Currently this won't speed up performance much, if at all, since we
don't even really support state changes. It sets the stage for using
a pipeline cache later, with better state change support. It also
makes implementing descriptor sets for Textures a bit simpler, since
we can just update descriptor sets when the dirty bits tell us of a
Texture change.

Add cache structures to VertexArrayVk and ContextVk so we only need
to update the structures before we create a new pipeline. When we
support pipeline caching, we will most likely be updating a compact
representation for fast cache query.

BUG=angleproject:1898
BUG=angleproject:2167

Change-Id: Id545f2c67c06d8b6e8b7eb63ca70464f6b9a51f6
Reviewed-on: https://chromium-review.googlesource.com/713586
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Frank Henigman <fjhenigman@chromium.org>

This adds support for macOS 10.13 "High Sierra"

BUG=chromium:774309

Change-Id: I9b5b13750608f703dd0a3e0afeb1e314b437b8cc
Reviewed-on: https://chromium-review.googlesource.com/721520
Commit-Queue: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>

The std::pair constructor isn't constexpr there.

BUG=

Change-Id: I7711185c6c991fd5793cc6c785e9905710660cd9
Reviewed-on: https://chromium-review.googlesource.com/723849Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>

Add the code that finds the right memory information into a shared
location. This will allow it to be reused with other points in the
code.

BUG=angleproject:2167

Change-Id: I72d0522e134666d0a00921cb5ef6bc093473afdf
Reviewed-on: https://chromium-review.googlesource.com/717189
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>

This was generating a warning for missing virtual destructor in
some configurations.

BUG=angleproject:2167

Change-Id: I05e335de4ab1160c6066665d588b50527ea34cbd
Reviewed-on: https://chromium-review.googlesource.com/723980Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

This method is useful for the Vulkan back-end as well as D3D11. It can
produce a uniform block layout for the default uniform blocks as well
as for interface blocks. Put it in blocklayout.h in the translator.

BUG=angleproject:2167

Change-Id: I13160906921da439746c1811a623006250aaeefd
Reviewed-on: https://chromium-review.googlesource.com/713941Reviewed-by: Frank Henigman <fjhenigman@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

DEPS libpng into third_party/libpng/src, instead if third_party/libpng.
Then we'll be able to put its BUILD.gn into third_party/libpng instead
of the more distant build/secondary/... location.

BUG=angleproject:1569

Change-Id: I12b1c51d65682cba3954c26f1bd3f7491ac31b01
Reviewed-on: https://chromium-review.googlesource.com/722176
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>

These can be used instead of direct access of the arraySize field in
Chromium, so that arrays of arrays support can be implemented inside
ANGLE without breaking the Chromium build.

BUG=angleproject:2125
TEST=angle_unittests

Change-Id: I123c2a18549e8f6e40ade946870a5437faf2155e
Reviewed-on: https://chromium-review.googlesource.com/718201Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>

BUG=angleproject:1569

Change-Id: I2bcde845d70159c1047349673ec8417b7686eafc
Reviewed-on: https://chromium-review.googlesource.com/721257Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Frank Henigman <fjhenigman@chromium.org>

For testing this also converts two unimportant GLenums, gl::BufferUsage
and gl::CullModeFace.

BUG=angleproject:2169

Change-Id: If1e86a97d0fed3fd567303aca6506ec579503076
Reviewed-on: https://chromium-review.googlesource.com/688000
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>

These extra include dirs were not being rebased correctly in the
BUILD.gn side. Instead add them later in angle_deqp_support. The
GN build already had the correct deps in any case.

BUG=angleproject:1569

Change-Id: Ieb4fab47bd9948a3309da239f2efdafbfcc8782c
Reviewed-on: https://chromium-review.googlesource.com/722065Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

The code is refactored so that symbol nodes can only be initialized
with an unique id object. This prevents accidentally forgetting to
create an id for a symbol node.

This opens up possibilities for future optimization: For example the
names and types of symbols could be stored in a central location
inside the SymbolTable, and TIntermSymbol nodes would only need to
store the symbol id. The symbol id could be used to look up the name
and type of the node.

