|
ef637366
|
2021-11-19T22:38:56
|
|
Revert the order of texture sync and frame buffer object sync
This change reverts the sync order swap done in
CL a280c671f178daf73da447d1030773b58b534998,
to address a performance issue in TRex.
The crash that sync order swap was
trying to resolve was addressed by
another CL 0fcad6260a3e3943fb84657a3a7f488d1e155fb7.
Bug: angleproject:6014
Change-Id: Ie49628295b6dcd5d012dc795bf123865b1a7f893
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3292636
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Charlie Lao <cclao@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
9de37b58
|
2021-11-20T21:45:02
|
|
Vulkan: Fix draw with emulated xfb without xfb active
When transform feedback is not active, make sure the code that captures
xfb is not generated, even though it's guarded against with an if. This
is because no buffer is bound to the capture buffer, and by avoiding
generation of code that references that buffer, SwiftShader will not
attempt to access the missing descriptor set.
Currently, the actual declaration of the buffer is left in the SPIR-V,
which seems to be benign. If that causes issues in the future, the
SPIR-V transformer can be modified to remove that variable as well.
Bug: chromium:1209285
Change-Id: I83da8d1dea5390ecacc681f8ada9f586932deaa4
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3294661
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
1f5eb6b8
|
2021-11-17T16:42:20
|
|
Avoid Android vkEnumerateDeviceExtensionProperties() bug
This works around a race-condition during Android start-up, when ANGLE
is used as the default GLES driver and when render engine (RE) is
using SkiaGL (which uses ANGLE, which uses Vulkan). The race
condition occassionally results in different numbers of extensions
between ANGLE's first and second calls to
vkEnumerateDeviceExtensionProperties(). In that case, the second call
would return VK_INCOMPLETE instead of VK_SUCCESS. That caused ANGLE
to fail to initialize, causing RE to fail to initialize.
This change works around this problem by increasing the number of
extensions asked for in the second call to
vkEnumerateDeviceExtensionProperties().
Background: Surface Flinger uses Hardware Composer (HWC) for
hardware-based composition (e.g. using overlays), and RE for GPU
composition (e.g. rendering to combine multiple app and system windows
together). SF, RE, and HWC all start about the same time. HWC sets a
property if it can support display timing. This gets passed through
SF to RE's Vulkan loader. The Vulkan loader uses that property to
determine whether to enable the VK_GOOGLE_display_timing extension.
The Vulkan loader used to make a synchronous call to SF in
vkEnumerateDeviceExtensionProperties() in order to get this property.
That took some number of milliseconds to complete and affected the
start-up time of every Vulkan/ANGLE app. To eliminate that
performance problem, the property now propogates in an asynchronous
manner. At that time, it was thought that RE would always get the
property in time. However, a partner's experience is that
VK_INCOMPLETE is happening 0.5% of the time.
ANGLE doesn't need to use the VK_GOOGLE_display_timing extension.
This is because the Android EGL loader provides the related
EGL_ANDROID_get_frame_timestamps extension. The issue that ANGLE is
working around is that it shouldn't fail to initialize in this
situation.
Bug: angleproject:6715
Bug: b/206733351
Change-Id: I4eb2197cdcc9692518b1bf5984d06fc8a1a7d145
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3290506
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
Commit-Queue: Ian Elliott <ianelliott@google.com>
|
|
5f755c29
|
2021-11-17T16:29:37
|
|
Vulkan: Lock around handle counter.
TSAN showed we could have a data race when multiple threads
were releasing objects because of the singleton handle
counter.
Bug: angleproject:6714
Change-Id: I23b5d343bec421a663198e7efc30c78dab2bde8f
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3288328
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
8700d9cc
|
2021-11-17T16:17:59
|
|
Vulkan: Lock around debug annotator.
TSAN showed thread contention could happen during make current
calls and when inserting debug markers.
Bug: angleproject:6714
Change-Id: I343b0bd6efbd46621f9e69bd8a329c135aed0d90
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3288327
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
06d34598
|
2021-11-17T16:16:27
|
|
Vulkan: Add mutex around cache stats.
TSAN showed thread contention could happen in multithreading
tests because of the singleton cache stats object.
Bug: angleproject:6714
Change-Id: I701aa0079b5eb9d60ef557b13ce57efba17df2fa
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3288326
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
c664c9fc
|
2021-11-16T12:22:47
|
|
Vulkan: Generalize barrier VVL suppression.
Seems to pop up in other tests as well. Seems to only affect
SwiftShader configs.
Run capture_replay_tests.py with the below test.
Test: BuiltinVariableVertexIdTest.LineLoop/ES3_Vulkan_SwiftShader
Bug: angleproject:6701
Change-Id: I857d4785d6e04edf96fb137a492548ed4d9fbcc4
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3285445
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
b9c0194c
|
2021-11-16T11:36:52
|
|
Add EVENT0 markers for why we call vkAcquireNextImage
This will help when looking at AGI system traces.
Bug: angleproject:6696
Change-Id: Ic4b06f1a1ec155076cfcb47299afe451cd325327
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3283428
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
27bc56c6
|
2021-11-15T18:18:53
|
|
Vulkan: MAP_UNSYNCHRONIZED_BIT: Skip ghosting/idling
Respect the following spec language:
No GL error is generated if pending operations which source or modify
the buffer overlap the mapped region, but the result of such previous
and any subsequent operations is undefined
Test: cpu time improves in unsync case in perf-tests/MapBufferRange.cpp
Bug: angleproject:6680
Change-Id: I6133952546735aced6e6ee8468ef2ac695316fb6
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3272018
Commit-Queue: Lingfeng Yang <lfy@google.com>
Reviewed-by: Charlie Lao <cclao@google.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
|
|
|
647a703e
|
2021-11-12T13:48:06
|
|
Vulkan: Reorder logic in BufferVk::mapRange
This CL flattens the logic, ordering read case first, then write, and
simpler cases before more complex ones.
This is to prepare for an optimization where we ignore certain paths if
MAP_UNSYNCHRONIZED_BIT is set. No change in functionality or performance
is expected.
Bug: angleproject:6680
Change-Id: I0a2e9ee969216c90353eac7af6dabf648dea2173
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3279615
Commit-Queue: Lingfeng Yang <lfy@google.com>
Reviewed-by: Tim Van Patten <timvp@google.com>
Reviewed-by: Charlie Lao <cclao@google.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
|
316dcb56
|
2021-10-06T09:31:49
|
|
Add SwANGLE Skia Gold testing.
Required updating our system info helper to be able to retrieve
SwiftShader device information.
Bug: angleproject:6496
Change-Id: Ib38ea4da65d199433e17b87df2630c3fd77cb619
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3208646
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Cody Northrop <cnorthrop@google.com>
|
|
|
d27552f2
|
2021-11-11T11:43:30
|
|
Re-land: "Vulkan: Allow SystemInfo to pick ICD."
Re-land fixes build on iOS and Android with ANGLE/Vulkan.
This will be used in conjunction with SwiftShader.
Bug: angleproject:6496
Change-Id: Id38403da1e377bba293dc8368d1c1aac29bf56a9
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3282426
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
3a84ef5e
|
2021-11-12T17:40:40
|
|
Add EGL_VULKAN_GET_INSTANCE_PROC_ADDR
Add EGL_VULKAN_GET_INSTANCE_PROC_ADDR for query Vulkan
vkGetInstanceProcAddr function pointer associate with the
Vulkan EGLDevice.
Bug: chromium:1264439
Change-Id: I4ff14e2aef7b8ca651e13b4d2867a540aed1b321
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3279100
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Peng Huang <penghuang@chromium.org>
|
|
|
3128c055
|
2021-11-12T14:59:46
|
|
Vulkan: Add wait semaphores to queueSubmitOneOff
For use in follow up change.
Bug: angleproject:3966
Change-Id: I5bfac51ef9d47a6df5d52268d3ce4863b848b1d1
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3279226
Reviewed-by: Charlie Lao <cclao@google.com>
Reviewed-by: Lingfeng Yang <lfy@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
573f9415
|
2021-11-12T12:22:39
|
|
Vulkan: Prefer aggregate barriers on SwiftShader
SwiftShader is not currently sensitivel to stage masks, so aggregating
barriers leads to more efficient rendering.
