Skip to content

[CUDA] Matrix tests now support legacy pass-manager #5475

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 2 commits into from
Feb 9, 2022

Conversation

JackAKirk
Copy link
Contributor

@JackAKirk JackAKirk commented Feb 4, 2022

Device code matrix tests were failing when using -fno-legacy-pass-manager : #5473

The device code names of the reinterpret_cast<int32_t*> variables are different when using the -fno-legacy-pass-manager clang option. This solution allows var1 from var1 = reinterpret_cast<t>(var2); to be named %call.ascast.i.i{{.*}}.i in LLVM device code which covers all cases when using both the legacy and new pass managers.

Signed-off-by: jack.kirk jack.kirk@codeplay.com

Signed-off-by: jack.kirk <jack.kirk@codeplay.com>
bader
bader previously approved these changes Feb 5, 2022
Copy link
Contributor

@bader bader left a comment

Choose a reason for hiding this comment

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

LGTM.
Just one minor suggestion, but it doesn't block the merge.

Switched test back on.

Signed-off-by: jack.kirk <jack.kirk@codeplay.com>
@JackAKirk JackAKirk changed the title [CUDA] Disable failing matrix device code tests. [CUDA] Matrix tests now support legacy pass-manager Feb 7, 2022
@vladimirlaz vladimirlaz merged commit b6dd5a2 into intel:sycl Feb 9, 2022
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.

3 participants