BUG=angleproject:1490
TEST=angle_unittests

Change-Id: Ib8c8675d31493037a5a28c7b36bb9d1113cc10f6
Reviewed-on: https://chromium-review.googlesource.com/580955Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>

Also SubDataOffsetPreservesContents.

BUG=angleproject:2185

Change-Id: I73401e4b511313f14ee9659f3d20c8086e3d04a3
Reviewed-on: https://chromium-review.googlesource.com/719890
Commit-Queue: Yuly Novikov <ynovikov@chromium.org>
Reviewed-by: Yuly Novikov <ynovikov@chromium.org>

This seems to be a new thing with VS2017.

BUG=angleproject:1569

Change-Id: Ic46566b979e975e5e6eb3a1f80eac9e44b66f2d4
Reviewed-on: https://chromium-review.googlesource.com/719296Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

There are many types of statements after which a semicolon is not
needed. Skip writing the semicolon in HLSL output in these cases to
make the output code more readable.

BUG=angleproject:1013
TEST=angle_end2end_tests

Change-Id: I8f6a5e4ecfe5def456fdf19cca5ca451c13d7f35
Reviewed-on: https://chromium-review.googlesource.com/718420
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>

The multiview transform feedback test was not compliant with the GLES
spec for transform feedback. The issue had to do with this part of
section 2.15.2 of GLES 3.0.5 spec:

"The error INVALID_OPERATION is also generated by
BeginTransformFeedback if no binding points would be used, either
because no program object is active or because the active program
object has specified no output variables to record."

Fix this.

BUG=angleproject:2184
TEST=angle_end2end_tests on NVIDIA 387.92 drivers

Change-Id: I24816d2c24df0072179f21ead892bd2c9ba696d2
Reviewed-on: https://chromium-review.googlesource.com/718702
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>

The incomplete texture handling is similar between the D3D and Vulkan
back-ends. We create 1x1 textures, initialize them to black, and bind
them when we detect incomplete textures. We would also bind incomplete
textures when we detect feedback loops. In the GL back-end, we
wouldn't detect feedback loops, and would allow the driver to handle
incompleteness.

Instead implement this in a shared helper class, and do the feedback
loop detection in the front-end for every back-end. This makes our
behaviour more consistent between back-ends, and prevents undefined
behaviour.

Because initializing multisample textures is tricky (they
can't be updated with TexImage calls) we do a bit of a workaround so
the back-end can clear the incomplete multisample texture initially.

This progresses the initial Vulkan textures implementation.

BUG=angleproject:2167

Change-Id: I79ddcc0711fcc986f2578a52ac6f701231d241ac
Reviewed-on: https://chromium-review.googlesource.com/700993Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

Actually failing on Win Intel D3D11,
but not possible to test for Intel D3D11 since GL_RENDERER is NULL.

BUG=angleproject:2183
TBR=jmadill@chromium.org

Change-Id: I894b2772af1154a86354ca755a06f146b33a4ed8
Reviewed-on: https://chromium-review.googlesource.com/716801
Commit-Queue: Yuly Novikov <ynovikov@chromium.org>
Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>

Also adds a test for an integer overflow in DrawArrays validation.

BUG=602737

Change-Id: I71c5bd155cf919964f46e2b0b48267253cab91eb
Reviewed-on: https://chromium-review.googlesource.com/716841Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>

We can use Python's "endswith" more easily than using the current
checks.

BUG=angleproject:1390
BUG=angleproject:2167
    

Change-Id: I5a35602c91b989270cff1d42a9d5ede542039fbd
Reviewed-on: https://chromium-review.googlesource.com/717188Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

If variables are declared inside a GLSL switch statement, they are
scoped until the end of the switch statement. This is not compatible
with HLSL rules, where the scoping is until the end of the case. To
work around this, wrap switch statements in a block that declares
the variables in HLSL.

This is done after most other transformations done to the AST are
complete, since some of the other transformations may introduce
temporary variables.

