daa9bfc5e9014d1b6572853d1d78ce6a6d9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Nov 5 19:29:04 2008 +0000 [trace] Only build if we have zlib. Use the configure check for libz and do not attempt to build the trace unless we have zlib. configure.ac | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) commit 34564aa84a4642dceba75efdeef438be6c6896c8 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Nov 5 19:27:49 2008 +0000 [test/pdf2png] Remove dependency on GdkPixbuf It's appears to be dropped from the current poppler trunk, so just use our own venerable cairo_surface_write_ton_png(). configure.ac | 2 +- test/pdf2png.c | 32 +++++++++++++++++++------------- 2 files changed, 20 insertions(+), 14 deletions(-) commit 564d64a1323c5cbcde2dd9365ac790fe8aa1c5a6 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Nov 5 18:47:34 2008 +0000 [png] Complete the idempotent read_png() -> write_png() Write out the original PNG mime-data if attached to the surface during cairo_surface_write_to_png(). Similar to how the compressed alternate representations are handled by the other backends. Note: by automatically attaching and using the mime-data in preference to the image data, we break the read_from_png(); draw(); write_to_png() cycle. src/cairo-png.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+) commit ec559822cfe6df0006ca2db2aa3a11699326865c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Nov 5 18:11:36 2008 +0000 [trace] Use the mime-type image representation When emitting image data, first check to see if we have a pre-compressed alternate representation. util/cairo-trace/trace.c | 303 +++++++++++++++++++++++++++++++++-------------- 1 file changed, 214 insertions(+), 89 deletions(-) commit d63267e4e7e148836dcfc4c8e2a8396ddaab70d8 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Nov 5 17:18:51 2008 +0000 [test] Update mime-data to check image/png Add a "image/png" mime-type test. test/Makefile.am | 1 + test/mime-data.c | 67 +++++++++++++++++++++++++++++---------------- test/mime-data.pdf.ref.png | Bin 4345 -> 4466 bytes test/mime-data.ps.ref.png | Bin 4345 -> 4466 bytes test/mime-data.ref.png | Bin 127 -> 155 bytes test/mime-data.svg.ref.png | Bin 0 -> 6153 bytes test/png.png | Bin 0 -> 2096 bytes 7 files changed, 45 insertions(+), 23 deletions(-) commit e4ec5c762f6d01cc5af28dc0a256e268a04101aa Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Nov 5 16:44:49 2008 +0000 [svg] Embed original PNG data. Embed the attached PNG representation of a surface in preference to re-encoding the surface. src/cairo-svg-surface.c | 38 ++++++++++++++++++++++++++++++++++++++ 1 file changed, 38 insertions(+) commit e40d62a0fe96b8b937017a3bc2f18766c411ec41 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Nov 5 16:38:34 2008 +0000 [png] Attach the png representation to cairo_surface_create_from_png(). Attach the original png data as an alternate representation for image surfaces created by cairo_surface_create_from_png(). src/cairo-output-stream-private.h | 5 ++ src/cairo-output-stream.c | 28 ++++++++- src/cairo-png.c | 127 +++++++++++++++++++++++--------------- src/cairo-surface-fallback.c | 20 +++--- src/cairo.h | 1 + 5 files changed, 122 insertions(+), 59 deletions(-) commit 89616dee8f11c6a7de3fa476b13661420648786f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Nov 5 16:41:13 2008 +0000 [surface] Don't allocate a structure for mime_data == NULL. If the user attempts to clear the attached mime data representation, just clear the user data slot and do not allocate an empty structure. src/cairo-surface.c | 17 ++++++++++------- 1 file changed, 10 insertions(+), 7 deletions(-) commit a1ab11df01962bea5752f5a5b53926a26a6f34ae Author: Behdad Esfahbod <behdad@behdad.org> Date: Wed Nov 5 11:50:20 2008 -0500 Define RepeatNone, etc if the available Render version doesn't (#18385) src/cairo-xlib-xrender-private.h | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+) commit 43cfaec39cc742ddfbf566b36391d620400e10be Author: Behdad Esfahbod <behdad@behdad.org> Date: Wed Nov 5 11:50:20 2008 -0500 Define RepeatNone, etc if the available Render version doesn't (#18385) src/cairo-xlib-xrender-private.h | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+) commit 199c0e71139fe9baf83e74ea69c01629ace5f9a2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Nov 5 15:12:19 2008 +0000 [svg] Embed jpeg data. Support jpeg embedding for svg output. src/cairo-svg-surface.c | 50 +++++++++++++++++++++++++++++++++++++++++++++---- test/Makefile.am | 1 + 2 files changed, 47 insertions(+), 4 deletions(-) commit 2261590875b4be7aa258c51e766f68974750e9e7 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Nov 5 15:11:32 2008 +0000 [trace] Trim a few bytes from glyph arrays Remove some redundant whitespace from the glyph arrays to improve readability and shrink the output file size. util/cairo-trace/trace.c | 65 ++++++++++++++++++++++++++++-------------------- 1 file changed, 38 insertions(+), 27 deletions(-) commit 248dd97075b50cd91619022098ed5dd35a66b5df Author: Adrian Johnson <ajohnson@redneon.com> Date: Thu Nov 6 00:17:53 2008 +1030 win32-printing: Implement JPEG support src/cairo-win32-printing-surface.c | 91 ++++++++++++++++++++++++++++++++++---- src/cairo-win32-private.h | 3 ++ 2 files changed, 85 insertions(+), 9 deletions(-) commit 29621bd3995c5269fd6f73ab501383433bd29768 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Nov 5 12:16:47 2008 +0000 [matrix] Remove stray offset from previous commit. I moved the pixel centre to xc,yc but forgot to remove it during compensation - as caught by the test suite. Refresh a couple of reference images that depend upon exact pixel-centre rounding conditions. src/cairo-matrix.c | 8 ++++---- test/rotate-image-surface-paint.ref.png | Bin 232 -> 190 bytes test/surface-pattern.pdf.ref.png | Bin 14776 -> 14791 bytes test/surface-pattern.ref.png | Bin 11044 -> 11100 bytes test/surface-pattern.svg.ref.png | Bin 16151 -> 16069 bytes 5 files changed, 4 insertions(+), 4 deletions(-) commit e811173311ebe76d47eebcf5c6c5c798f9d3ea72 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Nov 5 10:59:15 2008 +0000 [matrix] Compensate pixman_matrix in device space. We wish to reduce the visible error when converting to a pixman matrix, so perform the compensation in device space instead of pattern space. src/cairo-image-surface.c | 4 ++-- src/cairo-matrix.c | 53 ++++++++++++++++++++++++++++------------------- 2 files changed, 34 insertions(+), 23 deletions(-) commit c95eebc92316ad50840fab1cec6c14b6c3e50fa5 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Nov 5 09:56:02 2008 +0000 [pattern] Split the translation between the matrix and [xy]_offset pixman limits the src] co-ordinates (and thus [xy]_offset] to 16bits, so we need to be careful how much of the translation vector to push into [xy]_offset. Since the range is the same for both, split the integer component between the matrix and the offset. test/scale-offset* now at least shows the source image, even if it is misplaced. src/cairo-pattern.c | 7 +++++-- test/scale-offset-image.ps.ref.png | Bin 7445 -> 7289 bytes test/scale-offset-image.ref.png | Bin 10005 -> 9953 bytes test/scale-offset-similar.ps.ref.png | Bin 7819 -> 7860 bytes test/scale-offset-similar.ref.png | Bin 10005 -> 9953 bytes 5 files changed, 5 insertions(+), 2 deletions(-) commit d1b8186fd75922e73e62ef8f2ebb011e334ebe36 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Nov 5 01:16:25 2008 +0000 [trace] Correct emission of set_font_face and set_source set_font_face was not consuming it's operand but blithely placing an undefined font_face onto the operand stack, whereas set_source was performing invalid exchanges on the stack. util/cairo-trace/trace.c | 36 +++++++++++++++++++++++++++--------- 1 file changed, 27 insertions(+), 9 deletions(-) commit 4a4b10271a614ee9bf982994b8f9ec6b6102ee10 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Nov 5 00:02:09 2008 +0000 [trace] Capture foreign drawables. If we attempt to use a surface as a source before we write to it, record the surface contents. util/cairo-trace/trace.c | 102 ++++++++++++++++++++++++++++++++++++++++++----- 1 file changed, 93 insertions(+), 9 deletions(-) commit 432fe1ec2792153ec2559ef52aece3fcdc9c5df4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Nov 4 15:30:23 2008 +0000 [trace] Mark filter mode as immediate. Be consistent and use "//" for the filter mode to indicate a constant. util/cairo-trace/trace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 68309481aa9295026e3e7e6407c793f899b5e600 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Nov 5 08:03:01 2008 +0000 [test] Mark targets with is_meta? Allow tests to skip targets based on whether they are a meta surface or not. boilerplate/cairo-boilerplate.c | 40 ++++++++++++++++++++++++---------------- boilerplate/cairo-boilerplate.h | 1 + test/cairo-test.c | 13 ++++++++++++- 3 files changed, 37 insertions(+), 17 deletions(-) commit 8855f9583e84ae2e6d981e21133f590bff2065ab Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Nov 4 20:04:02 2008 +0000 [test] Add scale-offset-(similar|image) Add a test case for the scaling bug reported by Michel Iwaniec: http://lists.cairographics.org/archives/cairo/2008-November/015660.html test/Makefile.am | 6 ++ test/scale-offset-image.c | 142 ++++++++++++++++++++++++++++++++++ test/scale-offset-image.ps.ref.png | Bin 0 -> 7445 bytes test/scale-offset-image.ref.png | Bin 0 -> 10005 bytes test/scale-offset-similar.c | 143 +++++++++++++++++++++++++++++++++++ test/scale-offset-similar.ps.ref.png | Bin 0 -> 7819 bytes test/scale-offset-similar.ref.png | Bin 0 -> 10005 bytes 7 files changed, 291 insertions(+) commit aad980900314bd032468f41a3216769c66e62097 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Nov 4 19:11:55 2008 +0000 [test] Only depend on any2ppm if we build it. Do not add a dependency to any2ppm if it has been deconfigured. test/Makefile.am | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) commit 59bdeba9abd7bef15c6855d707c4cf92b623cb00 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Nov 4 18:42:05 2008 +0000 [matrix] Minimise pixman conversion error. Minimise the error of the pixman matrix about the centre of the displayed area rather than the origin. src/cairo-image-surface.c | 25 +++++++++++++++++-------- src/cairo-matrix.c | 24 ++++++++++++++---------- src/cairo-pattern.c | 3 ++- src/cairo-xlib-surface.c | 33 +++++++++++++++++++++++---------- src/cairoint.h | 4 +++- test/Makefile.am | 3 +-- test/surface-pattern.pdf.ref.png | Bin 15996 -> 14776 bytes test/surface-pattern.ref.png | Bin 11439 -> 11044 bytes test/surface-pattern.svg.ref.png | Bin 0 -> 16151 bytes test/surface-pattern.svg11.ref.png | Bin 17663 -> 0 bytes test/surface-pattern.svg12.ref.png | Bin 17663 -> 0 bytes 11 files changed, 60 insertions(+), 32 deletions(-) commit 068d465bb3541cc0ffc3cb9daa0848365c0a6a24 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Nov 4 14:57:21 2008 +0000 [trace] Fix emission of cairo_scaled_font_create() In the case where the font face was no longer on the operand stack, the font face would not have been passed to cairo_scaled_font_create(). util/cairo-trace/trace.c | 27 +++++++++++++-------------- 1 file changed, 13 insertions(+), 14 deletions(-) commit a3d56c60e85176526d6cce139441b6921acc9218 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Nov 4 14:06:02 2008 +0000 [trace] Use a common directory by default. If called directly (e.g. via /etc/ld.so.preload) put the output trace into a central directory, and not the pwd. util/cairo-trace/Makefile.am | 10 ++++++++-- util/cairo-trace/trace.c | 8 ++++++-- 2 files changed, 14 insertions(+), 4 deletions(-) commit 3f7a21c5c93d8648ded097757852df44fc9fcaeb Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Nov 4 13:46:10 2008 +0000 [test] Add jpeg.jpg to EXTRA_DIST Forgot to add the new file to the distribution with the previous commit. test/Makefile.am | 1 + 1 file changed, 1 insertion(+) commit 9900a2adf3e43e752bd421f00e81873b41db4c30 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Nov 4 13:41:02 2008 +0000 [test] Update mime-data jpeg reference data. <adrianj> ickle_: If we are going to use a different image for jpeg in mime-data maybe we could create a jpg that contains the text "jpeg". That way when support for the other image formats is added the mime-data test could have one image for each type with each image and it is easy to see that each image is the correct one. test/jpeg.jpg | Bin 0 -> 2316 bytes test/mime-data.c | 6 +++--- test/mime-data.pdf.ref.png | Bin 96 -> 4345 bytes test/mime-data.ps.ref.png | Bin 368 -> 4345 bytes test/mime-data.ref.png | Bin 263 -> 127 bytes 5 files changed, 3 insertions(+), 3 deletions(-) commit c13a09ddb411287224c15728b14f23fef472f7d0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Nov 4 13:00:36 2008 +0000 [test] Use a different jpeg image for mime-data Detect when we successfully encode the JPEG data from the mime-type by using a completely image. test/Makefile.am | 3 +++ test/mime-data.c | 6 +++++- test/mime-data.pdf.ref.png | Bin 0 -> 96 bytes test/mime-data.ps.ref.png | Bin 0 -> 368 bytes test/scarab.jpg | Bin 0 -> 9650 bytes 5 files changed, 8 insertions(+), 1 deletion(-) commit 8407470409ad9cb93263a1c2da7dd7446d3f897e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Nov 4 12:45:47 2008 +0000 [test] Adapt pdf-mime-data for non-srcdir build. Need to look for reference/input images from srcdir as well. test/Makefile.am | 2 +- test/pdf-mime-data.c | 44 ++++++++++++++++++++++++++++++++------------ 2 files changed, 33 insertions(+), 13 deletions(-) commit 8007618837c0292b6ebc6a4f954c5049d4e99d8a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Nov 4 12:31:10 2008 +0000 [test] Fix compilation of imagediff. imagediff broke once again. test/Makefile.am | 3 ++- test/imagediff.c | 17 +++++++++++++++++ 2 files changed, 19 insertions(+), 1 deletion(-) commit 282d7744275f623ce550638df953ef279d2209f0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Nov 4 12:53:15 2008 +0000 [mime-type] Request the mime-data for the source surface, not ourselves! Fixup a typo I introduced that caused us to request the jpeg data for the destination surface and not the source. src/cairo-pdf-surface.c | 2 +- src/cairo-ps-surface.c | 5 ++--- 2 files changed, 3 insertions(+), 4 deletions(-) commit aa616abc9ab02e333cf0cad4f524b15c710bf449 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Nov 4 11:18:22 2008 +0000 [png] Use RGB for opaque images. If the ARGB32 is opaque, discard the alpha channel - so we should generate byte identical output to the reference images. src/cairo-png.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) commit 30976635d7072c06d00e39b106e41be7b08c6c49 Author: Adrian Johnson <ajohnson@redneon.com> Date: Tue Nov 4 22:52:45 2008 +1030 PS: Implement JPEG embedding src/cairo-ps-surface.c | 104 +++++++++++++++++++++++++++++++++++++++++++++++-- 1 file changed, 101 insertions(+), 3 deletions(-) commit 73bc278c7a4630cd9299c974771ffd6e9245d0ac Author: Adrian Johnson <ajohnson@redneon.com> Date: Tue Nov 4 22:25:26 2008 +1030 Add pdf mime data test To test that images in PDF files correctly embedded the mime data. test/Makefile.am | 2 + test/pdf-mime-data.c | 151 +++++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 153 insertions(+) commit 3f18d38fbdc661d34e0fd951e0fd3aa1aa2cbcc9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Nov 4 10:31:48 2008 +0000 [win32] Compile fix for extend-pad extents. Update the show_glyphs() prototype to include the extents argument. src/cairo-win32-printing-surface.c | 3 ++- src/cairo-win32-private.h | 3 ++- 2 files changed, 4 insertions(+), 2 deletions(-) commit 5ba95b7e1287ade0b8a1acf2f5cb73fa1f44c451 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Nov 4 09:22:28 2008 +0000 [trace] Convert [1 0 0 1 0 0] to identity Simplify the trace/replay by replacing the identity array with the identity operator. util/cairo-trace/trace.c | 34 +++++++++++++++++++++++++--------- 1 file changed, 25 insertions(+), 9 deletions(-) commit 3664b32195b101ebab9a5f9087095306f8969c80 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Nov 4 10:38:41 2008 +0000 [trace] Remove more @..@ from Makefile.am Use $(..) for make variable substitution as opposed to incorrectly using @..@ for substitution by configure. util/cairo-trace/Makefile.am | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) commit 1d8ad69abb88f6e0283b6ce2aadc5d801b8e3722 Author: Behdad Esfahbod <behdad@behdad.org> Date: Tue Nov 4 01:51:06 2008 -0500 [aclocal.dolt.m4] Fix build with bash versions <= 3.0 (bug #18363) build/aclocal.dolt.m4 | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 310026ab42c2078e2749c886c74c5b38cab41671 Author: Behdad Esfahbod <behdad@behdad.org> Date: Tue Nov 4 01:51:06 2008 -0500 [aclocal.dolt.m4] Fix build with bash versions <= 3.0 (bug #18363) build/aclocal.dolt.m4 | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 90217550120f129004bc555b59ced28b4b25a4d4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Nov 3 23:45:12 2008 +0000 [mime-data] Copy a reference to the mime-data on snapshotting. Instead of doing a full-copy of the mime data (which can be 10K-100K, or even larger) just copy a reference to the original mime to the snapshot surface (as suggested by Behdad). src/cairo-surface-fallback.c | 29 +++-------------- src/cairo-surface.c | 77 +++++++++++++++++++++++++++++++------------- src/cairo-types-private.h | 2 ++ src/cairoint.h | 5 +++ 4 files changed, 66 insertions(+), 47 deletions(-) commit 4f032ca35a7086b76775e4b53c6b99ba4e1eb3cb Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Nov 3 23:23:09 2008 +0000 [trace] Install cairo-trace.so into $(libdir)/cairo Install the auxiliary library into the cairo subdirectory within the lib path, so that it doesn't clutter $(libdir) and sets a precedent for future auxiliary libraries. util/cairo-trace/Makefile.am | 7 ++++--- util/cairo-trace/cairo-trace.in | 13 ++++++++++--- 2 files changed, 14 insertions(+), 6 deletions(-) commit c80a1c68c19f876c9f55b7c6d62b18ae110b11d0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Nov 3 23:16:09 2008 +0000 [configure] Make trace consistent. Remove the debianism from the comments for HAVE_BFD and comply with Behdad's guidelines on using $(...) within Makefile.am. configure.ac | 5 ++--- util/cairo-trace/Makefile.am | 6 +++--- 2 files changed, 5 insertions(+), 6 deletions(-) commit 43e2370b024f66d995c514fd53414d3d8588a481 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Nov 3 14:20:35 2008 +0000 [test] Update extend-pad. extend-pad was not a clear demonstration of the EXTEND_PAD mode, so revamp it to show the filter extending a 4 pixel surface to cover the entire output. However, this hides a discrepancy with the vector surfaces that we cannot prevent the external renders from applying an interpolation to the border pixels, so we copy the original test to extend-pad-border to check the desired behaviour on boundary pixels. test/Makefile.am | 4 ++ test/extend-pad-border.c | 95 ++++++++++++++++++++++++++++++++++++++++ test/extend-pad-border.ref.png | Bin 0 -> 616 bytes test/extend-pad-similar.c | 82 ++++++++++++++++++++++++++++++++++ test/extend-pad-similar.ref.png | Bin 0 -> 315 bytes test/extend-pad.c | 50 ++++++++------------- test/extend-pad.ref.png | Bin 616 -> 315 bytes 7 files changed, 200 insertions(+), 31 deletions(-) commit b3462c5616ae24fd391ad0872d2fbb98c6cd0c92 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Nov 3 13:10:42 2008 +0000 [test] Convert a few residual '-out.*' Catch a few -out.* hiding in boilerplate. boilerplate/cairo-boilerplate-pdf.c | 2 +- boilerplate/cairo-boilerplate-ps.c | 2 +- boilerplate/cairo-boilerplate-svg.c | 2 +- boilerplate/cairo-boilerplate-win32-printing.c | 2 +- test/Makefile.am | 6 +++--- 5 files changed, 7 insertions(+), 7 deletions(-) commit 34586860ef08e6aab988ba227997dfe376f9d319 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Nov 3 12:11:14 2008 +0000 [trace] Big-endian compile fixes. A few typos. util/cairo-trace/trace.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) commit ca83df55a63eab1821dbc009f65b3b09d61f5b62 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Nov 3 11:46:31 2008 +0000 [trace] Wrap test surfaces. Wrap the test surfaces to avoid crashes whilst tracing the test suite. util/cairo-trace/trace.c | 116 +++++++++++++++++++++++++++++++++++++++++++---- 1 file changed, 108 insertions(+), 8 deletions(-) commit 2c08f3f83b1acd168cd74b300272970658179a0f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Nov 3 11:30:38 2008 +0000 [trace] Autodetect -lbfd during configure Stop being lazy and detect libbfd during configure. configure.ac | 10 ++++++++++ util/cairo-trace/Makefile.am | 2 +- util/cairo-trace/lookup-symbol.c | 6 +++--- 3 files changed, 14 insertions(+), 4 deletions(-) commit 8fc3d0ffebea5622332327cdef9222486cc85581 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Nov 3 11:01:22 2008 +0000 [mime-data] Allow embedding of arbitrary mime-types. Use the surface user-data array allow to store an arbitrary set of alternate image representations keyed by an interned string (which ensures that it has a unique key in the user-visible namespace). Update the API to mirror that of cairo_surface_set_user_data() [i.e. return a status indicator] and switch internal users of the mime-data to the public functions. src/cairo-debug.c | 2 + src/cairo-jpeg-info-private.h | 6 +-- src/cairo-jpeg-info.c | 19 ++++--- src/cairo-misc.c | 98 ++++++++++++++++++++++++++++++++++++ src/cairo-mutex-list-private.h | 1 + src/cairo-pdf-surface.c | 16 +++--- src/cairo-surface-fallback.c | 37 ++++++++++---- src/cairo-surface-private.h | 4 -- src/cairo-surface.c | 91 ++++++++++++++++++++++++++------- src/cairo-types-private.h | 7 +++ src/cairo.h | 4 +- src/cairoint.h | 8 +++ test/Makefile.am | 3 ++ test/mime-data.c | 111 +++++++++++++++++++++++++++++++++++++++++ test/mime-data.ref.png | Bin 0 -> 263 bytes test/romedalen.jpg | Bin 0 -> 11400 bytes util/cairo-trace/trace.c | 2 +- 17 files changed, 351 insertions(+), 58 deletions(-) commit afef83a1fae028b45cefd6b4698e7b32ad4d6fc7 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Nov 3 09:40:35 2008 +0000 Compile tidy. Fix a few minor compiler warnings. src/cairo-pdf-surface.c | 2 +- src/cairo-ps-surface.c | 2 +- src/cairo-surface.c | 6 +++--- 3 files changed, 5 insertions(+), 5 deletions(-) commit f77723fc88c12ebf6f5bff5a1797649d134ecc5d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Nov 2 19:59:53 2008 +0000 [pdf] Missing status check. Check that stream was successfully opened before attempting to write to it. src/cairo-pdf-surface.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) commit 04e3bb932eeac4f403b512087b96a164d082d52c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Nov 2 21:51:37 2008 +0000 [trace] Emit set_mime_data(). Wrap the new cairo_surface_set_mime_data() function. util/cairo-trace/trace.c | 30 ++++++++++++++++++++++++++++-- 1 file changed, 28 insertions(+), 2 deletions(-) commit 350fa7b98e5045a4fa05f64a1ad7d8167036be34 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Nov 1 13:12:43 2008 +0000 [trace] Fix boundary terminations Don't increment the terminator! util/cairo-trace/trace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit f3cbc5cf4580da83b7edacfd381afee3d571c659 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 31 23:47:16 2008 +0000 [trace] Use utf8 len in show_text_glyphs(). Honour the parameter specifying the length of the utf8 string when emitting show_text_glyphs. util/cairo-trace/trace.c | 36 ++++++++++++++++++++---------------- 1 file changed, 20 insertions(+), 16 deletions(-) commit 6434cddb899d750b815d71f8f51ae3e4568dd3c0 Author: Adrian Johnson <ajohnson@redneon.com> Date: Sun Nov 2 11:42:33 2008 +1030 type1-subset: return unsupported on FT errors and let type1-fallback handle it. This fixes the bug reported by Peter Weilbacher in http://lists.cairographics.org/archives/cairo/2008-October/015569.html src/cairo-type1-subset.c | 12 ++++-------- 1 file changed, 4 insertions(+), 8 deletions(-) commit 1685bbee1bf1ebc015ada94f77756d8d10c7818d Author: Adrian Johnson <ajohnson@redneon.com> Date: Sun Nov 2 19:50:59 2008 +1030 test/README: fix typo test/README | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 540ac11113015f0fd6856d016cb38fec5282b6ce Author: Adrian Johnson <ajohnson@redneon.com> Date: Sun Nov 2 11:42:33 2008 +1030 type1-subset: return unsupported on FT errors and let type1-fallback handle it. This fixes the bug reported by Peter Weilbacher in http://lists.cairographics.org/archives/cairo/2008-October/015569.html src/cairo-type1-subset.c | 12 ++++-------- 1 file changed, 4 insertions(+), 8 deletions(-) commit 3707178fa48e23b85c5640f3cee72e19f49c700b Author: Adrian Johnson <ajohnson@redneon.com> Date: Mon Sep 8 10:26:58 2008 +0930 PDF: Implement JPEG image embedding src/Makefile.sources | 2 + src/cairo-jpeg-info-private.h | 54 ++++++++++++++++ src/cairo-jpeg-info.c | 142 ++++++++++++++++++++++++++++++++++++++++++ src/cairo-pdf-surface.c | 55 ++++++++++++++++ 4 files changed, 253 insertions(+) commit 3c684347f49a581bfba35202ec61a5f6334acd4a Author: Adrian Johnson <ajohnson@redneon.com> Date: Sun Sep 7 20:46:20 2008 +0930 Add cairo_surface_set_mime_data() Currently only handles jpeg data. src/cairo-surface-fallback.c | 15 ++++++++++ src/cairo-surface-private.h | 4 +++ src/cairo-surface.c | 67 ++++++++++++++++++++++++++++++++++++++++++++ src/cairo.h | 15 ++++++++++ 4 files changed, 101 insertions(+) commit a39a4f52ac4a057f579e881a9a929029bcc514d1 Author: Adrian Johnson <ajohnson@redneon.com> Date: Wed Oct 1 21:02:26 2008 +0930 PS: meta-surface patterns with EXTEND_PAD are unsupported We have to use fallback images for this. src/cairo-ps-surface.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) commit 19ee13ca2b5b1e08a7fc965516be514d2fe40232 Author: Adrian Johnson <ajohnson@redneon.com> Date: Wed Oct 1 21:01:17 2008 +0930 PS: Implement EXTEND_PAD for image patterns Images with EXTEND_PAD are painted into a new image the size of the operation extents. The new image is then embedded in the PS file with the pattern matrix adjusted to ensure the image origin is in the correct location. src/cairo-ps-surface-private.h | 1 + src/cairo-ps-surface.c | 99 ++++++++++++++++++++++++++++++++++-------- 2 files changed, 83 insertions(+), 17 deletions(-) commit e639b536711c07bc09355205f476ab3f73012268 Author: Adrian Johnson <ajohnson@redneon.com> Date: Wed Oct 1 21:00:30 2008 +0930 PDF: meta-surface patterns with EXTEND_PAD are unsupported We have to use fallback images for this. src/cairo-pdf-surface.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) commit 6c92edd9a3b0c79857cfa8dee482efa56f40c714 Author: Adrian Johnson <ajohnson@redneon.com> Date: Wed Oct 1 20:56:49 2008 +0930 PDF: Implement EXTEND_PAD for image patterns Images with EXTEND_PAD are painted into a new image the size of the operation extents. The new image is then embedded in the PDF file with the pattern matrix adjusted to ensure the image origin is in the correct location. src/cairo-pdf-surface.c | 70 +++++++++++++++++++++++++++++++++++++++++++------ 1 file changed, 62 insertions(+), 8 deletions(-) commit d146cb4056ab54cf85454a6fe9d36282ca7a3f2e Author: Adrian Johnson <ajohnson@redneon.com> Date: Wed Oct 1 20:56:01 2008 +0930 PDF: Store the operation extents in each pattern The extents will be used by EXTEND_PAD patterns as well as any other pattern that can benefit from knowing the extents of the operation it will be used with. src/cairo-pdf-surface-private.h | 1 + src/cairo-pdf-surface.c | 30 ++++++++++++++++++++++++------ 2 files changed, 25 insertions(+), 6 deletions(-) commit fb7cfdd94d4e436e066d884f6dc294efd896b344 Author: Adrian Johnson <ajohnson@redneon.com> Date: Wed Oct 1 20:52:08 2008 +0930 Make meta-surface store and replay extents of each operation To be able to provide the extents of each operation to the backend during the render phase the meta-surface needs to store the extents computed by the analysis surface during the analysis phase. The extents argument is either a pointer to the extents of the operation stored in the meta-surface or NULL. During analysis the analysis surface writes the extents to the meta-surface. During the render phase the extents is made available to paginated surface backends. src/cairo-analysis-surface.c | 12 ++++++++++++ src/cairo-meta-surface-private.h | 1 + src/cairo-meta-surface.c | 33 +++++++++++++++++++++++++++------ 3 files changed, 40 insertions(+), 6 deletions(-) commit d682d275b90f7326df76d2764c513c8b6d5b551b Author: Adrian Johnson <ajohnson@redneon.com> Date: Wed Oct 1 20:50:53 2008 +0930 Add an extents argument to the high level surface backend functions Add a "cairo_rectangle_int_t *extents" argument to to the following backend functions: paint mask, stroke fill show_glyphs show_text_glyphs This will be used to pass the extents of each operation computed by the analysis surface to the backend. This is required for implementing EXTEND_PAD. src/cairo-analysis-surface.c | 49 +++++++++++++++++++++++--------------- src/cairo-directfb-surface.c | 3 ++- src/cairo-gstate.c | 12 +++++----- src/cairo-meta-surface.c | 27 ++++++++++++--------- src/cairo-paginated-surface.c | 27 ++++++++++++--------- src/cairo-pdf-surface.c | 18 +++++++++----- src/cairo-ps-surface.c | 12 ++++++---- src/cairo-quartz-surface.c | 15 ++++++++---- src/cairo-surface.c | 45 +++++++++++++++++++--------------- src/cairo-svg-surface.c | 24 ++++++++++++------- src/cairo-type3-glyph-surface.c | 17 ++++++++----- src/cairo-win32-printing-surface.c | 12 ++++++---- src/cairo-win32-surface.c | 3 ++- src/cairo-xlib-surface.c | 8 ++++--- src/cairoint.h | 39 ++++++++++++++++++++---------- src/test-meta-surface.c | 25 +++++++++++-------- src/test-paginated-surface.c | 25 +++++++++++-------- 17 files changed, 223 insertions(+), 138 deletions(-) commit ed2081d97401741db10b0244eaba7ff31ae63346 Author: Adrian Johnson <ajohnson@redneon.com> Date: Sun Nov 2 15:05:07 2008 +1030 Specify a background color for XFAIL text To improve readability on terminals with a white background. test/cairo-test.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 81c3009c077ddab40df052bffaa646526e2b8dfc Author: Behdad Esfahbod <behdad@behdad.org> Date: Fri Oct 31 16:45:58 2008 -0400 [boilerplate/Makefile.win32] Only build source files, not headers boilerplate/Makefile.win32 | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) commit d5f88c28021907ad761a1bf1c311bfd206db617e Author: Behdad Esfahbod <behdad@behdad.org> Date: Fri Oct 31 13:43:37 2008 -0400 [.gitignore] Add util/cairo-trace/.gitignore | 1 + 1 file changed, 1 insertion(+) commit 49bd253fd36dc93219a5414c6e7a6daeaa9c63b9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 31 16:40:49 2008 +0000 [tessellator] Refine the math comments. First of a simple substitution for -?-, as they are very confusing in context with other minus signs floating around. Carl has promised to go over these docs with me at the HackFest in order to improve them (and verify them). src/cairo-bentley-ottmann.c | 30 +++++++++++++++--------------- 1 file changed, 15 insertions(+), 15 deletions(-) commit ab8a0bfd82f8f474714d7aae4ceff4e7848c9b4b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 31 16:09:13 2008 +0000 Add a COPYING file to each aux. source directory Include a COPYING inside perf/, test/, util/ to clarify the licensing conditions beneath the respective directories. This is because cairo itself (libcairo.so) is LGPL-2.1/MPL-1.1 but that only relates to src/. The auxiliary source files are under a mix of free licenses and we wish to be clear just what license applies to each file. In particular, cairo-trace needs to include the GPL terms and conditions. COPYING | 17 +- perf/COPYING | 5 + perf/Makefile.am | 2 +- test/COPYING | 5 + test/Makefile.am | 2 +- util/COPYING | 4 + util/Makefile.am | 1 + util/cairo-trace/COPYING | 5 + util/cairo-trace/COPYING-GPL-3 | 674 +++++++++++++++++++++++++++++++++++++++++ util/cairo-trace/Makefile.am | 5 +- 10 files changed, 709 insertions(+), 11 deletions(-) commit c554f18d78fff71aff2d79309b19155de407363d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 31 15:37:58 2008 +0000 [util] Add cairo-trace. This tool can be used to trace all the cairo function calls made by an applications. This is useful for either extracting a test case triggering a bug from an application, or simply to get a general idea of how an application is using cairo. After make install, cairo-trace program arguments, will print out all the cairo calls to the terminal and also capture theme in ./program.