On Linux FYI Debug (NVIDIA), a single test is crashing: 
1 test crashed:
    ParallelShaderCompileTest.LinkAndDrawManyPrograms/ES2_OPENGL

Sample stack trace:
Received signal 11 SEGV_MAPERR 7f679b747008
#0 0x7f67a9a0de89 base::debug::StackTrace::StackTrace()
#1 0x7f67a9856aa5 base::debug::StackTrace::StackTrace()
#2 0x7f67a9a0d91a base::debug::(anonymous namespace)::StackDumpSignalHandler()
#3 0x7f67a8dec330 <unknown>
#4 0x7f67a9a6afb2 GetStackTrace()
#5 0x7f67a9a7527d MallocBlockQueueEntry::MallocBlockQueueEntry()
#6 0x7f67a9a731fd MallocBlock::ProcessFreeQueue()
#7 0x7f67a9a76634 MallocBlock::Deallocate()
#8 0x7f67a9a6fdb5 DebugDeallocate()
#9 0x7f67a9a9ccd3 tc_free
#10 0x7f67a9a3537d (anonymous namespace)::TCFree()
#11 0x7f67a9a34f3f ShimFree
#12 0x7f67a9a34999 free

On Mac FYI Debug (Intel), a single (but different) test is crashing:

1 test crashed:
    GLSLTest_ES3.LargeNumberOfFloat4Parameters/ES3_OPENGL 

It seems to die before a stack trace can be captured:
[46849:68099:0116/191648.240065:24362686907778:ERROR:kill_posix.cc(84)] Unable to terminate process group 46850: No such process (3)

Hum. Maybe a memory leak or something? I'll try to repro.

TSAN is reporting data races (more detail to come). From the Mac bot, it looks like the failures began with this ANGLE commit:

"ParallelCompile: add GL backend support" (a100d8f471f79b9f88d387164992cc5bd9c6ee9f)

parent_got_angle_revision on the same bot just prior to the failure is bb4be1ad1034193de209d9346dbe12b0ddea513b, which is the parent commit to the one above.

(The Chrome commits on the first Mac fail look pretty innocuous, but the range is r623523-r623536 just in case I missed something.)

Same data race from the TSAN bot:

Entering AtomicCounterBufferTest31.AtomicCounterIncrementAndDecrement/ES3_1_OPENGL
==================
WARNING: ThreadSanitizer: data race (pid=18302)
  Write of size 8 at 0x7b500000f0a8 by main thread:
    #0 free /b/swarming/w/ir/kitchen-workdir/src/third_party/llvm/compiler-rt/lib/tsan/rtl/tsan_interceptors.cc:720 (angle_end2end_tests+0x4bf734)
    #1 vk_icdGetInstanceProcAddr ??:? (libGLX_nvidia.so.0+0xb67b0)
    #2 mapRange third_party/angle/src/libANGLE/Buffer.cpp:168 (libGLESv2.so+0x2f5810)
    #3 mapBufferRange third_party/angle/src/libANGLE/Context.cpp:4429 (libGLESv2.so+0x32e3be)
    #4 MapBufferRange third_party/angle/src/libGLESv2/entry_points_gles_3_0_autogen.cpp:1271 (libGLESv2.so+0x24b9b1)
    #5 TestBody third_party/angle/src/tests/gl_tests/AtomicCounterBufferTest.cpp:211 (angle_end2end_tests+0x53281b)
    #6 non-virtual thunk to (anonymous namespace)::AtomicCounterBufferTest31_AtomicCounterIncrementAndDecrement_Test::TestBody() third_party/angle/src/tests/gl_tests/AtomicCounterBufferTest.cpp:? (angle_end2end_tests+0x532ce0)
    #7 HandleExceptionsInMethodIfSupported<testing::Test, void> third_party/googletest/src/googletest/src/gtest.cc:? (angle_end2end_tests+0xf2da6f)
    #8 Run third_party/googletest/src/googletest/src/gtest.cc:2522 (angle_end2end_tests+0xf2da6f)
    #9 Run third_party/googletest/src/googletest/src/gtest.cc:2703 (angle_end2end_tests+0xf2eae8)
    #10 Run third_party/googletest/src/googletest/src/gtest.cc:2825 (angle_end2end_tests+0xf2f376)
    #11 RunAllTests third_party/googletest/src/googletest/src/gtest.cc:5227 (angle_end2end_tests+0xf406b6)
    #12 HandleExceptionsInMethodIfSupported<testing::internal::UnitTestImpl, bool> third_party/googletest/src/googletest/src/gtest.cc:? (angle_end2end_tests+0xf3ff89)
    #13 Run third_party/googletest/src/googletest/src/gtest.cc:4835 (angle_end2end_tests+0xf3ff89)
    #14 RUN_ALL_TESTS third_party/googletest/src/googletest/include/gtest/gtest.h:2369 (angle_end2end_tests+0x10a3b62)
    #15 Run base/test/test_suite.cc:294 (angle_end2end_tests+0x10a3b62)
    #16 (anonymous namespace)::RunHelper(base::TestSuite*) gpu/angle_end2end_tests_main.cc:16 (angle_end2end_tests+0xd70d62)
    #17 Invoke<int (*)(base::TestSuite *), base::TestSuite *> base/bind_internal.h:416 (angle_end2end_tests+0xd70db4)
    #18 MakeItSo<int (*)(base::TestSuite *), base::TestSuite *> base/bind_internal.h:616 (angle_end2end_tests+0xd70db4)
    #19 RunImpl<int (*)(base::TestSuite *), std::__1::tuple<base::internal::UnretainedWrapper<base::TestSuite> >, 0> base/bind_internal.h:689 (angle_end2end_tests+0xd70db4)
    #20 RunOnce base/bind_internal.h:658 (angle_end2end_tests+0xd70db4)
    #21 Run base/callback.h:99 (angle_end2end_tests+0x10a7ff8)
    #22 LaunchUnitTestsInternal base/test/launcher/unit_test_launcher.cc:225 (angle_end2end_tests+0x10a7ff8)
    #23 LaunchUnitTestsWithOptions base/test/launcher/unit_test_launcher.cc:596 (angle_end2end_tests+0x10a873f)
    #24 main gpu/angle_end2end_tests_main.cc:30 (angle_end2end_tests+0xd70ce3)

  Previous read of size 8 at 0x7b500000f0a8 by thread T4:
    #0 memcpy /b/swarming/w/ir/kitchen-workdir/src/third_party/llvm/compiler-rt/lib/tsan/../sanitizer_common/sanitizer_common_interceptors.inc:801 (angle_end2end_tests+0x4ca60d)
    #1 _nv042glcore ??:? (libnvidia-glcore.so.384.90+0x11eb744)

