In case the initializer of an unsized array is not an array, the array
size still needs to be set to some value > 0 in order to not hit
asserts in the code that parses accessing the array. An error was
already being generated in the case an unsized array has a non-array
initializer, but the variable will still have an array type in the
symbol table.

BUG=chromium:661592
TEST=angle_unittests

Change-Id: I4a11527eab0404ba9e59ebb7227faef13dbea62c
Reviewed-on: https://chromium-review.googlesource.com/407256Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>

We can't fix them yet because our continuous testing doesn't compile
with the updated definitions yet.

BUG=angleproject:1598

Change-Id: I40bdeaa6bafbd03b5e3e6c16ac8485e3f488fe59
Reviewed-on: https://chromium-review.googlesource.com/406452
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>

BUG=angleproject:1332

Change-Id: I69a5780521a4810ae93567fbe875a8c5e361850e
Reviewed-on: https://chromium-review.googlesource.com/406632Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>

These flags can come from both attributes and bitfield flags.

Also move the robust access check to an UNIMPLEMENTED in Context, dEQP ES 3.1
tests attempt to use this flag and it's convenient to be able test by disabling
UNIMPLEMENTED assertions.

BUG=angleproject:1442

Change-Id: I1641fdb3afee308235351ebd60be680765b16b3b
Reviewed-on: https://chromium-review.googlesource.com/405774
Commit-Queue: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>

It is not feasible to run dEQP against ES 3.1 on the bots without removing this
assertion.  Messages will still be logged and UNREACHABLE will still generate
assertion failures when the behaviour is needed.  This matches the behaviour
of Chromium's NOTIMPLEMENTED macro.

Remove the ANGLE_TEST_CONFIG macro, it was not doing anything useful anymore.

BUG=angleproject:1442

Change-Id: I76376817a87e1a75ad1a938e1db1d7763e094372
Reviewed-on: https://chromium-review.googlesource.com/406928Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>

This should fix any potential out of bounds reads/writes.

BUG=chromium:660854

Change-Id: Iffa00e4551d7362115cbf023a09b1d0e15f724c8
Reviewed-on: https://chromium-review.googlesource.com/405816
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>

Use common helper functions instead of manually creating temporary
variable nodes. Also clean up the interface provided by the traverser.

BUG=angleproject:1597
Test=WebGL conformance tests

Change-Id: Ifd8d3815ff9e75e1a2040d65db9d4b3d6a9a9273
Reviewed-on: https://chromium-review.googlesource.com/403950Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>

Prefixing internal names will avoid clashes with user-defined names.
We use prefix webgl_angle_ so that the internal names are both
reserved in WebGL on one hand and clearly separated from hashed names
starting with just webgl_ on the other hand.

BUG=angleproject:1597
TEST=WebGL conformance tests

Change-Id: I7deef9c1a38105c8b57eda13c84eec13e515a91a
Reviewed-on: https://chromium-review.googlesource.com/406247Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>

This removes the global std::set initialization.

BUG=angleproject:1389
BUG=angleproject:1459

Change-Id: I6a7f4211905ea4a83e0e2337977e2f6fb375a7dd
Reviewed-on: https://chromium-review.googlesource.com/405368Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

This brings it in line with the new style.

BUG=angleproject:747

Change-Id: Ie4b223d018999be18c3b19fae9cc825a78de09a9
Reviewed-on: https://chromium-review.googlesource.com/405815Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

Update code to use gl::Version in as many places as possible to ease ES 3.1
support.

BUG=angleproject:1588

Change-Id: I3490b53a81027cf849dac551a9cc66ce04506144
Reviewed-on: https://chromium-review.googlesource.com/404946Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>

The fuzzer was generating cases where floating point constants would
someone return as valid when they were beyond max float. I was unable
to reproduce this in a standalone test, but confirmed this check
fixes the fuzzer error.

BUG=660702

Change-Id: I9e6b883958013638ea509e38b30e812a9e74fbc1
Reviewed-on: https://chromium-review.googlesource.com/406268Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

BUG=angleproject:1442

Change-Id: I9005a54d30988a472a3f512f394a4a352a4732e9
Reviewed-on: https://chromium-review.googlesource.com/405297
Commit-Queue: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>

BUG=angleproject:1144

Change-Id: I6186381d1aad13bbb07af47b3cf94d6ec7cbb566
Reviewed-on: https://chromium-review.googlesource.com/405693Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>

This removes another static global map initializer.

BUG=angleproject:1389
BUG=angleproject:1459

Change-Id: I302f349ea4e95d079077866a07b0d57758c40892
Reviewed-on: https://chromium-review.googlesource.com/405488
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>

This removes a static global map initialization.