Bug: angleproject:4633
Change-Id: I262d1e98d4b08a1ca134a1c5c64ae964ceb6cd16
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3279020
Reviewed-by: Alexis Hétu <sugoi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
6fe13477
|
2021-11-11T00:47:14
|
|
Vulkan: Add external's pNext to ANGLE_external_objects_flags
ANGLE was chaining VkImageFormatListCreateInfoKHR to
VkImageCreateInfo::pNext to support sRGB extensions. For external
images, it was unknown whether that was valid because there was no way
to know if external used an identical chain of pNexts. This was causing
a discrepancy between images created by Chrome and those created by
ANGLE as part of an import.
This change updates ANGLE_external_objects_flags to take in the pNext
chain external has used to create the image so ANGLE could create the
image identically.
Bug: chromium:1266094
Change-Id: I479b9e7ff39d437425dc91c79834880749766f99
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3274177
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Peng Huang <penghuang@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
eb3ef08e
|
2021-11-11T17:41:48
|
|
Vulkan: Differentiate perf warnings and RP closure events
... and application events, by using different debug source enums,
translating to different colors in API debuggers.
Bug: angleproject:2472
Change-Id: Ice422c7ae59fb508a10500b9fbed79d4c9664c11
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3275840
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
|
dbc0c646
|
2021-11-06T01:09:26
|
|
Vulkan: Output the reason for RP closure in command buffer
To make it easier when viewing the command buffer in a graphics
debugger, this change inserts a marker just before closing the render
pass that specifies why the render pass was closed.
Bug: angleproject:2472
Change-Id: I862e500cd58332d6e199c853315c560fe6a73dc2
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3265609
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
a9f2e87e
|
2021-11-12T15:02:50
|
|
Revert "Vulkan: Allow SystemInfo to pick ICD."
This reverts commit b5adbe272e4829d77105e49607bf8ddafd385f17.
Reason for revert: Fails GN check on Android/iOS in Chrome:
https://chromium-review.googlesource.com/c/chromium/src/+/3276779/
Original change's description:
> Vulkan: Allow SystemInfo to pick ICD.
>
> This will be used in conjunction with SwiftShader.
>
> Bug: angleproject:6496
> Change-Id: I894aa4cf3b9473738b549de9941eb82e09234121
> Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3276515
> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
> Commit-Queue: Jamie Madill <jmadill@chromium.org>
TBR=syoussefi@chromium.org,jmadill@chromium.org,angle-scoped@luci-project-accounts.iam.gserviceaccount.com
Change-Id: I198802d6bc93993d3544ed4627e131cb7b93f381
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: angleproject:6496
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3276045
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
|
|
|
d2d3a546
|
2021-11-11T12:22:04
|
|
Vulkan: Write perf warnings in command buffer
It's much easier to understand what command the perf warning refers to
when it's visible in the command buffer using a graphics API debugger.
This change creates ANGLE_VK_PERF_WARNING which gives the warning both
to the application (through ANGLE_PERF_WARNING) and inserts it in the
command buffer.
Bug: angleproject:2472
Change-Id: Ie84feed53eca5cda93e1f2bc653fcbf9bcd57b56
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3275839
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
2aa62964
|
2021-11-11T13:24:27
|
|
Vulkan: Reset mFramebuffer when mFramebufferCache is cleared
The mFramebuffer pointer becomes stale when mFramebufferCache
is cleared. Set mFramebuffer to nullptr when this happens.
Test: --deqp-surface-type=fbo --deqp-case=KHR-GLES31.*
Bug: angleproject:6682
Change-Id: I5fd21a64f0f935de04e2934e794c915ccf880c16
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3276701
Reviewed-by: Tim Van Patten <timvp@google.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Mohan Maiya <m.maiya@samsung.com>
Commit-Queue: Brandon Schade <b.schade@samsung.com>
|
|
|
558981c1
|
2021-11-10T23:13:03
|
|
Vulkan: Make write-after-invalidate checks more precise
Previously, the size of the command buffer was used as indication for
whether the render pass attachments might have been modified after
glInvalidateFramebuffer. In that case, the invalidate was undone. This
is made more precise by making sure only vkCmdClearAttachments and
vkCmdDraw* calls are counted for this purpose.
For example, inserting event markers after glInvalidateFramebuffer now
retains the invalidation.
Note that this can be even further optimized by tracking real writes to
attachments. For example, currently a draw call with depth test
disabled still undoes the invalidation of the depth buffer, but it
shouldn't.
Bug: angleproject:5079
Change-Id: I6257b4116a73213884b919bc7f3c86ff39b6aeed
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3274176
Reviewed-by: Ian Elliott <ianelliott@google.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
b5adbe27
|
2021-11-11T11:43:30
|
|
Vulkan: Allow SystemInfo to pick ICD.
This will be used in conjunction with SwiftShader.
Bug: angleproject:6496
Change-Id: I894aa4cf3b9473738b549de9941eb82e09234121
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3276515
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
|
|
|
96b587b0
|
2021-11-10T15:10:09
|
|
Vulkan: Fix rendering to RGB external images
Based on the provided usage flags, the texture's image access mode is
updated to be renderable.
Bug: chromium:1266094
Change-Id: I5d092a2786fc12537ef0ad1f224137cc1fd8f864
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3270980
Reviewed-by: Craig Stout <cstout@google.com>
Reviewed-by: Charlie Lao <cclao@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
4e26eac5
|
2021-11-10T21:19:35
|
|
Vulkan: Fix VVL warnings on extensions that are not enabled
Bug: chromium:1266094
Change-Id: Idbae6cbe2c5593665438d118f17d184f8a81f0d9
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3274932
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
0fcad626
|
2021-11-09T14:02:08
|
|
Vulkan: Fix edge case with changing base level.
Bug: chromium:1267624
Change-Id: I36b983fdbbb258454215abe827837517df5a5aff
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3270971
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
cd822868
|
2021-10-28T13:00:09
|
|
Vulkan: MAP_INVALIDATE_RANGE_BIT: shadow or exclude
One method of dealing with glMapBufferRange + range invalidation; treat
it like bufferSubData and stage the update.
Another method is to ghost the buffer but copy only memory outside the
invalidated range.
This CL pursues a policy where if less than half of the buffer is
invalidated, we stage. Otherwise, we ghost and copy only memory outside
the invalidated range.
DynamicBuffer is chosen over DynamicShadowBuffer because it turns out to
end up implicitly tracking all active invalidate ranges (through its
freelist), and performs buffer copy on GPU. if we use a
DynamicShadowBuffer and then BufferVk::stagedUpdate, it's the same thing
but more work (an extra memcpy into the staging buffer). To make this
clear, we split the logic of stagedUpdate into two parts, the
allocation/map, and the flush, and reuse one half in glMapBufferRange,
and the other half in glUnmapBuffer.
Test: Faster performance in MapBufferRange perf test, no non-noisy
regress in trace tests
Bug: angleproject:6634
Change-Id: Ie2e6a9586824b8cb59a97419bb8052acd1de2033
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3251686
Reviewed-by: Tim Van Patten <timvp@google.com>
Reviewed-by: Charlie Lao <cclao@google.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Lingfeng Yang <lfy@google.com>
|
|
|
ef93b32c
|
2021-11-09T00:26:13
|
|
Vulkan: Fix deferred clears vs invalidate
In this scenario:
- Clear color
- Invalidate depth
- Clear color
The invalidate step flushed the deferred color clear, but the following
clear did not expect an open render pass without any draw calls in it.
This change fixes this issue, while simultaneously optimizing invalidate
by making sure the clears accumulated during syncState() are redeferred
instead of flushed.
This issue was discovered in
https://chromium-review.googlesource.com/c/angle/angle/+/3266176 where,
as part of an unrelated fix, an accidental render pass closure is
removed.
Bug: chromium:1267424
Change-Id: Icfc0a53dbf84e6339ee23960ed847444830054e6
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3266178
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
5c16f77b
|
2021-10-01T20:07:09
|
|
Vulkan: Add flushCommandsAndEndRenderPassWithoutQueueSubmit()
Add flushCommandsAndEndRenderPassWithoutQueueSubmit() to allow
ContextVk::flushAndGetSerial() to only issue a single vkQueueSubmit().
Bug: angleproject:6546
Change-Id: Ia3fc2bb93e88583ebd80ee153749dc8ca42245dc
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3200673
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Tim Van Patten <timvp@google.com>
|
|
|
53371cc0
|
2021-11-01T20:25:17
|
|
Vulkan: Use optimalBufferCopyOffsetAlignment
optimalBufferCopyOffsetAlignment is the optimal buffer offset alignment
in bytes for vkCmdCopyBufferToImage2KHR, vkCmdCopyBufferToImage,
vkCmdCopyImageToBuffer2KHR, and vkCmdCopyImageToBuffer. The per texel
alignment requirements are enforced, but applications should use the
optimal alignment for optimal performance and power use.
