summaryrefslogtreecommitdiffstats
path: root/gfx
Commit message (Collapse)AuthorAgeLines
* Revert "Bug 1388020. r=nical, a=RyanVM"wolfbeast2018-08-30-41/+6
| | | | This reverts commit e7189e33f533f9b974b22c2110b522a13bc4c7f6.
* Remove unused variabletrav902018-08-19-1/+0
|
* Bug 1335296 - Expand about:support WebGL informationjanekptacijarabaci2018-08-01-0/+68
|
* ANGLE: Initialize uniform blocks before detaching shaderswolfbeast2018-07-24-1/+11
| | | | This resolves #668
* Add pref to control CanOptimizeToImageLayer for large downscale factors.wolfbeast2018-07-14-0/+1
| | | | This resolves #626
* Switch to Lanczos scaling from Hamming to get acceptable fast downscaling.wolfbeast2018-07-14-15/+10
| | | | | | | | | | In visual tests we see that Hamming-1 is not as good as Lanczos-2, however it is about 40% faster, and Lanczos-2 itself is about 30% faster than Lanczos-3. The use of Hamming-1 has been deemed an unacceptable trade-off between quality and speed due to the limited pixel space it operates in, so we pick Lanczos-2 here. On modern hardware, Lanczos-2 doesn't have any noticeable impact in normal use.
* Ad workaround for broken B5G6R5 format in Intel driver.wolfbeast2018-07-11-2/+16
|
* Re-apply genericDraw* fixeswolfbeast2018-07-11-6/+6
|
* Roll back to ANGLE/2845wolfbeast2018-07-11-55178/+21143
|
* Bug 604026 - Sync event loops in workers should be created only if ↵janekptacijarabaci2018-07-05-1/+1
| | | | compatible with the worker shutdown status
* Rework gfxFontUtils::MapCharToGlyphFormat4 to be more robust. DiDwolfbeast2018-07-01-42/+57
|
* Bug 1464039 - Only reject qcms transform with invalid grid size if the ↵Nicolas Silva2018-06-30-2/+8
| | | | transform function uses the grid size. r=Bas a=jcristau
* Bug 1464039 - Reject some invalid transforms in qcms. r=mwoodrow, a=RyanVMNicolas Silva2018-06-30-0/+4
|
* Improve the SSSE3 scaler.wolfbeast2018-06-07-9/+25
|
* Account for offsets when creating buffers for image textures.wolfbeast2018-06-07-3/+44
|
* Bug 1462682 - Skia path bounds rounding fix. r=rhunt, a=RyanVMLee Salzman2018-06-07-14/+8
|
* Remove SPS profiler.wolfbeast2018-05-24-60/+2
| | | | | - Conditionals and code blocks. (MOZ_ENABLE_PROFILER_SPS) - Stub out several profiler-only functions.
* Remove IPC CrashReporterClient/Hostwolfbeast2018-05-22-15/+0
| | | | Tag #20.
* Remove gfxCrashReporterUtils.wolfbeast2018-05-20-240/+0
| | | | Tag #20.
* Fix missing include in Skia SafeMathwolfbeast2018-05-17-0/+1
|
* Add omitted Skia "Safe Math" includes.wolfbeast2018-05-15-0/+2
|
* Remove other gonk widget conditionals and unused files.wolfbeast2018-05-13-2/+2
| | | | Tag #288.
* Remove MOZ_B2G leftovers and some dead B2G-only components.wolfbeast2018-05-12-45/+5
|
* Bug 1454692 - Backport some upstream Skia fixes to ESR52. r=rhunt, a=abillingsLee Salzman2018-05-10-4/+17
| | | | | --HG-- extra : histedit_source : 0fcd64cabe6f54a2286083d6518e4e6451183a19%2C37f5e7f9dbbfc01102631c33b23329d2af5aa71b
* Font fallback should detect emoji range for color emoji font.wolfbeast2018-05-09-50/+44
|
* Fix build bustagewolfbeast2018-05-09-4/+3
| | | | Fix nsAtom->nsIAtom and FindAndAddFamilies()
* Use font.name-list.emoji preference for emoji presenration.wolfbeast2018-05-09-19/+88
|
* Issue #325 Part 3: Remove support for using legacy Gecko normalization in ↵wolfbeast2018-05-04-34/+6
| | | | gfxHarfBuzzShaper.
