Issue 8680 in angleproject: Numerous ANGLE tests failing on newer Linux/Intel config
4 views
Skip to first unread message
bshe… via monorail
Apr 29, 2024, 4:58:12 PM4/29/24
to angleproj...@googlegroups.com
Status: Available
Owner: ----
OS: Linux
Priority: Medium
Type: Defect
New issue 8680 by bsh...@chromium.org: Numerous ANGLE tests failing on newer Linux/Intel config
https://bugs.chromium.org/p/angleproject/issues/detail?id=8680
We're in the process of qualifying Ubuntu 22 with Mesa 23.2.1 for our Intel UHD 630 GPUs (up from Ubuntu 18 with Mesa 20.0.8), but quite a few ANGLE tests are failing on the newer config. Sample failed build: https://ci.chromium.org/ui/p/angle/builders/try/linux-exp-test/5/infra
The biggest issue is that all (or at least almost all) of the pixel tests fail due to:
Failing test because of unexpected error:
vk_renderer.cpp:824 (DebugUtilsMessenger): [ VUID-VkSwapchainCreateInfoKHR-presentMode-02839 ] Validation Error: [ VUID-VkSwapchainCreateInfoKHR-presentMode-02839 ] | MessageID = 0xfb70ccdb | vkCreateSwapchainKHR(): pCreateInfo->minImageCount 3, which is outside the bounds returned by vkGetPhysicalDeviceSurfaceCapabilitiesKHR() (i.e. minImageCount = 5, maxImageCount = 0). The Vulkan spec states: If presentMode is not VK_PRESENT_MODE_SHARED_DEMAND_REFRESH_KHR nor VK_PRESENT_MODE_SHARED_CONTINUOUS_REFRESH_KHR, then minImageCount must be greater than or equal to the value returned in the minImageCount member of the VkSurfaceCapabilitiesKHR structure returned by vkGetPhysicalDeviceSurfaceCapabilitiesKHR for the surface (https://www.khronos.org/registry/vulkan/specs/1.3-extensions/html/vkspec.html#VUID-VkSwapchainCreateInfoKHR-presentMode-02839)
Sample task: https://chromium-swarm.appspot.com/task?id=69303141fcf31110
There are a few other occurrences of this failure in other suites, though, e.g. in this angle_perftests suite https://chromium-swarm.appspot.com/task?id=69303141fd5b3a10:
[==========] Running 1 test from 1 test suite.
[----------] Global test environment set-up.
[----------] 1 test from BlitFramebufferPerf
[ RUN ] BlitFramebufferPerf.Run/vulkan_depth_stencil_2_samples
../../src/tests/perf_tests/ANGLEPerfTest.h:131: Failure
Failed
Failing test because of unexpected error:
vk_renderer.cpp:824 (DebugUtilsMessenger): [ VUID-VkSwapchainCreateInfoKHR-presentMode-02839 ] Validation Error: [ VUID-VkSwapchainCreateInfoKHR-presentMode-02839 ] | MessageID = 0xfb70ccdb | vkCreateSwapchainKHR(): pCreateInfo->minImageCount 3, which is outside the bounds returned by vkGetPhysicalDeviceSurfaceCapabilitiesKHR() (i.e. minImageCount = 5, maxImageCount = 0). The Vulkan spec states: If presentMode is not VK_PRESENT_MODE_SHARED_DEMAND_REFRESH_KHR nor VK_PRESENT_MODE_SHARED_CONTINUOUS_REFRESH_KHR, then minImageCount must be greater than or equal to the value returned in the minImageCount member of the VkSurfaceCapabilitiesKHR structure returned by vkGetPhysicalDeviceSurfaceCapabilitiesKHR for the surface (https://www.khronos.org/registry/vulkan/specs/1.3-extensions/html/vkspec.html#VUID-VkSwapchainCreateInfoKHR-presentMode-02839)
ERR: vk_renderer.cpp:824 (DebugUtilsMessenger): [ VUID-VkSwapchainCreateInfoKHR-presentMode-02839 ] Validation Error: [ VUID-VkSwapchainCreateInfoKHR-presentMode-02839 ] | MessageID = 0xfb70ccdb | vkCreateSwapchainKHR(): pCreateInfo->minImageCount 3, which is outside the bounds returned by vkGetPhysicalDeviceSurfaceCapabilitiesKHR() (i.e. minImageCount = 5, maxImageCount = 0). The Vulkan spec states: If presentMode is not VK_PRESENT_MODE_SHARED_DEMAND_REFRESH_KHR nor VK_PRESENT_MODE_SHARED_CONTINUOUS_REFRESH_KHR, then minImageCount must be greater than or equal to the value returned in the minImageCount member of the VkSurfaceCapabilitiesKHR structure returned by vkGetPhysicalDeviceSurfaceCapabilitiesKHR for the surface (https://www.khronos.org/registry/vulkan/specs/1.3-extensions/html/vkspec.html#VUID-VkSwapchainCreateInfoKHR-presentMode-02839)
