BUG=angleproject:1356

Change-Id: Id563997d2921cf558c52a781ae66d8bde58d1f2f
Reviewed-on: https://chromium-review.googlesource.com/339847Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>

This eliminates another global std::map, and replaces it with a json
generated switch. This should be better for threading, better on
memory use, and faster.

BUG=angleproject:1389
BUG=angleproject:1459

Change-Id: I1d289637c00783690ec8ea743ea2aa17b0ab8e50
Reviewed-on: https://chromium-review.googlesource.com/394234Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

Subsequent refactoring of this code will be easier to review if there
won't be unrelated style changes that git cl format insists on doing.

BUG=angleproject:1490
TEST=angle_unittests

Change-Id: I102fd73bd92317ab438e1676422212f644d2859b
Reviewed-on: https://chromium-review.googlesource.com/394649Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>

This change is two-fold: first it allows us to remove the std::map for
D3D11 format info, which was only being used in a single place.

Secondly, it fixed the support table to use the correct 9_3 caps for
texture support, determined by running the D3D11FormatTablesTest.
This should reduce the number of format queries we need to run on 9_3
startup.

BUG=angleproject:1389
BUG=angleproject:1459

Change-Id: I2435a47fc221e73c96337bc7c1c4ed37b527bc1f
Reviewed-on: https://chromium-review.googlesource.com/392210Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

This query is currently a part of a std::map. Split it off to its
own switch using auto-generation based on the format string.

Also introduce a DXGI-to-angle format map. This map is not totally
complete because some more esoteric formats don't have corresponding
ANGLE formats yet. We should add these other formats (EG NV12) if we
need them.

BUG=angleproject:1389
BUG=angleproject:1459

Change-Id: I49e045bdc04a52166299c9e22a4c4d3f24069dbc
Reviewed-on: https://chromium-review.googlesource.com/392209Reviewed-by: Geoff Lang <geofflang@chromium.org>

There were two bugs here that were masking each other:

1. Case nodes were not being added to AST traversal path.
2. RecordConstantPrecision did not check if a constant was a child of
   a case node.

Fix these.

BUG=angleproject:1490
TEST=WebGL 2 conformance tests

Change-Id: Ic6cb7942bf57870fa26eb6a011919a2d7951cc85
Reviewed-on: https://chromium-review.googlesource.com/394588Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>

This change adds implementation of eglGetSyncValuesCHROMIUM extension
on D3D11 with Direct Composition. This should work on Windows 8.1 and
above.

The implementation is based on IDXGISwapChain::GetFrameStatistics.

Extension documentation:
https://chromium.googlesource.com/chromium/src/gpu/+/master/GLES2/extensions/CHROMIUM/EGL_CHROMIUM_get_sync_values.txt

BUG=angleproject:1402

Change-Id: I306434dd8d85d618b14edfa38fc2a22e50fddacc
Reviewed-on: https://chromium-review.googlesource.com/390351
Commit-Queue: Stanislav Chiknavaryan <stanisc@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>

The ASSERT macro is enabled with ANGLE_RELEASE_ASSERT, so we should
use the same behaviour as the normal ASSERT for the other debug
macros.

Also remove UNREACHABLEs in the format validation code that were
being hit silently.

Also make some EGL tests skip instead of fail, since they were hitting
UNIMPLEMENTED.

BUG=angleproject:1332
BUG=angleproject:1540
BUG=angleproject:1340

Change-Id: Ie3347f2bf2459a5831b66827871998d8fbe7d4b2
Reviewed-on: https://chromium-review.googlesource.com/392490Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

The fuzzer uncovered a case where we were trying to emulate precision
on HLSL 3.0, causing an ASSERT crash.

BUG=chromium:653276

Change-Id: I2e666a1ff4f605541e25f04264146063559cb835
Reviewed-on: https://chromium-review.googlesource.com/394237Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Olli Etuaho <oetuaho@nvidia.com>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

In some code path found by a fuzzer, we would have a zero sized
array, which would trigger a zero divide. I was unable to extract
the failing shader from the fuzzer case trivially.