To improve efficiency, this CL updates ContextVk::mStagingBuffer's
alignment to the max of:
- minMemoryMapAlignment
- nonCoherentAtomSize
- optimalBufferCopyOffsetAlignment
On ARM, this is not expected to have any affect, since all three values
are 0x40, but other platforms may see a benefit.
Bug: angleproject:4297
Change-Id: I9185da111e09c5d782eb1dedb10369727cb9bf51
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3256007
Commit-Queue: Tim Van Patten <timvp@google.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Charlie Lao <cclao@google.com>
|
|
|
0e20c680
|
2021-11-08T15:24:09
|
|
Sync framebuffer bindings in glInvalidateFramebuffer
If a framebuffer binding change is followed by glInvalidateFramebuffer,
ANGLE was not syncing the framebuffer binding.
- This means that invalidation was being done on the previous
framebuffer.
- Paired with deferred clears, this was causing ContextVk to start a
render pass on the previous, potentially deleted, framebuffer.
Bug: chromium:1267027
Change-Id: I092a0c8dd764db9e49258b694c970babb19cf24b
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3266175
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
|
79a5dc0a
|
2021-11-05T23:26:43
|
|
Vulkan: Fix spammy best practices message
A GPU WRITE->HOST barrier frequently issued when the GPU outputs to a
host-visible buffer used the ALL_COMMANDS stage mask causing a best
practices message. This message is suppressed in ANGLE, but shows up in
RenderDoc and clutters the output.
Bug: angleproject:5070
Change-Id: I59aa7f8c7b3bf2788a3f7ae0ab6abdb4d3cd175e
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3265606
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
5d8a89e4
|
2021-11-02T19:57:04
|
|
Vulkan: Always override surface format GL_RGB8 to GL_RGBA8
If an app requests to create a surface with GL_RGB8, override it to be
GL_RGBA8 for Android.
This is to workaround an issue with the Android Vulkan loader which
limits which formats can be used with swapchains.
This CL also adds GL_RGB8 back to DisplayVkAndroid::generateConfigs(),
effectively reverting the following CL:
https://chromium-review.googlesource.com/c/angle/angle/+/3235466
This is being done with this CL (rather than reverting) since these
changes are required to handle surfaces created with GL_RGB8.
Bug: angleproject:6277
Bug: angleproject:6651
Change-Id: Iad78ea0d7bdf12e1e309ed6a7181f08fac38b9de
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3258143
Reviewed-by: Tim Van Patten <timvp@google.com>
Reviewed-by: Ian Elliott <ianelliott@google.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Tim Van Patten <timvp@google.com>
|
|
|
4fd4eb50
|
2021-11-05T11:14:47
|
|
Vulkan: Minor cleanups.
Makes some VkFlags into VkAccessFlags to aid with code search. Also
corrects a few typos.
Bug: angleproject:6566
Change-Id: Ia4d14d38bb6d4e8c4c71fed6038c12edae112be9
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3264224
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
d7090f5d
|
2021-10-29T10:45:19
|
|
Add more attributes for EGL_ANGLE_device_vulkan
Bug: chromium:1264439
Change-Id: Ibc9676cef16411b17c833abe9559721811016eef
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3264164
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Peng Huang <penghuang@chromium.org>
|
|
|
f7bcd910
|
2021-11-05T11:24:34
|
|
Vulkan: Skip VVL barrier warning on SwiftShader.
This gets tripped up from our barrier handling code.
Bug: angleproject:6656
Change-Id: Iefb1c898ca6506e81a19a0928beaf221dcdf8f4e
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3264225
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
0777af70
|
2021-11-02T11:14:06
|
|
Env var for EXT_framebuffer_fetch_non_coherent
* Added an environment variable/Android property to remove the
restriction on exposing GL_EXT_shader_framebuffer_fetch_non_coherent
on ARM and Qualcomm.
* Environment variable: ANGLE_ENABLE_EXT_SHADER_FRAMEBUFFER
_FETCH_NON_COHERENT_OVERRIDE
* Android property: debug.angle.enable.ext_shader_framebuffer
_fetch_non_coherent_override
Bug: angleproject:6519
Change-Id: I62e80f348c2ffda7d1d63496e3a2aedad9d69229
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3258363
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Charlie Lao <cclao@google.com>
|
|
|
afaa1287
|
2021-10-26T19:29:44
|
|
Vulkan: Don't submit XFB queries when XFB is inactive
Consider the following GL commands:
glBeginQuery(GL_TRANSFORM_FEEDBACK_PRIMITIVES_WRITTEN)
glBeginTransformFeedback()
glDrawArrays()
glPauseTransformFeedback()
glDrawArrays()
glResumeTransformFeedback()
glDrawArrays()
glEndTransformFeedback()
glEndQuery(GL_TRANSFORM_FEEDBACK_PRIMITIVES_WRITTEN)
The renderpass is broken for each of the draw calls, leading to
pauseRenderPassQueriesIfActive() and resumeRenderPassQueriesIfActive()
being called when the RP is stopped/started. This leads ANGLE to
generate 3 TransformFeedbackPrimitivesWritten queries when looping
through the "active" queries, since no consideration is made for XFB
actually being active/unpaused.
ARM doesn't support the TransformFeedbackPrimitivesWritten query when
XFB is disabled. Instead, the query is never completed and the results
are never available. It could probably be argued that this is an ARM
bug, and they should just output '0' if XFB isn't active during the
query.
Regardless, ANGLE shouldn't be issuing the queries when XFB is no
active, since it's wasteful.
Bug: angleproject:6622
Test: dEQP.GLES3/functional_transform_feedback*
Test: TransformFeedbackTest.TransformFeedbackQueryPausedDrawThenResume
Change-Id: I380fb60405d7f538b802bb0df72bf609f0bc58e9
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3246532
Reviewed-by: Charlie Lao <cclao@google.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Tim Van Patten <timvp@google.com>
|
|
ab426735
|
2021-10-29T13:54:09
|
|
Template gl::Rectangle so it can be used for float
Bug: angleproject:6598
Change-Id: I8cf5894f0e34c56a6ad856c978be93ea9d5ae113
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3253131
Commit-Queue: Gregg Tavares <gman@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
d2b4d9ae
|
2021-10-30T00:27:56
|
|
Revert "Vulkan: Don't expose MSRTT on SwiftShader"
This reverts commit fb77afd3c76301e11f5c7397d9a63d283f86036c.
Reason for revert: Breaks ANGLE roll into Chromium:
https://chromium-review.googlesource.com/c/chromium/src/+/3253966/
Please follow the directions here to update the reference images:
https://chromium.googlesource.com/chromium/src/+/HEAD/docs/testing/web_test_expectations.md
Original change's description:
> Vulkan: Don't expose MSRTT on SwiftShader
>
> Emulating this extension on SwiftShader is neither useful nor efficient
> (on the contrary, it hurts performance (which is true for all non-tiling
> architectures)). This extension was exposed purely to increase testing
> coverage, but is being disabled now as OOMs resulting from allocating
> the implict multisampled image is confusing Chrome. Note that
> SwiftShader exposes only a limited amount of memory.
>
> Bug: chromium:1263046
> Change-Id: Ia69b7fc63804169431885352a8f637a95ab08bd0
> Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3251581
> Reviewed-by: Geoff Lang <geofflang@chromium.org>
> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Bug: chromium:1263046
Change-Id: I554df85109282db840ed6d2dbf0b153da5a7a98f
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3254007
Reviewed-by: Tim Van Patten <timvp@google.com>
Commit-Queue: Tim Van Patten <timvp@google.com>
|
|
|
4a2446c6
|
2021-10-19T11:55:17
|
|
Vulkan: Implement robust shader outputs.
In this CL we change ANGLE to pass the mask of missing shader
outputs down to pipeline creation. We then use the color mask
bits to block SwiftShader writing to unused outputs.
This fixes the undefined behaviour present in Genshin Impact.
Note that the other GLES implementations we tested don't seem
to modify outputs even if they're unused.
It was easier to mask out the color attachments in initialize
rather than set up the pipeline desc to mask out the attachments.
This was because we manipulate the color mask in a fairly complex
way before we initialize the pipeline desc.
Bug: angleproject:6566
Change-Id: Ie659fcd511cd286fa573fd25e3e6a0b9e123ebd6
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3232435
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
|
|
|
800c6885
|
2021-10-25T13:37:39
|
|
Vulkan: Only add samplable formats to outCompressedTextureFormats
Only samplable texture formats should be added to
outCompressedTextureFormats in FormatTable::initialize(), so move that
to after the check of mActualSampleOnlyImageFormatID.
