Skip to content

Conversation

ianayl
Copy link
Contributor

@ianayl ianayl commented Oct 9, 2025

This PR adds precompiled CTS binaries to sycl-prebuilt-e2e-container.yml.

Note that this PR also renames some fields in devops/actions/run-tests/cts/action.yml to match that of run-tests/e2e/action.yml. If this is unwanted I will undo the changes.

@ianayl ianayl requested a review from a team as a code owner October 9, 2025 19:43
@sarnex sarnex requested a review from aelovikov-intel October 9, 2025 19:45
@ianayl
Copy link
Contributor Author

ianayl commented Oct 9, 2025

@aelovikov-intel ping for awareness

@ianayl
Copy link
Contributor Author

ianayl commented Oct 9, 2025

Test run at https://github.com/intel/llvm/actions/runs/18386413684

CTS tests are running as expected, but there seems to be a single failure. It doesn't seem to be related to the workflow, but I still need to look into why it is failing.

@KornevNikita
Copy link
Contributor

Let's also update the description in sycl-prebuilt-e2e-container.yml:

# The purpose of this is to build E2E tests with the latest release toolchain
# and then run them with the trunk SYCL RT (libsycl.so and friends) to verify
# that ABI compatibility hasn't been broken.

Copy link
Contributor

@aelovikov-intel aelovikov-intel left a comment

Choose a reason for hiding this comment

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

Looks reasonable in general, just a few inline comments.

default:
cts_ref:
type: choice
description: tag/sha to use for CTS -- hardcoded hash is last known good commit.
Copy link
Contributor

Choose a reason for hiding this comment

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

I don't think "good" is appropriate description. I think it's what our sycl-rel-smth branch is "known" to work with. Maybe @KornevNikita or @AlexeySachkov could suggest better wording here.

Copy link
Contributor

Choose a reason for hiding this comment

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

I like the word "stable", but is anything ever truly stable?

Copy link
Contributor

Choose a reason for hiding this comment

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

We take the sycl-cts hash closest to the last sycl commit that was added to the sycl-rel-* branch.
I think stable sounds good enough. Or may be some thing like "hardcoded hash is the SYCL-CTS version that successfully passes with the latest release toolchain"

Copy link
Contributor Author

Choose a reason for hiding this comment

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

How's "tag/sha to use for SYCL-CTS -- hardcoded hash is last known commit to pass with latest release toolchain"?

Copy link
Contributor

Choose a reason for hiding this comment

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

IMO, "SYCL CTS revision matching sycl-rel-N_M branch"

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 currently have "tag/sha to use for SYCL-CTS -- hardcoded hash is CTS revision matching sycl-rel-N_M branch."

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.

4 participants