BUG=angleproject:2179
TEST=angle_end2end_tests

Change-Id: Id0bb89affe103177fd3d6a6b2f3619b5e1ada0a6
Reviewed-on: https://chromium-review.googlesource.com/716381Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>

BUG=602688

Change-Id: I49a1a9b3753159c91058a7754534383fbc8a778e
Reviewed-on: https://chromium-review.googlesource.com/713740Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>

RemoveSwitchFallThrough now treats cases where break; or return; is
nested inside a block as non-fallthrough to avoid unnecessary
duplication of code. For example, the case 1 below would previously
get treated as fall-through:

switch(foo)
{
    case 1:
    {
        break;
    }
    default:
        break;
}

Now RemoveSwitchFallThrough doesn't do anything to this code.

BUG=chromium:772695
TEST=angle_end2end_tests

Change-Id: Iafab6d8b05c63bcdb5f54834dbc1f41192c31dd4
Reviewed-on: https://chromium-review.googlesource.com/709197Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>

Some users of ANGLE might want to use the D3D11 backend on older
Windows 7 platforms that only support DXGI 1.1. DXGI 1.1 doesn't
recognize the FL 11.1 enum passed to D3D11CreateDevice, so special
fall-back code is needed to fall back to FL 11.0 or lower in this
case.

BUG=angleproject:2173

Change-Id: Id77401d0e2a16786ed44d117ca746908e8d0892c
Reviewed-on: https://chromium-review.googlesource.com/716356Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>

Fix resulting warnings.

Change-Id: Ie251012ee01aa484a8c489f9deeec385a50528f9
Reviewed-on: https://chromium-review.googlesource.com/713695
Commit-Queue: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>

Make NodeInsertMultipleEntry private and clarify some comments in
IntermTraverse.h.

BUG=angleproject:2100
TEST=angle_unittests, angle_end2end_tests

Change-Id: Iae60a46714c8b5cb9ad1e9d70aa6776f9deaf3d5
Reviewed-on: https://chromium-review.googlesource.com/715718Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>

The grammar needs to generate AST nodes even for no-op statements,
since they might be the last statement in a switch statement that is
required for switch statement validity. Change the grammar to generate
nodes from empty blocks and empty declarations.

We also need to do some further processing of the AST. This is because
PruneEmptyDeclarations will still remove empty declarations, and at
least the NVIDIA driver GLSL compiler doesn't accept some types of
no-op statements as the last statement inside a switch statement. So
after parsing has finished we do rudimentary dead code elimination to
remove dead cases from the end of switch statements.

BUG=angleproject:2181
TEST=angle_unittests

Change-Id: I586f2e4a3ac2171e65f1f0ccb7a7de220e3cc225
Reviewed-on: https://chromium-review.googlesource.com/712574
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>

The way that the strings were generated was erroneously marking 2D
multisample textures as "2D" instead of "2D_MULTISAMPLE". Fix this
by changing how the types are parsed and generated.

Fixes incomplete texture handling of 2D multisample textures.

BUG=angleproject:1390
BUG=angleproject:2167

Change-Id: I8ffffca119cc7ca654ed1fe81cba37b758f9ef9f
Reviewed-on: https://chromium-review.googlesource.com/713585Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Frank Henigman <fjhenigman@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

Update to the version of googletest current in chromium.  Googlemock has
merged into googletest so we get rid of the separate DEPs entry and point
everything inside googletest.

We put it in 'third_party' instead of 'testing' to prepare for the
upcoming stand-alone gn build which borrows chromium build files that
require other stuff in the 'testing' directory.

BUG=angleproject:1569

Change-Id: I51d1675efd5d0104bb6390721928010221a5d6fc
Reviewed-on: https://chromium-review.googlesource.com/713466Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Frank Henigman <fjhenigman@chromium.org>

In case the last case inside a switch statement is not terminated in
a branch statement, RemoveSwitchFallThrough needs to add it before
calling handlePreviousCase. This ensures that all preceding
fall-through cases will get a copy of the branch statement and so will
not fall through.

This also fixes running RemoveSwitchFallThrough so that it's only
executed once per each switch statement.

The error was not caught by the dEQP tests, so a new ANGLE test is
added.

BUG=angleproject:2178
TEST=angle_end2end_tests

Change-Id: I26b6989aa4d32de2d74cde56d72ee24f61195445
Reviewed-on: https://chromium-review.googlesource.com/709196Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>

Fix the resulting compile errors.