Bug: angleproject:6277
Change-Id: Id91908a04d7e94f2d6dd1f0fa499ebea6a1f107e
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3243068
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Charlie Lao <cclao@google.com>
Commit-Queue: Tim Van Patten <timvp@google.com>
|
|
7bda9d73
|
2021-10-22T11:53:20
|
|
DisplayVkAndroid: Always expose EGL_RECORDABLE_ANDROID.
Expose EGL_RECORDABLE_ANDROID in the config for all Vulkan drivers
on Android to provide app compatiblity with "My Talking Tom 2" and
"My Talking Tom Friends".
Don't require config to be unavailable in EGLRecordableTest and request
a config with EGL_RECORDABLE_ANDROID enabled, but do not check for
results.
Bug: angleproject:6612
Change-Id: I25113b7c7f9a4a3b13dc625f01d96ba91cb3c6d7
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3237001
Commit-Queue: Lubosz Sarnecki <lubosz.sarnecki@collabora.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Cody Northrop <cnorthrop@google.com>
|
|
|
fb77afd3
|
2021-10-28T11:07:53
|
|
Vulkan: Don't expose MSRTT on SwiftShader
Emulating this extension on SwiftShader is neither useful nor efficient
(on the contrary, it hurts performance (which is true for all non-tiling
architectures)). This extension was exposed purely to increase testing
coverage, but is being disabled now as OOMs resulting from allocating
the implict multisampled image is confusing Chrome. Note that
SwiftShader exposes only a limited amount of memory.
Bug: chromium:1263046
Change-Id: Ia69b7fc63804169431885352a8f637a95ab08bd0
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3251581
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
81d964a1
|
2021-10-28T10:50:51
|
|
vulkan: implement EGLDevice for vulkan backend
It allows ANGLE clients to get and use VkDevice and VkQueue
used by vulkan backend.
Bug: chromium:1264439
Change-Id: I338ac08152cfec50bb34c5025730e5e6368efba9
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3250964
Reviewed-by: Peng Huang <penghuang@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Peng Huang <penghuang@chromium.org>
|
|
|
32b3c963
|
2021-10-28T11:00:03
|
|
Vulkan: Remove MSRTT assertion in the presence of OOM
When the MSRTT transient multisampled image is created, the emulated
clear staged update is removed (as it's also staged for the single
sampled image). This was done after allocating memory for the image,
which can fail, leaving the update staged and leading to an ASSERT
failure when the texture is used.
Technically, the texture should not be used, and that's a Chrome bug
where the OOM result from ANGLE is being ignored. This change makes
sure the staged update is removed before allocating memory so the
multisampled image is in a more consistent state.
Bug: chromium:1263046
Change-Id: Ida8923ccbc517f740d76b4c2880086a039fa8a1f
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3251580
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
bae19e06
|
2021-10-26T13:35:57
|
|
Vulkan: Avoid unnecessary wait if mapBufferRange indicates read only
When we call BufferVk::mapRangeImpl(), both from internal code paths for
data reads or due to glMapBufferRange call, we are not passing the
access bit to the call. This CL passes the proper access bits to the
call and only wait for GPU writes to finish if access is for read only.
This CL also adds access bitfield to the BufferVk::mapImpl() API and
have various callers pass in the proper access bits as well.
Bug: b/203582620
Change-Id: Ica8493c902dbd7b15996266c81ce0fd4dbfc2520
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3245487
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
Commit-Queue: Charlie Lao <cclao@google.com>
|
|
|
191c236a
|
2021-10-14T17:01:45
|
|
Vulkan: Remove "current queue serial".
Queries, semaphores, and pipelines instead use the normal
vk::Resource design to track their lifetimes. Removes the current
serial APIs from all classes. Current serials are still tracked
internally in the command queue classes.
Bug: b/169788986
Change-Id: Idcd2c2a93bc8225c6f3f7c247eb8fcfb76be1030
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3223644
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
|
|
|
3a9f18f1
|
2021-10-18T10:44:38
|
|
Refactor program pipeline handling.
In preparation for moving more code from gl::Program to
gl::ProgramExecutable so it can be shared with ProgramPipeline.
Bug: angleproject:6566
Change-Id: Icb7ecccb37ae8e0d7d5fef8968f0dd7ef6fe6150
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3226305
Reviewed-by: Tim Van Patten <timvp@google.com>
Reviewed-by: Cody Northrop <cnorthrop@google.com>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
|
|
6ce67886
|
2021-10-28T13:45:18
|
|
Drop -Wweak-template-vtables
Clang is dropping suport for this flag, see bug.
Bug: chromium:1264351
Change-Id: I09b3d42889eecb8e786c52c205f28507fc797c5a
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3250885
Reviewed-by: Nico Weber <thakis@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
ca5e6f68
|
2021-10-27T11:28:08
|
|
Vulkan: Fix accessing stale FB cached variable.
This would happen when we start a query after deleting a
Framebuffer.
Bug: chromium:1262091
Change-Id: I595360bf55fe1757779669f168c95be802b70da5
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3248142
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
421dbf20
|
2021-10-26T19:06:19
|
|
Revert "Vulkan: Use different strategy for buffer memory allocation"
This reverts commit 0dbe308d91a3526e28fecd9014d873c649f4ef71.
Reason for revert: crbug.com/1253325
Original change's description:
> Vulkan: Use different strategy for buffer memory allocation
>
> This CL uses different memory allocation strategy based on the requested
> size. If the requested size exceeds 1M, we use dedicated memory
> allocation to avoid memory waste associated with the sub-allocator.
> Otherwise we uses VMA's sub-allocator pool. This CL creates two sets of
> customized pool so that we uses different allocation strategy for each
> set of pool: the small pool uses buddy algorithm which favors speed over
> memory and large pool uses default algorithm that favors memory saving
> over speed. This CL also replaces vmaFindMemoryTypeIndexForBufferInfo
> with vmaFindMemoryTypeIndex to avoid create and destroy VkBuffer object
> just try to find memoryTypeIndex.
>
> Bug: b/195588159
> Change-Id: I2bddbfffd77ba2ce6b9389d83a31051c4b748c4d
> Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2939490
> Commit-Queue: Charlie Lao <cclao@google.com>
> Reviewed-by: Jamie Madill <jmadill@chromium.org>
> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
> Reviewed-by: Tim Van Patten <timvp@google.com>
Bug: b/195588159
Change-Id: I2c8fe8cb2930d16f5212570d32be68e7c6a6e5f3
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3244258
Reviewed-by: Charlie Lao <cclao@google.com>
Commit-Queue: Charlie Lao <cclao@google.com>
Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
|
|
|
6b315a78
|
2021-10-26T19:04:06
|
|
Revert "Vulkan: Let BufferVk call into VMA for allocation when possible"
This reverts commit 894ce75fb2f75e718ce76e466b8938524f65ac07.
Reason for revert: crbug.com/1253325
Original change's description:
> Vulkan: Let BufferVk call into VMA for allocation when possible
>
> Previously BufferVk class maintains a DynamicBuffer pool per BufferVk
> object. This CL makes BufferVk skip DynamicBuffer pool in most cases and
> do its own BufferHelper allocation directly. DynamicBuffer pool is only
> used when desired, which is controled by a flag. With this CL, only
> UBO/SSBO/AtomicBuffer will still use DynamicBuffer pool if the buffer
> has to be allocated more than once.
>
> Bug: b/195588159
> Change-Id: I3aa08cef10ee9ee9f01f16403c6fbb99b37f4a8a
> Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2901241
> Commit-Queue: Charlie Lao <cclao@google.com>
> Reviewed-by: Tim Van Patten <timvp@google.com>
> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Bug: b/195588159
Change-Id: Iecda3baa6bc887fa0caa86ab076994cae7c10f93
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3244257
Reviewed-by: Charlie Lao <cclao@google.com>
Commit-Queue: Charlie Lao <cclao@google.com>
Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
|
|
|
9ac2409e
|
2021-10-26T00:55:06
|
|
Reland "Vulkan: Flush descriptor set updates during flush*Commands()"
This reverts commit 420e77a64e890ad4c585d72789ca59501be3ac75.
Reason for revert: The crash is in ProgramExecutableVk::getOrAllocateShaderResourcesDescriptorSet(), which is unrelated to this change. Creating a CL to reland this (the reland button is failing), to try and investigate further.