BUG=chromium:653274

Change-Id: Ia2558ae828fa73615c901fd1cda9ddaa3a72f1a9
Reviewed-on: https://chromium-review.googlesource.com/394238Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

Desktop OpenGL before 4.4 doesn't handle SRGB blits the same way OpenGL
ES does. Emulate them by drawing a quad.

BUG=angleproject:1492
BUG=chromium:634525

Change-Id: I9f2992d9b373941b10f19f8a51564f0f756cc4df
Reviewed-on: https://chromium-review.googlesource.com/389853Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>

This allows us to delete some duplicated code in the D3D11-side.

BUG=angleproject:1389
BUG=angleproject:1459

Change-Id: Ifdcfcd4a56e06ff2ae8f5ca0bda72281d52c2964
Reviewed-on: https://chromium-review.googlesource.com/392208
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>

This test appears to be flaky.

BUG=angleproject:1541

Change-Id: I2a91626f6dc309d2dc23098cbaf290adac7f4859
Reviewed-on: https://chromium-review.googlesource.com/394217Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

This fails because of issues with Intel UBOs and copying.

BUG=chromium:651493
BUG=chromium:593024

Change-Id: Ia4c634ebecd29082927a9a812929c102d0d629b2
Reviewed-on: https://chromium-review.googlesource.com/394131Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

The correct test was the multi-context version.

BUG=angleproject:1541

This reverts commit d0614b85.

Change-Id: I4f4a9e3b1e0301eef34ce8c06cd649bff66fcdb1
Reviewed-on: https://chromium-review.googlesource.com/394216Reviewed-by: Jamie Madill <jmadill@chromium.org>

This test appears to be flaky.

BUG=angleproject:1541

Change-Id: I947dbc759e0fef505ce9b11e501f2dbe35ff1e5c
Reviewed-on: https://chromium-review.googlesource.com/394329Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

This map lives in formatutils.cpp, and maps from a {format,type} to
an internal format. Convert this into generated code.

BUG=angleproject:1389

Change-Id: Ib12750d83d3f8cf8794d6668874cb025f856b5d4
Reviewed-on: https://chromium-review.googlesource.com/392207
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>

BUG=angleproject:1468

Change-Id: Ief4b0e7d7a0420e59e547cd24fde53deb3718ccb
Reviewed-on: https://chromium-review.googlesource.com/388845
Commit-Queue: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>

BUG=angleproject:1468

Change-Id: I2ed2b65a4b6f6ea3dda61c0467aef95e407cd38c
Reviewed-on: https://chromium-review.googlesource.com/388844Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>

This mainly affects RewriteElseBlocks, which was the only piece of
code still adding TIntermIfElse nodes directly as children of other
TIntermIfElse nodes.

BUG=angleproject:1490
TEST=angle_unittests

Change-Id: I5b25c2fb9c642424417cd6c29e37c20482c6ffaf
Reviewed-on: https://chromium-review.googlesource.com/392847Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>

BUG=angleproject:1332

Change-Id: Ie970466c4528cb80b2363d0c0683da2d9c23c4f4
Reviewed-on: https://chromium-review.googlesource.com/392267
Commit-Queue: John Bauman <jbauman@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>

Evaluating integer expressions in the ESSL preprocessor may result in
overflowing the signed integer range. Implement wrapping overflow for
preprocessor expressions in a way that doesn't hit any undefined
behavior.  In the ESSL spec, preprocessor expressions are defined to
have mostly the same semantics as in C++. Since C++ doesn't define
what happens on signed integer overflow, we choose to make most of the
operators wrap on overflow for backward compatibility and consistency
with the rest of the ESSL spec.

We reuse the existing wrapping overflow helpers that are
used for constant folding. To be able to do this, the type used in the
preprocessor expression parser is changed from 64-bit to 32-bit.

Shifting negative numbers is implemented as a logical shift. This
cannot be disallowed since dEQP requires shaders shifting negative
numbers to pass compilation.

Undefined bitwise shifts where the offset is greater than 31 will now
result in a compile-time error.

A couple of test cases are now covered by the preprocessor tests
rather than full compilation tests. This isolates the tests better and
they run faster.

