When using a vertex buffer with DYNAMIC usage, with robust buffer
access enabled, we would sometimes read out-of-bounds when using very
large values for the index range. An unchecked signed addition would
overflow and lead to reading a negative offset.

Fix this problem by keeping the value size_t whenever possible. Also do
clamped casts when converting to a smaller values.

Also adds a regression test.

Also combined with 26b0bfb46: Fix warnings from size_t conversions.

Bug: chromium:842028
Change-Id: Ie1a8f476f3e97149362eb9855f08450c067ff807
Reviewed-on: https://chromium-review.googlesource.com/1064721Reviewed-by: Jamie Madill <jmadill@chromium.org>

BUG=836131

Change-Id: I1206c8eda465da563e15cf43f2e5c9320bb65eae
Reviewed-on: https://chromium-review.googlesource.com/1026460Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>
(cherry picked from commit f2807385)
Reviewed-on: https://chromium-review.googlesource.com/1030890Reviewed-by: Geoff Lang <geofflang@chromium.org>

This seems to have caused a regression in Chrome. I'm reverting the
change and suppressing the test that is failing on our end because of
it. Once we can confirm this solves the problem, we'll need some help
investigating further.

This reverts commit c0db9add.

Bug: angleproject:2321
Bug: chromium:833809
Change-Id: I5e40364217e15ae6117f5288a4754b25d983ca0a
Reviewed-on: https://chromium-review.googlesource.com/1017763
Commit-Queue: Luc Ferron <lucferron@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
(cherry picked from commit 1d97a4d0)
Reviewed-on: https://chromium-review.googlesource.com/1028490Reviewed-by: Geoff Lang <geofflang@chromium.org>

When using glCopyTexImage2D clearing of the mip level needs to happen
when running in WebGL or robust resource init mode and any pixel would
be sampled outside of the framebuffer. Previously the code was using
"setImage" for this purpose, causing issues when a PIXEL_UNPACK_BUFFER
was bound.

Also add a regression test.

BUG=chromium:827667

Change-Id: I03be20d8272730ab30afdab2f8919be853e729b6
Reviewed-on: https://chromium-review.googlesource.com/1000182
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
(cherry picked from commit 56c8577b)
Reviewed-on: https://chromium-review.googlesource.com/1014425Reviewed-by: Corentin Wallez <cwallez@chromium.org>

Stop tying the viewport and the scissor together. Instead, we simply
use a very large (0->maxInt) scissor when scissor isn't enabled
and we use the clipped scissor to the renderArea size when its
enabled.

Bug: angleproject:2443

Change-Id: If7454793a050b1833c7d3166ea6b380192085c8f
Reviewed-on: https://chromium-review.googlesource.com/1006996
Commit-Queue: Luc Ferron <lucferron@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>

This is moving all ANGLE Windows tryservers to run on LUCI only.

BUG=chromium:815092

Change-Id: I737b0883dca7385bdf1ec664affe53f606ca9401
Reviewed-on: https://chromium-review.googlesource.com/1008066Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>

In the case there are any dirty inactive attributes, we were
continually invalidating the Vertex Buffer and Input Layout
bindings. This fixes the invalidation to only happen when
there are any dirty active attributes.

This regressed in "Move Buffer Subject/Observer to front end."

Bug: chromium:829906
Bug: angleproject:2389
Change-Id: I8ed616bb696e0be548344192037ad6cc6f9c595c
Reviewed-on: https://chromium-review.googlesource.com/1006998Reviewed-by: Luc Ferron <lucferron@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

The Vulkan back-end will be using this helper struct, so make it
available everywhere. This cleans up a lot of the code for D3D.

Potentially in the future we could generate these in the entry
points if we supported packing multiple arguments into one.

Also changes a few parameter types to GLint for compatibility.

Also updates the Vulkan Texture implementation check the stored
vk::Format on setImage changes.

Bug: angleproject:2318
Change-Id: I57cea4a42483ab51859229517d783b58f206b8e7
Reviewed-on: https://chromium-review.googlesource.com/985203
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Reviewed-by: Luc Ferron <lucferron@chromium.org>

This more closely follows the general pattern laid out by the naming
in vk_helpers.h. It also changes the dynamic buffer that the helper
wraps to be stored by-value since the header include order problem is
fixed.