$pid.trace The format of the output is CairoScript, watch this space for more cairo-script tools! build/configure.ac.system | 2 + configure.ac | 17 + util/Makefile.am | 6 + util/README | 8 - util/cairo-trace/Makefile.am | 21 + util/cairo-trace/cairo-trace.in | 58 + util/cairo-trace/lookup-symbol.c | 290 ++++ util/cairo-trace/lookup-symbol.h | 24 + util/cairo-trace/trace.c | 3345 ++++++++++++++++++++++++++++++++++++++ 9 files changed, 3763 insertions(+), 8 deletions(-) commit 992f74d884c9ea83f7b51c6959dd93718027b99a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 31 13:50:55 2008 +0000 [test] Use '.' as the field separator in the names We frequently use '-' within the test name or format name and so we encounter confusion as '-' is also used as the field separator. At times this has caused a new test to break an old test because the new test would match one of the old test's target specific reference images. So switch everything over to use '.' between fields (test name, target, format, subtest, etc.). test/.gitignore | 8 +- test/Makefile.am | 1402 ++++++++++---------- ...mage-sample-ref.png => a1-image-sample.ref.png} | Bin test/{a1-mask-ref.png => a1-mask.ref.png} | Bin ...raps-sample-ref.png => a1-traps-sample.ref.png} | Bin test/{a8-mask-ref.png => a8-mask.ref.png} | Bin ...alpha-similar-ref.png => alpha-similar.ref.png} | Bin ...r-rgb24-ref.png => alpha-similar.rgb24.ref.png} | Bin .../{big-line-ps2-ref.png => big-line.ps2.ref.png} | Bin ...s2-rgb24-ref.png => big-line.ps2.rgb24.ref.png} | Bin .../{big-line-ps3-ref.png => big-line.ps3.ref.png} | Bin ...s3-rgb24-ref.png => big-line.ps3.rgb24.ref.png} | Bin ...line-quartz-ref.png => big-line.quartz.ref.png} | Bin ...rgb24-ref.png => big-line.quartz.rgb24.ref.png} | Bin test/{big-line-ref.png => big-line.ref.png} | Bin ...g-line-rgb24-ref.png => big-line.rgb24.ref.png} | Bin ...bilevel-image-ref.png => bilevel-image.ref.png} | Bin test/{bitmap-font-ref.png => bitmap-font.ref.png} | Bin ...ont-rgb24-ref.png => bitmap-font.rgb24.ref.png} | Bin test/cairo-test.c | 68 +- test/cairo-test.h | 1 + ...rtz-ref.png => caps-joins-alpha.quartz.ref.png} | Bin ...oins-alpha-ref.png => caps-joins-alpha.ref.png} | Bin ...vg11-ref.png => caps-joins-alpha.svg11.ref.png} | Bin ...vg12-ref.png => caps-joins-alpha.svg12.ref.png} | Bin ...ps-joins-ps2-ref.png => caps-joins.ps2.ref.png} | Bin ...ps-joins-ps3-ref.png => caps-joins.ps3.ref.png} | Bin test/{caps-joins-ref.png => caps-joins.ref.png} | Bin ...ps-sub-paths-ref.png => caps-sub-paths.ref.png} | Bin test/{clip-all-ref.png => clip-all.ref.png} | Bin test/{clip-empty-ref.png => clip-empty.ref.png} | Bin ...ef.png => clip-fill-rule-pixel-aligned.ref.png} | Bin ... => clip-fill-rule-pixel-aligned.rgb24.ref.png} | Bin ...2-ref.png => clip-fill-rule.ps2.argb32.ref.png} | Bin ...24-ref.png => clip-fill-rule.ps2.rgb24.ref.png} | Bin ...2-ref.png => clip-fill-rule.ps3.argb32.ref.png} | Bin ...24-ref.png => clip-fill-rule.ps3.rgb24.ref.png} | Bin ...ip-fill-rule-ref.png => clip-fill-rule.ref.png} | Bin ...-rgb24-ref.png => clip-fill-rule.rgb24.ref.png} | Bin ...b32-ref.png => clip-nesting.ps2.argb32.ref.png} | Bin ...gb24-ref.png => clip-nesting.ps2.rgb24.ref.png} | Bin ...b32-ref.png => clip-nesting.ps3.argb32.ref.png} | Bin ...gb24-ref.png => clip-nesting.ps3.rgb24.ref.png} | Bin ...-quartz-ref.png => clip-nesting.quartz.ref.png} | Bin ...4-ref.png => clip-nesting.quartz.rgb24.ref.png} | Bin .../{clip-nesting-ref.png => clip-nesting.ref.png} | Bin ...ng-rgb24-ref.png => clip-nesting.rgb24.ref.png} | Bin ...32-ref.png => clip-operator.pdf.argb32.ref.png} | Bin ...b24-ref.png => clip-operator.pdf.rgb24.ref.png} | Bin ...b24-ref.png => clip-operator.ps2.rgb24.ref.png} | Bin ...rator-ps3-ref.png => clip-operator.ps3.ref.png} | Bin ...b24-ref.png => clip-operator.ps3.rgb24.ref.png} | Bin ...quartz-ref.png => clip-operator.quartz.ref.png} | Bin ...-ref.png => clip-operator.quartz.rgb24.ref.png} | Bin ...clip-operator-ref.png => clip-operator.ref.png} | Bin ...r-rgb24-ref.png => clip-operator.rgb24.ref.png} | Bin ...-ref.png => clip-push-group.ps2.argb32.ref.png} | Bin ...4-ref.png => clip-push-group.ps2.rgb24.ref.png} | Bin ...-ref.png => clip-push-group.ps3.argb32.ref.png} | Bin ...4-ref.png => clip-push-group.ps3.rgb24.ref.png} | Bin ...artz-ref.png => clip-push-group.quartz.ref.png} | Bin ...-push-group-ref.png => clip-push-group.ref.png} | Bin ...rgb32-ref.png => clip-twice.ps2.argb32.ref.png} | Bin ...-rgb24-ref.png => clip-twice.ps2.rgb24.ref.png} | Bin ...rgb32-ref.png => clip-twice.ps3.argb32.ref.png} | Bin ...-rgb24-ref.png => clip-twice.ps3.rgb24.ref.png} | Bin ...ce-quartz-ref.png => clip-twice.quartz.ref.png} | Bin ...b24-ref.png => clip-twice.quartz.rgb24.ref.png} | Bin test/{clip-twice-ref.png => clip-twice.ref.png} | Bin ...wice-rgb24-ref.png => clip-twice.rgb24.ref.png} | Bin ...group-ps2-ref.png => clipped-group.ps2.ref.png} | Bin ...group-ps3-ref.png => clipped-group.ps3.ref.png} | Bin ...clipped-group-ref.png => clipped-group.ref.png} | Bin ...ped-surface-ref.png => clipped-surface.ref.png} | Bin ...ose-path-ps2-ref.png => close-path.ps2.ref.png} | Bin ...ose-path-ps3-ref.png => close-path.ps3.ref.png} | Bin test/{close-path-ref.png => close-path.ref.png} | Bin ...site-integer-translate-over-repeat.ps2.ref.png} | Bin ...site-integer-translate-over-repeat.ps3.ref.png} | Bin ...omposite-integer-translate-over-repeat.ref.png} | Bin ...> composite-integer-translate-over.ps2.ref.png} | Bin ...> composite-integer-translate-over.ps3.ref.png} | Bin ...ng => composite-integer-translate-over.ref.png} | Bin ...composite-integer-translate-source.ps2.ref.png} | Bin ...composite-integer-translate-source.ps3.ref.png} | Bin ... => composite-integer-translate-source.ref.png} | Bin ...copy-path-ps2-ref.png => copy-path.ps2.ref.png} | Bin ...copy-path-ps3-ref.png => copy-path.ps3.ref.png} | Bin test/{copy-path-ref.png => copy-path.ref.png} | Bin test/create-for-stream.c | 2 +- test/create-from-png-stream.c | 2 +- ...ream-ref.png => create-from-png-stream.ref.png} | Bin ...alpha-ref.png => create-from-png.alpha.ref.png} | Bin test/create-from-png.c | 16 +- ...-ref.png => create-from-png.gray-alpha.ref.png} | Bin ...g-gray-ref.png => create-from-png.gray.ref.png} | Bin ...f.png => create-from-png.indexed-alpha.ref.png} | Bin ...xed-ref.png => create-from-png.indexed.ref.png} | Bin ...te-from-png-ref.png => create-from-png.ref.png} | Bin ...-ref.png => dash-caps-joins.ps2.argb32.ref.png} | Bin ...4-ref.png => dash-caps-joins.ps2.rgb24.ref.png} | Bin ...-ref.png => dash-caps-joins.ps3.argb32.ref.png} | Bin ...4-ref.png => dash-caps-joins.ps3.rgb24.ref.png} | Bin ...artz-ref.png => dash-caps-joins.quartz.ref.png} | Bin ...-caps-joins-ref.png => dash-caps-joins.ref.png} | Bin ...sh-curve-ps2-ref.png => dash-curve.ps2.ref.png} | Bin ...sh-curve-ps3-ref.png => dash-curve.ps3.ref.png} | Bin ...ve-quartz-ref.png => dash-curve.quartz.ref.png} | Bin test/{dash-curve-ref.png => dash-curve.ref.png} | Bin .../{dash-no-dash-ref.png => dash-no-dash.ref.png} | Bin ...gative-ref.png => dash-offset-negative.ref.png} | Bin ...rgb32-ref.png => dash-scale.ps2.argb32.ref.png} | Bin ...-rgb24-ref.png => dash-scale.ps2.rgb24.ref.png} | Bin ...rgb32-ref.png => dash-scale.ps3.argb32.ref.png} | Bin ...-rgb24-ref.png => dash-scale.ps3.rgb24.ref.png} | Bin ...le-quartz-ref.png => dash-scale.quartz.ref.png} | Bin test/{dash-scale-ref.png => dash-scale.ref.png} | Bin ...sh-state-ps2-ref.png => dash-state.ps2.ref.png} | Bin ...sh-state-ps3-ref.png => dash-state.ps3.ref.png} | Bin ...te-quartz-ref.png => dash-state.quartz.ref.png} | Bin test/{dash-state-ref.png => dash-state.ref.png} | Bin ...th-ps2-ref.png => dash-zero-length.ps2.ref.png} | Bin ...-ref.png => dash-zero-length.ps2.rgb24.ref.png} | Bin ...th-ps3-ref.png => dash-zero-length.ps3.ref.png} | Bin ...-ref.png => dash-zero-length.ps3.rgb24.ref.png} | Bin ...ero-length-ref.png => dash-zero-length.ref.png} | Bin ...gb24-ref.png => dash-zero-length.rgb24.ref.png} | Bin ...-arc-ps2-ref.png => degenerate-arc.ps2.ref.png} | Bin ...-arc-ps3-ref.png => degenerate-arc.ps3.ref.png} | Bin ...generate-arc-ref.png => degenerate-arc.ref.png} | Bin ...nerate-dash-ref.png => degenerate-dash.ref.png} | Bin ...-ref.png => degenerate-path.ps2.argb32.ref.png} | Bin ...4-ref.png => degenerate-path.ps2.rgb24.ref.png} | Bin ...-ref.png => degenerate-path.ps3.argb32.ref.png} | Bin ...4-ref.png => degenerate-path.ps3.rgb24.ref.png} | Bin ...artz-ref.png => degenerate-path.quartz.ref.png} | Bin ...ef.png => degenerate-path.quartz.rgb24.ref.png} | Bin ...nerate-path-ref.png => degenerate-path.ref.png} | Bin ...rgb24-ref.png => degenerate-path.rgb24.ref.png} | Bin ...-pen-ps2-ref.png => degenerate-pen.ps2.ref.png} | Bin ...-pen-ps3-ref.png => degenerate-pen.ps3.ref.png} | Bin ...uartz-ref.png => degenerate-pen.quartz.ref.png} | Bin ...generate-pen-ref.png => degenerate-pen.ref.png} | Bin ...ef.png => device-offset-fractional.pdf.ref.png} | Bin ...ef.png => device-offset-fractional.ps2.ref.png} | Bin ...ef.png => device-offset-fractional.ps3.ref.png} | Bin ...al-ref.png => device-offset-fractional.ref.png} | Bin ...tive-ref.png => device-offset-positive.ref.png} | Bin ...ef.png => device-offset-positive.rgb24.ref.png} | Bin ...t-scale-ref.png => device-offset-scale.ref.png} | Bin ...device-offset-ref.png => device-offset.ref.png} | Bin ...t-rgb24-ref.png => device-offset.rgb24.ref.png} | Bin test/{extend-pad-ref.png => extend-pad.ref.png} | Bin ...-ref.png => extend-reflect-similar.ps2.ref.png} | Bin ...-ref.png => extend-reflect-similar.ps3.ref.png} | Bin ...lect-ref.png => extend-reflect-similar.ref.png} | Bin ...ilar-ps2-ref.png => extend-reflect.ps2.ref.png} | Bin ...ilar-ps3-ref.png => extend-reflect.ps3.ref.png} | Bin ...lect-similar-ref.png => extend-reflect.ref.png} | Bin ...2-ref.png => extend-repeat-similar.ps2.ref.png} | Bin ...3-ref.png => extend-repeat-similar.ps3.ref.png} | Bin ...epeat-ref.png => extend-repeat-similar.ref.png} | Bin ...milar-ps2-ref.png => extend-repeat.ps2.ref.png} | Bin ...milar-ps3-ref.png => extend-repeat.ps3.ref.png} | Bin ...epeat-similar-ref.png => extend-repeat.ref.png} | Bin test/fallback-resolution.c | 11 +- ....png => fallback-resolution.ppi150x150.ref.png} | Bin ....png => fallback-resolution.ppi150x300.ref.png} | Bin ...png => fallback-resolution.ppi150x37.5.ref.png} | Bin ....png => fallback-resolution.ppi150x600.ref.png} | Bin ...f.png => fallback-resolution.ppi150x72.ref.png} | Bin ...f.png => fallback-resolution.ppi150x75.ref.png} | Bin ....png => fallback-resolution.ppi300x150.ref.png} | Bin ....png => fallback-resolution.ppi300x300.ref.png} | Bin ...png => fallback-resolution.ppi300x37.5.ref.png} | Bin ....png => fallback-resolution.ppi300x600.ref.png} | Bin ...f.png => fallback-resolution.ppi300x72.ref.png} | Bin ...f.png => fallback-resolution.ppi300x75.ref.png} | Bin ...png => fallback-resolution.ppi37.5x150.ref.png} | Bin ...png => fallback-resolution.ppi37.5x300.ref.png} | Bin ...ng => fallback-resolution.ppi37.5x37.5.ref.png} | Bin ...png => fallback-resolution.ppi37.5x600.ref.png} | Bin ....png => fallback-resolution.ppi37.5x72.ref.png} | Bin ....png => fallback-resolution.ppi37.5x75.ref.png} | Bin ....png => fallback-resolution.ppi600x150.ref.png} | Bin ....png => fallback-resolution.ppi600x300.ref.png} | Bin ...png => fallback-resolution.ppi600x37.5.ref.png} | Bin ....png => fallback-resolution.ppi600x600.ref.png} | Bin ...f.png => fallback-resolution.ppi600x72.ref.png} | Bin ...f.png => fallback-resolution.ppi600x75.ref.png} | Bin ...f.png => fallback-resolution.ppi72x150.ref.png} | Bin ...f.png => fallback-resolution.ppi72x300.ref.png} | Bin ....png => fallback-resolution.ppi72x37.5.ref.png} | Bin ...f.png => fallback-resolution.ppi72x600.ref.png} | Bin ...ef.png => fallback-resolution.ppi72x72.ref.png} | Bin ...ef.png => fallback-resolution.ppi72x75.ref.png} | Bin ...f.png => fallback-resolution.ppi75x150.ref.png} | Bin ...f.png => fallback-resolution.ppi75x300.ref.png} | Bin ....png => fallback-resolution.ppi75x37.5.ref.png} | Bin ...f.png => fallback-resolution.ppi75x600.ref.png} | Bin ...ef.png => fallback-resolution.ppi75x72.ref.png} | Bin ...ef.png => fallback-resolution.ppi75x75.ref.png} | Bin ...f.png => fill-alpha-pattern.pdf.argb32.ref.png} | Bin ...ef.png => fill-alpha-pattern.pdf.rgb24.ref.png} | Bin ...-ps3-ref.png => fill-alpha-pattern.ps3.ref.png} | Bin ...-pattern-ref.png => fill-alpha-pattern.ref.png} | Bin test/{fill-alpha-ref.png => fill-alpha.ref.png} | Bin ...ng => fill-and-stroke-alpha-add.quartz.ref.png} | Bin ...d-ref.png => fill-and-stroke-alpha-add.ref.png} | Bin ...ef.png => fill-and-stroke-alpha.quartz.ref.png} | Bin ...alpha-ref.png => fill-and-stroke-alpha.ref.png} | Bin ...-ref.png => fill-and-stroke.ps2.argb32.ref.png} | Bin ...4-ref.png => fill-and-stroke.ps2.rgb24.ref.png} | Bin ...-ref.png => fill-and-stroke.ps3.argb32.ref.png} | Bin ...4-ref.png => fill-and-stroke.ps3.rgb24.ref.png} | Bin ...artz-ref.png => fill-and-stroke.quartz.ref.png} | Bin ...ef.png => fill-and-stroke.quartz.rgb24.ref.png} | Bin ...-and-stroke-ref.png => fill-and-stroke.ref.png} | Bin ...rgb24-ref.png => fill-and-stroke.rgb24.ref.png} | Bin ...g => fill-degenerate-sort-order.quartz.ref.png} | Bin ...ill-degenerate-sort-order.quartz.rgb24.ref.png} | Bin ...-ref.png => fill-degenerate-sort-order.ref.png} | Bin ...ng => fill-degenerate-sort-order.rgb24.ref.png} | Bin ...ref.png => fill-missed-stop.ps2.argb32.ref.png} | Bin ...-ref.png => fill-missed-stop.ps2.rgb24.ref.png} | Bin ...ref.png => fill-missed-stop.ps3.argb32.ref.png} | Bin ...-ref.png => fill-missed-stop.ps3.rgb24.ref.png} | Bin ...issed-stop-ref.png => fill-missed-stop.ref.png} | Bin ...gb24-ref.png => fill-missed-stop.rgb24.ref.png} | Bin ...argb32-ref.png => fill-rule.ps2.argb32.ref.png} | Bin ...2-rgb24-ref.png => fill-rule.ps2.rgb24.ref.png} | Bin ...argb32-ref.png => fill-rule.ps3.argb32.ref.png} | Bin ...3-rgb24-ref.png => fill-rule.ps3.rgb24.ref.png} | Bin ...ule-quartz-ref.png => fill-rule.quartz.ref.png} | Bin ...gb24-ref.png => fill-rule.quartz.rgb24.ref.png} | Bin test/{fill-rule-ref.png => fill-rule.ref.png} | Bin ...-rule-rgb24-ref.png => fill-rule.rgb24.ref.png} | Bin ...ref.png => filter-bilinear-extents.pdf.ref.png} | Bin ...ref.png => filter-bilinear-extents.ps2.ref.png} | Bin ...ref.png => filter-bilinear-extents.ps3.ref.png} | Bin ...nts-ref.png => filter-bilinear-extents.ref.png} | Bin ...f-ref.png => filter-nearest-offset.pdf.ref.png} | Bin ...2-ref.png => filter-nearest-offset.ps2.ref.png} | Bin ...3-ref.png => filter-nearest-offset.ps3.ref.png} | Bin ...ffset-ref.png => filter-nearest-offset.ref.png} | Bin ...ref.png => filter-nearest-offset.svg11.ref.png} | Bin ...ref.png => filter-nearest-offset.svg12.ref.png} | Bin ....png => filter-nearest-transformed.pdf.ref.png} | Bin ...-ref.png => filter-nearest-transformed.ref.png} | Bin ...ng => filter-nearest-transformed.svg11.ref.png} | Bin ...ng => filter-nearest-transformed.svg12.ref.png} | Bin ...ref.png => finer-grained-fallbacks.ps2.ref.png} | Bin ...g => finer-grained-fallbacks.ps2.rgb24.ref.png} | Bin ...ref.png => finer-grained-fallbacks.ps3.ref.png} | Bin ...g => finer-grained-fallbacks.ps3.rgb24.ref.png} | Bin ...cks-ref.png => finer-grained-fallbacks.ref.png} | Bin ...f.png => finer-grained-fallbacks.rgb24.ref.png} | Bin ... => font-matrix-translation.ps2.argb32.ref.png} | Bin ...g => font-matrix-translation.ps2.rgb24.ref.png} | Bin ... => font-matrix-translation.ps3.argb32.ref.png} | Bin ...g => font-matrix-translation.ps3.rgb24.ref.png} | Bin ....png => font-matrix-translation.quartz.ref.png} | Bin ...ion-ref.png => font-matrix-translation.ref.png} | Bin ...f.png => font-matrix-translation.svg11.ref.png} | Bin ...f.png => font-matrix-translation.svg12.ref.png} | Bin ....png => ft-show-glyphs-positioning.pdf.ref.png} | Bin ....png => ft-show-glyphs-positioning.ps2.ref.png} | Bin ....png => ft-show-glyphs-positioning.ps3.ref.png} | Bin ...-ref.png => ft-show-glyphs-positioning.ref.png} | Bin ...ng => ft-show-glyphs-positioning.svg11.ref.png} | Bin ...ng => ft-show-glyphs-positioning.svg12.ref.png} | Bin ...s2-ref.png => ft-show-glyphs-table.ps2.ref.png} | Bin ...s3-ref.png => ft-show-glyphs-table.ps3.ref.png} | Bin ...-table-ref.png => ft-show-glyphs-table.ref.png} | Bin ...g => ft-text-antialias-none.ps2.argb32.ref.png} | Bin ...g => ft-text-antialias-none.ps3.argb32.ref.png} | Bin ...none-ref.png => ft-text-antialias-none.ref.png} | Bin ...g => ft-text-vertical-layout-type1.pdf.ref.png} | Bin ...g => ft-text-vertical-layout-type1.ps2.ref.png} | Bin ...g => ft-text-vertical-layout-type1.ps3.ref.png} | Bin ...f.png => ft-text-vertical-layout-type1.ref.png} | Bin ...=> ft-text-vertical-layout-type1.svg11.ref.png} | Bin ...=> ft-text-vertical-layout-type1.svg12.ref.png} | Bin ...g => ft-text-vertical-layout-type3.pdf.ref.png} | Bin ...g => ft-text-vertical-layout-type3.ps2.ref.png} | Bin ...g => ft-text-vertical-layout-type3.ps3.ref.png} | Bin ...f.png => ft-text-vertical-layout-type3.ref.png} | Bin ...=> ft-text-vertical-layout-type3.svg11.ref.png} | Bin ...=> ft-text-vertical-layout-type3.svg12.ref.png} | Bin ...oup-target-ref.png => get-group-target.ref.png} | Bin ...source-ref.png => glitz-surface-source.ref.png} | Bin ...s2-ref.png => glyph-cache-pressure.ps2.ref.png} | Bin ...s3-ref.png => glyph-cache-pressure.ps3.ref.png} | Bin ...ref.png => glyph-cache-pressure.quartz.ref.png} | Bin ...essure-ref.png => glyph-cache-pressure.ref.png} | Bin ...2-ref.png => gradient-alpha.pdf.argb32.ref.png} | Bin ...24-ref.png => gradient-alpha.pdf.rgb24.ref.png} | Bin ...2-ref.png => gradient-alpha.ps2.argb32.ref.png} | Bin ...24-ref.png => gradient-alpha.ps2.rgb24.ref.png} | Bin ...2-ref.png => gradient-alpha.ps3.argb32.ref.png} | Bin ...24-ref.png => gradient-alpha.ps3.rgb24.ref.png} | Bin ...adient-alpha-ref.png => gradient-alpha.ref.png} | Bin ...-rgb24-ref.png => gradient-alpha.rgb24.ref.png} | Bin ... => gradient-constant-alpha.pdf.argb32.ref.png} | Bin ...g => gradient-constant-alpha.pdf.rgb24.ref.png} | Bin ...ref.png => gradient-constant-alpha.ps3.ref.png} | Bin ...g => gradient-constant-alpha.ps3.rgb24.ref.png} | Bin ...pha-ref.png => gradient-constant-alpha.ref.png} | Bin ...f.png => gradient-constant-alpha.rgb24.ref.png} | Bin ...o-stops-ref.png => gradient-zero-stops.ref.png} | Bin ...4-ref.png => gradient-zero-stops.rgb24.ref.png} | Bin test/{group-paint-ref.png => group-paint.ref.png} | Bin ...attern-pdf-ref.png => huge-pattern.pdf.ref.png} | Bin ...gb24-ref.png => huge-pattern.pdf.rgb24.ref.png} | Bin ...attern-ps3-ref.png => huge-pattern.ps3.ref.png} | Bin .../{huge-pattern-ref.png => huge-pattern.ref.png} | Bin ...s2-ref.png => image-surface-source.ps2.ref.png} | Bin ...s3-ref.png => image-surface-source.ps3.ref.png} | Bin ...source-ref.png => image-surface-source.ref.png} | Bin ...-join-ps2-ref.png => infinite-join.ps2.ref.png} | Bin ...-join-ps3-ref.png => infinite-join.ps3.ref.png} | Bin ...infinite-join-ref.png => infinite-join.ref.png} | Bin test/{large-clip-ref.png => large-clip.ref.png} | Bin test/{large-font-ref.png => large-font.ref.png} | Bin .../{large-source-ref.png => large-source.ref.png} | Bin ...rgb32-ref.png => leaky-dash.ps2.argb32.ref.png} | Bin ...-rgb24-ref.png => leaky-dash.ps2.rgb24.ref.png} | Bin ...rgb32-ref.png => leaky-dash.ps3.argb32.ref.png} | Bin ...-rgb24-ref.png => leaky-dash.ps3.rgb24.ref.png} | Bin ...sh-quartz-ref.png => leaky-dash.quartz.ref.png} | Bin test/{leaky-dash-ref.png => leaky-dash.ref.png} | Bin ...-ref.png => leaky-dashed-rectangle.ps2.ref.png} | Bin ...-ref.png => leaky-dashed-rectangle.ps3.ref.png} | Bin ...ngle-ref.png => leaky-dashed-rectangle.ref.png} | Bin ...ps2-ref.png => leaky-dashed-stroke.ps2.ref.png} | Bin ...ps3-ref.png => leaky-dashed-stroke.ps3.ref.png} | Bin ...-stroke-ref.png => leaky-dashed-stroke.ref.png} | Bin ...lygon-ps2-ref.png => leaky-polygon.ps2.ref.png} | Bin ...lygon-ps3-ref.png => leaky-polygon.ps3.ref.png} | Bin ...leaky-polygon-ref.png => leaky-polygon.ref.png} | Bin ...le-ps2-ref.png => line-width-scale.ps2.ref.png} | Bin ...le-ps3-ref.png => line-width-scale.ps3.ref.png} | Bin ...rtz-ref.png => line-width-scale.quartz.ref.png} | Bin ...idth-scale-ref.png => line-width-scale.ref.png} | Bin test/{line-width-ref.png => line-width.ref.png} | Bin ... => linear-gradient-reflect.pdf.argb32.ref.png} | Bin ...g => linear-gradient-reflect.pdf.rgb24.ref.png} | Bin ...ref.png => linear-gradient-reflect.ps3.ref.png} | Bin ....png => linear-gradient-reflect.quartz.ref.png} | Bin ...ect-ref.png => linear-gradient-reflect.ref.png} | Bin ...ent-pdf-ref.png => linear-gradient.pdf.ref.png} | Bin ...ent-ps3-ref.png => linear-gradient.ps3.ref.png} | Bin ...artz-ref.png => linear-gradient.quartz.ref.png} | Bin ...ar-gradient-ref.png => linear-gradient.ref.png} | Bin ...svg11-ref.png => linear-gradient.svg11.ref.png} | Bin ...svg12-ref.png => linear-gradient.svg12.ref.png} | Bin ...s-ps2-ref.png => long-dashed-lines.ps2.ref.png} | Bin ...s-ps3-ref.png => long-dashed-lines.ps3.ref.png} | Bin ...tz-ref.png => long-dashed-lines.quartz.ref.png} | Bin ...hed-lines-ref.png => long-dashed-lines.ref.png} | Bin test/{long-lines-ref.png => long-lines.ref.png} | Bin ...32-ref.png => mask-alpha.quartz.argb32.ref.png} | Bin test/{mask-alpha-ref.png => mask-alpha.ref.png} | Bin ...lpha-rgb24-ref.png => mask-alpha.rgb24.ref.png} | Bin ...b32-ref.png => mask-alpha.svg11.argb32.ref.png} | Bin ...b32-ref.png => mask-alpha.svg12.argb32.ref.png} | Bin test/{mask-ctm-ref.png => mask-ctm.ref.png} | Bin ...sk-ctm-rgb24-ref.png => mask-ctm.rgb24.ref.png} | Bin ...rgb32-ref.png => mask-ctm.svg11.argb32.ref.png} | Bin ...rgb32-ref.png => mask-ctm.svg12.argb32.ref.png} | Bin ...urface-ctm-ref.png => mask-surface-ctm.ref.png} | Bin ...gb24-ref.png => mask-surface-ctm.rgb24.ref.png} | Bin ...f.png => mask-surface-ctm.svg11.argb32.ref.png} | Bin ...f.png => mask-surface-ctm.svg12.argb32.ref.png} | Bin ...mage-ref.png => mask-transformed-image.ref.png} | Bin ...ef.png => mask-transformed-similar.pdf.ref.png} | Bin ...ar-ref.png => mask-transformed-similar.ref.png} | Bin ....png => mask-transformed-similar.svg11.ref.png} | Bin ....png => mask-transformed-similar.svg12.ref.png} | Bin ...-pdf-argb32-ref.png => mask.pdf.argb32.ref.png} | Bin ...sk-pdf-rgb24-ref.png => mask.pdf.rgb24.ref.png} | Bin test/{mask-quartz-ref.png => mask.quartz.ref.png} | Bin ...rtz-rgb24-ref.png => mask.quartz.rgb24.ref.png} | Bin test/{mask-ref.png => mask.ref.png} | Bin test/{mask-rgb24-ref.png => mask.rgb24.ref.png} | Bin ...11-argb32-ref.png => mask.svg11.argb32.ref.png} | Bin ...vg11-rgb24-ref.png => mask.svg11.rgb24.ref.png} | Bin ...12-argb32-ref.png => mask.svg12.argb32.ref.png} | Bin ...vg12-rgb24-ref.png => mask.svg12.rgb24.ref.png} | Bin ...df-ref.png => meta-surface-pattern.pdf.ref.png} | Bin ....png => meta-surface-pattern.pdf.rgb24.ref.png} | Bin ...png => meta-surface-pattern.ps2.argb32.ref.png} | Bin ....png => meta-surface-pattern.ps2.rgb24.ref.png} | Bin ...png => meta-surface-pattern.ps3.argb32.ref.png} | Bin ....png => meta-surface-pattern.ps3.rgb24.ref.png} | Bin ...ref.png => meta-surface-pattern.quartz.ref.png} | Bin ...g => meta-surface-pattern.quartz.rgb24.ref.png} | Bin ...attern-ref.png => meta-surface-pattern.ref.png} | Bin ...-ref.png => meta-surface-pattern.rgb24.ref.png} | Bin ...g => meta-surface-pattern.svg11.argb32.ref.png} | Bin ...ng => meta-surface-pattern.svg11.rgb24.ref.png} | Bin ...g => meta-surface-pattern.svg12.argb32.ref.png} | Bin ...ng => meta-surface-pattern.svg12.rgb24.ref.png} | Bin ...ion-ps2-ref.png => miter-precision.ps2.ref.png} | Bin ...ion-ps3-ref.png => miter-precision.ps3.ref.png} | Bin ...r-precision-ref.png => miter-precision.ref.png} | Bin ...urface-ref.png => move-to-show-surface.ref.png} | Bin test/multi-page.c | 4 +- ...b32-ref.png => new-sub-path.ps2.argb32.ref.png} | Bin ...gb24-ref.png => new-sub-path.ps2.rgb24.ref.png} | Bin ...b32-ref.png => new-sub-path.ps3.argb32.ref.png} | Bin ...gb24-ref.png => new-sub-path.ps3.rgb24.ref.png} | Bin ...-quartz-ref.png => new-sub-path.quartz.ref.png} | Bin ...4-ref.png => new-sub-path.quartz.rgb24.ref.png} | Bin .../{new-sub-path-ref.png => new-sub-path.ref.png} | Bin ...th-rgb24-ref.png => new-sub-path.rgb24.ref.png} | Bin test/{nil-surface-ref.png => nil-surface.ref.png} | Bin ...ace-rgb24-ref.png => nil-surface.rgb24.ref.png} | Bin ...erator-alpha-ref.png => operator-alpha.ref.png} | Bin ...-rgb24-ref.png => operator-alpha.rgb24.ref.png} | Bin ...2-ref.png => operator-clear.pdf.argb32.ref.png} | Bin ...24-ref.png => operator-clear.pdf.rgb24.ref.png} | Bin ...uartz-ref.png => operator-clear.quartz.ref.png} | Bin ...ref.png => operator-clear.quartz.rgb24.ref.png} | Bin ...erator-clear-ref.png => operator-clear.ref.png} | Bin ...-rgb24-ref.png => operator-clear.rgb24.ref.png} | Bin ...-ref.png => operator-source.pdf.argb32.ref.png} | Bin ...4-ref.png => operator-source.pdf.rgb24.ref.png} | Bin ...artz-ref.png => operator-source.quartz.ref.png} | Bin ...ef.png => operator-source.quartz.rgb24.ref.png} | Bin ...ator-source-ref.png => operator-source.ref.png} | Bin ...rgb24-ref.png => operator-source.rgb24.ref.png} | Bin test/{operator-ref.png => operator.ref.png} | Bin ...erator-rgb24-ref.png => operator.rgb24.ref.png} | Bin ...ef.png => over-above-source.ps2.argb32.ref.png} | Bin ...ef.png => over-above-source.ps3.argb32.ref.png} | Bin ...tz-ref.png => over-above-source.quartz.ref.png} | Bin ....png => over-above-source.quartz.rgb24.ref.png} | Bin ...ve-source-ref.png => over-above-source.ref.png} | Bin ...b24-ref.png => over-above-source.rgb24.ref.png} | Bin ...f.png => over-around-source.ps2.argb32.ref.png} | Bin ...ef.png => over-around-source.ps2.rgb24.ref.png} | Bin ...f.png => over-around-source.ps3.argb32.ref.png} | Bin ...ef.png => over-around-source.ps3.rgb24.ref.png} | Bin ...z-ref.png => over-around-source.quartz.ref.png} | Bin ...png => over-around-source.quartz.rgb24.ref.png} | Bin ...d-source-ref.png => over-around-source.ref.png} | Bin ...24-ref.png => over-around-source.rgb24.ref.png} | Bin ...ef.png => over-below-source.ps2.argb32.ref.png} | Bin ...ef.png => over-below-source.ps3.argb32.ref.png} | Bin ...ow-source-ref.png => over-below-source.ref.png} | Bin ...b24-ref.png => over-below-source.rgb24.ref.png} | Bin ....png => over-between-source.ps2.argb32.ref.png} | Bin ....png => over-between-source.ps3.argb32.ref.png} | Bin ...-ref.png => over-between-source.quartz.ref.png} | Bin ...ng => over-between-source.quartz.rgb24.ref.png} | Bin ...-source-ref.png => over-between-source.ref.png} | Bin ...4-ref.png => over-between-source.rgb24.ref.png} | Bin .../{paint-repeat-ref.png => paint-repeat.ref.png} | Bin ...f.png => paint-source-alpha.pdf.argb32.ref.png} | Bin ...-pdf-ref.png => paint-source-alpha.pdf.ref.png} | Bin ...ce-alpha-ref.png => paint-source-alpha.ref.png} | Bin ...11-ref.png => paint-source-alpha.svg11.ref.png} | Bin ...12-ref.png => paint-source-alpha.svg12.ref.png} | Bin ...ha-pdf-ref.png => paint-with-alpha.pdf.ref.png} | Bin ...with-alpha-ref.png => paint-with-alpha.ref.png} | Bin ...vg11-ref.png => paint-with-alpha.svg11.ref.png} | Bin ...vg12-ref.png => paint-with-alpha.svg12.ref.png} | Bin test/{paint-ref.png => paint.ref.png} | Bin .../{pass-through-ref.png => pass-through.ref.png} | Bin ...gh-rgb24-ref.png => pass-through.rgb24.ref.png} | Bin ...ern-getters-ref.png => pattern-getters.ref.png} | Bin test/pdf-features.c | 2 +- test/pdf-surface-source.c | 2 +- ...e-source-ref.png => pdf-surface-source.ref.png} | Bin ...pixman-rotate-ref.png => pixman-rotate.ref.png} | Bin ...e-rgb24-ref.png => pixman-rotate.rgb24.ref.png} | Bin test/png.c | 2 +- test/ps-features.c | 2 +- test/ps-surface-source.c | 2 +- ...ce-source-ref.png => ps-surface-source.ref.png} | Bin ...sh-group-pdf-ref.png => push-group.pdf.ref.png} | Bin ...-rgb24-ref.png => push-group.pdf.rgb24.ref.png} | Bin test/{push-group-ref.png => push-group.ref.png} | Bin ...roup-rgb24-ref.png => push-group.rgb24.ref.png} | Bin ...b32-ref.png => push-group.svg11.argb32.ref.png} | Bin ...b32-ref.png => push-group.svg12.argb32.ref.png} | Bin ...ent-pdf-ref.png => radial-gradient.pdf.ref.png} | Bin ...artz-ref.png => radial-gradient.quartz.ref.png} | Bin ...al-gradient-ref.png => radial-gradient.ref.png} | Bin ...svg11-ref.png => radial-gradient.svg11.ref.png} | Bin ...svg12-ref.png => radial-gradient.svg12.ref.png} | Bin ...s2-ref.png => random-intersections.ps2.ref.png} | Bin ...s3-ref.png => random-intersections.ps3.ref.png} | Bin ...ref.png => random-intersections.quartz.ref.png} | Bin ...ctions-ref.png => random-intersections.ref.png} | Bin ...or-ref.png => rectangle-rounding-error.ref.png} | Bin ...inear-fill-ref.png => rectilinear-fill.ref.png} | Bin ...ref.png => rectilinear-miter-limit.ps2.ref.png} | Bin ...ref.png => rectilinear-miter-limit.ps3.ref.png} | Bin ...mit-ref.png => rectilinear-miter-limit.ref.png} | Bin ...r-stroke-ref.png => rectilinear-stroke.ref.png} | Bin ...ke-ps2-ref.png => reflected-stroke.ps2.ref.png} | Bin ...ke-ps3-ref.png => reflected-stroke.ps3.ref.png} | Bin ...rtz-ref.png => reflected-stroke.quartz.ref.png} | Bin ...ted-stroke-ref.png => reflected-stroke.ref.png} | Bin ...s2-rgb24-ref.png => rel-path.ps2.rgb24.ref.png} | Bin ...s3-rgb24-ref.png => rel-path.ps3.rgb24.ref.png} | Bin ...path-quartz-ref.png => rel-path.quartz.ref.png} | Bin ...rgb24-ref.png => rel-path.quartz.rgb24.ref.png} | Bin test/{rel-path-ref.png => rel-path.ref.png} | Bin ...l-path-rgb24-ref.png => rel-path.rgb24.ref.png} | Bin ...re-alpha-ref.png => rgb24-ignore-alpha.ref.png} | Bin ... rotate-image-surface-paint.pdf.argb32.ref.png} | Bin ...> rotate-image-surface-paint.pdf.rgb24.ref.png} | Bin ....png => rotate-image-surface-paint.ps2.ref.png} | Bin ....png => rotate-image-surface-paint.ps3.ref.png} | Bin ...g => rotate-image-surface-paint.quartz.ref.png} | Bin ...-ref.png => rotate-image-surface-paint.ref.png} | Bin ...ng => rotate-image-surface-paint.svg11.ref.png} | Bin ...ng => rotate-image-surface-paint.svg12.ref.png} | Bin ...png => scale-down-source-surface-paint.ref.png} | Bin ... scale-source-surface-paint.pdf.argb32.ref.png} | Bin ...> scale-source-surface-paint.pdf.rgb24.ref.png} | Bin ...-ref.png => scale-source-surface-paint.ref.png} | Bin ...ng => scale-source-surface-paint.rgb24.ref.png} | Bin ...cale-source-surface-paint.svg11.argb32.ref.png} | Bin ...scale-source-surface-paint.svg11.rgb24.ref.png} | Bin ...cale-source-surface-paint.svg12.argb32.ref.png} | Bin ...scale-source-surface-paint.svg12.rgb24.ref.png} | Bin ...ce-ps2-ref.png => select-font-face.ps2.ref.png} | Bin ...ce-ps3-ref.png => select-font-face.ps3.ref.png} | Bin ...rtz-ref.png => select-font-face.quartz.ref.png} | Bin ...-font-face-ref.png => select-font-face.ref.png} | Bin ...y-overlap-ref.png => self-copy-overlap.ref.png} | Bin ...b24-ref.png => self-copy-overlap.rgb24.ref.png} | Bin ...self-copy-ps2-ref.png => self-copy.ps2.ref.png} | Bin ...self-copy-ps3-ref.png => self-copy.ps3.ref.png} | Bin test/{self-copy-ref.png => self-copy.ref.png} | Bin ...ersecting-ref.png => self-intersecting.ref.png} | Bin ...b24-ref.png => self-intersecting.rgb24.ref.png} | Bin test/{set-source-ref.png => set-source.ref.png} | Bin ...urce-rgb24-ref.png => set-source.rgb24.ref.png} | Bin ...b32-ref.png => set-source.svg11.argb32.ref.png} | Bin ...b32-ref.png => set-source.svg12.argb32.ref.png} | Bin ...lyphs-many-ref.png => show-glyphs-many.ref.png} | Bin ...ref.png => show-text-current-point.ps2.ref.png} | Bin ...ref.png => show-text-current-point.ps3.ref.png} | Bin ....png => show-text-current-point.quartz.ref.png} | Bin ...int-ref.png => show-text-current-point.ref.png} | Bin ...xtreme-ps2-ref.png => skew-extreme.ps2.ref.png} | Bin ...xtreme-ps3-ref.png => skew-extreme.ps3.ref.png} | Bin .../{skew-extreme-ref.png => skew-extreme.ref.png} | Bin ...ask-fill-pdf-ref.png => smask-fill.pdf.ref.png} | Bin test/{smask-fill-ref.png => smask-fill.ref.png} | Bin ...fill-svg11-ref.png => smask-fill.svg11.ref.png} | Bin ...fill-svg12-ref.png => smask-fill.svg12.ref.png} | Bin ...sk-pdf-ref.png => smask-image-mask.pdf.ref.png} | Bin ...image-mask-ref.png => smask-image-mask.ref.png} | Bin ...ask-mask-pdf-ref.png => smask-mask.pdf.ref.png} | Bin test/{smask-mask-ref.png => smask-mask.ref.png} | Bin ...mask-svg11-ref.png => smask-mask.svg11.ref.png} | Bin ...mask-svg12-ref.png => smask-mask.svg12.ref.png} | Bin ...k-paint-pdf-ref.png => smask-paint.pdf.ref.png} | Bin test/{smask-paint-ref.png => smask-paint.ref.png} | Bin ...int-svg11-ref.png => smask-paint.svg11.ref.png} | Bin ...int-svg12-ref.png => smask-paint.svg12.ref.png} | Bin ...stroke-pdf-ref.png => smask-stroke.pdf.ref.png} | Bin .../{smask-stroke-ref.png => smask-stroke.ref.png} | Bin ...ask-text-pdf-ref.png => smask-text.pdf.ref.png} | Bin ...ask-text-ps2-ref.png => smask-text.ps2.ref.png} | Bin ...ask-text-ps3-ref.png => smask-text.ps3.ref.png} | Bin test/{smask-text-ref.png => smask-text.ref.png} | Bin ...text-svg11-ref.png => smask-text.svg11.ref.png} | Bin ...text-svg12-ref.png => smask-text.svg12.ref.png} | Bin test/{smask-pdf-ref.png => smask.pdf.ref.png} | Bin test/{smask-ps2-ref.png => smask.ps2.ref.png} | Bin test/{smask-ps3-ref.png => smask.ps3.ref.png} | Bin test/{smask-ref.png => smask.ref.png} | Bin test/{smask-svg11-ref.png => smask.svg11.ref.png} | Bin test/{smask-svg12-ref.png => smask.svg12.ref.png} | Bin ...-ref.png => solid-pattern-cache-stress.ref.png} | Bin ...e-pdf-ref.png => source-clip-scale.pdf.ref.png} | Bin ...ef.png => source-clip-scale.ps2.argb32.ref.png} | Bin ...ref.png => source-clip-scale.ps2.rgb24.ref.png} | Bin ...ef.png => source-clip-scale.ps3.argb32.ref.png} | Bin ...ref.png => source-clip-scale.ps3.rgb24.ref.png} | Bin ...tz-ref.png => source-clip-scale.quartz.ref.png} | Bin ...lip-scale-ref.png => source-clip-scale.ref.png} | Bin ...g11-ref.png => source-clip-scale.svg11.ref.png} | Bin ...g12-ref.png => source-clip-scale.svg12.ref.png} | Bin test/{source-clip-ref.png => source-clip.ref.png} | Bin ...-ref.png => source-surface-scale-paint.ref.png} | Bin ...ng => source-surface-scale-paint.rgb24.ref.png} | Bin ...df-ref.png => spline-decomposition.pdf.ref.png} | Bin ...s2-ref.png => spline-decomposition.ps2.ref.png} | Bin ...s3-ref.png => spline-decomposition.ps3.ref.png} | Bin ...sition-ref.png => spline-decomposition.ref.png} | Bin ...-ref.png => spline-decomposition.svg11.ref.png} | Bin ...-ref.png => spline-decomposition.svg12.ref.png} | Bin ...aps-ps2-ref.png => stroke-ctm-caps.ps2.ref.png} | Bin ...aps-ps3-ref.png => stroke-ctm-caps.ps3.ref.png} | Bin ...artz-ref.png => stroke-ctm-caps.quartz.ref.png} | Bin ...ke-ctm-caps-ref.png => stroke-ctm-caps.ref.png} | Bin ...-image-pdf-ref.png => stroke-image.pdf.ref.png} | Bin ...-image-ps2-ref.png => stroke-image.ps2.ref.png} | Bin ...-image-ps3-ref.png => stroke-image.ps3.ref.png} | Bin ...-quartz-ref.png => stroke-image.quartz.ref.png} | Bin .../{stroke-image-ref.png => stroke-image.ref.png} | Bin ....png => surface-pattern-big-scale-down.ref.png} | Bin ... surface-pattern-scale-down.pdf.argb32.ref.png} | Bin ...> surface-pattern-scale-down.pdf.rgb24.ref.png} | Bin ....png => surface-pattern-scale-down.ps2.ref.png} | Bin ....png => surface-pattern-scale-down.ps3.ref.png} | Bin ...g => surface-pattern-scale-down.quartz.ref.png} | Bin ...