The logic for extending the lifetimes of hoisted values is yet again not
quite right. We do need to track the proper lifetime of these values,
and only extend them to infinity if we actually hoist them. The JIT
has a no-hoist fallback plan for JITting when under register pressure,
but we're actually undermining that by treating these values as eternal
today, burning those registers where they should be recycled.
So I think the answer to this is either,
1) add a used_in_loop bit, and treat as eternal only if hoist && used_in_loop
2) track two death points, one when hoisted, one when not
I think the first is cleaner, and maybe easier too.
Change-Id: I09704ecaf12896d242850c2e827b19e31fb1dd70
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/242591
Reviewed-by: Mike Klein <mtklein@google.com>
Reviewed-by: Herb Derby <herb@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
Commit-Queue: Herb Derby <herb@google.com>
Auto-Submit: Mike Klein <mtklein@google.com>
Gives enough information to locate variables by name (using the same
scheme as glGetUniformLocation), and provide hints about type and size.
Bug: skia:9513
Change-Id: I9444f1042471967a79c9f05167dcdb78eca41bad
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/244502
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Change-Id: Iee2f376f7270200b5732726e9c6a3015b0cd4139
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/248928
Reviewed-by: Ben Wagner aka dogben <benjaminwagner@google.com>
Commit-Queue: Stephen White <senorblanco@chromium.org>
This requires some changes to dawnSetProcs() and some build file changes.
Change-Id: I8a643ec0690857d53db7c1f8f5223077e30f69e8
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/248860
Commit-Queue: Stephen White <senorblanco@chromium.org>
Reviewed-by: Robert Phillips <robertphillips@google.com>
The idea/dream here is to be able to (optionally) tessellate in onPrePrepareDraws and then just copy the prePrepared data into the vertex buffer.
Change -Id: I899eebce36dc9516bfc4b0fce5e657458167c772
Change-Id: I899eebce36dc9516bfc4b0fce5e657458167c772
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/248804
Reviewed-by: Michael Ludwig <michaelludwig@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
2d0e5b554d..0b9ebe58e3
git log 2d0e5b554d4f..0b9ebe58e380 --date=short --no-merges --format='%ad %ae %s'
2019-10-15 jmadill@chromium.org Vulkan: Add "ImageViewHelper".
2019-10-15 lsalzman@mozilla.com Use image upload workaround for Intel Ivy Bridge with D3D11 renderer.
Created with:
gclient setdep -r third_party/externals/angle2@0b9ebe58e380
If this roll has caused a breakage, revert this CL and stop the roller
using the controls here:
https://autoroll.skia.org/r/angle-skia-autoroll
Please CC csmartdalton@google.com on the revert to ensure that a human
is aware of the problem.
To report a problem with the AutoRoller itself, please file a bug:
https://bugs.chromium.org/p/skia/issues/entry?template=Autoroller+Bug
Documentation for the AutoRoller is here:
https://skia.googlesource.com/buildbot/+/master/autoroll/README.md
CQ_INCLUDE_TRYBOTS=skia.primary:Build-Debian9-Clang-x86_64-Release-ANGLE;skia.primary:Test-Win10-Clang-AlphaR2-GPU-RadeonR9M470X-x86_64-Debug-All-ANGLE;skia.primary:Test-Win10-Clang-Golo-GPU-QuadroP400-x86_64-Debug-All-ANGLE;skia.primary:Test-Win10-Clang-NUC5i7RYH-GPU-IntelIris6100-x86_64-Debug-All-ANGLE;skia.primary:Test-Win10-Clang-NUC6i5SYK-GPU-IntelIris540-x86_64-Debug-All-ANGLE;skia.primary:Test-Win10-Clang-NUC8i5BEK-GPU-IntelIris655-x86_64-Debug-All-ANGLE;skia.primary:Test-Win10-Clang-NUCD34010WYKH-GPU-IntelHD4400-x86_64-Debug-All-ANGLE
Bug: None
TBR=csmartdalton@google.com
Change-Id: I14bfd6949fe01f6b97fcb7bb6dbbcae9df9d94a4
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/248898
Reviewed-by: skia-autoroll <skia-autoroll@skia-public.iam.gserviceaccount.com>
Commit-Queue: skia-autoroll <skia-autoroll@skia-public.iam.gserviceaccount.com>
As of revision 7924d9a4ae (https://skia-review.googlesource.com/c/skia/+/222277), SkTextBlob::getIntercepts no longer considers each glyph's vertical offset. So getIntercepts will only work properly if each glyph has a vertical offset of 0. This patch restores the original behavior of offsetting the bounds from each glyph's position.
