This reverts commit fad918f8.

Reason for revert: attempt to fix anglebug.com/2537

Original change's description:
> ANGLETest: Reuse test windows per-renderer
> 
> When running angle_end2end_tests unfiltered with the OpenGL and Vulkan
> backends enabled, the test window was recreated all the time and grabbed
> focus every-time it was created. This made it impossible to do anything
> with the machine running the tests.
> 
> Fix this by having one OSWindow per renderer group that's lazily created:
> this solves most of the issue since only a couple windows end up being
> created, and at the beginning of the test suite.
> 
> BUG=
> 
> Change-Id: I7a51300f0d59d8b6bb79e54d20b3acbf01068002
> Reviewed-on: https://chromium-review.googlesource.com/1038433
> Commit-Queue: Corentin Wallez <cwallez@chromium.org>
> Reviewed-by: Jamie Madill <jmadill@chromium.org>

TBR=geofflang@chromium.org,jmadill@chromium.org,cwallez@chromium.org

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

Change-Id: I93bdfa38757cbe2a6ce939c0c3e3da806307e7dd
Reviewed-on: https://chromium-review.googlesource.com/1050326Reviewed-by: Frank Henigman <fjhenigman@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Frank Henigman <fjhenigman@chromium.org>

This reverts commit c009255c.

Reason for revert: Seems to be some kind of infra or goma error with
the mac configs that wasn't showing up last night.

Original change's description:
> Update cq.cfg with new builders.
> 
> These builders are replacing the old standalone builders and the old
> compile-only builders. They should be much faster. They also can be
> extended in the future when we support running tests from ANGLE
> standalone.
> 
> Bug: chromium:833999
> Change-Id: Ice44c0fb8cb32d8be573f81d5df858509b00a107
> Reviewed-on: https://chromium-review.googlesource.com/1049959
> Reviewed-by: Jamie Madill <jmadill@chromium.org>

TBR=fjhenigman@chromium.org,geofflang@chromium.org,jmadill@chromium.org

Change-Id: I8bf895d7473da7fe44a3bb866a2026b59a374868
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: chromium:833999
Reviewed-on: https://chromium-review.googlesource.com/1050325Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

These builders are replacing the old standalone builders and the old
compile-only builders. They should be much faster. They also can be
extended in the future when we support running tests from ANGLE
standalone.

Bug: chromium:833999
Change-Id: Ice44c0fb8cb32d8be573f81d5df858509b00a107
Reviewed-on: https://chromium-review.googlesource.com/1049959Reviewed-by: Jamie Madill <jmadill@chromium.org>

BUG=angleproject:2464

Change-Id: Ia0f0788a1bc4d2ecd883b058f15c629cac5fd166
Reviewed-on: https://chromium-review.googlesource.com/1036063
Commit-Queue: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>

quadVertices.data() pointer needs to remain valid until drawArrays
is called. To guarantee that quadVertices should stay in the stack
until the end of drawQuad().

BUG=angleproject:2533
TEST=angle_end2end_tests

Change-Id: I263262db9a9ed9c9cd0d8679addd09677ad7179e
Reviewed-on: https://chromium-review.googlesource.com/1046600
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>

When running angle_end2end_tests unfiltered with the OpenGL and Vulkan
backends enabled, the test window was recreated all the time and grabbed
focus every-time it was created. This made it impossible to do anything
with the machine running the tests.

Fix this by having one OSWindow per renderer group that's lazily created:
this solves most of the issue since only a couple windows end up being
created, and at the beginning of the test suite.

BUG=

Change-Id: I7a51300f0d59d8b6bb79e54d20b3acbf01068002
Reviewed-on: https://chromium-review.googlesource.com/1038433
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>

This patch is the first one of the series that organize
implementation dependent resource limits on every type
of shader into ShaderMap and clean up all the related
code.

In the next patch all such resource limits are put in
the corresponding ShaderMaps.

BUG=angleproject:2169

Change-Id: I40cb58c55b2e82df33221ddb36eff0abcd7e8b22
Reviewed-on: https://chromium-review.googlesource.com/1034108
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>

BUG=angleproject:2527

Change-Id: I000f68366de4119375b6d3c79bc2eff3ebd5db9e
Reviewed-on: https://chromium-review.googlesource.com/1042885
Commit-Queue: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>

This works around the most common instances of a bug that reproduces
on some NVIDIA OpenGL drivers prior to version 397.31.

