-
Notifications
You must be signed in to change notification settings - Fork 769
[UR] Move adapter tests over to lit #18397
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
Conversation
39ef280
to
ddc25a9
Compare
ddc25a9
to
5f5257e
Compare
5f5257e
to
264dd01
Compare
264dd01
to
9467ceb
Compare
9467ceb
to
92db588
Compare
92db588
to
461abd8
Compare
461abd8
to
bba7032
Compare
bba7032
to
e57306b
Compare
e57306b
to
883e108
Compare
https://github.com/intel/llvm/actions/runs/15252687628/job/42893633907 Example of an (artificial) failure from this patch. |
883e108
to
ac415e2
Compare
@@ -26,6 +26,8 @@ set(INSTALL_GTEST OFF) | |||
FetchContent_MakeAvailable(googletest) | |||
enable_testing() | |||
|
|||
find_program(VALGRIND valgrind) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This seems to have just been moved from unified-runtime/test/adapters/CMakeLists.txt, but what do we need valgrind for? Is it to propagate it in the lit configuration further down?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Correct, it's so the .in file can be populated with the path to the valgrind binary. In theory I could add a .in file in test/adapters
, but that feels like more effort than it's worth.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Sounds reasonable, but a comment might be good.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Added a comment.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
One small open, but otherwise LGTM!
ac415e2
to
bf36d88
Compare
bf36d88
to
fc02196
Compare
This moves the CUDA/HIP and Level Zero adapter tests over to lit. New targets are available, `check-unified-runtime-level_zero`, `check-unified-runtime-hip` and `check-unified-runtime-cuda` where applicable. The generic `check-unified-runtime-adapter` can be used to run it all the above.
fc02196
to
9d40aff
Compare
@pbalcer Just noticed that the ctest invocation has |
Yes, the memcheck issue shouldn't block this patch from proceeding. If I had to guess this is an issue this particular version of memcheck in CI is not working well with the particular version of the driver in CI. Thanks for asking! |
@intel/llvm-gatekeepers Please merge. |
Change intel#18397 switches a filter for "memcheck" to an explicit path. However, that filter would have also hit deferred_kernel_memcheck.test. Since that test is also timing out randomly, disable it too.
Change #18397 switches a filter for "memcheck" to an explicit path. However, that filter would have also hit deferred_kernel_memcheck.test. Since that test is also timing out randomly, disable it too.
This moves the CUDA/HIP and Level Zero adapter tests over to
lit. New targets are available,
check-unified-runtime-level_zero
,check-unified-runtime-hip
andcheck-unified-runtime-cuda
whereapplicable. The generic
check-unified-runtime-adapter
can beused to run it all the above.