New issue
Advanced search Search tips

Issue 894200 link

Starred by 4 users
Status: Fixed
Owner:
khushals...@chromium.org
Closed: Jan 9
Cc:
gov...@chromium.org
vmi...@chromium.org
ericrk@chromium.org
enne@chromium.org
Components:
EstimatedDays: ----
NextAction: 2018-10-22
OS: Linux , Android , Windows , Chrome , Mac
Pri: 1
Type: Task

Blocked on:
issue 895765
issue 897507
issue 898292


Sign in to add a comment

Tracking bug for OOPR serialization improvements and metrics.

Project Member Reported by khushals...@chromium.org, Oct 10 
The current metrics measure raster duration for work on the raster worker threads, which doesn't capture the work on the GPU main thread. With more of this moving to the GPU main thread with OOPR, the existing metric is even more inaccurate.

This change adds the base metric: https://chromium-review.googlesource.com/c/chromium/src/+/1267235

We'd likely want to follow up with another metric which captures cpu time using ThreadTicks on platforms where it is available. Since 71 branch point is close, filing a bug to remember to merge this to 71 if necessary.
Showing comments 2 - 101 of 101 Older ›
Project Member

Comment 2 by bugdroid1@chromium.org, Oct 16 

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/01a6817cb9a1091f1a9763a8c69307cff2d1bc8d

commit 01a6817cb9a1091f1a9763a8c69307cff2d1bc8d
Author: Takashi Sakamoto <tasak@google.com>
Date: Tue Oct 16 09:07:21 2018

Revert "gpu: Add UMA metric for measuring total raster task duration."

This reverts commit c3e902dd12c2ae20f00dcf2fe521fefffd0034b7.

Reason for revert: this cl causes linux MSAN cc_unittests failures:

RasterizeWithGpuRasterizationCreatesResources.RunMultiThread_DelegatingRenderer
RasterBufferProviderTests/RasterBufferProviderTest.MeasureGpuRasterDuration/2
LayerTreeHostPictureTestResizeViewportWithGpuRaster.RunSingleThread_DelegatingRenderer

  [ RUN      ] RasterizeWithGpuRasterizationCreatesResources.RunMultiThread_DelegatingRenderer
  ==2667==WARNING: MemorySanitizer: use-of-uninitialized-value
      #0 0x5974de0 in Do<long> ./../../base/numerics/safe_math_clang_gcc_impl.h:74:13
      #1 0x5974de0 in Do<long> ./../../base/numerics/checked_math_impl.h:54:0
      #2 0x5974de0 in MathOp<CheckedAddOp, long> ./../../base/numerics/checked_math.h:245:0
      #3 0x5974de0 in operator+=<long> ./../../base/numerics/checked_math.h:339:0
      #4 0x5974de0 in base::time_internal::SaturatedAdd(base::TimeDelta, long) ./../../base/time/time.cc:70:0
      #5 0x49b3d8e in operator+ ./../../base/time/time.h:207:22
      #6 0x49b3d8e in cc::GpuRasterBufferProvider::CheckRasterFinishedQueries() ./../../cc/raster/gpu_raster_buffer_provider.cc:558:0

Seems, GetQueryObjectuivEXT doesn't always initialize gpu_duration:

GLuint gpu_duration;
    ri->GetQueryObjectuivEXT(it->query_id, GL_QUERY_RESULT_EXT, &gpu_duration);

Original change's description:
> gpu: Add UMA metric for measuring total raster task duration.
> 
> Add an UMA metric which reports the accumulated time for a raster task
> including the time from raster worker thread and the GPU main thread.
> The GPU side timer takes care of only measuring time slices when the
> decoder is processing commands.
> 
> R=​vmiura@chromium.org, piman@chromium.org
> 
> BUG:  894200 
> Cq-Include-Trybots: luci.chromium.try:android_optional_gpu_tests_rel;luci.chromium.try:linux_optional_gpu_tests_rel;luci.chromium.try:mac_optional_gpu_tests_rel;luci.chromium.try:win_optional_gpu_tests_rel;master.tryserver.blink:linux_trusty_blink_rel
> Change-Id: If02a76420dbdf5ffb5041b8290412d144d787fb4
> Reviewed-on: https://chromium-review.googlesource.com/c/1267235
> Commit-Queue: Khushal <khushalsagar@chromium.org>
> Reviewed-by: Sunny Sachanandani <sunnyps@chromium.org>
> Reviewed-by: Steven Holte <holte@chromium.org>
> Reviewed-by: Antoine Labour <piman@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#599870}

TBR=enne@chromium.org,vmiura@chromium.org,holte@chromium.org,sunnyps@chromium.org,khushalsagar@chromium.org,piman@chromium.org

Change-Id: I25b54841dca13aaa94cfc8dded8a40254f90d8e0
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Cq-Include-Trybots: luci.chromium.try:android_optional_gpu_tests_rel;luci.chromium.try:linux_optional_gpu_tests_rel;luci.chromium.try:mac_optional_gpu_tests_rel;luci.chromium.try:win_optional_gpu_tests_rel;master.tryserver.blink:linux_trusty_blink_rel
Reviewed-on: https://chromium-review.googlesource.com/c/1282519
Reviewed-by: Takashi Sakamoto <tasak@google.com>
Commit-Queue: Takashi Sakamoto <tasak@google.com>
Cr-Commit-Position: refs/heads/master@{#599926}
[modify] https://crrev.com/01a6817cb9a1091f1a9763a8c69307cff2d1bc8d/cc/raster/bitmap_raster_buffer_provider.cc
[modify] https://crrev.com/01a6817cb9a1091f1a9763a8c69307cff2d1bc8d/cc/raster/bitmap_raster_buffer_provider.h
[modify] https://crrev.com/01a6817cb9a1091f1a9763a8c69307cff2d1bc8d/cc/raster/gpu_raster_buffer_provider.cc
[modify] https://crrev.com/01a6817cb9a1091f1a9763a8c69307cff2d1bc8d/cc/raster/gpu_raster_buffer_provider.h
[modify] https://crrev.com/01a6817cb9a1091f1a9763a8c69307cff2d1bc8d/cc/raster/one_copy_raster_buffer_provider.cc
[modify] https://crrev.com/01a6817cb9a1091f1a9763a8c69307cff2d1bc8d/cc/raster/one_copy_raster_buffer_provider.h
[modify] https://crrev.com/01a6817cb9a1091f1a9763a8c69307cff2d1bc8d/cc/raster/raster_buffer_provider.h
[modify] https://crrev.com/01a6817cb9a1091f1a9763a8c69307cff2d1bc8d/cc/raster/raster_buffer_provider_unittest.cc
[modify] https://crrev.com/01a6817cb9a1091f1a9763a8c69307cff2d1bc8d/cc/raster/zero_copy_raster_buffer_provider.cc
[modify] https://crrev.com/01a6817cb9a1091f1a9763a8c69307cff2d1bc8d/cc/raster/zero_copy_raster_buffer_provider.h
[modify] https://crrev.com/01a6817cb9a1091f1a9763a8c69307cff2d1bc8d/cc/test/cc_test_suite.cc
[modify] https://crrev.com/01a6817cb9a1091f1a9763a8c69307cff2d1bc8d/cc/test/fake_raster_buffer_provider.cc
[modify] https://crrev.com/01a6817cb9a1091f1a9763a8c69307cff2d1bc8d/cc/test/fake_raster_buffer_provider.h
[modify] https://crrev.com/01a6817cb9a1091f1a9763a8c69307cff2d1bc8d/cc/tiles/tile_manager.cc
[modify] https://crrev.com/01a6817cb9a1091f1a9763a8c69307cff2d1bc8d/cc/tiles/tile_manager.h
[modify] https://crrev.com/01a6817cb9a1091f1a9763a8c69307cff2d1bc8d/cc/tiles/tile_manager_unittest.cc
[modify] https://crrev.com/01a6817cb9a1091f1a9763a8c69307cff2d1bc8d/gpu/GLES2/gl2extchromium.h
[modify] https://crrev.com/01a6817cb9a1091f1a9763a8c69307cff2d1bc8d/gpu/command_buffer/build_gles2_cmd_buffer.py
[modify] https://crrev.com/01a6817cb9a1091f1a9763a8c69307cff2d1bc8d/gpu/command_buffer/build_raster_cmd_buffer.py
[modify] https://crrev.com/01a6817cb9a1091f1a9763a8c69307cff2d1bc8d/gpu/command_buffer/client/gles2_implementation.cc
[modify] https://crrev.com/01a6817cb9a1091f1a9763a8c69307cff2d1bc8d/gpu/command_buffer/client/query_tracker.cc
[modify] https://crrev.com/01a6817cb9a1091f1a9763a8c69307cff2d1bc8d/gpu/command_buffer/client/query_tracker.h
[modify] https://crrev.com/01a6817cb9a1091f1a9763a8c69307cff2d1bc8d/gpu/command_buffer/client/query_tracker_unittest.cc
[modify] https://crrev.com/01a6817cb9a1091f1a9763a8c69307cff2d1bc8d/gpu/command_buffer/client/raster_implementation.cc
[modify] https://crrev.com/01a6817cb9a1091f1a9763a8c69307cff2d1bc8d/gpu/command_buffer/common/gles2_cmd_utils_implementation_autogen.h
[modify] https://crrev.com/01a6817cb9a1091f1a9763a8c69307cff2d1bc8d/gpu/command_buffer/service/gles2_cmd_decoder.cc
[modify] https://crrev.com/01a6817cb9a1091f1a9763a8c69307cff2d1bc8d/gpu/command_buffer/service/gles2_cmd_validation_implementation_autogen.h
[modify] https://crrev.com/01a6817cb9a1091f1a9763a8c69307cff2d1bc8d/gpu/command_buffer/service/query_manager.cc
[modify] https://crrev.com/01a6817cb9a1091f1a9763a8c69307cff2d1bc8d/gpu/command_buffer/service/query_manager.h
[modify] https://crrev.com/01a6817cb9a1091f1a9763a8c69307cff2d1bc8d/gpu/command_buffer/service/raster_cmd_validation_implementation_autogen.h
[modify] https://crrev.com/01a6817cb9a1091f1a9763a8c69307cff2d1bc8d/gpu/command_buffer/service/raster_decoder.cc
[modify] https://crrev.com/01a6817cb9a1091f1a9763a8c69307cff2d1bc8d/tools/metrics/histograms/histograms.xml
[modify] https://crrev.com/01a6817cb9a1091f1a9763a8c69307cff2d1bc8d/ui/gl/gl_bindings.h

Comment 3 by kbr@chromium.org, Oct 16

Blockedon: 895765

Comment 4 by kbr@chromium.org, Oct 16 

Please note that https://chromium-review.googlesource.com/1267235 also caused somewhat reliable failures in the context_lost tests per  Issue 895765 . On Windows Debug the following stack trace was seen:

  Stack Trace:
  ********************************************************************************
  	Last event: 1160.1b74: Access violation - code c0000005 (first/second chance not available)
  	  debugger time: Mon Oct 15 23:05:01.906 2018 (UTC - 7:00)
  	ChildEBP RetAddr  Args to Child              
  	1377e108 5d7f854c 1377e154 5d7f08a0 1377e13c cc!cc::TileManager::CheckRasterFinishedQueries+0x5d
  	1377e118 5d7f848f 5d7f08a0 1377e154 1377e154 cc!??$Invoke@P8TileManager@cc@@AEXXZPAV12@$$V@?$FunctorTraits@P8TileManager@cc@@AEXXZX@internal@base@@SAXP8TileManager@cc@@AEXXZ$$QAPAV34@@Z+0x1c
  	1377e13c 5d7f8425 0ec2da4c 1377e154 0ec2da4c cc!base::internal::InvokeHelper<0,void>::MakeItSo<void (__thiscall cc::TileManager::*const &)(void),cc::TileManager *>+0x4f
  	1377e15c 5d7f830f 0ec2da4c 0ec2da50 00000084 cc!base::internal::Invoker<base::internal::BindState<void (__thiscall cc::TileManager::*)(void),base::internal::UnretainedWrapper<cc::TileManager> >,void __cdecl(void)>::RunImpl<void (__thiscall cc::TileManager::*const &)(void),std::tuple<base::internal::UnretainedWrapper<cc::TileManager> > const &,0>+0x55
  	1377e184 5d62f122 0ec2da38 5d7f82d0 0e9d3310 cc!base::internal::Invoker<base::internal::BindState<void (__thiscall cc::TileManager::*)(void),base::internal::UnretainedWrapper<cc::TileManager> >,void __cdecl(void)>::Run+0x3f
  	1377e1a0 5d62efbf 0e9d3310 1377e1bc 5d62f411 cc!base::RepeatingCallback<void __cdecl(void)>::Run+0x32
  	1377e1ac 5d62f411 0ec70d80 5d62efb0 1377e1e0 cc!??$ForwardRepeating@$$V@?$CancelableCallbackImpl@V?$RepeatingCallback@$$A6AXXZ@base@@@internal@base@@AAEXXZ+0xf
  	1377e1bc 5d62f326 5d62efb0 0ec70d80 0ec70d80 cc!??$Invoke@P8?$CancelableCallbackImpl@V?$RepeatingCallback@$$A6AXXZ@base@@@internal@base@@AEXXZABV?$WeakPtr@V?$CancelableCallbackImpl@V?$RepeatingCallback@$$A6AXXZ@base@@@internal@base@@@3@$$V@?$FunctorTraits@P8?$CancelableCallbackImpl@V?$RepeatingCallback@$$A6AXXZ@base@@@internal@base@@AEXXZX@internal@base@@SAXP8?$CancelableCallbackImpl@V?$RepeatingCallback@$$A6AXXZ@base@@@12@AEXXZABV?$WeakPtr@V?$CancelableCallbackImpl@V?$RepeatingCallback@$$A6AXXZ@base@@@internal@base@@@2@@Z+0x21
  	1377e1e0 5d62f2ba 0ec70d7c 0ec70d80 0ec70d80 cc!??$MakeItSo@ABQ8?$CancelableCallbackImpl@V?$RepeatingCallback@$$A6AXXZ@base@@@internal@base@@AEXXZABV?$WeakPtr@V?$CancelableCallbackImpl@V?$RepeatingCallback@$$A6AXXZ@base@@@internal@base@@@3@$$V@?$InvokeHelper@$00X@internal@base@@SAXABQ8?$CancelableCallbackImpl@V?$RepeatingCallback@$$A6AXXZ@base@@@12@AEXXZABV?$WeakPtr@V?$CancelableCallbackImpl@V?$RepeatingCallback@$$A6AXXZ@base@@@internal@base@@@2@@Z+0x66
 

Please investigate this before re-landing. Thanks.