BUG=chromium:798117
TEST=angle_end2end_tests

Change-Id: Iafc6a9a64e56fa98b42117149fe6867040e932e5
Reviewed-on: https://chromium-review.googlesource.com/1042190
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>

This patch adds the link validation on the maximum combined shader
output resources required in OpenGL ES 3.1 SPEC.

OpenGL ES 3.1 SPEC has restrictions on the sum of the number of all
active images, shader storage blocks and fragment shader outputs. A
link error will be generated if this sum exceeds the implementation-
dependent value of MAX_COMBINED_SHADER_OUTPUT_RESOURCES.

In order not to affect the existing image tests, this patch also
sets a temporary value for maxCombinedShaderOutputResources on D3D11
back-ends. We will set more accurate values for all the UAV related
resource limits in the next patch.

BUG=angleproject:2345
TEST=dEQP-GLES31.functional.state_query.integer.max_combined_shader_output_resources_*

Change-Id: Ib83a19ef0ae0b9af3422b5c970c7c07d96b2359d
Reviewed-on: https://chromium-review.googlesource.com/1039155
Commit-Queue: Jiawei Shao <jiawei.shao@intel.com>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>

Add generation of libGLESv2.cpp and libGLESv2.def to generate_entry_points.py

Bug: angleproject:2476
Change-Id: I86c7fb31f73ccbbbc32b28acca179b11527dff9e
Reviewed-on: https://chromium-review.googlesource.com/1040647
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>

The D3D11 backend used to calculate the blit areas incorrectly when
the source blit area provided via the API stretched outside the source
framebuffer in all directions.

This is now fixed and the code is also simplified.

BUG=angleproject:2521
TEST=angle_end2end_tests

Change-Id: I46d5de27464a8455b3956c0be598f75550ca1616
Reviewed-on: https://chromium-review.googlesource.com/1032787Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>

avoid1BitAlphaTextureFormats workaround was added to angle in 288584
due to an old driver bug on Intel and AMD that 1-bit alpha always
rounds up when converting from float to unsigned int. The workaround
uses rgba8 instead of *a1 to avoid the driver bug and optimizes precision.
However, this workaround brings a new issue because 1-bit alpha can only
represent u0 or u1 while 8 bits alpha covers from u0 to u255. For
example, if we expect to render to a rgb5_a1 renderbuffer, the expected
alpha value should be 0 or 255, but actually get a value between 0 and
255 which is incorrect. The current Intel drivers have fixed the old
driver bug. So we suggest to disable this workaround for Intel drivers.

TEST=dEQP_GLES2.Default/functional_fbo_render_color_clear_rbo_rgb5_a1_depth_component16_stencil_index8
--deqp-egl-display-type=angle-gl
BUG=angleproject:2349

Change-Id: I14933f92fa27031ff7442fa437f77a3c67f2f1db
Reviewed-on: https://chromium-review.googlesource.com/1034163Reviewed-by: Jiajia Qin <jiajia.qin@intel.com>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>

The test fails on D3D11, which seems to ignore floor() calls if
they're nested in a particular way with magic multipliers inside.

BUG=chromium:838885
TEST=angle_end2end_tests

Change-Id: I2c8749defa923553bfee69149da15f8e4ddf5096
Reviewed-on: https://chromium-review.googlesource.com/1041959Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>

Bug: angleproject:2418
Change-Id: Ia146c202cff96b1de9cb55ee44fa5479cef36b53
Reviewed-on: https://chromium-review.googlesource.com/1039777Reviewed-by: Courtney Goeltzenleuchter <courtneygo@google.com>
Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Commit-Queue: Tobin Ehlis <tobine@google.com>

Renames the macro to be more ANGLE-ey.

Refactoring only.

Bug: angleproject:1671
Change-Id: I8f2dd227c7e2025886ec66e85efa877ea261d0ad
Reviewed-on: https://chromium-review.googlesource.com/1036209Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

BUG=angleproject:2464

Change-Id: I59fd38c626f7076b4065f25601de3e53c1a446ad
Reviewed-on: https://chromium-review.googlesource.com/1040051Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>

BUG=angleproject:2169

Change-Id: I129a9d8e295859daa071a298dab9fe1895315cc0
Reviewed-on: https://chromium-review.googlesource.com/957318Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>

This implements masked color clear using clear shaders. The shaders
themselves were introduced in a prior patch. In order to get the
right setup for the draw call to trigger the shaders, we create
an internal pipeline from the pipeline cache. We also use a special
pipeline layout with only uniform buffers. The masked out color
channels are disabled via settings on the pipeline.