BUG=chromium:652223
TEST=angle_unittests

Change-Id: I84be40d404c10ecd0846c5d477e626a94a2a8587
Reviewed-on: https://chromium-review.googlesource.com/392146
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>

Our for-loop logic was confused so that some unreferenced
transform feedback varyings might get dropped.

BUG=chromium:651493

Change-Id: Id283230da0a47fc647b2a3862da60be5538e439e
Reviewed-on: https://chromium-review.googlesource.com/391945
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>

This should fix a null D3D11 backend draw call performance regression.

BUG=651101

Change-Id: I2eb10cddd15f0e7b25b886c89eccd2906e988c72
Reviewed-on: https://chromium-review.googlesource.com/392227Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>

On some drivers, the WGL_ARB_create_context_robustness extension is present but
the created context does not expose the extensions.  Be more defensive and only
expose robustness if we're able to create contexts that have it.

BUG=angleproject:1538

Change-Id: I3d02b2071c7b2c576abe66cc283cee467e400fd3
Reviewed-on: https://chromium-review.googlesource.com/391833Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: John Bauman <jbauman@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>

The change to initialize a constant buffer immediately after
calling BufferData in D3D11 led to a bug where we would map
the UBO for writing with the wrong map bits. Fix this by
using the same map method as the rest of the code.

The D3D11 runtime seems to allow arbitrarily large constant buffers
on Windows 10, but not Windows 7. Thus this CL also fixes a bug in
our constant buffer size clamping to not copy more than the available
buffer size for uniform buffers.

BUG=chromium:651493

Change-Id: I876767691d02db90ecb08a8fa78199f03339a35e
Reviewed-on: https://chromium-review.googlesource.com/391167Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

Some traversers that insert code to main() assume that the main()
function has a non-null body node in place. This assumption was
previously wrong, since functions could be missing the body node
in case the function body was empty.

Fix possible invalid dereferencing of missing function body nodes by
always adding an empty sequence node to represent the body of
functions that have an empty body in the ESSL source. This also
enables simplifying some tree traversers that used to take the
possibility of missing function body nodes into account.

Also fix AddDefaultReturnStatements to check the last statement inside
the function body for a return statement, instead of checking the
first statement.

BUG=angleproject:1539
TEST=angle_unittests, angle_end2end_tests

Change-Id: I2fbd18c78653fa2f1a96dbd9a619accc4874030d
Reviewed-on: https://chromium-review.googlesource.com/392046Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>

This reverts commit 9aa83fe3.

The new test UniformBufferTest.ActiveUniformNumberAndName/ES3_OPENGL is failing on multiple platforms. Examples:
https://build.chromium.org/p/chromium.gpu.fyi/builders/Mac%2010.10%20Release%20%28ATI%29/builds/12285
https://build.chromium.org/p/chromium.gpu.fyi/builders/Linux%20Release%20(ATI)

Change-Id: I78b1a4d58e9a291e40ad304eb32f990e0518f7ee
Reviewed-on: https://chromium-review.googlesource.com/391049Reviewed-by: Kenneth Russell <kbr@chromium.org>

This reverts commit ca40662a.

https://chromium.googlesource.com/angle/angle/+/ca40662aa81f7735d3510fa52e55d28503c7f9ec broke angle_end2end_tests on the Win Debug bots. Examples: 

https://build.chromium.org/p/chromium.gpu.fyi/builders/Win7%20Debug%20%28NVIDIA%29/builds/18718 
https://build.chromium.org/p/chromium.gpu.fyi/builders/Win8%20Debug%20%28NVIDIA%29/builds/15905 

Tests failing: 
Texture2DTest.QueryBinding/ES2_D3D11 
Texture2DTest.QueryBinding/ES2_D3D9 
Texture2DTest.QueryBinding/ES2_D3D11_9_3

Change-Id: If9da864f49e0de3a134d43ae36c2a99e9b5a2a53
Reviewed-on: https://chromium-review.googlesource.com/391130Reviewed-by: John Bauman <jbauman@chromium.org>

BUG=angleproject:1510

Change-Id: I9ed7c877767909ba93e2ea5ef09c4735b0d0bcc7
Reviewed-on: https://chromium-review.googlesource.com/382652
Commit-Queue: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>

BUG=angleproject:1332

Change-Id: I591b9520bc1122eb6232803ac86b6d0322f8c349
Reviewed-on: https://chromium-review.googlesource.com/390323
Commit-Queue: John Bauman <jbauman@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>

This allows specifying data size to all GL functions that provide a pointer to
client memory and a length parameter for all functions in which the driver
writes to client memory.