  Thread T4 (tid=18323, finished) created by main thread at:
    #0 pthread_create /b/swarming/w/ir/kitchen-workdir/src/third_party/llvm/compiler-rt/lib/tsan/rtl/tsan_interceptors.cc:980 (angle_end2end_tests+0x4c07e5)
    #1 __libcpp_thread_create buildtools/third_party/libc++/trunk/include/__threading_support:328 (libGLESv2.so+0x41ca42)
    #2 thread<void (std::__1::__async_assoc_state<void, std::__1::__async_func<(lambda at ../../third_party/angle/src/libANGLE/WorkerThread.cpp:181:54)> >::*)(), std::__1::__async_assoc_state<void, std::__1::__async_func<(lambda at ../../third_party/angle/src/libANGLE/WorkerThread.cpp:181:54)> > *, void> buildtools/third_party/libc++/trunk/include/thread:368 (libGLESv2.so+0x41ca42)
    #3 __make_async_assoc_state<void, std::__1::__async_func<(lambda at ../../third_party/angle/src/libANGLE/WorkerThread.cpp:181:54)> > buildtools/third_party/libc++/trunk/include/future:2296 (libGLESv2.so+0x41ca42)
    #4 async<(lambda at ../../third_party/angle/src/libANGLE/WorkerThread.cpp:181:54)> buildtools/third_party/libc++/trunk/include/future:2345 (libGLESv2.so+0x41ca42)
    #5 checkToRunPendingTasks third_party/angle/src/libANGLE/WorkerThread.cpp:181 (libGLESv2.so+0x41ca42)
    #6 postWorkerTask third_party/angle/src/libANGLE/WorkerThread.cpp:153 (libGLESv2.so+0x41c6ba)
    #7 LinkEventGL third_party/angle/src/libANGLE/renderer/gl/ProgramGL.cpp:158 (libGLESv2.so+0x57217c)
    #8 make_unique<rx::ProgramGL::LinkEventGL, std::__1::shared_ptr<angle::WorkerThreadPool> &, std::__1::shared_ptr<rx::ProgramGL::LinkTask> &, (lambda at ../../third_party/angle/src/libANGLE/renderer/gl/ProgramGL.cpp:382:29) &> buildtools/third_party/libc++/trunk/include/memory:3118 (libGLESv2.so+0x57217c)
    #9 link third_party/angle/src/libANGLE/renderer/gl/ProgramGL.cpp:432 (libGLESv2.so+0x57217c)
    #10 link third_party/angle/src/libANGLE/Program.cpp:1263 (libGLESv2.so+0x3a3a4c)
    #11 linkProgram third_party/angle/src/libANGLE/Context.cpp:5889 (libGLESv2.so+0x339a88)
    #12 LinkProgram third_party/angle/src/libGLESv2/entry_points_gles_2_0_autogen.cpp:1583 (libGLESv2.so+0x244cc2)
    #13 CompileComputeProgram(char const*, bool) third_party/angle/util/shader_utils.cpp:240 (libangle_util.so+0x4971c)
    #14 makeCompute third_party/angle/src/tests/test_utils/gl_raii.h:143 (angle_end2end_tests+0x532453)
    #15 TestBody third_party/angle/src/tests/gl_tests/AtomicCounterBufferTest.cpp:194 (angle_end2end_tests+0x532453)
    #16 non-virtual thunk to (anonymous namespace)::AtomicCounterBufferTest31_AtomicCounterIncrementAndDecrement_Test::TestBody() third_party/angle/src/tests/gl_tests/AtomicCounterBufferTest.cpp:? (angle_end2end_tests+0x532ce0)
    #17 HandleExceptionsInMethodIfSupported<testing::Test, void> third_party/googletest/src/googletest/src/gtest.cc:? (angle_end2end_tests+0xf2da6f)
    #18 Run third_party/googletest/src/googletest/src/gtest.cc:2522 (angle_end2end_tests+0xf2da6f)
    #19 Run third_party/googletest/src/googletest/src/gtest.cc:2703 (angle_end2end_tests+0xf2eae8)
    #20 Run third_party/googletest/src/googletest/src/gtest.cc:2825 (angle_end2end_tests+0xf2f376)
    #21 RunAllTests third_party/googletest/src/googletest/src/gtest.cc:5227 (angle_end2end_tests+0xf406b6)
    #22 HandleExceptionsInMethodIfSupported<testing::internal::UnitTestImpl, bool> third_party/googletest/src/googletest/src/gtest.cc:? (angle_end2end_tests+0xf3ff89)
    #23 Run third_party/googletest/src/googletest/src/gtest.cc:4835 (angle_end2end_tests+0xf3ff89)
    #24 RUN_ALL_TESTS third_party/googletest/src/googletest/include/gtest/gtest.h:2369 (angle_end2end_tests+0x10a3b62)
    #25 Run base/test/test_suite.cc:294 (angle_end2end_tests+0x10a3b62)
    #26 (anonymous namespace)::RunHelper(base::TestSuite*) gpu/angle_end2end_tests_main.cc:16 (angle_end2end_tests+0xd70d62)
    #27 Invoke<int (*)(base::TestSuite *), base::TestSuite *> base/bind_internal.h:416 (angle_end2end_tests+0xd70db4)
    #28 MakeItSo<int (*)(base::TestSuite *), base::TestSuite *> base/bind_internal.h:616 (angle_end2end_tests+0xd70db4)
    #29 RunImpl<int (*)(base::TestSuite *), std::__1::tuple<base::internal::UnretainedWrapper<base::TestSuite> >, 0> base/bind_internal.h:689 (angle_end2end_tests+0xd70db4)
    #30 RunOnce base/bind_internal.h:658 (angle_end2end_tests+0xd70db4)
    #31 Run base/callback.h:99 (angle_end2end_tests+0x10a7ff8)
    #32 LaunchUnitTestsInternal base/test/launcher/unit_test_launcher.cc:225 (angle_end2end_tests+0x10a7ff8)
    #33 LaunchUnitTestsWithOptions base/test/launcher/unit_test_launcher.cc:596 (angle_end2end_tests+0x10a873f)
    #34 main gpu/angle_end2end_tests_main.cc:30 (angle_end2end_tests+0xd70ce3)