Without this fix, Firefox has no way to implement proper intercepts for underlining decorations when glyphs are offset...
Change-Id: I06fc4b63bd57c9d70e3b07a95ead69f3caa8b33a
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/248724
Reviewed-by: Herb Derby <herb@google.com>
Commit-Queue: Herb Derby <herb@google.com>
Auto-Submit: Lee Salzman <lsalzman@mozilla.com>
Remove optional param to SetIndexBuffer().
Update SetVertexBuffers -> SetVertexBuffer.
PipelineStageDescriptor -> ProgrammableStageDescriptor.
Update past generator changes, ring buffer changes, etc.
Add ErrorScopeTracker.cpp/.h to the build.
Add Vulkan MemoryResourceAllocator files.
Fix vertexShader ShaderStageDescriptor.
Fix spirv-cross include path.
TextureUsageBit -> TextureUsage, etc.
DawnErrorCallback fixes.
Removal of texture.CreateDefaultView.
Fix GL supported_extesions mumbo jumbo.
Update past ChromeOS change.
Add PassResourceUsageTracker.cpp/.h to build.
Add GLFormat.cpp/.h to build.
Add Extensions and Toggles to the build.
Add EncodingContext, AttachmentState to build.
Add RenderEncoderBase to Dawn build.
gn format dawn BUILD.gn, spirv-cross BUILD.gn.
Change-Id: I26538d63c93668647048814aad6ad456ae323679
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/248261
Commit-Queue: Stephen White <senorblanco@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
This should also fix a bug in the check generated file housekeeper bot.
Change-Id: I70eeb1f8783c760fcba1e81ea79b2734cae54b25
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/248856
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Commit-Queue: Greg Daniel <egdaniel@google.com>
Change-Id: I511f15f056e15beddeaaf203414a1519569bfbe4
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/248815
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Jim Van Verth <jvanverth@google.com>
This is a reland of 3f55695a04
Original change's description:
> ccpr: Unblacklist Qualcomm with msaa
>
> Bug: skia:
> Change-Id: I1ab69fe87136955b01a329675f1d0affec3700ec
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/245530
> Commit-Queue: Chris Dalton <csmartdalton@google.com>
> Reviewed-by: Brian Salomon <bsalomon@google.com>
Bug: skia:
Change-Id: Icbd0ad7546bc089c538afafed515b22b05130e47
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/248702
Reviewed-by: Ravi Mistry <rmistry@google.com>
Commit-Queue: Chris Dalton <csmartdalton@google.com>
These checks are no longer needed as we have enough checks in other parts
of the code now. Additionally we want to remove any GrColorType and format
compatibility checks in proxy provider since proxies will not have knowledge
of color types in the future.
This also lets us get ride of some used of GrPixelConfig.
Change-Id: Ice835cba69ff783fd7e7d9b57321882b1e50093a
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/248803
Commit-Queue: Greg Daniel <egdaniel@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
This is a reland of 7e6d3b00ad
Original change's description:
> Use GL_QCOM_TILED_RENDERING to explicitly discard stencil
>
> Bug: skia:
> Change-Id: I7a6234f6a9955dbda83ab2d8fc6930fe680e102b
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/245529
> Commit-Queue: Chris Dalton <csmartdalton@google.com>
> Reviewed-by: Brian Salomon <bsalomon@google.com>
Bug: skia:
Change-Id: Ifb2ac1a595be0ca55d75d53d85bf76f7d98e1c15
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/246282
Commit-Queue: Chris Dalton <csmartdalton@google.com>
Reviewed-by: Greg Daniel <egdaniel@google.com>
This change also allows for the remove of GrPixelConfigIsOpaque function.
Bug: skia:6718
Change-Id: I7e7b70f02d911eda67640d648fb6348091e0f55d
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/248698
Reviewed-by: Robert Phillips <robertphillips@google.com>
Commit-Queue: Greg Daniel <egdaniel@google.com>
Now using dispatch_semaphore_t, maybe it already knows?