This fixes the dEQP masked color clear tests. It doesn't handle
masked color clears combined with the depth clear bit. It's likely
we don't have test coverage for this case.

Bug: angleproject:2455
Change-Id: I513248cc0f7e58f490fc16ac9afb40119d730ccc
Reviewed-on: https://chromium-review.googlesource.com/1031373
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>

This makes the API easy to use with internal shaders and pipelines.
This is useful for the implementation of masked color clear.

Also renames the serials as shader serials. This is more precise than
program serials.

Bug: angleproject:2455
Change-Id: Ie6247d1212ed4df856b561a5e9f16c0378202588
Reviewed-on: https://chromium-review.googlesource.com/1032857
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>

Even on the highest optimization level MSVC was still generating extra
code in Release for ASSERT macros. This realigns the code more with
Chromium by using a streaming eating macro with the expression at the
end. Now no code should be generating. I believe this did not affect
Clang builds.

Bug: angleproject:1671
Change-Id: I099c59cf0b28ae3ffc08b864982e67559901724e
Reviewed-on: https://chromium-review.googlesource.com/1036208Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

This re-uses the same machinery as the GL front-end. It saves a lot of
custom casting and array sizing. Currently we only support vertex and
fragment shaders in Vulkan. Because of the Pipeline cache sizing, it's
easier to stick with just VS/FS and introduce the full set of shaders
when we move to packing the cache better.

Bug: angleproject:2522
Bug: angleproject:2455
Change-Id: I21432a335c741885af87970d8ee52b4a36338304
Reviewed-on: https://chromium-review.googlesource.com/1036927Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Jiawei Shao <jiawei.shao@intel.com>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

BUG=angleproject:1874

Change-Id: I8898c1a467c1fec25963f45a31376a6b02e1b6c4
Reviewed-on: https://chromium-review.googlesource.com/1037584Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>

We can validate to avoid triggering driver issues even if the native
GLES spec is not telling us to.

This will fix WebGL in Chromium when it's being run with the
--use-cmd-decoder=validating --use-angle=gl config.

BUG=chromium:830046
TEST=WebGL conformance tests

Change-Id: I2d61182cb6cbe46b52e1d9b7ed6b4035defee082
Reviewed-on: https://chromium-review.googlesource.com/1033743
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>

Simplify handling of reversing the blitted area and calculate shader
parameters in a way that doesn't require lengthy explanation.

BUG=chromium:830046
TEST=angle_end2end_tests

Change-Id: Ie9bce812be4ef04a969153fc7c484039fc48a198
Reviewed-on: https://chromium-review.googlesource.com/1030172
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>

This allows Framebuffer syncState methods to handle dirty bits
themselves. This will be useful for the implementation of masked color
clear.

Bug: angleproject:2455
Change-Id: I65ad9c61e89e317c820f6be5550edd8185b52afc
Reviewed-on: https://chromium-review.googlesource.com/1032856
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Luc Ferron <lucferron@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>

Vulkan intenal shaders are stored in a ShaderLibrary, and this is
owned by the RendererVk. This way the shaders are reused between all
the different Contexts. They are initialized lazily to keep init time
low. They also have an associated Serial (called a ProgramSerial) so
they can be identified in a PipelineDesc (used by the Pipeline cache).

We use a python script to build and invoke the glslang validator, that
also produces SPIR-V binary code snippets. These snippets are gathered
into an auto-generated file that is exposed via an auto-generated
header file. The InternalShaderID enum class gives access to the
internal shaders that are shared through the Vulkan back-end.

This also adds simple clear shaders to be used in masked color clears.
The patch doesn't add any functionality but it is split off from the
color clear functionality to keep the code size down.

Bug: angleproject:2339
Bug: angleproject:2455
Change-Id: Ie83043eda217c9f013817b198c92a3b7ba0878b4
Reviewed-on: https://chromium-review.googlesource.com/1031372
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>

Bug: angleproject:2474

Change-Id: Ia558c2e266422f8f8b55523c9542379688a7058d
Reviewed-on: https://chromium-review.googlesource.com/1037164
Commit-Queue: Luc Ferron <lucferron@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>

1- Link program with a shader and draw with it, relink with another shader and draw again.
2- Release program that is reading from a uniform.