Bug: angleproject:2318
Change-Id: I1de9e1edee2125d3afd490b4f9c99cf70c61215c
Reviewed-on: https://chromium-review.googlesource.com/1001654Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Reviewed-by: Luc Ferron <lucferron@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

Bug: angleproject:2364

Change-Id: I316db1f6796a4a2efe344dd4b0a4a6e7bf2ead90
Reviewed-on: https://chromium-review.googlesource.com/1005556
Commit-Queue: Luc Ferron <lucferron@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>

syncPrimitiveTopology checks the transform feedback state, so it needs
to be called whenever the transform feedback state changes. This fixes
flakes in the WebGL 2 conformance test
transform_feedback/simultaneous_binding.html

Bug: 696345
Change-Id: I4e17bbf60b4a387cc23dc55bd5a051f5da9fa66e
Reviewed-on: https://chromium-review.googlesource.com/1006489
Commit-Queue: James Darpinian <jdarpinian@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>

This makes ANGLE to follow GLSL ES 3.00.6 spec section 12.46. The spec
requires that all attributes are taken into account when checking for
aliasing, regardless of if they are active or not. WebGL 2.0 spec was
also recently changed to reflect GLSL ES 3.00.6 correctly. Aliasing
checks for GLSL ES 1.00 shaders are left as-is.

BUG=chromium:829541
TEST=angle_end2end_tests, WebGL conformance tests

Change-Id: I71c36ac123f18dadf075e81f93af29321c15136b
Reviewed-on: https://chromium-review.googlesource.com/1005077Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>

BUG=angleproject:2306

Change-Id: I178b051c23da51d8eaf24c2a0df97f01ae5f3aaa
Reviewed-on: https://chromium-review.googlesource.com/1002914
Commit-Queue: Lingfeng Yang <lfy@google.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>

BUG=angleproject:2306

Change-Id: I46fe961e6eb87b2f41c39afe1e943d9146a0abfa
Reviewed-on: https://chromium-review.googlesource.com/1002913
Commit-Queue: Lingfeng Yang <lfy@google.com>
Reviewed-by: Geoff Lang <geofflang@chromium.org>

Bug: None
Change-Id: Iedf69564f465ad1a48fd5cebfe384ace6787a720
Reviewed-on: https://chromium-review.googlesource.com/1006050Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

Flaky.

BUG=angleproject:2408

Change-Id: I20a6cbf8c4daa722c9d3f91aa3b55227cfff94b5
Reviewed-on: https://chromium-review.googlesource.com/1005811Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Commit-Queue: Yuly Novikov <ynovikov@chromium.org>

- The scissor when clearing should not be clipped against the viewport.
- The render pass was created with a render area == to the current
viewport, but the viewport can be changed by the tests multiple times
per render pass, so we should always keep the renderArea to the full
framebuffer instead.
- Enables an additional 163 dEQP tests.
- We should clip the scissor to the framebuffer dimensions instead of
the viewport. Its valid to do a cmdClearAttachments outside the
viewport, but not outside the full framebuffer's dimension.

Bug: angleproject:2443

Change-Id: I79168e9f0c782d6dec77470fef938b85ad7a8794
Reviewed-on: https://chromium-review.googlesource.com/998448
Commit-Queue: Luc Ferron <lucferron@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>

Bug: angleproject:2447

Change-Id: Ib5a8176e2836374fbeb70ccd05586495c388680b
Reviewed-on: https://chromium-review.googlesource.com/1005438Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Commit-Queue: Luc Ferron <lucferron@chromium.org>

Bug: angleproject:2447

Change-Id: I92a6ed730dd2048b055d32d1f149a29172c7f38a
Reviewed-on: https://chromium-review.googlesource.com/999822Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Luc Ferron <lucferron@chromium.org>

- Also enables 174 additional dEQP tests.

