|
611bbaab
|
2018-12-06T01:59:53
|
|
Vulkan: Convert vertex attributes in compute
In this commit, VertexArrayVk::convertVertexBuffer() is renamed to
VertexArrayVk::convertVertexBufferCpu() to explicitly show it does a CPU
readback. A new VertexArrayVk::convertVertexBuffer() function is added
that has the same functionality in gpu (with some assumptions, where the
CPU fallback is used should those assumptions fail). Currently, the
only requirement is that buffer offset/stride are divided by the
component size.
ConvertVertex.comp is the shader responsible for this conversion, and it
implements the functionality in renderer/copyvertex.inc, minus a few
functions that are not used in the Vulkan backend.
Bug: angleproject:2958, angleproject:3009
Change-Id: I8ec9a5f4672509bcf7b9e352cd27663970ad4653
Reviewed-on: https://chromium-review.googlesource.com/c/1364451
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
|
|
573f76b3
|
2018-05-03T11:10:44
|
|
Debug: Add Systrace Markers
Update ScopedPerfEventHelper class to add systrace markers by default.
This change unifies ANGLE EVENT* macro system so that at the base level
in LoggingAnnotator class, systrace markers will be added by default.
Modify the base DebugLogger to use char* by default and move any
conversions to wchar_t to the Windows specializations where wchar is
used. This limits type conversions to only where they're needed.
This change also includes some new TRACE_EVENT() calls in the VK
backend which will result in systrace markers for those calls on the
Android platform.
The new build flag "angle_enable_trace" is added to enable the tracing
calls.
Bug: angleproject:2528
Change-Id: Icefc197d4407e1cd31338710e37865abae6a0b15
Reviewed-on: https://chromium-review.googlesource.com/c/1042785
Commit-Queue: Tobin Ehlis <tobine@google.com>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
|
|
fde74c07
|
2018-11-18T16:12:02
|
|
Vulkan: Apply point size clamping workaround in NV.
It seems to be a regression in recent drivers. Olli contributed this
workaround for the GL back-end some time ago. This CL enables it for
Vulkan as well on the affected drivers.
Bug: angleproject:2970
Change-Id: I37f0caf8d9db073cb880aa1500d1ff7a1eee9d34
Reviewed-on: https://chromium-review.googlesource.com/c/1341108
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
|
|
|
d856ca48
|
2018-10-31T16:55:12
|
|
Vulkan: add clear test for emulated stencil or depth formats
S8_UINT and D24_UNORM_X8_UINT are the only formats currently that are
single-aspect and are possibly emulated with a packed depth-stencil
format if it's not supported. A flag to FeaturesVk has been added as a
way to force this behavior for the sake of testing.
This test is added to ensure the correct clear algorithm is used for
this case. Additionally, this case is detected and the other aspect is
forcefully cleared to 0 whenever the original aspect is cleared.
Bug: angleproject:2815
Change-Id: Ief3039d66bbf46468213b9e3224f7cc7541c3a2e
Reviewed-on: https://chromium-review.googlesource.com/c/1312453
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
|
1bf963b6
|
2018-07-19T14:56:37
|
|
Clean up PlatformMethods attribute enum
Centralize definition of EGL_PLATFORM_ANGLE_PLATFORM_METHODS_ANGLEX
in Platform.h.
Bug: angleproject:2528
Change-Id: Id2b022c4ea08d2e98cbd6c3b248b58d096c20dfb
Reviewed-on: https://chromium-review.googlesource.com/1144265
Reviewed-by: Tobin Ehlis <tobine@google.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
|
|
|
5ba37427
|
2018-06-26T10:45:53
|
|
Move PlatformMethods constructor to header
Move the default constructor for PlatformMethods struct from
Platform.cpp into Platform.h. This allows other projects to include
Platform.h and use PlatformMethods without having to link against ANGLE
Bug: angleproject:2528
Change-Id: Icb4b9e149e78e9ec7be85804d711d8a2216e61c4
Reviewed-on: https://chromium-review.googlesource.com/1115212
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
e13f7ebe
|
2018-01-26T17:08:49
|
|
Make NVIDIA constant register zero skip specific to VS
It was found that the driver issue only affects vertex
shaders, so the workaround can be simplified to affect only
vertex shaders.
