Skip to content

[CI][sycl-rel] Upd workflow #19293

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.

Sign up for GitHub

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

Jump to bottom
Merged
merged 4 commits into from
Jul 7, 2025
Merged

[CI][sycl-rel] Upd workflow #19293

merged 4 commits into from
Jul 7, 2025

Conversation

KornevNikita
Copy link
Contributor

@KornevNikita KornevNikita commented Jul 3, 2025 •
edited
Loading

Align with regular nightly.
Remove aws-cuda job, I believe it's not required here.
Also update tests failed on DG2 due to old driver.

@KornevNikita
[CI][sycl-rel] Upd workflow
5f8f10c 
Align with regular nightly.
Remove aws-cuda job, I believe it's not required here.
@KornevNikita KornevNikita requested a review from a team as a code owner July 3, 2025 14:56
@KornevNikita
Copy link
Contributor Author

KornevNikita commented Jul 3, 2025 •
edited
Loading

@AlexeySachkov FYI
test run: https://github.com/intel/llvm/actions/runs/16052924994

sarnex
sarnex approved these changes Jul 3, 2025
View reviewed changes
Copy link
Contributor

@sarnex sarnex left a comment

Choose a reason for hiding this comment

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

lgtm pending offline question about arc

KornevNikita and others added 3 commits July 4, 2025 13:10
@KornevNikita
Remove XFAIL
063c678
@sarnex @KornevNikita
[SYCL][ESIMD][E2E] Query max workgroups in group barrier test (#18161)
5057bdb 
We need to query the runtime.

Signed-off-by: Sarnie, Nick <[email protected]>
@bb-sycl @actions-user
@sarnex @KornevNikita
[GHA] Uplift Linux GPU RT version to 25.18.33578.6 (#18673)
4615064 
Scheduled drivers uplift

---------

Signed-off-by: Sarnie, Nick <[email protected]>
Co-authored-by: GitHub Actions <[email protected]>
Co-authored-by: Sarnie, Nick <[email protected]>
@KornevNikita KornevNikita requested review from a team as code owners July 4, 2025 12:00
@KornevNikita
Copy link
Contributor Author

Test run - https://github.com/intel/llvm/actions/runs/16072503870
test_event is flaky, unrelated to this patch.

@KornevNikita
Copy link
Contributor Author

@intel/dpcpp-esimd-reviewers @intel/sycl-graphs-reviewers @intel/sycl-matrix-reviewers @konradkusiak97 could you please approve? Just cherry-picks

@KornevNikita
Copy link
Contributor Author

KornevNikita commented Jul 7, 2025 •
edited
Loading

@EwanC could you please approve?
nevermind, I didn't notice Konrad is in this group

@KornevNikita KornevNikita mentioned this pull request Jul 7, 2025
Merged
dkhaldi
dkhaldi reviewed Jul 7, 2025
View reviewed changes
sycl/test-e2e/Matrix/joint_matrix_bf16_fill_k_cache_SLM.cpp
@@ -8,7 +8,7 @@
// UNSUPPORTED: target-nvidia, target-amd
// UNSUPPORTED-INTENDED: aspect-ext_intel_matrix isn't currently supported for
// other triples
// XFAIL: run-mode && gpu-intel-dg2
// XFAIL: run-mode && igc-dev
Copy link
Contributor

@dkhaldi dkhaldi Jul 7, 2025 •
edited
Loading

Choose a reason for hiding this comment

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

You need to keep gpu-intel-dg2 here. Otherwise, test will xfail on other gpus as well.

Copy link
Contributor Author

Choose a reason for hiding this comment

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

But it unexpectedly passes on DG2, see - https://github.com/intel/llvm/actions/runs/16052924994

Copy link
Contributor

Choose a reason for hiding this comment

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

If it passes on DG2, should not we remove XFAIL all together?

Copy link
Contributor Author

Choose a reason for hiding this comment

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

You mean should we remove the whole XFAIL string?

Copy link
Contributor

Choose a reason for hiding this comment

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

Correct, remove // XFAIL: run-mode && gpu-intel-dg2

Copy link
Contributor Author

Choose a reason for hiding this comment

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

I just aligned this string with the sycl branch: https://github.com/intel/llvm/blob/sycl/sycl/test-e2e/Matrix/joint_matrix_bf16_fill_k_cache_SLM.cpp
So now this file looks same.

Copy link
Contributor

Choose a reason for hiding this comment

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

In this case, you need to update the description for this test. If you are aligning with the sycl branch, keep XFAIL for this test.
However, for the other two tests, I see the xfail is only set for DG2 in the sycl branch. If you see they are xpassing now, it makes sense to remove the whole XFAIL string for the two remaining matrix tests.

Copy link
Contributor Author

Choose a reason for hiding this comment

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

I aligned all 3 tests with the sycl branch:

llvm/sycl/test-e2e/Matrix/joint_matrix_bf16_fill_k_cache_arg_dim.cpp

Lines 16 to 19 in 322cd13

// Waiting for the commit in IGC to be pulled into the driver to resolve the
// test.
// XFAIL: run-mode && igc-dev
// XFAIL-TRACKER: GSD-10510

Could you please clarify, why can't we just keep it this way?

dkhaldi
dkhaldi reviewed Jul 7, 2025
View reviewed changes
sycl/test-e2e/Matrix/joint_matrix_bf16_fill_k_cache_arg_dim.cpp
@@ -17,7 +17,7 @@

// Waiting for the commit in IGC to be pulled into the driver to resolve the
// test.
// XFAIL: gpu-intel-dg2 && run-mode
// XFAIL: run-mode && igc-dev
Copy link
Contributor

Choose a reason for hiding this comment

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

Suggested change
// XFAIL: run-mode && igc-dev
// XFAIL: gpu-intel-dg2 && run-mode && igc-dev

dkhaldi
dkhaldi reviewed Jul 7, 2025
View reviewed changes
sycl/test-e2e/Matrix/joint_matrix_bf16_fill_k_cache_runtime_dim.cpp
@@ -18,7 +18,7 @@

// Waiting for the commit in IGC to be pulled into the driver to resolve the
// test.
// XFAIL: gpu-intel-dg2 && run-mode
// XFAIL: run-mode && igc-dev
Copy link
Contributor

Choose a reason for hiding this comment

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

Suggested change
// XFAIL: run-mode && igc-dev
// XFAIL: gpu-intel-dg2 && run-mode && igc-dev

dkhaldi
dkhaldi approved these changes Jul 7, 2025
View reviewed changes
Copy link
Contributor

@dkhaldi dkhaldi left a comment

Choose a reason for hiding this comment

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

LGTM

@AlexeySachkov AlexeySachkov merged commit 219439c into sycl-rel-6_2 Jul 7, 2025
23 of 26 checks passed
@AlexeySachkov AlexeySachkov deleted the upd-sycl-rel-nightly branch July 7, 2025 15:22
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@sarnex sarnex sarnex approved these changes

@dkhaldi dkhaldi dkhaldi approved these changes

@MrSidims MrSidims MrSidims approved these changes

@konradkusiak97 konradkusiak97 konradkusiak97 approved these changes

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
7 participants
@KornevNikita @sarnex @dkhaldi @MrSidims @konradkusiak97 @AlexeySachkov @bb-sycl