Change-Id: I65499ceaee195293541a37704ae424bbee7c8740
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/248456
Auto-Submit: Mike Klein <mtklein@google.com>
Commit-Queue: Ben Wagner <bungeman@google.com>
Reviewed-by: Ben Wagner <bungeman@google.com>
Simplify burst handling. Scripts should just add to burst (if
they want to handle programmatic bursting, as well).
Update most effects to handle dynamic updates to position better,
and add a sample effect meant to be used with mouse tracking.
Change-Id: Ia302e1d04e62e2b07974807c44067786cc10a8ad
Bug: skia:9513
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/248798
Commit-Queue: Brian Osman <brianosman@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
This also changes how the Metal backend computes its key (removing the # of mipMap levels) and removes the use of the # of mipMap levels when setting up the MTLSamplerDescriptor
Change-Id: I29fedc7c7a12773f3efeea824bca98cbcde98be8
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/248565
Reviewed-by: Jim Van Verth <jvanverth@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
fede5cb664..2d0e5b554d
git log fede5cb664d9..2d0e5b554d4f --date=short --no-merges --format='%ad %ae %s'
2019-10-15 mihu@microsoft.com GL_EXT_multisampled_render_to_texture extension. Part 2.
2019-10-15 jdarpinian@chromium.org Fix registry_xml.py dependency
2019-10-14 jmadill@chromium.org Metal: Use -weak_framework to support OS 10.10.
2019-10-14 bsheedy@chromium.org Add .yapfignore file
Created with:
gclient setdep -r third_party/externals/angle2@2d0e5b554d4f
If this roll has caused a breakage, revert this CL and stop the roller
using the controls here:
https://autoroll.skia.org/r/angle-skia-autoroll
Please CC csmartdalton@google.com on the revert to ensure that a human
is aware of the problem.
To report a problem with the AutoRoller itself, please file a bug:
https://bugs.chromium.org/p/skia/issues/entry?template=Autoroller+Bug
Documentation for the AutoRoller is here:
https://skia.googlesource.com/buildbot/+/master/autoroll/README.md
CQ_INCLUDE_TRYBOTS=skia.primary:Build-Debian9-Clang-x86_64-Release-ANGLE;skia.primary:Test-Win10-Clang-AlphaR2-GPU-RadeonR9M470X-x86_64-Debug-All-ANGLE;skia.primary:Test-Win10-Clang-Golo-GPU-QuadroP400-x86_64-Debug-All-ANGLE;skia.primary:Test-Win10-Clang-NUC5i7RYH-GPU-IntelIris6100-x86_64-Debug-All-ANGLE;skia.primary:Test-Win10-Clang-NUC6i5SYK-GPU-IntelIris540-x86_64-Debug-All-ANGLE;skia.primary:Test-Win10-Clang-NUC8i5BEK-GPU-IntelIris655-x86_64-Debug-All-ANGLE;skia.primary:Test-Win10-Clang-NUCD34010WYKH-GPU-IntelHD4400-x86_64-Debug-All-ANGLE
Bug: None
TBR=csmartdalton@google.com
Change-Id: I6c9e0cb23a6445be884be46e6daf139f087cac37
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/248731
Reviewed-by: skia-autoroll <skia-autoroll@skia-public.iam.gserviceaccount.com>
Commit-Queue: skia-autoroll <skia-autoroll@skia-public.iam.gserviceaccount.com>
I'm gearing up to add a GrBackendFormat to GrPrimitiveProcessor::TextureSampler so this CL just reduces the API surface area I'll need to alter.
The GrFragmentProcess::TextureSampler changes just keep it aligned with its "twin".
Change-Id: Ia9ece03ca76b4f6c8ebdaf0e0ba0061ecde6c5f4
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/248558
Reviewed-by: Greg Daniel <egdaniel@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
depot_tools w/vpython now seems to require the depot_tools dir
to be on PATH, and also python and curl need to be installed.
* Creates a new base image that has all the build tools we need for
building Skia: skia-build-tools.
* Moves skia-release to use that new base image.
* Moves both the skia-release and wasm builds to use fetch and gclient sync.
* Changes gcr.io/skia-public/emsdk-base:prod to depend on skia-build-tools.
Bug: skia:993053
Change-Id: Ib6218d8c18dd59d43de84dcf063f204cfc8556d0
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/248556
Commit-Queue: Joe Gregorio <jcgregorio@google.com>
Reviewed-by: Eric Boren <borenet@google.com>
This is a reland of 4e385e21a4
Original change's description:
> Minimal iOS app: Perform present subsequent to flush
>
> Currently the backbuffer present is not on the same MTLCommandQueue
> as the flush from Skia, which probably means that it's happening
> concurrently. This can overwhelm the GPU. This CL submits the present
> to the same queue as the flush.