BUG=angleproject:2294
TEST=WebGL conformance tests on passthrough command buffer,
angle_end2end_tests
Change-Id: Ie7fb4e75a3999e1de3d5c20d3de21e7ebb08f148
Reviewed-on: https://chromium-review.googlesource.com/889099
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
|
|
|
36937a64
|
2018-01-03T14:03:44
|
|
Revert "Enable depth buffer to workaround driver bug on Intel windows"
This reverts commit 63ba357c093954d2bc475867e1296fbf5afc1d37.
Reason for revert: May be causing Windows GPU.FYI bot failures.
https://ci.chromium.org/buildbot/chromium.gpu.fyi/Win10%20Release%20%28Intel%20HD%20630%29/1443
maps_pixel_test on Intel GPU:
Unexpected Failures:
* gpu_tests.maps_integration_test.MapsIntegrationTest.Maps_maps
webgl2_conformance_tests on Intel GPU:
Unexpected Failures:
* gpu_tests.webgl_conformance_integration_test.WebGLConformanceIntegrationTest.WebglConformance_deqp_functional_gles3_fboinvalidate_default
* gpu_tests.webgl_conformance_integration_test.WebGLConformanceIntegrationTest.WebglConformance_deqp_functional_gles3_fboinvalidate_sub
Bug: chromium:798757
Original change's description:
> Enable depth buffer to workaround driver bug on Intel windows
>
> Rendering with depth buffer disabled and stencil buffer enabled
> leads to memory leak if we set viewport a large size on Intel
> windows platforms. So we enable depth buffer if stencil buffer
> is enabled to workaround this issue.
>
> TEST=gl_test.exe --gtest_filter=GLClearFramebufferTestWithParam/GLClearFramebufferTest.ClearDepthStencil/0
> TEST=conformance/rendering/rendering-stencil-large-viewport.html
> TEST=RenderStencilBufferTest.DrawWithLargeViewport/ES3_D3D11
>
> BUG=782317
>
> Change-Id: Idb185db296f13e3fa897534514e198651a56439f
> Reviewed-on: https://chromium-review.googlesource.com/809574
> Commit-Queue: Jamie Madill <jmadill@chromium.org>
> Reviewed-by: Jamie Madill <jmadill@chromium.org>
TBR=zmo@chromium.org,geofflang@chromium.org,jmadill@chromium.org,kbr@chromium.org,cwallez@chromium.org,yunchao.he@intel.com,jiajia.qin@intel.com,xinghua.cao@intel.com,bryan.bernhart@intel.com,yizhou.jiang@intel.com,yang.gu@intel.com
# Not skipping CQ checks because original CL landed > 1 day ago.
Bug: 782317
Change-Id: I7e29da683b22fd4640c230598c7b220cfae6a177
Reviewed-on: https://chromium-review.googlesource.com/848133
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
|
d8724a94
|
2017-12-29T18:40:36
|
|
Start D3D constant register allocations from 1 on NVIDIA
Recent NVIDIA drivers have a bug where a specific optimized path
inside the driver doesn't handle constant register 0 correctly. Work
around this by starting constant register allocations from 1. This
should make sure that the bug doesn't trigger if the ordering of
uniforms is changed on the D3D backend.
The repro case seems to require some specific driver state to be set
that's used inside Chromium. Because of this we have not been able to
develop a standalone test case so far.
The maximum number of available uniform slots is reduced accordingly.
This should not take them below required minimums in the spec.
BUG=angleproject:2294
TEST=WebGL tests on passthrough command buffer,
angle_end2end_tests --gtest_filter=*GLSLTest*Uniform*
Change-Id: I92fff71efe5432ea7f15a7e90d497492514c65dc
Reviewed-on: https://chromium-review.googlesource.com/847481
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
63ba357c
|
2017-12-06T10:25:40
|
|
Enable depth buffer to workaround driver bug on Intel windows
Rendering with depth buffer disabled and stencil buffer enabled
leads to memory leak if we set viewport a large size on Intel
windows platforms. So we enable depth buffer if stencil buffer
is enabled to workaround this issue.