Bug: angleproject:2397

Change-Id: Icb4211c5cf71efaf41833d9d5afd83ce8410c598
Reviewed-on: https://chromium-review.googlesource.com/1028580
Commit-Queue: Luc Ferron <lucferron@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>

This was causing very slow builds/test runs.

Bug: chromium:837166
Bug: chromium:834269
Change-Id: If2e5665455d4a8af13cbc732a65a07550ace8304
Reviewed-on: https://chromium-review.googlesource.com/1036220Reviewed-by: Jamie Madill <jmadill@chromium.org>

Bug: angleproject:2418

Change-Id: I793d330a2cc92a510b189fa23d14f41f48a7d986
Reviewed-on: https://chromium-review.googlesource.com/1035422Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Commit-Queue: Courtney Goeltzenleuchter <courtneygo@google.com>

The mtime values I was seeing on my file system were sometimes
floating point numbers that were very very close but not exactly
the same. This might be due to numerical errors. Increasing the
tolerance to something still very small, precision is less than
a second. It will lead to fewer rebuilds of non-dirty files.

Bug: angleproject:2455
Change-Id: I95dc3214ee91af7a70a20cc625405e0e2bc18698
Reviewed-on: https://chromium-review.googlesource.com/1032855Reviewed-by: Luc Ferron <lucferron@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

These weren't needed in many places.

Also renames one FramebufferVk method.

Bug: angleproject:2455
Change-Id: Idb641094fa3e180a85f357533d86bd0b19db4ec8
Reviewed-on: https://chromium-review.googlesource.com/1024826Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Luc Ferron <lucferron@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

This patch intends to remove redundant computations on combined
interface block counts in ValidateGraphicsInterfaceBlocks.

In this patch, we compute and check the number of combined interface
blocks by the result of ValidateInterfaceBlocksCount directly instead
of re-compute it, and if the check on max combined interface blocks
failed, Program::link() can early return without entering
ValidateGraphicsInterfaceBlocks.

BUG=angleproject:2345

Change-Id: I7573f7c645993b4d75230a8471203a305127f2a3
Reviewed-on: https://chromium-review.googlesource.com/1031852Reviewed-by: Jiajia Qin <jiajia.qin@intel.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Jiawei Shao <jiawei.shao@intel.com>

This patch implements GL_REFERENCED_BY_GEOMETRY_SHADER as a valid
property of program interfaces.

BUG=angleproject:1941
TEST=angle_end2end_tests
     dEQP-GLES31.functional.geometry_shading.query.referenced_by_geometry_shader

Change-Id: Id9659313d371dbfc7d00bf9d816df4449fbf29ec
Reviewed-on: https://chromium-review.googlesource.com/1025281Reviewed-by: Jiajia Qin <jiajia.qin@intel.com>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Jiawei Shao <jiawei.shao@intel.com>

The index range was being resolved in the call to
VertexArray11::syncStateForDraw but the code which invalidates the input layout
was being executed before this.

BUG=837002

Change-Id: I4e00b53cdc51f8758b0e2ba9e2dfc93b5e22556c
Reviewed-on: https://chromium-review.googlesource.com/1032633
Commit-Queue: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>

BUG=angleproject:1472
BUG=621240

Change-Id: I723839bd7961167adddcccc680638ae066eacea3
Reviewed-on: https://chromium-review.googlesource.com/1025645Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>

BUG=837717

Change-Id: I1ac659b18bd9aef2f24f798d23f49d1efbb81938
Reviewed-on: https://chromium-review.googlesource.com/1033478Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>

Scaling and flipping of the framebuffer with blitFramebuffer are now
covered by end2end_tests. The test configurations are set so that the
BlitGL shader-based blit is also covered.

The tests reveal a bug in the D3D11 backend implementation of
blitFramebuffer when using an oversized source area.

BUG=angleproject:2486
BUG=angleproject:2519
BUG=angleproject:2521
TEST=angle_end2end_tests

Change-Id: Ica48e564a5b4180241ec4c4c6be06c58cc6dc29b
Reviewed-on: https://chromium-review.googlesource.com/1030171Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>

This patch refactors ProgramD3D by storing all shader information
into ShaderMap to simplify the code structure.

This patch also fixes a bug on getting the number of maximum uniform
blocks.

BUG=angleproject:2169

Change-Id: I5b9fbfd70a18f8731ce19efed0df88037d495389
Reviewed-on: https://chromium-review.googlesource.com/1024749
Commit-Queue: Jiawei Shao <jiawei.shao@intel.com>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>