Skip to content

[SYCL] Move SPIRV test into proper directory #6589

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

Merged
merged 1 commit into from
Aug 16, 2022

Conversation

aelovikov-intel
Copy link
Contributor

Something bad happened during SPIRV pulldown, fix it.

Something bad happened during SPIRV pulldown, fix it.
@aelovikov-intel
Copy link
Contributor Author

These

Failed Tests (1):
  SYCL :: XPTI/kernel/content.cpp

********************
Unexpectedly Passed Tests (2):

  SYCL :: SubGroup/reduce_spirv13.cpp
  SYCL :: SubGroup/reduce_spirv13_fp64.cpp

In OCL Linux testing are unrelated. I'll try to see if they have been reported already.

Copy link
Contributor

@asudarsa asudarsa left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM. Thanks

@aelovikov-intel
Copy link
Contributor Author

intel/llvm-test-suite#1040 seems to take care of the failing test.

@aelovikov-intel
Copy link
Contributor Author

@intel/llvm-gatekeepers PR is ready.

@steffenlarsen
Copy link
Contributor

Seems to need approval from @pvchupin as code owner.

@pvchupin pvchupin merged commit 2ad29fc into intel:sycl Aug 16, 2022
@pvchupin
Copy link
Contributor

@dbudanov-cmplr, can you follow up on this?
Would be good to know how it happened.

@pvchupin
Copy link
Contributor

intel/llvm-test-suite#1040 seems to take care of the failing test.

It used to fail on OpenCL CPU and GPU, but in your testing fails on CPU only.
I hope CPU testing just picked up this change, and GPU didn't. Otherwise we may have different behaviors per target.
Can watch in next PRs.
Tagging @steffenlarsen

@steffenlarsen
Copy link
Contributor

I hope CPU testing just picked up this change, and GPU didn't. Otherwise we may have different behaviors per target. Can watch in next PRs.

I am currently attempting to reproduce the failure.

@dbudanov-cmplr
Copy link
Contributor

dbudanov-cmplr commented Aug 17, 2022

@pvchupin

The problem was introduced by commit 36a19b6.
Which is actually a pick of KhronosGroup/SPIRV-LLVM-Translator@a827fb0

The only reasonable explanation of what happened is that there were 2 patches one for lib/SPIRV/SPIRVUtil.cpp which was taken and applied properly. And the second for test/transcoding/SPV_INTEL_joint_matrix/sycl_2020_namespace.ll which was taken relativity to llvm-spirv but applied to llvm.
I'm not sure if this is an automation issues or something else.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

6 participants