TEST=gl_test.exe --gtest_filter=GLClearFramebufferTestWithParam/GLClearFramebufferTest.ClearDepthStencil/0
TEST=conformance/rendering/rendering-stencil-large-viewport.html
TEST=RenderStencilBufferTest.DrawWithLargeViewport/ES3_D3D11
BUG=782317
Change-Id: Idb185db296f13e3fa897534514e198651a56439f
Reviewed-on: https://chromium-review.googlesource.com/809574
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
|
740aa41d
|
2017-11-14T10:35:35
|
|
Fix angle::Platform export definitions.
This was only showing up in the GYP build on Windows.
BUG=chromium:778918
Change-Id: Ib9db0c928474be887ec1f46d7e99c0a3dedfb9e0
Reviewed-on: https://chromium-review.googlesource.com/768947
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Olli Etuaho <oetuaho@nvidia.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
|
ff7aac5e
|
2017-11-13T16:57:14
|
|
Re-land "Export angle::PlatformMethods."
This should fix undefined function call sanitizer warnings.
Re-land fixes the build and uses proper export annotation. Also fixes
the missing source includes in BUILD.gn for libANGLE so future
changes to Platform.h (and other includes) will get picked up by the
dependency analyzer step.
BUG=chromium:778918
Change-Id: I92cc8a43bbbdc70de50a67f73b223d8f903cd6cd
Reviewed-on: https://chromium-review.googlesource.com/767111
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Antoine Labour <piman@chromium.org>
|
|
|
c2eb32ba
|
2017-11-13T21:51:48
|
|
Revert "Export angle::PlatformMethods."
This reverts commit 8b920b211d7363daef0d5fb8d2bd69f3253ab033.
Reason for revert: This had a syntax error and skipped the compile step, so went undetected.
BUG=chromium:778918
Original change's description:
> Export angle::PlatformMethods.
>
> This should fix undefined function call sanitizer warnings.
>
> BUG=chromium:778918
>
> Change-Id: I718d58edd7b75a3fca1089edc4df7f3251d85153
> Reviewed-on: https://chromium-review.googlesource.com/767152
> Reviewed-by: Jamie Madill <jmadill@chromium.org>
> Commit-Queue: Jamie Madill <jmadill@chromium.org>
TBR=jmadill@chromium.org,piman@chromium.org
Change-Id: I1825ce88bc82867a983bb08cd6732caff54f061c
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: chromium:778918
Reviewed-on: https://chromium-review.googlesource.com/767128
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
|
|
|
8b920b21
|
2017-11-13T15:09:56
|
|
Export angle::PlatformMethods.
This should fix undefined function call sanitizer warnings.
BUG=chromium:778918
Change-Id: I718d58edd7b75a3fca1089edc4df7f3251d85153
Reviewed-on: https://chromium-review.googlesource.com/767152
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
|
|
|
38d92b59
|
2017-09-13T13:47:52
|
|
Workaround Intel driver bug on D3D when renderering with no render target
When rendering with no render target on D3D, two bugs lead to
incorrect behavior on Intel drivers < 4815. The rendering samples
always pass neglecting discard statement in pixel shader.
Two bugs are listed bellow,
1. When a framebuffer has no attachments, the pixel shader will be
recompiled to drop 'SV_TARGET'. On Intel drivers, when using a pixel
shader with no 'SV_TARGET' in a draw, pixels are always generated even
if they should be discard by 'discard' statements.
2. When a framebuffer has no attachments, ID3D11BlendState.RenderTarget
[].RenderTargetWriteMask were set to 0 in angle. If RenderTargetWriteMask
is 0 and rendertarget is not set, then rendering samples also pass
neglecting discard statement in pixel shader on Intel.
So we add a dummy texture as render target to workaround this issue.