Comment 5 by khushals...@chromium.org, Oct 16 

I see what's going on here. CheckRasterFinishedQueries assumes we always have a raster buffer provider but when we lose the context, we give up all the resources and destroy the raster buffer provider in TileManager. That cleanup work also invalidates any weak ptrs to TileManager callbacks depending on these resources but I didn't add cancelling this callback in that step. Thanks gpu tests!
Project Member

Comment 6 by bugdroid1@chromium.org, Oct 16 

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/cd8fbb77cfa164eef66378204a588443af9bb875

commit cd8fbb77cfa164eef66378204a588443af9bb875
Author: Khushal <khushalsagar@chromium.org>
Date: Tue Oct 16 22:46:14 2018

Reland gpu: Add UMA metric for measuring total raster task duration.

Add an UMA metric which reports the accumulated time for a raster task
including the time from raster worker thread and the GPU main thread.
The GPU side timer takes care of only measuring time slices when the
decoder is processing commands.

TBR=holte@chromium.org, piman@chromium.org, sunnyps@chromium.org

BUG:  894200 
Cq-Include-Trybots: luci.chromium.try:android_optional_gpu_tests_rel;luci.chromium.try:linux_optional_gpu_tests_rel;luci.chromium.try:mac_optional_gpu_tests_rel;luci.chromium.try:win_optional_gpu_tests_rel;master.tryserver.blink:linux_trusty_blink_rel
Change-Id: I7ed515daba4b5bdc701a20e5ff33dc770d170b2b
Reviewed-on: https://chromium-review.googlesource.com/c/1283487
Commit-Queue: Khushal <khushalsagar@chromium.org>
Reviewed-by: Khushal <khushalsagar@chromium.org>
Cr-Commit-Position: refs/heads/master@{#600158}
[modify] https://crrev.com/cd8fbb77cfa164eef66378204a588443af9bb875/cc/raster/bitmap_raster_buffer_provider.cc
[modify] https://crrev.com/cd8fbb77cfa164eef66378204a588443af9bb875/cc/raster/bitmap_raster_buffer_provider.h
[modify] https://crrev.com/cd8fbb77cfa164eef66378204a588443af9bb875/cc/raster/gpu_raster_buffer_provider.cc
[modify] https://crrev.com/cd8fbb77cfa164eef66378204a588443af9bb875/cc/raster/gpu_raster_buffer_provider.h
[modify] https://crrev.com/cd8fbb77cfa164eef66378204a588443af9bb875/cc/raster/one_copy_raster_buffer_provider.cc
[modify] https://crrev.com/cd8fbb77cfa164eef66378204a588443af9bb875/cc/raster/one_copy_raster_buffer_provider.h
[modify] https://crrev.com/cd8fbb77cfa164eef66378204a588443af9bb875/cc/raster/raster_buffer_provider.h
[modify] https://crrev.com/cd8fbb77cfa164eef66378204a588443af9bb875/cc/raster/raster_buffer_provider_unittest.cc
[modify] https://crrev.com/cd8fbb77cfa164eef66378204a588443af9bb875/cc/raster/zero_copy_raster_buffer_provider.cc
[modify] https://crrev.com/cd8fbb77cfa164eef66378204a588443af9bb875/cc/raster/zero_copy_raster_buffer_provider.h
[modify] https://crrev.com/cd8fbb77cfa164eef66378204a588443af9bb875/cc/test/cc_test_suite.cc
[modify] https://crrev.com/cd8fbb77cfa164eef66378204a588443af9bb875/cc/test/fake_raster_buffer_provider.cc
[modify] https://crrev.com/cd8fbb77cfa164eef66378204a588443af9bb875/cc/test/fake_raster_buffer_provider.h
[modify] https://crrev.com/cd8fbb77cfa164eef66378204a588443af9bb875/cc/tiles/tile_manager.cc
[modify] https://crrev.com/cd8fbb77cfa164eef66378204a588443af9bb875/cc/tiles/tile_manager.h
[modify] https://crrev.com/cd8fbb77cfa164eef66378204a588443af9bb875/cc/tiles/tile_manager_unittest.cc
[modify] https://crrev.com/cd8fbb77cfa164eef66378204a588443af9bb875/gpu/GLES2/gl2extchromium.h
[modify] https://crrev.com/cd8fbb77cfa164eef66378204a588443af9bb875/gpu/command_buffer/build_gles2_cmd_buffer.py
[modify] https://crrev.com/cd8fbb77cfa164eef66378204a588443af9bb875/gpu/command_buffer/build_raster_cmd_buffer.py
[modify] https://crrev.com/cd8fbb77cfa164eef66378204a588443af9bb875/gpu/command_buffer/client/gles2_implementation.cc
[modify] https://crrev.com/cd8fbb77cfa164eef66378204a588443af9bb875/gpu/command_buffer/client/query_tracker.cc
[modify] https://crrev.com/cd8fbb77cfa164eef66378204a588443af9bb875/gpu/command_buffer/client/query_tracker.h
[modify] https://crrev.com/cd8fbb77cfa164eef66378204a588443af9bb875/gpu/command_buffer/client/query_tracker_unittest.cc
[modify] https://crrev.com/cd8fbb77cfa164eef66378204a588443af9bb875/gpu/command_buffer/client/raster_implementation.cc
[modify] https://crrev.com/cd8fbb77cfa164eef66378204a588443af9bb875/gpu/command_buffer/common/gles2_cmd_utils_implementation_autogen.h
[modify] https://crrev.com/cd8fbb77cfa164eef66378204a588443af9bb875/gpu/command_buffer/service/gles2_cmd_decoder.cc
[modify] https://crrev.com/cd8fbb77cfa164eef66378204a588443af9bb875/gpu/command_buffer/service/gles2_cmd_validation_implementation_autogen.h
[modify] https://crrev.com/cd8fbb77cfa164eef66378204a588443af9bb875/gpu/command_buffer/service/query_manager.cc
[modify] https://crrev.com/cd8fbb77cfa164eef66378204a588443af9bb875/gpu/command_buffer/service/query_manager.h
[modify] https://crrev.com/cd8fbb77cfa164eef66378204a588443af9bb875/gpu/command_buffer/service/raster_cmd_validation_implementation_autogen.h
[modify] https://crrev.com/cd8fbb77cfa164eef66378204a588443af9bb875/gpu/command_buffer/service/raster_decoder.cc
[modify] https://crrev.com/cd8fbb77cfa164eef66378204a588443af9bb875/tools/metrics/histograms/histograms.xml
[modify] https://crrev.com/cd8fbb77cfa164eef66378204a588443af9bb875/ui/gl/gl_bindings.h
Project Member

Comment 7 by bugdroid1@chromium.org, Oct 17 

The following revision refers to this bug:
  https://skia.googlesource.com/skia/+/38c7a2e7d5bb770e65244a1b901986485b203986

commit 38c7a2e7d5bb770e65244a1b901986485b203986
Author: Khushal <khushalsagar@chromium.org>
Date: Wed Oct 17 16:18:37 2018

fonts: Cache serialized typeface wire data for remote fonts.

The serialization showed up as a significant cost on profiling,
primarily because of the lock acquired for query the glyph count of a
typeface and the SkData allocation each time. Avoid this by caching the
data for each typeface.

Note that since the typeface is already cached on the client side, we
shouldn't need to resend the typeface data each time. But serializing a
typeface doesn't guarentee that it will be sent, because of which we
can not make any assumptions about client state after serialize is
called.

Bug:  894200 
Change-Id: I45e43e297d817826646ab08bd4937b2625603858
Reviewed-on: https://skia-review.googlesource.com/c/162841
Reviewed-by: Herb Derby <herb@google.com>
Commit-Queue: Khusal Sagar <khushalsagar@chromium.org>
Auto-Submit: Khusal Sagar <khushalsagar@chromium.org>

[modify] https://crrev.com/38c7a2e7d5bb770e65244a1b901986485b203986/src/core/SkRemoteGlyphCache.h
[modify] https://crrev.com/38c7a2e7d5bb770e65244a1b901986485b203986/src/core/SkRemoteGlyphCache.cpp
Project Member

Comment 8 by bugdroid1@chromium.org, Oct 17 

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/b0e3983f7a7a814c285c2302b16ffd345ba37815

commit b0e3983f7a7a814c285c2302b16ffd345ba37815
Author: chromium-autoroll <chromium-autoroll@skia-public.iam.gserviceaccount.com>
Date: Wed Oct 17 18:47:56 2018

Roll src/third_party/skia f20c3493eb57..e92aae6b3612 (6 commits)

https://skia.googlesource.com/skia.git/+log/f20c3493eb57..e92aae6b3612


git log f20c3493eb57..e92aae6b3612 --date=short --no-merges --format='%ad %ae %s'
2018-10-17 reed@google.com make legacy shadercontext optional
2018-10-17 skia-autoroll@skia-public.iam.gserviceaccount.com Roll third_party/externals/angle2 1d8a783c9db2..c20ead32ad6d (2 commits)
2018-10-17 khushalsagar@chromium.org fonts: Cache serialized typeface wire data for remote fonts.
2018-10-17 herb@google.com Fix logic error for GrTextBlob mustRegen
2018-10-17 skia-autoroll@skia-public.iam.gserviceaccount.com Roll skia/third_party/skcms fa2161cf791f..c851ac96663d (4 commits)
2018-10-17 skia-recreate-skps@skia-swarming-bots.iam.gserviceaccount.com Update go_deps asset


Created with:
  gclient setdep -r src/third_party/skia@e92aae6b3612

The AutoRoll server is located here: https://autoroll.skia.org/r/skia-autoroll

Documentation for the AutoRoller is here:
https://skia.googlesource.com/buildbot/+/master/autoroll/README.md

If the roll is causing failures, please contact the current sheriff, who should
be CC'd on the roll, and stop the roller if necessary.

CQ_INCLUDE_TRYBOTS=luci.chromium.try:android_optional_gpu_tests_rel;luci.chromium.try:linux-chromeos-compile-dbg;luci.chromium.try:linux_optional_gpu_tests_rel;luci.chromium.try:mac_optional_gpu_tests_rel;luci.chromium.try:win_optional_gpu_tests_rel;master.tryserver.blink:linux_trusty_blink_rel

BUG= chromium:894200 
TBR=egdaniel@chromium.org

Change-Id: I03baf792befb769de37ac16efc3537325c358819
Reviewed-on: https://chromium-review.googlesource.com/c/1286112
Reviewed-by: chromium-autoroll <chromium-autoroll@skia-public.iam.gserviceaccount.com>
Commit-Queue: chromium-autoroll <chromium-autoroll@skia-public.iam.gserviceaccount.com>
Cr-Commit-Position: refs/heads/master@{#600498}
[modify] https://crrev.com/b0e3983f7a7a814c285c2302b16ffd345ba37815/DEPS

Comment 9 by khushals...@chromium.org, Oct 17

Summary: Tracking bug for OOPR serialization improvements and metrics. (was: Add UMA metrics for measuring total raster duration with GPU raster.)
Project Member

Comment 11 by 42576172...@developer.gserviceaccount.com, Oct 18 

📍 Job complete. See results below.
https://pinpoint-dot-chromeperf.appspot.com/job/140494f4e40000
Project Member

Comment 13 by 42576172...@developer.gserviceaccount.com, Oct 18 

📍 Job complete. See results below.
https://pinpoint-dot-chromeperf.appspot.com/job/13a9a45ce40000
Project Member

Comment 14 by bugdroid1@chromium.org, Oct 18 

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/5d4e09620ff79ac668c72ad3677f993b75e617ce

commit 5d4e09620ff79ac668c72ad3677f993b75e617ce
Author: Khushal <khushalsagar@chromium.org>
Date: Thu Oct 18 18:04:07 2018

cc: Avoid acquiring the context lock if unnecessary.