Original change's description:
> Revert "Vulkan: Flush descriptor set updates during flush*Commands()"
>
> This reverts commit 02b73c2fd738b237f1ab3ecd400feec222903a48.
>
> Reason for revert: Causes test to crash on Linux/Intel: anglebug.com/6591
>
> Original change's description:
> > Vulkan: Flush descriptor set updates during flush*Commands()
> >
> > The intent of this CL is to reduce the number of descriptor set updates
> > by delaying the work until all of the GLES commands that could trigger a
> > re-update have been performed and the command stream is being flushed.
> > To achieve this, flushDescriptorSetUpdates() is being moved from
> > setupDraw()/setupDispatch() to
> > flushRenderPassCommands()/flushOutsideRPCommands().
> >
> > This change also exposed an issue where the BufferView handles were not
> > being preserved until flushDescriptorSetUpdates() was called. To resolve
> > this, flushDescriptorSetUpdates() is also being called during
> > BufferViewHelper::release() before the BufferView memory is released.
> >
> > Bug: angleproject:5706
> > Change-Id: I61e19af9c0fac891aa2115d72391459b80d22f19
> > Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2939385
> > Reviewed-by: Charlie Lao <cclao@google.com>
> > Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
> > Reviewed-by: Jamie Madill <jmadill@chromium.org>
> > Commit-Queue: Tim Van Patten <timvp@google.com>
>
> Bug: angleproject:5706
> Bug: angleproject:6591
> Change-Id: I9d1ee3fcb3d1aebc86e60896e0065cab847a92b4
> No-Presubmit: true
> No-Tree-Checks: true
> No-Try: true
> Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3233901
> Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Bug: angleproject:5706
Bug: angleproject:6591
Change-Id: Ibc7e934b82e6cd90e766f43135583f7c45acfe39
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3244024
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Tim Van Patten <timvp@google.com>
|
|
|
894ce75f
|
2021-05-21T10:59:14
|
|
Vulkan: Let BufferVk call into VMA for allocation when possible
Previously BufferVk class maintains a DynamicBuffer pool per BufferVk
object. This CL makes BufferVk skip DynamicBuffer pool in most cases and
do its own BufferHelper allocation directly. DynamicBuffer pool is only
used when desired, which is controled by a flag. With this CL, only
UBO/SSBO/AtomicBuffer will still use DynamicBuffer pool if the buffer
has to be allocated more than once.
Bug: b/195588159
Change-Id: I3aa08cef10ee9ee9f01f16403c6fbb99b37f4a8a
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2901241
Commit-Queue: Charlie Lao <cclao@google.com>
Reviewed-by: Tim Van Patten <timvp@google.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
|
0dbe308d
|
2021-07-16T14:26:20
|
|
Vulkan: Use different strategy for buffer memory allocation
This CL uses different memory allocation strategy based on the requested
size. If the requested size exceeds 1M, we use dedicated memory
allocation to avoid memory waste associated with the sub-allocator.
Otherwise we uses VMA's sub-allocator pool. This CL creates two sets of
customized pool so that we uses different allocation strategy for each
set of pool: the small pool uses buddy algorithm which favors speed over
memory and large pool uses default algorithm that favors memory saving
over speed. This CL also replaces vmaFindMemoryTypeIndexForBufferInfo
with vmaFindMemoryTypeIndex to avoid create and destroy VkBuffer object
just try to find memoryTypeIndex.
Bug: b/195588159
Change-Id: I2bddbfffd77ba2ce6b9389d83a31051c4b748c4d
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2939490
Commit-Queue: Charlie Lao <cclao@google.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
|
|
|
48b605d8
|
2021-10-25T09:46:16
|
|
Vulkan: Fix UBSAN error in CommandBatch
CommandBatch doesn't initialize all fields, causing UBSAN errors in the
move constructor.
Bug: chromium:1261861
Change-Id: I46b963a415ccc08d3d9db40de98d1968eea311e8
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3241544
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
ee672193
|
2021-10-06T16:51:42
|
|
SurfaceVk: Use vkGetPhysicalDeviceSurfaceFormats2KHR.
Use VK_KHR_get_surface_capabilities2 to determine if the requested
color space and surface format combination is supported.
Create a DoesSurfaceSupportFormatAndColorspace helper function
and use vkGetPhysicalDeviceSurfaceFormatsKHR as a fallback
in a new DoesSurfaceSupportFormat helper function.
Treat the surface format as not supported when a non-sRGB
colorspace was requested and VK_KHR_get_surface_capabilities2
is not available.
Move MapEglColorSpaceToVkColorSpace to anonymous namespace.
Emit angle::Result::Incomplete if a suitable surface format is not
found.
Bug: angleproject:2514
Bug: b/201993125
Change-Id: I28baf904cf8846ca575eaf83c93b3f593c1b63a2
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3208690
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Lubosz Sarnecki <lubosz.sarnecki@collabora.com>
|
|
|
52934a5d
|
2021-10-04T22:19:30
|
|
Vulkan: Use the UNDEFINED layout for ExternalPreInitialized
According to the spec, the UNDEFINED layout does not invalidate the
memory of external images.
Bug: angleproject:6460
Change-Id: I682e8c3501cc52431268ff2c3a063467ba9d2a65
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3203796
Reviewed-by: Charlie Lao <cclao@google.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
4c2c6bd2
|
2021-10-20T16:43:51
|
|
vulkan: fix read overrun in commonDebugUtilsLabel
The size of the incoming string should be the 3rd argument to
storePointerParameter(), not the aligned string size value. If we use
the latter, we read past the end of the incoming string. This tripped
the AddressSanitizer during the object label tests.
Bug: angleproject:6600
Change-Id: Idd72353803661684057753f5b01bb69af1d4a141
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3235900
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
66e8faf7
|
2021-10-20T14:06:08
|
|
Remove ProgramExecutable's "isCompute" property.
This consolidates the lists of uniforms, ssbos, etc into one.
Requires a few checks to change from graphics shader stages into
all shaders.
Bug: angleproject:6596
Change-Id: Ic8f6bfc4fa295c3bea9f5f1ded11e8fbca1c3164
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3233361
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
aac5d678
|
2021-10-20T11:48:57
|
|
Make "isCompute" private to ProgramExecutable.
This eliminates uses of "isCompute" is the Vulkan back-end. Instead
of checking the state flag, we can use the context of the current
command to determine if we're running a compute or a graphics
command. This will eventually lead to us being able to compile the
program pipeline objects before we run a draw or dispatch command.
Changes the driver uniforms descriptor desc to bind to both graphics
and compute shader stages to simplify the code. This could have
theoretical but low-risk performance implications.
Bug: angleproject:6595
Change-Id: Ie30d419b6ece5b33f5066a034d3805fe96519b36
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3233903
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
|
|
|
bdc633a8
|
2021-10-20T12:33:42
|
|
Pass command type down to ContextImpl::syncState.
This mirrors the command type passed to the object sync functions.
It will be useful to determine if we're syncing for a draw or a
dispatch call.
Bug: angleproject:6595
Change-Id: Ia04bd14a3c2dd2eb211c47a6e55f8ddcbfedfaaa
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3233904
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
|
|
|
78ae5d70
|
2021-10-20T12:15:33
|
|
Vulkan: Remove support for GL_RGB8 from DisplayVkAndroid
The Android Vulkan loader doesn't support GL_RGB8, so remove it from
DisplayVkAndroid::generateConfigs().
The list of supported swapchain formats is available at:
https://cs.android.com/android/platform/superproject/+/master:frameworks/native/vulkan/libvulkan/swapchain.cpp;l=465-486?q=GetNativePixelFormat
This fixes launching the Zillow app.
Bug: angleproject:6277
Change-Id: I34f06e2b45a1de77864701462bf11712285b7ae2
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3235466
Commit-Queue: Tim Van Patten <timvp@google.com>
Reviewed-by: Cody Northrop <cnorthrop@google.com>
Reviewed-by: Ian Elliott <ianelliott@google.com>
|
|
|
420e77a6
|
2021-10-20T15:02:45
|
|
Revert "Vulkan: Flush descriptor set updates during flush*Commands()"
This reverts commit 02b73c2fd738b237f1ab3ecd400feec222903a48.
Reason for revert: Causes test to crash on Linux/Intel: anglebug.com/6591
Original change's description:
> Vulkan: Flush descriptor set updates during flush*Commands()
>
> The intent of this CL is to reduce the number of descriptor set updates
> by delaying the work until all of the GLES commands that could trigger a
> re-update have been performed and the command stream is being flushed.