BUG=angleproject:1389
BUG=angleproject:1459

Change-Id: I6eb19f6649280e72682c6806ea0bf6eb70173984
Reviewed-on: https://chromium-review.googlesource.com/405487Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

This should prevent further unexpected bot breakage due to
unreferenced variables in the ASSERT expression.

Also remove the no longer needed variable referencing macro.

BUG=angleproject:1586

Change-Id: I127695165bdfe39c51fe8d17e00daf6bf2fa8252
Reviewed-on: https://chromium-review.googlesource.com/404948Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

The system memory storage is needed to back UBOs larger than the max
constant buffer size in D3D11 Windows 7. Because readback from UBOs
can be tricky, the system memory storage keeps the canonical copy of
the data in this case.

We can also extend this to add a workaround to fix the outstanding
failure cases of UBOs on Intel.

BUG=chromium:660670

Change-Id: Ia3119d3064d10c4262def4c5a967972b4de2d10f
Reviewed-on: https://chromium-review.googlesource.com/405367Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

BUG=chromium:659472

Change-Id: Icfeada9f09fff5d28eb3f4fafb2c94d57598af1f
Reviewed-on: https://chromium-review.googlesource.com/403849
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>

BUG=

Change-Id: I7ca6e6c0fc4b178a1df6d6acc95dde3b6ed3c479
Reviewed-on: https://chromium-review.googlesource.com/403089Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>

On some Intel D3D drivers on Skylake, calling clear may take no effect.
To work around this bug, we call clear() twice when we attempt to do clear
on these platforms.

BUG=chromium:655534

Change-Id: I3b05892e80e5797d268861e8f52e080f5175855b
Reviewed-on: https://chromium-review.googlesource.com/398522
Commit-Queue: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>

This relands https://chromium-review.googlesource.com/#/c/400005/.

ESSL and GLSL are not consistent on invariant matching in vertex shader
and fragment shader. See the following rules:
ESSL 1.00 - input and output must match
ESSL 3.00 - only output, inputs cannot be declared as invariant.

GLSL 1.10.59 - does not exist
GLSL 1.20.8  - input and output must match
GLSL 1.30.10 - input and output must match
GLSL 1.40.8  - input and output must match
GLSL 1.50.11 - input and output must match
GLSL 3.30.6  - input and output must match
GLSL 4.00.9  - input and output must match
GLSL 4.10.6  - input and output must match
GLSL 4.20.11 - input can omit invariant
GLSL 4.30.8  - input can omit invariant
GLSL 4.40.9  - input can omit invariant
GLSL 4.50.5  - input can omit invariant

Since GLSL 4.20, invariant qualifier description were changed to:
"
Only variables output from a shader (including those that are then input
to a subsequent shader) can be candidates for invariance. This includes
user-defined output variables and the built-in output variables. As only
outputs need be declared with invariant, an output from one shader stage
will still match an input of a subsequent stage without the input being
declared as invariant.
"
It's not very clear if input in fragment can be declared as invariant.
Mesa driver disallows use of input declared as invariant in fragment
shader, while other drivers may allow it. This CL removes invariant
declaration for input in fragment shader except AMD driver in Linux.
AMD's driver obviously contradicts the spec by forcing invariance to
match between vertex and fragment shaders.

BUG=chromium:639760, chromium:659326
TEST=conformance/glsl/misc/shaders-with-invariance.html and
conformance/glsl/bugs/invariant-does-not-leak-across-shaders.html

Change-Id: I0aa9be14f0cee7a11a249c91fba27c570c52ca1b
Reviewed-on: https://chromium-review.googlesource.com/404228Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Yuly Novikov <ynovikov@chromium.org>

When mapping uniform buffers for very large buffers, we could end
up copying past buffer bounds. This would trigger crashes on some
configs. We can fix this by clamping the copy size correctly.

BUG=chromium:659892

Change-Id: I9d1af984da34867692d4c7b8908c016ebec7a63b
Reviewed-on: https://chromium-review.googlesource.com/404931Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

The best practice is to expose a non-constexpr query function and
hide the constexpr table in a cpp, so it won't be included verbatim
in multiple other source files.

BUG=angleproject:1389

Change-Id: Ic4e19d14c832643be1552c5a14e441826c41970f
Reviewed-on: https://chromium-review.googlesource.com/404478Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

The DEPS for buildtools changed to not recurse automatically. This
change mirrors the change in Chromium.