Bug: angleproject:2446

Change-Id: I07cce46815d2f7a80fd48e7fd9407f7d528b397f
Reviewed-on: https://chromium-review.googlesource.com/998406Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Luc Ferron <lucferron@chromium.org>

This defers the actual Image initialization until the Image is used
as either a Framebuffer Attachment or OpenGL Texture object. This
will allow us to construct an Image from multiple sub resources, like
when we're initializing a mip chain, or a cube map texture.

Also adds a helper "hasDepthOrStencilBits" function to angle::Format.

Bug: angleproject:2318
Change-Id: Ife861560216581a90fc6da32a583f69886c7daea
Reviewed-on: https://chromium-review.googlesource.com/985202Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

Using FixedVector, we can remove the need for dynamic allocation in
most of the use cases. Only when we exceed a certain amount of fixed
storage do we need to spill into the heap.

Bug: angleproject:2389
Bug: chromium:829906
Change-Id: Ib5f7073b58876bfd2400bd87b743bddcee5e5dc8
Reviewed-on: https://chromium-review.googlesource.com/1002884Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Reviewed-by: Luc Ferron <lucferron@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

BUG=angleproject:2169

Change-Id: I192c2e3c453540c8a6d7b0d066218ea3c9fbaab2
Reviewed-on: https://chromium-review.googlesource.com/989411
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>

Create context and validation stubs for unimplemented extensions. This
will allow us to use an autogeneration script in the future to generate
entry_points_gles_2_0_ext.cpp.

Bug:angleproject:2263
Change-Id: If3ad3ebd823d65085c7f143cce9e09187e65d4da
Reviewed-on: https://chromium-review.googlesource.com/998440Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>

Will be useful for an optimization to the Buffer Subject/Observer
pattern.

Also cleans up an ASSERT.

Bug: angleproject:2389
Bug: chromium:829906
Change-Id: I2f8313ab531bca61947a51cc2396c04fb5d4bb1d
Reviewed-on: https://chromium-review.googlesource.com/1002883Reviewed-by: Luc Ferron <lucferron@chromium.org>
Reviewed-by: Lingfeng Yang <lfy@google.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

Invariant declarations didn't affect static use before, but now they
are also skipped in CollectVariables so an invariant declaration is
not enough in itself to mark a variable as active. This fixes an
assert in CollectVariables.

BUG=chromium:829553
TEST=angle_unittests

Change-Id: I3e51d2916f091bcc283af136a4abc846ff71447d
Reviewed-on: https://chromium-review.googlesource.com/999532Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>

Bug: 829773
Change-Id: I9bfe3c7b585acb7c91303f59ee448ce2d2dc2786
Reviewed-on: https://chromium-review.googlesource.com/999181Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Victor Costan <pwnall@chromium.org>

One field determines if a format is a compressed format
(called Block format for angle::Format, but basically the same).
The second field is the number of bytes in the format, so we
can do calculations on pixel size.

Both of these will make life cleaner in the Vulkan back-end.

Also solves some TODOs in the vk format table init.

Bug: angleproject:2318
Bug: angleproject:2358
Change-Id: I8b021b959c5892c86635e0225012295e5e830256
Reviewed-on: https://chromium-review.googlesource.com/987524
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Frank Henigman <fjhenigman@chromium.org>

This removes passing the Serial around to several methods, so that
dependency management is a bit more automatic.

This makes life a bit easier when dealing with state updates when
resources are in use by Vulkan.

The FramebuffeVk no longer stores an extra serial of the last draw,
instead it will trigger creation of a new writing node on a state
change update.

Bug: angleproject:2318
Change-Id: Ie58ec66e6e8644ba4d402c509255c3795d363dd3
Reviewed-on: https://chromium-review.googlesource.com/985201
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Frank Henigman <fjhenigman@chromium.org>

This renames ResourceVk to vk::CommandGraphResource, which should help
clarify its usage. This also moves LineLoopHandler, ImageHelper, and
the DynamicBuffer and DynamicCommandPool classes into a new vk_helpers
module. This file contains helper classes that manage other resources.

Also this makes DynamicBuffer and DynamicDescriptorPool no longer
inherit from CommandGraphResource. In the future, only Impl objects
will be allowed to be graph resources.