> To achieve this, flushDescriptorSetUpdates() is being moved from
> setupDraw()/setupDispatch() to
> flushRenderPassCommands()/flushOutsideRPCommands().
>
> This change also exposed an issue where the BufferView handles were not
> being preserved until flushDescriptorSetUpdates() was called. To resolve
> this, flushDescriptorSetUpdates() is also being called during
> BufferViewHelper::release() before the BufferView memory is released.
>
> Bug: angleproject:5706
> Change-Id: I61e19af9c0fac891aa2115d72391459b80d22f19
> Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2939385
> Reviewed-by: Charlie Lao <cclao@google.com>
> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
> Reviewed-by: Jamie Madill <jmadill@chromium.org>
> Commit-Queue: Tim Van Patten <timvp@google.com>
Bug: angleproject:5706
Bug: angleproject:6591
Change-Id: I9d1ee3fcb3d1aebc86e60896e0065cab847a92b4
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3233901
Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
d8c18ac6
|
2021-10-08T13:00:19
|
|
Re-land: "Isolate commit_id.h from other code."
Re-land limits the angle_version deps to Windows to fix an iOS
visibility rule violation in GN.
Instead of recompiling multiple source files, we can seal off the
files that include commit_id to prevent recompilations when the
commit changes and source files don't change.
Bug: angleproject:2551
Change-Id: I033f00ec7afe4bfd01e29e0eea8848eea27747a0
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3233899
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
|
|
|
4b75345a
|
2021-10-06T22:47:16
|
|
Vulkan: Don't create pipeline with depthWrite if !depthTest
depthWrite is ineffective without depthTest. This change makes sure
that depthWrite/depthTest combinations of true/false and false/false
generate the same Vulkan pipeline.
Bug: angleproject:6410
Change-Id: I47e1524379d40dfb956f026903e0affc7f4fc59d
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3210628
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Charlie Lao <cclao@google.com>
|
|
|
02b73c2f
|
2021-07-14T15:45:47
|
|
Vulkan: Flush descriptor set updates during flush*Commands()
The intent of this CL is to reduce the number of descriptor set updates
by delaying the work until all of the GLES commands that could trigger a
re-update have been performed and the command stream is being flushed.
To achieve this, flushDescriptorSetUpdates() is being moved from
setupDraw()/setupDispatch() to
flushRenderPassCommands()/flushOutsideRPCommands().
This change also exposed an issue where the BufferView handles were not
being preserved until flushDescriptorSetUpdates() was called. To resolve
this, flushDescriptorSetUpdates() is also being called during
BufferViewHelper::release() before the BufferView memory is released.
Bug: angleproject:5706
Change-Id: I61e19af9c0fac891aa2115d72391459b80d22f19
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2939385
Reviewed-by: Charlie Lao <cclao@google.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Tim Van Patten <timvp@google.com>
|
|
|
d3753039
|
2021-10-14T17:21:47
|
|
Reland "RendererVk: Enable VK_KHR_image_format_list on SwiftShader."
This is a reland of 8559d50b3b6fcc71b939be4a2c0ca3b48a797d61
Original change's description:
> RendererVk: Enable VK_KHR_image_format_list on SwiftShader.
>
> This reverts commit 08b42714f51dc87e16d52757b4b115037218a4ba.
>
> Bug: angleproject:5281
> Bug: chromium:1245774
> Change-Id: I773d95225301ccf3c2807fd2269b6d7da1195a16
> Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3223679
> Commit-Queue: Lubosz Sarnecki <lubosz.sarnecki@collabora.com>
> Commit-Queue: Jamie Madill <jmadill@chromium.org>
> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Bug: angleproject:5281
Bug: chromium:1245774
Bug: chromium:1260941
Change-Id: Ic6992bcc4b44aa16a2931271350a6f6d2a23d9c8
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3232555
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
|
|
|
e637e4c9
|
2021-10-18T13:54:00
|
|
Vulkan: Optimize updating blend state in pipeline desc
Updating blend funcs and equations always updated all 8 slots. Now
that's only done for the attachments that are present.
Bug: angleproject:6298
Change-Id: I58fa7e4dfa27d05fef54cc9d56c7b2aa5ef43dd8
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3202550
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
|
|
|
7defdb60
|
2021-10-19T14:07:07
|
|
Revert "Isolate commit_id.h from other code."
This reverts commit 6d7ebf656d06f1152967aa5f4457d3787a24732c.
Reason for revert: Suspected reason for failing ANGLE roll into Chromium
Original change's description:
> Isolate commit_id.h from other code.
>
> Instead of recompiling multiple source files, we can seal off the
> files that include commit_id to prevent recompilations when the
> commit changes and source files don't change.
>
> Bug: angleproject:2551
> Change-Id: I98800c5a8e04ec7d4c3e57ea0eb3b636d3b224bb
> Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3212895
> Commit-Queue: Jamie Madill <jmadill@chromium.org>
> Reviewed-by: Geoff Lang <geofflang@chromium.org>
> Reviewed-by: Jonah Ryan-Davis <jonahr@google.com>
Bug: angleproject:2551
Bug: angleproject:6588
Change-Id: I19f41141f57d0683efb9a4ec043e078a9f7dc925
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3226056
Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
|
|
|
4bb0a7af
|
2021-10-18T12:14:04
|
|
Vulkan: Async Queue: Fix signal of external semaphores
Per the Vulkan spec:
> ... when a semaphore wait operation is submitted to a queue:
>
> - A binary semaphore must be signaled, or have an associated semaphore
> signal operation that is pending execution.
It is incorrect for ANGLE to defer the submission that signals an
external semaphore, because then it's impossible for the application to
know when it can wait on the semaphore.
Bug: angleproject:6574
Bug: b/172704839
Change-Id: I5469b500b2f7d402acec31d9848585a9947843c6
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3226308
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
c9e0da8e
|
2021-10-18T13:54:00
|
|
Vulkan: Improve wording in DeferredClears.md
Bug: angleproject:1944
Change-Id: I268e8191cfb7aeb16e33c1f9b39382461c5ac3e7
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3226310
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Mohan Maiya <m.maiya@samsung.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
|
bb0e0dea
|
2021-10-18T21:46:44
|
|
Roll vulkan-deps from 5be75e2ae595 to a9d8c46f2b61 (6 revisions)
https://chromium.googlesource.com/vulkan-deps.git/+log/5be75e2ae595..a9d8c46f2b61
Changed dependencies:
* spirv-tools: https://chromium.googlesource.com/external/github.com/KhronosGroup/SPIRV-Tools.git/+log/06ebc4806b..35fd0e17d0
* vulkan-loader: https://chromium.googlesource.com/external/github.com/KhronosGroup/Vulkan-Loader.git/+log/6d9e5b07d6..61688e09fe
* vulkan-validation-layers: https://chromium.googlesource.com/external/github.com/KhronosGroup/Vulkan-ValidationLayers.git/+log/77cd59ba80..78b1f89454
If this roll has caused a breakage, revert this CL and stop the roller
using the controls here:
https://autoroll.skia.org/r/vulkan-deps-angle-autoroll
Please CC syoussefi@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 report a problem with the AutoRoller itself, please file a bug:
https://bugs.chromium.org/p/skia/issues/entry?template=Autoroller+Bug
Documentation for the AutoRoller is here:
https://skia.googlesource.com/buildbot/+doc/main/autoroll/README.md
Bug: angleproject:6584
Change-Id: I8da9da9f82ccaf781d89a4f7975fb7520d3b6b28
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3226052
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
670b58d8
|
2021-10-18T19:50:10
|
|
Revert "RendererVk: Enable VK_KHR_image_format_list on SwiftShader."
This reverts commit 8559d50b3b6fcc71b939be4a2c0ca3b48a797d61.
Reason for revert: Causes MSAN failures. See crbug.com/1260941
Original change's description:
> RendererVk: Enable VK_KHR_image_format_list on SwiftShader.
>
> This reverts commit 08b42714f51dc87e16d52757b4b115037218a4ba.
>
> Bug: angleproject:5281
> Bug: chromium:1245774
> Change-Id: I773d95225301ccf3c2807fd2269b6d7da1195a16
> Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3223679
> Commit-Queue: Lubosz Sarnecki <lubosz.sarnecki@collabora.com>
> Commit-Queue: Jamie Madill <jmadill@chromium.org>
> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Bug: angleproject:5281
Bug: chromium:1245774
Bug: chromium:1260941
Change-Id: Id6712aed68ed04d2525549ae016594aa05243a5d
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3226312
Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
eb1d2474
|
2021-10-04T20:23:26
|
|
Vulkan: Add R8G8B8_UNORM to vk_format_map.json
The map of R8G8B8_UNORM -> VK_FORMAT_R8G8B8_UNORM is missing from
vk_format_map.json. This leads to attempting to get format bits for
angle::FormatID::NONE / VK_FORMAT_UNDEFINED.
