This instruments the plumbing for the extension without adding any
functionality or exposing the extensions string.

The extension text is also updated to reflect the new entry point
design and naming. Also this corrects a few mistakes.

This will be followed up by the tests (which won't run) and then
the extension functionality in ANGLE.

BUG=angleproject:1897

Change-Id: I5b009e23bc27da06b067375525bd6fc574027702

The validation was iterating over maxDrawBuffers attachments when
default framebuffers only have one attachment. Use the framebuffer's
drawBufferCount instead.

Also adds a regression test in the form of a WebGLComptibility test for
glClearBuffer with the default framebuffer.

BUG=angleproject:2091

Change-Id: I07ee524db1fcb8a99dab4043248c0885100fd216

The patch enables viewport selection for multiview rendering in the
GLSL/ESSL vertex shader through the use of the GL_NV_viewport_array2
extension. The AST is modified only for GLSL and ESSL to include the
viewport selection expression after ViewID_OVR's initialization.

BUG=angleproject:2062
TEST=angle_unittests

Change-Id: Iee05bb5a4b687ed53ddbdd466f1572227b1f0cde

Previously, arrays were being incorrectly accepted as array indices.
This was because the isScalar() check only checked that the type was
not a vector or matrix, but still returned true for scalar arrays.

This patch changes the isScalar() check so that it returns false for
arrays. This makes usage of the term "scalar" more consistent in the
shader translator. Most of the code using isScalar() was compatible
with this change. Code in util.cpp that used to assume that isScalar()
doesn't care about arrayness is refactored to work with the new
behavior.

BUG=angleproject:2102
TEST=angle_unittests

Change-Id: I2a7f4c30fca7917d1099d0400efe3de859338b2a

This includes asserts in TConstantUnion to reveal incorrect usage of
union - reading a different field of an union that has last been set
is undefined behavior in C++.

Existing issues with accessing incorrect fields of constant unions
are fixed.

BUG=angleproject:1490
TEST=angle_unittests

Change-Id: Idd6b7a871d73e2928f117a9348c92043612fab82

Most of the functions were just simple wrappers around node
constructors. Dropping this extra redirection makes the code simpler.

The fold() functions of node types are simplified, so that if the node
can't be folded the pointer to the node itself is returned. This makes
the code in ParseContext more straightforward.

The few remaining functions in Intermediate are a bit more complex so
they should be handled separately, but they'll be removed eventually
as well.

BUG=angleproject:1490
TEST=angle_unittests

Change-Id: I85e11919d1f62358cfba9c011b841e32bc25402f
Reviewed-on: https://chromium-review.googlesource.com/563393Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>

The patch adds a function in MatchOutputCodeTest to get the position
of the first occurrence of an expression in the translated output code.
This can help design more complicated tests like a test which checks
for the order of simple expressions.

BUG=angleproject:2062
TEST=angle_unittests

Change-Id: I1249d4762c247848c4eec64ecb8c1357b5e8d40a
Reviewed-on: https://chromium-review.googlesource.com/563659Reviewed-by: Olli Etuaho <oetuaho@nvidia.com>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>

The copied data is not supposed to have sRGB conversions applied to it
when written to the destination texture but an sRGB SRV is used by Blit11.
Instead of creating multiple sRGB and non-sRGB SRVs for textures, simply
fall back to the CPU copy path for this format for now.

Clip color channels that should not exist in the destination texture
formats in Image11::CopyImage.  This works around issues with texture
formats with emulated channels.

TEST=conformance2/textures/image_bitmap_from_canvas/tex-2d-srgb8-rgb-unsigned_byte
TEST=conformance2/textures/image_bitmap_from_canvas/tex-2d-srgb8_alpha8-rgba-unsigned_byte

BUG=angleproject:1932

Change-Id: Ieeda3569f80d016fda781e7eb498acd3b97568d0
Reviewed-on: https://chromium-review.googlesource.com/559857Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>

queueReplacement is always called to replace the node that's currently
being visited in the traverser. The currently visited node can be
fetched automatically from the traversal path so it can be removed
from parameters of queueReplacement.

BUG=angleproject:2100
TEST=angle_unittests

Change-Id: I62ab6d1cd9c0d2b4c260af9f7c85bc156fb3f349
Reviewed-on: https://chromium-review.googlesource.com/562336Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>

Two simple fixes to make libANGLE compile for Fuchsia:
1. Added stubs typedefs for platform-specific primitives in
   EGL/eglplatform.h for Fuchsia.
2. src/common/platform.h updated to define ANGLE_PLATFORM_POSIX on
   Fuchsia.