Acquiring the context lock marks the context as busy and can delay
cleanup of gpu memory. Avoid acquiring it if there are no pending raster
queries.

R=ericrk@chromium.org

Bug: 896260,  894200 
Cq-Include-Trybots: luci.chromium.try:android_optional_gpu_tests_rel;master.tryserver.blink:linux_trusty_blink_rel
Change-Id: I91b9ea85ef11b0b31e9d14599800efb0d2ca00e0
Reviewed-on: https://chromium-review.googlesource.com/c/1287931
Commit-Queue: Khushal <khushalsagar@chromium.org>
Reviewed-by: Eric Karl <ericrk@chromium.org>
Cr-Commit-Position: refs/heads/master@{#600824}
[modify] https://crrev.com/5d4e09620ff79ac668c72ad3677f993b75e617ce/cc/raster/gpu_raster_buffer_provider.cc
[modify] https://crrev.com/5d4e09620ff79ac668c72ad3677f993b75e617ce/cc/raster/gpu_raster_buffer_provider.h

Comment 15 by khushals...@chromium.org, Oct 18

Labels: Merge-Request-71
Making a merge request for the change in #6 and #14. I'll wait for 14 to go out in today's canary before merging. Its good to merge by tomorrow so this makes it in the next 71 dev release.
Project Member

Comment 16 by sheriffbot@chromium.org, Oct 18

Labels: -Merge-Request-71 Hotlist-Merge-Review Merge-Review-71
This bug requires manual review: DEPS changes referenced in bugdroid comments.
Please contact the milestone owner if you have questions.
Owners: benmason@(Android), kariahda@(iOS), kbleicher@(ChromeOS), govind@(Desktop)

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot

Comment 17 by khushals...@chromium.org, Oct 18

Cc: gov...@chromium.org

Comment 18 by gov...@chromium.org, Oct 18

NextAction: 2018-10-19
Pls update bug with canary result tomorrow morning.

Comment 20 by khushals...@chromium.org, Oct 18 

Yet another thing I missed which will have to merge: https://chromium-review.googlesource.com/c/chromium/src/+/1289958. :/
Project Member

Comment 21 by bugdroid1@chromium.org, Oct 19 

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/da82bcce69bf5384435976a1945c9cb3270c5abb

commit da82bcce69bf5384435976a1945c9cb3270c5abb
Author: Khushal <khushalsagar@chromium.org>
Date: Fri Oct 19 01:01:30 2018

gpu: Fix raster implementation error for no flush result available query

The enum GL_QUERY_RESULT_AVAILABLE_NO_FLUSH_CHROMIUM_EXT was not being
handled, and as a result we were doing a check before the result was
available and causing a flush.

R=piman@chromium.org

Bug:  896830 , 894200 
Cq-Include-Trybots: luci.chromium.try:android_optional_gpu_tests_rel;luci.chromium.try:linux_optional_gpu_tests_rel;luci.chromium.try:mac_optional_gpu_tests_rel;luci.chromium.try:win_optional_gpu_tests_rel
Change-Id: Ic4dc1ec22b2768d7d47ee63c61cc410554be563e
Reviewed-on: https://chromium-review.googlesource.com/c/1289958
Reviewed-by: Antoine Labour <piman@chromium.org>
Commit-Queue: Khushal <khushalsagar@chromium.org>
Cr-Commit-Position: refs/heads/master@{#601003}
[modify] https://crrev.com/da82bcce69bf5384435976a1945c9cb3270c5abb/gpu/command_buffer/client/raster_implementation.cc
[modify] https://crrev.com/da82bcce69bf5384435976a1945c9cb3270c5abb/gpu/command_buffer/client/raster_implementation_unittest.cc
Project Member

Comment 22 by 42576172...@developer.gserviceaccount.com, Oct 19 

📍 Job complete. See results below.
https://pinpoint-dot-chromeperf.appspot.com/job/16263bd4e40000

Comment 23 by monor...@bugs.chromium.org, Oct 19 

The NextAction date has arrived: 2018-10-19

Comment 24 by gov...@chromium.org, Oct 19 

This is not a bug, Type: Task. 
Could you pls justify why we need these merges to M71 and how safe they are?

Comment 25 by khushals...@chromium.org, Oct 19

NextAction: 2018-10-22
Thanks for the ping Krishna. The merge is for a metric to make sure we are measuring the impact of a feature being rolled out in 71. The patch is mostly safe but since I've had a few perf regressions from it, I'll give it another few days for all perf bots to go through it and ensure its okay to merge. Lets check back again on Monday.
Project Member

Comment 26 by bugdroid1@chromium.org, Oct 19 

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/e898b99eac0cf5db379cf8c7e216d772095a7448

commit e898b99eac0cf5db379cf8c7e216d772095a7448
Author: Khushal <khushalsagar@chromium.org>
Date: Fri Oct 19 22:25:16 2018

cc: Rate limit raster metric queries.

Since queries for measuring GPU side raster duration is too expensive,
rate limit the tiles for which we measure this metric to 1%.

R=piman@chromium.org

Bug:  894200 ,896821
Change-Id: I0713e537447a8ca7ebec30747314810e1a881d6e
Reviewed-on: https://chromium-review.googlesource.com/c/1292110
Reviewed-by: Antoine Labour <piman@chromium.org>
Commit-Queue: Khushal <khushalsagar@chromium.org>
Cr-Commit-Position: refs/heads/master@{#601319}
[modify] https://crrev.com/e898b99eac0cf5db379cf8c7e216d772095a7448/cc/raster/gpu_raster_buffer_provider.cc
[modify] https://crrev.com/e898b99eac0cf5db379cf8c7e216d772095a7448/cc/raster/gpu_raster_buffer_provider.h
[modify] https://crrev.com/e898b99eac0cf5db379cf8c7e216d772095a7448/cc/raster/raster_buffer_provider_unittest.cc
Project Member

Comment 27 by bugdroid1@chromium.org, Oct 19 

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/464808c91afd87d4531e1073d6de03367b383e7c

commit 464808c91afd87d4531e1073d6de03367b383e7c
Author: Khushal <khushalsagar@chromium.org>
Date: Fri Oct 19 23:45:59 2018

cc/paint: Remove paint wrapping of SkTextBlob and friends.

This was initially added to ensure that the typefaces embedded in a blob
were stored in a sidelist on the blob, in order to enable serialization
of typefaces for OOP raster. This is no longer required and the wrapping
is just unnecessary overhead.

R=drott@chromium.org,enne@chromium.org

Bug:  816445 , 894200 
Cq-Include-Trybots: luci.chromium.try:android_optional_gpu_tests_rel;luci.chromium.try:linux_layout_tests_layout_ng;luci.chromium.try:linux_layout_tests_slimming_paint_v2;master.tryserver.blink:linux_trusty_blink_rel
Change-Id: I76f106d13ce1aa04a50aadb0d4faa98dafe7a688
Reviewed-on: https://chromium-review.googlesource.com/c/1289571
Commit-Queue: Khushal <khushalsagar@chromium.org>
Reviewed-by: enne <enne@chromium.org>
Reviewed-by: Michael Wasserman <msw@chromium.org>
Reviewed-by: Chris Harrelson <chrishtr@chromium.org>
Reviewed-by: Dominik Röttsches <drott@chromium.org>
Cr-Commit-Position: refs/heads/master@{#601359}
[modify] https://crrev.com/464808c91afd87d4531e1073d6de03367b383e7c/cc/paint/BUILD.gn
[modify] https://crrev.com/464808c91afd87d4531e1073d6de03367b383e7c/cc/paint/oop_pixeltest.cc
[modify] https://crrev.com/464808c91afd87d4531e1073d6de03367b383e7c/cc/paint/paint_canvas.h
[delete] https://crrev.com/31ba7eade907515bc50df8e564c75372b1b2f969/cc/paint/paint_font.cc
[delete] https://crrev.com/31ba7eade907515bc50df8e564c75372b1b2f969/cc/paint/paint_font.h
[modify] https://crrev.com/464808c91afd87d4531e1073d6de03367b383e7c/cc/paint/paint_op_buffer.cc
[modify] https://crrev.com/464808c91afd87d4531e1073d6de03367b383e7c/cc/paint/paint_op_buffer.h
[modify] https://crrev.com/464808c91afd87d4531e1073d6de03367b383e7c/cc/paint/paint_op_buffer_unittest.cc
[modify] https://crrev.com/464808c91afd87d4531e1073d6de03367b383e7c/cc/paint/paint_op_perftest.cc
[modify] https://crrev.com/464808c91afd87d4531e1073d6de03367b383e7c/cc/paint/paint_op_reader.cc
[modify] https://crrev.com/464808c91afd87d4531e1073d6de03367b383e7c/cc/paint/paint_op_reader.h
[modify] https://crrev.com/464808c91afd87d4531e1073d6de03367b383e7c/cc/paint/paint_op_writer.cc
[modify] https://crrev.com/464808c91afd87d4531e1073d6de03367b383e7c/cc/paint/paint_op_writer.h
[delete] https://crrev.com/31ba7eade907515bc50df8e564c75372b1b2f969/cc/paint/paint_text_blob.cc
[delete] https://crrev.com/31ba7eade907515bc50df8e564c75372b1b2f969/cc/paint/paint_text_blob.h
[delete] https://crrev.com/31ba7eade907515bc50df8e564c75372b1b2f969/cc/paint/paint_text_blob_builder.cc
[delete] https://crrev.com/31ba7eade907515bc50df8e564c75372b1b2f969/cc/paint/paint_text_blob_builder.h
[modify] https://crrev.com/464808c91afd87d4531e1073d6de03367b383e7c/cc/paint/record_paint_canvas.cc
[modify] https://crrev.com/464808c91afd87d4531e1073d6de03367b383e7c/cc/paint/record_paint_canvas.h
[modify] https://crrev.com/464808c91afd87d4531e1073d6de03367b383e7c/cc/paint/skia_paint_canvas.cc
[modify] https://crrev.com/464808c91afd87d4531e1073d6de03367b383e7c/cc/paint/skia_paint_canvas.h
[modify] https://crrev.com/464808c91afd87d4531e1073d6de03367b383e7c/cc/test/paint_op_helper.h
[modify] https://crrev.com/464808c91afd87d4531e1073d6de03367b383e7c/third_party/blink/public/platform/web_font_render_style.h
[modify] https://crrev.com/464808c91afd87d4531e1073d6de03367b383e7c/third_party/blink/renderer/platform/BUILD.gn
[modify] https://crrev.com/464808c91afd87d4531e1073d6de03367b383e7c/third_party/blink/renderer/platform/fonts/font.cc
[modify] https://crrev.com/464808c91afd87d4531e1073d6de03367b383e7c/third_party/blink/renderer/platform/fonts/font_platform_data.cc
[modify] https://crrev.com/464808c91afd87d4531e1073d6de03367b383e7c/third_party/blink/renderer/platform/fonts/font_platform_data.h
[modify] https://crrev.com/464808c91afd87d4531e1073d6de03367b383e7c/third_party/blink/renderer/platform/fonts/mac/font_platform_data_mac.mm
[delete] https://crrev.com/31ba7eade907515bc50df8e564c75372b1b2f969/third_party/blink/renderer/platform/fonts/paint_font.h
[delete] https://crrev.com/31ba7eade907515bc50df8e564c75372b1b2f969/third_party/blink/renderer/platform/fonts/paint_text_blob.h
[modify] https://crrev.com/464808c91afd87d4531e1073d6de03367b383e7c/third_party/blink/renderer/platform/fonts/shaping/harfbuzz_face.cc
[modify] https://crrev.com/464808c91afd87d4531e1073d6de03367b383e7c/third_party/blink/renderer/platform/fonts/shaping/shape_result_bloberizer.cc
[modify] https://crrev.com/464808c91afd87d4531e1073d6de03367b383e7c/third_party/blink/renderer/platform/fonts/shaping/shape_result_bloberizer.h
[modify] https://crrev.com/464808c91afd87d4531e1073d6de03367b383e7c/third_party/blink/renderer/platform/fonts/simple_font_data.cc
[modify] https://crrev.com/464808c91afd87d4531e1073d6de03367b383e7c/third_party/blink/renderer/platform/fonts/web_font_render_style.cc
[modify] https://crrev.com/464808c91afd87d4531e1073d6de03367b383e7c/third_party/blink/renderer/platform/fonts/win/font_platform_data_win.cc
[modify] https://crrev.com/464808c91afd87d4531e1073d6de03367b383e7c/third_party/blink/renderer/platform/graphics/test/mock_paint_canvas.h
[modify] https://crrev.com/464808c91afd87d4531e1073d6de03367b383e7c/ui/gfx/render_text.cc
Project Member

Comment 29 by 42576172...@developer.gserviceaccount.com, Oct 20 

😿 Pinpoint job stopped with an error.
https://pinpoint-dot-chromeperf.appspot.com/job/1599a45ce40000

Buildbucket says the build completed successfully, but Pinpoint can't find the isolate hash.
Project Member

Comment 30 by bugdroid1@chromium.org, Oct 20 

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/57df06767601e98a1b51239d3a5f3862e3d000dd

commit 57df06767601e98a1b51239d3a5f3862e3d000dd
Author: Khushal <khushalsagar@chromium.org>
Date: Sat Oct 20 04:34:07 2018

cc: Cache text blobs in the GPU service.