BUG=angleproject:2152
TEST=FramebufferTest_ES31.RenderingLimitToDefaultFBOSizeWithNoAttachments/ES3_1_D3D11
TEST=dEQP-GLES31.functional.fbo.no_attachments.*
TEST=dEQP-GLES31.functional.state_query.integer.max_framebuffer*
TEST=dEQP-GLES31.functional.state_query.integer.max_color_texture_samples_*
TEST=dEQP-GLES31.functional.state_query.integer.max_depth_texture_samples_*
TEST=dEQP-GLES31.functional.state_query.integer.max_integer_samples_*
Change-Id: I1cb974703b6c05c39b731d147f7c8c4fb7b5fe68
Reviewed-on: https://chromium-review.googlesource.com/741544
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
|
|
20c285ec
|
2017-10-31T19:01:09
|
|
callClearTwiceOnSmallTarget is needed for all sizes.
Gecko bug: https://bugzilla.mozilla.org/show_bug.cgi?id=1325733
BUG=angleproject:2214
Change-Id: I6837a0fbf0bbbd0877f333f0625729445a5999aa
Reviewed-on: https://chromium-review.googlesource.com/748320
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>
|
|
|
9abe07c6
|
2017-09-19T00:25:00
|
|
Move Platform Context pointer to start of struct.
This will ensure that new methods (placed at the end) don't collide
with the context pointer when using a mismatched platform.
BUG=angleproject:1892
Change-Id: Ice6a3ccaf3cc2a1e36a04bc217386cf096612176
Reviewed-on: https://chromium-review.googlesource.com/672147
Reviewed-by: Frank Henigman <fjhenigman@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
|
|
c1d4e550
|
2017-08-21T12:01:10
|
|
D3D11: Select view in vertex shader
View selection can happen in the vertex shader through the optional
feature VPAndRTArrayIndexFromAnyShaderFeedingRasterizer.
BUG=angleproject:2062
TEST=angle_end2end_tests
Change-Id: Iaf65685e04f828b0936295fea867f6f6cbe69bee
Reviewed-on: https://chromium-review.googlesource.com/628419
Commit-Queue: Martin Radev <mradev@nvidia.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
|
360daeef
|
2017-06-29T10:36:19
|
|
Add platform hook for program cache updates.
This will need to be matched with a corresponding browser-side CL.
It will enable writing out binary shaders to disk.
BUG=angleproject:1897
Change-Id: I443281086050b9711b92a034cf37f808dd919007
Reviewed-on: https://chromium-review.googlesource.com/542963
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
|
|
|
98de826c
|
2017-05-29T13:01:02
|
|
Platform: Fix initialization order in tests.
Because of the platform refactor in abf38572062b91 we broke the way
the tests override the D3D workarounds for the tiny depth/stencil bug.
This change passes a pointer to the platform directly in the EGL init
logic, which solves this issue. It also removes all decltype code in
the platform header which might fix the UBSAN problem we were seeing
previously.
Also change a present mode selection error into a warning in the Vk
back-end, since this was being triggered on AMD, but is safe.
BUG=angleproject:2042
Change-Id: Ibbd0c69ce11a840cf4b33c616f56020001e553aa
Reviewed-on: https://chromium-review.googlesource.com/513519
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
|
|
331f6dbd
|
2017-04-13T13:57:59
|
|
Tighten workaround on unary minus for Intel D3D drivers
This driver bug has been fixed in 20.19.15.4624.
BUG=chromium:644033
Change-Id: I6f621d7c4d4b823da29255f869af84097f4f7c2c
Reviewed-on: https://chromium-review.googlesource.com/476118
Reviewed-by: Yunchao He <yunchao.he@intel.com>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>
|
|
|
abe89c7d
|
2017-02-16T10:24:06
|
|
Tweak platform method signatures.
This works around a limitation in UBSAN which can't handle decltype.
Instead use void * and typedef where appropriate.
BUG=chromium:692274
Change-Id: I4eab796db3aa2e51c0fc558170eb2af61f07223d
Reviewed-on: https://chromium-review.googlesource.com/443885
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
|
|
|
abf38572
|
2017-02-14T16:47:59
|
|
Remove old C++ ANGLE platform.