Change-Id: I5f053de46632ac3f12bb93e57eeb94c3a34c867c
Reviewed-on: https://chromium-review.googlesource.com/563758Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>

Using a depth buffer format requires that the DSV and RTV have equal
quality levels; otherwise, the bound render target will discard writes.

BUG=angleproject:1917

Change-Id: Ife25b0a8958fa2b31b43a0d877d27e440916a9bf
Reviewed-on: https://chromium-review.googlesource.com/560716Reviewed-by: Bryan Bernhart <bryan.bernhart@intel.com>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>

The OVR_multiview specification states that gl_ViewID_OVR is visible
at each pipeline stage. Previously to this patch the ViewID_OVR
impostor was declared only in the vertex shader and occurrences of
gl_ViewID_OVR in the fragment shader were not being handled. The
patch addresses the issue by declaring the ViewID_OVR variable as
a vertex output in the vertex shader and as a fragment input
in the fragment shader.

BUG=angleproject:2062
TEST=angle_unittests

Change-Id: I895953e81d3632d9bb873e8ac081fdf36f63f6b7
Reviewed-on: https://chromium-review.googlesource.com/559337
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>

The patch adds functionality to determine whether compiler output belongs
to the ESSL, GLSL, HLSL or Vulkan output family. The new functions can
be now used in other parts of the compiler in which code paths are
selected based on the compiler output.

BUG=angleproject:2062
TEST=angle_unittests

Change-Id: I45ccf63f0a756c60df47a679c2da9f60856d5918
Reviewed-on: https://chromium-review.googlesource.com/558990Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Olli Etuaho <oetuaho@nvidia.com>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>

Define TIntermTraverser and TIntermLValueTrackingTraverser in a
separate header file. hash() function is moved out from
TIntermTraverser as it is not related to the core functionality
of traversing and transforming ASTs.

Also reorganize some traversers to follow common conventions:

- Intermediate output is now in OutputTree.h/.cpp
- Max tree depth check is now in IsASTDepthBelowLimit.h/.cpp

BUG=angleproject:1490
TEST=angle_unittests

Change-Id: Id4968aa9d4e24d0c5bac90dc147fc9f310de0184
Reviewed-on: https://chromium-review.googlesource.com/559531Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>

The previous code was using a completely unrelated variable to compute
max combined uniform blocks limit.

BUG=angleproject:2099

Change-Id: I9e56b50a92790f525dda50adca52b6ac5edfc95a
Reviewed-on: https://chromium-review.googlesource.com/562276
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>

While compiling ESSL1 shaders, with the compiler having both
SH_INITIALIZE_BUILTINS_FOR_INSTANCED_MULTIVIEW and SH_VARIABLES
set, variable collection terminates with an assertion failure. The reason
behind this is that SH_INITIALIZE_BUILTINS_FOR_INSTANCED_MULTIVIEW
adds gl_InstanceID to the AST to initialize the multiview builtins, but
the variable collection pass cannot find gl_InstanceID information in
the symbol table because the builtin is only available in ESSL 3.00 and
greater.

To address this the patch populates the gl_InstanceID attribute
information explicitly in the variable collection pass instead of
retrieving it from the symbol table.

BUG=angleproject:2062
TEST=angle_unittests

Change-Id: I5ecb9967ebe6658e956d17a2637090f9b685ef33
Reviewed-on: https://chromium-review.googlesource.com/559669Reviewed-by: Olli Etuaho <oetuaho@nvidia.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>

Temporary symbols used to all have symbol id 0. Now they get assigned
unique symbol ids. This makes it possible to keep track of them
according to the symbol id instead of their name, paving way to more
robust AST handling in the future.

BUG=angleproject:1490
TEST=angle_unittests

Change-Id: I292e2e483cc39173524fd30a30b48c4c808442e5
Reviewed-on: https://chromium-review.googlesource.com/559335Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>

TEST=conformance2/glsl3/no-attribute-vertex-shader
TEST=deqp/functional/gles3/instancedrendering

BUG=angleproject:2012

Change-Id: I234410fabf6a8fcd87040c8085ca5dce82fa8932
Reviewed-on: https://chromium-review.googlesource.com/559851
Commit-Queue: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>

Reserved function names are now caught if the function is just
declared without being called in the shader source. Actually, function
calls don't need to be checked for reserved names, since that just
generates a redundant error message if function declarations are being
checked.

Includes some cleanup of ParseContext::checkIsNotReserved. It doesn't
need special handling of built-in symbols, as they are never passed to
the function.

BUG=chromium:739448
TEST=angle_unittests

Change-Id: I7115e1a7509626b5109b5c054c0704b0c3c19c58
Reviewed-on: https://chromium-review.googlesource.com/561457Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>

Also split the ANGLE modifications to the header into a separate
include.