../../src/tests/perf_tests/ANGLEPerfTest.cpp:928: Skipped
Failing test because of unexpected error:
vk_renderer.cpp:824 (DebugUtilsMessenger): [ VUID-VkSwapchainCreateInfoKHR-presentMode-02839 ] Validation Error: [ VUID-VkSwapchainCreateInfoKHR-presentMode-02839 ] | MessageID = 0xfb70ccdb | vkCreateSwapchainKHR(): pCreateInfo->minImageCount 3, which is outside the bounds returned by vkGetPhysicalDeviceSurfaceCapabilitiesKHR() (i.e. minImageCount = 5, maxImageCount = 0). The Vulkan spec states: If presentMode is not VK_PRESENT_MODE_SHARED_DEMAND_REFRESH_KHR nor VK_PRESENT_MODE_SHARED_CONTINUOUS_REFRESH_KHR, then minImageCount must be greater than or equal to the value returned in the minImageCount member of the VkSurfaceCapabilitiesKHR structure returned by vkGetPhysicalDeviceSurfaceCapabilitiesKHR for the surface (https://www.khronos.org/registry/vulkan/specs/1.3-extensions/html/vkspec.html#VUID-VkSwapchainCreateInfoKHR-presentMode-02839)
[ FAILED ] BlitFramebufferPerf.Run/vulkan_depth_stencil_2_samples, where GetParam() = vulkan_depth_stencil_2_samples (61 ms)
[----------] 1 test from BlitFramebufferPerf (61 ms total)
--
You received this message because:
1. The project was configured to send all issue notifications to this address
You may adjust your notification preferences at:
https://bugs.chromium.org/hosting/settings
Owner: ----
OS: Linux
Priority: Medium
Type: Defect
New issue 8680 by bsh...@chromium.org: Numerous ANGLE tests failing on newer Linux/Intel config
https://bugs.chromium.org/p/angleproject/issues/detail?id=8680
We're in the process of qualifying Ubuntu 22 with Mesa 23.2.1 for our Intel UHD 630 GPUs (up from Ubuntu 18 with Mesa 20.0.8), but quite a few ANGLE tests are failing on the newer config. Sample failed build: https://ci.chromium.org/ui/p/angle/builders/try/linux-exp-test/5/infra
The biggest issue is that all (or at least almost all) of the pixel tests fail due to:
Failing test because of unexpected error:
vk_renderer.cpp:824 (DebugUtilsMessenger): [ VUID-VkSwapchainCreateInfoKHR-presentMode-02839 ] Validation Error: [ VUID-VkSwapchainCreateInfoKHR-presentMode-02839 ] | MessageID = 0xfb70ccdb | vkCreateSwapchainKHR(): pCreateInfo->minImageCount 3, which is outside the bounds returned by vkGetPhysicalDeviceSurfaceCapabilitiesKHR() (i.e. minImageCount = 5, maxImageCount = 0). The Vulkan spec states: If presentMode is not VK_PRESENT_MODE_SHARED_DEMAND_REFRESH_KHR nor VK_PRESENT_MODE_SHARED_CONTINUOUS_REFRESH_KHR, then minImageCount must be greater than or equal to the value returned in the minImageCount member of the VkSurfaceCapabilitiesKHR structure returned by vkGetPhysicalDeviceSurfaceCapabilitiesKHR for the surface (https://www.khronos.org/registry/vulkan/specs/1.3-extensions/html/vkspec.html#VUID-VkSwapchainCreateInfoKHR-presentMode-02839)
Sample task: https://chromium-swarm.appspot.com/task?id=69303141fcf31110
There are a few other occurrences of this failure in other suites, though, e.g. in this angle_perftests suite https://chromium-swarm.appspot.com/task?id=69303141fd5b3a10:
[==========] Running 1 test from 1 test suite.