Since skia's internal caching is keyed on text blobs, its important to
reuse the same object. Make sure we cache text blobs in the
ServiceTransferCache to get a cache hit in skia.

R=enne@chromium.org

Bug:  894200 
Change-Id: I8808d649ca3520211d034e07e7d4670ffb39094c
Reviewed-on: https://chromium-review.googlesource.com/c/1292722
Commit-Queue: Khushal <khushalsagar@chromium.org>
Reviewed-by: enne <enne@chromium.org>
Cr-Commit-Position: refs/heads/master@{#601404}
[modify] https://crrev.com/57df06767601e98a1b51239d3a5f3862e3d000dd/cc/paint/BUILD.gn
[modify] https://crrev.com/57df06767601e98a1b51239d3a5f3862e3d000dd/cc/paint/paint_op_reader.cc
[modify] https://crrev.com/57df06767601e98a1b51239d3a5f3862e3d000dd/cc/paint/paint_op_writer.cc
[add] https://crrev.com/57df06767601e98a1b51239d3a5f3862e3d000dd/cc/paint/textblob_transfer_cache_entry.cc
[add] https://crrev.com/57df06767601e98a1b51239d3a5f3862e3d000dd/cc/paint/textblob_transfer_cache_entry.h
[modify] https://crrev.com/57df06767601e98a1b51239d3a5f3862e3d000dd/cc/paint/transfer_cache_entry.cc
[modify] https://crrev.com/57df06767601e98a1b51239d3a5f3862e3d000dd/cc/paint/transfer_cache_entry.h
Project Member

Comment 32 by 42576172...@developer.gserviceaccount.com, Oct 21 

😿 Pinpoint job stopped with an error.
https://pinpoint-dot-chromeperf.appspot.com/job/10ece69ce40000

Buildbucket says the build completed successfully, but Pinpoint can't find the isolate hash.
Project Member

Comment 34 by 42576172...@developer.gserviceaccount.com, Oct 21 

😿 Pinpoint job stopped with an error.
https://pinpoint-dot-chromeperf.appspot.com/job/15ad1132e40000

Buildbucket says the build completed successfully, but Pinpoint can't find the isolate hash.

Comment 35 by jmad...@chromium.org, Oct 21

Blockedon: 897507
Project Member

Comment 36 by bugdroid1@chromium.org, Oct 21 

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/e99280223d68ac7886976c45c562ce4dcbd58d54

commit e99280223d68ac7886976c45c562ce4dcbd58d54
Author: Jamie Madill <jmadill@chromium.org>
Date: Sun Oct 21 15:26:29 2018

Revert "cc: Cache text blobs in the GPU service."

This reverts commit 57df06767601e98a1b51239d3a5f3862e3d000dd.

Reason for revert: Seems to be causing a crash on Mac.

https://ci.chromium.org/p/chromium/builders/luci.chromium.ci/Mac%20FYI%2010.14%20Release%20%28Intel%29/327
https://ci.chromium.org/p/chromium/builders/luci.chromium.ci/Mac%20FYI%20Experimental%20Release%20%28Intel%29/6072

Crash in __ZN13GrCCPathCache5evictEP18GrCCPathCacheEntry

Bug:  897507 

Original change's description:
> cc: Cache text blobs in the GPU service.
> 
> Since skia's internal caching is keyed on text blobs, its important to
> reuse the same object. Make sure we cache text blobs in the
> ServiceTransferCache to get a cache hit in skia.
> 
> R=​enne@chromium.org
> 
> Bug:  894200 
> Change-Id: I8808d649ca3520211d034e07e7d4670ffb39094c
> Reviewed-on: https://chromium-review.googlesource.com/c/1292722
> Commit-Queue: Khushal <khushalsagar@chromium.org>
> Reviewed-by: enne <enne@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#601404}

TBR=enne@chromium.org,khushalsagar@chromium.org

# Not skipping CQ checks because original CL landed > 1 day ago.

Bug:  894200 
Change-Id: I5e0a151c5a539aa8a9b6cac5f9bcf710ba2f405a
Reviewed-on: https://chromium-review.googlesource.com/c/1292384
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Cr-Commit-Position: refs/heads/master@{#601451}
[modify] https://crrev.com/e99280223d68ac7886976c45c562ce4dcbd58d54/cc/paint/BUILD.gn
[modify] https://crrev.com/e99280223d68ac7886976c45c562ce4dcbd58d54/cc/paint/paint_op_reader.cc
[modify] https://crrev.com/e99280223d68ac7886976c45c562ce4dcbd58d54/cc/paint/paint_op_writer.cc
[delete] https://crrev.com/d2a371815c3881a02c05c2de63f33daf1f9bd4af/cc/paint/textblob_transfer_cache_entry.cc
[delete] https://crrev.com/d2a371815c3881a02c05c2de63f33daf1f9bd4af/cc/paint/textblob_transfer_cache_entry.h
[modify] https://crrev.com/e99280223d68ac7886976c45c562ce4dcbd58d54/cc/paint/transfer_cache_entry.cc
[modify] https://crrev.com/e99280223d68ac7886976c45c562ce4dcbd58d54/cc/paint/transfer_cache_entry.h

Comment 37 by monor...@bugs.chromium.org, Oct 22 

The NextAction date has arrived: 2018-10-22

Comment 38 by khushals...@chromium.org, Oct 22 

All the telemetry regressions are addressed and the changes have been in canary for a few days. I think it should be safe to merge now.
And these are the changes that will need to be merged (https://chromium-review.googlesource.com/c/chromium/src/+/1283487, https://chromium-review.googlesource.com/c/chromium/src/+/1287931, https://chromium-review.googlesource.com/c/chromium/src/+/1289958 and https://chromium-review.googlesource.com/c/chromium/src/+/1292110).

Comment 40 by khushals...@chromium.org, Oct 22 

Chatted with govind@ offline.

Why multiple changes need merge?
The main change for merging is https://chromium-review.googlesource.com/c/chromium/src/+/1283487. The rest are minor follow up fixes from telemetry regressions. These have gone through the perf bots now and everything is resolved.

How critical are they?
This is an important metric to understand the impact of OOP raster in M71 and required for the experiment on stable.

And safety?
This is only adding a new metric. And the changes have gone through enough perf testing to flush out all problems.

govind@ approved the merges on chat.
Project Member

Comment 41 by bugdroid1@chromium.org, Oct 22

Labels: merge-merged-3578
The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/94af0f596f7950f0cfdc752170ed6391b6e1e5c4

commit 94af0f596f7950f0cfdc752170ed6391b6e1e5c4
Author: Khushal <khushalsagar@chromium.org>
Date: Mon Oct 22 18:19:20 2018

Reland gpu: Add UMA metric for measuring total raster task duration.

Add an UMA metric which reports the accumulated time for a raster task
including the time from raster worker thread and the GPU main thread.
The GPU side timer takes care of only measuring time slices when the
decoder is processing commands.

TBR=holte@chromium.org, piman@chromium.org, sunnyps@chromium.org

BUG:  894200 
Cq-Include-Trybots: luci.chromium.try:android_optional_gpu_tests_rel;luci.chromium.try:linux_optional_gpu_tests_rel;luci.chromium.try:mac_optional_gpu_tests_rel;luci.chromium.try:win_optional_gpu_tests_rel;master.tryserver.blink:linux_trusty_blink_rel
Change-Id: I7ed515daba4b5bdc701a20e5ff33dc770d170b2b
Reviewed-on: https://chromium-review.googlesource.com/c/1283487
Commit-Queue: Khushal <khushalsagar@chromium.org>
Reviewed-by: Khushal <khushalsagar@chromium.org>
Cr-Original-Commit-Position: refs/heads/master@{#600158}(cherry picked from commit cd8fbb77cfa164eef66378204a588443af9bb875)
Reviewed-on: https://chromium-review.googlesource.com/c/1294393
Cr-Commit-Position: refs/branch-heads/3578@{#227}
Cr-Branched-From: 4226ddf99103e493d7afb23a4c7902ee496108b6-refs/heads/master@{#599034}
[modify] https://crrev.com/94af0f596f7950f0cfdc752170ed6391b6e1e5c4/cc/raster/bitmap_raster_buffer_provider.cc
[modify] https://crrev.com/94af0f596f7950f0cfdc752170ed6391b6e1e5c4/cc/raster/bitmap_raster_buffer_provider.h
[modify] https://crrev.com/94af0f596f7950f0cfdc752170ed6391b6e1e5c4/cc/raster/gpu_raster_buffer_provider.cc
[modify] https://crrev.com/94af0f596f7950f0cfdc752170ed6391b6e1e5c4/cc/raster/gpu_raster_buffer_provider.h
[modify] https://crrev.com/94af0f596f7950f0cfdc752170ed6391b6e1e5c4/cc/raster/one_copy_raster_buffer_provider.cc
[modify] https://crrev.com/94af0f596f7950f0cfdc752170ed6391b6e1e5c4/cc/raster/one_copy_raster_buffer_provider.h
[modify] https://crrev.com/94af0f596f7950f0cfdc752170ed6391b6e1e5c4/cc/raster/raster_buffer_provider.h
[modify] https://crrev.com/94af0f596f7950f0cfdc752170ed6391b6e1e5c4/cc/raster/raster_buffer_provider_unittest.cc
[modify] https://crrev.com/94af0f596f7950f0cfdc752170ed6391b6e1e5c4/cc/raster/zero_copy_raster_buffer_provider.cc
[modify] https://crrev.com/94af0f596f7950f0cfdc752170ed6391b6e1e5c4/cc/raster/zero_copy_raster_buffer_provider.h
[modify] https://crrev.com/94af0f596f7950f0cfdc752170ed6391b6e1e5c4/cc/test/cc_test_suite.cc
[modify] https://crrev.com/94af0f596f7950f0cfdc752170ed6391b6e1e5c4/cc/test/fake_raster_buffer_provider.cc
[modify] https://crrev.com/94af0f596f7950f0cfdc752170ed6391b6e1e5c4/cc/test/fake_raster_buffer_provider.h
[modify] https://crrev.com/94af0f596f7950f0cfdc752170ed6391b6e1e5c4/cc/tiles/tile_manager.cc
[modify] https://crrev.com/94af0f596f7950f0cfdc752170ed6391b6e1e5c4/cc/tiles/tile_manager.h
[modify] https://crrev.com/94af0f596f7950f0cfdc752170ed6391b6e1e5c4/cc/tiles/tile_manager_unittest.cc
[modify] https://crrev.com/94af0f596f7950f0cfdc752170ed6391b6e1e5c4/gpu/GLES2/gl2extchromium.h
[modify] https://crrev.com/94af0f596f7950f0cfdc752170ed6391b6e1e5c4/gpu/command_buffer/build_gles2_cmd_buffer.py
[modify] https://crrev.com/94af0f596f7950f0cfdc752170ed6391b6e1e5c4/gpu/command_buffer/build_raster_cmd_buffer.py
[modify] https://crrev.com/94af0f596f7950f0cfdc752170ed6391b6e1e5c4/gpu/command_buffer/client/gles2_implementation.cc
[modify] https://crrev.com/94af0f596f7950f0cfdc752170ed6391b6e1e5c4/gpu/command_buffer/client/query_tracker.cc
[modify] https://crrev.com/94af0f596f7950f0cfdc752170ed6391b6e1e5c4/gpu/command_buffer/client/query_tracker.h
[modify] https://crrev.com/94af0f596f7950f0cfdc752170ed6391b6e1e5c4/gpu/command_buffer/client/query_tracker_unittest.cc
[modify] https://crrev.com/94af0f596f7950f0cfdc752170ed6391b6e1e5c4/gpu/command_buffer/client/raster_implementation.cc
[modify] https://crrev.com/94af0f596f7950f0cfdc752170ed6391b6e1e5c4/gpu/command_buffer/common/gles2_cmd_utils_implementation_autogen.h
[modify] https://crrev.com/94af0f596f7950f0cfdc752170ed6391b6e1e5c4/gpu/command_buffer/service/gles2_cmd_decoder.cc
[modify] https://crrev.com/94af0f596f7950f0cfdc752170ed6391b6e1e5c4/gpu/command_buffer/service/gles2_cmd_validation_implementation_autogen.h
[modify] https://crrev.com/94af0f596f7950f0cfdc752170ed6391b6e1e5c4/gpu/command_buffer/service/query_manager.cc
[modify] https://crrev.com/94af0f596f7950f0cfdc752170ed6391b6e1e5c4/gpu/command_buffer/service/query_manager.h
[modify] https://crrev.com/94af0f596f7950f0cfdc752170ed6391b6e1e5c4/gpu/command_buffer/service/raster_cmd_validation_implementation_autogen.h
[modify] https://crrev.com/94af0f596f7950f0cfdc752170ed6391b6e1e5c4/gpu/command_buffer/service/raster_decoder.cc
[modify] https://crrev.com/94af0f596f7950f0cfdc752170ed6391b6e1e5c4/tools/metrics/histograms/histograms.xml
[modify] https://crrev.com/94af0f596f7950f0cfdc752170ed6391b6e1e5c4/ui/gl/gl_bindings.h
Project Member

Comment 42 by bugdroid1@chromium.org, Oct 22 

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/2ed1221f50733c2b53fa6270d0aeea7c97a88475

commit 2ed1221f50733c2b53fa6270d0aeea7c97a88475
Author: Khushal <khushalsagar@chromium.org>
Date: Mon Oct 22 18:19:43 2018

cc: Avoid acquiring the context lock if unnecessary.

