-
Notifications
You must be signed in to change notification settings - Fork 796
[SYCL][E2E][Bindless] Add new VK/DX12 interop tests for L0 backend #14945
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[SYCL][E2E][Bindless] Add new VK/DX12 interop tests for L0 backend #14945
Conversation
L0 backend recently added support of following VkFormat: VK_FORMAT_R16G16B16_SFLOAT VK_FORMAT_R16G16_SFLOAT VK_FORMAT_R32_SFLOAT VK_FORMAT_R16G16B16A16_SFLOAT VK_FORMAT_R8G8B8_UNORM VK_FORMAT_R8G8B8A8_UNORM In order to validate the support for oneAPI release, * Add vulkan_interop/sampled_images_l0.cpp to test above formats. * Update dx12_interop/read_write_unsampled.cpp to test above formats. * Update dx12_interop/read_write_unsampled.cpp to not test semaphore.
|
Note new test sampled_images_l0.cpp has compfail in CUDA AOT (GTX 1060, Windows): |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
We tested this PR on CUDA on Windows and Linux, seems to pass.
sycl/test-e2e/bindless_images/vulkan_interop/sampled_images_l0.cpp
Outdated
Show resolved
Hide resolved
….cpp Co-authored-by: Peter Žužek <peterzuzek@gmail.com>
|
Changes in commit 798faa4:
@ProGTX please review again, thanks. |
@intel/bindless-images-reviewers please review this new commit, thank you. |
Co-authored-by: cppchedy <najjarchedy@gmail.com>
|
@intel/llvm-gatekeepers please merge, thanks |
|
@wenju-he This PR seems to be causing an off by one error in |
thank you @DBDuncan . I reproduced the fail. I had tested this PR on CUDA windows, but somehow I missed the
The test was passing before this PR is because the output value after multiplying 10.1f is always larger than the maximum value of the type. I've submitted a fix: #15086 |
|
No problem. Thanks for the fix! |
…es (#15086) globalSize.size() may be larger than maximum value of int8/int16 types. Resolve #14945 (comment)
…ntel#14945) L0 backend recently added support of following VkFormat: VK_FORMAT_R16G16_SFLOAT VK_FORMAT_R32_SFLOAT VK_FORMAT_R16G16B16A16_SFLOAT VK_FORMAT_R8G8B8A8_UNORM In order to validate the support for upcoming oneAPI release, * Update vulkan_interop/sampled_images.cpp to test above formats. * Update vulkan_interop/unsampled_images.cpp to test above formats. * Update vulkan_interop/read_write_unsampled.cpp to not test semaphore. * Update dx12_interop/read_write_unsampled.cpp to test above formats. --------- Co-authored-by: Peter Žužek <peterzuzek@gmail.com> Co-authored-by: cppchedy <najjarchedy@gmail.com>
…es (intel#15086) globalSize.size() may be larger than maximum value of int8/int16 types. Resolve intel#14945 (comment)
L0 backend recently added support of following VkFormat:
VK_FORMAT_R16G16_SFLOAT
VK_FORMAT_R32_SFLOAT
VK_FORMAT_R16G16B16A16_SFLOAT
VK_FORMAT_R8G8B8A8_UNORM
In order to validate the support for upcoming oneAPI release,