This CL adds R8G8B8_UNORM to vk_format_map.json, so devices that support
it, like ARM, can use it. This allows those devices to import AHBs
created with R8G8B8_UNORM.
This CL also adds some additional ASSERT() and UNREACHABLE() calls to
catch missing formats sooner.
This allows the EGL tests to pass on ARM devices, but the tests are
still skipped due to Qualcomm not having the necessary support for
optimal tiling.
Bug: angleproject:6277
Test: dEQP-EGL.functional.image*
Change-Id: I41b7ead97a10ae80be4b6048d36d0d85ad71784f
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3204152
Commit-Queue: Tim Van Patten <timvp@google.com>
Reviewed-by: Charlie Lao <cclao@google.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
6d7ebf65
|
2021-10-08T13:00:19
|
|
Isolate commit_id.h from other code.
Instead of recompiling multiple source files, we can seal off the
files that include commit_id to prevent recompilations when the
commit changes and source files don't change.
Bug: angleproject:2551
Change-Id: I98800c5a8e04ec7d4c3e57ea0eb3b636d3b224bb
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3212895
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Jonah Ryan-Davis <jonahr@google.com>
|
|
|
ddedcdaf
|
2021-10-14T23:35:45
|
|
Vulkan: Recycle vkAcquireNextImageKHR semaphores
This change optimizes CPU performance by recycling semaphores used for
ANI instead of creating and deleting them every frame.
Bug: angleproject:6580
Change-Id: I151ea227870ed7b8bd123cbf629a65723c4696d2
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3225085
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Ian Elliott <ianelliott@google.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
|
f3334d03
|
2021-09-30T23:44:54
|
|
Vulkan: Fix importing layered AHBs
Bug: angleproject:6475
Change-Id: Iceb0880cadc54552d3f01593d2e12088cafa10cc
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3198733
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
22ecd454
|
2021-07-19T17:28:05
|
|
Vulkan: Move BufferMemory class from vk_helpers.h to vk_utils.h
In preparation for the StagingBuffer may also use BufferMemory, this CL
moves the class to vk_utils.h so that it can be used by any class
without create kinks on header file inclusion.
Bug: b/195588159
Change-Id: I97ab136be691322fe6284fd1b4b7336fdf1f904f
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3040106
Commit-Queue: Charlie Lao <cclao@google.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
|
|
|
f5d65922
|
2021-10-14T15:58:38
|
|
Vulkan: Optimize QueryVk::getResult()
This function always incurred a check for completed commands to ensure
forward progress for applications (or rather, tests) that busy-wait
until the query results become available.
This change instead does an initial isCurrentlyInUse() check and only if
that's true will it check to see if there are any finished commands.
This should reduce one checkCompletedCommands() call when the
application queries the status of the Query well after it's issued.
Based on patch authored by Igor Nazarov <i.nazarov@samsung.com>
Bug: angleproject:6573
Change-Id: Iad4ed6b7eb6ae5316707e6d916e5ca3054d6c5aa
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3225083
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
8ef1b2a2
|
2021-07-15T15:27:15
|
|
Vulkan: Add BufferMemoryAllocator class
BufferMemoryAllocator class will be responsible for memory allocation
for vkBuffers. The original Allocator class remains the wrapper for
vmaAllocator.
Bug: b/195588159
Change-Id: I26d016e2f8d13ba459d0aa0bd4e1ebf4d0d48251
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3036256
Commit-Queue: Charlie Lao <cclao@google.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
|
|
|
7aba3bc5
|
2021-10-14T23:14:37
|
|
Vulkan: Use a CircularBuffer helper class in SurfaceVk
There are two instances of circular buffer in SurfaceVk. A common
helper class is introduced by this change to consolidate the
implementation.
This is in preparation for a follow up change that introduces yet
another circular buffer in this class.
Bug: angleproject:6401
Change-Id: Id01b585567310a2528888114239a5c80ff93f1ee
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3225084
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Ian Elliott <ianelliott@google.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
|
123ba58d
|
2021-10-14T11:56:35
|
|
Vulkan: Remove "last submitted serial".
This fixes race conditions with the async command processor. Instead
of querying specific serial numbers, we ask the command queue to
either wait for idle, or return the answer to "are you busy" directly.
Bug: b/172704839
Change-Id: I06a8268d9b58d8c33b783af00ca74979ee158316
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3223641
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
1dc2b702
|
2021-10-13T16:02:37
|
|
Vulkan: Pass shader module map to init pipelines.
Instead of pulling out the shader modules from the shader map when
we start the call chain, pull them out right in the init call. This
saves a bunch of boilerplate code.
Refactoring change only.
Bug: angleproject:6566
Change-Id: Ib8d79bd7284d7ddb83522270f3d4df9086ab7300
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3221134
Reviewed-by: Tim Van Patten <timvp@google.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
|
|
|
8559d50b
|
2021-10-14T17:21:47
|
|
RendererVk: Enable VK_KHR_image_format_list on SwiftShader.
This reverts commit 08b42714f51dc87e16d52757b4b115037218a4ba.
Bug: angleproject:5281
Bug: chromium:1245774
Change-Id: I773d95225301ccf3c2807fd2269b6d7da1195a16
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3223679
Commit-Queue: Lubosz Sarnecki <lubosz.sarnecki@collabora.com>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
b95cdec3
|
2021-10-12T11:26:15
|
|
Vulkan: Switch to `std::unordered_map` to cache renderpass
ANGLE use `flat_hash_map` by default,while it will not retain pointer
stability which causes random null pointer issue when running MH31 with
asynccommandqueue.This change switch to using `std::unordered_map` to
cache renderpass.
Bug: angleproject:6436
Bug: angleproject:6569
Change-Id: Ia2829a41251eea39ce2a79a9a42b6fe9dff6b120
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3219010
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
f3d5dac3
|
2021-08-23T17:25:15
|
|
Vulkan: SPIR-V Gen: Drop dependency to glslang
The SPIR-V gen path is now made default. Compilation through glslang is
still supported for debugging, and is enabled on the GLSL* end2end tests
for smoke testing. On release builds, glslang is not supported.
To test with glslang, add the following gn arg (only necessary if dcheck
is disabled):
angle_enable_spirv_gen_through_glslang = true
Then enable the generateSPIRVThroughGlslang feature. This can be done
by setting an environment variable:
ANGLE_FEATURE_OVERRIDES_ENABLED=generateSPIRVThroughGlslang ./angle_deqp_gles2_tests
Binary size saving:
- 1.3MB on Linux (SPIR-V gen itself: 240KB)
- 730KB on Android (SPIR-V gen itself: 140KB)
Perf tests:
- LinkProgramBenchmark.Run/vulkan_compile_single_thread
* Through glslang:
truncated mean: 1287033.36
* Direct SPIR-V Gen:
truncated mean: 244495.91 (~80% reduction)
- LinkProgramBenchmark.Run/vulkan_compile_multi_thread
* Through glslang:
truncated mean: 4565894.83
* Direct SPIR-V Gen:
truncated mean: 1158164.10 (~75% reduction)
Bug: angleproject:4889
Bug: angleproject:6210
Change-Id: I486342702977c8114e90073b97183aba115a8b2d
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3115140
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
|
|
|
ea580484
|
2021-10-05T13:40:01
|
|
Vulkan: Add feature flag to prefer CPU copy instead of staged update
For ARM GPU, use GPU to do buffer to buffer copy has performance penalty
due to potential bubble in the vertex pipeline. This CL adds a feature
flag preferCPUForBufferDataSubData so that we can enable this behavior
for ARM GPUs.
This CL also tracks if GPU has referenced this BufferVk's storage since
it got new storage. Due to sub-allocation, we may get a new sub-range of
the same BufferHelper object when allocating new storage. But we
currently do not have a way to track GPU progress of the sub-range of a
buffer. So we will end up using BufferHelper's queueSerial to decide if
it is still GPU busy or not. This CL adds mHasBeenReferencedByGPU
boolean variable that will set to false when we got a new allocation and
set to true as soon as buffer is been referenced by any GPU command. We
use this to avoid checking queueSerial if it never been referenced by
GPU. This is a temporary workaround for the bug, the full fix is tracked
by https://issuetracker.google.com/201826021
Bug: b/200067929
Change-Id: I231fb0a678b0165a2ce1775d0aa4dbe7512fb4a8
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3183398
Commit-Queue: Charlie Lao <cclao@google.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
|
|
|
6793d50e
|
2021-10-06T15:48:55
|
|
Vulkan: Remove ImageHelper::setFirstAllocatedLevel and Add assertion
ImageHelper::mFirstAllocatedLevel is set via ImageHelper::initExternal.
