-
Notifications
You must be signed in to change notification settings - Fork 130
[SYCL-MLIR] Start with only the single_task.cpp test case #1330
Conversation
f8dc503
to
3828374
Compare
Signed-off-by: Tsang, Whitney <[email protected]>
Signed-off-by: Tsang, Whitney <[email protected]>
f4a0cb3
to
a3914f5
Compare
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.
Just a few nits in the test-case.
On a more general note, I'm wondering if introducing a sub-directory with new tests is the most efficient way to test SYCL-MLIR.
The repository already contains a rather large set of SYCL tests that we could potentially re-use instead of introducing lots of new tests.
We could create a white-list of tests that should be working on our branch and alter the CMake target (or create a new one) to only run the tests in the white-list using the --filter
feature of llvm-lit
.
What do you think?
Signed-off-by: Tsang, Whitney <[email protected]>
1aade43
to
39af20f
Compare
The buildbots are failing because |
Signed-off-by: Tsang, Whitney <[email protected]>
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.
Changes look good, thanks.
IMHO we should just make sure to notify the user that Nvidia and AMD are currently not supported by the SYCL-MLIR-based compilation flow (other than just failing the test).
Signed-off-by: Tsang, Whitney <[email protected]>
Signed-off-by: Tsang, Whitney <[email protected]>
Signed-off-by: Tsang, Whitney <[email protected]>
Signed-off-by: Tsang, Whitney <[email protected]>
Signed-off-by: Tsang, Whitney <[email protected]>
Signed-off-by: Tsang, Whitney <[email protected]>
Signed-off-by: Tsang, Whitney <[email protected]>
Signed-off-by: Tsang, Whitney [email protected]