>
> Change-Id: Ibaf805553931c9dc46368b362a2391425ae3e60e
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/248557
> Reviewed-by: Hal Canary <halcanary@skia.org>
> Commit-Queue: Jim Van Verth <jvanverth@google.com>
Change-Id: I118047630b8936838d5da8ef4307ad5c5e8d2ade
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/248562
Reviewed-by: Hal Canary <halcanary@skia.org>
Commit-Queue: Jim Van Verth <jvanverth@google.com>
It seems like we see quite a few "force quarantined" bots when pushing
resources to the device fails. Rewrite copy_directory_contents_to_device
to use the _adb method, which automatically retries.
Change-Id: Iae36f8e42e85a58fdddc8c9dc80e693a371fab8b
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/248560
Commit-Queue: Eric Boren <borenet@google.com>
Auto-Submit: Ben Wagner aka dogben <benjaminwagner@google.com>
Reviewed-by: Eric Boren <borenet@google.com>
I'd rather catch uninitialized input here than
somewhere deep down in zlib or libpng.
Bug: chromium:1013368
Cq-Include-Trybots: skia.primary:Test-Debian9-Clang-GCE-CPU-AVX2-x86_64-Debug-All-MSAN
Change-Id: I37f5a66a8fc12c7e0ce6d3100de94021bb5343ef
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/248551
Reviewed-by: Leon Scroggins <scroggo@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
Update to GrProgramInfo change.
Implement GrCaps::bytesPerPixel().
Change-Id: Ic695fe85f2cd1389918724d17368df60f88ee2db
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/247796
Commit-Queue: Stephen White <senorblanco@chromium.org>
Reviewed-by: Robert Phillips <robertphillips@google.com>
This reverts commit 4e385e21a4.
Reason for revert: Causing issues with iOS Skottie app.
Original change's description:
> Minimal iOS app: Perform present subsequent to flush
>
> Currently the backbuffer present is not on the same MTLCommandQueue
> as the flush from Skia, which probably means that it's happening
> concurrently. This can overwhelm the GPU. This CL submits the present
> to the same queue as the flush.
>
> Change-Id: Ibaf805553931c9dc46368b362a2391425ae3e60e
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/248557
> Reviewed-by: Hal Canary <halcanary@skia.org>
> Commit-Queue: Jim Van Verth <jvanverth@google.com>
TBR=jvanverth@google.com,halcanary@google.com,halcanary@skia.org
Change-Id: I56a9f1f12ebceeb750948e459ff9f90e06b7b2e4
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/248561
Reviewed-by: Jim Van Verth <jvanverth@google.com>
Commit-Queue: Jim Van Verth <jvanverth@google.com>
If we don't care about any of the existing content in the framebuffer
(which we don't in the case of blurs) then fullscreen clears are
always a big win on tilers.
Bug: chromium:1008192
Change-Id: I538ce295ab91c56e88b6cb44773a35644980a162
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/247798
Reviewed-by: Michael Ludwig <michaelludwig@google.com>
Commit-Queue: Chris Dalton <csmartdalton@google.com>
Our base supported SDK is the one available in Xcode 10, so we don't
need these any more.
Bug: skia:9408
Change-Id: I71f4b6ee77a1f9b0a3b2dfa902c1bf23f2d66696
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/247365
Reviewed-by: Greg Daniel <egdaniel@google.com>
Commit-Queue: Jim Van Verth <jvanverth@google.com>
Currently the backbuffer present is not on the same MTLCommandQueue
as the flush from Skia, which probably means that it's happening
concurrently. This can overwhelm the GPU. This CL submits the present
to the same queue as the flush.
Change-Id: Ibaf805553931c9dc46368b362a2391425ae3e60e
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/248557
Reviewed-by: Hal Canary <halcanary@skia.org>
Commit-Queue: Jim Van Verth <jvanverth@google.com>
Ideally the GrProgramDesc would be a simple program key. We currently can't remove usage of the KeyHeader for "hasPointSize" but this, at least, removes the other uses.