(I could be misinterpreting, but it's interesting that these seem to be Vulkan backend stack traces on a GL backend change.)

OK. Maybe we should revert the commit. It looks like the D3D11 failures might be happening in the call to SafeRelease(mDebug); in Renderer11.cpp. That would make sense why it only happens on Debug. I can repro with the flag 

--gtest_filter=Texture2DArrayCopy*D3D11

I'll do a bisect to find the CL.

The TSAN bot is pinpointing the same ANGLE change (a100d8).

Ah, I think I see at least one culprit:

Re-enable 2DArray Copy Texture Tests by Brandon Jones from Intel.

Can probably revert that as well. Or investigate more locally.

Link: https://chromium-review.googlesource.com/c/1308735

The following revision refers to this bug:
  https://chromium.googlesource.com/angle/angle/+/c66cd693c167dfadef16a6d5086a85c735900119

commit c66cd693c167dfadef16a6d5086a85c735900119
Author: Geoff Lang <geofflang@chromium.org>
Date: Thu Jan 17 16:41:24 2019

GL: Temporarily disable parallel shader compile.

The Skia bots have been unable to roll ANGLE since parallel compile landed in
ANGLE.  Speculatively disable parallel compile using a workaround.
Failed roll: https://skia-review.googlesource.com/c/skia/+/184705

Also seen instability on Linux, Mac and Windows in 922936.

BUG=922936
BUG=849576

Change-Id: I7116b3f6541425135984c52a8077bdb98d961784
Reviewed-on: https://chromium-review.googlesource.com/c/1415729
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>

[modify] https://crrev.com/c66cd693c167dfadef16a6d5086a85c735900119/src/libANGLE/renderer/gl/RendererGL.cpp
[modify] https://crrev.com/c66cd693c167dfadef16a6d5086a85c735900119/src/libANGLE/renderer/gl/WorkaroundsGL.h
[modify] https://crrev.com/c66cd693c167dfadef16a6d5086a85c735900119/src/libANGLE/renderer/gl/renderergl_utils.cpp

From one failed Skia roll:
https://task-scheduler.skia.org/job/20190117T040145.103624731Z_0000000000721e4a

and this shard from "Test-Win10-Clang-NUC6i5SYK-GPU-IntelIris540-x86_64-Debug-All-ANGLE":
https://chromium-swarm.appspot.com/task?id=4272d884c4980110&refresh=10

this compile failure is reported:

start unit test  makeBackendTexture
GLSL compilation error
----------------------
GLSL:
   1	#version 100
   2	
   3	precision mediump float;
   4	varying mediump vec4 vcolor_Stage0;
   5	void main() {
   6	    mediump vec4 outputColor_Stage0;
   7	    {
   8	        outputColor_Stage0 = vcolor_Stage0;
   9	    }
  10	    {
  11	        gl_FragColor = outputColor_Stage0;
  12	    }
  13	}
  14	
../../../src/gpu/gl/builders/GrGLShaderStringBuilder.cpp(158): fatal error: "GLSL compilation failed!"