BUG=angleproject:1354

Change-Id: Ia68be1576b957cb529c87b5e0d1bd638c7dbd371
Reviewed-on: https://chromium-review.googlesource.com/382012
Commit-Queue: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>

BUG=angleproject:1510

Change-Id: I42355607235a05fe276342084acd1eb359c16de4
Reviewed-on: https://chromium-review.googlesource.com/390414Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>

This avoids creating a weird aggregate node with a sequence of
constant union nodes to store the offsets. They're stored neatly
inside a vector instead. This makes code that needs to iterate
over the swizzle offsets much simpler.

BUG=angleproject:1490
TEST=angle_unittests

Change-Id: I156b95723529ee05a94d30295ffb6d0952a98564
Reviewed-on: https://chromium-review.googlesource.com/390832Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>

After experiencing a device loss, Renderer11::resetDevice can fail to
recreate mDevice, leaving it null. In a following call to
testDeviceLost, mDevice will be dereferenced, causing a segfault.

BUG=angleproject:1457

Change-Id: Ib5be64542929441ceb7088a328c3245ae7e52f17
Reviewed-on: https://chromium-review.googlesource.com/364120Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>

On some Mac drivers with shader version 4.1, they will
treat unused std140 and shared uniform blocks' members as inactive. However,
WebGL2.0 based on OpenGL ES3.0.4 requires all members of a named uniform block
declared with a shared or std140 layout qualifier to be considered active.
The uniform block itself is also considered active.

This workaround is to reference all members of unused std140 and shared uniform blocks
at the beginning of the vertex/fragment shader's main().

BUG=chromium:618464
TEST=UniformBufferTest.ActiveUniformBlockNumber

Change-Id: I1d2c5e3e8da04786ac6a37fd26f7bb9c14cd76ed
Reviewed-on: https://chromium-review.googlesource.com/387169
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>

BUG=angleproject:1512

Change-Id: I4c60472d216bfc5198e635d70fd197a5738dde98
Reviewed-on: https://chromium-review.googlesource.com/390133
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>

On some Intel drivers, calling function isnan() on highp float will get wrong
answer. This patch work arounds this bug by using an expression to emulate
this function.

BUG=chromium:650547

Change-Id: I5bc5e0352c434f42cd2c55103a74f9f7ba51a72c
Reviewed-on: https://chromium-review.googlesource.com/389834
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>

Negating the minimum representable integer overflows, so it
has undefined behavior in C++. Handle this as a special case in the
code.

BUG=chromium:637050
TEST=angle_unittests

Change-Id: Ic6e6d638faddad9b70b5d1637bb4b42ef4f43784
Reviewed-on: https://chromium-review.googlesource.com/390551Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>

Integer operations that overflow are defined to wrap in the ESSL
3.00.6 spec. Constant folding that happens inside the shader
translator should also follow the wrapping rules.

The new implementations of wrapping integer addition and subtraction
use unsigned integers to perform calculations. Unsigned integers are
defined to implement arithmetic in modulo 2^n in the C++ spec. This
behavior is also leveraged to implement wrapping unsigned integer
multiplication.

The implementation of wrapping signed integer multiplication is
slightly trickier. The operands are casted to a wider type to perform
the multiplication in a way that doesn't overflow, and then the result
is truncated and casted back to the narrower integer type.

Incorrect tests that expected errors to be generated from integer
overflow in constant folding are removed.

BUG=chromium:637050
TEST=angle_unittests

Change-Id: I0de7e25881d254803455fbf22907c192f49d09ff
Reviewed-on: https://chromium-review.googlesource.com/390252
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>