-ref.png => surface-pattern-scale-down.ref.png} | Bin ...=> surface-pattern-scale-up.pdf.argb32.ref.png} | Bin ... => surface-pattern-scale-up.pdf.rgb24.ref.png} | Bin ...ef.png => surface-pattern-scale-up.ps2.ref.png} | Bin ...ef.png => surface-pattern-scale-up.ps3.ref.png} | Bin ...up-ref.png => surface-pattern-scale-up.ref.png} | Bin ...ern-pdf-ref.png => surface-pattern.pdf.ref.png} | Bin ...ern-ps2-ref.png => surface-pattern.ps2.ref.png} | Bin ...ern-ps3-ref.png => surface-pattern.ps3.ref.png} | Bin ...ace-pattern-ref.png => surface-pattern.ref.png} | Bin ...svg11-ref.png => surface-pattern.svg11.ref.png} | Bin ...svg12-ref.png => surface-pattern.svg12.ref.png} | Bin test/svg-clip.c | 2 +- test/svg-surface-source.c | 2 +- ...e-source-ref.png => svg-surface-source.ref.png} | Bin test/svg-surface.c | 2 +- ...-ref.png => text-antialias-gray.quartz.ref.png} | Bin ...as-gray-ref.png => text-antialias-gray.ref.png} | Bin ...-ref.png => text-antialias-none.quartz.ref.png} | Bin ...as-none-ref.png => text-antialias-none.ref.png} | Bin ....png => text-antialias-subpixel.quartz.ref.png} | Bin ...xel-ref.png => text-antialias-subpixel.ref.png} | Bin ...ge-ps2-ref.png => text-glyph-range.ps2.ref.png} | Bin ...ge-ps3-ref.png => text-glyph-range.ps3.ref.png} | Bin ...lyph-range-ref.png => text-glyph-range.ref.png} | Bin ...b32-ref.png => text-pattern.pdf.argb32.ref.png} | Bin ...gb24-ref.png => text-pattern.pdf.rgb24.ref.png} | Bin ...b32-ref.png => text-pattern.ps3.argb32.ref.png} | Bin ...gb24-ref.png => text-pattern.ps3.rgb24.ref.png} | Bin ...-quartz-ref.png => text-pattern.quartz.ref.png} | Bin ...4-ref.png => text-pattern.quartz.rgb24.ref.png} | Bin .../{text-pattern-ref.png => text-pattern.ref.png} | Bin ...rn-rgb24-ref.png => text-pattern.rgb24.ref.png} | Bin ...2-ref.png => text-pattern.svg11.argb32.ref.png} | Bin ...24-ref.png => text-pattern.svg11.rgb24.ref.png} | Bin ...2-ref.png => text-pattern.svg12.argb32.ref.png} | Bin ...24-ref.png => text-pattern.svg12.rgb24.ref.png} | Bin ...-rotate-pdf-ref.png => text-rotate.pdf.ref.png} | Bin ...-rotate-ps2-ref.png => text-rotate.ps2.ref.png} | Bin ...-rotate-ps3-ref.png => text-rotate.ps3.ref.png} | Bin ...e-quartz-ref.png => text-rotate.quartz.ref.png} | Bin test/{text-rotate-ref.png => text-rotate.ref.png} | Bin ...ate-svg11-ref.png => text-rotate.svg11.ref.png} | Bin ...ate-svg12-ref.png => text-rotate.svg12.ref.png} | Bin ...form-pdf-ref.png => text-transform.pdf.ref.png} | Bin ...form-ps2-ref.png => text-transform.ps2.ref.png} | Bin ...form-ps3-ref.png => text-transform.ps3.ref.png} | Bin ...xt-transform-ref.png => text-transform.ref.png} | Bin ...ansforms-ps2-ref.png => transforms.ps2.ref.png} | Bin ...ansforms-ps3-ref.png => transforms.ps3.ref.png} | Bin test/{transforms-ref.png => transforms.ref.png} | Bin ...face-ref.png => translate-show-surface.ref.png} | Bin ...argb32-ref.png => trap-clip.pdf.argb32.ref.png} | Bin ...f-rgb24-ref.png => trap-clip.pdf.rgb24.ref.png} | Bin ...argb32-ref.png => trap-clip.ps2.argb32.ref.png} | Bin ...2-rgb24-ref.png => trap-clip.ps2.rgb24.ref.png} | Bin ...argb32-ref.png => trap-clip.ps3.argb32.ref.png} | Bin ...3-rgb24-ref.png => trap-clip.ps3.rgb24.ref.png} | Bin ...lip-quartz-ref.png => trap-clip.quartz.ref.png} | Bin ...gb24-ref.png => trap-clip.quartz.rgb24.ref.png} | Bin test/{trap-clip-ref.png => trap-clip.ref.png} | Bin ...-clip-rgb24-ref.png => trap-clip.rgb24.ref.png} | Bin test/{twin-ps2-ref.png => twin.ps2.ref.png} | Bin test/{twin-ps3-ref.png => twin.ps3.ref.png} | Bin test/{twin-ref.png => twin.ref.png} | Bin test/{twin-svg11-ref.png => twin.svg11.ref.png} | Bin test/{twin-svg12-ref.png => twin.svg12.ref.png} | Bin ...ref.png => unantialiased-shapes.quartz.ref.png} | Bin ...shapes-ref.png => unantialiased-shapes.ref.png} | Bin ...z-ref.png => unbounded-operator.quartz.ref.png} | Bin ...png => unbounded-operator.quartz.rgb24.ref.png} | Bin ...operator-ref.png => unbounded-operator.ref.png} | Bin ...24-ref.png => unbounded-operator.rgb24.ref.png} | Bin ...mask-pdf-ref.png => user-font-mask.pdf.ref.png} | Bin ...mask-ps2-ref.png => user-font-mask.ps2.ref.png} | Bin ...mask-ps3-ref.png => user-font-mask.ps3.ref.png} | Bin ...er-font-mask-ref.png => user-font-mask.ref.png} | Bin ...-svg11-ref.png => user-font-mask.svg11.ref.png} | Bin ...oxy-pdf-ref.png => user-font-proxy.pdf.ref.png} | Bin ...oxy-ps2-ref.png => user-font-proxy.ps2.ref.png} | Bin ...oxy-ps3-ref.png => user-font-proxy.ps3.ref.png} | Bin ...-font-proxy-ref.png => user-font-proxy.ref.png} | Bin ...svg11-ref.png => user-font-proxy.svg11.ref.png} | Bin ...svg12-ref.png => user-font-proxy.svg12.ref.png} | Bin ...e-ps2-ref.png => user-font-rescale.ps2.ref.png} | Bin ...e-ps3-ref.png => user-font-rescale.ps3.ref.png} | Bin ...t-rescale-ref.png => user-font-rescale.ref.png} | Bin ...g11-ref.png => user-font-rescale.svg11.ref.png} | Bin ...g12-ref.png => user-font-rescale.svg12.ref.png} | Bin ...user-font-ps2-ref.png => user-font.ps2.ref.png} | Bin ...user-font-ps3-ref.png => user-font.ps3.ref.png} | Bin test/{user-font-ref.png => user-font.ref.png} | Bin ...-font-svg11-ref.png => user-font.svg11.ref.png} | Bin ...-font-svg12-ref.png => user-font.svg12.ref.png} | Bin test/xlib-expose-event.c | 4 +- ...ose-event-ref.png => xlib-expose-event.ref.png} | Bin ...ps2-ref.png => xlib-surface-source.ps2.ref.png} | Bin ...ps3-ref.png => xlib-surface-source.ps3.ref.png} | Bin ...-source-ref.png => xlib-surface-source.ref.png} | Bin test/{zero-alpha-ref.png => zero-alpha.ref.png} | Bin 715 files changed, 782 insertions(+), 752 deletions(-) commit e90073f7ddc6f461a935bc360c409b04f1fe9f74 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Sep 3 16:38:03 2008 +0100 [test] Build test suite into single binary. Avoid calling libtool to link every single test case, by building just one binary from all the sources. This binary is then given the task of choosing tests to run (based on user selection and individual test requirement), forking each test into its own process and accumulating the results. Makefile.am | 1 + boilerplate/cairo-boilerplate.c | 658 +++++++++------- boilerplate/cairo-boilerplate.h | 1 + build/configure.ac.noversion | 12 + build/configure.ac.system | 57 ++ build/configure.ac.version | 8 - configure.ac | 16 + doc/public/tmpl/cairo-surface.sgml | 1 + test/.gitignore | 237 +----- test/Makefile.am | 572 +++++++------- test/README | 28 +- test/a1-image-sample.c | 20 +- test/a1-mask.c | 38 +- test/a1-traps-sample.c | 20 +- test/a8-mask.c | 37 +- test/alpha-similar.c | 23 +- test/big-line.c | 22 +- test/big-trap.c | 22 +- test/bilevel-image.c | 20 +- test/bitmap-font.c | 22 +- test/cairo-test-private.h | 74 ++ test/cairo-test-runner.c | 907 ++++++++++++++++++++++ test/cairo-test.c | 594 +++++++------- test/cairo-test.h | 99 ++- test/caps-joins-alpha.c | 26 +- test/caps-joins.c | 27 +- test/caps-sub-paths.c | 21 +- test/clip-all.c | 21 +- test/clip-empty.c | 20 +- test/clip-fill-rule-pixel-aligned.c | 22 +- test/clip-fill-rule.c | 20 +- test/clip-nesting.c | 20 +- test/clip-operator.c | 21 +- test/clip-push-group.c | 20 +- test/clip-twice.c | 20 +- test/clip-zero.c | 20 +- test/clipped-group.c | 20 +- test/clipped-surface.c | 19 +- test/close-path.c | 18 +- test/composite-integer-translate-over-repeat.c | 20 +- test/composite-integer-translate-over.c | 22 +- test/composite-integer-translate-source.c | 22 +- test/copy-path.c | 33 +- test/create-for-stream.c | 53 +- test/create-from-png-stream.c | 20 +- test/create-from-png.c | 82 +- test/dash-caps-joins.c | 26 +- test/dash-curve.c | 20 +- test/dash-no-dash.c | 20 +- test/dash-offset-negative.c | 20 +- test/dash-scale.c | 26 +- test/dash-state.c | 22 +- test/dash-zero-length.c | 20 +- test/degenerate-arc.c | 20 +- test/degenerate-dash.c | 20 +- test/degenerate-path.c | 21 +- test/degenerate-pen.c | 20 +- test/device-offset-fractional.c | 20 +- test/device-offset-positive.c | 20 +- test/device-offset-scale.c | 21 +- test/device-offset.c | 20 +- test/extend-pad.c | 20 +- test/extend-reflect-similar.c | 22 +- test/extend-reflect.c | 22 +- test/extend-repeat-similar.c | 22 +- test/extend-repeat.c | 22 +- test/fallback-resolution.c | 28 +- test/fill-alpha-pattern.c | 20 +- test/fill-alpha.c | 20 +- test/fill-and-stroke-alpha-add.c | 20 +- test/fill-and-stroke-alpha.c | 20 +- test/fill-and-stroke.c | 20 +- test/fill-degenerate-sort-order.c | 20 +- test/fill-missed-stop.c | 19 +- test/fill-rule.c | 20 +- test/filter-bilinear-extents.c | 22 +- test/filter-nearest-offset.c | 21 +- test/filter-nearest-transformed.c | 19 +- test/finer-grained-fallbacks.c | 20 +- test/font-face-get-type.c | 28 +- test/font-matrix-translation.c | 20 +- test/font-options.c | 17 +- test/ft-font-create-for-ft-face.c | 21 +- test/ft-show-glyphs-positioning.c | 20 +- test/ft-show-glyphs-table.c | 21 +- test/ft-text-antialias-none.c | 20 +- test/ft-text-vertical-layout-type1.c | 23 +- test/ft-text-vertical-layout-type3.c | 20 +- test/get-and-set.c | 20 +- test/get-clip.c | 21 +- test/get-group-target.c | 20 +- test/get-path-extents.c | 20 +- test/get-xrender-format.c | 35 +- test/glyph-cache-pressure.c | 20 +- test/gradient-alpha.c | 20 +- test/gradient-constant-alpha.c | 20 +- test/gradient-zero-stops.c | 22 +- test/group-paint.c | 20 +- test/huge-pattern.c | 21 +- test/image-surface-source.c | 8 +- test/in-fill-empty-trapezoid.c | 24 +- test/in-fill-trapezoid.c | 20 +- test/infinite-join.c | 22 +- test/invalid-matrix.c | 20 +- test/large-clip.c | 22 +- test/large-font.c | 22 +- test/large-source.c | 20 +- test/leaky-dash.c | 22 +- test/leaky-dashed-rectangle.c | 20 +- test/leaky-dashed-stroke.c | 20 +- test/leaky-polygon.c | 20 +- test/line-width-scale.c | 20 +- test/line-width-zero.c | 20 +- test/line-width.c | 20 +- test/linear-gradient-reflect.c | 21 +- test/linear-gradient.c | 20 +- test/long-dashed-lines.c | 20 +- test/long-lines.c | 24 +- test/make-cairo-test-constructors.pl | 18 + test/mask-alpha.c | 20 +- test/mask-ctm.c | 21 +- test/mask-surface-ctm.c | 20 +- test/mask-transformed-image.c | 19 +- test/mask-transformed-similar.c | 19 +- test/mask.c | 23 +- test/meta-surface-pattern.c | 20 +- test/miter-precision.c | 22 +- test/move-to-show-surface.c | 20 +- test/multi-page.c | 30 +- test/new-sub-path.c | 22 +- test/nil-surface.c | 20 +- test/operator-alpha.c | 21 +- test/operator-clear.c | 20 +- test/operator-source.c | 20 +- test/operator.c | 21 +- test/over-above-source.c | 20 +- test/over-around-source.c | 20 +- test/over-below-source.c | 20 +- test/over-between-source.c | 20 +- test/paint-repeat.c | 20 +- test/paint-source-alpha.c | 20 +- test/paint-with-alpha.c | 20 +- test/paint.c | 20 +- test/pattern-get-type.c | 21 +- test/pattern-getters.c | 20 +- test/pdf-features.c | 31 +- test/pdf-surface-source.c | 10 +- test/pixman-rotate.c | 20 +- test/png.c | 56 +- test/ps-features.c | 29 +- test/ps-surface-source.c | 10 +- test/pthread-show-text.c | 41 +- test/push-group.c | 20 +- test/radial-gradient.c | 22 +- test/random-intersections.c | 21 +- test/rectangle-rounding-error.c | 22 +- test/rectilinear-fill.c | 20 +- test/rectilinear-miter-limit.c | 20 +- test/rectilinear-stroke.c | 21 +- test/reflected-stroke.c | 21 +- test/rel-path.c | 20 +- test/rgb24-ignore-alpha.c | 20 +- test/rotate-image-surface-paint.c | 22 +- test/scale-down-source-surface-paint.c | 20 +- test/scale-source-surface-paint.c | 20 +- test/select-font-face.c | 28 +- test/select-font-no-show-text.c | 20 +- test/self-copy-overlap.c | 22 +- test/self-copy.c | 20 +- test/self-intersecting.c | 22 +- test/set-source.c | 20 +- test/show-glyphs-many.c | 20 +- test/show-text-current-point.c | 22 +- test/skew-extreme.c | 20 +- test/smask-fill.c | 20 +- test/smask-image-mask.c | 20 +- test/smask-mask.c | 20 +- test/smask-paint.c | 20 +- test/smask-stroke.c | 20 +- test/smask-text.c | 20 +- test/smask.c | 20 +- test/solid-pattern-cache-stress.c | 20 +- test/source-clip-scale.c | 20 +- test/source-clip.c | 20 +- test/source-surface-scale-paint.c | 20 +- test/spline-decomposition.c | 20 +- test/stroke-ctm-caps.c | 20 +- test/stroke-image.c | 20 +- test/surface-finish-twice.c | 20 +- test/surface-pattern-big-scale-down.c | 20 +- test/surface-pattern-scale-down-pdf-rgb24-ref.png | Bin 0 -> 2386 bytes test/surface-pattern-scale-down.c | 21 +- test/surface-pattern-scale-up.c | 21 +- test/surface-pattern.c | 24 +- test/surface-source.c | 15 +- test/svg-clip.c | 26 +- test/svg-surface-source.c | 11 +- test/svg-surface.c | 26 +- test/text-antialias-gray.c | 20 +- test/text-antialias-none.c | 20 +- test/text-antialias-subpixel.c | 20 +- test/text-cache-crash.c | 40 +- test/text-glyph-range-ps2-ref.png | Bin 0 -> 1389 bytes test/text-glyph-range-ps3-ref.png | Bin 0 -> 1389 bytes test/text-glyph-range-ref.png | Bin 1810 -> 2211 bytes test/text-glyph-range-rgb24-ref.png | Bin 1645 -> 0 bytes test/text-glyph-range.c | 62 +- test/text-pattern.c | 54 +- test/text-rotate.c | 22 +- test/text-transform.c | 22 +- test/text-zero-len.c | 20 +- test/toy-font-face.c | 16 +- test/transforms.c | 20 +- test/translate-show-surface.c | 20 +- test/trap-clip.c | 22 +- test/truetype-tables.c | 18 +- test/twin.c | 20 +- test/unantialiased-shapes.c | 20 +- test/unbounded-operator.c | 20 +- test/user-data.c | 17 +- test/user-font-mask.c | 28 +- test/user-font-proxy.c | 28 +- test/user-font-rescale.c | 20 +- test/user-font.c | 28 +- test/xlib-expose-event.c | 28 +- test/xlib-surface-source.c | 8 +- test/xlib-surface.c | 36 +- test/zero-alpha.c | 20 +- 228 files changed, 3770 insertions(+), 4123 deletions(-) commit 8457972d40088cda165f31fdd7bd9b4c19c6e095 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 31 02:16:54 2008 +0000 [type3] Whitespace. Tightly scope the output stream. src/cairo-type3-glyph-surface.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit 08f4d49a9464bc19bc2ca7f2356a6eb8d62f5daf Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 31 02:03:41 2008 +0000 [xlib] Intialise clip_dirty. Eeek, forgot to set the clip_dirty to 0 on construction. src/cairo-xlib-surface.c | 1 + 1 file changed, 1 insertion(+) commit b722d5b2b988d43931648d8377a9433d75f7d95f Merge: 05fce5bce 710f7d30d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 31 02:24:25 2008 +0000 Merge branch '1.8' commit 710f7d30d348517602293f123525aef7677d69cf Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 31 02:14:53 2008 +0000 [scaled-font] Fix typo that prevented meta_surface lookup. test/text-glyph-range was crashing since we requested the meta_surface but the glyph wrongly believed it already had the meta_surface but was checking for a path instead. src/cairo-scaled-font.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 05fce5bced9718ffa803f5a87df00f96809d1f53 Author: Adrian Johnson <ajohnson@redneon.com> Date: Sat Apr 5 13:32:51 2008 +1030 PS/PDF: Set image Interpolation flag If the filter mode is anything other than DEFAILT, FAST or NEAREST set the Interpolate flag in the image dictionary so that a smoothing filter is applied when rasterising the vector file. As we have no control over the implementation of the Interpolate filter (the PS/PDF specifications leave it undefined) we need to capture the output of poppler/GS and update our reference images. (For a couple of tests, the filtering is irrelevant so for those we set the filter to NEAREST.) Note that GhostScript's Interpolate filter does not work on rotated images (and a variety of other transformations) so several of the PS reference images have use nearest-neighbour sampling instead of a bilinear filter. src/cairo-pdf-surface.c | 28 +++++++++-- src/cairo-ps-surface.c | 28 +++++++++-- test/Makefile.am | 54 +++++++++++++++------ test/bilevel-image.c | 1 + test/composite-integer-translate-over-ps2-ref.png | Bin 0 -> 15783 bytes test/composite-integer-translate-over-ps3-ref.png | Bin 0 -> 15783 bytes ...osite-integer-translate-over-repeat-ps2-ref.png | Bin 0 -> 448 bytes ...osite-integer-translate-over-repeat-ps3-ref.png | Bin 0 -> 448 bytes .../composite-integer-translate-source-ps2-ref.png | Bin 0 -> 15783 bytes .../composite-integer-translate-source-ps3-ref.png | Bin 0 -> 15783 bytes test/create-from-png-stream.c | 1 + test/create-from-png.c | 1 + test/device-offset-fractional-ps2-ref.png | Bin 201 -> 200 bytes test/device-offset-fractional-ps3-ref.png | Bin 201 -> 200 bytes test/extend-reflect-ps2-ref.png | Bin 153322 -> 146990 bytes test/extend-reflect-ps3-ref.png | Bin 153322 -> 146990 bytes test/extend-reflect-similar-ps2-ref.png | Bin 153322 -> 146990 bytes test/extend-reflect-similar-ps3-ref.png | Bin 153322 -> 146990 bytes test/extend-repeat-ps2-ref.png | Bin 0 -> 119246 bytes test/extend-repeat-ps3-ref.png | Bin 0 -> 119246 bytes test/extend-repeat-similar-ps2-ref.png | Bin 0 -> 119246 bytes test/extend-repeat-similar-ps3-ref.png | Bin 0 -> 119246 bytes test/filter-bilinear-extents-ps2-ref.png | Bin 173 -> 556 bytes test/filter-bilinear-extents-ps3-ref.png | Bin 173 -> 556 bytes test/image-surface-source-ps2-ref.png | Bin 0 -> 376 bytes test/image-surface-source-ps3-ref.png | Bin 0 -> 376 bytes test/image-surface-source-ref.png | Bin 268 -> 332 bytes test/stroke-image-ps2-ref.png | Bin 1786 -> 2121 bytes test/stroke-image-ps3-ref.png | Bin 1786 -> 2121 bytes test/surface-pattern-pdf-argb32-ref.png | Bin 14880 -> 0 bytes test/surface-pattern-pdf-ref.png | Bin 0 -> 15996 bytes test/surface-pattern-ps2-argb32-ref.png | Bin 2520 -> 0 bytes test/surface-pattern-ps2-ref.png | Bin 0 -> 2364 bytes test/surface-pattern-ps3-argb32-ref.png | Bin 2520 -> 0 bytes test/surface-pattern-ps3-ref.png | Bin 0 -> 2364 bytes test/surface-pattern-scale-down-pdf-argb32-ref.png | Bin 1565 -> 2386 bytes test/surface-pattern-scale-down-ps2-argb32-ref.png | Bin 1400 -> 0 bytes test/surface-pattern-scale-down-ps2-ref.png | Bin 0 -> 1324 bytes test/surface-pattern-scale-down-ps3-argb32-ref.png | Bin 1400 -> 0 bytes test/surface-pattern-scale-down-ps3-ref.png | Bin 0 -> 1324 bytes test/surface-pattern-scale-down.c | 3 +- test/surface-pattern-scale-up-pdf-argb32-ref.png | Bin 3886 -> 4247 bytes test/surface-pattern-scale-up-pdf-rgb24-ref.png | Bin 0 -> 4247 bytes test/surface-pattern-scale-up-ps2-argb32-ref.png | Bin 1011 -> 0 bytes test/surface-pattern-scale-up-ps2-ref.png | Bin 0 -> 943 bytes test/surface-pattern-scale-up-ps3-argb32-ref.png | Bin 1011 -> 0 bytes test/surface-pattern-scale-up-ps3-ref.png | Bin 0 -> 943 bytes test/surface-pattern-scale-up-ref.png | Bin 3825 -> 4061 bytes test/surface-pattern-scale-up.c | 6 ++- test/surface-pattern.c | 6 ++- test/xlib-surface-source-ps2-ref.png | Bin 0 -> 376 bytes test/xlib-surface-source-ps3-ref.png | Bin 0 -> 376 bytes 52 files changed, 98 insertions(+), 30 deletions(-) commit 2c53bdb82a955c672c8fee2b1b4c991af2654ae0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Sep 29 21:12:53 2008 +0100 [paginated] Emit fallback images using FILTER_NEAREST. Fallback images should (in theory) be emitted at native resolution, so disable filtering to avoid introducing potential artifacts into cairo's ideal output. src/cairo-paginated-surface.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) commit f03d3f9f7f923730cdc4c5f21364f752f705086a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Oct 30 19:24:29 2008 +0000 [boilerplate] Remove #if VERSION>1.9.0 Remove the obsolete #if statement. boilerplate/cairo-boilerplate.c | 2 -- 1 file changed, 2 deletions(-) commit ef98eb26f7484019b023cd4c93dce44f23e3a594 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Oct 30 19:23:17 2008 +0000 [test] Reenable device offset testing. Restore device offset testing that was temporarily disable in the run up to the 1.8 release. test/cairo-test.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 4f2f46ef1bd793a712df045178b3d00d027e3ba6 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Oct 30 19:14:07 2008 +0000 Restore the ability to choose the internal font. Behdad wants to include the feature with 1.10, so we enable it as early as possible in 1.9 dev cycle to generate as much feedback as possible. The first change is to use "<cairo>" as being a name unlikely to clash with any real font names. This reverts commits: a824d284be23793a5c48b9ae833dcb7b2d5fff80, 292233685534aed712dfd45e8ccf498b792ce496, e0046aaf417a61da008dc6374871fa3687ba94ab, f534bd549e1e2283735d1eabb60c015a5949a735. perf/Makefile.am | 1 + perf/cairo-perf.c | 1 + perf/cairo-perf.h | 1 + perf/twin.c | 50 ++++++++++++++++++++++++++++++++++++++++ src/cairo-font-face.c | 6 +++-- src/cairoint.h | 1 + test/.gitignore | 1 + test/Makefile.am | 6 +++++ test/twin-ps2-ref.png | Bin 0 -> 1167 bytes test/twin-ps3-ref.png | Bin 0 -> 1167 bytes test/twin-ref.png | Bin 0 -> 1836 bytes test/twin-svg11-ref.png | Bin 0 -> 1773 bytes test/twin-svg12-ref.png | Bin 0 -> 1773 bytes test/twin.c | 60 ++++++++++++++++++++++++++++++++++++++++++++++++ 14 files changed, 125 insertions(+), 2 deletions(-) commit 52c3fc58b52c77282998f9ad75657a6bec5956f8 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Oct 8 13:04:37 2008 +0100 [tessellator] Simplify dequeuing by using a sentinel value. Instead of maintaining an index and comparing it to the count, just mark the last startstop event with NULL and stop dequeuing events upon seeing that sentinel value. (Removes an unreadable line, and cachegrind hints that it may be a tiny bit faster.) src/cairo-bentley-ottmann.c | 32 ++++++++++++++++---------------- 1 file changed, 16 insertions(+), 16 deletions(-) commit ef9e0a5d1d74ac92a1fcde5a657c866a8e6509e6 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Oct 7 22:09:37 2008 +0100 [tessellator] Use a combsort for sorting the event queue. In my experiments using cairo-perf, the inlined combsort is ~20% quicker than the library qsort. src/Makefile.sources | 1 + src/cairo-bentley-ottmann.c | 39 ++++++++++++------------ src/cairo-combsort-private.h | 71 ++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 92 insertions(+), 19 deletions(-) commit b1461308416fa83d1de0016a9d4804b68a1f1d8f Author: Behdad Esfahbod <behdad@behdad.org> Date: Thu Oct 30 14:18:18 2008 -0400 [Makefile.am.releasing] Symlink /manual and /cairo-manual.tar.gz only for releases And not for snapshots. Also symlink /cairo-manual-X.Y.Z.tar.gz for all versions. build/Makefile.am.releasing | 13 +++++++++++-- 1 file changed, 11 insertions(+), 2 deletions(-) commit e3a7f522a6b96729b2a0122f8c430c24dc17fc5a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Oct 7 23:33:32 2008 +0100 [tessellator] Perform cheap checks before computing intersect. First check if we can reject the intersection without having to perform the full divides, based on analysing: t * (ady*bdx - bdy*adx) = bdy * (ax - bx) - bdx * (ay - by) s * (ady*bdx - bdy*adx) = ady * (ax - bx) - adx * (ay - by) and excluding the result if by inspection we know that (s, t) <= 0 || (s, t) => 1. Doing so virtually eliminates all division and speeds up the strokes (when performing self-intersection elimination using the tessellator) perf cases by about 5%. src/cairo-bentley-ottmann.c | 51 ++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 50 insertions(+), 1 deletion(-) commit 553fde4bb3e913de7e26bf416166d69bae4d02e1 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Oct 4 13:15:08 2008 +0100 [tessellator] Simplify special cases of edges to compare. Use our prior knowledge of the inputs and trivial conditions to simplify the edge equations and in many common conditions, such as vertical edges and common points, reduce the operations down to a just returning the non-degenerate 32 bit value. This adds an overhead of a few conditionals, but on the fast paths we actually generate fewer branches and many fewer arithmetic operations such that it improves performance of the fill performance tests by around 10%. src/cairo-bentley-ottmann.c | 132 +++++++++++++++++++++++++++++++++++--------- 1 file changed, 106 insertions(+), 26 deletions(-) commit cc109df2a70e953d71e3e6fc4e6e54cce4ba0d47 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Oct 30 15:50:20 2008 +0000 [test] Update .gitignore Add a couple of new programs to gitignore. test/.gitignore | 2 ++ 1 file changed, 2 insertions(+) commit 7d546bba213c3b89404bb3c58a5665343abfcd51 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Oct 29 15:46:10 2008 +0000 [xlib] Eliminate shadowed variable. Cleanup a minor -Wshadow warning. src/cairo-xlib-surface.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit e3388b8180e38f79e0d5b66cc640c3d5c37db2a8 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Oct 29 15:45:09 2008 +0000 [cff-subset] Eliminate shadowed variable. Cleanup a minor -Wshadow warning. src/cairo-cff-subset.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) commit 3e18114aff21acfa73a7d13a419719375ae9b5a6 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Oct 29 15:43:23 2008 +0000 [user-font] Eliminate shadowed variable Cleanup a minor -Wshadow warning. src/cairo-user-font.c | 1 - 1 file changed, 1 deletion(-) commit d073537e82b57738068a7e1cf1d36ce452876804 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Oct 29 15:41:44 2008 +0000 [scaled-font] Eliminate shadowed variable. Cleanup a minor -Wshadow warning. src/cairo-scaled-font.c | 1 - 1 file changed, 1 deletion(-) commit 6429e3f39468ef4f865681b3b5988d35906deeb0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Oct 28 17:38:09 2008 +0000 [xlib] Only set Picture clip as necessary. Only set ClipRectangles on a new Picture, avoiding redundant calls to clear the non-existent ClipMask otherwise. src/cairo-xlib-surface.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) commit cd45258c5bf6a06f7bd3f2bfaf928e423a9b70e1 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Oct 28 16:54:10 2008 +0000 [xlib] Only clear the GC clip mask as necessary. Avoid redundant calls to XSetClipMask() for clean GCs. src/cairo-xlib-private.h | 10 ++++++++-- src/cairo-xlib-screen.c | 8 ++++++-- src/cairo-xlib-surface.c | 7 ++++--- 3 files changed, 18 insertions(+), 7 deletions(-) commit d384f864992fbf7513446a70158135a08e704678 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Oct 28 16:33:51 2008 +0000 [xlib] Propagate real status from get_screen_info(). Return the real error from _cairo_xlib_screen_info_get() in order to avoid having to create a fake NO_MEMORY error. src/cairo-xlib-private.h | 6 +++-- src/cairo-xlib-screen.c | 66 +++++++++++++++++++++++++++--------------------- src/cairo-xlib-surface.c | 6 ++--- 3 files changed, 44 insertions(+), 34 deletions(-) commit e25b106e9c33e220fb104afb0929e766647b6de9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Oct 28 16:27:58 2008 +0000 [xlib] Propagate real status from get_display() Avoid throwing away the error and inventing a new NO_MEMORY error for _cairo_xlib_display_get(). src/cairo-xlib-display.c | 13 ++++++++----- src/cairo-xlib-private.h | 4 ++-- src/cairo-xlib-surface.c | 14 ++++++++------ 3 files changed, 18 insertions(+), 13 deletions(-) commit 2555f83b11ee23829cfdeaae703420ca3e572fa1 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Oct 28 16:22:49 2008 +0000 [xlib] Avoid repeated calls to XRenderQueryVersion Use the value determined during display initialisation in order to avoid redundant XRenderQueryFormats requests. src/cairo-xlib-display.c | 6 ++++-- src/cairo-xlib-private.h | 2 ++ src/cairo-xlib-surface.c | 7 ++----- 3 files changed, 8 insertions(+), 7 deletions(-) commit 0d0c6a199c5b631299c72dce80d66ac0f4936a64 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Oct 26 12:53:29 2008 +0000 [matrix] Optimise invert for simple scaling|translation matrices. The matrix is quite often just a simple scale and translate (or even identity!). For this class of matrix, we can skip the full adjoint rearrangement and determinant calculation and just compute the inverse directly. src/cairo-matrix.c | 26 +++++++++++++++++++++++++- 1 file changed, 25 insertions(+), 1 deletion(-) commit 74876b00cd969d5cfc209fa903dfad50bda59361 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Oct 25 11:33:52 2008 +0100 [xlib] Exploit compaction of XRenderFillRectangle() libXrender amalgamates sequences of XRenderFillRectangle() into a single XRenderFillRectangles request (when permissible). Since it is common for a cairo application to draw rectangles individually in order to exploit fast paths within cairo [rectilinear fills], it is a reasonably common pattern. src/cairo-xlib-surface.c | 50 +++++++++++++++++++++++++++++++----------------- 1 file changed, 32 insertions(+), 18 deletions(-) commit 64726ccfb99ba964420f3a840fd707998927e902 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 24 17:06:04 2008 +0100 [traps] Whitespace. Fixup whitespace in line with CODING_STYLE and rest of file. src/cairo-region.c | 6 +++--- src/cairo-traps.c | 33 ++++++++++++++++++++------------- 2 files changed, 23 insertions(+), 16 deletions(-) commit d5543005e7b7cc08f97578f301182ef4807a05fe Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 24 09:00:45 2008 +0100 [surface] Only copy font options if the target surface has them. No need to copy font options if the similar surface is from the same backend and no special options have been applied by the user. Doing so breaks lazy initialisation of backend specific options. src/cairo-surface.c | 18 +++++++++++------- 1 file changed, 11 insertions(+), 7 deletions(-) commit a0023758104b700751ba8d7a66d75db139eea6cb Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 24 08:39:29 2008 +0100 [xlib] Defer querying of font options until first use Constructing the font options cause the initialisation of Xlc and invoke several round-trips to the X server, significantly delaying the creation of the first surface. By deferring that operation until the first use of fonts then we avoid that overhead for very simple applications (like the test suite) and should improve start-up latency for larger application. src/cairo-xlib-private.h | 4 ++++ src/cairo-xlib-screen.c | 28 +++++++++++++++++++++++++--- src/cairo-xlib-surface.c | 2 +- 3 files changed, 30 insertions(+), 4 deletions(-) commit 6706590d4e7ad19dae0b8e3efe6f573d5688e19a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Sep 29 16:22:48 2008 +0100 [pattern] Reduce likelihood of range overflow with large downscaling. Exploit the auxiliary offset vector in the attributes to reduce likelihood of range overflow in the translation components when converting the pattern matrix to fixed-point pixman_matrix_t. An example of this is bug 9148 Bug 9148 - invalid rendering when painting large scaled-down surfaces (https://bugs.freedesktop.org/show_bug.cgi?id=9148) but the issue is perhaps even more likely with high resolution fallback images. src/cairo-pattern.c | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+) commit c0af8c70635d641fc5770afc0cd6e9285122fd72 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Oct 20 16:49:23 2008 +0100 [pattern] Tweak REFLECT HACK In order to workaound a directfb bug, tweak the reflect->repeat pattern so that it covers the destination rectangle. Although the number of paint() increases, the number of read/written pixels remain the same so that performance should not deteriorate, but instead be improved by using a cloned source. The early return of the REFLECT surface is discarded so that the latter optimisations for surface sources can be applied. One side effect of this is that acquire_source_image() is removed due to its lax reference counting which thereby exposes the ROI optimisations for image destinations as well. src/cairo-pattern.c | 279 +++++++++++++++++++++++++--------------------------- src/cairoint.h | 1 - 2 files changed, 136 insertions(+), 144 deletions(-) commit 234623b3d5884b02a0619de8ddd57bbed36ef4fd Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Oct 23 14:55:26 2008 +0100 [pattern] Support unbounded surfaces. An unbounded surface should just report an infinite pattern extent and not return UNSUPPORTED from _cairo_pattern_get_extents(). src/cairo-pattern.c | 3 +++ 1 file changed, 3 insertions(+) commit 2836be6f7565684e86b3cc8a3d2db4ba690b3718 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Oct 23 14:42:47 2008 +0100 Cleanup 'status && status != UNSUPPORTED' Replace instances of 'status && status != UNSUPPORTED' with the more readable _cairo_status_is_error(). src/cairo-analysis-surface.c | 14 +++++++------- src/cairo-clip.c | 4 ++-- src/cairo-pdf-surface.c | 8 ++++---- src/cairo-ps-surface.c | 4 ++-- src/cairo-scaled-font-subsets.c | 4 ++-- 5 files changed, 17 insertions(+), 17 deletions(-) commit 13ba43eb8ffc3bb6c8881d91824102a8df86d928 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Oct 23 12:28:25 2008 +0100 [matrix] Optimize finding the bounding box under an orthogonal matrix. We frequently need to find the bounds of a pattern under an identity matrix, or a simple scale+translation. For these cases we do not need to transform each corner and search for the bounds as the matrix is x/y separable and so allows us to inspect the results for the extreme x/y points independently. src/cairo-matrix.c | 42 +++++++++++++++++++++++++++++++++++++++++- 1 file changed, 41 insertions(+), 1 deletion(-) commit 6ed957fc242d7890117f4dea121c03ec2523470c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Oct 22 22:02:05 2008 +0100 [gstate] Use _cairo_pattern_black for the default source. Avoid allocating a default source pattern by using the static black pattern object. The one complication is that we need to ensure that the static pattern does leak to the application, so we replace it with an allocated solid pattern within _cairo_gstate_get_source(). src/cairo-gstate.c | 9 +++++++-- src/cairo-pattern.c | 12 +++++++----- src/cairo.c | 6 +++--- src/cairoint.h | 2 +- 4 files changed, 18 insertions(+), 11 deletions(-) commit 794460186459981cd43ca38f3eba07e3f6bf8908 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Oct 22 19:24:44 2008 +0100 [pattern] Avoid needless copying of patterns. Only copy the pattern if we need to modify it, e.g. preserve a copy in a snapshot or a soft-mask, or to modify the matrix. Otherwise we can continue to use the original pattern and mark it as const in order to generate compiler warnings if we do attempt to write to it. src/cairo-analysis-surface.c | 37 +++---- src/cairo-directfb-surface.c | 33 +++---- src/cairo-glitz-surface.c | 42 ++++---- src/cairo-gstate.c | 117 ++++++++++++++-------- src/cairo-image-surface.c | 6 +- src/cairo-meta-surface.c | 12 +-- src/cairo-paginated-surface.c | 12 +-- src/cairo-pattern.c | 70 ++++++-------- src/cairo-pdf-surface.c | 88 +++++++++++------ src/cairo-ps-surface.c | 39 ++++---- src/cairo-quartz-surface.c | 10 +- src/cairo-scaled-font.c | 2 +- src/cairo-sdl-surface.c | 22 ++--- src/cairo-surface-fallback-private.h | 45 ++++----- src/cairo-surface-fallback.c | 65 ++++++------- src/cairo-surface.c | 181 +++++++++++++++++++++-------------- src/cairo-svg-surface.c | 50 +++++----- src/cairo-type3-glyph-surface.c | 18 ++-- src/cairo-win32-font.c | 26 ++--- src/cairo-win32-printing-surface.c | 51 +++++----- src/cairo-win32-private.h | 2 +- src/cairo-win32-surface.c | 6 +- src/cairo-xcb-surface.c | 42 ++++---- src/cairo-xlib-surface.c | 14 +-- src/cairoint.h | 74 +++++++------- src/test-meta-surface.c | 28 +++--- src/test-paginated-surface.c | 40 ++++---- 27 files changed, 605 insertions(+), 527 deletions(-) commit 9d2189afbd2aa1413f620a4d35016ef3be3fc0e9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Oct 22 18:29:02 2008 +0100 [pattern] Do not recompute the inverse if setting an identical matrix. More micro-optimisation. src/cairo-pattern.c | 3 +++ 1 file changed, 3 insertions(+) commit 2c277ddbea42222130b8b1c2a2b4934503b40188 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Oct 22 18:17:49 2008 +0100 [matrix] Avoid error correction overhead for translation matrices. We can only correct rounding errors between cairo and pixman matrices for scaled matrices - so skip the inversion and point transformation overhead for simple translation matrices. src/cairo-matrix.