Caught exception 2147483651 EXCEPTION_BREAKPOINT	unit test  makeBackendTexture
	unit test  makeBackendTexture
WARN: initializeD3DDevice(771): Failed creating Debug D3D11 device - falling back to release r
Caught exception 2147483651 EXCEPTION_BREAKPOINT, was running:
	unit test  makeBackendTexture
Likely culprit:
	unit test  makeBackendTexture
step returned non-zero exit code: 0x80000003

Great investigation senorblanco@! Thanks for your quick triage!

Looks like the current build on Mac FYI Debug (Intel) and Linux FYI GPU TSAN Release (which include Geoff's disable) have gone green on angle_end2end_tests.

Win10 FYI Debug (NVIDIA) actually went green before Geoff's patch, so that failure must be unrelated (or flaky).

Windows issue tracked in  bug 922296  .

The TSAN race in #6 may have the symbol "vk_icdGetInstanceProcAddr" in the stack but I think that's a red herring.

The most likely cause of the failure as well as some of the other crashes or assertion failures on the other platforms is probably a memory error causing heap corruption.

Jie, may I please assign this to you to try to reproduce some of these failures with parallel shader compilation enabled?

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/74ee536827fa1558a29601be3959c1f109da5fd0

commit 74ee536827fa1558a29601be3959c1f109da5fd0
Author: chromium-autoroll <chromium-autoroll@skia-public.iam.gserviceaccount.com>
Date: Fri Jan 18 00:39:32 2019

Roll src/third_party/angle b3bdd2acc4f0..5fe7c5b92642 (3 commits)

https://chromium.googlesource.com/angle/angle.git/+log/b3bdd2acc4f0..5fe7c5b92642


git log b3bdd2acc4f0..5fe7c5b92642 --date=short --no-merges --format='%ad %ae %s'
2019-01-17 ynovikov@chromium.org Include common/platform.h where used
2019-01-17 geofflang@chromium.org Reland "Don't call Texture::onDestroy twice for the zero textures."
2019-01-17 geofflang@chromium.org GL: Temporarily disable parallel shader compile.


Created with:
  gclient setdep -r src/third_party/angle@5fe7c5b92642

The AutoRoll server is located here: https://autoroll.skia.org/r/angle-chromium-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_optional_gpu_tests_rel;luci.chromium.try:mac_optional_gpu_tests_rel;luci.chromium.try:win_optional_gpu_tests_rel

BUG= chromium:922443 , chromium:922296 ,chromium:922936,chromium:849576
TBR=cwallez@chromium.org

Change-Id: I970bdb5c71c8dd228901354565301076ca794675
Reviewed-on: https://chromium-review.googlesource.com/c/1418126
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@{#623929}
[modify] https://crrev.com/74ee536827fa1558a29601be3959c1f109da5fd0/DEPS

Sorry for the troubles. I am still looking into these failures. The failed Linux bots were both with Nvidia GPU. I haven't found a machine to reproduce the failures successfully. Probably we can enable the native extension support to work around it, as it's available on Nvidia driver.

The GLSLTest_ES3.LargeNumberOfFloat4Parameters/ES3_OPENGL crashed on Mac also occurred on trybots. I once found a workaround for it. But it seemed useless on FYI bots. I will look into it further. 

The following revision refers to this bug:
  https://chromium.googlesource.com/angle/angle/+/828acb39df5fe60953dc1eca00b8d3cd25a63ab7

commit 828acb39df5fe60953dc1eca00b8d3cd25a63ab7
Author: jchen10 <jie.a.chen@intel.com>
Date: Wed Jan 23 00:36:54 2019

Enable ParallelCompile for Mac

Linked in parallel, GLSLTest_ES3.LargeNumberOfFloat4Parameters
fails on the Mac asan bots. It looks a driver bug caused by
handling some long GLSL expressions. This works around it by
breaking down the expression in the test from:
    return a0 + a1 + ... + alast;
to:
    vec4 sum = vec4(0, 0, 0, 0);
    sum += a0;
    sum += a1;
    ...
    sum += alast;
    return sum;

This also fixes a CGLPixelFormat leak, although it's irrelevant
to the bot failures.