[----------] Global test environment set-up.
[----------] 1 test from BlitFramebufferPerf
[ RUN ] BlitFramebufferPerf.Run/vulkan_depth_stencil_2_samples
../../src/tests/perf_tests/ANGLEPerfTest.h:131: Failure
Failed
Failing test because of unexpected error:
vk_renderer.cpp:824 (DebugUtilsMessenger): [ VUID-VkSwapchainCreateInfoKHR-presentMode-02839 ] Validation Error: [ VUID-VkSwapchainCreateInfoKHR-presentMode-02839 ] | MessageID = 0xfb70ccdb | vkCreateSwapchainKHR(): pCreateInfo->minImageCount 3, which is outside the bounds returned by vkGetPhysicalDeviceSurfaceCapabilitiesKHR() (i.e. minImageCount = 5, maxImageCount = 0). The Vulkan spec states: If presentMode is not VK_PRESENT_MODE_SHARED_DEMAND_REFRESH_KHR nor VK_PRESENT_MODE_SHARED_CONTINUOUS_REFRESH_KHR, then minImageCount must be greater than or equal to the value returned in the minImageCount member of the VkSurfaceCapabilitiesKHR structure returned by vkGetPhysicalDeviceSurfaceCapabilitiesKHR for the surface (https://www.khronos.org/registry/vulkan/specs/1.3-extensions/html/vkspec.html#VUID-VkSwapchainCreateInfoKHR-presentMode-02839)
ERR: vk_renderer.cpp:824 (DebugUtilsMessenger): [ VUID-VkSwapchainCreateInfoKHR-presentMode-02839 ] Validation Error: [ VUID-VkSwapchainCreateInfoKHR-presentMode-02839 ] | MessageID = 0xfb70ccdb | vkCreateSwapchainKHR(): pCreateInfo->minImageCount 3, which is outside the bounds returned by vkGetPhysicalDeviceSurfaceCapabilitiesKHR() (i.e. minImageCount = 5, maxImageCount = 0). The Vulkan spec states: If presentMode is not VK_PRESENT_MODE_SHARED_DEMAND_REFRESH_KHR nor VK_PRESENT_MODE_SHARED_CONTINUOUS_REFRESH_KHR, then minImageCount must be greater than or equal to the value returned in the minImageCount member of the VkSurfaceCapabilitiesKHR structure returned by vkGetPhysicalDeviceSurfaceCapabilitiesKHR for the surface (https://www.khronos.org/registry/vulkan/specs/1.3-extensions/html/vkspec.html#VUID-VkSwapchainCreateInfoKHR-presentMode-02839)
../../src/tests/perf_tests/ANGLEPerfTest.cpp:928: Skipped
Failing test because of unexpected error:
vk_renderer.cpp:824 (DebugUtilsMessenger): [ VUID-VkSwapchainCreateInfoKHR-presentMode-02839 ] Validation Error: [ VUID-VkSwapchainCreateInfoKHR-presentMode-02839 ] | MessageID = 0xfb70ccdb | vkCreateSwapchainKHR(): pCreateInfo->minImageCount 3, which is outside the bounds returned by vkGetPhysicalDeviceSurfaceCapabilitiesKHR() (i.e. minImageCount = 5, maxImageCount = 0). The Vulkan spec states: If presentMode is not VK_PRESENT_MODE_SHARED_DEMAND_REFRESH_KHR nor VK_PRESENT_MODE_SHARED_CONTINUOUS_REFRESH_KHR, then minImageCount must be greater than or equal to the value returned in the minImageCount member of the VkSurfaceCapabilitiesKHR structure returned by vkGetPhysicalDeviceSurfaceCapabilitiesKHR for the surface (https://www.khronos.org/registry/vulkan/specs/1.3-extensions/html/vkspec.html#VUID-VkSwapchainCreateInfoKHR-presentMode-02839)