Acquiring the context lock marks the context as busy and can delay
cleanup of gpu memory. Avoid acquiring it if there are no pending raster
queries.

R=​ericrk@chromium.org

Bug: 896260,  894200 
Cq-Include-Trybots: luci.chromium.try:android_optional_gpu_tests_rel;master.tryserver.blink:linux_trusty_blink_rel
Change-Id: I91b9ea85ef11b0b31e9d14599800efb0d2ca00e0
Reviewed-on: https://chromium-review.googlesource.com/c/1287931
Commit-Queue: Khushal <khushalsagar@chromium.org>
Reviewed-by: Eric Karl <ericrk@chromium.org>
Cr-Original-Commit-Position: refs/heads/master@{#600824}(cherry picked from commit 5d4e09620ff79ac668c72ad3677f993b75e617ce)
Reviewed-on: https://chromium-review.googlesource.com/c/1294394
Reviewed-by: Khushal <khushalsagar@chromium.org>
Cr-Commit-Position: refs/branch-heads/3578@{#228}
Cr-Branched-From: 4226ddf99103e493d7afb23a4c7902ee496108b6-refs/heads/master@{#599034}
[modify] https://crrev.com/2ed1221f50733c2b53fa6270d0aeea7c97a88475/cc/raster/gpu_raster_buffer_provider.cc
[modify] https://crrev.com/2ed1221f50733c2b53fa6270d0aeea7c97a88475/cc/raster/gpu_raster_buffer_provider.h
Project Member

Comment 43 by bugdroid1@chromium.org, Oct 22 

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/fa6aa1fe175cee9ca0db6e9f142b2f1b4c7c97e6

commit fa6aa1fe175cee9ca0db6e9f142b2f1b4c7c97e6
Author: Khushal <khushalsagar@chromium.org>
Date: Mon Oct 22 18:20:12 2018

gpu: Fix raster implementation error for no flush result available query

The enum GL_QUERY_RESULT_AVAILABLE_NO_FLUSH_CHROMIUM_EXT was not being
handled, and as a result we were doing a check before the result was
available and causing a flush.

R=​piman@chromium.org

Bug:  896830 , 894200 
Cq-Include-Trybots: luci.chromium.try:android_optional_gpu_tests_rel;luci.chromium.try:linux_optional_gpu_tests_rel;luci.chromium.try:mac_optional_gpu_tests_rel;luci.chromium.try:win_optional_gpu_tests_rel
Change-Id: Ic4dc1ec22b2768d7d47ee63c61cc410554be563e
Reviewed-on: https://chromium-review.googlesource.com/c/1289958
Reviewed-by: Antoine Labour <piman@chromium.org>
Commit-Queue: Khushal <khushalsagar@chromium.org>
Cr-Original-Commit-Position: refs/heads/master@{#601003}(cherry picked from commit da82bcce69bf5384435976a1945c9cb3270c5abb)
Reviewed-on: https://chromium-review.googlesource.com/c/1294396
Reviewed-by: Khushal <khushalsagar@chromium.org>
Cr-Commit-Position: refs/branch-heads/3578@{#229}
Cr-Branched-From: 4226ddf99103e493d7afb23a4c7902ee496108b6-refs/heads/master@{#599034}
[modify] https://crrev.com/fa6aa1fe175cee9ca0db6e9f142b2f1b4c7c97e6/gpu/command_buffer/client/raster_implementation.cc
[modify] https://crrev.com/fa6aa1fe175cee9ca0db6e9f142b2f1b4c7c97e6/gpu/command_buffer/client/raster_implementation_unittest.cc
Project Member

Comment 44 by bugdroid1@chromium.org, Oct 22 

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/b81eb60bb4b9483e5c41834a143a02f6f5132176

commit b81eb60bb4b9483e5c41834a143a02f6f5132176
Author: Khushal <khushalsagar@chromium.org>
Date: Mon Oct 22 18:20:31 2018

cc: Rate limit raster metric queries.

Since queries for measuring GPU side raster duration is too expensive,
rate limit the tiles for which we measure this metric to 1%.

R=​piman@chromium.org

Bug:  894200 ,896821
Change-Id: I0713e537447a8ca7ebec30747314810e1a881d6e
Reviewed-on: https://chromium-review.googlesource.com/c/1292110
Reviewed-by: Antoine Labour <piman@chromium.org>
Commit-Queue: Khushal <khushalsagar@chromium.org>
Cr-Original-Commit-Position: refs/heads/master@{#601319}(cherry picked from commit e898b99eac0cf5db379cf8c7e216d772095a7448)
Reviewed-on: https://chromium-review.googlesource.com/c/1294397
Reviewed-by: Khushal <khushalsagar@chromium.org>
Cr-Commit-Position: refs/branch-heads/3578@{#230}
Cr-Branched-From: 4226ddf99103e493d7afb23a4c7902ee496108b6-refs/heads/master@{#599034}
[modify] https://crrev.com/b81eb60bb4b9483e5c41834a143a02f6f5132176/cc/raster/gpu_raster_buffer_provider.cc
[modify] https://crrev.com/b81eb60bb4b9483e5c41834a143a02f6f5132176/cc/raster/gpu_raster_buffer_provider.h
[modify] https://crrev.com/b81eb60bb4b9483e5c41834a143a02f6f5132176/cc/raster/raster_buffer_provider_unittest.cc

Comment 45 by khushals...@chromium.org, Oct 22

Labels: -ReleaseBlock-Stable

Comment 46 by gov...@chromium.org, Oct 22

Labels: Merge-Approved-71
Applying "Merge-Approved-71" label per comment #40.

Comment 47 by gov...@chromium.org, Oct 22

Labels: -Merge-Approved-71 -Merge-Review-71
And removing "Merge-Approved-71" as this is already merged to M71.
Project Member

Comment 48 by 42576172...@developer.gserviceaccount.com, Oct 22 

😿 Pinpoint job stopped with an error.
https://pinpoint-dot-chromeperf.appspot.com/job/12e13c06e40000

Buildbucket says the build completed successfully, but Pinpoint can't find the isolate hash.
Project Member

Comment 49 by bugdroid1@chromium.org, Oct 22 

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/bd34e88ade5706c80b60b78ead85af8a251d390a

commit bd34e88ade5706c80b60b78ead85af8a251d390a
Author: Khushal <khushalsagar@chromium.org>
Date: Mon Oct 22 20:46:08 2018

cc: Ensure bitmap backed images are tracked as discardable.

Since GPU raster uses the GpuImageDecodeCache for image uploads,
ensuring that the image is tracked as discardable avoids at-raster
caching for them which has more ref/unref overhead.

R=ericrk@chromium.org

Bug:  894200 
Change-Id: I9d40ab5e0d7c2c2e4a12d7ab8be318cfcba2f590
Reviewed-on: https://chromium-review.googlesource.com/c/1292813
Reviewed-by: Eric Karl <ericrk@chromium.org>
Commit-Queue: Khushal <khushalsagar@chromium.org>
Cr-Commit-Position: refs/heads/master@{#601717}
[modify] https://crrev.com/bd34e88ade5706c80b60b78ead85af8a251d390a/cc/layers/recording_source_unittest.cc
[modify] https://crrev.com/bd34e88ade5706c80b60b78ead85af8a251d390a/cc/paint/discardable_image_map.cc
[modify] https://crrev.com/bd34e88ade5706c80b60b78ead85af8a251d390a/cc/paint/oop_pixeltest.cc
[modify] https://crrev.com/bd34e88ade5706c80b60b78ead85af8a251d390a/cc/paint/paint_filter.cc
[modify] https://crrev.com/bd34e88ade5706c80b60b78ead85af8a251d390a/cc/paint/paint_filter_unittest.cc
[modify] https://crrev.com/bd34e88ade5706c80b60b78ead85af8a251d390a/cc/paint/paint_image.h
[modify] https://crrev.com/bd34e88ade5706c80b60b78ead85af8a251d390a/cc/paint/paint_op_buffer.cc
[modify] https://crrev.com/bd34e88ade5706c80b60b78ead85af8a251d390a/cc/paint/paint_shader.cc
[modify] https://crrev.com/bd34e88ade5706c80b60b78ead85af8a251d390a/cc/test/skia_common.cc
[modify] https://crrev.com/bd34e88ade5706c80b60b78ead85af8a251d390a/cc/test/skia_common.h

Comment 50 by gov...@chromium.org, Oct 22 

Is cl listed at $49 need a merge to M71 once baked in canary?

Comment 51 by khushals...@chromium.org, Oct 22 

Nope, the changes associated with adding a new metric have been merged. No other change needs a merge for 71.

Comment 52 by gov...@chromium.org, Oct 22 

Thank you for confirmation.
Project Member

Comment 53 by cr-audit...@appspot.gserviceaccount.com, Oct 23

Labels: Merge-Merged-71-3578
The following revision refers to this bug: 
https://chromium.googlesource.com/chromium/src.git/+/2ed1221f50733c2b53fa6270d0aeea7c97a88475

Commit: 2ed1221f50733c2b53fa6270d0aeea7c97a88475
Author: khushalsagar@chromium.org
Commiter: khushalsagar@chromium.org
Date: 2018-10-22 18:19:43 +0000 UTC

cc: Avoid acquiring the context lock if unnecessary.

Acquiring the context lock marks the context as busy and can delay
cleanup of gpu memory. Avoid acquiring it if there are no pending raster
queries.

R=​ericrk@chromium.org

Bug: 896260,  894200 
Cq-Include-Trybots: luci.chromium.try:android_optional_gpu_tests_rel;master.tryserver.blink:linux_trusty_blink_rel
Change-Id: I91b9ea85ef11b0b31e9d14599800efb0d2ca00e0
Reviewed-on: https://chromium-review.googlesource.com/c/1287931
Commit-Queue: Khushal <khushalsagar@chromium.org>
Reviewed-by: Eric Karl <ericrk@chromium.org>
Cr-Original-Commit-Position: refs/heads/master@{#600824}(cherry picked from commit 5d4e09620ff79ac668c72ad3677f993b75e617ce)
Reviewed-on: https://chromium-review.googlesource.com/c/1294394
Reviewed-by: Khushal <khushalsagar@chromium.org>
Cr-Commit-Position: refs/branch-heads/3578@{#228}
Cr-Branched-From: 4226ddf99103e493d7afb23a4c7902ee496108b6-refs/heads/master@{#599034}
Project Member

Comment 54 by cr-audit...@appspot.gserviceaccount.com, Oct 23 

The following revision refers to this bug: 
https://chromium.googlesource.com/chromium/src.git/+/94af0f596f7950f0cfdc752170ed6391b6e1e5c4

Commit: 94af0f596f7950f0cfdc752170ed6391b6e1e5c4
Author: khushalsagar@chromium.org
Commiter: khushalsagar@chromium.org
Date: 2018-10-22 18:19:20 +0000 UTC

Reland gpu: Add UMA metric for measuring total raster task duration.

Add an UMA metric which reports the accumulated time for a raster task
including the time from raster worker thread and the GPU main thread.
The GPU side timer takes care of only measuring time slices when the
decoder is processing commands.

TBR=holte@chromium.org, piman@chromium.org, sunnyps@chromium.org

BUG:  894200 
Cq-Include-Trybots: luci.chromium.try:android_optional_gpu_tests_rel;luci.chromium.try:linux_optional_gpu_tests_rel;luci.chromium.try:mac_optional_gpu_tests_rel;luci.chromium.try:win_optional_gpu_tests_rel;master.tryserver.blink:linux_trusty_blink_rel
Change-Id: I7ed515daba4b5bdc701a20e5ff33dc770d170b2b
Reviewed-on: https://chromium-review.googlesource.com/c/1283487
Commit-Queue: Khushal <khushalsagar@chromium.org>
Reviewed-by: Khushal <khushalsagar@chromium.org>
Cr-Original-Commit-Position: refs/heads/master@{#600158}(cherry picked from commit cd8fbb77cfa164eef66378204a588443af9bb875)
Reviewed-on: https://chromium-review.googlesource.com/c/1294393
Cr-Commit-Position: refs/branch-heads/3578@{#227}
Cr-Branched-From: 4226ddf99103e493d7afb23a4c7902ee496108b6-refs/heads/master@{#599034}
Project Member

Comment 55 by cr-audit...@appspot.gserviceaccount.com, Oct 23 

The following revision refers to this bug: 
https://chromium.googlesource.com/chromium/src.git/+/b81eb60bb4b9483e5c41834a143a02f6f5132176

Commit: b81eb60bb4b9483e5c41834a143a02f6f5132176
Author: khushalsagar@chromium.org
Commiter: khushalsagar@chromium.org
Date: 2018-10-22 18:20:31 +0000 UTC

cc: Rate limit raster metric queries.