BUG=angleproject:1897

Change-Id: I06cdcc26bab6bfa4b5bcb488c46558583cab5a83
Reviewed-on: https://chromium-review.googlesource.com/560020Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

BUG=angleproject:1815

Change-Id: Ia200cfe4ce4e6e3edd9ad2de841acbd1df49eb85
Reviewed-on: https://chromium-review.googlesource.com/561636
Commit-Queue: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>

Ensure the underlying GL does not modify areas of the texture that
correspond to areas outside the framebuffer, as required for WebGL.
Also zero out the texture in WebGL mode because CopyTexImage must return
zeroes for areas outside the framebuffer.

Enable corresponding test.

BUG=angleproject:1815

Change-Id: I51b1221dbf0dda0952e2ae89ee6ac925b5d1d4a4
Reviewed-on: https://chromium-review.googlesource.com/551535
Commit-Queue: Frank Henigman <fjhenigman@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>

The patch fixes the bug of having the OVR_multiview extension directive
being outputted by the ESSL translator whenever the
SH_INITIALIZE_BUILTINS_FOR_INSTANCED_MULTIVIEW option is enabled. The
directive should not be outputted because the extension is emulated
through that option.

BUG=angleproject:2062
TEST=angle_unittests

Change-Id: I95d0a651ace6db42d496de08e774ec7ceca4c197
Reviewed-on: https://chromium-review.googlesource.com/558981
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>

The EGL extension will allow for querying the cache and populating it
with shaders on startup. The GL extension adds some language to the
spec and allows for querying the program binary cache status of a
Context.

BUG=angleproject:1897

Change-Id: I7233844e41008f097efbb30cf9fe9c89227085fc
Reviewed-on: https://chromium-review.googlesource.com/556383
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>

Also move the ANGLE modifications to eglext.h to a separate file
so we can more easily update and maintain the official header.

This is in preparation for adding the new program cache control
extension.

BUG=angleproject:1897

Change-Id: I3546e0c82f6536381301f62a115b42d580747a08
Reviewed-on: https://chromium-review.googlesource.com/559917
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>

WebGL CopyTexImage needs to zero the part of the texture corresponding
to area outside the framebuffer, so we zero the whole texture then clip
the read area.
The clipping also avoids problems with code lower down that isn't prepared
for read areas not entirely within the framebuffer.

Enable corresponding test.

BUG=angleproject:1815

Change-Id: Ia7e0243ca72fa7c8f5bacda4d2022061d6a6d4f0
Reviewed-on: https://chromium-review.googlesource.com/551056Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Frank Henigman <fjhenigman@chromium.org>

This change replaces common string literals used for
error messages with const string variables mapped in
a new header file.

Additionally, more validation for WebGL naming scenarios
has been added, along with unit tests.

BUG=:angleproject:1644

Change-Id: Icff44a456aa78221c6df12b0454a7cc147a7d26e
Reviewed-on: https://chromium-review.googlesource.com/535974
Commit-Queue: Brandon1 Jones <brandon1.jones@intel.com>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>

In GL, ReadPixels() is allowed to modify memory that corresponds to
pixels outside the framebuffer.
In WebGL it must not do that, so clip the read area to the framebuffer.

Enable corresponding test.

BUG=angleproject:1815

Change-Id: I8113ae417dee7834e63498aec8291ce711bd7513
Reviewed-on: https://chromium-review.googlesource.com/536434
Commit-Queue: Frank Henigman <fjhenigman@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>

Structs with different names but identical members are treated as
ambiguous by the native HLSL compiler when looking up user-defined
functions. Add the struct name to the function name to work around
this limitation.

BUG=chromium:731324
TEST=angle_end2end_tests

Change-Id: Ie80ac0f1374bc5ac05dfebef3f94e2da7cdfc581
Reviewed-on: https://chromium-review.googlesource.com/558929
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>

Re-landing with a fallback for failed internal format queries to work
around issue seen on Shield TV. Also fixed wrong handling of integer
RG formats in isRequiredRenderbufferFormat.

Some NVIDIA GL drivers expose non-conformant multisampling modes. The
conformance of multisampling modes can be queried using the extension
NV_internalformat_sample_query. Use it to filter out the
non-conformant modes from the modes that are exposed by ANGLE.

The MAX_SAMPLES value and other similar values stored in caps also
need to be lowered to match the maximum number of samples exposed
for required formats.

There seems to be an NVIDIA driver bug related to querying
STENCIL_INDEX8 multisample format. Work around this by querying
DEPTH24_STENCIL8 instead.