No-Try: true
Change-Id: I88e74f240b29ecf7b3883b0c5880dcb871e4c3cd
Reviewed-on: https://chromium-review.googlesource.com/710415
Commit-Queue: Ben Wagner <benjaminwagner@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>

Strcmp in SampleApplication.cpp is a cstring function. So some
compilers (at least the default gcc compiler in my Ubuntu 17.04)
would think that strcmp is not declared and lead to build failure,
even we have include <string> in header file.

Now that we use std::string variable in SampleApplication.h, it is
better to include <string> in that header file as what it is.
Then add <string.h> to include cstring functions in .cpp file to
fix the build issue.

BUG=angleproject:2180

Change-Id: Id97a6956480637c1218314c04d9b331031c7b911
Reviewed-on: https://chromium-review.googlesource.com/711396Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>

RemoveSwitchFallThrough now correctly records the existence of
declaration and swizzle statements inside switch statements.

BUG=angleproject:2177
TEST=angle_end2end_tests

Change-Id: I1ef83997db7ae510ded002a9568c29272c00c2fe
Reviewed-on: https://chromium-review.googlesource.com/709195Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>

This should slightly reduce draw call overhead.

BUG=angleproject:1898

Change-Id: I0e515bf2868f237f1d6948c12942f8cb6637c0c0
Reviewed-on: https://chromium-review.googlesource.com/707690Reviewed-by: Frank Henigman <fjhenigman@chromium.org>
Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

Refactoring change only.

BUG=angleproject:1898

Change-Id: I9f55651f923ff930c395a9bb575b4f86ad5d9cbd
Reviewed-on: https://chromium-review.googlesource.com/707689Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

This cleans up the API provided by RemoveSwitchFallThrough.h, and adds
documentation about caveats of RemoveSwitchFallThrough. This change is
pure refactoring without any functional changes.

BUG=angleproject:2177
TEST=angle_end2end_tests

Change-Id: I2646e4fe3b53130b07977823cb1344e5096f67e4
Reviewed-on: https://chromium-review.googlesource.com/709194Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>

Array indices are sorted so that the outermost index is in the back.
This is because we want to be consistent with future arrays of arrays
parsing code. In parsing we'll have a utility function to make a
TType object into an array, and there it's most natural to push the
new outermost sizes to the back of the vector.

Further patches will still be needed to parse arrays of arrays and
add support to arrays of arrays into the API.

BUG=angleproject:2125
TEST=angle_unittests, angle_end2end_tests

Change-Id: I6c88edabf68ae9dbd803ec6d20543016c408b702
Reviewed-on: https://chromium-review.googlesource.com/686414Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>

This patch will remove unused functions syncVertexAttributes and
hasDirtyOrDynamicAttrib, and private member mAppliedIBChanged.
Meanwhile, it changes hasDynamicAttrib to hasActiveDynamicAttrib
to avoid unnecessary update in some cases.

BUG=angleproject:1155

Change-Id: I29a0aa7fbc13874b7b070cf1ac5ec4134728f519
Reviewed-on: https://chromium-review.googlesource.com/707014Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>

This gives us the ability to test Vulkan without a special recompile,
and keeps the default setting to the default ANGLE renderer. Only
implemented for the hello_triangle sample currently.

Also includes an extra error message for missing EGL config, which is
where the Vulkan init fails right now, due to missing depth/stencil
support.

Also includes a hack to get Vulkan samples running without depth/
stencil.

BUG=angleproject:2167

Change-Id: I6925b9b84956fb69cbf602a828ea95d8c7125b68
Reviewed-on: https://chromium-review.googlesource.com/707688Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

1) timeout is unsigned, and clients tend to pass large unsigned values
   (like -1) to mean "wait as long as possible". Casting to signed at
   the start of the math meant that everything was negative, so we would
   never actually wait for fences to be signalled.

2) When going from nanoseconds to seconds, we should divide by 10^9, not
   multiply by 10^6.

3) Even with all of this, it's possible for a sufficiently large counter
   frequency to still cause overflow, so detect that case and clamp.

Change-Id: I9e728aac72d8dc0b15582732b6fd4d87c90bd140
Reviewed-on: https://chromium-review.googlesource.com/707202Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>