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) commit 57a1d932f3e9d79f3259a536ec2f45f020d46de5 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Oct 22 18:12:10 2008 +0100 [array] Rearrange user_data_fini() to optimize common case. Micro-optimisation to avoid the _cairo_array_index() for the common case of 0 elements. src/cairo-array.c | 16 ++++++++++------ 1 file changed, 10 insertions(+), 6 deletions(-) commit 2852127c72fa6cfe246e6ece0433be6c3f2a93ab Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Oct 22 18:04:30 2008 +0100 [cairoint] Add a few cairo_pure. Start marking up the prototypes with function attributes - first the obvious pure functions. src/cairoint.h | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) commit cf072c7203dfce989ee60e56352f548329d3d616 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Oct 30 14:35:49 2008 +0000 [sdl] Add new backend. Add a new backend to allow easy interoperability with the Simple DirectMedia Layer. boilerplate/Makefile.sources | 3 + boilerplate/cairo-boilerplate-sdl-private.h | 56 ++++ boilerplate/cairo-boilerplate-sdl.c | 69 +++++ boilerplate/cairo-boilerplate.c | 13 + build/configure.ac.features | 1 + configure.ac | 8 + src/Makefile.sources | 3 + src/cairo-sdl-surface.c | 418 ++++++++++++++++++++++++++++ src/cairo-sdl.h | 54 ++++ src/cairo.h | 4 +- 10 files changed, 628 insertions(+), 1 deletion(-) commit 31ada1ea15d14c0a843728b105340c270e4e4613 Merge: 30f2df1de ab15d7627 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Oct 30 16:56:07 2008 +0000 Merge branch '1.8' commit ab15d7627506a837fea7101661425bb923ef5b4f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Oct 30 16:54:23 2008 +0000 [test] Remove XFAIL from user-font-mask user-font-mask should PASS modulo the scaling anomalies in the external renderers, so remove it from the XFAIL list. test/Makefile.am | 15 ++++++--------- test/user-font-mask-pdf-ref.png | Bin 0 -> 1927 bytes test/user-font-mask-ps2-ref.png | Bin 0 -> 1927 bytes test/user-font-mask-ps3-ref.png | Bin 0 -> 1927 bytes test/user-font-mask-svg11-ref.png | Bin 0 -> 2224 bytes 5 files changed, 6 insertions(+), 9 deletions(-) commit 540f555840ba4076e3ab86669daf9bee5695afa2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Oct 23 14:38:44 2008 +0100 [analysis] Only limit to mask extends if bounded by mask. The extents of cairo_mask() is only limited to the mask if the operation is bounded by the mask. src/cairo-analysis-surface.c | 8 ++++++-- test/clip-operator-pdf-argb32-ref.png | Bin 11145 -> 12125 bytes test/clip-operator-pdf-rgb24-ref.png | Bin 6530 -> 7367 bytes test/operator-clear-pdf-argb32-ref.png | Bin 1438 -> 1614 bytes test/operator-clear-pdf-rgb24-ref.png | Bin 1170 -> 1402 bytes test/operator-source-pdf-argb32-ref.png | Bin 4694 -> 5149 bytes test/operator-source-pdf-rgb24-ref.png | Bin 3848 -> 4354 bytes 7 files changed, 6 insertions(+), 2 deletions(-) commit 4b2998893930e4bab19646e6f0a0c8d2a47f12a2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Oct 23 14:34:30 2008 +0100 Review users of cairo_rectangle_int_t for incorrect unsigned promotion. Adrian Johnson discovered cases where we mistakenly compared the result of unsigned arithmetic where we need signed quantities. Look for similar cases in the users of cairo_rectangle_int_t. src/cairo-analysis-surface.c | 48 +++++++++++++++++++++++-------------- src/cairo-clip.c | 43 +++++++++++++++++----------------- src/cairo-gstate.c | 4 ++-- src/cairo-pattern.c | 13 ++++++---- src/cairo-rectangle.c | 6 ++++- src/cairo-region-private.h | 3 ++- src/cairo-region.c | 3 ++- src/cairo-surface-fallback.c | 31 +++++++++++++++--------- src/cairo-surface.c | 27 +++++++++++---------- src/cairo-win32-surface.c | 4 ++-- src/cairo-xcb-surface.c | 56 ++++++++++++++------------------------------ src/cairo-xlib-surface.c | 52 ++++++++++++++-------------------------- src/cairoint.h | 2 +- 13 files changed, 144 insertions(+), 148 deletions(-) commit 0e4156121f05b14f800289ea34c5382de1f20869 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Oct 23 13:37:41 2008 +0100 [rectangle] Fix unsigned promotion whilst computing intersect. _cairo_rectangle_intersect() incorrectly allows unsigned promotion during its arithmetic. src/cairo-rectangle.c | 31 +++++++++++++++++-------------- src/cairoint.h | 3 ++- 2 files changed, 19 insertions(+), 15 deletions(-) commit 2464b8a0a9f7db7421c257eb4e3ac0d98af34761 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Oct 23 13:36:19 2008 +0100 [pattern] Allow the projected surface extents to be negative. In order to handle projection of analysis surface with user-fonts we need to accommodate patterns extending into negative coordinate space. src/cairo-pattern.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) commit d5d29075bdbb7f412fa6dc182573da18fd81af54 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Oct 23 15:24:13 2008 +0100 [gstate] Allocate temporary variable for backend_to_user transform. _cairo_gstate_backend_to_user_rectangle() requires that its input arguments are non-NULL and describe the input rectangle to be transformed. However, we were passing through output parameters from the public API which were allowed to be NULL. So we need to allocate temporary variables in which to compute the output rectangle, but only copy them as required. src/cairo-gstate.c | 69 +++++++++++++++++++++++++++++++++++++++--------------- 1 file changed, 50 insertions(+), 19 deletions(-) commit 30f2df1dec419a10408b53ecd26cebfcf607bf27 Author: Carl Worth <cworth@cworth.org> Date: Thu Oct 30 08:40:35 2008 -0700 Fix typo in bug URL. Thanks to AndrewR in IRC for pointing out the typo. NEWS | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit b598dcd1b1726184506de6a9d253f6091a000922 Author: Carl Worth <cworth@cworth.org> Date: Thu Oct 30 08:29:51 2008 -0700 Fix release-publish to also update the current manual. That is, the vesion that appears as: http://cairographics.org/manual and: http://cairographics.org/cairo-manual.tar.gz It was silly that we previously required a manual step to upload the documentation (which we regularly forgot to do) and that it uploaded with a date in the name rather than a version. So we just drop the old doc-publish Makefile target now as it's just not useful anymore. RELEASING | 3 ++- build/Makefile.am.releasing | 18 ++++-------------- 2 files changed, 6 insertions(+), 15 deletions(-) commit 91183a503e3b7bed11e0149783d3974e26daf623 Author: Carl Worth <cworth@cworth.org> Date: Thu Oct 30 08:26:55 2008 -0700 RELEASING: Mention that a versioned manual is uploaded. RELEASING | 2 ++ 1 file changed, 2 insertions(+) commit 5037fa238e820419824773fa2ea475c949e23a8c Author: Carl Worth <cworth@cworth.org> Date: Thu Oct 30 08:23:06 2008 -0700 Increment version to 1.9.1 We're finalyl opening the 1.9 development series in order to land lots of exciting new features that people have been working on. cairo-version.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit d53537e8b5e3a8345983d8a0529cb230fa2886d1 Author: Carl Worth <cworth@cworth.org> Date: Thu Oct 30 08:20:59 2008 -0700 Increment version to 1.8.3 after the 1.8.2 release. Thanks to Chris Wilson (who else?) for the last-minute bug fix. cairo-version.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit f7c958d97221375fdcbb6c58c3b58c07676b7589 Author: Carl Worth <cworth@cworth.org> Date: Wed Oct 29 21:02:04 2008 -0700 Increment cairo version to 1.8.2. Hurrah! We're finally there. cairo-version.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 42711a5586cba5db5451ce2400ee5fe655700391 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Oct 30 09:54:47 2008 +0000 [xlib] Fix _draw_image_surface() with opaque images. If the image was opaque with no alpha channel, we filled the output alpha with 0. Typically, the destination surface for dithering is an RGB window, so this bug went unnoticed. However, test/xlib-expose-event is an example where we generate an intermediate alpha-only pixmap for use as a stencil and this was failing as the mask was left completely transparent. The simple solution is to ensure that for opaque images, the output alpha is set to the maximum permissible value. src/cairo-xlib-surface.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) commit c3940d342ac506055c2ce6b7e9b27f92d8a63999 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Oct 30 09:59:48 2008 +0000 [xlib] whitespace Tweak the whitespace to lose some unnecessary line wrapping, casts and blanks. src/cairo-xlib-surface.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) commit e51648b601de274559800c63575c7f7e0288f596 Author: Behdad Esfahbod <behdad@behdad.org> Date: Thu Oct 30 04:43:45 2008 -0400 [test/xlib-expose-event] Save the output image to disk test/xlib-expose-event.c | 1 + 1 file changed, 1 insertion(+) commit f534bd549e1e2283735d1eabb60c015a5949a735 Author: Carl Worth <cworth@cworth.org> Date: Wed Oct 29 21:28:56 2008 -0700 Remove test/twin-ref.png. I missed this in a previous commit, (I think I had used a pattern of twin-*-ref.png which of course didn't match this file). test/twin-ref.png | Bin 1836 -> 0 bytes 1 file changed, 0 insertions(+), 0 deletions(-) commit cf97966c07bc64ff4b169d1d8239407883a26747 Author: Carl Worth <cworth@cworth.org> Date: Wed Oct 29 21:01:01 2008 -0700 NEWS: Finish the entry for the 1.8.2 release. Add some summarizing paragraphs and organize bug-fixes and optimizations into separate sections. NEWS | 89 ++++++++++++++++++++++++++++++++++++++++++++++---------------------- 1 file changed, 60 insertions(+), 29 deletions(-) commit b6741cffef6e86d59fa4809d32de14b9c5e8afbf Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Oct 30 01:15:45 2008 +0000 [NEWS] Spot another couple of typos. Proof-reading is difficult, even with the squiggly lines. NEWS | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) commit d20cc827791e177eb73a9003c6704b7a5bc49f71 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Oct 30 01:06:08 2008 +0000 [NEWS] Correct a couple of typos. Whilst Carl's not looking, fix a couple of typos -- in particular the one calling me a bug! ;-) NEWS | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) commit 0ceda81dfdfc4adcec28c71be7822caa0627d818 Author: Carl Worth <cworth@cworth.org> Date: Wed Oct 29 17:38:36 2008 -0700 NEWS: Add long list of bugs fixed for 1.8.2. (Still need to add a paragraph or two summarizing the release.) NEWS | 110 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 110 insertions(+) commit e0046aaf417a61da008dc6374871fa3687ba94ab Author: Carl Worth <cworth@cworth.org> Date: Wed Oct 29 16:25:00 2008 -0700 Remove twin perf case. This performance test relied on the recently-removed ability to select the internal twin-based font family with a name of "cairo". Presumably, we'll want to bring this performance case back when some other means of requesting that font face is added. perf/Makefile.am | 1 - perf/cairo-perf.c | 1 - perf/cairo-perf.h | 1 - perf/twin.c | 50 -------------------------------------------------- 4 files changed, 53 deletions(-) commit 292233685534aed712dfd45e8ccf498b792ce496 Author: Carl Worth <cworth@cworth.org> Date: Wed Oct 29 16:25:00 2008 -0700 Remove twin test case. This test relied on the recently-removed ability to select the internal twin-based font family with a name of "cairo". Presumably, we'll want to bring this test case back when some other means of requesting that font face is added. test/.gitignore | 1 - test/Makefile.am | 6 ----- test/twin-ps2-ref.png | Bin 1167 -> 0 bytes test/twin-ps3-ref.png | Bin 1167 -> 0 bytes test/twin-svg11-ref.png | Bin 1773 -> 0 bytes test/twin-svg12-ref.png | Bin 1773 -> 0 bytes test/twin.c | 60 ------------------------------------------------ 7 files changed, 67 deletions(-) commit a824d284be23793a5c48b9ae833dcb7b2d5fff80 Author: Carl Worth <cworth@cworth.org> Date: Wed Oct 29 16:21:42 2008 -0700 Remove the ability to select the internal font face with a name of "cairo". It's not fair to steal this name from the namespace of family names. There are definitely cairo.ttf files that exist out there, and people may already be using these, (or may use them in the future), with cairo_select_font_face and a family name of "cairo". In place of this, we'll want to come up with some other new, and documented API for selecting the internal font face. src/cairo-font-face.c | 6 ++---- src/cairoint.h | 1 - 2 files changed, 2 insertions(+), 5 deletions(-) commit f039da8d9172a6fb60082901c8dc56c7159ecd24 Author: Carl Worth <cworth@cworth.org> Date: Wed Oct 29 16:17:24 2008 -0700 Add doltcompile and doltlibtool to .gitignore. These generated files were recently added to our build system. We definitely don't want to see them in git-status output. .gitignore | 2 ++ 1 file changed, 2 insertions(+) commit 81439b3bc8656b34c13b3cc7b5b6a9f03a3c8c54 Author: Carl Worth <cworth@cworth.org> Date: Wed Oct 29 16:15:12 2008 -0700 Add missing files to CLEANFILES and DISTCLEANFILES. Without these files in these lists, "make distcheck" is unhappy. Makefile.am | 4 +++- test/Makefile.am | 9 +++++++++ 2 files changed, 12 insertions(+), 1 deletion(-) commit dd0416f5d47a946498219616f53abdad81d4fc38 Author: Carl Worth <cworth@cworth.org> Date: Tue Oct 28 18:01:09 2008 -0700 Add recently added reference images to REFERENCE_IMAGES list. This is just part of the make-distcheck routine for me. I know Behdad added a test to check for missing images in the list, but it doesn't seem to be getting run automatically as part of 'make test' nor 'make distcheck', (or if it it, then I'm not noticing its output). test/Makefile.am | 4 ++++ 1 file changed, 4 insertions(+) commit 9cd9644c1d993f86c6eb7bee692fe642d6d280b1 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Oct 29 21:56:36 2008 +0000 [test] Add a pass-through test. Check that colour values are correctly passed through all the backends. Simple test of the most fundamental functionality. test/.gitignore | 1 + test/Makefile.am | 3 ++ test/pass-through-ref.png | Bin 0 -> 221 bytes test/pass-through-rgb24-ref.png | Bin 0 -> 179 bytes test/pass-through.c | 99 ++++++++++++++++++++++++++++++++++++++++ 5 files changed, 103 insertions(+) commit cec689eab628b59ee1d69e241c5d336760304ea0 Author: Carl Worth <cworth@cworth.org> Date: Tue Oct 28 17:18:34 2008 -0700 RELEASING: Suggest running 'make distcheck' against Xvfb. This avoids hitting driver-specific bugs in the X server, and is better than doing 'DISPLAY= make distcheck' which simply disables all xlib testing completely. RELEASING | 5 +++++ 1 file changed, 5 insertions(+) commit 0ca6ba18e398db7ed8f8bca5faa93d48aa3a4878 Author: Carl Worth <cworth@cworth.org> Date: Tue Oct 28 17:15:42 2008 -0700 Add note to test/README on running with Xvfb. This X server has been tested to run through the test suite with no unexpected failures, and it avoids hitting any X-driver-specific bugs. test/README | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) commit eed67717649c343e42ce102a4d78a245dba48925 Author: Carl Worth <cworth@cworth.org> Date: Tue Oct 28 17:12:15 2008 -0700 Disable the svg12 target in boilerplate. This target was added to the boilerplate during 1.8.1. It currently shows many failures in the test suite. These failures likely fall into three different classes: * Tests needing new svg12-specific reference images * Tests exercising bugs in librsvg * Tests exercising existing cairo bugs We haven't gone through the effort to separate these, but even for the tests that are exercising actual cairo bugs, these are likely bugs that existed in the cairo 1.8.0 release and not regressions. Because of that, in this commit I'm conditionally disabling the testing of the svg12 target. As soon as we increment the cairo version to 1.9.0 or higher, this target will get re-enabled automatically and we can begin the work to separate the tests as described above and also fix the bugs. boilerplate/cairo-boilerplate.c | 5 +++++ 1 file changed, 5 insertions(+) commit a94928ddfa066f6e6824775b9d0b6008f3fcbf26 Author: Carl Worth <cworth@cworth.org> Date: Tue Oct 28 15:55:38 2008 -0700 Add svg-specific reference images for the twin test. A bit annoying that I have to add the same image as both -svg11 and -svg12 but that's all the support we have in the current test suite. I suppose I could avoid doing that by figuring out why this test case cannot successfully roundtrip through librsvg and back through cairo. test/twin-svg11-ref.png | Bin 0 -> 1773 bytes test/twin-svg12-ref.png | Bin 0 -> 1773 bytes 2 files changed, 0 insertions(+), 0 deletions(-) commit 53554ea8892893098b29a27a23c069f0ee584a06 Author: Carl Worth <cworth@cworth.org> Date: Tue Oct 28 15:50:44 2008 -0700 Add ps2-specific reference images for trap-clip test. These are quite similar to the existing ps3-specific reference images. I definitely don't see any reason why this output should be considered a failure. test/trap-clip-ps2-argb32-ref.png | Bin 0 -> 5690 bytes test/trap-clip-ps2-rgb24-ref.png | Bin 0 -> 5617 bytes 2 files changed, 0 insertions(+), 0 deletions(-) commit 10fd16a05d8549753179ab5cca855539fb7dec94 Author: Carl Worth <cworth@cworth.org> Date: Tue Oct 28 15:41:40 2008 -0700 Mark user-font-mask as XFAIL. This test is expected to fail due to a couple of known bugs. Chris has fixes for both bugs, but is holding off on them until after 1.8.2 to prevent introducing any possible new bugs with his fixes. test/Makefile.am | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) commit 655546243f57ae2a5c1813bd45994b18061f1c74 Author: Carl Worth <cworth@cworth.org> Date: Tue Oct 28 15:35:35 2008 -0700 Rename user-font-image test to user-font-mask. Otherwise the reference image (user-font-image-ref.png) gets interpreted as an image-specific reference image for the user-font test case resulting in a bogus failure. test/Makefile.am | 4 ++-- test/{user-font-image-ref.png => user-font-mask-ref.png} | Bin test/{user-font-image.c => user-font-mask.c} | 4 ++-- 3 files changed, 4 insertions(+), 4 deletions(-) commit 42d2d1b6d3b187b5dbdb630783d7c9bf90c66fb2 Author: Carl Worth <cworth@cworth.org> Date: Tue Oct 28 12:48:03 2008 -0700 Correct NEWS to say 'release' instead of 'snapshot' for 1.8.0 A copy-and-paste bug strikes again. NEWS | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 1327ec232cfca675647fb03876487c92fb638354 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Oct 28 10:00:38 2008 +0000 [image] Remove invalid assert. The assert can fail for an error surface. TODO: Decide what values should be returned from getters for error surfaces. src/cairo-image-surface.c | 2 -- 1 file changed, 2 deletions(-) commit 9481d999df8d399543bdbb45b85bd24b1725bece Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Oct 28 09:59:01 2008 +0000 [image] Check create_for_data() to ensure a valid minimum stride. Double check that the user is not being silly by passing in a stride that is too small for the width. evince/poppler is one such example... src/cairo-image-surface.c | 12 ++++++++++++ 1 file changed, 12 insertions(+) commit 88f628b3972819c0f6ce8e5f06d0f7e6abb9d661 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Oct 28 09:11:23 2008 +0000 [perf] Fix rectangular case of unaligned-clip. Janoos spotted that the unaligned clip actually degenerated to an empty clip due to a typo when constructing the second rectangle. Simply use a cairo_rectangle() instead. perf/unaligned-clip.c | 9 +++------ 1 file changed, 3 insertions(+), 6 deletions(-) commit 1522805f39c505ff05231e67de5a6e6686081d46 Author: Behdad Esfahbod <behdad@behdad.org> Date: Mon Oct 27 20:57:56 2008 -0400 [Makefile.am.releasing] Fix typo build/Makefile.am.releasing | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 3386794d96d2b1616af9db5b7df1fc5c7a5796a7 Author: Carl Worth <cworth@cworth.org> Date: Mon Oct 27 14:09:19 2008 -0700 Add creation of a versioned manual to the release-publish target. This arranges so that a versioned manual will be available of the form: http://cairographics.org/manual-X.Y.Z for all future releases and snapshots. We're going through the process of manually doing all former releases, such as: http://cairographics.org/manual-1.2.0/ which the LSB folks wanted to reference. build/Makefile.am.releasing | 27 ++++++++++++++++++--------- 1 file changed, 18 insertions(+), 9 deletions(-) commit a1a7ec6a056fa36389a48c0f53884ae7e92ebc1a Author: Adrian Johnson <ajohnson@redneon.com> Date: Mon Oct 27 23:13:05 2008 +1030 Change ps to ps3 in test/README The ps target no longer exists. test/README | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 670d942fe3dafbed5ceca234e760f29f793233cb Author: Adrian Johnson <ajohnson@redneon.com> Date: Mon Oct 27 23:04:16 2008 +1030 Change user-font-image test to use a pattern This can be used to expose a bug in _cairo_rectangle_intersect() by changing: fixed_scale = 1024 to 1 in cairo-user-font.c and cairo_matrix_translate (&matrix, 0, -8) to (&matrix, 0, -9) in user-font-image.c This will cause cairo_text_extents (cr, text, &extents) in user-font-image.c to return a height of 8388683. test/user-font-image-ref.png | Bin 5502 -> 6011 bytes test/user-font-image.c | 13 +++++++++++-- 2 files changed, 11 insertions(+), 2 deletions(-) commit cdacf55e1b16a6982c681fab212405fc4af4f1db Author: Adrian Johnson <ajohnson@redneon.com> Date: Mon Oct 27 22:07:04 2008 +1030 Add user-font-image test Draws bitmap glyphs using cairo_mask(). This test exposes a bug in the calculation of the glyph extents. test/Makefile.am | 2 + test/user-font-image-ref.png | Bin 0 -> 5502 bytes test/user-font-image.c | 245 +++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 247 insertions(+) commit 22e2dac819de3c082424ebf3ce8f64c9c4797e8a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Oct 26 10:21:37 2008 +0000 [stroke] Ensure we record the first face for a dashed path. If the first face was outside the bounds then we skipped it, and so a close would incorrectly connect to the first visible face. src/cairo-path-stroke.c | 1 + 1 file changed, 1 insertion(+) commit 64fb3290f00bafc4a10e6432226f669a51fa0bc6 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Oct 26 09:54:03 2008 +0000 [stroke] _cairo_stroker_line_to_dashed() whitespace A couple of comment spelling mistakes and rearrange whitespace to more closely match CODING_STYLE. src/cairo-path-stroke.c | 57 +++++++++++++++++++++++++++++++++---------------- 1 file changed, 39 insertions(+), 18 deletions(-) commit 88935947d11762c91e282fb129c493afc6f2f1a6 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Oct 26 08:19:31 2008 +0000 [test] Add leaky-dashed-stroke A test case for a leak whilst closing a dashed stroke extracted from the report by Jeff Muizelaar who found the artifact whilst looking at firefox http://people.mozilla.com/~jmuizelaar/BerlinDistricts-check.svg test/.gitignore | 1 + test/Makefile.am | 4 + test/leaky-dashed-stroke-ps2-ref.png | Bin 0 -> 5511 bytes test/leaky-dashed-stroke-ps3-ref.png | Bin 0 -> 5511 bytes test/leaky-dashed-stroke-ref.png | Bin 0 -> 9214 bytes test/leaky-dashed-stroke.c | 174 +++++++++++++++++++++++++++++++++++ 6 files changed, 179 insertions(+) commit 195fd27cc64c7aaf18a6d48ccd9d94548331fd4d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Oct 22 22:04:53 2008 +0100 [build] Use AC_LINK_IFELSE for testing linker flags. Richard Hult reported that -Wl,--allow-shlib-undefined was incorrectly being identified as supported by gcc under Mac OS/X: configure:25103: checking whether gcc supports -Wl,--allow-shlib-undefined configure:25117: gcc -c -Werror -Wl,--allow-shlib-undefined conftest.c >&5 i686-apple-darwin9-gcc-4.0.1: --allow-shlib-undefined: linker input file unused because linking not done So instead of just checking whether the compiler accepts the flag, check that we can actually link a dummy file. build/aclocal.cairo.m4 | 20 ++++++++++++++++++++ build/configure.ac.warnings | 2 +- 2 files changed, 21 insertions(+), 1 deletion(-) commit d692284ab8f90d7402621f1a07aaddcf11d87ef5 Author: Behdad Esfahbod <behdad@behdad.org> Date: Thu Oct 23 01:45:35 2008 -0400 [test/user-font] Exercise glyph 0 Previously there was a bug in cairo and/or poppler causing glyph 0 of user fonts to not show up. This is fixed now, so we exercise glyph 0. test/user-font.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) commit e6219f7116311d999054dc31df081aa66a0cd2d3 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Oct 22 17:53:05 2008 +0100 [win32] Correct error paths in text_to_glyphs(). Carl spotted that 1db894 introduced an error into text_to_glyphs() such that following an allocation error the code would attempt to thaw an unfrozen glyph cache, leading to an attempt to unlock an unlocked mutex - which causes a deadlock under win32. src/cairo-win32-font.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) commit 09651a1c1a4856607423e9b06c1f7c397ba4f304 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Oct 8 22:39:48 2008 +0100 Twin perf case perf/Makefile.am | 3 ++- perf/cairo-perf.c | 1 + perf/cairo-perf.h | 1 + perf/twin.c | 50 ++++++++++++++++++++++++++++++++++++++++++++++++++ 4 files changed, 54 insertions(+), 1 deletion(-) commit 954ebacb71071c53c4e4092b469417f01478bc2d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Oct 22 15:43:56 2008 +0100 Map toy font face to implementation. Quartz fonts and user fonts use an indirect font face when creating a scaled font for the toy font face. This means that they insert a scaled font into the font map that has a different font face to the one that is initially searched upon. The result is that when we try to create an identical scaled font, we fail to find the existing scaled font and attempt to insert a duplicate into the hash table - which triggers an assert. In order to avoid creating duplicate fonts, we add a new method to the font backends that allows cairo_scaled_font_create() to peek at the font_face that will be used to actually implement the scaled font constructor - thus we are able to use the correct font_face as part of the hash key. src/cairo-font-face.c | 34 ++++++++++++++++++++++++++++++ src/cairo-ft-font.c | 2 ++ src/cairo-quartz-font.c | 55 +++++++++++++++++++++++++++++++++++++++---------- src/cairo-scaled-font.c | 11 +++++++++- src/cairo-user-font.c | 37 +++++++++++++++++++++++++-------- src/cairo-win32-font.c | 2 ++ src/cairoint.h | 8 +++++++ 7 files changed, 128 insertions(+), 21 deletions(-) commit 5e4a1cb0b830f069d99c9512563c82ad628587e3 Author: Adrian Johnson <ajohnson@redneon.com> Date: Wed Oct 22 21:36:42 2008 +1030 Don't add a notdef glyph to Type 3 font subsets This allows user-fonts to use glyph 0 when embedding in PS/PDF. According to ISO32000 Type 3 fonts do not require a notdef glyph. src/cairo-pdf-surface.c | 19 ++++------- src/cairo-ps-surface.c | 21 ++++-------- src/cairo-scaled-font-subsets.c | 59 +++++++++++++++++++-------------- src/cairo-type3-glyph-surface-private.h | 6 ---- src/cairo-type3-glyph-surface.c | 16 --------- 5 files changed, 48 insertions(+), 73 deletions(-) commit 170686d4b0ed913a1a175f3f354ec5796814b416 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Oct 22 01:01:06 2008 +0100 [cairoint] Remove the duplicate prototype. In the previous commit I added comments and a duplicate prototype. Remove the duplicate - but keep the comments. src/cairoint.h | 3 --- 1 file changed, 3 deletions(-) commit 1db8949f2baf1e620e1d5ef73a66de211420bd0a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Oct 21 22:48:17 2008 +0100 Ensure that the scaled font is frozen for the lifetime of the scaled glyph. After discussing the scaled font locking with Behdad, it transpired that it is not sufficient for a font to be locked for the lifetime of a scaled glyph, but that the scaled font's glyph cache must be frozen for the glyph' lifetime. If the cache is not frozen, then there is a possibility that the glyph may be evicted before the reference goes out of scope i.e. the glyph becomes invalid whilst we are trying to use it. Since the freezing of the cache is the stronger barrier, we remove the locking/unlocking of the mutex from the backends and instead move the mutex acquisition into the freeze/thaw routines. Then update the rule on acquiring glyphs to enforce that the cache is frozen and review the usage of freeze/thaw by all the backends to ensure that the cache is frozen for the lifetime of the glyph. src/cairo-directfb-surface.c | 4 +- src/cairo-glitz-surface.c | 2 +- src/cairo-gstate.c | 4 - src/cairo-meta-surface.c | 2 - src/cairo-mutex-impl-private.h | 3 +- src/cairo-mutex-type-private.h | 3 + src/cairo-paginated-surface.c | 2 - src/cairo-pdf-surface.c | 2 - src/cairo-scaled-font-subsets.c | 30 +++--- src/cairo-scaled-font.c | 81 ++++++++++----- src/cairo-surface-fallback.c | 1 + src/cairo-surface.c | 4 - src/cairo-svg-surface.c | 4 +- src/cairo-type1-fallback.c | 9 +- src/cairo-type3-glyph-surface.c | 18 ++-- src/cairo-user-font.c | 1 - src/cairo-win32-font.c | 12 ++- src/cairo-xcb-surface.c | 216 ++++++++++++++++++++++++---------------- src/cairo-xlib-surface.c | 57 ++++++----- src/cairoint.h | 4 + src/test-meta-surface.c | 25 +---- src/test-paginated-surface.c | 25 +---- 22 files changed, 280 insertions(+), 229 deletions(-) commit b2cbbceb4ca57816a498c2c1e676b19182c34e12 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Oct 21 19:44:42 2008 +0100 [test/meta-surface-pattern] Propagate errors. Use cairo_get_target() to propagate any errors from the secondary context. test/meta-surface-pattern.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) commit 9d4262ebe07bacacd6e743dea616b37ac51eac01 Author: Behdad Esfahbod <behdad@behdad.org> Date: Tue Oct 21 17:44:47 2008 -0400 [build] Use dolt Dolt is a hack to speed up libtool-based build systems: http://dolt.freedesktop.org/ It's completely transparent to the user. Brings make time of cairo from 70s down to 20s. Yay! build/aclocal.dolt.m4 | 177 ++++++++++++++++++++++++++++++++++++++++++++++++++ configure.ac | 1 + 2 files changed, 178 insertions(+) commit 5a822046cc66d8405618370c1071433016969020 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Oct 21 11:22:48 2008 +0100 [test/xlib-expose-event] Propagate errors. Use cairo_get_target() to propagate errors from secondary contexts. test/xlib-expose-event.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) commit 21223cc12ca00c52c0b39dc60db410f29e4e76c9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Oct 21 11:29:12 2008 +0100 [test/unbounded-operator] Propagate errors. Use cairo_get_target() to propagate errors from secondary contexts. test/unbounded-operator.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) commit 4c8de6d99d337338d6fd57134247515f9815b5e5 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Oct 21 11:21:37 2008 +0100 [test/surface-pattern-scale-up] Propagate errors. Use cairo_get_target() to propagate errors from the secondary context. test/surface-pattern-scale-up.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) commit e8de7d1c3605fa6d89640fb4e0631c07e5366348 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Oct 21 11:20:54 2008 +0100 [test/surface-pattern-scale-down] Propagate errors. Use cairo_get_target() to propagate errors from the secondary context. test/surface-pattern-scale-down.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) commit b5551cfce2969358746dadae90cbe1add012e29c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Oct 21 11:19:40 2008 +0100 [test/rotate-image-surface-paint] Proapgate errors. Use cairo_get_target() to propagate errors from the secondary context. test/rotate-image-surface-paint.