[ FAILED ] BlitFramebufferPerf.Run/vulkan_depth_stencil_2_samples, where GetParam() = vulkan_depth_stencil_2_samples (61 ms)
[----------] 1 test from BlitFramebufferPerf (61 ms total)
--
You received this message because:
1. The project was configured to send all issue notifications to this address
You may adjust your notification preferences at:
https://bugs.chromium.org/hosting/settings
bshe… via monorail
Apr 29, 2024, 4:58:24 PM4/29/24
to angleproj...@googlegroups.com
Updates:
Cc: ynovi...@chromium.org
Comment #1 on issue 8680 by bsh...@chromium.org: Numerous ANGLE tests failing on newer Linux/Intel config
https://bugs.chromium.org/p/angleproject/issues/detail?id=8680#c1
(No comment was entered for this change.)
Cc: ynovi...@chromium.org
Comment #1 on issue 8680 by bsh...@chromium.org: Numerous ANGLE tests failing on newer Linux/Intel config
https://bugs.chromium.org/p/angleproject/issues/detail?id=8680#c1
(No comment was entered for this change.)
ynovi… via monorail
Apr 30, 2024, 9:02:04 AM4/30/24
to angleproj...@googlegroups.com
Updates:
Blockedon: 8672
Labels: Renderer-Vulkan
Owner: cc...@google.com
Comment #2 on issue 8680 by ynovi...@chromium.org: Numerous ANGLE tests failing on newer Linux/Intel config
https://bugs.chromium.org/p/angleproject/issues/detail?id=8680#c2
Charlie, looks like https://chromium-review.googlesource.com/c/angle/angle/+/5467850 didn't fix Linux Intel.
Could you PTAL?
Blockedon: 8672
Labels: Renderer-Vulkan
Owner: cc...@google.com
Comment #2 on issue 8680 by ynovi...@chromium.org: Numerous ANGLE tests failing on newer Linux/Intel config
https://bugs.chromium.org/p/angleproject/issues/detail?id=8680#c2
Charlie, looks like https://chromium-review.googlesource.com/c/angle/angle/+/5467850 didn't fix Linux Intel.
Could you PTAL?
ynovi… via monorail
Apr 30, 2024, 9:05:32 AM4/30/24
to angleproj...@googlegroups.com
Comment #3 on issue 8680 by ynovi...@chromium.org: Numerous ANGLE tests failing on newer Linux/Intel config
https://bugs.chromium.org/p/angleproject/issues/detail?id=8680#c3
Looks like this is investigated VVL-side in https://github.com/KhronosGroup/Vulkan-ValidationLayers/issues/7921.
I think we can suppress VUID-VkSwapchainCreateInfoKHR-presentMode-02839 in the meanwhile.
ynovi… via monorail
Apr 30, 2024, 10:42:55 AM4/30/24
to angleproj...@googlegroups.com
Comment #4 on issue 8680 by ynovi...@chromium.org: Numerous ANGLE tests failing on newer Linux/Intel config
https://bugs.chromium.org/p/angleproject/issues/detail?id=8680#c4
Brian, could you re-apply https://chromium-review.googlesource.com/c/angle/angle/+/5466741 to proceed with the experiment?