Since queries for measuring GPU side raster duration is too expensive,
rate limit the tiles for which we measure this metric to 1%.

R=​piman@chromium.org

Bug:  894200 ,896821
Change-Id: I0713e537447a8ca7ebec30747314810e1a881d6e
Reviewed-on: https://chromium-review.googlesource.com/c/1292110
Reviewed-by: Antoine Labour <piman@chromium.org>
Commit-Queue: Khushal <khushalsagar@chromium.org>
Cr-Original-Commit-Position: refs/heads/master@{#601319}(cherry picked from commit e898b99eac0cf5db379cf8c7e216d772095a7448)
Reviewed-on: https://chromium-review.googlesource.com/c/1294397
Reviewed-by: Khushal <khushalsagar@chromium.org>
Cr-Commit-Position: refs/branch-heads/3578@{#230}
Cr-Branched-From: 4226ddf99103e493d7afb23a4c7902ee496108b6-refs/heads/master@{#599034}
Project Member

Comment 56 by cr-audit...@appspot.gserviceaccount.com, Oct 23 

The following revision refers to this bug: 
https://chromium.googlesource.com/chromium/src.git/+/fa6aa1fe175cee9ca0db6e9f142b2f1b4c7c97e6

Commit: fa6aa1fe175cee9ca0db6e9f142b2f1b4c7c97e6
Author: khushalsagar@chromium.org
Commiter: khushalsagar@chromium.org
Date: 2018-10-22 18:20:12 +0000 UTC

gpu: Fix raster implementation error for no flush result available query

The enum GL_QUERY_RESULT_AVAILABLE_NO_FLUSH_CHROMIUM_EXT was not being
handled, and as a result we were doing a check before the result was
available and causing a flush.

R=​piman@chromium.org

Bug:  896830 , 894200 
Cq-Include-Trybots: luci.chromium.try:android_optional_gpu_tests_rel;luci.chromium.try:linux_optional_gpu_tests_rel;luci.chromium.try:mac_optional_gpu_tests_rel;luci.chromium.try:win_optional_gpu_tests_rel
Change-Id: Ic4dc1ec22b2768d7d47ee63c61cc410554be563e
Reviewed-on: https://chromium-review.googlesource.com/c/1289958
Reviewed-by: Antoine Labour <piman@chromium.org>
Commit-Queue: Khushal <khushalsagar@chromium.org>
Cr-Original-Commit-Position: refs/heads/master@{#601003}(cherry picked from commit da82bcce69bf5384435976a1945c9cb3270c5abb)
Reviewed-on: https://chromium-review.googlesource.com/c/1294396
Reviewed-by: Khushal <khushalsagar@chromium.org>
Cr-Commit-Position: refs/branch-heads/3578@{#229}
Cr-Branched-From: 4226ddf99103e493d7afb23a4c7902ee496108b6-refs/heads/master@{#599034}

Comment 57 by kbr@chromium.org, Oct 23

Blockedon: 898292
Project Member

Comment 58 by bugdroid1@chromium.org, Oct 24 

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/8ae18f7b2f42e8528aef24f434db91fb3e15d112

commit 8ae18f7b2f42e8528aef24f434db91fb3e15d112
Author: Adrienne Walker <enne@chromium.org>
Date: Wed Oct 24 02:46:59 2018

gpu: Don't reset state on more raster decoder commands

In looking at some sampling profiles, it turned out that these commands
(set/reset url and the query commands) were all causing unnecessary
RestoreState in RasterDecoder.  As these don't touch gl state,
whitelist them so that they don't cause extra work.

Bug:  894200 
Change-Id: I54837830d7df3b5333b3a3e8e42d9232ef8e8b41
Reviewed-on: https://chromium-review.googlesource.com/c/1297592
Commit-Queue: enne <enne@chromium.org>
Reviewed-by: Antoine Labour <piman@chromium.org>
Cr-Commit-Position: refs/heads/master@{#602226}
[modify] https://crrev.com/8ae18f7b2f42e8528aef24f434db91fb3e15d112/gpu/command_buffer/service/raster_decoder.cc
Project Member

Comment 59 by bugdroid1@chromium.org, Oct 24 

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/02acf1bba75bd6d345b8592c4a9b0ce61e88be56

commit 02acf1bba75bd6d345b8592c4a9b0ce61e88be56
Author: Khushal <khushalsagar@chromium.org>
Date: Wed Oct 24 19:59:07 2018

Reland "cc: Cache text blobs in the GPU service."

This reverts commit e99280223d68ac7886976c45c562ce4dcbd58d54.

Reason for revert: Speculative fixes have landed in skia.

Original change's description:
> Revert "cc: Cache text blobs in the GPU service."
> 
> This reverts commit 57df06767601e98a1b51239d3a5f3862e3d000dd.
> 
> Reason for revert: Seems to be causing a crash on Mac.
> 
> https://ci.chromium.org/p/chromium/builders/luci.chromium.ci/Mac%20FYI%2010.14%20Release%20%28Intel%29/327
> https://ci.chromium.org/p/chromium/builders/luci.chromium.ci/Mac%20FYI%20Experimental%20Release%20%28Intel%29/6072
> 
> Crash in __ZN13GrCCPathCache5evictEP18GrCCPathCacheEntry
> 
> Bug:  897507 
> 
> Original change's description:
> > cc: Cache text blobs in the GPU service.
> > 
> > Since skia's internal caching is keyed on text blobs, its important to
> > reuse the same object. Make sure we cache text blobs in the
> > ServiceTransferCache to get a cache hit in skia.
> > 
> > R=​enne@chromium.org
> > 
> > Bug:  894200 
> > Change-Id: I8808d649ca3520211d034e07e7d4670ffb39094c
> > Reviewed-on: https://chromium-review.googlesource.com/c/1292722
> > Commit-Queue: Khushal <khushalsagar@chromium.org>
> > Reviewed-by: enne <enne@chromium.org>
> > Cr-Commit-Position: refs/heads/master@{#601404}
> 
> TBR=enne@chromium.org,khushalsagar@chromium.org
> 
> # Not skipping CQ checks because original CL landed > 1 day ago.
> 
> Bug:  894200 
> Change-Id: I5e0a151c5a539aa8a9b6cac5f9bcf710ba2f405a
> Reviewed-on: https://chromium-review.googlesource.com/c/1292384
> Reviewed-by: Jamie Madill <jmadill@chromium.org>
> Commit-Queue: Jamie Madill <jmadill@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#601451}

TBR=jmadill@chromium.org,enne@chromium.org,khushalsagar@chromium.org

# Not skipping CQ checks because original CL landed > 1 day ago.

Bug:  897507 ,  894200 
Change-Id: I345adbbfaff91ab201d68e73e861dd4ccce41775
Reviewed-on: https://chromium-review.googlesource.com/c/1298235
Reviewed-by: Khushal <khushalsagar@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Khushal <khushalsagar@chromium.org>
Cr-Commit-Position: refs/heads/master@{#602406}
[modify] https://crrev.com/02acf1bba75bd6d345b8592c4a9b0ce61e88be56/cc/paint/BUILD.gn
[modify] https://crrev.com/02acf1bba75bd6d345b8592c4a9b0ce61e88be56/cc/paint/paint_op_reader.cc
[modify] https://crrev.com/02acf1bba75bd6d345b8592c4a9b0ce61e88be56/cc/paint/paint_op_writer.cc
[add] https://crrev.com/02acf1bba75bd6d345b8592c4a9b0ce61e88be56/cc/paint/textblob_transfer_cache_entry.cc
[add] https://crrev.com/02acf1bba75bd6d345b8592c4a9b0ce61e88be56/cc/paint/textblob_transfer_cache_entry.h
[modify] https://crrev.com/02acf1bba75bd6d345b8592c4a9b0ce61e88be56/cc/paint/transfer_cache_entry.cc
[modify] https://crrev.com/02acf1bba75bd6d345b8592c4a9b0ce61e88be56/cc/paint/transfer_cache_entry.h
Project Member

Comment 61 by 42576172...@developer.gserviceaccount.com, Oct 25 

📍 Job complete. See results below.
https://pinpoint-dot-chromeperf.appspot.com/job/10ac8081e40000
Project Member

Comment 62 by bugdroid1@chromium.org, Oct 25 

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/2fdde1d37db07cdb8313802d22fd43e95b11bac4

commit 2fdde1d37db07cdb8313802d22fd43e95b11bac4
Author: Adrienne Walker <enne@chromium.org>
Date: Thu Oct 25 01:45:51 2018

cc: reduce allocations for rtrees in oop-r

To avoid repeatedly allocating a vector for the results in the rtree
results during oop-r, hold onto this vector between raster commands.
This was showing up minorly in profiles and is an easy alloc to remove.

Bug:  894200 
Change-Id: Ia3f283c9ce20ad6db730c9b8589b030f0612e36d
Reviewed-on: https://chromium-review.googlesource.com/c/1297645
Commit-Queue: enne <enne@chromium.org>
Reviewed-by: Khushal <khushalsagar@chromium.org>
Reviewed-by: Antoine Labour <piman@chromium.org>
Cr-Commit-Position: refs/heads/master@{#602566}
[modify] https://crrev.com/2fdde1d37db07cdb8313802d22fd43e95b11bac4/cc/base/rtree.h
[modify] https://crrev.com/2fdde1d37db07cdb8313802d22fd43e95b11bac4/cc/base/rtree_perftest.cc
[modify] https://crrev.com/2fdde1d37db07cdb8313802d22fd43e95b11bac4/cc/base/rtree_unittest.cc
[modify] https://crrev.com/2fdde1d37db07cdb8313802d22fd43e95b11bac4/cc/paint/discardable_image_map.cc
[modify] https://crrev.com/2fdde1d37db07cdb8313802d22fd43e95b11bac4/cc/paint/discardable_image_map_unittest.cc
[modify] https://crrev.com/2fdde1d37db07cdb8313802d22fd43e95b11bac4/cc/paint/display_item_list.cc
[modify] https://crrev.com/2fdde1d37db07cdb8313802d22fd43e95b11bac4/gpu/command_buffer/client/raster_implementation.cc
[modify] https://crrev.com/2fdde1d37db07cdb8313802d22fd43e95b11bac4/gpu/command_buffer/client/raster_implementation.h
Project Member

Comment 63 by bugdroid1@chromium.org, Oct 25 

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/7c03b1e331419980ec21187b5b3878d04c22f8e6

commit 7c03b1e331419980ec21187b5b3878d04c22f8e6
Author: Khushal <khushalsagar@chromium.org>
Date: Thu Oct 25 06:57:24 2018

cc/gpu: Inline transfer cache entries instead of deferring them.

This change updates the approach for writing transfer cache entries into
the transfer buffer from defering the entry until the end of
RasterCHROMIUM to inlining them during the RasterCHROMIUM serialization
itself. It avoid an extra allocation and copy for these entries.

R=enne@chromium.org, ericrk@chromium.org

Bug:  894200 
Cq-Include-Trybots: luci.chromium.try:android_optional_gpu_tests_rel;luci.chromium.try:linux_optional_gpu_tests_rel;luci.chromium.try:mac_optional_gpu_tests_rel;luci.chromium.try:win_optional_gpu_tests_rel;master.tryserver.blink:linux_trusty_blink_rel
Change-Id: If12167d36ca074750f114e7af8c715d40a3a7dcc
Reviewed-on: https://chromium-review.googlesource.com/c/1287092
Commit-Queue: Khushal <khushalsagar@chromium.org>
Reviewed-by: Antoine Labour <piman@chromium.org>
Reviewed-by: enne <enne@chromium.org>
Cr-Commit-Position: refs/heads/master@{#602626}
[modify] https://crrev.com/7c03b1e331419980ec21187b5b3878d04c22f8e6/cc/paint/oop_pixeltest.cc
[modify] https://crrev.com/7c03b1e331419980ec21187b5b3878d04c22f8e6/cc/paint/paint_op_reader.cc
[modify] https://crrev.com/7c03b1e331419980ec21187b5b3878d04c22f8e6/cc/paint/paint_op_writer.cc
[modify] https://crrev.com/7c03b1e331419980ec21187b5b3878d04c22f8e6/cc/paint/path_transfer_cache_entry.cc
[modify] https://crrev.com/7c03b1e331419980ec21187b5b3878d04c22f8e6/cc/paint/raw_memory_transfer_cache_entry.cc
[modify] https://crrev.com/7c03b1e331419980ec21187b5b3878d04c22f8e6/cc/paint/transfer_cache_serialize_helper.cc
[modify] https://crrev.com/7c03b1e331419980ec21187b5b3878d04c22f8e6/cc/paint/transfer_cache_serialize_helper.h
[modify] https://crrev.com/7c03b1e331419980ec21187b5b3878d04c22f8e6/cc/test/transfer_cache_test_helper.cc
[modify] https://crrev.com/7c03b1e331419980ec21187b5b3878d04c22f8e6/cc/test/transfer_cache_test_helper.h
[modify] https://crrev.com/7c03b1e331419980ec21187b5b3878d04c22f8e6/gpu/command_buffer/client/client_transfer_cache.cc
[modify] https://crrev.com/7c03b1e331419980ec21187b5b3878d04c22f8e6/gpu/command_buffer/client/client_transfer_cache.h
[modify] https://crrev.com/7c03b1e331419980ec21187b5b3878d04c22f8e6/gpu/command_buffer/client/raster_implementation.cc
[modify] https://crrev.com/7c03b1e331419980ec21187b5b3878d04c22f8e6/gpu/command_buffer/client/raster_implementation.h
[modify] https://crrev.com/7c03b1e331419980ec21187b5b3878d04c22f8e6/gpu/command_buffer/client/raster_implementation_unittest.cc
[modify] https://crrev.com/7c03b1e331419980ec21187b5b3878d04c22f8e6/gpu/command_buffer/client/transfer_buffer.cc
[modify] https://crrev.com/7c03b1e331419980ec21187b5b3878d04c22f8e6/gpu/command_buffer/client/transfer_buffer.h
Project Member

Comment 64 by bugdroid1@chromium.org, Oct 26 

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/da1fb833922299d715bdb95a478c500040eb8676

commit da1fb833922299d715bdb95a478c500040eb8676
Author: Khushal <khushalsagar@chromium.org>
Date: Fri Oct 26 19:26:22 2018

gpu: Don't resend the last active URL for a context.