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) commit 6d41ffbbed2ad3bab2b1a58e1e5d9e1570a37926 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Oct 21 11:18:37 2008 +0100 [test/pixman-rotate] Propagate errors Use cairo_get_target() to propagate errors from the secondary context. test/pixman-rotate.c | 33 ++++++++++++++++----------------- 1 file changed, 16 insertions(+), 17 deletions(-) commit 64949b81735ee4e43772224d10031f7edfbc8068 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Oct 21 11:17:13 2008 +0100 [test/operator-source] Propagate errors. Use cairo_get_target() to propagate errors from the secondary context. test/operator-source.c | 12 ++++-------- 1 file changed, 4 insertions(+), 8 deletions(-) commit 971f030b7dca7ce24e54d0465900e6984020345e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Oct 21 11:27:17 2008 +0100 [test/operator-clear] Propagate errors. Use cairo_get_target() to propagate errors from secondary contexts. test/operator-clear.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) commit 9b919b0e91b6c64acbcbfc05c72610188327be67 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Oct 21 11:14:51 2008 +0100 [test/extend-pad] Propagate errors. Use cairo_get_target() to propagate errors from the secondary context. test/extend-pad.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) commit 69415b4fc6387acda7498847f8181c802c8b0982 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Oct 21 11:25:38 2008 +0100 [test/device-offset-scale] Propagate errors Use cairo_get_target() to propagate errors from the secondary contexts. test/device-offset-scale.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) commit 9ab296123ef49259a67e8dc43708023da2759283 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Oct 21 11:14:01 2008 +0100 [test/device-offset-positive] Propagate errors. Use cairo_get_target() to propagate errors from the secondary context. test/device-offset-positive.c | 12 +++++------- 1 file changed, 5 insertions(+), 7 deletions(-) commit 594fa0836f2526e0a7f44fc7184f7f09de7dd41f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Oct 21 11:12:11 2008 +0100 [test/device-offset] Propagate error. Use cairo_get_target() to propagate errors from the secondary context. test/device-offset.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) commit 29da47be60b0373ff93769f9f373a20b15a01e0f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Oct 21 10:42:41 2008 +0100 [ps] Check for an empty font subset. Return early before trying to emit an empty subset into the document - avoids a potential use of an uninitialised status variable. src/cairo-ps-surface.c | 3 +++ 1 file changed, 3 insertions(+) commit fd1f3c27e093b1f51f0c1a381cc02cbf8f9889d7 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Oct 21 09:17:56 2008 +0100 [configure] Check for atomic xchg. Bug 18140 identifies a case where we have an atomic increment, but not an atomic exchange. We need both to implement atomic reference counting, so add a second check to detect whether __sync_val_compare_and_swap generates a non-atomic instruction. Fixes http://bugs.freedesktop.org/show_bug.cgi?id=18140. build/aclocal.cairo.m4 | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) commit 71e4f7e3a10e6c9794360473d0114b27b42bcead Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Oct 20 23:54:57 2008 +0100 Add API documentation for NULL filenames and write_funcs. Mention in the API docs that you can pass a NULL filename to cairo_(pdf|ps|svg)_surface_create in order to construct a queryable surface without generating any temporary files. Similarly when passing a NULL write_func to cairo_pdf_surface_create_for_stream et al. src/cairo-pdf-surface.c | 10 ++++++++-- src/cairo-ps-surface.c | 10 ++++++++-- src/cairo-svg-surface.c | 10 ++++++++-- 3 files changed, 24 insertions(+), 6 deletions(-) commit 6cb2b29434ffd5e718868d90e31cb050904b3ea7 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Oct 20 23:22:16 2008 +0100 [mutex] s/HOLDS_MUTEX/MUTEX_IS_LOCKED/ Behdad prefers the latter to keep a clean namespace. src/cairo-mutex-impl-private.h | 2 +- src/cairo-mutex-type-private.h | 4 ++-- src/cairo-scaled-font.c | 18 +++++++++--------- 3 files changed, 12 insertions(+), 12 deletions(-) commit ae9d17c015867002895c8aa4e1c6dd1ef370dd20 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Oct 20 22:31:17 2008 +0100 [directfb] Unbounded operators are unsupported. Currently the emulation of Porter-Duff operators does not correctly handle the unbounded operators. src/cairo-directfb-surface.c | 10 ++++++++++ 1 file changed, 10 insertions(+) commit 1f35f31c4ee50dcc2f7c8616b9bbb5c45b4ed707 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Oct 20 22:14:41 2008 +0100 [directfb] Fallback for SATURATE We're unable to satisfactorily emulate SATURATE using the src/dst blend modes, so fallback. (Might just be a bug in libdirectfb...) src/cairo-directfb-surface.c | 3 +++ 1 file changed, 3 insertions(+) commit c7951c4a368baaeeadca4e2a18d3ad45225380a0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Oct 20 22:04:58 2008 +0100 [directfb] Perfom text fallback if emulating clip regions. Fallback if we have a clip that is not a simple region. src/cairo-directfb-surface.c | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 deletions(-) commit 98933fd4b8579b68623a8212015769b058db43f5 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Oct 20 20:59:14 2008 +0100 [directfb] When blitting check if we need the un-premultiplied color. When blitting whether we need to use the premultiplied color is dependent upon the destination surface capabilities. src/cairo-color.c | 3 ++ src/cairo-directfb-surface.c | 75 ++++++++++++++++++++++++++++++-------------- 2 files changed, 54 insertions(+), 24 deletions(-) commit 4af8aa5f4d31141a6a4ba914cc860aff5d342d5b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Oct 20 20:15:12 2008 +0100 [directfb] Apply clip to release_dest_image() We need to respect the current clip when copying the fallback image back to the target surface. src/cairo-directfb-surface.c | 23 ++++++++++++++--------- 1 file changed, 14 insertions(+), 9 deletions(-) commit 0ba5085ba06fd1c0ff1ebc88111d4573675a2e17 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Oct 20 19:41:20 2008 +0100 [directfb] Fix OBO in clip. DirectFB seems to use inclusive upper bounds. src/cairo-directfb-surface.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 2743e84171fd2a548ac37b1fdd3611d319acded0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Oct 20 19:32:21 2008 +0100 [directfb] Force NEAREST. The current version of DirectFB does not support any filters, so set NEAREST on the source patterns so that we can fully optimize the pattern inside core. src/cairo-directfb-surface.c | 5 +++++ 1 file changed, 5 insertions(+) commit f1669d31c0064b8ffbc7db2527fed85488fb305c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Oct 20 19:03:33 2008 +0100 [directfb] Track the empty clip rectangle. A region with no extents means everything should be clipped out, so we need an extra flag to disambiguate when we have 0 clip rectangles. src/cairo-directfb-surface.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) commit ebe3048f9b74ff949795a807a7076ec756150ede Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Oct 20 18:59:03 2008 +0100 [directfb] Cap the maximum surface size to use for the font cache. Fallback if we need to create a cache larger than the maximum usable surface. src/cairo-directfb-surface.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) commit 6f35c2d13e4cb736634bf446d3b8ee38cc6a6b9b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Oct 20 18:39:45 2008 +0100 [directfb] Only use the pure BLIT for integer translations. If we have non-integer translation on a source pattern then we need to use STRECTHBLIT (the core takes care to optimize NEAREST patterns to integer translations so that this will only be used when interpolation is required). src/cairo-directfb-surface.c | 30 +++++++++++++++++++++++------- 1 file changed, 23 insertions(+), 7 deletions(-) commit 55bb7087b0236fe27980efda829f7b72e047b316 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Oct 20 16:47:41 2008 +0100 [directfb] Tweak _categorize_operation(). Explicitly list the extend modes that are supported/unsupported and immediately check for an unsupported operation. src/cairo-directfb-surface.c | 37 +++++++++++++++++++++++++------------ 1 file changed, 25 insertions(+), 12 deletions(-) commit edb65213ec975e517165bc3b623f676a12917d0c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Oct 20 16:46:08 2008 +0100 [directfb] Support ROI cloning. Only clone the ROI in order to efficiently handle large sources. src/cairo-directfb-surface.c | 11 ++++------- 1 file changed, 4 insertions(+), 7 deletions(-) commit 4006e49ac1e96a983346f97be97bfd850f247ab0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Oct 20 16:45:06 2008 +0100 [directfb] Rename backend structure. Use _cairo_directfb_surface_backend for consistency with other backends. src/cairo-directfb-surface.c | 37 ++++++++++++++++++++++++------------- 1 file changed, 24 insertions(+), 13 deletions(-) commit 7e31ddbb08a6a7cf1db06c950c3768e54a5f4cfa Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Oct 20 13:04:20 2008 +0100 [directfb] Whitespace. Trim lots of trailing whitespace and fix up according to CODING_STYLE. src/cairo-directfb-surface.c | 1478 +++++++++++++++++++++--------------------- 1 file changed, 733 insertions(+), 745 deletions(-) commit d49563280b80dd85275057b6442d2f4ece8d724c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Oct 20 12:49:52 2008 +0100 [directfb] Simplify return from _directfb_get_operator(). Used only as a predicate, so simply return a boolean. src/cairo-directfb-surface.c | 138 +++++++++++++++++++++---------------------- 1 file changed, 68 insertions(+), 70 deletions(-) commit b17d0ded8cd42d84e97ca2c1fbd4aefedd570b8e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Oct 20 12:48:02 2008 +0100 [directfb] Use the solid pattern cache. The dfb backend recreated a solid color surface for every colour, completely bypassing the solid pattern cache. src/cairo-directfb-surface.c | 72 +++++++++++--------------------------------- 1 file changed, 18 insertions(+), 54 deletions(-) commit 786bea48df0dc6d76b2ad00409804a803fe2f4b6 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Oct 20 12:19:50 2008 +0100 [boilerplate/directfb] Create ARGB surfaces on demand. Create an RGB or ARGB surface depending upon the content type of the test target, with the result that the directfb rgb24 target no longer unconditionally fails. boilerplate/cairo-boilerplate-directfb.c | 23 +++++++++++++---------- 1 file changed, 13 insertions(+), 10 deletions(-) commit 4ba9ccc6ab5926d59696f86193f42df2195d7cda Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Oct 20 23:07:38 2008 +0100 [configure.ac.analysis] s/safe/_save/ Rename the temporary variable in line with the convention used by the rest of the build system. build/configure.ac.analysis | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) commit 3a45b07af9743a26305c2576f838b4f3ef83bbb8 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Oct 20 10:37:46 2008 +0100 [perf] Free images for composite-checker Fix memleak of the image surfaces. perf/composite-checker.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit 86356a9578dc248a755aceb7a04aef67a6c529b3 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Oct 20 17:12:21 2008 +0100 [test] Add degenerate-dash to XFAIL. PS (ghostscript) interprets the degenerate end-caps quite differently, and perhaps slightly more rationally... Add this to XFAIL, until we have a definitive specification on how we should behave. test/Makefile.am | 4 ++++ 1 file changed, 4 insertions(+) commit 83d2552e4cafc9f86a8055abdf6caa685ed4922b Author: Behdad Esfahbod <behdad@behdad.org> Date: Mon Oct 20 17:59:09 2008 -0400 Remove CAIRO_BEGIN_DECLS uses that ickle introduced while merging src/cairo-mutex-impl-private.h | 2 -- src/cairo-mutex-type-private.h | 2 -- 2 files changed, 4 deletions(-) commit eab37f76b0f26a7b007dd51debf4d4901310c8b0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Oct 19 22:12:18 2008 +0100 [directfb] Return an error surface on create() failure. Do not return NULL but an NO_MEMORY error surface if we fail to allocate the surface during creation. src/cairo-directfb-surface.c | 21 ++++++++++----------- 1 file changed, 10 insertions(+), 11 deletions(-) commit 6a02f53dd191605986b15e4757b16f599fe8de5f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Oct 19 14:21:15 2008 +0100 [directfb] Simplifiy ADD_TRI Simplify the ADD_TRI macro to make the code more readable. src/cairo-directfb-surface.c | 30 +++++++++++++----------------- 1 file changed, 13 insertions(+), 17 deletions(-) commit aba457fe64f12598043b11ae076f0a93fe05eba7 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Oct 19 14:03:24 2008 +0100 [directfb] Add missing error status We jumped to the ERROR path without setting an error status - assume NO_MEMORY. src/cairo-directfb-surface.c | 1 + 1 file changed, 1 insertion(+) commit 610451e4a7b079d30a73384a76f0aaab250c2970 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Oct 19 12:45:45 2008 +0100 [directfb] Correct fixed-to-double conversion. The trapezoid calculations had not been updated after the 16.16->24.8 switch. Use _cairo_fixed_to_double() rather than open-coding. src/cairo-directfb-surface.c | 110 +++++++++++++++++++++---------------------- 1 file changed, 54 insertions(+), 56 deletions(-) commit 8e0950ced8837a35b9dc0254860250f9fa1a9613 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Oct 19 12:11:13 2008 +0100 [boilerplate/directfb] Gracefully handle failure to create surface. Simply return NULL indicating failure to create the target surface rather than aborting the test (allowing other targets to be tested). boilerplate/cairo-boilerplate-directfb.c | 230 +++++++++++++++++-------------- 1 file changed, 130 insertions(+), 100 deletions(-) commit 801df1b87a585ebf71e6d3d74b815f3470b81ae9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Oct 19 11:55:20 2008 +0100 [directfb] Fix error propagation from surface_create() Do not mask fatal errors, but correctly propagate the error status. src/cairo-directfb-surface.c | 208 ++++++++++++++++++++++++------------------- 1 file changed, 114 insertions(+), 94 deletions(-) commit 476d17fa788fcf173e58b963540c6557f846ceed Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Oct 19 11:35:27 2008 +0100 [directfb] Compile fix. Fixup compilation after 901f0b. src/cairo-directfb-surface.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit d72969289db62b38ca024a41c0d8e0df480e7f8c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Oct 19 09:47:33 2008 +0100 [image] Remove unused variable. Cleanup the compiler warning for the unused status after code shuffling. src/cairo-image-surface.c | 1 - 1 file changed, 1 deletion(-) commit f2ff7944264c23cbec856be3e85f240a93184f80 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Oct 16 11:56:19 2008 +0100 [perf] A crude tool to visualise performance changes across a series. Generate a cairo-perf-diff graph for a series of commits in order to be able to identify significant commits. Still very crude, but minimally functional. configure.ac | 3 + perf/.gitignore | 1 + perf/Makefile.am | 26 +- perf/Makefile.win32 | 6 +- perf/cairo-perf-diff | 12 +- perf/cairo-perf-diff-files.c | 458 ------------------------------- perf/cairo-perf-graph | 205 ++++++++++++++ perf/cairo-perf-graph-files.c | 593 +++++++++++++++++++++++++++++++++++++++++ perf/cairo-perf-graph-widget.c | 450 +++++++++++++++++++++++++++++++ perf/cairo-perf-graph.h | 63 +++++ perf/cairo-perf-report.c | 456 +++++++++++++++++++++++++++++++ perf/cairo-perf.c | 1 + perf/cairo-perf.h | 61 ++++- perf/cairo-stats.h | 8 - 14 files changed, 1865 insertions(+), 478 deletions(-) commit 41c8eefc6d432ab213f6f405c3d6346adb7f7931 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Oct 15 22:24:32 2008 +0100 [output-stream] Protect against NULL write_func. Allow the user to specify a NULL write_func for the output stream so that a dummy surface can be created, for example, for querying target font options or font extents. Currently we do not perform any sanity checks at the user entry point and will generate a mysterious SEGV during cairo_surface_finish() - which may not immediately be obvious that it is due to a NULL write_func. src/cairo-output-stream.c | 3 +++ 1 file changed, 3 insertions(+) commit 84e4a825fffe3d78e95dec3406a86da6a23144c2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Oct 15 22:15:17 2008 +0100 [output-stream] Accept a NULL filename Principally to support creating a dummy vector surface (i.e. cairo_ps_surface_create (NULL, 1, 1)) that can be used to determine font extents (or target font options) before opening an output file, but also because we currently fail to do any sanity checking at the entry point. src/cairo-output-stream.c | 3 +++ 1 file changed, 3 insertions(+) commit 52ea38f97b1b90e395c667619770482b10e5672e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Oct 18 00:44:29 2008 +0100 [meta] Acquire scaled_font mutex for glyph_path(). The caller of _cairo_scaled_font_glyph_path() is expected to be holding the scaled_font->mutex. src/cairo-meta-surface.c | 2 ++ 1 file changed, 2 insertions(+) commit 299ea0580a646dc55fd0156f1904fe4b45ec5725 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Oct 18 00:37:38 2008 +0100 [user-font] Review locks under error conditions. Simplify the error handling by only relinquishing the global scaled_font_map mutex if we successfully insert the placeholder font. The result is that on the error path, there are no changes to global state and thus we can entirely skip the user-font initialisation and re-registration. src/cairo-scaled-font.c | 44 ++++++++++++++++++++++---------------------- src/cairo-user-font.c | 39 +++++++++++++++++++++++---------------- src/cairoint.h | 2 +- 3 files changed, 46 insertions(+), 39 deletions(-) commit c76a8481f372fab8981231b257fdcc69466263d2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Oct 18 00:36:34 2008 +0100 [analysis] Check for error surfaces. If the target surface is an error surface, ensure that we return the appropriate error surface. Likewise, avoid writing to error surfaces. src/cairo-analysis-surface.c | 16 +++++++++++++--- 1 file changed, 13 insertions(+), 3 deletions(-) commit 5b28b0b903cb2fdb8a5614659d528bf12488389e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Oct 18 00:35:48 2008 +0100 [ps] Destroy type3 surface on error. After an error, ensure that the local type3 surface is destroyed. src/cairo-ps-surface.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) commit 6b17c6da47c42dd04ed2acad723c25b6da41b51c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Oct 18 00:34:17 2008 +0100 [type3] Acquire scaled_font mutex whilst looking up glyphs. When looking up the glyph in the shared scaled_font glyph cache, the caller is required to have taken the scaled_font->mutex. src/cairo-type3-glyph-surface.c | 50 ++++++++++++++++++++++++----------------- 1 file changed, 30 insertions(+), 20 deletions(-) commit f56cf93181e73b4ba74f25ce14f7ed6c7cf36e83 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 17 22:13:53 2008 +0100 [user-font] Check for error objects. Check that the user has not passed in an inert error object before any attempts to write to it and return the default values. src/cairo-user-font.c | 71 ++++++++++++++++++++++++++++++++++++++++++--------- 1 file changed, 59 insertions(+), 12 deletions(-) commit 1f9f9d936b296dbe796b1436c7da7fa3462f7d59 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 17 15:39:56 2008 +0100 [pdf] Acquire scaled_font mutex for show_text_glyphs(). We need to be holding the scaled font mutex over a call to draw text glyphs from within an smask group. src/cairo-pdf-surface.c | 2 ++ 1 file changed, 2 insertions(+) commit 12fb8c9b7c808ab20bee466aa28ee368559fd902 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 17 15:33:07 2008 +0100 [type1] Acquire scaled_font mutex. In order to perform glyph lookups we need to be holding the mutex for the scaled_font. src/cairo-type1-fallback.c | 2 ++ 1 file changed, 2 insertions(+) commit ca5f868a73b35eda737ae6596efff91f82eeea75 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 17 15:26:20 2008 +0100 [scaled-font] Zero font extents for an error surface. Do not attempt to read from the error object, but just return zero extents. src/cairo-scaled-font.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-) commit 1ddf0b2a5c1f3d20b9a91acf7aae021f6b738485 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 17 10:35:38 2008 +0100 [surface] Reorder asserts to make no assumptions about error objects. If the surface is in error, then we cannot assume anything about the validity of its contents other than the error status (and reference count). This is for the cases were the surface is replaced by a nil surface, and in future where the error surface may be replaced by a tiny error object. src/cairo-surface.c | 94 ++++++++++++++++++++++++++--------------------------- 1 file changed, 47 insertions(+), 47 deletions(-) commit 644e78ca4e6f72499fcf763acd53fa7235b0226d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Oct 16 20:34:13 2008 +0100 [svg] Lock the scaled_font whilst emitting glyphs. We need to hold the scaled_font mutex whilst looking upon glyphs so lock the font whilst iterating over the font subset. src/cairo-svg-surface.c | 2 ++ 1 file changed, 2 insertions(+) commit 4587e9e4f5c46656108b05f7a4841f78df14de26 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Oct 16 19:11:09 2008 +0100 [scaled-font-subsets] Add locking to unscaled font. During map_glyph() we are passed a locked scaled_font for which we are asked to add a glyph from that font into the active subsets for the surface. This means that we eventually call scaled_glyph_lookup() to load the glyph. But first, we attempt to find an existing an existing sub_font for glyph, creating a new sub_font as necessary (possibly using an entirely different unhinted scaled font). So before accessing the glyph cache we need to make sure that we are holding the appropriate mutexes. src/cairo-scaled-font-subsets.c | 23 +++++++++++++++++------ 1 file changed, 17 insertions(+), 6 deletions(-) commit d3a6651237a276c410885578c41fb505f83372ed Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Oct 16 14:37:42 2008 +0100 [xlib] Set return code after failing to allocate glyph surface. Although we checked for an allocation error, we missed propagating the status to the return error code. src/cairo-xlib-surface.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) commit bfc3a72cff1a3f9641dae7c9f121598a14eb9a5a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Oct 16 14:06:03 2008 +0100 [gstate] Propagate error from clip. During clip transformation we query the clip extents which can trigger a fatal error. Check and propagate. src/cairo-gstate.c | 33 +++++++++++++++++++++++---------- 1 file changed, 23 insertions(+), 10 deletions(-) commit 7975cf3f2c0bc28b8b3c916d5ba725abb38b300d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Oct 16 13:21:28 2008 +0100 [paginated] Free local reference to target on error path. On the error path we must destroy the local reference to the target surface. src/cairo-paginated-surface.c | 1 + 1 file changed, 1 insertion(+) commit 4662204a2a02d50e674f121d82bbb4fe1b8f1436 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Oct 16 12:44:45 2008 +0100 [test-paginated] Fix double free of surface along error path. The ownership of the surface is transferred to the paginated surface, so we do not need to destroy it ourselves along the error path. src/test-paginated-surface.c | 16 +++++++++------- 1 file changed, 9 insertions(+), 7 deletions(-) commit bf3202fcfd281be3fa62c2d6719377a8f2a0dec4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Oct 16 12:33:22 2008 +0100 [xlib] Fix double free of Pixmap along error path. If the paint fails then we attempt to free the Pixmap ourselves and via the cairo_surface_destroy (as it also believes that it owns the Pixmap). src/cairo-xlib-surface.c | 16 ++++++++++------ 1 file changed, 10 insertions(+), 6 deletions(-) commit 901f0b540c764e131dd9745def329308af61b3d5 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Oct 16 12:19:09 2008 +0100 Review backend->create_similar() Avoid masking fatal errors by enforcing any error to be returned via an error surface, so that the NULL return only means UNSUPPORTED. A few backends called their create_similar() directly, without correctly checking for a potential NULL (for example, the directfb backend was a timebomb, since it used NULL to indicate out-of-memory). src/cairo-directfb-surface.c | 18 ++++++++++-------- src/cairo-glitz-surface.c | 17 +++++++++++------ src/cairo-quartz-surface.c | 4 ++-- src/cairo-surface.c | 8 ++------ src/cairo-win32-surface.c | 16 ++++++++++++---- src/cairo-xcb-surface.c | 2 ++ 6 files changed, 39 insertions(+), 26 deletions(-) commit 9529699028a5c7a3b0f81f945d25f26285247a11 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Oct 19 09:35:07 2008 +0100 [test/user-font-rescale] Check and propagate errors. We need to check the error status on set_user_data() to be sure that the data has been set and error otherwise. test/user-font-rescale.c | 63 ++++++++++++++++++++++++++++++------------------ 1 file changed, 40 insertions(+), 23 deletions(-) commit 683de2fea2d8cdeea9759cbf3ab7492d634e7af2 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Oct 18 00:01:39 2008 +0100 [test/user-font[-proxy]] Check for error from set_user_data(). Check in case we fail to attach our user_data. test/user-font-proxy.c | 69 ++++++++++++++++++++++++++++---------------------- test/user-font.c | 24 ++++++++++++++---- 2 files changed, 58 insertions(+), 35 deletions(-) commit 198392990082e01443bbda5c0cbe6dc8bf090baf Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 17 22:14:49 2008 +0100 [test/bitmap] Leak on error path. Free the FcPattern on error. test/bitmap-font.c | 8 ++++++++ 1 file changed, 8 insertions(+) commit 0409be426cb7f67974346a93213dd4675b59776c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Oct 16 14:26:15 2008 +0100 [test/font-matrix-translation] Check for OOM First check that querying the extents did not trigger NO_MEMORY. test/font-matrix-translation.c | 5 +++++ 1 file changed, 5 insertions(+) commit 21360bf6748b30cf58d54561e40a1aa711d8d7af Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 17 20:44:51 2008 +0100 [test/source-surface-big-scaled-down] Propagate error. Propagate error to source. test/surface-pattern-big-scale-down.c | 43 ++++++++++++++++++++--------------- 1 file changed, 25 insertions(+), 18 deletions(-) commit 794160c35baf39593942ba71b45e4e75d2a87103 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 17 20:37:43 2008 +0100 [test/surface-pattern] Propagate error. Use cairo_get_target() to propagate error from secondary context. test/surface-pattern.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) commit 0f48cce8d90cf95afcf86d71a932ab4e504b909b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 17 20:31:57 2008 +0100 [test/source-clip-scale] Propagate error. Use cairo_get_target() to propagate error from secondary context. test/source-clip-scale.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) commit 6cdbd132ceba159bb7e3ac5612aa79293b21e95a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 17 20:22:49 2008 +0100 [test/source-clip] Propagate error. Use cairo_get_target() to propagate error from the secondary context. test/source-clip.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit f86f233b3e49c3f401b2fa02987b2193485e8be5 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 17 20:14:06 2008 +0100 [test/smask] Propagate error. Use cairo_get_target() to propagate errors from the secondary context. test/smask.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) commit 4490a6c363a8745ba89bfa890519a570d098086e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 17 19:15:17 2008 +0100 [test/stroke-image] Propagate error Use cairo_get_target() to propagate errors from the secondary context. test/stroke-image.c | 19 +++++++++++-------- 1 file changed, 11 insertions(+), 8 deletions(-) commit 6e6d7e52a3c16787633e5059ae36f2fb0d03148a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 17 19:02:31 2008 +0100 [test/mask] Propagate error from secondary context. Reset the source every time so that the error is propagate from the secondary context using cairo_get_target(). test/mask.c | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) commit a56d4530d6149e2c4fc884a5cfeede2d6ff6d3b6 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 17 15:54:21 2008 +0100 [test/smask-text] Propagate error. Use cairo_get_target() to propagate error from the secondary context. test/smask-text.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) commit b1e884cd89ab2b83d3b8e072414e4e3f63e3a023 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 17 15:52:55 2008 +0100 [test/smask-stroke] Propagate error. Use cairo_get_target() to propagate the error from the secondary context. test/smask-stroke.c | 13 ++++++------- 1 file changed, 6 insertions(+), 7 deletions(-) commit 546fbc2c0cf87abd627477056bf19063c5aa001a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 17 15:50:25 2008 +0100 [test/smask-paint] Propagate error. Use cairo_get_target() to propagate errors from the secondary context. test/smask-paint.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) commit 8bd69132827fa539d9dfb65bd00e00c079827360 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 17 15:49:02 2008 +0100 [test/smask-mask] Propagate error. Use cairo_get_target() to propagate errors from the secondary contexts. test/smask-mask.c | 46 ++++++++++++++++++++++------------------------ 1 file changed, 22 insertions(+), 24 deletions(-) commit 9c0fe6c61e92044e8599f49e75d38edcdd6be960 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 17 15:45:07 2008 +0100 [test/smask-image-mask] Propagate error. Use cairo_get_target() to propagate any error status from the secondary context. test/smask-image-mask.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) commit 373d6c76b65e826f81873b51461bdf3c30024c06 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 17 15:43:26 2008 +0100 [test/smask-fill] Propagate error status. Use cairo_get_target() to propagate the error from the secondary context. test/smask-fill.c | 13 ++++++------- 1 file changed, 6 insertions(+), 7 deletions(-) commit bb05beaab626d2fca6d219bfd8148dde71073b39 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 17 15:27:11 2008 +0100 [test/show-glyphs-many] Check for NO_MEMORY As we do a manual status check, we need to perform a full check. test/show-glyphs-many.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) commit c0a4ef76e17dfd43d248c13b7f0665da9b847316 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 17 14:57:42 2008 +0100 [test/composite-integer-translate-over-repeat] Propagate error. Use cairo_get_target() to propagate any errors from the secondary context. test/composite-integer-translate-over-repeat.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) commit 4a9e0f0e5ba17a322bbf0f38b2fb418211571d48 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Oct 16 16:50:32 2008 +0100 [test/mask] Propagate error from secondary context. Use cairo_get_target() to infect the primary context with any errors raised whilst drawing the mask. test/mask.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) commit eaec1caa94ac871eb881ac354c4442ddd2a308ea Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Oct 16 16:20:28 2008 +0100 [test/ft-text-*] Check for errors from fontconfig. If a match cannot be found, or if any other error occured, then fontconfig will return NULL. Check and propagate. test/ft-text-antialias-none.c | 31 ++++++++++++++++++++++++++----- test/ft-text-vertical-layout-type1.c | 31 ++++++++++++++++++++++++++----- test/ft-text-vertical-layout-type3.c | 31 ++++++++++++++++++++++++++----- 3 files changed, 78 insertions(+), 15 deletions(-) commit 336eddfdbb31865e3669ceb775089c63ce1db27f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Oct 16 17:19:12 2008 +0100 [test/get-clip] Check the status on the secondary context. As we solely use a secondary context, we must manually report NO_MEMORY errors whilst running under memfault. test/cairo-test.c | 14 +++++++- test/cairo-test.h | 4 +++ test/get-clip.c | 98 ++++++++++++++++++++++++++++--------------------------- 3 files changed, 67 insertions(+), 49 deletions(-) commit d48f9340514c258cfece1b72d2f7e3f7b2d3c7b1 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Oct 16 14:05:14 2008 +0100 [test/clip-operator] Propagate error from secondary context. Use the cairo_get_target() to propagate any errors from the secondary context. test/clip-operator.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) commit d11014386f739f43ec5f290714d7c51cc638f172 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 5 13:37:07 2007 +0100 Add support for lockdep. lockdep is a valgrind skin which performs pthread locking correctness validation. In particular it allows one to write assert(HOLDS_LOCK(mutex)) which both documents the preconditions for a function and enforces them when the program is run under lockdep. As an aide to lockdep (as it works by intercepting the pthread functions), all the mutexes should be initialised and destroyed using pthread_mutex_init() and pthread_mutex_destroy() rather than using static initializers and no-ops. src/cairo-mutex-impl-private.h | 14 ++++++++++++++ src/cairo-mutex-type-private.h | 6 +++++- src/cairo-scaled-font.c | 13 +++++++++++++ 3 files changed, 32 insertions(+), 1 deletion(-) commit bccfdf7d93c2a92a342127fc212770f4053cb2cf Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Oct 16 13:47:27 2008 +0100 [test] Add support for memfault. Add the core support to cairo-test for running the test-suite under a malloc fault injector. This commit contains the adjustments to cairo_test_run() to repeat the test if it detects a failure due to fault injection and complains if it detects unreported faults or memory leaks. build/configure.ac.analysis | 17 +++ test/cairo-test.c | 271 ++++++++++++++++++++++++++++++++++---------- test/cairo-test.h | 10 +- 3 files changed, 239 insertions(+), 59 deletions(-) commit 974fabfe059f9744116d20fd9f8b59f1b8c34b70 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 17 18:52:51 2008 +0100 [ft] Add more comments about FT_Face lifetime issues. Currently we only have an example of how to call FT_Done after the last reference to the FT cairo_font_face_t has been dropped, but do not actually explain why this needs to be done. So add a couple of sentences to clarify the likely lifetime of the FT_Face once exposed to cairo. src/cairo-ft-font.