There's also some confusion around whether RGB9_E5 should be
renderable. Once the floating point texture extensions got rolled
into the core GL spec, it was eventually made clear that RGB9_E5
is intended not to be renderable. The extension specs that predate
float textures in the core spec do suggest that it would be
renderable, but in practice drivers that advertise the extension
strings don't reliably implement RGB9_E5 as renderable. Solve this
by disabling it as a renderable format and adding an explanatory
comment.

BUG=chromium:682815
TEST=angle_end2end_tests,
     dEQP-GLES31.functional.state_query.internal_format.renderbuffer.*

Change-Id: I727f03045a1534d6764b571e6d839243705d25b3
Reviewed-on: https://chromium-review.googlesource.com/551957Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

We should give the app layer more control over the cache before we
enable it, to save on memory usage in Chrome.

BUG=angleproject:1897

Change-Id: I532c05c3042cb0a2d9c62f362f25d6064042ca2c
Reviewed-on: https://chromium-review.googlesource.com/558370
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>

The workaround which was previously defined to only apply on load
also seems to affect save on some AMD drivers.

BUG=angleproject:1637
BUG=angleproject:1897

Change-Id: Ia01a1420a484f3c2682ce97eaab18baccfb66a50
Reviewed-on: https://chromium-review.googlesource.com/558008Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>

In case gl_Position is statically used in the input shader, setting
the INIT_OUTPUT_VARIABLES flag will initialize gl_Position. Avoid
redundant initialization of gl_Position in this case.

Includes cleaning up memory management in InitOutputVariables_test:
all the pool-allocated variables will be freed at the end of each test
when the memory pool is cleared, so manual memory management is not
needed.

Also includes making the zero node check used in unit tests stricter
so that the tests are more reliable and moving it to
ShaderCompileTreeTest.h so that it can be reused in the future.

BUG=angleproject:2092
TEST=angle_unittests

Change-Id: I323a0a094afa6cea95c8a64e681d9fc485137423
Reviewed-on: https://chromium-review.googlesource.com/549418
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>

The validation would generate the correct errors to satisfy tests but
would skip clears.

TEST=conformance2/reading/format-r11f-g11f-b10f.html
TEST=conformance2/reading/read-pixels-from-fbo-test.html
TEST=conformance2/rendering/clearbuffer-sub-source.html
TEST=deqp/functional/gles3/fbocolorbuffer/clear.html
TEST=deqp/functional/gles3/fboinvalidate/sub.html
TEST=deqp/functional/gles3/framebufferblit/default_framebuffer_*.html

BUG=angleproject:1954

Change-Id: I0f220d06f98b630be5d27c7ffb1837d4ef29e0ac
Reviewed-on: https://chromium-review.googlesource.com/558786Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>

In GL, ReadPixels() is allowed to modify memory that corresponds to
pixels outside the framebuffer.
In WebGL it must not do that, so clip the read area to the framebuffer.

Enable corresponding test.

BUG=angleproject:1815

Change-Id: Ie99401a2102c352ffb1193a57aa66a5b96c184aa
Reviewed-on: https://chromium-review.googlesource.com/540556
Commit-Queue: Frank Henigman <fjhenigman@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>

With the SH_INIT_OUTPUT_VARIABLES option enabled, vertex and fragment
shader outputs get initialized with zeros at the beginning of main.
However, previous to this patch the lvalues in the binary expression did
not receive the correct type. This can lead to incorrect modifications
of the AST in subsequent stages or incorrect output code from the
translator. The patch addresses the issue by copying the type
information from the symbol table.

BUG=angleproject:2081
TEST=angle_unittests
TEST=angle_end2end_tests

Change-Id: I9e062376bcfad7d57b637a5248caebce1c9a0688
Reviewed-on: https://chromium-review.googlesource.com/544982
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>

GLSL ES spec accepts the case where an initializer of a variable calls
a function with the same name as the variable. The HLSL compiler
doesn't accept that. Work around this limitation in the HLSL compiler
by disambiguating user-defined functions from variables with a
different prefix.

BUG=angleproject:2095
TEST=angle_end2end_test, angle_unittests

Change-Id: I41b32a3fcc6fd4c548e8dc3aa680d1b07fcf8719
Reviewed-on: https://chromium-review.googlesource.com/557872Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>

This patch intends to fix a bug in ANGLE GLSL compiler.

In TCompiler::Init(resources), we should initialize maxUniformVectors by
resource.maxComputeUniformComponents / 4 when we attempt to initialize a
compiler for compute shader instead of resource.maxFragmentUniformVectors.

BUG=angleproject:2083

Change-Id: I4901f71ef5ac4f5770e2d5f8ee21786fcf19fbca
Reviewed-on: https://chromium-review.googlesource.com/545190
Commit-Queue: Jiawei Shao <jiawei.shao@intel.com>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>