We can then land it if there is no quick fix VVL-side.
bshe… via monorail
Apr 30, 2024, 2:30:39 PM4/30/24
to angleproj...@googlegroups.com
Comment #5 on issue 8680 by bsh...@chromium.org: Numerous ANGLE tests failing on newer Linux/Intel config
https://bugs.chromium.org/p/angleproject/issues/detail?id=8680#c5
The tests look a lot happier with #c4 applied. Still some test failures, but a more reasonable amount, and many of them seem to just be pixel diffs.
Git Watcher via monorail
May 2, 2024, 2:47:18 PM5/2/24
to angleproj...@googlegroups.com
Comment #6 on issue 8680 by Git Watcher: Numerous ANGLE tests failing on newer Linux/Intel config
https://bugs.chromium.org/p/angleproject/issues/detail?id=8680#c6
The following revision refers to this bug:
https://chromium.googlesource.com/angle/angle/+/e560647104bef1f2fe99e772b71bc36ac7f6efa0
commit e560647104bef1f2fe99e772b71bc36ac7f6efa0
Author: Brian Sheedy <bsh...@chromium.org>
Date: Thu Apr 25 21:48:53 2024
Start Linux/Intel experiment
Bug: chromium:41496254, angleproject:8680, angleproject:8682
Bug: angleproject:8683
Change-Id: I1956ac642969cae0c17fff87388e899237b9df2d
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/5491789
Auto-Submit: Brian Sheedy <bsh...@chromium.org>
Reviewed-by: Yuly Novikov <ynov...@chromium.org>
Commit-Queue: Brian Sheedy <bsh...@chromium.org>
[modify] https://crrev.com/e560647104bef1f2fe99e772b71bc36ac7f6efa0/infra/specs/angle.json
[modify] https://crrev.com/e560647104bef1f2fe99e772b71bc36ac7f6efa0/src/libANGLE/renderer/vulkan/vk_renderer.cpp
[modify] https://crrev.com/e560647104bef1f2fe99e772b71bc36ac7f6efa0/infra/config/generated/luci-milo.cfg
[modify] https://crrev.com/e560647104bef1f2fe99e772b71bc36ac7f6efa0/src/tests/perf_tests/TracePerfTest.cpp
[modify] https://crrev.com/e560647104bef1f2fe99e772b71bc36ac7f6efa0/src/tests/angle_end2end_tests_expectations.txt
[modify] https://crrev.com/e560647104bef1f2fe99e772b71bc36ac7f6efa0/infra/specs/waterfalls.pyl
[modify] https://crrev.com/e560647104bef1f2fe99e772b71bc36ac7f6efa0/infra/config/main.star
[modify] https://crrev.com/e560647104bef1f2fe99e772b71bc36ac7f6efa0/infra/specs/mixins.pyl
Git Watcher via monorail
May 7, 2024, 4:44:18 PM5/7/24
to angleproj...@googlegroups.com
Comment #7 on issue 8680 by Git Watcher: Numerous ANGLE tests failing on newer Linux/Intel config
https://bugs.chromium.org/p/angleproject/issues/detail?id=8680#c7
The following revision refers to this bug:
https://chromium.googlesource.com/angle/angle/+/e29d643e5215c42f1aba7c5cd9925129212b6e05
commit e29d643e5215c42f1aba7c5cd9925129212b6e05
Author: Shahbaz Youssefi <syou...@chromium.org>
Date: Tue May 07 17:40:18 2024
Vulkan: Fix missing per-present-mode query
... if VK_EXT_surface_maintenance1 is supported but
VK_EXT_swapchain_maintenance1 isn't. The code mistakenly made the
appropriate query conditional to the presence of swapchain_maint1
instead of surface_maint1. On devices where surface_maint1 was
available but swapchain_maint1 wasn't, this caused a VVL error that is
now fixed.
