Skip to content

Fix multi-GPU lit issue by using first detected agent #2031

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

Open
djramic wants to merge 3 commits into
base: develop
Choose a base branch
from
+5 −6
Open

Fix multi-GPU lit issue by using first detected agent #2031

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
926c3de
Set config.arch from first GPU only
djramic Oct 10, 2025
a7f2548
Merge branch 'develop' into lit_fix_multiple_gpu
djramic Oct 10, 2025
69bf3c9
Merge branch 'develop' into lit_fix_multiple_gpu
djramic Oct 14, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
11 changes: 5 additions & 6 deletions mlir/test/lit.site.cfg.py.in
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -80,12 +80,11 @@ config.features = None
if config.rocm_path:
try:
agents = get_agents()
config.arch = ','.join(agents)
for x in agents:
if not config.features:
config.features, config.arch_support_mfma, config.arch_support_wmma = get_arch_features(x)
config.substitutions.append(('%features', config.features))
if not config.arch:
if agents:
Copy link
Contributor

@justinrosner justinrosner Oct 10, 2025 •
edited
Loading

Choose a reason for hiding this comment

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

I know here is probably not the place to do this, but is there somewhere where we can emit a warning to the user that we are only using one of the available archs?

Copy link
Contributor Author

@djramic djramic Oct 14, 2025

Choose a reason for hiding this comment

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

I'm not sure what the best place for that is. This seems like it might make most sense to me:
https://github.com/ROCm/rocMLIR/blob/develop/mlir/test/common_utils/common.py#L58

config.arch = sorted(agents)[0]
Copy link
Contributor

@dhernandez0 dhernandez0 Oct 13, 2025

Choose a reason for hiding this comment

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

why are we sorting? let's imagine the user sets HIP_VISIBLE_DEVICES="1,0", I guess HIP will return gpu 1 first? we want to run on that device, right?

config.features, config.arch_support_mfma, config.arch_support_wmma = get_arch_features(config.arch)
config.substitutions.append(('%features', config.features))
else:
config.no_AMD_GPU = True
except subprocess.CalledProcessError:
config.no_AMD_GPU = True
Expand Down