The transfer buffer allocation on the client and GURL allocation on the
service was showing up in profiles. Avoid that by caching the last
active URL on the client and only send an update if it changes. Also
hook this up for GPU raster.

R=vmiura@chromium.org

Bug:  894200 
Change-Id: If9519f3203a9f53ac16ccbeba685b239339aa5cc
Reviewed-on: https://chromium-review.googlesource.com/c/1298642
Reviewed-by: enne <enne@chromium.org>
Reviewed-by: Victor Miura <vmiura@chromium.org>
Commit-Queue: Khushal <khushalsagar@chromium.org>
Cr-Commit-Position: refs/heads/master@{#603165}
[modify] https://crrev.com/da1fb833922299d715bdb95a478c500040eb8676/components/viz/common/gpu/raster_context_provider.cc
[modify] https://crrev.com/da1fb833922299d715bdb95a478c500040eb8676/components/viz/common/gpu/raster_context_provider.h
[add] https://crrev.com/da1fb833922299d715bdb95a478c500040eb8676/gpu/GLES2/extensions/CHROMIUM/CHROMIUM_active_url.txt
[modify] https://crrev.com/da1fb833922299d715bdb95a478c500040eb8676/gpu/GLES2/gl2chromium_autogen.h
[modify] https://crrev.com/da1fb833922299d715bdb95a478c500040eb8676/gpu/command_buffer/build_gles2_cmd_buffer.py
[modify] https://crrev.com/da1fb833922299d715bdb95a478c500040eb8676/gpu/command_buffer/build_raster_cmd_buffer.py
[modify] https://crrev.com/da1fb833922299d715bdb95a478c500040eb8676/gpu/command_buffer/client/gles2_c_lib_autogen.h
[modify] https://crrev.com/da1fb833922299d715bdb95a478c500040eb8676/gpu/command_buffer/client/gles2_cmd_helper_autogen.h
[modify] https://crrev.com/da1fb833922299d715bdb95a478c500040eb8676/gpu/command_buffer/client/gles2_implementation.cc
[modify] https://crrev.com/da1fb833922299d715bdb95a478c500040eb8676/gpu/command_buffer/client/gles2_implementation.h
[modify] https://crrev.com/da1fb833922299d715bdb95a478c500040eb8676/gpu/command_buffer/client/gles2_implementation_autogen.h
[modify] https://crrev.com/da1fb833922299d715bdb95a478c500040eb8676/gpu/command_buffer/client/gles2_implementation_unittest.cc
[modify] https://crrev.com/da1fb833922299d715bdb95a478c500040eb8676/gpu/command_buffer/client/gles2_interface_autogen.h
[modify] https://crrev.com/da1fb833922299d715bdb95a478c500040eb8676/gpu/command_buffer/client/gles2_interface_stub_autogen.h
[modify] https://crrev.com/da1fb833922299d715bdb95a478c500040eb8676/gpu/command_buffer/client/gles2_interface_stub_impl_autogen.h
[modify] https://crrev.com/da1fb833922299d715bdb95a478c500040eb8676/gpu/command_buffer/client/gles2_trace_implementation_autogen.h
[modify] https://crrev.com/da1fb833922299d715bdb95a478c500040eb8676/gpu/command_buffer/client/gles2_trace_implementation_impl_autogen.h
[modify] https://crrev.com/da1fb833922299d715bdb95a478c500040eb8676/gpu/command_buffer/client/raster_cmd_helper_autogen.h
[modify] https://crrev.com/da1fb833922299d715bdb95a478c500040eb8676/gpu/command_buffer/client/raster_implementation.cc
[modify] https://crrev.com/da1fb833922299d715bdb95a478c500040eb8676/gpu/command_buffer/client/raster_implementation.h
[modify] https://crrev.com/da1fb833922299d715bdb95a478c500040eb8676/gpu/command_buffer/client/raster_implementation_autogen.h
[modify] https://crrev.com/da1fb833922299d715bdb95a478c500040eb8676/gpu/command_buffer/client/raster_implementation_gles.cc
[modify] https://crrev.com/da1fb833922299d715bdb95a478c500040eb8676/gpu/command_buffer/client/raster_implementation_gles.h
[modify] https://crrev.com/da1fb833922299d715bdb95a478c500040eb8676/gpu/command_buffer/client/raster_implementation_unittest.cc
[modify] https://crrev.com/da1fb833922299d715bdb95a478c500040eb8676/gpu/command_buffer/client/raster_interface_autogen.h
[modify] https://crrev.com/da1fb833922299d715bdb95a478c500040eb8676/gpu/command_buffer/common/gles2_cmd_format_autogen.h
[modify] https://crrev.com/da1fb833922299d715bdb95a478c500040eb8676/gpu/command_buffer/common/gles2_cmd_format_test_autogen.h
[modify] https://crrev.com/da1fb833922299d715bdb95a478c500040eb8676/gpu/command_buffer/common/gles2_cmd_ids_autogen.h
[modify] https://crrev.com/da1fb833922299d715bdb95a478c500040eb8676/gpu/command_buffer/common/raster_cmd_format_autogen.h
[modify] https://crrev.com/da1fb833922299d715bdb95a478c500040eb8676/gpu/command_buffer/common/raster_cmd_format_test_autogen.h
[modify] https://crrev.com/da1fb833922299d715bdb95a478c500040eb8676/gpu/command_buffer/common/raster_cmd_ids_autogen.h
[modify] https://crrev.com/da1fb833922299d715bdb95a478c500040eb8676/gpu/command_buffer/gles2_cmd_buffer_functions.txt
[modify] https://crrev.com/da1fb833922299d715bdb95a478c500040eb8676/gpu/command_buffer/raster_cmd_buffer_functions.txt
[modify] https://crrev.com/da1fb833922299d715bdb95a478c500040eb8676/gpu/command_buffer/service/decoder_client.h
[modify] https://crrev.com/da1fb833922299d715bdb95a478c500040eb8676/gpu/command_buffer/service/gles2_cmd_decoder.cc
[modify] https://crrev.com/da1fb833922299d715bdb95a478c500040eb8676/gpu/command_buffer/service/gles2_cmd_decoder_passthrough.cc
[modify] https://crrev.com/da1fb833922299d715bdb95a478c500040eb8676/gpu/command_buffer/service/raster_decoder.cc
[modify] https://crrev.com/da1fb833922299d715bdb95a478c500040eb8676/gpu/command_buffer/service/raster_decoder_autogen.h
[modify] https://crrev.com/da1fb833922299d715bdb95a478c500040eb8676/gpu/ipc/service/raster_command_buffer_stub.cc
[modify] https://crrev.com/da1fb833922299d715bdb95a478c500040eb8676/gpu/ipc/service/raster_command_buffer_stub.h

Comment 65 by gov...@chromium.org, Oct 26 

Are any more merges needed to M71?

Comment 66 by khushals...@chromium.org, Oct 26 

Nope, no merges needed for 71 anymore.
Project Member

Comment 68 by 42576172...@developer.gserviceaccount.com, Oct 29 

😿 Pinpoint job stopped with an error.
https://pinpoint-dot-chromeperf.appspot.com/job/14f9b755e40000

Buildbucket says the build completed successfully, but Pinpoint can't find the isolate hash.
Project Member

Comment 70 by 42576172...@developer.gserviceaccount.com, Oct 30 

📍 Job complete. See results below.
https://pinpoint-dot-chromeperf.appspot.com/job/14d9f3bde40000

Comment 72 by gov...@chromium.org, Oct 30

Labels: Merge-Approved-71
Approving merge to M71 branch 3578 based on comment #71. Pls merge before 1:00 PM PT, tomorrow (Tuesday) so we can pick it up for this week beta release. Thank you.
Project Member

Comment 73 by bugdroid1@chromium.org, Oct 30

Labels: -merge-approved-71
The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/3c169d2cc2b7aef6aca79cd84e72a8d8de49fbc1

commit 3c169d2cc2b7aef6aca79cd84e72a8d8de49fbc1
Author: Adrienne Walker <enne@chromium.org>
Date: Tue Oct 30 03:40:32 2018

gpu: Don't reset state on more raster decoder commands

In looking at some sampling profiles, it turned out that these commands
(set/reset url and the query commands) were all causing unnecessary
RestoreState in RasterDecoder.  As these don't touch gl state,
whitelist them so that they don't cause extra work.

Bug:  894200 
Change-Id: I54837830d7df3b5333b3a3e8e42d9232ef8e8b41
Reviewed-on: https://chromium-review.googlesource.com/c/1297592
Commit-Queue: enne <enne@chromium.org>
Reviewed-by: Antoine Labour <piman@chromium.org>
Cr-Original-Commit-Position: refs/heads/master@{#602226}(cherry picked from commit 8ae18f7b2f42e8528aef24f434db91fb3e15d112)
Reviewed-on: https://chromium-review.googlesource.com/c/1307063
Reviewed-by: Khushal <khushalsagar@chromium.org>
Cr-Commit-Position: refs/branch-heads/3578@{#382}
Cr-Branched-From: 4226ddf99103e493d7afb23a4c7902ee496108b6-refs/heads/master@{#599034}
[modify] https://crrev.com/3c169d2cc2b7aef6aca79cd84e72a8d8de49fbc1/gpu/command_buffer/service/raster_decoder.cc
Project Member

Comment 74 by cr-audit...@appspot.gserviceaccount.com, Oct 30 

The following revision refers to this bug: 
https://chromium.googlesource.com/chromium/src.git/+/3c169d2cc2b7aef6aca79cd84e72a8d8de49fbc1

Commit: 3c169d2cc2b7aef6aca79cd84e72a8d8de49fbc1
Author: enne@chromium.org
Commiter: khushalsagar@chromium.org
Date: 2018-10-30 03:40:32 +0000 UTC

gpu: Don't reset state on more raster decoder commands

In looking at some sampling profiles, it turned out that these commands
(set/reset url and the query commands) were all causing unnecessary
RestoreState in RasterDecoder.  As these don't touch gl state,
whitelist them so that they don't cause extra work.

Bug:  894200 
Change-Id: I54837830d7df3b5333b3a3e8e42d9232ef8e8b41
Reviewed-on: https://chromium-review.googlesource.com/c/1297592
Commit-Queue: enne <enne@chromium.org>
Reviewed-by: Antoine Labour <piman@chromium.org>
Cr-Original-Commit-Position: refs/heads/master@{#602226}(cherry picked from commit 8ae18f7b2f42e8528aef24f434db91fb3e15d112)
Reviewed-on: https://chromium-review.googlesource.com/c/1307063
Reviewed-by: Khushal <khushalsagar@chromium.org>
Cr-Commit-Position: refs/branch-heads/3578@{#382}
Cr-Branched-From: 4226ddf99103e493d7afb23a4c7902ee496108b6-refs/heads/master@{#599034}
Project Member

Comment 76 by bugdroid1@chromium.org, Oct 31 

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/7acd0617a387674dfdfc522d2a2c47fafc9edf0f

commit 7acd0617a387674dfdfc522d2a2c47fafc9edf0f
Author: Khushal <khushalsagar@chromium.org>
Date: Wed Oct 31 00:25:11 2018

gpu: Make the list of commands which require GL state restore explicit.

Make the list of commands which require consistent GL state in the
RasterDecoder explicit instead of defaulting to it for unhandled
commands. Its easy to add a new command and miss updating this list.
This makes sure you get a compile time error to avoid this.

R=piman@chromium.org

Bug:  894200 
Change-Id: I71407d33c3339342d9ebf26b6ce94e475324a37e
Reviewed-on: https://chromium-review.googlesource.com/c/1308714
Commit-Queue: Khushal <khushalsagar@chromium.org>
Reviewed-by: Antoine Labour <piman@chromium.org>
Cr-Commit-Position: refs/heads/master@{#604076}
[modify] https://crrev.com/7acd0617a387674dfdfc522d2a2c47fafc9edf0f/gpu/command_buffer/service/raster_decoder.cc

Comment 77 by gov...@chromium.org, Oct 31 

Is cl listed at #76 need a merge to M71?