Bug: angleproject:8680
Change-Id: Ide9a87f0e50887572f693d741d5476361320ea19
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/5522756
Reviewed-by: Cody Northrop <cnor...@google.com>
Reviewed-by: Charlie Lao <cc...@google.com>
Commit-Queue: Shahbaz Youssefi <syou...@chromium.org>
[modify] https://crrev.com/e29d643e5215c42f1aba7c5cd9925129212b6e05/include/platform/autogen/FeaturesVk_autogen.h
[modify] https://crrev.com/e29d643e5215c42f1aba7c5cd9925129212b6e05/src/libANGLE/renderer/vulkan/SurfaceVk.cpp
[modify] https://crrev.com/e29d643e5215c42f1aba7c5cd9925129212b6e05/src/libANGLE/renderer/vulkan/vk_renderer.cpp
[modify] https://crrev.com/e29d643e5215c42f1aba7c5cd9925129212b6e05/util/autogen/angle_features_autogen.h
[modify] https://crrev.com/e29d643e5215c42f1aba7c5cd9925129212b6e05/include/platform/vk_features.json
[modify] https://crrev.com/e29d643e5215c42f1aba7c5cd9925129212b6e05/util/autogen/angle_features_autogen.cpp
commit e29d643e5215c42f1aba7c5cd9925129212b6e05
Author: Shahbaz Youssefi <syou...@chromium.org>
Date: Tue May 07 17:40:18 2024
Vulkan: Fix missing per-present-mode query
... if VK_EXT_surface_maintenance1 is supported but
VK_EXT_swapchain_maintenance1 isn't. The code mistakenly made the
appropriate query conditional to the presence of swapchain_maint1
instead of surface_maint1. On devices where surface_maint1 was
available but swapchain_maint1 wasn't, this caused a VVL error that is
now fixed.
Bug: angleproject:8680
Change-Id: Ide9a87f0e50887572f693d741d5476361320ea19
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/5522756
Reviewed-by: Cody Northrop <cnor...@google.com>
Reviewed-by: Charlie Lao <cc...@google.com>
Commit-Queue: Shahbaz Youssefi <syou...@chromium.org>
[modify] https://crrev.com/e29d643e5215c42f1aba7c5cd9925129212b6e05/include/platform/autogen/FeaturesVk_autogen.h
[modify] https://crrev.com/e29d643e5215c42f1aba7c5cd9925129212b6e05/src/libANGLE/renderer/vulkan/SurfaceVk.cpp
[modify] https://crrev.com/e29d643e5215c42f1aba7c5cd9925129212b6e05/src/libANGLE/renderer/vulkan/vk_renderer.cpp
[modify] https://crrev.com/e29d643e5215c42f1aba7c5cd9925129212b6e05/util/autogen/angle_features_autogen.h
[modify] https://crrev.com/e29d643e5215c42f1aba7c5cd9925129212b6e05/include/platform/vk_features.json
[modify] https://crrev.com/e29d643e5215c42f1aba7c5cd9925129212b6e05/util/autogen/angle_features_autogen.cpp
syous… via monorail
May 7, 2024, 9:50:16 PM5/7/24
to angleproj...@googlegroups.com
Updates:
Owner: syou...@chromium.org
Status: Fixed
Comment #8 on issue 8680 by syou...@chromium.org: Numerous ANGLE tests failing on newer Linux/Intel config
https://bugs.chromium.org/p/angleproject/issues/detail?id=8680#c8
Owner: syou...@chromium.org
Status: Fixed
Comment #8 on issue 8680 by syou...@chromium.org: Numerous ANGLE tests failing on newer Linux/Intel config
https://bugs.chromium.org/p/angleproject/issues/detail?id=8680#c8
(No comment was entered for this change.)
Reply all
Reply to author
Forward
0 new messages