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) commit 5fcfc7ebc9a2d1d947898d269be7ba91d27c5261 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Oct 16 17:13:45 2008 +0100 [test/README] Document all fonts used. Do a little leg-work to answer a query in a bug report and document the results. test/README | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) commit c1164a574d3e2cc4746384875d5c2781114c0ea4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Oct 16 08:49:13 2008 +0100 [type1] Add comment to warn about read() as macro. In case we are ever tempted to simplify the function pointer, warn that it may alias with a macro. src/cairo-type1-subset.c | 5 +++++ 1 file changed, 5 insertions(+) commit ff9ee74b37a88cba2e996bf0cdae1af6b6c6fe30 Author: Adrian Johnson <ajohnson@redneon.com> Date: Thu Oct 16 17:29:33 2008 +1030 Revert unintended changes to test/fallback-resolution.c in the previous commit test/fallback-resolution.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 5911c6966de832fa350a6e9a4d587b7c3b3b4825 Author: Adrian Johnson <ajohnson@redneon.com> Date: Thu Oct 16 17:22:48 2008 +1030 Use correct inverse ctm for stroke in meta surface replay This fixes #9189 Dashed strokes too long in fallback images src/cairo-meta-surface.c | 2 +- test/fallback-resolution.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) commit 6afcd821182137e8c3f128f9600e97cff79978b8 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Oct 15 22:41:26 2008 +0100 [test] Add a spline test. Test the decomposition of 5 different types of spline curve. Useful test for future experiments in improving the decomposition algorithm. Note: the vector targets all need separate reference images due to their lack of support for cairo_set_tolerance(). Also GS strokes the Bezier curve differently using offset curves and opposed to transcribing the outline of a pen. test/.gitignore | 1 + test/Makefile.am | 7 + test/spline-decomposition-pdf-ref.png | Bin 0 -> 20404 bytes test/spline-decomposition-ps2-ref.png | Bin 0 -> 8957 bytes test/spline-decomposition-ps3-ref.png | Bin 0 -> 8957 bytes test/spline-decomposition-ref.png | Bin 0 -> 20402 bytes test/spline-decomposition-svg11-ref.png | Bin 0 -> 20404 bytes test/spline-decomposition-svg12-ref.png | Bin 0 -> 20404 bytes test/spline-decomposition.c | 416 ++++++++++++++++++++++++++++++++ 9 files changed, 424 insertions(+) commit 91b0dc92c8ede189d942188f92068f189907e239 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Oct 15 22:27:53 2008 +0100 [build] Add -fno-common. Generate a warning if two files define a global variable with the same name, instead of silently merging them to reference the same location. build/configure.ac.warnings | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit b94b600b34793cecd9a66d37dd34a6619738e2c2 Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Wed Oct 15 13:41:03 2008 -0400 [quartz] Completely remove CAIRO_HAS_QUARTZ_IMAGE_SURFACE check Instead of fixing the check in d36b02dc662e623bee58a69c32b149a242cbf52b, just remove it. Conditionally compilation will keep cairo-quartz-image-surface.c from being compiled in when it is not configured to be. Suggested by Behdad. src/cairo-quartz-image-surface.c | 4 ---- 1 file changed, 4 deletions(-) commit d36b02dc662e623bee58a69c32b149a242cbf52b Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Wed Oct 15 12:45:11 2008 -0400 [quartz] Rebalance 'CAIRO_HAS_QUARTZ_IMAGE_SURFACE' #ifdef Fixes compilation by re-adding '#ifdef CAIRO_HAS_QUARTZ_IMAGE_SURFACE' which was, I assume, unintentionally removed by 2cf4b3161c19b9b0349bd2aa94855fd0822968be. src/cairo-quartz-image-surface.c | 2 ++ 1 file changed, 2 insertions(+) commit 83162360a44c4d4701f0ade2e2ee813f79186439 Author: Jeff Muizelaar <jmuizelaar@mozilla.com> Date: Wed Oct 15 10:34:12 2008 -0400 Fix quratz_image typo in Makefile.sources. src/Makefile.sources | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 877e77fa3e85c09c29e64ca5ae7329346db029cc Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Oct 15 12:20:00 2008 +0100 [glitz] Disable caching of solid surfaces. glitz currently has no mechanism to release resources upon connection closure (ala XCloseDisplay) and so will attempt invalid accesses when evicting old surfaces from the solid pattern cache. src/cairo-glitz-surface.c | 7 +++++++ 1 file changed, 7 insertions(+) commit fc04a1e0dbc3cefa51c7602e0a23dba886813fe5 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Oct 15 12:13:40 2008 +0100 [boilerplate] Fixup compilation for gliitz. Convert a few old references to cairo_boilerplate_surface_set_user_data. boilerplate/cairo-boilerplate-glitz.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) commit e17f02add019d9acf461a1ad1245ba5be00cf336 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Oct 15 10:52:33 2008 +0100 [type1] Protect against read macro POSIX allows for any of its functions to be implemented as a macro, inflicting damage upon innocent function pointers that happen to share the same name. face->stream->read() is one such example. As it's outside of our control (being a FT_Face) we cannot simply rename the data member, so we need to explicitly dereference it as a function pointer in order to disambiguate it for the compiler. src/cairo-type1-subset.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) commit bc872a5f5fb9b6318dc78d132da46a6b201f75c7 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Oct 15 10:21:05 2008 +0100 [xlib] Handle 4,20,24,28 depth XImages Bug 9102 cairo doesn't support 24 bits per pixel mode on X11 (https://bugs.freedesktop.org/show_bug.cgi?id=9102) is a reminder that that we need to support many obscure XImage formats. With Carl's and Behdad's work to support psuedocolor we have a mechanism in place to handle any format that is not natively handled by pixman. The only piece we were missing was extending the swapper to handle all-known formats and putting in defensive checks that pixels were correctly aligned in accordance with pixman's requirements. src/cairo-image-surface.c | 13 ++++----- src/cairo-xlib-surface.c | 70 +++++++++++++++++++++++++++++++++++++++-------- src/cairoint.h | 2 +- 3 files changed, 66 insertions(+), 19 deletions(-) commit 1728fa352011c570ce1bcc3bd14897e789c85252 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Oct 15 09:23:01 2008 +0100 [svg] Add notes about why we shouldn't add DOCTYPE. Emmanuel Pacaud did some research into adding DOCTYPE after we had a complaint that our SVG output failed to be validated by batik. The conclusion he came to was that the validation of SVG was incomplete and misleading, i.e. our output might generate false negatives leading to more confusion. src/cairo-svg-surface.c | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+) commit e9f5ee6efe7725c1e0e32fa9e85cca61ecb93440 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Oct 14 14:42:48 2008 +0100 [doc] Fix a few gtk-doc errors. gtk-doc insists on a non-empty long description, even for trivial functions that are fully described by their input arguments and return value. Grrr. src/cairo-gstate.c | 4 ++++ src/cairo-os2-surface.c | 2 ++ src/cairo-scaled-font.c | 14 +++++++------- src/cairo-surface.c | 4 +++- src/cairo-unicode.c | 6 ++++-- 5 files changed, 20 insertions(+), 10 deletions(-) commit 595414bb787d570ac95fc91c58c370558e9d5a54 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Oct 14 14:24:49 2008 +0100 [test/fallback-resolution] Extend to cover separate ppi x/y. Adrian Johnson proposed a hack that fixed the current test by simply scaling the dashed stroke by device_transform.xx. Obviously this exposes a deficiency in the test case. So add an outer loop to iterate over several common ppi for both x and y and in doing so reveal more ugly artifacts in the current fallback code. test/Makefile.am | 42 ++++- ....png => fallback-resolution-ppi150x150-ref.png} | Bin test/fallback-resolution-ppi150x300-ref.png | Bin 0 -> 6848 bytes test/fallback-resolution-ppi150x37.5-ref.png | Bin 0 -> 13176 bytes test/fallback-resolution-ppi150x600-ref.png | Bin 0 -> 6338 bytes test/fallback-resolution-ppi150x72-ref.png | Bin 0 -> 8370 bytes test/fallback-resolution-ppi150x75-ref.png | Bin 0 -> 10445 bytes test/fallback-resolution-ppi300x150-ref.png | Bin 0 -> 6851 bytes ....png => fallback-resolution-ppi300x300-ref.png} | Bin test/fallback-resolution-ppi300x37.5-ref.png | Bin 0 -> 12870 bytes test/fallback-resolution-ppi300x600-ref.png | Bin 0 -> 4848 bytes test/fallback-resolution-ppi300x72-ref.png | Bin 0 -> 7053 bytes test/fallback-resolution-ppi300x75-ref.png | Bin 0 -> 9617 bytes test/fallback-resolution-ppi37.5x150-ref.png | Bin 0 -> 12728 bytes test/fallback-resolution-ppi37.5x300-ref.png | Bin 0 -> 11981 bytes ...ng => fallback-resolution-ppi37.5x37.5-ref.png} | Bin test/fallback-resolution-ppi37.5x600-ref.png | Bin 0 -> 11654 bytes test/fallback-resolution-ppi37.5x72-ref.png | Bin 0 -> 12990 bytes test/fallback-resolution-ppi37.5x75-ref.png | Bin 0 -> 13787 bytes test/fallback-resolution-ppi600x150-ref.png | Bin 0 -> 6096 bytes test/fallback-resolution-ppi600x300-ref.png | Bin 0 -> 4812 bytes test/fallback-resolution-ppi600x37.5-ref.png | Bin 0 -> 12401 bytes ....png => fallback-resolution-ppi600x600-ref.png} | Bin test/fallback-resolution-ppi600x72-ref.png | Bin 0 -> 6398 bytes test/fallback-resolution-ppi600x75-ref.png | Bin 0 -> 8974 bytes test/fallback-resolution-ppi72x150-ref.png | Bin 0 -> 8069 bytes test/fallback-resolution-ppi72x300-ref.png | Bin 0 -> 6803 bytes test/fallback-resolution-ppi72x37.5-ref.png | Bin 0 -> 13147 bytes test/fallback-resolution-ppi72x600-ref.png | Bin 0 -> 6245 bytes ...ef.png => fallback-resolution-ppi72x72-ref.png} | Bin test/fallback-resolution-ppi72x75-ref.png | Bin 0 -> 10628 bytes test/fallback-resolution-ppi75x150-ref.png | Bin 0 -> 9940 bytes test/fallback-resolution-ppi75x300-ref.png | Bin 0 -> 9003 bytes test/fallback-resolution-ppi75x37.5-ref.png | Bin 0 -> 14020 bytes test/fallback-resolution-ppi75x600-ref.png | Bin 0 -> 8455 bytes test/fallback-resolution-ppi75x72-ref.png | Bin 0 -> 10202 bytes ...ef.png => fallback-resolution-ppi75x75-ref.png} | Bin test/fallback-resolution.c | 199 +++++++++++---------- 38 files changed, 139 insertions(+), 102 deletions(-) commit 0d3e152d2fe28b22a27dd6b9f82e2047aea3be94 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Oct 14 13:44:47 2008 +0100 [check-doc-syntax.sh] Allow quoting. func_regexp was incorrectly complaining that cairo_pure was a function without parenthesis. The simplest solution appeared to be allow quoting of 'cairo_pure'. src/cairo-compiler-private.h | 22 +++++++++++----------- src/check-doc-syntax.sh | 6 +++--- 2 files changed, 14 insertions(+), 14 deletions(-) commit 02a7ca80f9f9b62deff6c8ba4dc58fee0cebcaa6 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Oct 14 11:08:43 2008 +0100 [pattern] Correctly optimize away fractional translations. As is so often the case, reading the commit log gives you fresh insight in the problem - often called confessional debugging... We can simplify the problem by ignoring attr->[xy]_offset, for the time being, and focus on computing the correct matrix. This is comparatively simple as all we need to do is perform the appropriate rounding on the translation vector. src/cairo-pattern.c | 18 ++++++++++-------- 1 file changed, 10 insertions(+), 8 deletions(-) commit 9886cb3353eb02ce5b99d555a35b13b8347f8e87 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Oct 14 10:23:41 2008 +0100 [pattern] Only perform non-integer optimization for identity matrices. A complication I realised after pushing 3eb4bc3 was handling larger sampled areas. Extending the test case revealed that the optimization was broken for anything but the identity transform (after removing the translation). Correctness first, leaving the "pixel-exact" solution for interested reader... src/cairo-pattern.c | 16 ++++++---------- test/filter-nearest-transformed-pdf-ref.png | Bin 373 -> 548 bytes test/filter-nearest-transformed-ref.png | Bin 151 -> 570 bytes test/filter-nearest-transformed-svg11-ref.png | Bin 217 -> 559 bytes test/filter-nearest-transformed-svg12-ref.png | Bin 217 -> 559 bytes test/filter-nearest-transformed.c | 24 ++++++++++++++++++++++++ 6 files changed, 30 insertions(+), 10 deletions(-) commit 8dc3deecb8e0ca2f84d996096230f27ab4c24c18 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Oct 14 08:17:23 2008 +0100 Add documentation for function attributes 'pure' and 'const'. Carl suggested that cairo_pure and cairo_const are pretty opaque, even to the developer who added them, so it is extremely important that they have a good description so that they are used correctly and perhaps ported to other compilers. src/cairo-compiler-private.h | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+) commit 6aa1143d5a8ec45fc793102a64e956d7d1c7ad01 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Oct 13 23:49:17 2008 +0100 Update documentation to remove references to cairo_has_show_text_glyphs(). With the addition of cairo_show_text_glyphs() came a couple of functions to query whether the target supported the extended attributes. However, at Carl's request cairo_has_show_text_glyphs() was removed - but the documentation was not updated to reflect that. NEWS | 2 +- src/cairo-surface.c | 3 --- src/cairo.c | 6 +++--- 3 files changed, 4 insertions(+), 7 deletions(-) commit b2053945df35d4d7eea9fe3200e9c018705309d1 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Oct 13 23:39:47 2008 +0100 [scaled-font] Correct documentation for cairo_scaled_font_create(). Fix the documentation not to suggest that NULL is a legal value for font_options (as we temporarily made it during the 1.[57] dev cycle). src/cairo-scaled-font.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) commit 3eb4bc37577e905d93a1935e7a9cd33ae8dbda15 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Oct 13 23:26:04 2008 +0100 [pattern] Optimize away fractional translation for NEAREST patterns. As identified in bug 15479, Unpredictable performance of cairo-xlib with non-integer translations of a source surface pattern (https://bugs.freedesktop.org/show_bug.cgi?id=15479), source surfaces with a fractional translation hit slow paths for some drivers, causing seemingly random performance variations. As a work-around Owen Taylor proposed that cairo could convert non-integer translations on NEAREST sources patterns to their integer equivalents. The messy detail involved here is replicating the rounding mode used by pixman for the sample offset, but otherwise the conversion is fairly trivial. src/cairo-pattern.c | 35 +++++++++ test/.gitignore | 1 + test/Makefile.am | 5 ++ test/filter-nearest-transformed-pdf-ref.png | Bin 0 -> 373 bytes test/filter-nearest-transformed-ref.png | Bin 0 -> 151 bytes test/filter-nearest-transformed-svg11-ref.png | Bin 0 -> 217 bytes test/filter-nearest-transformed-svg12-ref.png | Bin 0 -> 217 bytes test/filter-nearest-transformed.c | 103 ++++++++++++++++++++++++++ 8 files changed, 144 insertions(+) commit ceab32e2c3d51c0dd29b105a5a390cdd8aa58f8f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Oct 12 12:11:12 2008 +0100 [win32] win32 also uses font subsets. Include win32 in the list of font subset users, fixes build when cross-compiling. src/cairoint.h | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) commit 1d3453eee50e4343995c5d3f225d18289fbcebec Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Oct 12 11:21:14 2008 +0100 [pattern] Rescale the linear pattern matrix to avoid overflow. As proof-of-principle, compute a scale factor to avoid overflow when converting a linear pattern to pixman_fixed_t. Fixes test/huge-pattern, but the principle should be extended to handle more cases of overflow. src/cairo-pattern.c | 48 +++++++++++++++++++++++++++++++----- test/Makefile.am | 5 ++-- test/huge-pattern-pdf-ref.png | Bin 0 -> 2716 bytes test/huge-pattern-pdf-rgb24-ref.png | Bin 0 -> 2817 bytes test/huge-pattern-ps3-ref.png | Bin 0 -> 1786 bytes 5 files changed, 45 insertions(+), 8 deletions(-) commit ab1d106cba7aa4abe5f7253c089aadd15e0bb06d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Oct 11 19:52:28 2008 +0100 [xlib] Use the cached xrender formats. Use the cached formats for consistency and simplify several double pointer dereferences. src/cairo-xlib-surface.c | 64 ++++++++++++++++++++++++++++++------------------ 1 file changed, 40 insertions(+), 24 deletions(-) commit 6736faba3ebe4eae4c1d2aeeb3153eb6f12a32de Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Oct 11 18:10:16 2008 +0100 [test/any2ppm] Do not attempt to compile PS without spectre Since CAN_TEST_PS_SURFACE does not currently require spectre, we were attempting to compile in spectre support for any2ppm even on systems without libspectre installed. Fix that by adding a separate flag for CAIRO_HAS_SPECTRE. configure.ac | 5 ++++- test/Makefile.am | 2 +- test/any2ppm.c | 28 ++++++++++++++++++++-------- 3 files changed, 25 insertions(+), 10 deletions(-) commit b35871116243724b59738fce00b512818c83ea17 Author: Benjamin Otte <otte@gnome.org> Date: Sat Oct 11 00:18:11 2008 +0100 [test] Add huge pattern. Add a test case to exercise range overflow during gradient construction. test/.gitignore | 1 + test/Makefile.am | 4 +++ test/huge-pattern-ref.png | Bin 0 -> 1619 bytes test/huge-pattern.c | 74 ++++++++++++++++++++++++++++++++++++++++++++++ 4 files changed, 79 insertions(+) commit 77070270813d2ca40d422e5301f75222d19e1d8f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 10 22:24:01 2008 +0100 [test] Use _POSIX_C_SOURCE for flockfile. From bug 18010 (https://bugs.freedesktop.org/show_bug.cgi?id=18010), in order to make flockfile() available we need to set _POSIX_C_SOURCE and according to the man page, the appropriate feature check is _POSIX_THREAD_SAFE_FUNCTIONS. test/cairo-test.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) commit 34ff7e4ac42acfb2ab9921da21933113518bd8a0 Author: Adrian Johnson <ajohnson@redneon.com> Date: Sat Oct 11 23:44:37 2008 +1030 Check that reads in truetype_reverse_cmap are within valid data Bulia Byak reported a bug where cairo was crashing with a particular font. The font had an incorrect entry in the cmap table that caused cairo to read from outside of the buffer allocated for the cmap. src/cairo-truetype-subset.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) commit a16ef6ead2b02120ca482aa3d2667e51b0c67e78 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 10 18:03:49 2008 +0100 [INSTALL] Mention pkg-config requirement Mention that building the configure script requires at least version 0.16 of pkg-config. See the old bug: Bug 4702 PKG_PROG_PKG_CONFIG: command not found (https://bugs.freedesktop.org/show_bug.cgi?id=4702) INSTALL | 1 + 1 file changed, 1 insertion(+) commit eba9d87d541d89804f70e56874538a249de08502 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 10 16:51:07 2008 +0100 [image] Warn about a potential NULL return after finish(). Improve the wording as suggested by https://bugs.freedesktop.org/show_bug.cgi?id=12659 to warn about the possibility of returning NULL even if the passed in surface was an image surface. src/cairo-image-surface.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit ac3e4b4bea574c94de77ec15878d116b619ee3f4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Oct 10 14:00:32 2008 +0100 [xlib] Disable XRender support if the surface has no xrender_format Bug 11734: XRender crashes due to NULL pointer from Cairo on SGI O2 (https://bugs.freedesktop.org/show_bug.cgi?id=1173) is an example of a case where we try to perform an XRender operation on a surface with a format that was not supported by the XRender extension. By marking the extension version as -1 on those surfaces, the current checks for SURFACE_RENDER_HAS_* always correctly return false and prevent us try to create a Picture with a NULL xrender_format. src/cairo-xlib-surface.c | 37 ++++++++++++++++++++++++------------- 1 file changed, 24 insertions(+), 13 deletions(-) commit 8ac8e8c523abaa5db24c5303c671ff3487bf0801 Author: Carlos Garcia Campos <carlosgc@gnome.org> Date: Sun Sep 21 14:29:30 2008 +0200 [test] Add ps2png check program using libspectre Add a simple program to compliment pdf2png and svg2png. configure.ac | 1 + test/Makefile.am | 8 ++++ test/ps2png.c | 113 +++++++++++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 122 insertions(+) commit 3e6afb353da1fee624b519f5a96b3303c7eb91ae Author: Carlos Garcia Campos <carlosgc@gnome.org> Date: Thu Oct 9 12:11:51 2008 +0100 [test/any2ppm] Enable PS conversion using libspectre. Complete the vector trilogy using libspectre to provide a similar interface (to poppler and librsvg) around GhostScript. AUTHORS | 1 + configure.ac | 20 ++++++++++++---- test/Makefile.am | 4 ++-- test/any2ppm.c | 73 ++++++++++++++++++++++++++++++++++++++++++++++++++++++-- 4 files changed, 90 insertions(+), 8 deletions(-) commit ddd1615a1777181c6e8db1dbafacb68535ed163a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Oct 8 23:13:03 2008 +0100 [pattern] Remove incorrect assert. It's possible to reach that point without setting the filter to NEAREST, for example if using FAST or GAUSSIAN. src/cairo-pattern.c | 1 - 1 file changed, 1 deletion(-) commit c647c0bbca8694a14f5f5807b8349f0b0bb4987d Author: Benjamin Otte <otte@gnome.org> Date: Thu Oct 9 13:21:27 2008 +0200 [TEST] make imagediff work again It seems to have bitrotten test/imagediff.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) commit f644d78dc138a7ba920a108dfb058e9f0fdb5302 Author: Björn Lindqvist <bjourne@gmail.com> Date: Wed Oct 8 20:08:25 2008 +0100 [perf] Add composite performance test. Add a new test case to Cairo for checking the performance of Cairo's equivalent to GDK's gdk_pixbuf_composite_color() operation. That is an operation that happens to be extremely useful when viewing or editing transparent images so I think it is important that it is as fast as possible. AUTHORS | 1 + perf/Makefile.am | 1 + perf/cairo-perf.c | 1 + perf/cairo-perf.h | 1 + perf/composite-checker.c | 108 +++++++++++++++++++++++++++++++++++++++++++++++ 5 files changed, 112 insertions(+) commit 96b083d3f0e47f919cb6d0f06735427eaea21565 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Oct 8 18:53:46 2008 +0100 [test] Exercise degenerate dashes. Add a test case to capture the current behaviour when a segment ends on an off/on dash transition. Originally filed as bug: Miter artifacts for dashed strokes https://bugs.freedesktop.org/show_bug.cgi?id=17973 test/.gitignore | 1 + test/Makefile.am | 2 + test/degenerate-dash-ref.png | Bin 0 -> 2005 bytes test/degenerate-dash.c | 97 +++++++++++++++++++++++++++++++++++++++++++ 4 files changed, 100 insertions(+) commit 6487a147a2129f588aae900b048b2a801ef95ec0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Oct 8 15:26:03 2008 +0100 [cairo.h] Be defensive when checking for _MSC_VER Since cairo.h is a public header file, we need to be careful so that it can be compiled by random compilers and even users specifying "-Werror -Wundef" (mentioning no names, Company). So replace the bare (and legal) #if _MSC_VER with #if defined (_MSC_VER) just in case. src/cairo.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 59141e56a723f741c87e2eaa37bf96e243ac5728 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Oct 8 13:50:28 2008 +0100 [win32] Restore fine-grained fallback support for printing. Missed updating win32 when implementing the new paginated API to query support for the fine-grained fallbacks. (Thanks to Adrian for spotting this oversight.) src/cairo-win32-printing-surface.c | 8 ++++++++ 1 file changed, 8 insertions(+) commit 565c02a37e2d140a20867081eea9bb5de9999b1c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Oct 8 11:06:42 2008 +0100 [svg] Use finer-grained fallbacks for SVG 1.2 The use of fine-grained fallbacks requires the native support of the SOURCE operator applied to an image on the target surface. SVG 1.2 introduces the "comp-op:src" mode fulfilling this criteria - so we can enable fine-grained fallbacks for 1.2+. Update test/fine-grained-fallbacks to exercise this pathway in SVG 1.2 - as SVG natively supported all the current operations within that test. This reveals yet another librsvg bug in handling SVG 1.2. src/cairo-paginated-private.h | 9 ++- src/cairo-paginated-surface.c | 47 +++++--------- src/cairo-pdf-surface.c | 14 ++++- src/cairo-ps-surface.c | 8 +++ src/cairo-svg-surface.c | 34 +++++++--- test/Makefile.am | 4 +- test/finer-grained-fallbacks-ps2-argb32-ref.png | Bin 1047 -> 0 bytes test/finer-grained-fallbacks-ps2-ref.png | Bin 0 -> 1356 bytes test/finer-grained-fallbacks-ps2-rgb24-ref.png | Bin 819 -> 1096 bytes test/finer-grained-fallbacks-ps3-argb32-ref.png | Bin 1047 -> 0 bytes test/finer-grained-fallbacks-ps3-ref.png | Bin 0 -> 1356 bytes test/finer-grained-fallbacks-ps3-rgb24-ref.png | Bin 819 -> 1096 bytes test/finer-grained-fallbacks-ref.png | Bin 796 -> 1111 bytes test/finer-grained-fallbacks-rgb24-ref.png | Bin 590 -> 1114 bytes test/finer-grained-fallbacks.c | 79 +++++++++++++++++++++++- 15 files changed, 147 insertions(+), 48 deletions(-) commit 0c777a3e0de0d48289432a3d66be68f8779b08a0 Author: Vladimir Vukicevic <vladimir@pobox.com> Date: Tue Oct 7 15:20:33 2008 -0700 [win32] Use wide-char versions of some API functions, for compat with Windows Mobile src/cairo-win32-surface.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) commit cf572b58e1197dac0ff1795b27b142c3e044cf45 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Oct 7 09:55:03 2008 +0100 [tessellator] Compile fixes for !HAVE_INT64_T Fixup a couple of instances of implicit down-casting to 32bits from a 64bit wide integer and add a new is_zero() predicate. src/cairo-bentley-ottmann.c | 14 +++++++------- src/cairo-wideint-private.h | 10 +++++++++- 2 files changed, 16 insertions(+), 8 deletions(-) commit 6b8c0559620ab23c4df1f381d2e95ffc307d2e2f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Oct 7 09:43:45 2008 +0100 [tessellator] Avoid implicit promotion to 64bit integer. Avoid passing a 32bit integer as a cairo_int64_t in case we do not have a 64bit native integral type. As a side-effect this means we can also use a narrower multiply. src/cairo-bentley-ottmann.c | 20 ++++++++++---------- 1 file changed, 10 insertions(+), 10 deletions(-) commit eaa4bd13926728e9da97a23df8a465ef2296049a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Oct 7 21:09:16 2008 +0100 [pattern] After cloning adjust [xy]_offset if possible. For the simple case where the pattern matrix only contains an integer translation then care is taken to convert that to a identity source matrix with the translation applied to the [xy]_offsets. 5b97ee6525 broke this guarantee by applying the clone offsets to the source matrix. So when the source matrix is identity we can simply adjust the [xy]_offsets and preserve the identity matrix. (This idea can be extended further by removing any integer translation from the source matrix and storing it in the [xy]_offsets as a means to extend the limited precision in pixman_matrix_t - encountered when downscaling large images offset onto the target surface.) src/cairo-pattern.c | 16 +++++++++++++--- 1 file changed, 13 insertions(+), 3 deletions(-) commit 552cc09e6be2b704dc32f986c84640d50316c25c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Oct 7 21:05:57 2008 +0100 [xlib] Check integer translation for XCopyArea. A precondition for using the core XCopyArea protocol is that the source attributes contain only integer translations. However, we failed to apply any integer translations from the source matrix to the XCopyArea offsets. This worked prior to 5b97ee6525 as _cairo_pattern_acquire_surface_for_surface() was careful to only generate an identity matrix if the pattern matrix only contained an integer translation (and thus we would use XCopyArea in the xlib backend). src/cairo-xlib-surface.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) commit 308485a3fc70bc4455a01990443c1f954b8a66f8 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Oct 7 20:19:10 2008 +0100 [test] Add a test for a reported regression by Michael Natterer. mitch reported on irc that expose events in the gimp were suffering from artifacts which he tracked down to a bug with clipping and source surfaces. This is the cairo test case for that regression. test/.gitignore | 1 + test/Makefile.am | 2 ++ test/clipped-surface-ref.png | Bin 0 -> 401 bytes test/clipped-surface.c | 70 +++++++++++++++++++++++++++++++++++++++++++ 4 files changed, 73 insertions(+) commit 28a3b546b91290f3377777adedcf7ab71e599188 Author: Vladimir Vukicevic <vladimir@pobox.com> Date: Tue Oct 7 10:25:46 2008 -0700 [win32] create_similar should fall back to DIBs if DDB creation fail src/cairo-win32-surface.c | 43 +++++++++++++++++-------------------------- 1 file changed, 17 insertions(+), 26 deletions(-) commit 4ae37a97b3e415591caa611533f44db5a29cfb72 Author: Vladimir Vukicevic <vladimir@pobox.com> Date: Tue Oct 7 10:18:41 2008 -0700 [quartz] initialize do_reflect field correctly for pattern draws src/cairo-quartz-surface.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) commit 8cf399b944294645833088ee2ac1313093948b66 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Oct 6 18:59:15 2008 +0100 [test] Add a test to exercise the internal twin font Although Behdad promises that the rendering is subject to improvement, introduce a test to check consistency across platforms and update as required. test/.gitignore | 1 + test/Makefile.am | 4 ++++ test/twin-ps2-ref.png | Bin 0 -> 1167 bytes test/twin-ps3-ref.png | Bin 0 -> 1167 bytes test/twin-ref.png | Bin 0 -> 1836 bytes test/twin.c | 60 ++++++++++++++++++++++++++++++++++++++++++++++++++ 6 files changed, 65 insertions(+) commit 1f813ff04022a36a91bf9d1b0f7f2ae1c63b399f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Oct 6 17:55:37 2008 +0100 [array] Silence gtk-doc complaints. gtk-doc fails make check for array as it insists that even the simplest functions must have a long description and cannot be entirely described by their arguments and return value. src/cairo-array.c | 26 ++++++++++++++++++++------ 1 file changed, 20 insertions(+), 6 deletions(-) commit 7cb9ccb4cb2106bb59e181ee284f8dcc5504eb7a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Oct 6 17:44:03 2008 +0100 Fixup a couple of trivials doc warnings. Correct the 'Returns' statement to match what gtk-doc expects. src/cairo-cache.c | 3 +-- src/cairo-pattern.c | 5 ++--- 2 files changed, 3 insertions(+), 5 deletions(-) commit d6f0351b6cbb0d542a069eb5d0a7377eb85a6e4e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Oct 6 17:17:04 2008 +0100 [skiplist] Avoid repeated calls to compare on the same element when inserting. During insertion we must traverse the skiplist in order to find the insertion point for the new element. As we descend each level, the next element in the chain for this level is sometimes the same as the one we just compared against (and know that the new element is greater than). Hence we can skip the search on that level and descend to the next. During world_map this reduces the number of calls into _sweep_line_elt_compare() by ~2.5% (and when performing trapezoidation on strokes gives a similar speed up of about 2% - not bad for the addition of a single line.) src/cairo-skiplist.c | 16 ++++++++++------ 1 file changed, 10 insertions(+), 6 deletions(-) commit 1440399625ae0579d0748475fc924cfe74339a21 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Oct 6 16:15:29 2008 +0100 [hull] Replace open-coding of 64bit arithmetic. Use primitives from cairo-wideint-private.h - in this case it helps to make the code more readable as well as reduce dependence on native 64bit integers. src/cairo-hull.c | 31 +++++++++++++++++++------------ 1 file changed, 19 insertions(+), 12 deletions(-) commit c6a6bf580fa7036e4c28137f8c038beb5c347244 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Oct 6 16:13:21 2008 +0100 [slope] Replace open-coding 64bit integer arithmetic. Prefer to use the operations form cairo-wideint-private.h in order to improve readability and reduce our assumptions on the availability of 64bit integers. src/cairo-slope.c | 14 ++++++-------- 1 file changed, 6 insertions(+), 8 deletions(-) commit 5e3fcb7934f568bb7e304a1c434a9370b693df17 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Oct 6 17:02:54 2008 +0100 [wideint] Declare the wideint arithmetic functions as const. 