Bug: angleproject:2318
Change-Id: I0fa23da2ac853d90f3c822547a4a314f247cc757
Reviewed-on: https://chromium-review.googlesource.com/985200
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Frank Henigman <fjhenigman@chromium.org>

The old code was accidentally using letters offset by 10 when writing
out hex values >= 10. Now the letters a to f are used as they should.

This is one issue that changed shader output when ImmutableString was
introduced, so it is a potential cause for a regression detailed in
bug 824062, though this has not been verified.

BUG=chromium:824062
TEST=angle_unittests

Change-Id: Idb871dffba32a3ab20df0fe17b4b1a98ec00b7fa
Reviewed-on: https://chromium-review.googlesource.com/999480
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>

BUG=angleproject:2306

Change-Id: I96498aebbbc62ebd53e5320db17ef6a54d20d2dc
Reviewed-on: https://chromium-review.googlesource.com/998308
Commit-Queue: Lingfeng Yang <lfy@google.com>
Reviewed-by: Geoff Lang <geofflang@chromium.org>

Good for implementing stacks of fixed size (in the pop operation).

+ fix an issue where push_back() only allows adding up to max_size() - 1
elements.

BUG=angleproject:2306

Change-Id: I99b3c8416055f069fa563b684c102fd6aec116bc
Reviewed-on: https://chromium-review.googlesource.com/1000031Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Lingfeng Yang <lfy@google.com>

BUG=angleproject:2306

Change-Id: I83e15990c10d9354c2db00766ddc7b0ab960aa5c
Reviewed-on: https://chromium-review.googlesource.com/996019
Commit-Queue: Lingfeng Yang <lfy@google.com>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>

Refactors ANGLE extensions context calls into a common format for future
autogeneration. Any work that occurred in the entry point has been moved into
its own context member function.

Bug:angleproject:2263
Change-Id: I91cb3aca6a8c135ed3ae80a723d6cf16b54aeda6
Reviewed-on: https://chromium-review.googlesource.com/990809Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

This can now use the VertexArrayVk dirty bits. It also
seems as though there are a couple caching bugs with the
LineLoopHandler.

Bug: angleproject:2389
Change-Id: I8af73f4acf56768ed9c68395349ba96acfbe9666
Reviewed-on: https://chromium-review.googlesource.com/989259
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Frank Henigman <fjhenigman@chromium.org>
Reviewed-by: Luc Ferron <lucferron@google.com>

Instead of the ContextVk owning the translations for the various
attributes, make the VertexArrayVk own them. This way they can
handle the dirty bit state notifications directly instead of
needing their own Observers.

Bug: angleproject:2389
Change-Id: I5e571ba6c563e820a4c0d5f92db35031e6f2428a
Reviewed-on: https://chromium-review.googlesource.com/989258
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Yuly Novikov <ynovikov@chromium.org>

This works around a bug in Vulkan where the VertexArrayVk still
had resources in use when the Context was being destroyed. An
alternative approach is to call finish() from Context::onDestroy,
but this seems heavyweight. We might have to adopt something like
this eventually though.

Bug: angleproject:2389
Change-Id: Id99ca00404972a0b857c0e7e36f37b8ac4fd5c5e
Reviewed-on: https://chromium-review.googlesource.com/997743Reviewed-by: Frank Henigman <fjhenigman@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>

Compiling fails with some combinations of gcc and glibc.
'size_t' does not name a type in proc_table.h

Change-Id: Id79e033dc72d836788551e076ced6dbd706ce902
Reviewed-on: https://chromium-review.googlesource.com/995422Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Commit-Queue: Yuly Novikov <ynovikov@chromium.org>

In the situation when we try to clear the depth OR the stencil
buffer but we have a packed depth/stencil attachment, we were still
clearing both of them since Vulkan doesn't allow setting just the depth
or the stencil image aspect bit on the clear call. The workaround is to
use the vkCmdClearAttachments command that allows us to clear one or
the other easily.

Bug: angleproject:2443

Change-Id: I6814e1927c019d9ec9255d79c6bc7b913a8c99e9
Reviewed-on: https://chromium-review.googlesource.com/997752
Commit-Queue: Luc Ferron <lucferron@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>