There isn't a good use to set the mFirstAllocatedLevel when mImage is
invalid, and semantically it is wrong to try to get mFirstAllocatedLevel
or format while its storage is invalid (i.e., not allocated). This CL
removes ImageHelper::setFirstAllocatedLevel() API all together and added
assertion that image is valid when ImageHelper::getFirstAllocatedLevel
and getActualFormat and various other get APIs whose value are set
during initExternal call.
Bug: angleproject:6532
Change-Id: I9c97f488b3e959443fec632540e7f0906a49dc13
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3209838
Commit-Queue: Charlie Lao <cclao@google.com>
Reviewed-by: Tim Van Patten <timvp@google.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
185e48a7
|
2021-10-06T18:53:08
|
|
Vulkan: Make TextureVk::mRedefinedLevels bitmask of gl::LevelIndex
We have gl::LevelIndex and vk::LevelIndex. gl::LevelIndex means level
index at OpenGL concept and is used in TextureVk code. vk::LevelIndex
means already adjusted with baseLevel and is used by ImageHelper object.
This mRedefinedLevel is gl::TexLevelMask type, but it actually is a bit
mask of vk::LevelIndex right now. This is inconsistent and confusing.
This CL changes it to bit mask of gl::LevelIndex to be consistent with
the namespace of "gl".
Bug: angleproject:6532
Change-Id: Ic88e373f70d3bd3e3fea676765c82fd0941e245c
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3210726
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
Commit-Queue: Charlie Lao <cclao@google.com>
|
|
|
2ab7a178
|
2021-10-06T13:22:19
|
|
Vulkan: Remove suppressions for SPIR-V validation
A bug in VVL produced validation errors regarding texel gather offset,
but it seems to be fixed now.
Bug: angleproject:6491
Bug: angleproject:6492
Change-Id: Ia20816ea0be65aa185a8ee89cf58edd33e452875
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3209086
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
9ba5dcf6
|
2021-10-07T06:56:14
|
|
Vulkan: Fix inconsistency in half render area width/height
The half render area rotation matrix should be removed.
DrawableArea is constant over rotations.
The frag coord rotation matrix is still needed because
the frag coord generated by GPU is still rotated.
Test:
DISPLAY=:0 out/Debug/angle_deqp_gles3_rotate90_tests
Using test config with:
linux
nvidia
debug
vulkan
prerotation
prerotation270
Test: Cuttlefish SwANGLE
./cts-tradefed run cts -m CtsDeqpTestCases -t
dEQP-GLES3.functional.shaders.builtin_variable.fragcoord_xyz
Bug: angleproject:6504
Change-Id: I37a6308fdb7586200b19a6138affc78c2c329a96
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3212427
Commit-Queue: Lingfeng Yang <lfy@google.com>
Reviewed-by: Tim Van Patten <timvp@google.com>
Reviewed-by: Charlie Lao <cclao@google.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
|
6cb599f9
|
2021-10-06T13:07:27
|
|
Vulkan: Change dest to dst for consistency
Cleanup only, no functional change. dst aligns better with src.
Bug: angleproject:6502
Change-Id: I69821b1aae50a7ce647c7cc876468b6de309eec8
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3208514
Commit-Queue: Charlie Lao <cclao@google.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Cody Northrop <cnorthrop@google.com>
|
|
|
53388e54
|
2021-10-08T15:43:34
|
|
Vulkan: Remove VVL suppresions for VK_EXT_provoking_vertex
Suppressions were added as ANGLE was using an unreleased extension.
This extension is released and supported by VVL now.
Bug: angleproject:4063
Change-Id: Ief23ac1ac3a3e6ecbd7b04a6b583302bbdf48398
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3213298
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
4ad55d07
|
2021-10-05T15:45:05
|
|
Vulkan: Turn 1-view multiview FBOs into no-multiview
When only 1 view is enabled, ANGLE continued to append a
VkRenderPassMultiviewCreateInfo struct to the render pass. VVL produced
an error when transform feedback was used in this case, quoting that
multiview and transform feedback cannot mix.
This change makes sure that 1-view framebuffers are threated as if they
were not multiview.
Bug: angleproject:6478
Change-Id: If079c9a052f822342a49a9cc880be2577a356b64
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3206269
Reviewed-by: Charlie Lao <cclao@google.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
4c23b817
|
2021-10-08T19:29:08
|
|
Roll vulkan-deps from d78bd80b30f6 to 651804d6bf55 (11 revisions)
Suppressing SYNC-HAZARD-WRITE_AFTER_WRITE messages after
https://github.com/KhronosGroup/Vulkan-ValidationLayers/pull/3403
https://chromium.googlesource.com/vulkan-deps.git/+log/d78bd80b30f6..651804d6bf55
Changed dependencies:
* glslang: https://chromium.googlesource.com/external/github.com/KhronosGroup/glslang.git/+log/c8cdbac705..b9ba4c5743
* vulkan-loader: https://chromium.googlesource.com/external/github.com/KhronosGroup/Vulkan-Loader.git/+log/8c0ee92eb6..aeffbbd6b9
* vulkan-validation-layers: https://chromium.googlesource.com/external/github.com/KhronosGroup/Vulkan-ValidationLayers.git/+log/7a6a52723b..51279399ea
If this roll has caused a breakage, revert this CL and stop the roller
using the controls here:
https://autoroll.skia.org/r/vulkan-deps-angle-autoroll
Please CC ynovikov@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 report a problem with the AutoRoller itself, please file a bug:
https://bugs.chromium.org/p/skia/issues/entry?template=Autoroller+Bug
Documentation for the AutoRoller is here:
https://skia.googlesource.com/buildbot/+doc/main/autoroll/README.md
Bug: angleproject:6550, angleproject:6551
Tbr: ynovikov@google.com
Change-Id: I8b85fd82fc1042a5632027b8d6fada123533f276
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3214695
Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Commit-Queue: Yuly Novikov <ynovikov@chromium.org>
|
|
|
d134581b
|
2021-10-07T16:09:25
|
|
Vulkan: Remove SyncHelper::mEvent
Now that inserting a sync object incurs a flush, we can use the serial
to perform synchronization.
Bug: angleproject:6464
Bug: angleproject:6481
Change-Id: I8c82a12855b6497861ce34854d165af82956b384
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3212573
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
|
|
|
74b2886f
|
2021-09-22T13:49:57
|
|
Vulkan: Try to use CPU to copy when preserving old buffer data
When glBufferSubData is called, we may acquire a new buffer if buffer is
still GPU busy. When this happened, we have to preserve buffer content
if old buffer has valid data in it. Instead of always use GPU to do
copy, this CL will check if GPU is not writing to the buffer, we will
just use CPU to do the copy form old buffer to new buffer from the
ranges outside subData, controlled by the feature flag
preferCPUWhenPreservingBufferData.
Bug: b/200067929
Change-Id: I42053104b2be8da5f399cca92e934254988f2fd8
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3177322
Reviewed-by: Tim Van Patten <timvp@google.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Charlie Lao <cclao@google.com>
|
|
|
e437c4ad
|
2021-09-22T09:15:19
|
|
Vulkan: Only preserve buffer data when BufferVk has valid data
When we receive glBufferSubData call and GPU is still accessing the
buffer, we have two code paths to update data. If subData is more than
half of the entire buffer range, we choose to acquire a new buffer and
use DMA to copy the rest of buffer that outside of subData range from
the old buffer back to the new buffer so that existing buffer data is
being preserved. Otherwise we stage subData to use GPU buffer to buffer
copy later on when buffer is been used. The reasoning behind is to
minimize the amount of data copy. The improvement here is that if
previously app called glBufferData with null pointer, we really do not
have any valid data in the buffer and there is no need to preserve the
existing buffer data. This CL tracks whether buffer has any valid data
or not and also put this into consideration when we pick which code path
to go. We also use this information to avoid preserve the existing data
in BufferVk::acquireAndUpdate
Bug: b/200067929
Change-Id: I266dd93bed2d3c07e3a5af3e4e613e7f6023b393
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3176500
Commit-Queue: Charlie Lao <cclao@google.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
|