'const' is a stricter form of 'pure' in that functions declared with that attribute do not access any values other than their arguments (in contrast to 'pure' which is allowed to read from global memory). src/cairo-compiler-private.h | 2 ++ src/cairo-wideint-private.h | 2 +- 2 files changed, 3 insertions(+), 1 deletion(-) commit 9430bf57ebc86c79e28128e54aeb5a194c9cad03 Author: Carl Worth <cworth@cworth.org> Date: Mon Oct 6 14:00:37 2008 -0700 Fix scrambled version number in generated documentation. In the cairo 1.8.0 release the documentation would get generated with the second and third version components transposed, (so it would say 1.0.8). Fix the obviously mistaken transposition. doc/public/Makefile.am | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 5beae8f530fe7056fa965cb5ca7b345a667e8122 Author: Behdad Esfahbod <behdad@behdad.org> Date: Mon Oct 6 12:43:42 2008 -0400 [Makefile.am.releasing] Make 'make snapshot' print a banner when done Copied from "make distcheck" implementation. Prints something like: ============================================================= cairo-1.8.1-20081006-38de89 archives ready for distribution: cairo-1.8.1-20081006-38de89.tar.gz ============================================================= build/Makefile.am.releasing | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) commit 38de895d3863bd2e6fe921ad6cc6eeff15102701 Author: Behdad Esfahbod <behdad@behdad.org> Date: Mon Oct 6 12:11:18 2008 -0400 [RELEASING] Fix sample distcheck output to reflect current versioning RELEASING | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) commit f8018cf193823356270c8e7f21f3108fbe44679c Author: Behdad Esfahbod <behdad@behdad.org> Date: Mon Oct 6 11:45:08 2008 -0400 [boilerplate] Fix "make dist" The cairo-boilerplate-ps/pdf/svg.h files where removed in commit 9841d9d58ea286f798626d325d50a85bf3f02c8f but the Makefile.sources listing was not updated. boilerplate/Makefile.sources | 3 --- 1 file changed, 3 deletions(-) commit 6578ac148ecdd236b0860923e5dee87291c9966c Author: Behdad Esfahbod <behdad@behdad.org> Date: Mon Oct 6 11:39:47 2008 -0400 [twin-font] Make the stroke lighter I'm still to fine-tune the font and add hinting and bold/oblique support. In the mean time though, a lighter stroke produces much nicer glyphs. src/cairo-font-face-twin.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 2fb7753903daf5c207b22034682a250161af22ad Author: Behdad Esfahbod <behdad@behdad.org> Date: Mon Oct 6 11:39:00 2008 -0400 [Makefile.am.releasing] Exclude uid/gid from tarball By setting GNU tar options --owner=0 --group=0 build/Makefile.am.releasing | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) commit 589564c3b02d5b0505806924bd051323717791a3 Author: Behdad Esfahbod <behdad@behdad.org> Date: Sat Oct 4 01:36:07 2008 -0400 [user-font] Fix comment re backend interface src/cairo-user-font.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit c98c1cb5325bbc05c88ea350ff99d6eaf56fd339 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Oct 5 10:39:10 2008 +0100 [wideint] Mark functions as pure Take advantage of the gcc function attribute 'pure', which tells gcc that the function result only depends upon its arguments and it has zero side effects (e.g. it does not clobber memory). This gives gcc greater opportunity to rearrange and optimize the wideint arithmetic. src/cairo-compiler-private.h | 6 ++++++ src/cairo-wideint-private.h | 2 +- 2 files changed, 7 insertions(+), 1 deletion(-) commit 76dd4603d01068b1b377312ced6b44fe5419794f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Oct 5 10:15:49 2008 +0100 [tessellator] Replace open-coding _cairo_int64_cmp(). We often use the construct: if (_cairo_int64_lt (A, B) return -1; if (_cairo_int64_gt (A, B) return 1; return 0; to compare two large integers (int64, or int128) which does twice the required work on CPUs without large integer support. So replace it with a single wideint function _cairo_int64_cmp() and therefore allow opportunities to both shrink the code size and write a more efficient comparison. (The primarily motivation is to simply replace each block with a single more expressive line.) src/cairo-bentley-ottmann.c | 19 +++--------------- src/cairo-wideint-private.h | 10 +++++++++- src/cairo-wideint.c | 48 +++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 60 insertions(+), 17 deletions(-) commit 6eead4a5f746e182eabfcda9959cd9cc53d95a89 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Oct 4 12:39:21 2008 +0100 [stroke] Simplify dash-off logic. Remove a redundant level of if-else branching (and remove a duplicate code) block by testing dash_on first. src/cairo-path-stroke.c | 65 +++++++++++++++++++++---------------------------- 1 file changed, 28 insertions(+), 37 deletions(-) commit 4aa0991a11f03a47735dcfafe43d9505b939554e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Oct 4 20:22:18 2008 +0100 [test] Mark fallback-resolution as XFAIL. There appears to be no simple solution here, as it seems to be a fundamental flaw in the design of the meta-surface wrt to replaying into a fallback image. (I may be wrong, but if Carl found no easy solution then I feel no shame for my own failure ;-) test/Makefile.am | 13 +++++++++++++ 1 file changed, 13 insertions(+) commit ae87382a84770f8656c369d258f705b8ac20049c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Oct 4 10:54:25 2008 +0100 [tessellator] Special case edge comparisons when on either end-point. If the sweep-line is currently on an end-point of a line, then we know its precise x value and can use a cheaper comparator. Considering that we often need to compare events at end-points (for instance on a start event), this happens frequently enough to warrant special casing. src/cairo-bentley-ottmann.c | 100 ++++++++++++++++++++++++++++++++++++++++++-- 1 file changed, 96 insertions(+), 4 deletions(-) commit ab23c2995356821537b9a0facdff87c339a05d2a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Oct 1 12:18:16 2008 +0100 [tessellator] Direct comparison of result in edges_compare_x_for_y. We need to compare the x-coordinate of a line at a for a particular y, without loss of precision. The x-coordinate along an edge for a given y is: X = A_x + (Y - A_y) * A_dx / A_dy So the inequality we wish to test is: A_x + (Y - A_y) * A_dx / A_dy -?- B_x + (Y - B_y) * B_dx / B_dy, where -?- is our inequality operator. By construction we know that A_dy and B_dy (and (Y - A_y), (Y - B_y)) are all positive, so we can rearrange it thus without causing a sign change: A_dy * B_dy * (A_x - B_x) -?- (Y - B_y) * B_dx * A_dy - (Y - A_y) * A_dx * B_dy Given the assumption that all the deltas fit within 32 bits, we can compute this comparison directly using 128 bit arithmetic. src/cairo-bentley-ottmann.c | 97 +++++++++++++++++++++++++-------------------- src/cairo-wideint-private.h | 1 + 2 files changed, 56 insertions(+), 42 deletions(-) commit 7db03ac68cd556c903c07a2e2f8b75ec51263d12 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Oct 1 12:34:36 2008 +0100 [tessellator] Use abort() instead of exit(). More friendly when debugging, as the debug will (by default) catch the SIGTRAP and break at the offending test. src/cairo-bentley-ottmann.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) commit 59e569576d00e9c1cb66a77cf447c3cc3fb038e7 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Sep 28 19:04:39 2008 +0100 [traps] Discard trivially empty trapezoid. The convex_quad tessellator (and possibly even the more general polygon tessellator) will generate empty trapezoids when given a rectangle which can be trivially discarded before inserting into traps. src/cairo-traps.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) commit 7a2329e9c8afbfecb88c6c50bd63aa03ea7f9f81 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Oct 1 09:49:45 2008 +0100 [traps] Reset extents on clearing. When clearing the array of current trapezoids, reset the extents to infinite so that they are properly recomputed. src/cairo-traps.c | 2 ++ 1 file changed, 2 insertions(+) commit 8ec24a443d45b012df9b1a14b00a0b5b1c43e2ea Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Sep 30 23:19:01 2008 +0100 [xlib] Share the common conditions for starting a new xGlyphElt using a macro. Move the predicate for starting a new glyph elt into a macro so that it can be shared between _cairo_xlib_surface_emit_glyphs() and _emit_glyph_chunks() without code duplication. src/cairo-xlib-surface.c | 39 ++++++++++++++++++++++----------------- 1 file changed, 22 insertions(+), 17 deletions(-) commit a593338b2c2cdaff808947436f8fb9a52d74c695 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Oct 1 16:24:33 2008 +0100 [xcb] Return false from _cairo_xcb_surface_is_similar(). _cairo_xcb_surface_is_similar() is currently only used by the pattern cache to determine whether to keep the surface in the solid color cache. This is fundamentally broken without hooking into Display closure as it keeps a reference to an expired picture. So in order to prevent spurious application crashes, disable the caching for xcb. src/cairo-xcb-surface.c | 6 ++++++ 1 file changed, 6 insertions(+) commit 65f1575f6b2392d59410f40281a52654053bd2a8 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Oct 1 16:16:45 2008 +0100 [xcb] Compile fix. Must compile xcb more often. Especially after copying code over from xlib. src/cairo-xcb-surface.c | 41 ++++++++++++++++++++++++----------------- 1 file changed, 24 insertions(+), 17 deletions(-) commit e76432f7544d5b3b9157f09e18acc167d9ee90a3 Author: Behdad Esfahbod <behdad@behdad.org> Date: Sat Oct 4 00:35:43 2008 -0400 [win32] Use -MDd and -LDd instead of -MD and -LD for debug build As reported by Damian Frank: "I ran into a hitch with the Makefile.win32 infrastructure. It uses -MD and -LD when linking regardless of the config, but it should be using -MDd and -LDd for the debug config. I believe both the Makefile.win32.common and src/Makefile.win32 files include erroneous declarations. This produces warnings at link time about a mismatch when linking against properly created debug libs (for instance, I had a zlib built as "LIB ASM Debug" that linked properly against the debug runtime). This problem applies to pixman too; can you pass this along to the maintainer?" Would be delighted to if someone commits a fix to pixman reading this commit message. build/Makefile.win32.common | 8 ++++++-- src/Makefile.win32 | 2 +- 2 files changed, 7 insertions(+), 3 deletions(-) commit 8ea9cb187a198c64bfb79af8001d95eab29715b4 Author: Carl Worth <cworth@cworth.org> Date: Wed Oct 1 14:12:18 2008 -0700 Use symbolic constants for Render repeat modes. Names like RepeatNone and RepeatNormal are much easier to read and understand than bare 0 and 1. src/cairo-xlib-surface.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 2c58a2c3851afac0386fcf0bf8504a937231185c Author: Behdad Esfahbod <behdad@behdad.org> Date: Tue Sep 30 17:55:01 2008 -0400 [xlib] Start a new glyph element every 128 glyphs Xrender has limits at 252 glyphs. We play safe and fast and limit elements to 128 glyphs. That's plenty, no measurable performance hit expected. src/cairo-xlib-surface.c | 26 ++++++++++++++++++++++---- 1 file changed, 22 insertions(+), 4 deletions(-) commit fd7e09c7e66876b8492424e1c7d1260c12cc17f3 Author: Behdad Esfahbod <behdad@behdad.org> Date: Tue Sep 30 17:46:06 2008 -0400 [xlib] Allow room for glyph element padding Xrender pads glyph elements to 4byte boundaries. We didn't consider that in our request size calculations. We do now. src/cairo-xlib-surface.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) commit c01fb77abbaf28c03aa6a21ebb997638dbdf950b Author: Behdad Esfahbod <behdad@behdad.org> Date: Tue Sep 30 17:41:51 2008 -0400 [xlib] s/_cairo_xlib_surface_emit_glyphs_chunk/_emit_glyphs_chunk/ For readability's sake. src/cairo-xlib-surface.c | 34 +++++++++++++++++----------------- 1 file changed, 17 insertions(+), 17 deletions(-) commit c2ba25df1aec1cebfc4ce85e06a4187950675820 Author: Behdad Esfahbod <behdad@behdad.org> Date: Tue Sep 30 17:40:56 2008 -0400 [xlib] Add comment about glyph chunk invariant src/cairo-xlib-surface.c | 4 ++++ 1 file changed, 4 insertions(+) commit e983458e1fba15153815430c83619da53929139d Author: Behdad Esfahbod <behdad@behdad.org> Date: Tue Sep 30 17:26:46 2008 -0400 Revert "[xlib] Correct calculation of XRenderComposite* request size." This reverts commit 0eb0c26474a19477554bfd580aa5f8ae77c29779. The change was too drastic and overlooked some subleties of the old code, but the main reason for the revert is that it introduced an ugly duplicated glyph flush block. I'm working on a more incremental fix. src/cairo-xlib-surface.c | 123 +++++++++++++++++------------------------------ 1 file changed, 45 insertions(+), 78 deletions(-) commit 32a158528884562f4b22a81da3fa352ee1061835 Author: Behdad Esfahbod <behdad@behdad.org> Date: Tue Sep 30 15:20:36 2008 -0400 [configure.ac] Require autoconf >= 2.59 and automake >= 1.9.6 These are the versions available on RHEL5 (two years old now), and we know cairo works with them. There's evidence that our build system does not work with older automake, and we've been requiring autoconf 2.58 but no one ever tested 2.58 with the new build system. It's very likely that 2.58 doesn't work and needs some macro backporting. In any case, no one reported that they have 2.58 when I asked on the list. configure.ac | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) commit 0eb0c26474a19477554bfd580aa5f8ae77c29779 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Sep 30 13:33:25 2008 +0100 [xlib] Correct calculation of XRenderComposite* request size. show-glyphs-many is triggering an assertion failure within xlib. The cause appears to be that we are submitting an overlong request. Reviewing the code and comparing with libXrender/src/Glyph.c I found two points to address. 1. When encountering the first 2-byte, or 4-byte glyph and thus triggering the recalculation of the current request_size, we did not check that there was enough room for the expanded request. In case there is not, we need to emit the current request (before expansion) and reset. 2. Subtleties in how XRenderCompositeText* constructs the binary protocol buffer require that xGlyphElts are aligned to 32-bit boundaries and that it inserts an additional xGlyphElt every 252 glyphs when width==1 or every 254 glyphs for width==2 || width==4. Thus we need to explicitly compute how many bytes would be required to add this glyph in accordance with the above. Considering the complexity (and apparent fragility since we require tight coupling to XRender) of the code, I'm sure there are more bugs to be found. src/cairo-xlib-surface.c | 123 ++++++++++++++++++++++++++++++----------------- 1 file changed, 78 insertions(+), 45 deletions(-) commit 02a56a4c84cd07a2c33134974680bad7f17f733d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Sep 30 15:08:54 2008 +0100 [test/show-glyphs-many] Exercise xlib boundary conditions. Within _cairo_xlib_surface_emit_glyphs() there are a number of complications to do with packing as many glyphs as possible into a single XRenderCompositeGlyph*() call. Essentially these consist of choosing the right function and packing for the current glyphs, describing runs of glyphs and ensuring that we do not exceed the maximum request size within a single call. So we add to the test case we an attempt to show 64k 2-byte glyphs and an attempt to mix 64k 1-byte and 2-byte glyphs, with the change-over point chosen to overflow the maximum request size, should _cairo_xlib_surface_emit_glyphs() naively resize the current request. test/show-glyphs-many.c | 83 ++++++++++++++++++++++++++++++++++++++----------- 1 file changed, 64 insertions(+), 19 deletions(-) commit 2a347a92b0a27a42840f9538cb98f792be12b277 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Sep 30 13:29:35 2008 +0100 [test/show-glyphs-many] Re-enable test to trigger crash in xlib. Moral of this story is bugs cluster. If we made a mistake, especially in a complicated bit of code that is interfacing with another library, then we are likely to make a similar mistake in future. Disabling this test hid a regression between 1.4 and 1.6. test/Makefile.am | 15 +-------------- test/show-glyphs-many.c | 9 +++++++++ 2 files changed, 10 insertions(+), 14 deletions(-) commit fade54e71a48919cc3e8311e202960c66eab74bf Author: Torsten Schönfeld <kaffeetisch@gmx.de> Date: Tue Sep 30 11:47:53 2008 +0100 [doc] Add links to solid pattern constructors. Add links for the solid pattern constructors to the language bindings guidelines in line with the other pattern types. doc/public/language-bindings.xml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit d9f50d7841fa0216df157ba632483c8b27c50119 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Sep 30 11:03:29 2008 +0100 [test/filter-nearest-offset] Remove XFAIL status I swear that when I said that it was a rounding error, I was looking at an image where the squares were overlapping the lines and had a listing of all the coordinates used. However, the current output on all the machines I have to hand is correct so I believe the underlying bug to be fixed. Update the reference images for the external renderers because (a) GS exhibits the same bug cairo had and (b) librvsg/PDF do not use NEAREST when applying surface patterns, so the squares are blurred as a result. test/Makefile.am | 8 +++++--- test/filter-nearest-offset-pdf-ref.png | Bin 0 -> 4537 bytes test/filter-nearest-offset-ps2-ref.png | Bin 0 -> 255 bytes test/filter-nearest-offset-ps3-ref.png | Bin 0 -> 255 bytes test/filter-nearest-offset-svg11-ref.png | Bin 0 -> 4739 bytes test/filter-nearest-offset-svg12-ref.png | Bin 0 -> 4739 bytes test/filter-nearest-offset.c | 3 +-- 7 files changed, 6 insertions(+), 5 deletions(-) commit 7d5e6a18d5e72366e19cb806cbdad5ed180b53da Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Sep 30 10:04:12 2008 +0100 [test/filter-nearest-offset] Paint a grey background Avoid requiring (an absent!) rgb24 ref image by using an opaque grey background. Confirm the reference image is identical to the old one (modulo the background change) using GIMP. test/filter-nearest-offset-ref.png | Bin 254 -> 248 bytes test/filter-nearest-offset.c | 6 ++++++ 2 files changed, 6 insertions(+) commit 35be08715595aa0bda48ce0b3cb11711eee55c64 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Sep 30 10:54:45 2008 +0100 [Makefile.am] Only rerun headers-standalone if a header file has changed Use a stamp to mark the time that the headers were last checked and only rerun the test if any have been modified since. src/.gitignore | 1 + src/Makefile.am.analysis | 4 +++- 2 files changed, 4 insertions(+), 1 deletion(-) commit 6da8d8ea62399d94281a2a262a630ef113062d0c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Sep 29 17:56:34 2008 +0100 [pattern] Fix up OBO introduced to cloned area In the midst of porting 5eec3e378afd6ff9991cea8e42b8478eb3e79773 I failed to include the pad in the floor() and ceil() which introduces two potential off-by-one errors into each dimension of the region of interest of the source surface. src/cairo-pattern.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) commit e3f9e40df74b0334c2b544790fd3485ba527bd55 Author: Behdad Esfahbod <behdad@behdad.org> Date: Mon Sep 29 11:49:04 2008 -0400 Fix residues of the Makefile.*.config Makefile.*.features renaming They just don't seem to want to die... boilerplate/Makefile.win32 | 2 +- src/Makefile.sources | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) commit d703f964aa7d1d4c8ec0b8a5dee437c222ce8629 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Sep 29 16:37:35 2008 +0100 Clean up compile warnings. A couple of unchecked status returns (which should always be CAIRO_STATUS_SUCCESS) and an unused variable. src/cairo-surface.c | 8 ++++++-- src/cairo-svg-surface.c | 1 - 2 files changed, 6 insertions(+), 3 deletions(-) commit 1310df44edcf0b9fee648c8f8688d4a1177cde68 Author: Behdad Esfahbod <behdad@behdad.org> Date: Mon Sep 29 11:34:19 2008 -0400 [text_to_glyphs] Further enhance the docs The most complex public cairo API deserves more doc love... src/cairo-scaled-font.c | 8 +++++--- src/cairo-user-font.c | 5 +++++ src/cairo.h | 2 +- 3 files changed, 11 insertions(+), 4 deletions(-) commit 41c66a9dfa9f4e7b9839db9dc362f5fa48797561 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Sep 29 15:28:28 2008 +0100 [test/xlib-surface-source] Fix ill-timed XCloseDisplay Sigh - finding a crasher in xlib seemed too good to be true. And it was. The bug is due to the coupling of the XCloseDisplay with the user_data on the source surface. This was an externally trigger XCloseDisplay whilst the user had live surfaces, so I consider this a gross application bug and therefore does not require graceful handling within cairo-xlib. However, I'm willing to listen to reason... test/surface-source.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) commit 63ee963c63a73cf6d9272e15a7d8f7b63f12e77b Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Sep 29 15:09:23 2008 +0100 [user-font] Rewrite text_to_glyphs allocation instructions. Resolve the seemingly conflicting two paragraphs that instruct the writer of a user-font on when to allocate the glyphs array during text_to_glyphs, so that it is consistent with itself and the code. In particular, I could find no indication in the code that num_glyphs is preset to -1 (it should be a user parameter reflecting the number of entries in the supplied glyph buffer) and the callback must allocate the array using cairo_glyph_allocate(). src/cairo.h | 33 ++++++++++++++------------------- 1 file changed, 14 insertions(+), 19 deletions(-) commit c23dbc4c616aed05dfe71901ce7ac0cadcbfb13d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Sep 29 13:54:12 2008 +0100 [test/surface-source] Modify to trigger a crash. Whilst investigating: Bug 7360 painting huge surfaces fails https://bugs.freedesktop.org/show_bug.cgi?id=7360 I found a particular combination of operations that cause a crash within xlib, so I'm committing the test for posterity. test/surface-source.c | 50 ++++++++++++++++++++++++++++++++++++++------------ 1 file changed, 38 insertions(+), 12 deletions(-) commit 3b33d49d37a5751e7848516c468b323e19c34bea Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Sep 29 08:14:24 2008 +0100 [test/filter-bilinear-extents] Remove XFAIL. With the reintroduction of Carl Worth's and Owen Taylor's work to expand pattern extents as necessary to include extra samples due to the filter radius, we expect filter-bilinear-extents to PASS. This patch series is important as it factors out our assumptions about filter radius into a single function and cleans up the code in its wake. However, since the external PS/PDF renderers do not necessarily use the same filter as cairo (and currently they only use NEAREST due to lack of /Interpolate emission in those backends) we can expect differences in test output. So add the respective reference images to capture current expectations and to highlight future changes. Fixes bugs: Bug 15349 - bad clipping with EXTEND_NONE [https://bugs.freedesktop.org/show_bug.cgi?id=15349], Bug 15367 -Improve filtering handling in cairo-pattern.c [https://bugs.freedesktop.org/show_bug.cgi?id=15367] test/Makefile.am | 35 ++++++++++++++++++++++++++++++- test/filter-bilinear-extents-pdf-ref.png | Bin 0 -> 798 bytes test/filter-bilinear-extents-ps2-ref.png | Bin 0 -> 173 bytes test/filter-bilinear-extents-ps3-ref.png | Bin 0 -> 173 bytes 4 files changed, 34 insertions(+), 1 deletion(-) commit 5eec3e378afd6ff9991cea8e42b8478eb3e79773 Author: Owen W. Taylor <otaylor@fishsoup.net> Date: Mon Sep 29 00:53:59 2008 +0100 [pattern] Improve handling of filter radius and optimize when possible. Factor out common filter analysis code from _cairo_pattern_get_extents() so that we can share it with _cairo_pattern_acquire_surface_for_surface() as well. During the analysis of the filter determine whether the pattern matrix maps source pixels exactly onto destination pixels and if so convert the filter to NEAREST - generalising the existing conversion to NEAREST. (Patch ported to master by Chris Wilson, all bugs are his.) src/cairo-matrix.c | 29 ++++++++++++++ src/cairo-pattern.c | 113 ++++++++++++++++++++++++++++++++++++---------------- src/cairoint.h | 3 ++ 3 files changed, 110 insertions(+), 35 deletions(-) commit 7d7057cf8896a3558fadc35d160d3bc667972800 Author: Carl Worth <cworth@cworth.org> Date: Fri Apr 4 19:00:28 2008 -0700 _cairo_pattern_get_extents: Fix to allow for expansion based on filter This fixes the filter-bilinear-extents test case and the related bug entry: bad clipping with EXTEND_NONE http://bugs.freedesktop.org/show_bug.cgi?id=15349 Though there are still differences in the PDF and PostScript backends, (primarily because we can't capture cairo's filter modes in those file formats). src/cairo-pattern.c | 22 +++++++++++++++++++++ test/paint-source-alpha-svg11-ref.png | Bin 505 -> 756 bytes ...scale-source-surface-paint-svg11-argb32-ref.png | Bin 182 -> 229 bytes .../scale-source-surface-paint-svg11-rgb24-ref.png | Bin 192 -> 243 bytes ...scale-source-surface-paint-svg12-argb32-ref.png | Bin 182 -> 229 bytes .../scale-source-surface-paint-svg12-rgb24-ref.png | Bin 192 -> 243 bytes 6 files changed, 22 insertions(+) commit 6ef4bd4ecfa30f00ad0fe6e5e5f1d1a0a5dfe8e3 Author: Owen W. Taylor <otaylor@fishsoup.net> Date: Sun Sep 28 23:30:22 2008 +0100 [pattern] Fix acquiring outside of source bounds Restrict the area we acquire from the source image to the bounds of the source image, even when we have an identity matrix. This handles circumstances where the pattern extents may be enlarged due to filtering, for example when applying a bilinear filter. This patch replaces the open-coded rectangle intersection which has already proven itself to be a source of bugs. (Patch ported to master by Chris Wilson, all bugs are his.) src/cairo-pattern.c | 63 +++++++++++++++++++++++++++-------------------------- 1 file changed, 32 insertions(+), 31 deletions(-) commit 6d0f2c4421e34a078469a81646a52dfa47b4dcf1 Author: Owen W. Taylor <otaylor@fishsoup.net> Date: Mon Sep 29 00:02:18 2008 +0100 [pattern] Remove the hack to support cloning surface patterns. Remove instances (abuses) of calling _cairo_pattern_acquire_surface() on a known-surface-pattern with a hack to say "give me the entire surface". If you know you just want the entire surface as an image surface, that can be done more simply. (Split from original patch on https://bugs.freedesktop.org/attachment.cgi?id=15703, by Chris Wilson - as usual all bugs are his.) src/cairo-pattern.c | 5 +---- src/cairo-svg-surface.c | 16 ++------------- src/cairo-win32-printing-surface.c | 41 ++++++++++++++++++-------------------- 3 files changed, 22 insertions(+), 40 deletions(-) commit 117f643e49615295eb37af24efffd8327429cbf9 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Sep 28 22:42:30 2008 +0100 [test/filter-bilinear-extents] Extend testing. First, explicitly set the filter mode to BILINEAR in case the default should ever change. And then draw a second pattern with extents that in theory are larger than the source surface in order to test handling of acquiring out-of-bounds extents. test/filter-bilinear-extents.c | 45 +++++++++++++++++++++++++++++------------- 1 file changed, 31 insertions(+), 14 deletions(-) commit 3ff86a4510b718a3154b56b052be786f9495a3b0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Sep 29 08:38:31 2008 +0100 [test] Disable caching of SVG fail/pass surfaces. As we use cairo to convert SVG files back to an image, that process is dependent upon changes within our library and so we cannot skip the conversion if the SVG file happens to match a previous run. Fortunately, librsvg is quick enough that this is not a major issue. boilerplate/cairo-boilerplate.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) commit 24b22f0d2161f7a4ce75ceb2a86a77737027ee56 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Mon Sep 29 07:52:09 2008 +0100 [glitz] Compile fix for typo. I wonder whether merely fixing a compile error in glitz will do anything towards its utility... Anyway fix the missing comma as reported by Shunichi Fuji and correct the asserts. src/cairo-glitz-surface.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) commit 4f449107afba8ca96cd16e8dec283db124105dc8 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Sep 28 21:05:36 2008 +0100 [test/fallback-resolution] Exercise a couple of outstanding bugs. Use dashes to exercise bugs: https://bugs.freedesktop.org/show_bug.cgi?id=9189 https://bugs.freedesktop.org/show_bug.cgi?id=17223 Note bug 17223 indicates that this is still relevant for win32 printing where fallbacks are used if the dash offset is non-zero. And use a pattern to exercise the (fixed) regression: https://bugs.launchpad.net/inkscape/+bug/234546 test/fallback-resolution-ppi150-ref.png | Bin 2397 -> 8043 bytes test/fallback-resolution-ppi300-ref.png | Bin 1591 -> 5637 bytes test/fallback-resolution-ppi37.5-ref.png | Bin 4681 -> 14098 bytes test/fallback-resolution-ppi600-ref.png | Bin 1121 -> 4104 bytes test/fallback-resolution-ppi72-ref.png | Bin 2497 -> 8409 bytes test/fallback-resolution-ppi75-ref.png | Bin 3575 -> 12101 bytes test/fallback-resolution.c | 40 +++++++++++++++++++++++++++++-- 7 files changed, 38 insertions(+), 2 deletions(-) commit 0a7d781ab020171e70bd1dc66ff1556106e5be3f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Sep 28 18:57:15 2008 +0100 [perf] Add rounded rectangle perf case. Add the performance test case to compare the speed of filling a rounded rectangle (one with camphered corners) as opposed to an ordinary rectangle. Since the majority of the pixels are identical, ideally the two cases would take similar times (modulo the additional overhead in the more complex path). perf/Makefile.am | 1 + perf/cairo-perf.c | 1 + perf/cairo-perf.h | 1 + perf/rectangles.c | 22 ++++++++- perf/rounded-rectangles.c | 111 ++++++++++++++++++++++++++++++++++++++++++++++ 5 files changed, 135 insertions(+), 1 deletion(-) commit d52b55cb1b2a4e9b5fd8dd12592a227f8f6a93fc Author: Karl Tomlinson <karlt+@karlt.net> Date: Sun Sep 28 17:51:23 2008 +0100 [xlib] _surfaces_compatible() can return true for different picture formats _surfaces_compatible in cairo-xlib-surface returns true for surfaces with different xrender_format when each has the same depth and no (NULL) visual. Common picture formats will not have the same depth, but it is possible to create a surface with a non-standard xrender_format having the same depth as another xrender_format with cairo_xlib_surface_create_with_xrender_format. Both cairo_xlib_surface_create_with_xrender_format and _cairo_xlib_surface_create_similar_with_format create surfaces with no visual. The same issue exists in the xcb backend. Fixes bug https://bugs.freedesktop.org/show_bug.cgi?id=16564. AUTHORS | 1 + src/cairo-xcb-surface.c | 4 +++- src/cairo-xlib-surface.c | 4 +++- 3 files changed, 7 insertions(+), 2 deletions(-) commit b6b54e2f2736bfa9f76b7b6fc0da5d35a71fe846 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Sep 28 17:13:11 2008 +0100 [svg] Compile fix for last commit. Sorry - badly merged patch (missed a chunk). src/cairo-svg-surface.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) commit cff3bcba68233101c5518801b64e37d2f0e70ae8 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Sep 28 17:06:32 2008 +0100 [svg] Check that the mask is supported. During the analyze phase we also need to check whether we can emit the mask without requiring fallbacks. src/cairo-svg-surface.c | 17 +++++++++++++++-- 1 file changed, 15 insertions(+), 2 deletions(-) commit 8ad3031359d74bdff4b1e3ef8d0af047821f1795 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Sep 28 17:04:15 2008 +0100 [svg] Enable the assertion that paint() is supported. The XXX comment before the disabled assert is misleading as the final fallback image paint operation should never trigger UNSUPPORTED. src/cairo-svg-surface.c | 13 ++++--------- 1 file changed, 4 insertions(+), 9 deletions(-) commit 95e637f9ac9e4256f7ba73e8f048f8e559631c82 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Sep 28 16:54:59 2008 +0100 [paginated] Set paginated mode before every replay. Previously we were replaying full page fallbacks without updating the mode - this was upsetting SVG. src/cairo-paginated-surface.c | 23 ++++++++++++++--------- 1 file changed, 14 insertions(+), 9 deletions(-) commit 36b98589911ab57762317ec95cbff41681c1e0b6 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Sep 28 16:24:40 2008 +0100 [svg] Kill _cairo_svg_surface_force_fallbacks. It is not used, but refuses to die! src/cairo-svg-surface.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) commit 57851464f920d52444d19d25e78293fa0f40c5f9 Author: Benjamin Otte <otte@gnome.org> Date: Sun Sep 28 17:02:23 2008 +0200 [SVG] make backend handle new operators gracefully Previously, the SVG backend would rash when new operators were added to cairo.h, now it returns UNSUPPORTED. Also unsupported operators can now be set as NULL, so image fallbacks can be used properly. Should use fallbacks instead of color-dodge for CAIRO_OPERATOR_STURATE? src/cairo-svg-surface.c | 45 +++++++++++++++++++++++++++++++-------------- 1 file changed, 31 insertions(+), 14 deletions(-) commit fda80cc852a109fe60e3277a1847bc5c24f35d8e Author: Benjamin Otte <otte@gnome.org> Date: Sun Sep 28 14:57:49 2008 +0200 [SVG] use _cairo_operator_bounded_by_source() previously, the code was duplicating its functionality src/cairo-svg-surface.c | 27 ++++++++++++--------------- 1 file changed, 12 insertions(+), 15 deletions(-) commit 3c18d95b293eddaa25232bc06398c78bc8043cb4 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Sep 28 15:57:37 2008 +0100 Compute device_transform_inverse. Because we can independently set either the device scale or the device offset, we need to be careful and recompute the inverse rather than simply assuming that the original contents of the device transform is identity. Fixes regression https://bugs.launchpad.net/inkscape/+bug/234546. src/cairo-surface.c | 10 ++++------ 1 file changed, 4 insertions(+), 6 deletions(-) commit 9841d9d58ea286f798626d325d50a85bf3f02c8f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sun Sep 28 13:34:50 2008 +0100 Automate error checking for fallback-resolution. For this we extend the boilerplate get_image() routines to extract a single page out of a paginated document and then proceed to manually check each page of the fallback-resolution test. (Well that's the theory, in practice SVG doesn't support multiple pages and so we just generate a new surface for each resolution. But the infrastructure is in place so that we can automate other tests, e.g. test/multi-pages.) boilerplate/cairo-boilerplate-pdf-private.h | 5 + boilerplate/cairo-boilerplate-pdf.c | 29 +- boilerplate/cairo-boilerplate-pdf.h | 33 -- boilerplate/cairo-boilerplate-ps-private.h | 9 +- boilerplate/cairo-boilerplate-ps.c | 33 +- boilerplate/cairo-boilerplate-ps.h | 33 -- boilerplate/cairo-boilerplate-svg-private.h | 5 + boilerplate/cairo-boilerplate-svg.c | 36 +- boilerplate/cairo-boilerplate-svg.h | 33 -- .../cairo-boilerplate-test-surfaces-private.h | 1 + boilerplate/cairo-boilerplate-test-surfaces.c | 7 +- boilerplate/cairo-boilerplate-win32-printing.c | 5 + boilerplate/cairo-boilerplate-win32-private.h | 1 + boilerplate/cairo-boilerplate.c | 74 ++-- boilerplate/cairo-boilerplate.h | 7 + src/cairo-svg-surface.c | 6 +- test/Makefile.am | 6 + test/cairo-test.c | 36 +- test/cairo-test.h | 12 + test/fallback-resolution-ppi150-ref.png | Bin 0 -> 2397 bytes test/fallback-resolution-ppi300-ref.png | Bin 0 -> 1591 bytes test/fallback-resolution-ppi37.5-ref.png | Bin 0 -> 4681 bytes test/fallback-resolution-ppi600-ref.png | Bin 0 -> 1121 bytes test/fallback-resolution-ppi72-ref.png | Bin 0 -> 2497 bytes test/fallback-resolution-ppi75-ref.png | Bin 0 -> 3575 bytes test/fallback-resolution.c | 413 +++++++++++++++------ 26 files changed, 486 insertions(+), 298 deletions(-) commit cca1fc6358e9d0213dd2b41a5bfd1629eec6511e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Sep 27 21:53:21 2008 +0100 [test] Fix target specific tests for recent SVG/PS version split. A few tests explicitly checked whether the "ps" or "svg" target was enabled and this broke because of the name change. So fixup, to run the generic test if either PS or SVG target is enabled as appropriate. test/create-for-stream.c | 8 ++++++-- test/fallback-resolution.c | 8 ++++++-- test/multi-page.c | 4 +++- test/ps-features.c | 4 +++- test/svg-clip.c | 4 +++- test/svg-surface.c | 4 +++- 6 files changed, 24 insertions(+), 8 deletions(-) commit 39e4a0bcd04191eecfc593653a4ba32a28fcac6a Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Sep 27 19:33:51 2008 +0100 [test] Update list of REFERENCE_IMAGES A few typos, deletions and additions from the inclusion of versioned SVG/PS testing. test/Makefile.am | 50 +++++++++++++++++++++++++++++++------------------- 1 file changed, 31 insertions(+), 19 deletions(-) commit 260362e5728e2591bb50f565ef04ba951d95de79 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Sep 27 19:16:34 2008 +0100 [cairo-test] Fix cairo_test_file_is_older(). It did not even compile, so no longer it wasn't deleting cached results if you updated the reference images. test/cairo-test.