BUG=922936
BUG=angleproject:3087
BUG=angleproject:3047

Change-Id: I20249ada43e9dd226f582a568ed4ed50a0e4375d
Reviewed-on: https://chromium-review.googlesource.com/c/1426430
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Jie A Chen <jie.a.chen@intel.com>

[modify] https://crrev.com/828acb39df5fe60953dc1eca00b8d3cd25a63ab7/src/tests/gl_tests/GLSLTest.cpp
[modify] https://crrev.com/828acb39df5fe60953dc1eca00b8d3cd25a63ab7/src/libANGLE/renderer/gl/renderergl_utils.cpp
[modify] https://crrev.com/828acb39df5fe60953dc1eca00b8d3cd25a63ab7/src/libANGLE/renderer/gl/cgl/DisplayCGL.mm

The following revision refers to this bug:
  https://chromium.googlesource.com/angle/angle/+/828acb39df5fe60953dc1eca00b8d3cd25a63ab7

commit 828acb39df5fe60953dc1eca00b8d3cd25a63ab7
Author: jchen10 <jie.a.chen@intel.com>
Date: Wed Jan 23 00:36:54 2019

Enable ParallelCompile for Mac

Linked in parallel, GLSLTest_ES3.LargeNumberOfFloat4Parameters
fails on the Mac asan bots. It looks a driver bug caused by
handling some long GLSL expressions. This works around it by
breaking down the expression in the test from:
    return a0 + a1 + ... + alast;
to:
    vec4 sum = vec4(0, 0, 0, 0);
    sum += a0;
    sum += a1;
    ...
    sum += alast;
    return sum;

This also fixes a CGLPixelFormat leak, although it's irrelevant
to the bot failures.

BUG=922936
BUG=angleproject:3087
BUG=angleproject:3047

Change-Id: I20249ada43e9dd226f582a568ed4ed50a0e4375d
Reviewed-on: https://chromium-review.googlesource.com/c/1426430
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Jie A Chen <jie.a.chen@intel.com>

[modify] https://crrev.com/828acb39df5fe60953dc1eca00b8d3cd25a63ab7/src/tests/gl_tests/GLSLTest.cpp
[modify] https://crrev.com/828acb39df5fe60953dc1eca00b8d3cd25a63ab7/src/libANGLE/renderer/gl/renderergl_utils.cpp
[modify] https://crrev.com/828acb39df5fe60953dc1eca00b8d3cd25a63ab7/src/libANGLE/renderer/gl/cgl/DisplayCGL.mm

The following revision refers to this bug:
  https://chromium.googlesource.com/angle/angle/+/828acb39df5fe60953dc1eca00b8d3cd25a63ab7

commit 828acb39df5fe60953dc1eca00b8d3cd25a63ab7
Author: jchen10 <jie.a.chen@intel.com>
Date: Wed Jan 23 00:36:54 2019

Enable ParallelCompile for Mac

Linked in parallel, GLSLTest_ES3.LargeNumberOfFloat4Parameters
fails on the Mac asan bots. It looks a driver bug caused by
handling some long GLSL expressions. This works around it by
breaking down the expression in the test from:
    return a0 + a1 + ... + alast;
to:
    vec4 sum = vec4(0, 0, 0, 0);
    sum += a0;
    sum += a1;
    ...
    sum += alast;
    return sum;

This also fixes a CGLPixelFormat leak, although it's irrelevant
to the bot failures.