Bug: skia:9455
Change-Id: Ie9e15997ee79c6ffe62b5cd2c96885ca06204383
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/247597
Commit-Queue: Robert Phillips <robertphillips@google.com>
Reviewed-by: Greg Daniel <egdaniel@google.com>
Bug: skia:6718
Change-Id: I4d9c31bc0ca42a80e652de29462531587efef031
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/248257
Reviewed-by: Michael Ludwig <michaelludwig@google.com>
Commit-Queue: Greg Daniel <egdaniel@google.com>
This defers the decision about whether pointSize is emitted until down in GrProgramDesc::Build. Although a bit wonky in GL's case, this anticipates the world where each cluster of meshes always has the same GrPrimitiveType.
Ultimately, we would like the GrPrimitiveType to be part of the GrProgramInfo and known far earlier.
Change-Id: I65566b2d77eab6a69f7af5aa60a069299b58db65
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/246018
Reviewed-by: Greg Daniel <egdaniel@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
67486ec245..fede5cb664
git log 67486ec24551..fede5cb664d9 --date=short --no-merges --format='%ad %ae %s'
2019-10-13 jmadill@chromium.org Vulkan: Remove resource destruction layer suppressions.
2019-10-12 jiajie.hu@intel.com Add sampler array index validation for ESSL 3.00 onwards
2019-10-12 jmadill@chromium.org Roll chromium_revision 43eb5958f7..9325340209 (704334:704874)
2019-10-12 jmadill@chromium.org Add a roll_deps script based on WebRTC.
2019-10-12 jdarpinian@chromium.org Extend ANGLE_request_extension with DisableExtension
2019-10-11 lehoangq@gmail.com Metal implementation pt 1: autogen resources
2019-10-11 cnorthrop@google.com Vulkan: Support binding texture levels as a rendertarget
2019-10-11 tobine@google.com Vulkan:Add support to stage D or S textures
2019-10-11 ianelliott@google.com Vulkan: Update GLES 3.0 expectations files
2019-10-11 b.schade@samsung.com Vulkan: finishToSerial only if the buffer is in use
2019-10-11 cnorthrop@google.com Vulkan: Handle 2Darray base/max level change
2019-10-11 jonahr@google.com GL: Missing check for texture format sample counts
Created with:
gclient setdep -r third_party/externals/angle2@fede5cb664d9
If this roll has caused a breakage, revert this CL and stop the roller
using the controls here:
https://autoroll.skia.org/r/angle-skia-autoroll
Please CC csmartdalton@google.com on the revert to ensure that a human
is aware of the problem.
To report a problem with the AutoRoller itself, please file a bug:
https://bugs.chromium.org/p/skia/issues/entry?template=Autoroller+Bug
Documentation for the AutoRoller is here:
https://skia.googlesource.com/buildbot/+/master/autoroll/README.md
CQ_INCLUDE_TRYBOTS=skia.primary:Build-Debian9-Clang-x86_64-Release-ANGLE;skia.primary:Test-Win10-Clang-AlphaR2-GPU-RadeonR9M470X-x86_64-Debug-All-ANGLE;skia.primary:Test-Win10-Clang-Golo-GPU-QuadroP400-x86_64-Debug-All-ANGLE;skia.primary:Test-Win10-Clang-NUC5i7RYH-GPU-IntelIris6100-x86_64-Debug-All-ANGLE;skia.primary:Test-Win10-Clang-NUC6i5SYK-GPU-IntelIris540-x86_64-Debug-All-ANGLE;skia.primary:Test-Win10-Clang-NUC8i5BEK-GPU-IntelIris655-x86_64-Debug-All-ANGLE;skia.primary:Test-Win10-Clang-NUCD34010WYKH-GPU-IntelHD4400-x86_64-Debug-All-ANGLE
Bug: None
TBR=csmartdalton@google.com
Change-Id: I6a583009ecdef9d9f1b4c03c3ce32b037e8615e6
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/248496
Reviewed-by: skia-autoroll <skia-autoroll@skia-public.iam.gserviceaccount.com>
Commit-Queue: skia-autoroll <skia-autoroll@skia-public.iam.gserviceaccount.com>
The dispatch_semaphore_t is a higher-level, more-efficient semaphore
primitive if the cross-process capabilities of semaphore_t are not
needed.
Bug: chromium:1012386
Change-Id: I7c39fa6b8c015bdf7b4b2275da8165165253c2d1
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/248259
Commit-Queue: Mike Klein <mtklein@google.com>
Reviewed-by: Mike Klein <mtklein@google.com>
