Tags: google/dawn

Suppress Xclipse failures on CI

This should be enough to get the bot green.

Bug: 543960651
Change-Id: Id304b3dac54ee00aa3857fa370f459245141fbb6
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/330855
Auto-Submit: James Price <jrprice@google.com>
Reviewed-by: Brandon Jones <bajones@chromium.org>
Commit-Queue: Brandon Jones <bajones@chromium.org>

[spirv] Gate NonPrivatePointer on cooperative matrix loads

SubgroupMatrixLoad() hardcodes SpvMemoryAccessNonPrivatePointerMask on
OpCooperativeMatrixLoadKHR. Ordinary accesses do not: the printer's
MemoryAccessMaskForPointer() requests it only under the Vulkan memory
model, for storage or workgroup pointers, and only when the access mode
is read_write. The same file already follows that rule for storage
textures, via NonPrivateTexel.

NonPrivatePointer makes the access take part in the memory model's
availability and visibility rules. A read-only pointer has nothing to
take part in, since no agent can write it for the life of the dispatch,
but honouring the operand still keeps the access out of storage that
other agents cannot observe.

Apply the same rule. Only its access mode term needs testing here: these
builtins accept nothing but storage and workgroup pointers, and subgroup
matrix already requires the memory model. SubgroupMatrixStore() is left
alone: storing needs a writable pointer, so the rule could never change
its result.

Measured with two builds differing only in this change, on Phi-4-mini
prefill at 512 tokens through ONNX Runtime's subgroup-matrix matmul
kernel, which reads its A operand from a var<storage, read> buffer: an
AMD Radeon RX 7900 XTX goes from 2300 to 5058 tok/s. On an NVIDIA
GeForce RTX 5080 and an Intel Arc 140V there was neither improvement nor
regression.

The added test covers a read-only pointer; the existing subgroup matrix
tests all use read_write, whose result is unchanged.

Bug: none
Change-Id: Iea60136b3aabe3274141bf1cdb93db2d7d4879a0
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/330695
Reviewed-by: Alan Baker <alanbaker@google.com>
SLSA-Policy-Verified: SLSA Policy Verification Service <devtools-gerritcodereview-exitgate@google.com>
Commit-Queue: Yang Gu <ygu@microsoft.com>

[dawn] Updates and spanifies CommandEncoder::WriteBuffer.

- Update the internal function to use the same argument
  types as Queue::WriteBuffer.
- Spanifies the native side of the function.
- Remove the handling for it on the wire side since it
  is trivially spanified once the argument types are
  updated.

Bug: 528925203, 526537254
Change-Id: I6329339e5f38502b470ee825dc5f476a6870f325
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/329015
Reviewed-by: Kai Ninomiya <kainino@chromium.org>
Commit-Queue: Loko Kung <lokokung@google.com>

[vulkan] Enqueue fence and increment serial atomically.

- The bug report notes a small window where we could
  theoretically race with a WaitAny on a submitted work
  done future queried and waited on after the serial was
  incremented, but before the fence was actually enqueued.
  This could result in us thinking that the serial
  completed because we use the existence of the fence as
  a signal. By making those things happen at the same time
  w.r.t the list of fences, we ensure that a race can't
  accidentally see the fence missing and assume that that
  serial had completed.

Bug: 536687607
Change-Id: I933b4e02e34081965977642b8496e86eb858925b
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/325975
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Auto-Submit: Loko Kung <lokokung@google.com>

AllocatorMemoryInstrumentationTest: skip test on WARP

We get spurious failures on the bots presumably because some of them are
running an older system WARP version. Skip for now.

Bug: 536069163
Change-Id: I62a8ba32e1892376ba6509b1a43b2211daaf8bb1
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/325596
Commit-Queue: Kai Ninomiya <kainino@chromium.org>
Reviewed-by: Kai Ninomiya <kainino@chromium.org>
Auto-Submit: Antonio Maiorano <amaiorano@google.com>

Fix build errors about implicit conversions

This patch fixes all the build errors about implicit conversions:
- error: implicit conversion increases floating-point precision:
         'const float' to 'double' [-Werror,-Wdouble-promotion]
- error: comparison of integers of different signs:
        'const unsigned int' and 'const int' [-Werror,-Wsign-compare]

Bug: none
Change-Id: Iceda4d9dbca241a244566b8d4db4c8a563e0c22e
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/323595
Commit-Queue: Shao, Jiawei <jiawei.shao@intel.com>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>

Move core type info to a Flag.

Currently we check the type info name for a prefix of `core::` to
determine if a type is core or not. Instead, add a new `kCore` Flag
which is set in all the types in core/type.

Change-Id: If24024088f84ac40b016a79fdaa1a835fa04ce9a
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/324435
Reviewed-by: Ryan Harrison <rharrison@chromium.org>
Reviewed-by: Steven Perron <stevenperron@google.com>
Commit-Queue: Kai Ninomiya <kainino@chromium.org>

[dawn][wire] Replace some raw allocations with HeapArray

Bug: 512465980
Change-Id: I9b2086af4c305ad5df60869df256bc6f7a7d3090
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/304235
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Auto-Submit: Kai Ninomiya <kainino@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>

GL: implement Feature::TransientAttachments

Implement support for TransientAttachment via Renderbuffers.
This is a partial reland of https://dawn-review.googlesource.com/c/dawn/+/308775
Renderbuffers are only used for texture formats containing depth or stencil.

Bug: 462577182

Change-Id: I92090e92d6e4e1ee1d732ce97d245aae66ea2ef6
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/323915
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Stephen White <senorblanco@chromium.org>

Roll ANGLE from 195369f6675e to b9aa07e9d122 (6 revisions)

https://chromium.googlesource.com/angle/angle.git/+log/195369f6675e..b9aa07e9d122

2026-06-24 cclao@google.com Vulkan: Fix the bug related to queue family global priority
2026-06-24 cclao@google.com Vulkan: Disable supportsGlobalPriority feature for all
2026-06-24 bsheedy@chromium.org Generate gn_isolate_map.pyl
2026-06-24 syoussefi@chromium.org MSL/AST: Handle non-unary/binary expr in loop-forward-progress
2026-06-24 angle-autoroll@skia-public.iam.gserviceaccount.com Roll vulkan-deps from e73c8345b94e to 5481b8c77757 (9 revisions)
2026-06-24 angle-autoroll@skia-public.iam.gserviceaccount.com Roll Chromium from e23ff548502b to 3ebaa3a7d74a (745 revisions)

If this roll has caused a breakage, revert this CL and stop the roller
using the controls here:
https://autoroll.skia.org/r/angle-dawn-autoroll
Please CC angle-team+autoroll@google.com,jrprice@google.com on the revert to ensure that a human
is aware of the problem.

To file a bug in ANGLE: https://bugs.chromium.org/p/angleproject/issues/entry
To file a bug in Dawn: https://bugs.chromium.org/p/dawn/issues/entry

To report a problem with the AutoRoller itself, please file a bug:
https://issues.skia.org/issues/new?component=1389291&template=1850622

Documentation for the AutoRoller is here:
https://skia.googlesource.com/buildbot/+doc/main/autoroll/README.md

Bug: None
Tbr: jrprice@google.com
Change-Id: If768b99981d6b3f3160e02258c363db489c1f430
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/319755
Commit-Queue: dawn-autoroll@skia-public.iam.gserviceaccount.com <dawn-autoroll@skia-public.iam.gserviceaccount.com>
Bot-Commit: dawn-autoroll@skia-public.iam.gserviceaccount.com <dawn-autoroll@skia-public.iam.gserviceaccount.com>