Now that the new platform is in place, we can remove the old methods.
Must be landed after https://codereview.chromium.org/2697463003/
BUG=angleproject:1892
BUG=chromium:678870
Change-Id: Ia29a3b120cf3521fc0409019c2e64e4dbc6f460d
Reviewed-on: https://chromium-review.googlesource.com/441274
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
|
|
|
dbffdfbd
|
2017-02-13T18:52:38
|
|
Fix counting of the new platform methods.
The context pointer was throwing off the count by one.
Also change the type to void * for ease of use.
BUG=angleproject:1892
BUG=chromium:678870
Change-Id: Iffac667f0aa896d9247e9ffddfeb3bc447a9692f
Reviewed-on: https://chromium-review.googlesource.com/441931
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
|
|
|
b1eeba1f
|
2017-02-13T16:36:40
|
|
Use a C API for the ANGLE platform.
The main purpose of this change is to fix a fuzzer bug where we would
trigger undefined behaviour calling between Chrome and ANGLE. It's
not specced how virtual function calls work if the shared objects are
not directly linked together, and ANGLE and Chrome are not linked.
Replace the old class-style API with a C dispatch table.
Follow-up work will make the Platform owned by the Display instead of
using global variables, but fixing this is a bit tricky.
BUG=angleproject:1892
BUG=chromium:678870
Change-Id: Iad188bc2e50f2b5e4a03ce0de233d686f569c705
Reviewed-on: https://chromium-review.googlesource.com/441273
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
|
|
4af4951f
|
2016-12-26T13:12:22
|
|
Tighten the workaround for B5G6R5 on Intel drivers
This driver bug has been fixed in 20.19.15.4539.
BUG=chromium:644610
TEST=dEQP-GLES3.functional.fbo.blit.default_framebuffer.rgb565*
Change-Id: I37556304ed4f70771fb88b792c01b6be5b3aa294
Reviewed-on: https://chromium-review.googlesource.com/424103
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
|
|
|
f165f0b4
|
2016-12-20T15:53:11
|
|
Tighten the workarounds on isnan for Intel drivers
This driver bug has been fixed in 21.20.16.4542.
BUG=chromium:650547
Change-Id: I2a5c7c01bf48d854bd6c4c6a380e57617e00529a
Reviewed-on: https://chromium-review.googlesource.com/422992
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
|
408293f8
|
2016-12-20T10:11:45
|
|
D3D11: Fix stale SRVs with tiny depth textures.
When the tiny depth texture would get re-created on change, it was
not freeing any cached SRVs which were pointing to the old tex.
Fix this by erasing the cache entry if the texure is recreated.
Also include a test which hooks into the workarounds to force
testing on every platform.
Also narrow the workaround to only apply to depth/stencil textures
which use these mips in GL, rather than D3D, where we always create
a full mip chain.
BUG=angleproject:1664
Change-Id: If0ac396b8847e1bf9b50165e5332da573e9bb3e4
Reviewed-on: https://chromium-review.googlesource.com/421567
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
|
|
464bfc86
|
2015-10-07T12:12:02
|
|
Use .def file to export functions using dllexport.
BUG=angleproject:1183
This is a reland of part of commit 566273222683314dfc8ac1cdb9bf6deb5d2b4c65.
The original commit caused warnings on 64-bit MSVC linker. This version
removes declspec(dllexport) from exported ANGLE APIs (for which we need
a new ANGLE_PLATFORM_EXPORT macro). It also makes those APIs stdcall on
Windows builds.
Change-Id: Iee2ab7e43ef0b98924f5759b5e1b333307052235
Reviewed-on: https://chromium-review.googlesource.com/304501
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Tested-by: Jamie Madill <jmadill@chromium.org>
|
|
|
e826b9f7
|
2015-10-09T16:41:06
|
|
Revert "Fixed compilation on mingw."