Comment 78 by khushals...@chromium.org, Oct 31 

Nope. Its just a safeguard to avoid errors like these in the future.

Comment 79 by gov...@chromium.org, Oct 31 

Ok, got it. Thank you.
Project Member

Comment 80 by 42576172...@developer.gserviceaccount.com, Oct 31 

📍 Job complete. See results below.
https://pinpoint-dot-chromeperf.appspot.com/job/17e31fade40000
Project Member

Comment 81 by bugdroid1@chromium.org, Nov 6 

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/3c3873a6c34f6c8a4454f669351cd49e494e5b70

commit 3c3873a6c34f6c8a4454f669351cd49e494e5b70
Author: Khushal <khushalsagar@chromium.org>
Date: Tue Nov 06 18:14:01 2018

cc: Use uniform distribution for sampling raster metric.

We currently keep a counter on the raster buffer provider for sampling
this metric, which can result in bias in some cases.

R=vmiura@chromium.org

Bug:  894200 
Change-Id: I943afeaf5653ca30baee611f7d1f8ebb50a3030c
Reviewed-on: https://chromium-review.googlesource.com/c/1319029
Reviewed-by: Victor Miura <vmiura@chromium.org>
Commit-Queue: Khushal <khushalsagar@chromium.org>
Cr-Commit-Position: refs/heads/master@{#605739}
[modify] https://crrev.com/3c3873a6c34f6c8a4454f669351cd49e494e5b70/cc/raster/gpu_raster_buffer_provider.cc
[modify] https://crrev.com/3c3873a6c34f6c8a4454f669351cd49e494e5b70/cc/raster/gpu_raster_buffer_provider.h
Project Member

Comment 85 by 42576172...@developer.gserviceaccount.com, Nov 7 

📍 Job complete. See results below.
https://pinpoint-dot-chromeperf.appspot.com/job/1260597ee40000
Project Member

Comment 86 by 42576172...@developer.gserviceaccount.com, Nov 7 

📍 Job complete. See results below.
https://pinpoint-dot-chromeperf.appspot.com/job/131cedf5e40000
Project Member

Comment 87 by 42576172...@developer.gserviceaccount.com, Nov 7 

📍 Job complete. See results below.
https://pinpoint-dot-chromeperf.appspot.com/job/132e5a85e40000
Project Member

Comment 89 by 42576172...@developer.gserviceaccount.com, Nov 8 

📍 Job complete. See results below.
https://pinpoint-dot-chromeperf.appspot.com/job/16b376bae40000
Project Member

Comment 90 by bugdroid1@chromium.org, Nov 8 

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/33205a780d892f1ae0af639fd57a339f44093f9a

commit 33205a780d892f1ae0af639fd57a339f44093f9a
Author: Khushal <khushalsagar@chromium.org>
Date: Thu Nov 08 10:12:29 2018

cc/gpu: Add client driven paint caching for OOP raster.

Currently text blobs are cached in the GPU process in the
ServiceTransferCache during OOP raster. This approach has 2 shortcomings:

1) While it gives us cache hits in skia's internal caching keyed on these
blobs, we still reserialize the blobs each time. This is because the
client side tracking of an entry in ClientTransferCache can get more
expensive than serializing the entry itself for these small entries.

2) The ServiceTransferCache budget is based on the assumption that it will
be used for images, which is much larger than needed for this use-case.
And because the eviction is also only done when we hit the cache limit,
even if the renderer creating the entries is now invisible, the cache gets
filled with stale data that may have already been purged on the client
and will never be reused.

This change adds a new client driven cache for these paint data type (text
blobs and paths), with a much smaller budget and something that is cleared
during our normal context idle cleanup or on visibility change. Lookup and
additions to the cache are also slightly cheaper than transfer cache since
it avoids allocating discardable handles for cross process ref-counting
and the need to track entries which need to be unref-ed after every raster.

TBR=tsepez@chromium.org
R=enne@chromium.org, piman@chromium.org

Change-Id: Ibc4039ec5ab20e45c1ae67aec20367c20371458f
Bug: 898910, 894200 
Reviewed-on: https://chromium-review.googlesource.com/c/1321190
Commit-Queue: Khushal <khushalsagar@chromium.org>
Reviewed-by: Antoine Labour <piman@chromium.org>
Cr-Commit-Position: refs/heads/master@{#606400}
[modify] https://crrev.com/33205a780d892f1ae0af639fd57a339f44093f9a/cc/BUILD.gn
[modify] https://crrev.com/33205a780d892f1ae0af639fd57a339f44093f9a/cc/paint/BUILD.gn
[modify] https://crrev.com/33205a780d892f1ae0af639fd57a339f44093f9a/cc/paint/image_transfer_cache_entry.cc
[add] https://crrev.com/33205a780d892f1ae0af639fd57a339f44093f9a/cc/paint/paint_cache.cc
[add] https://crrev.com/33205a780d892f1ae0af639fd57a339f44093f9a/cc/paint/paint_cache.h
[add] https://crrev.com/33205a780d892f1ae0af639fd57a339f44093f9a/cc/paint/paint_cache_unittest.cc
[modify] https://crrev.com/33205a780d892f1ae0af639fd57a339f44093f9a/cc/paint/paint_op_buffer.cc
[modify] https://crrev.com/33205a780d892f1ae0af639fd57a339f44093f9a/cc/paint/paint_op_buffer.h
[modify] https://crrev.com/33205a780d892f1ae0af639fd57a339f44093f9a/cc/paint/paint_op_buffer_fuzzer.cc
[modify] https://crrev.com/33205a780d892f1ae0af639fd57a339f44093f9a/cc/paint/paint_op_buffer_serializer.cc
[modify] https://crrev.com/33205a780d892f1ae0af639fd57a339f44093f9a/cc/paint/paint_op_buffer_serializer.h
[modify] https://crrev.com/33205a780d892f1ae0af639fd57a339f44093f9a/cc/paint/paint_op_buffer_unittest.cc
[modify] https://crrev.com/33205a780d892f1ae0af639fd57a339f44093f9a/cc/paint/paint_op_perftest.cc
[modify] https://crrev.com/33205a780d892f1ae0af639fd57a339f44093f9a/cc/paint/paint_op_reader.cc
[modify] https://crrev.com/33205a780d892f1ae0af639fd57a339f44093f9a/cc/paint/paint_op_writer.cc
[modify] https://crrev.com/33205a780d892f1ae0af639fd57a339f44093f9a/cc/test/test_options_provider.cc
[modify] https://crrev.com/33205a780d892f1ae0af639fd57a339f44093f9a/cc/test/test_options_provider.h
[modify] https://crrev.com/33205a780d892f1ae0af639fd57a339f44093f9a/gpu/command_buffer/build_raster_cmd_buffer.py
[modify] https://crrev.com/33205a780d892f1ae0af639fd57a339f44093f9a/gpu/command_buffer/client/raster_cmd_helper_autogen.h
[modify] https://crrev.com/33205a780d892f1ae0af639fd57a339f44093f9a/gpu/command_buffer/client/raster_implementation.cc
[modify] https://crrev.com/33205a780d892f1ae0af639fd57a339f44093f9a/gpu/command_buffer/client/raster_implementation.h
[modify] https://crrev.com/33205a780d892f1ae0af639fd57a339f44093f9a/gpu/command_buffer/common/raster_cmd_format_autogen.h
[modify] https://crrev.com/33205a780d892f1ae0af639fd57a339f44093f9a/gpu/command_buffer/common/raster_cmd_format_test_autogen.h
[modify] https://crrev.com/33205a780d892f1ae0af639fd57a339f44093f9a/gpu/command_buffer/common/raster_cmd_ids_autogen.h
[modify] https://crrev.com/33205a780d892f1ae0af639fd57a339f44093f9a/gpu/command_buffer/raster_cmd_buffer_functions.txt
[modify] https://crrev.com/33205a780d892f1ae0af639fd57a339f44093f9a/gpu/command_buffer/service/raster_decoder.cc
[modify] https://crrev.com/33205a780d892f1ae0af639fd57a339f44093f9a/gpu/command_buffer/service/raster_decoder_autogen.h
[modify] https://crrev.com/33205a780d892f1ae0af639fd57a339f44093f9a/services/viz/public/cpp/compositing/paint_filter_struct_traits.cc
Project Member

Comment 92 by 42576172...@developer.gserviceaccount.com, Nov 8 

📍 Job complete. See results below.
https://pinpoint-dot-chromeperf.appspot.com/job/149a52abe40000

Comment 93 by khushals...@chromium.org, Nov 8 

One more 71 merge request for the change in #81. Its a minor update for sampling of the metric and is extremely safe.

Comment 94 by gov...@chromium.org, Nov 8 

Approving merge for cl listed at #81 to M71 branch 3578 based on comment #93. Pls merge ASAP. Thank you.
Project Member

Comment 95 by cr-audit...@appspot.gserviceaccount.com, Nov 8 

The following revision refers to this bug: 
https://chromium.googlesource.com/chromium/src.git/+/010af68a87603021daee029b7bd3b7c423183907

Commit: 010af68a87603021daee029b7bd3b7c423183907
Author: khushalsagar@chromium.org
Commiter: khushalsagar@chromium.org
Date: 2018-11-08 23:48:37 +0000 UTC

cc: Use uniform distribution for sampling raster metric.

We currently keep a counter on the raster buffer provider for sampling
this metric, which can result in bias in some cases.

R=​vmiura@chromium.org

Bug:  894200 
Change-Id: I943afeaf5653ca30baee611f7d1f8ebb50a3030c
Reviewed-on: https://chromium-review.googlesource.com/c/1319029
Reviewed-by: Victor Miura <vmiura@chromium.org>
Commit-Queue: Khushal <khushalsagar@chromium.org>
Cr-Original-Commit-Position: refs/heads/master@{#605739}(cherry picked from commit 3c3873a6c34f6c8a4454f669351cd49e494e5b70)
Reviewed-on: https://chromium-review.googlesource.com/c/1327539
Reviewed-by: Khushal <khushalsagar@chromium.org>
Cr-Commit-Position: refs/branch-heads/3578@{#594}
Cr-Branched-From: 4226ddf99103e493d7afb23a4c7902ee496108b6-refs/heads/master@{#599034}
Project Member

Comment 96 by bugdroid1@chromium.org, Nov 8 

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/010af68a87603021daee029b7bd3b7c423183907

commit 010af68a87603021daee029b7bd3b7c423183907
Author: Khushal <khushalsagar@chromium.org>
Date: Thu Nov 08 23:48:37 2018

cc: Use uniform distribution for sampling raster metric.

We currently keep a counter on the raster buffer provider for sampling
this metric, which can result in bias in some cases.

R=​vmiura@chromium.org

Bug:  894200 
Change-Id: I943afeaf5653ca30baee611f7d1f8ebb50a3030c
Reviewed-on: https://chromium-review.googlesource.com/c/1319029
Reviewed-by: Victor Miura <vmiura@chromium.org>
Commit-Queue: Khushal <khushalsagar@chromium.org>
Cr-Original-Commit-Position: refs/heads/master@{#605739}(cherry picked from commit 3c3873a6c34f6c8a4454f669351cd49e494e5b70)
Reviewed-on: https://chromium-review.googlesource.com/c/1327539
Reviewed-by: Khushal <khushalsagar@chromium.org>
Cr-Commit-Position: refs/branch-heads/3578@{#594}
Cr-Branched-From: 4226ddf99103e493d7afb23a4c7902ee496108b6-refs/heads/master@{#599034}
[modify] https://crrev.com/010af68a87603021daee029b7bd3b7c423183907/cc/raster/gpu_raster_buffer_provider.cc
[modify] https://crrev.com/010af68a87603021daee029b7bd3b7c423183907/cc/raster/gpu_raster_buffer_provider.h
Project Member

Comment 98 by 42576172...@developer.gserviceaccount.com, Nov 9 

📍 Job complete. See results below.
https://pinpoint-dot-chromeperf.appspot.com/job/1295e6e3e40000
Project Member

Comment 100 by bugdroid1@chromium.org, Nov 16 

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/1be26f6e4aa49b40d19501e105b72ebf170b4afb

commit 1be26f6e4aa49b40d19501e105b72ebf170b4afb
Author: Austin Eng <enga@chromium.org>
Date: Fri Nov 16 19:37:08 2018

Rereun ui/gl/generate_bindings.py

Bug:  894200 ,  891059 
Change-Id: Ic3ad1be6bf9ce21cd672266d42ef4259fd47cd9d
Reviewed-on: https://chromium-review.googlesource.com/c/1340568
Reviewed-by: Kenneth Russell <kbr@chromium.org>
Commit-Queue: Kenneth Russell <kbr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#608893}
[modify] https://crrev.com/1be26f6e4aa49b40d19501e105b72ebf170b4afb/ui/gl/gl_enums_implementation_autogen.h

Comment 101 by khushals...@chromium.org, Jan 9

Status: Fixed (was: Started)
We are not pursuing any more serialization improvements, and the required metrics are in place.
Showing comments 2 - 101 of 101 Older ›
►

Sign in to add a comment