See https://chromium.googlesource.com/chromium/buildtools/+/3d2e47bf14e4e67816a53e304dea422fa18f9180%5E%21/#F0

BUG=None

Change-Id: Ib15520324ad7894d4a640a0e2013c3823849e846
Reviewed-on: https://chromium-review.googlesource.com/404969Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

In function resetOffscreenColorBuffer, the local variable 'result'
is never used in its scope, so when attempting to build a release
version of Chromium with latest ANGLE on Windows, ninja will report
a compile warning and stop the compilation: 'result': local variable
is initialized but not referenced.
The build failure starts at:
https://chromium-review.googlesource.com/#/c/316804/

Change-Id: I29c05653f0d2e4f34c2e0402848f403ec2af8e5a
Reviewed-on: https://chromium-review.googlesource.com/404515Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>

BUG=angleproject:1585

Change-Id: I550750e0e94bd15c0b076f911d05f8dd92e4a202
Reviewed-on: https://chromium-review.googlesource.com/404499Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Commit-Queue: Yuly Novikov <ynovikov@chromium.org>

This should guarantee the best memory access patterns.
It introduces some indirections for some format queries,
but most of these should be direct array lookups, or used
infrequently. We can optimize this later if necessary.

BUG=angleproject:1389

Change-Id: I5e2c8c530a07798494afd3ea36b6164d7564c02c
Reviewed-on: https://chromium-review.googlesource.com/403314
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Yuly Novikov <ynovikov@chromium.org>

BUG=angleproject:1547
BUG=655247

Change-Id: I3f04ddc7032e4a47eb21ff3b8586c5b47415bb64
Reviewed-on: https://chromium-review.googlesource.com/400958
Commit-Queue: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>

Precision qualification is now checked properly also for declarations
that don't have qualifiers.

BUG=angleproject:1574
TEST=angle_unittests

Change-Id: I3d186df0763e071614c1da9a355a6f6fefdc8091
Reviewed-on: https://chromium-review.googlesource.com/403949
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>

Allows creation of pbuffers from D3D texture objects.

BUG=540829
BUG=angleproject:1144

Change-Id: If8ea717ef011608cd01357c217837133d726d3ea
Reviewed-on: https://chromium-review.googlesource.com/316804
Commit-Queue: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>

This patch intends to add platform detection to ANGLE and tighten
the driver bug workarounds on Intel GPU.

BUG=angleproject:1548

Change-Id: I1ea57e174f688a175da8b658de4337295037fcab
Reviewed-on: https://chromium-review.googlesource.com/399914
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>

The new class TIntermDeclaration is now used for struct, interface
block and variable declarations. TIntermDeclaration nodes do not have
a type - rather the type is stored in each child node. The types may
differ in case the declaration is a series of array declarators with
mismatching sizes.

TIntermAggregate is still used for function calls, function
prototypes, function parameter lists and invariant declarations.

BUG=angleproject:1490
TEST=angle_unittests

Change-Id: I0457188f354481470855f61ac1c878fc2579b1d1
Reviewed-on: https://chromium-review.googlesource.com/400023
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>

BUG=angleproject:1354

Change-Id: I596ebd6d8d4d500942eeb31629113b0686933531
Reviewed-on: https://chromium-review.googlesource.com/403537Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>

BUG=658555

Change-Id: I578b8aff37a116fd7b2b387388311a27bb8a2809
Reviewed-on: https://chromium-review.googlesource.com/403848Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>

UpdateSubresource and CopySubresource cannot copy to depth stencil textures
in feature level 10.0.

BUG=angleproject:1572

Change-Id: Ic9f8cbfb2c69de8502da7df34f1c5935144a9387
Reviewed-on: https://chromium-review.googlesource.com/402554
Commit-Queue: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>

Introduced in cd2ed608.

BUG=angleproject:1570

Change-Id: I71ef8a06e379eca853e7729863976e868066df95
Reviewed-on: https://chromium-review.googlesource.com/403313Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

The DynamicHLSL code was using global strings to find stubs.
The math code was also using pow in a global, which can be replaced
with a simple bit shift.

BUG=angleproject:1459

Change-Id: Idb0602ab7640c221843385b0a0a4cfecd5fd3a26
Reviewed-on: https://chromium-review.googlesource.com/403289Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

This reverts commit d842a6b2.

Because of WebglConformance_conformance_glsl_bugs_invariant_does_not_leak_across_shaders failure

BUG=chromium:659326

Change-Id: I0602e24f3d34ccf852cda865f673c5c7634f82a6
Reviewed-on: https://chromium-review.googlesource.com/403230Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Commit-Queue: Yuly Novikov <ynovikov@chromium.org>

There are 6 crashes in SwapChain11::getSyncValues on crash server.
I looked at one of them and it is due to a nullptr mSwapChain
which appears to be a valid state that is handled in other
SwapChain11 methods.

This change adds a similar check / returning an error to
SwapChain11::getSyncValues.

BUG=angleproject:1402

Change-Id: I9c17b884704b8aaaddf2d802911cd0ca961ae0b4
Reviewed-on: https://chromium-review.googlesource.com/402613Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Stanislav Chiknavaryan <stanisc@chromium.org>