This causes some linker warnings. I don't think we should
add linker warnings to Chrome, so reverting this temporarily until
we can land your CL with the correct using of export keywords.
Please re-submit the fixed CL.
This reverts commit 566273222683314dfc8ac1cdb9bf6deb5d2b4c65.
Change-Id: I21fa623af31fac06de6afbefa50c9ea1ec157c15
Reviewed-on: https://chromium-review.googlesource.com/305000
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Jamie Madill <jmadill@chromium.org>
|
|
|
56627322
|
2015-10-05T16:03:10
|
|
Fixed compilation on mingw.
Change-Id: I5734f789f25fbc6b9b90f2043459a3a5f72008c4
Reviewed-on: https://chromium-review.googlesource.com/303602
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Geoff Lang <geofflang@chromium.org>
|
|
|
2e5797e3
|
2015-07-29T16:20:28
|
|
Log EGL initialize errors to the platform.
This will allow us to show more extended error info in about:gpu
in Chromium.
BUG=angleproject:966
BUG=515229
Change-Id: I24c8f33ac3576395ff13dbf7f4664c649af18e33
Reviewed-on: https://chromium-review.googlesource.com/289590
Tested-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Zhenyao Mo <zmo@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
|
|
|
72537ef1
|
2015-05-14T10:16:08
|
|
Add collection of D3D11 and DXGI stats.
Will require a chromium-side patch to enable boolean histograms.
BUG=angleproject:966
BUG=436191
Change-Id: I3026c73894fbbff521503b377955816a8e831df8
Reviewed-on: https://chromium-review.googlesource.com/271174
Tested-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Brandon Jones <bajones@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
|
|
|
ad0a486b
|
2015-04-20T16:32:13
|
|
Add D3D11CreateDevice timing histogram.
This should give us a bound on how much we can improve startup
time in ANGLE on D3D11 across all users.
BUG=angleproject:944
BUG=436191
Change-Id: Ie9047c0424429aecec5f6d7be8e119ebcc53fbe3
Reviewed-on: https://chromium-review.googlesource.com/266524
Tested-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Zhenyao Mo <zmo@chromium.org>
|
|
|
ae2d8a3a
|
2015-04-07T11:21:07
|
|
Add Platform methods for event tracing.
Event tracing requires pulling in timing counters, so we can sync
with Chromium's timestamps. It may require a bit more tricky stuff
to sync ANGLE's GPU timestamps with trace timestamps, but should
allow us to add GPU trace events in the future.
BUG=chromium:436191
BUG=angleproject:966
Change-Id: Ie1dc2981650d96888f988fa74b6fa435fbe8edc2
Reviewed-on: https://chromium-review.googlesource.com/263781
Reviewed-by: Brandon Jones <bajones@chromium.org>
Tested-by: Jamie Madill <jmadill@chromium.org>
|
|
|
6bd58312
|
2015-03-13T10:45:27
|
|
Make Platform methods global.
This will allow them to be imported dynamically without name
mangling. This is necessary because sometimes SwiftShader
overrides libGLESv2 and libEGL, so we need to determine at
run-time if we are running with "actual" ANGLE.
BUG=466735
Change-Id: I396d717b79066feb8ed0d577ee7386b33eb1d160
Reviewed-on: https://chromium-review.googlesource.com/259954
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Tested-by: Jamie Madill <jmadill@chromium.org>
|
|
|
5feea562
|
2015-02-17T16:03:16
|
|
Add ANGLE platform implementation template.
The platform implementation allows the app layer to pass in an
object, on which ANGLE can call virtual methods. Our current
platform will handle trace events, for app profiling, and
histogram records for statistics. The platform approach gives
a much more robust approach than using entry points for every
piece of functionality, and is based on the interop with Blink.
The default platform implementation does a no-op on every call.
The destructor is also private, to ensure we do not call the
destructor of the passed-in class.
BUG=436191
Change-Id: I05641b89a48a9cff81ced059518fceb5aa6c883b
Reviewed-on: https://chromium-review.googlesource.com/248631
Tested-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Zhenyao Mo <zmo@chromium.org>
|