* Merge pull request #295 from janekptacijarabaci/_native_52ESR_events_auxclick_1Moonchild2018-04-29-0/+1
|\ | | | | UI Events - implement auxclick
| * Implement auxclickjanekptacijarabaci2018-02-04-0/+1
| | | | | | | | | | | | Bug(s): https://bugzilla.mozilla.org/show_bug.cgi?id=1304044 (native in moebius)
* | Use SRWlocks for mutexes in cairo on Windows.wolfbeast2018-04-28-11/+12
| |
* | Bug 1393367 - Change MOZ_ASSERT to MOZ_RELEASE_ASSERT. r=mstange, r=fbraun, ↵Miko Mynttinen2018-04-19-1/+1
| | | | | | | | | | | | | | | | a=RyanVM --HG-- extra : source : 1908cd8ed88dd4f77a99dff39c193d7d3f435195 extra : intermediate-source : 9718d92fab4d9a39acdc2afb0302b6fcd7997f6c
* | Bug 1388020. r=nical, a=RyanVMAndrew Osmond2018-04-19-6/+41
| |
* | Bug 1441941 - Limit allocations in SkTDArray. r=jrmuizel, a=RyanVMLee Salzman2018-04-19-3/+145
| |
* | Bug 1440717 - Use RefPtr for CompositingRenderTargetOGL::mGL. r=Bas, a=rituNicolas Silva2018-04-19-2/+2
| | | | | | | | | | | | --HG-- extra : source : b6d2d55223d2aa5cb85bbdf33075d1d38f2a9a30 extra : intermediate-source : 0ec90964e0bc479412a4da8a61d48b665211736e
* | Rebase on masterwolfbeast2018-04-11-7/+7
|\ \
| * | Fix Twemoji GTK build bustagetrav902018-04-04-1/+1
| | | | | | | | | | | | Followup to b4e843a.
| * | Change embedded emoji font from EmojiOne to Twemoji.wolfbeast2018-04-04-6/+6
| | |
* | | Remove base conditional code for crash reporter and injector.wolfbeast2018-03-30-80/+0
|/ /
* | Fix line heights to properly scale to requested size for color-bitmapped SFNTs.wolfbeast2018-03-27-7/+23
| |
* | Allow bitmap fonts to force scaling and bypass tolerance check in ↵wolfbeast2018-03-27-4/+23
| | | | | | | | gfxFcPlatformFontList.
* | Use FC_OUTLINE instead of FC_SCALABLE with Fontconfig to check if a font is ↵wolfbeast2018-03-27-3/+3
| | | | | | | | scalable.
* | Bug 1388020. r=nical, a=RyanVMAndrew Osmond2018-03-14-41/+94
| |
* | Bug 1428947 - Check plane width & stride constraints. r=mattwoodrow, a=rituDan Glastonbury2018-03-14-10/+13
| | | | | | | | | | | | | | MozReview-Commit-ID: 328ETwMdVnq --HG-- extra : rebase_source : e16b28d137de080f9d8495c6937e24ac16b16ab1
* | Remove gfx/thebes checks for Windows 7+.wolfbeast2018-02-19-19/+6
| |
* | Remove some gfx/thebes checks for Windows Vista+.wolfbeast2018-02-19-23/+6
| |
* | Remove no longer relevant ClearType prefs.wolfbeast2018-02-18-65/+5
| | | | | | | | Tag #22
* | Stop shipping d3dcompiler_43.dllwolfbeast2018-02-18-33/+0
| |
* | Add more NS_BUILD_REFCNT_LOGGING in AtomicRefCountedWithFinalize.wolfbeast2018-02-08-2/+8
| |
* | Avoid potentially unsafe snprintf usage in FPSCounter.wolfbeast2018-02-08-6/+21
| | | | | | | | | | | | snprintf returns the number of bytes it would have written when it runs out of space. This patch makes sure we properly handle this unlikely event in FPSCounter. This patch also makes sure we don't print out the contents of an uninitialized buffer.