BUG=922936
BUG=angleproject:3087
BUG=angleproject:3047

Change-Id: I20249ada43e9dd226f582a568ed4ed50a0e4375d
Reviewed-on: https://chromium-review.googlesource.com/c/1426430
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Jie A Chen <jie.a.chen@intel.com>

[modify] https://crrev.com/828acb39df5fe60953dc1eca00b8d3cd25a63ab7/src/tests/gl_tests/GLSLTest.cpp
[modify] https://crrev.com/828acb39df5fe60953dc1eca00b8d3cd25a63ab7/src/libANGLE/renderer/gl/renderergl_utils.cpp
[modify] https://crrev.com/828acb39df5fe60953dc1eca00b8d3cd25a63ab7/src/libANGLE/renderer/gl/cgl/DisplayCGL.mm

The following revision refers to this bug:
  https://chromium.googlesource.com/angle/angle/+/828acb39df5fe60953dc1eca00b8d3cd25a63ab7

commit 828acb39df5fe60953dc1eca00b8d3cd25a63ab7
Author: jchen10 <jie.a.chen@intel.com>
Date: Wed Jan 23 00:36:54 2019

Enable ParallelCompile for Mac

Linked in parallel, GLSLTest_ES3.LargeNumberOfFloat4Parameters
fails on the Mac asan bots. It looks a driver bug caused by
handling some long GLSL expressions. This works around it by
breaking down the expression in the test from:
    return a0 + a1 + ... + alast;
to:
    vec4 sum = vec4(0, 0, 0, 0);
    sum += a0;
    sum += a1;
    ...
    sum += alast;
    return sum;

This also fixes a CGLPixelFormat leak, although it's irrelevant
to the bot failures.

BUG=922936
BUG=angleproject:3087
BUG=angleproject:3047

Change-Id: I20249ada43e9dd226f582a568ed4ed50a0e4375d
Reviewed-on: https://chromium-review.googlesource.com/c/1426430
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Jie A Chen <jie.a.chen@intel.com>

[modify] https://crrev.com/828acb39df5fe60953dc1eca00b8d3cd25a63ab7/src/tests/gl_tests/GLSLTest.cpp
[modify] https://crrev.com/828acb39df5fe60953dc1eca00b8d3cd25a63ab7/src/libANGLE/renderer/gl/renderergl_utils.cpp
[modify] https://crrev.com/828acb39df5fe60953dc1eca00b8d3cd25a63ab7/src/libANGLE/renderer/gl/cgl/DisplayCGL.mm

The following revision refers to this bug:
  https://chromium.googlesource.com/angle/angle/+/828acb39df5fe60953dc1eca00b8d3cd25a63ab7

commit 828acb39df5fe60953dc1eca00b8d3cd25a63ab7
Author: jchen10 <jie.a.chen@intel.com>
Date: Wed Jan 23 00:36:54 2019

Enable ParallelCompile for Mac

Linked in parallel, GLSLTest_ES3.LargeNumberOfFloat4Parameters
fails on the Mac asan bots. It looks a driver bug caused by
handling some long GLSL expressions. This works around it by
breaking down the expression in the test from:
    return a0 + a1 + ... + alast;
to:
    vec4 sum = vec4(0, 0, 0, 0);
    sum += a0;
    sum += a1;
    ...
    sum += alast;
    return sum;

This also fixes a CGLPixelFormat leak, although it's irrelevant
to the bot failures.

BUG=922936
BUG=angleproject:3087
BUG=angleproject:3047

Change-Id: I20249ada43e9dd226f582a568ed4ed50a0e4375d
Reviewed-on: https://chromium-review.googlesource.com/c/1426430
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Jie A Chen <jie.a.chen@intel.com>

[modify] https://crrev.com/828acb39df5fe60953dc1eca00b8d3cd25a63ab7/src/tests/gl_tests/GLSLTest.cpp
[modify] https://crrev.com/828acb39df5fe60953dc1eca00b8d3cd25a63ab7/src/libANGLE/renderer/gl/renderergl_utils.cpp
[modify] https://crrev.com/828acb39df5fe60953dc1eca00b8d3cd25a63ab7/src/libANGLE/renderer/gl/cgl/DisplayCGL.mm