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) commit ffbb67303c1a4d986096ba046bb3256314b9f1b1 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Sep 27 18:54:45 2008 +0100 [test/gradient-constant-alpha] Remove ps2 reference images. PostScript level 2 has no support for gradients and so should uses image fallbacks and hence do not require the GS specific reference images. test/Makefile.am | 2 -- test/gradient-constant-alpha-ps2-ref.png | Bin 124 -> 0 bytes test/gradient-constant-alpha-ps2-rgb24-ref.png | Bin 124 -> 0 bytes 3 files changed, 2 deletions(-) commit 4fa9f8c0608c137d9aa277039cb286c8294e6a2f Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Sep 27 17:42:06 2008 +0100 [test/large-source] Remove stale comment. Delete a reference to an old limitation of pixman. test/large-source.c | 3 --- 1 file changed, 3 deletions(-) commit 35f1e2fa4f15c562e5cc41578b413c6874e131bc Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Sep 27 17:37:32 2008 +0100 [test/operator*] Remove XFAIL descriptions. The tests should now pass, so remove the line claiming that they fail for unknown reasons. test/operator-alpha.c | 3 +-- test/operator.c | 3 +-- 2 files changed, 2 insertions(+), 4 deletions(-) commit 871f683367b1a1647b607f9221e1fc2eddc6e5dd Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Sep 27 17:24:57 2008 +0100 clone_similar(): s/device_offset/clone_offset/ A little bit of sleep and reflection suggested that the use of device_offset_[xy] was confusing and clone_offset_[xy] more consistent with the function naming. src/cairo-directfb-surface.c | 12 ++++++------ src/cairo-glitz-surface.c | 18 +++++++++--------- src/cairo-image-surface.c | 6 +++--- src/cairo-pattern.c | 6 +++--- src/cairo-quartz-surface.c | 16 ++++++++-------- src/cairo-surface-fallback-private.h | 4 ++-- src/cairo-surface-fallback.c | 8 ++++---- src/cairo-surface.c | 16 ++++++++-------- src/cairo-win32-private.h | 4 ++-- src/cairo-win32-surface.c | 8 ++++---- src/cairo-xcb-surface.c | 12 ++++++------ src/cairo-xlib-surface.c | 12 ++++++------ src/cairoint.h | 10 +++++----- src/test-fallback-surface.c | 8 ++++---- 14 files changed, 70 insertions(+), 70 deletions(-) commit 95c3c069326a77a28eb49053c0ace9f1977e0378 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Sep 27 17:13:32 2008 +0100 [surface] Only return early for no traps if the operator is bounded by a mask. If the operator is unbounded, then its area of effect extends beyond the definition of the mask by the trapezoids and so we must always perform the image composition. Fixes test/operator*. src/cairo-surface-fallback.c | 4 +--- test/Makefile.am | 2 -- 2 files changed, 1 insertion(+), 5 deletions(-) commit 922a5801164c2b2340a7b4a950820b07e72adadc Merge: 0c4317085 1e45da48b Author: Emmanuel Pacaud <emmanuel.pacaud@free.fr> Date: Sat Sep 27 11:55:50 2008 +0200 Merge branch 'master' of git+ssh://emmanuel@git.cairographics.org/git/cairo commit 0c4317085ecfc112214cdfe2abd66d9df8038282 Author: Emmanuel Pacaud <emmanuel.pacaud@free.fr> Date: Sat Sep 27 11:47:32 2008 +0200 [svg] Fix comp-op for paint, mask and show_glyphs operations. In paint and show_glyphs, the compositing operator was not emitted at all. In mask, the operator was also emitted for the mask itself, which is wrong. SVG clear and source differ from cairo as it also affects the destination if the source pixel are completely transparent. We need to emit an additional clip-to-self property. Librsvg does not support clip-to-self, so it renders the SVG test outputs incorrectly. This patch also remove a lot of useless spaces in the style property strings (I know, this should go in another commit). src/cairo-svg-surface.c | 196 ++++++++++++++++++++++++++++-------------------- 1 file changed, 113 insertions(+), 83 deletions(-) commit 1e45da48b555b707225c1521068011cabbcc976f Author: Adrian Johnson <ajohnson@redneon.com> Date: Sat Sep 27 14:38:22 2008 +0930 Update ref images in Makefile.am test/Makefile.am | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) commit c0535e8e3394f0b78867785e65cf89dbb31875d5 Author: Adrian Johnson <ajohnson@redneon.com> Date: Sat Sep 27 14:31:50 2008 +0930 Remove miter-precision from XFAIL The miter-precision PS output prints correctly on my LaserJet 4050. So file a Ghostscript bug, supply PS ref images and remove from XFAIL. test/Makefile.am | 1 - test/README | 5 +++++ test/miter-precision-ps2-ref.png | Bin 0 -> 865 bytes test/miter-precision-ps3-ref.png | Bin 0 -> 865 bytes test/miter-precision.c | 2 +- 5 files changed, 6 insertions(+), 2 deletions(-) commit 5b97ee65259cafb335c1a2c53f1a55dfcb175e20 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Sep 27 00:18:07 2008 +0100 Allow cloning sub-regions of similar surfaces. Previously the rule for clone_similar() was that the returned surface had exactly the same size as the original, but only the contents within the region of interest needed to be copied. This caused failures for very large images in the xlib-backend (see test/large-source). The obvious solution to allow cloning only the region of interest seemed to be to simply set the device offset on the cloned surface. However, this fails as a) nothing respects the device offset on the surface at that layer in the compositing stack and b) possibly returning references to the original source surface provides further confusion by mixing in another source of device offset. The second method was to add extra out parameters so that the device offset could be returned separately and, for example, mixed into the pattern matrix. Not as elegant, a couple of extra warts to the interface, but it works - one less XFAIL... src/cairo-clip.c | 6 ++++++ src/cairo-directfb-surface.c | 14 +++++++++----- src/cairo-glitz-surface.c | 10 ++++++++++ src/cairo-image-surface.c | 3 +++ src/cairo-pattern.c | 15 +++++++++++++-- src/cairo-quartz-surface.c | 10 +++++++++- src/cairo-surface-fallback-private.h | 2 ++ src/cairo-surface-fallback.c | 8 ++++++-- src/cairo-surface.c | 24 ++++++++++++++++++------ src/cairo-win32-private.h | 2 ++ src/cairo-win32-surface.c | 8 ++++++-- src/cairo-xcb-surface.c | 18 ++++++++++++++---- src/cairo-xlib-surface.c | 21 +++++++++++++++------ src/cairoint.h | 11 +++++++++-- src/test-fallback-surface.c | 4 ++++ test/Makefile.am | 1 - 16 files changed, 126 insertions(+), 31 deletions(-) commit 7f3a48f90b409653a2d4fb802779ecb7508e4d6e Author: Behdad Esfahbod <behdad@behdad.org> Date: Fri Sep 26 13:27:42 2008 -0400 [cairo.h] Don't define cairo_public to __declspec(dllimport) for static build That define should target win32 DLL builds only. We can't tell though, so we require user to define CAIRO_WIN32_STATIC_BUILD to signal that. src/cairo-system.c | 2 +- src/cairo.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) commit b6e910f4b636e947cefb310125979ce742e87730 Author: Behdad Esfahbod <behdad@behdad.org> Date: Fri Sep 26 12:36:33 2008 -0400 [README.win32] Update wording from Tor Lillqvist README.win32 | 19 ++++++++++++------- 1 file changed, 12 insertions(+), 7 deletions(-) commit fc12267405e98497c9f086d49fa6501d8f671a68 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Sep 26 17:10:52 2008 +0100 [scaled-font] Spelling fix. Arbitrary spelling fix. src/cairo-scaled-font.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit e0f214710760a0e0332207cb3009e3119867bd13 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Sep 26 16:49:57 2008 +0100 [boilerplate] Add a PS 2 target. Ensure that we can generate PS2 output when required by creating a PS level 2 target and using GhostScript for conformance testing. boilerplate/cairo-boilerplate-ps-private.h | 29 ++- boilerplate/cairo-boilerplate-ps.c | 55 +++- boilerplate/cairo-boilerplate.c | 23 +- test/Makefile.am | 286 ++++++++++++++------- test/{big-line-ps-ref.png => big-line-ps2-ref.png} | Bin ...ps-rgb24-ref.png => big-line-ps2-rgb24-ref.png} | Bin test/big-line-ps3-ref.png | Bin 0 -> 1148 bytes test/big-line-ps3-rgb24-ref.png | Bin 0 -> 830 bytes ...aps-joins-ps-ref.png => caps-joins-ps2-ref.png} | Bin test/caps-joins-ps3-ref.png | Bin 0 -> 1459 bytes ...2-ref.png => clip-fill-rule-ps2-argb32-ref.png} | Bin ...24-ref.png => clip-fill-rule-ps2-rgb24-ref.png} | Bin test/clip-fill-rule-ps3-argb32-ref.png | Bin 0 -> 310 bytes test/clip-fill-rule-ps3-rgb24-ref.png | Bin 0 -> 302 bytes ...b32-ref.png => clip-nesting-ps2-argb32-ref.png} | Bin ...gb24-ref.png => clip-nesting-ps2-rgb24-ref.png} | Bin test/clip-nesting-ps3-argb32-ref.png | Bin 0 -> 651 bytes test/clip-nesting-ps3-rgb24-ref.png | Bin 0 -> 636 bytes ...b24-ref.png => clip-operator-ps2-rgb24-ref.png} | Bin test/clip-operator-ps3-ref.png | Bin 0 -> 8188 bytes test/clip-operator-ps3-rgb24-ref.png | Bin 0 -> 3624 bytes ...-ref.png => clip-push-group-ps2-argb32-ref.png} | Bin ...4-ref.png => clip-push-group-ps2-rgb24-ref.png} | Bin test/clip-push-group-ps3-argb32-ref.png | Bin 0 -> 179 bytes test/clip-push-group-ps3-rgb24-ref.png | Bin 0 -> 179 bytes ...rgb32-ref.png => clip-twice-ps2-argb32-ref.png} | Bin ...-rgb24-ref.png => clip-twice-ps2-rgb24-ref.png} | Bin test/clip-twice-ps3-argb32-ref.png | Bin 0 -> 575 bytes test/clip-twice-ps3-rgb24-ref.png | Bin 0 -> 541 bytes ...-group-ps-ref.png => clipped-group-ps2-ref.png} | Bin test/clipped-group-ps3-ref.png | Bin 0 -> 343 bytes ...lose-path-ps-ref.png => close-path-ps2-ref.png} | Bin test/close-path-ps3-ref.png | Bin 0 -> 309 bytes ...{copy-path-ps-ref.png => copy-path-ps2-ref.png} | Bin test/copy-path-ps3-ref.png | Bin 0 -> 474 bytes ...-ref.png => dash-caps-joins-ps2-argb32-ref.png} | Bin ...4-ref.png => dash-caps-joins-ps2-rgb24-ref.png} | Bin test/dash-caps-joins-ps3-argb32-ref.png | Bin 0 -> 3870 bytes test/dash-caps-joins-ps3-rgb24-ref.png | Bin 0 -> 3870 bytes ...ash-curve-ps-ref.png => dash-curve-ps2-ref.png} | Bin test/dash-curve-ps3-ref.png | Bin 0 -> 27293 bytes ...rgb32-ref.png => dash-scale-ps2-argb32-ref.png} | Bin ...-rgb24-ref.png => dash-scale-ps2-rgb24-ref.png} | Bin test/dash-scale-ps3-argb32-ref.png | Bin 0 -> 6318 bytes test/dash-scale-ps3-rgb24-ref.png | Bin 0 -> 6316 bytes ...ash-state-ps-ref.png => dash-state-ps2-ref.png} | Bin test/dash-state-ps3-ref.png | Bin 0 -> 10219 bytes ...gth-ps-ref.png => dash-zero-length-ps2-ref.png} | Bin ...-ref.png => dash-zero-length-ps2-rgb24-ref.png} | Bin test/dash-zero-length-ps3-ref.png | Bin 0 -> 319 bytes test/dash-zero-length-ps3-rgb24-ref.png | Bin 0 -> 304 bytes ...e-arc-ps-ref.png => degenerate-arc-ps2-ref.png} | Bin test/degenerate-arc-ps3-ref.png | Bin 0 -> 585 bytes ...-ref.png => degenerate-path-ps2-argb32-ref.png} | Bin ...4-ref.png => degenerate-path-ps2-rgb24-ref.png} | Bin test/degenerate-path-ps3-argb32-ref.png | Bin 0 -> 390 bytes test/degenerate-path-ps3-rgb24-ref.png | Bin 0 -> 358 bytes ...e-pen-ps-ref.png => degenerate-pen-ps2-ref.png} | Bin test/degenerate-pen-ps3-ref.png | Bin 0 -> 865 bytes ...ef.png => device-offset-fractional-ps2-ref.png} | Bin test/device-offset-fractional-ps3-ref.png | Bin 0 -> 201 bytes ...flect-ps-ref.png => extend-reflect-ps2-ref.png} | Bin ...milar-ps-ref.png => extend-reflect-ps3-ref.png} | Bin test/extend-reflect-similar-ps2-ref.png | Bin 0 -> 153322 bytes test/extend-reflect-similar-ps3-ref.png | Bin 0 -> 153322 bytes ...n-ps-ref.png => fill-alpha-pattern-ps3-ref.png} | Bin ...-ref.png => fill-and-stroke-ps2-argb32-ref.png} | Bin ...4-ref.png => fill-and-stroke-ps2-rgb24-ref.png} | Bin test/fill-and-stroke-ps3-argb32-ref.png | Bin 0 -> 310 bytes test/fill-and-stroke-ps3-rgb24-ref.png | Bin 0 -> 319 bytes ...ref.png => fill-missed-stop-ps2-argb32-ref.png} | Bin ...-ref.png => fill-missed-stop-ps2-rgb24-ref.png} | Bin test/fill-missed-stop-ps3-argb32-ref.png | Bin 0 -> 564 bytes test/fill-missed-stop-ps3-rgb24-ref.png | Bin 0 -> 531 bytes ...argb32-ref.png => fill-rule-ps2-argb32-ref.png} | Bin ...s-rgb24-ref.png => fill-rule-ps2-rgb24-ref.png} | Bin test/fill-rule-ps3-argb32-ref.png | Bin 0 -> 2253 bytes test/fill-rule-ps3-rgb24-ref.png | Bin 0 -> 2031 bytes ... => finer-grained-fallbacks-ps2-argb32-ref.png} | Bin ...g => finer-grained-fallbacks-ps2-rgb24-ref.png} | Bin test/finer-grained-fallbacks-ps3-argb32-ref.png | Bin 0 -> 1047 bytes test/finer-grained-fallbacks-ps3-rgb24-ref.png | Bin 0 -> 819 bytes ... => font-matrix-translation-ps2-argb32-ref.png} | Bin ...g => font-matrix-translation-ps2-rgb24-ref.png} | Bin test/font-matrix-translation-ps3-argb32-ref.png | Bin 0 -> 748 bytes test/font-matrix-translation-ps3-rgb24-ref.png | Bin 0 -> 748 bytes ....png => ft-show-glyphs-positioning-ps2-ref.png} | Bin test/ft-show-glyphs-positioning-ps3-ref.png | Bin 0 -> 2269 bytes ...ps-ref.png => ft-show-glyphs-table-ps2-ref.png} | Bin test/ft-show-glyphs-table-ps3-ref.png | Bin 0 -> 5687 bytes ...g => ft-text-antialias-none-ps2-argb32-ref.png} | Bin test/ft-text-antialias-none-ps3-argb32-ref.png | Bin 0 -> 335 bytes ...g => ft-text-vertical-layout-type1-ps2-ref.png} | Bin test/ft-text-vertical-layout-type1-ps3-ref.png | Bin 0 -> 1846 bytes ...g => ft-text-vertical-layout-type3-ps2-ref.png} | Bin test/ft-text-vertical-layout-type3-ps3-ref.png | Bin 0 -> 1804 bytes ...ps-ref.png => glyph-cache-pressure-ps2-ref.png} | Bin test/glyph-cache-pressure-ps3-ref.png | Bin 0 -> 1454 bytes ...2-ref.png => gradient-alpha-ps2-argb32-ref.png} | Bin ...24-ref.png => gradient-alpha-ps2-rgb24-ref.png} | Bin test/gradient-alpha-ps3-argb32-ref.png | Bin 0 -> 134 bytes test/gradient-alpha-ps3-rgb24-ref.png | Bin 0 -> 130 bytes ...ref.png => gradient-constant-alpha-ps2-ref.png} | Bin ...g => gradient-constant-alpha-ps2-rgb24-ref.png} | Bin test/gradient-constant-alpha-ps3-ref.png | Bin 0 -> 124 bytes test/gradient-constant-alpha-ps3-rgb24-ref.png | Bin 0 -> 124 bytes ...e-join-ps-ref.png => infinite-join-ps2-ref.png} | Bin test/infinite-join-ps3-ref.png | Bin 0 -> 218 bytes ...rgb32-ref.png => leaky-dash-ps2-argb32-ref.png} | Bin ...-rgb24-ref.png => leaky-dash-ps2-rgb24-ref.png} | Bin test/leaky-dash-ps3-argb32-ref.png | Bin 0 -> 286 bytes test/leaky-dash-ps3-rgb24-ref.png | Bin 0 -> 284 bytes ...-ref.png => leaky-dashed-rectangle-ps2-ref.png} | Bin test/leaky-dashed-rectangle-ps3-ref.png | Bin 0 -> 444 bytes ...olygon-ps-ref.png => leaky-polygon-ps2-ref.png} | Bin test/leaky-polygon-ps3-ref.png | Bin 0 -> 354 bytes ...ale-ps-ref.png => line-width-scale-ps2-ref.png} | Bin test/line-width-scale-ps3-ref.png | Bin 0 -> 3911 bytes ...ient-ps-ref.png => linear-gradient-ps3-ref.png} | Bin ...ref.png => linear-gradient-reflect-ps3-ref.png} | Bin ...es-ps-ref.png => long-dashed-lines-ps2-ref.png} | Bin test/long-dashed-lines-ps3-ref.png | Bin 0 -> 1516 bytes ...png => meta-surface-pattern-ps2-argb32-ref.png} | Bin ....png => meta-surface-pattern-ps2-rgb24-ref.png} | Bin test/meta-surface-pattern-ps3-argb32-ref.png | Bin 0 -> 3257 bytes test/meta-surface-pattern-ps3-rgb24-ref.png | Bin 0 -> 3873 bytes ...b32-ref.png => new-sub-path-ps2-argb32-ref.png} | Bin ...gb24-ref.png => new-sub-path-ps2-rgb24-ref.png} | Bin test/new-sub-path-ps3-argb32-ref.png | Bin 0 -> 556 bytes test/new-sub-path-ps3-rgb24-ref.png | Bin 0 -> 423 bytes ...ef.png => over-above-source-ps2-argb32-ref.png} | Bin test/over-above-source-ps3-argb32-ref.png | Bin 0 -> 636 bytes ...f.png => over-around-source-ps2-argb32-ref.png} | Bin ...ef.png => over-around-source-ps2-rgb24-ref.png} | Bin test/over-around-source-ps3-argb32-ref.png | Bin 0 -> 632 bytes test/over-around-source-ps3-rgb24-ref.png | Bin 0 -> 538 bytes ...ef.png => over-below-source-ps2-argb32-ref.png} | Bin test/over-below-source-ps3-argb32-ref.png | Bin 0 -> 503 bytes ....png => over-between-source-ps2-argb32-ref.png} | Bin test/over-between-source-ps3-argb32-ref.png | Bin 0 -> 678 bytes ...ps-ref.png => random-intersections-ps2-ref.png} | Bin test/random-intersections-ps3-ref.png | Bin 0 -> 89253 bytes ...ref.png => rectilinear-miter-limit-ps2-ref.png} | Bin test/rectilinear-miter-limit-ps3-ref.png | Bin 0 -> 221 bytes ...oke-ps-ref.png => reflected-stroke-ps2-ref.png} | Bin test/reflected-stroke-ps3-ref.png | Bin 0 -> 3898 bytes ...ps-rgb24-ref.png => rel-path-ps2-rgb24-ref.png} | Bin test/rel-path-ps3-rgb24-ref.png | Bin 0 -> 243 bytes ....png => rotate-image-surface-paint-ps2-ref.png} | Bin test/rotate-image-surface-paint-ps3-ref.png | Bin 0 -> 212 bytes ...ace-ps-ref.png => select-font-face-ps2-ref.png} | Bin test/select-font-face-ps3-ref.png | Bin 0 -> 959 bytes ...{self-copy-ps-ref.png => self-copy-ps2-ref.png} | Bin test/self-copy-ps3-ref.png | Bin 0 -> 335 bytes ...ref.png => show-text-current-point-ps2-ref.png} | Bin test/show-text-current-point-ps3-ref.png | Bin 0 -> 1356 bytes ...extreme-ps-ref.png => skew-extreme-ps2-ref.png} | Bin test/skew-extreme-ps3-ref.png | Bin 0 -> 1048 bytes test/{smask-ps-ref.png => smask-ps2-ref.png} | Bin test/smask-ps3-ref.png | Bin 0 -> 3669 bytes ...mask-text-ps-ref.png => smask-text-ps2-ref.png} | Bin test/smask-text-ps3-ref.png | Bin 0 -> 2023 bytes ...ef.png => source-clip-scale-ps2-argb32-ref.png} | Bin ...ref.png => source-clip-scale-ps2-rgb24-ref.png} | Bin test/source-clip-scale-ps3-argb32-ref.png | Bin 0 -> 183 bytes test/source-clip-scale-ps3-rgb24-ref.png | Bin 0 -> 183 bytes ...caps-ps-ref.png => stroke-ctm-caps-ps2-ref.png} | Bin test/stroke-ctm-caps-ps3-ref.png | Bin 0 -> 1079 bytes ...e-image-ps-ref.png => stroke-image-ps2-ref.png} | Bin test/stroke-image-ps3-ref.png | Bin 0 -> 1786 bytes ...-ref.png => surface-pattern-ps2-argb32-ref.png} | Bin test/surface-pattern-ps3-argb32-ref.png | Bin 0 -> 2520 bytes ... surface-pattern-scale-down-ps2-argb32-ref.png} | Bin test/surface-pattern-scale-down-ps3-argb32-ref.png | Bin 0 -> 1400 bytes ...=> surface-pattern-scale-up-ps2-argb32-ref.png} | Bin test/surface-pattern-scale-up-ps3-argb32-ref.png | Bin 0 -> 1011 bytes ...b32-ref.png => text-pattern-ps3-argb32-ref.png} | Bin ...gb24-ref.png => text-pattern-ps3-rgb24-ref.png} | Bin ...t-rotate-ps-ref.png => text-rotate-ps2-ref.png} | Bin test/text-rotate-ps3-ref.png | Bin 0 -> 6851 bytes ...sform-ps-ref.png => text-transform-ps2-ref.png} | Bin test/text-transform-ps3-ref.png | Bin 0 -> 4154 bytes ...ransforms-ps-ref.png => transforms-ps2-ref.png} | Bin test/transforms-ps3-ref.png | Bin 0 -> 418 bytes ...argb32-ref.png => trap-clip-ps3-argb32-ref.png} | Bin ...s-rgb24-ref.png => trap-clip-ps3-rgb24-ref.png} | Bin ...roxy-ps-ref.png => user-font-proxy-ps2-ref.png} | Bin test/user-font-proxy-ps3-ref.png | Bin 0 -> 8171 bytes ...{user-font-ps-ref.png => user-font-ps2-ref.png} | Bin test/user-font-ps3-ref.png | Bin 0 -> 5043 bytes ...le-ps-ref.png => user-font-rescale-ps2-ref.png} | Bin test/user-font-rescale-ps3-ref.png | Bin 0 -> 8041 bytes 192 files changed, 277 insertions(+), 116 deletions(-) commit 0c030d3b768718eac5d429d0c819934f32896b25 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Fri Sep 26 15:26:40 2008 +0100 [boilerplate] Add svg-1.2 target. SVG 1.2 is sufficiently different from 1.1 that it has separate code paths within cairo-svg-surface and so justifies its own boilerplate target. boilerplate/cairo-boilerplate-svg-private.h | 30 +++++++--- boilerplate/cairo-boilerplate-svg.c | 48 +++++++++++++-- boilerplate/cairo-boilerplate.c | 24 ++++++-- test/Makefile.am | 66 ++++++++++++++------- ...-svg-ref.png => caps-joins-alpha-svg11-ref.png} | Bin test/caps-joins-alpha-svg12-ref.png | Bin 0 -> 2454 bytes ...f.png => font-matrix-translation-svg11-ref.png} | Bin test/font-matrix-translation-svg12-ref.png | Bin 0 -> 970 bytes ...ng => ft-show-glyphs-positioning-svg11-ref.png} | Bin test/ft-show-glyphs-positioning-svg12-ref.png | Bin 0 -> 6590 bytes ...=> ft-text-vertical-layout-type1-svg11-ref.png} | Bin test/ft-text-vertical-layout-type1-svg12-ref.png | Bin 0 -> 3973 bytes ...=> ft-text-vertical-layout-type3-svg11-ref.png} | Bin test/ft-text-vertical-layout-type3-svg12-ref.png | Bin 0 -> 3957 bytes ...t-svg-ref.png => linear-gradient-svg11-ref.png} | Bin test/linear-gradient-svg12-ref.png | Bin 0 -> 1000 bytes ...b32-ref.png => mask-alpha-svg11-argb32-ref.png} | Bin test/mask-alpha-svg12-argb32-ref.png | Bin 0 -> 615 bytes ...rgb32-ref.png => mask-ctm-svg11-argb32-ref.png} | Bin ...rgb32-ref.png => mask-ctm-svg12-argb32-ref.png} | Bin test/mask-surface-ctm-svg11-argb32-ref.png | Bin 0 -> 108 bytes test/mask-surface-ctm-svg12-argb32-ref.png | Bin 0 -> 108 bytes ...vg-argb32-ref.png => mask-svg11-argb32-ref.png} | Bin ...-svg-rgb24-ref.png => mask-svg11-rgb24-ref.png} | Bin test/mask-svg12-argb32-ref.png | Bin 0 -> 8527 bytes test/mask-svg12-rgb24-ref.png | Bin 0 -> 7935 bytes ....png => mask-transformed-similar-svg11-ref.png} | Bin test/mask-transformed-similar-svg12-ref.png | Bin 0 -> 4051 bytes ...g => meta-surface-pattern-svg11-argb32-ref.png} | Bin ...ng => meta-surface-pattern-svg11-rgb24-ref.png} | Bin test/meta-surface-pattern-svg12-argb32-ref.png | Bin 0 -> 3924 bytes test/meta-surface-pattern-svg12-rgb24-ref.png | Bin 0 -> 4593 bytes ...vg-ref.png => paint-source-alpha-svg11-ref.png} | Bin test/paint-source-alpha-svg12-ref.png | Bin 0 -> 505 bytes ...-svg-ref.png => paint-with-alpha-svg11-ref.png} | Bin test/paint-with-alpha-svg12-ref.png | Bin 0 -> 516 bytes ...b32-ref.png => push-group-svg11-argb32-ref.png} | Bin test/push-group-svg12-argb32-ref.png | Bin 0 -> 2935 bytes ...t-svg-ref.png => radial-gradient-svg11-ref.png} | Bin test/radial-gradient-svg12-ref.png | Bin 0 -> 91039 bytes ...ng => rotate-image-surface-paint-svg11-ref.png} | Bin test/rotate-image-surface-paint-svg12-ref.png | Bin 0 -> 248 bytes ...cale-source-surface-paint-svg11-argb32-ref.png} | Bin ...scale-source-surface-paint-svg11-rgb24-ref.png} | Bin ...scale-source-surface-paint-svg12-argb32-ref.png | Bin 0 -> 182 bytes .../scale-source-surface-paint-svg12-rgb24-ref.png | Bin 0 -> 192 bytes ...b32-ref.png => set-source-svg11-argb32-ref.png} | Bin test/set-source-svg12-argb32-ref.png | Bin 0 -> 104 bytes ...k-fill-svg-ref.png => smask-fill-svg11-ref.png} | Bin test/smask-fill-svg12-ref.png | Bin 0 -> 1166 bytes ...k-mask-svg-ref.png => smask-mask-svg11-ref.png} | Bin test/smask-mask-svg12-ref.png | Bin 0 -> 2529 bytes ...paint-svg-ref.png => smask-paint-svg11-ref.png} | Bin test/smask-paint-svg12-ref.png | Bin 0 -> 2612 bytes test/{smask-svg-ref.png => smask-svg11-ref.png} | Bin test/smask-svg12-ref.png | Bin 0 -> 3634 bytes ...k-text-svg-ref.png => smask-text-svg11-ref.png} | Bin test/smask-text-svg12-ref.png | Bin 0 -> 1886 bytes ...svg-ref.png => source-clip-scale-svg11-ref.png} | Bin test/source-clip-scale-svg12-ref.png | Bin 0 -> 119 bytes ...n-svg-ref.png => surface-pattern-svg11-ref.png} | Bin test/surface-pattern-svg12-ref.png | Bin 0 -> 17663 bytes ...2-ref.png => text-pattern-svg11-argb32-ref.png} | Bin ...24-ref.png => text-pattern-svg11-rgb24-ref.png} | Bin test/text-pattern-svg12-argb32-ref.png | Bin 0 -> 1733 bytes test/text-pattern-svg12-rgb24-ref.png | Bin 0 -> 1563 bytes ...otate-svg-ref.png => text-rotate-svg11-ref.png} | Bin test/text-rotate-svg12-ref.png | Bin 0 -> 18301 bytes ...y-svg-ref.png => user-font-proxy-svg11-ref.png} | Bin test/user-font-proxy-svg12-ref.png | Bin 0 -> 17902 bytes ...svg-ref.png => user-font-rescale-svg11-ref.png} | Bin test/user-font-rescale-svg12-ref.png | Bin 0 -> 15741 bytes ...er-font-svg-ref.png => user-font-svg11-ref.png} | Bin test/user-font-svg12-ref.png | Bin 0 -> 6134 bytes 74 files changed, 128 insertions(+), 40 deletions(-) commit 86e57671dd21078215505202ef8aa2141d10fb9c Author: Behdad Esfahbod <behdad@behdad.org> Date: Fri Sep 26 11:08:22 2008 -0400 [check-*.sh] Redirect error reports to stderr doc/public/check-doc-coverage.sh | 6 +++--- src/check-def.sh | 2 +- src/check-doc-syntax.sh | 8 ++++---- src/check-headers.sh | 2 +- src/check-plt.sh | 2 +- src/check-preprocessor-syntax.sh | 10 +++++----- 6 files changed, 15 insertions(+), 15 deletions(-) commit 096f9def0f599e527d97bb979f7cafbc87dcddc6 Author: Behdad Esfahbod <behdad@behdad.org> Date: Fri Sep 26 10:55:44 2008 -0400 Add slim markers to make check-plt.sh happy src/cairo-font-face.c | 4 ++++ src/cairo-user-font.c | 4 ++++ src/cairo.c | 5 +++++ src/cairoint.h | 13 +++++++++++++ 4 files changed, 26 insertions(+) commit 1de059b33bd2e6121e35ccf1c4a35ba7755c44d8 Author: Behdad Esfahbod <behdad@behdad.org> Date: Fri Sep 26 09:49:50 2008 -0400 [RELEASING] Add ftp-release@lists.freedesktop.org That list is monitored by various distros to pick up new releases. We should CC release announcements there. RELEASING | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) commit 9662be8ee77e219908b64b82e828d577dcf120f0 Author: Behdad Esfahbod <behdad@behdad.org> Date: Fri Sep 26 00:22:00 2008 -0400 [twin-font] Adjust ascent/descent The font data uses a 64 int high font space with baseline at 50. src/cairo-font-face-twin.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) commit 3883a371f2ba78b17a1302ba8c5df1a79f3e4075 Author: Behdad Esfahbod <behdad@behdad.org> Date: Fri Sep 26 00:20:21 2008 -0400 [scaled-font] Use full ctm matrix when comparing scaled-font keys We use the full matrix in hash computation, but only compare the non-translation items in equality check. This is no bug though, as we set the ctm translation components of a scaled font to zero explicitly. But the change makes the hash and equal functions consistent, which is good. src/cairo-scaled-font.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) commit 48099af9fed7497a230992e41d559a19c5dbdc60 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Sep 25 13:24:58 2008 +0100 [boilerplate] Retry conversion in process if first attempt fails. One possibility for a read failure whilst converting the image is if the external utility crashed. This information is important for the test suite as knowing input that causes the converter to crash is just as vital as identifying a crash within the library. boilerplate/cairo-boilerplate-pdf.c | 11 +---------- boilerplate/cairo-boilerplate-svg.c | 11 +---------- boilerplate/cairo-boilerplate.c | 35 +++++++++++++++++++++++++++++++++-- boilerplate/cairo-boilerplate.h | 11 +++++++++-- 4 files changed, 44 insertions(+), 24 deletions(-) commit f2c484d73ce61012a2d9925f15b55e7c51e34b85 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Sep 25 00:16:45 2008 +0100 [test/any2ppm] Handle short reads/writes. No excuse other than simple laziness - it manifested itself with random "error whilst reading" failures. boilerplate/cairo-boilerplate.c | 24 +++++++++++++++++--- test/any2ppm.c | 50 +++++++++++++++++++++++++++++++++++------ 2 files changed, 64 insertions(+), 10 deletions(-) commit 681424cbaf94556bf1804547b42ad642c0a066ab Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Sep 25 10:40:32 2008 +0100 [analysis] Merge two analysis status codes. Since there is an implicit precedence in the ranking of the analysis return codes, provide a function to centralize the logic within the analysis surface and isolate the backends from the complexity. src/cairo-analysis-surface-private.h | 3 +++ src/cairo-analysis-surface.c | 50 +++++++++++++++++++++++++++--------- src/cairo-pdf-surface.c | 31 +++++++--------------- 3 files changed, 50 insertions(+), 34 deletions(-) commit 06f0cc81d227a571e011e83cc192050a34d6ea13 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Sep 25 10:29:27 2008 +0100 Add a macro to determine whether a status is a fatal error. In a few places we need to check whether we have a fatal error or an internal return code, so provide a utility macro to centralise the check. src/cairo.c | 3 +-- src/cairoint.h | 3 +++ 2 files changed, 4 insertions(+), 2 deletions(-) commit ba18d57339dbedbbdb7f3ca35b06ce6b5dbcd972 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Thu Sep 25 09:39:41 2008 +0100 [pdf] Propagate status Check and return a few forgotten error codes from static functions. src/cairo-pdf-operators.c | 14 +++++++++++--- src/cairo-pdf-surface.c | 6 +++++- 2 files changed, 16 insertions(+), 4 deletions(-) commit 907f550a1b13b0a388f9ff22663911d4aabbe2d8 Author: Adrian Johnson <ajohnson@redneon.com> Date: Fri Sep 26 20:26:55 2008 +0930 Fix bitmap-font XFAIL Images in PDF are scaled to a unit square. In PS we set the ImageMatrix to do the same. When the image is painted we scale the graphics state to paint the image at the right size. In the case of Type 3 fonts consisting of bitmap images we want to paint the images at their original size so we scale the graphics state by the image width and height. The bug was that we were scaling by the width/height in the glyph metrics. For non rotated fonts this worked. However for rotated fonts the width/height of the glyph images may be larger than the width/height in the glyph metrics. This resulted in a Type 3 font where the glyph images were scaled slightly smaller than they should have been. src/cairo-type3-glyph-surface.c | 10 +++------- test/Makefile.am | 1 - test/bitmap-font-pdf-argb32-ref.png | Bin 1064 -> 0 bytes 3 files changed, 3 insertions(+), 8 deletions(-) commit f8542dc9dd4dd0685f68381f21fa72dbddd8d682 Author: Behdad Esfahbod <behdad@behdad.org> Date: Thu Sep 25 19:08:24 2008 -0400 [twin-font] Clean up font data by joining lines and closing paths Two changes here: * Replace move_to;line_to;move_to;line_to sequences with move_to;line_to;line_to when feasible. * Close paths for round glyphs. Both improve the stroke rendering of the joint. The first change also saves 3 bytes per joint (33 such joints). Which we have just left unused for now. To reclaim them one need to update the charset table. Something for a lazy Sunday afternoon scripting task. In the saving department, we can save further by: - Getting rid of the left/ascent/descent values as we compute glyph bounding box automatically. Then we can liberally use the right value to adjust glyph advance width. Saves three bytes per glyph (there's 96 glyphs in the font). - First operation is always a move_to. So we can remove the 'm' for that. Ugly though. And the charset has zeros for the first 32 entries. Can get rid of that too at the expense of handling it in the code... In total, combining the above we can save some 500 bytes. The font currently takes about 3.7kb. src/cairo-font-face-twin-data.c | 154 ++++++++++++++++++++-------------------- src/cairo-font-face-twin.c | 18 +++-- 2 files changed, 91 insertions(+), 81 deletions(-) commit d5a998387bcee6569d33375d592190f480f12712 Author: Behdad Esfahbod <behdad@behdad.org> Date: Thu Sep 25 04:27:11 2008 -0400 Add an internal font face The font data and rendering is adapted from Keith Packard's Twin window system. The hinting stuff is not ported yet, but hey, it renders! The implementation uses user fonts, and the user font backend is modified to use this font face (which we call "twin" font face internally) when a toy font is needed. The font face layer is then modified to use this font if: - The toy font face "cairo" is asked for, or - No native font backend is available, or - The preferred native font backend fails to return a font with STATUS_UNSUPPORTED. No font backend does this right now but the idea is to change FreeType to return it if no fonts found on the system. We also allow building with no font backends now! The new doc/tutorial/src/twin.c file tests the twin face at various sizes. build/configure.ac.features | 26 +- doc/tutorial/src/twin.c | 39 ++ src/Makefile.sources | 2 + src/cairo-font-face-twin-data.c | 1030 +++++++++++++++++++++++++++++++++++++++ src/cairo-font-face-twin.c | 199 ++++++++ src/cairo-font-face.c | 33 +- src/cairo-user-font.c | 53 +- src/cairoint.h | 18 +- 8 files changed, 1371 insertions(+), 29 deletions(-) commit dd7e2461ce748403e121a5de5e4e4c8890e39236 Author: Carl Worth <cworth@cworth.org> Date: Thu Sep 25 16:06:19 2008 -0700 Increment cairo version to 1.8.1 after the 1.8.0 release. cairo-version.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)