[NFC][OpenMP] Update a test that was failing on aarch64. #164456
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
The failure was reported here: llvm#164039 (comment) The test is checking for the "bad" behavior so as to keep track of it. It could just be updated to check the "good" behavior and marked as XFAIL, but then it would not be as informative. The update for now is to fix the pointer arithmetic. If that isn't sufficient, we can fall back to doing the above.
|
@llvm/pr-subscribers-offload Author: Abhinav Gaba (abhinavgaba) ChangesThe failure was reported here: The test is checking for the "bad" behavior so as to keep track of it. It could just be updated to check the "good" behavior and marked as XFAIL, but then it would not be as informative. The update for now is to fix the pointer arithmetic. If that isn't sufficient, we can fall back to doing the above. Full diff: https://github.com/llvm/llvm-project/pull/164456.diff 1 Files Affected:
diff --git a/offload/test/mapping/use_device_addr/target_data_use_device_addr_class_member_ref_with_map.cpp b/offload/test/mapping/use_device_addr/target_data_use_device_addr_class_member_ref_with_map.cpp
index 5e8769eb3079d..204e1c3407724 100644
--- a/offload/test/mapping/use_device_addr/target_data_use_device_addr_class_member_ref_with_map.cpp
+++ b/offload/test/mapping/use_device_addr/target_data_use_device_addr_class_member_ref_with_map.cpp
@@ -16,7 +16,7 @@ struct ST {
int m = 0;
void f6() {
- uintptr_t offset = (uintptr_t)&d - n;
+ intptr_t offset = (uintptr_t)&d - n;
#pragma omp target data map(to : m, d)
{
void *mapped_ptr = omp_get_mapped_ptr(&d, omp_get_default_device());
@@ -34,10 +34,11 @@ struct ST {
// ref/attach modifiers:
// &ref_ptee(this[0].[d])), &ref_ptee(this[0].d), TO | FROM
// &ref_ptr(this[0].d), &ref_ptee(this[0].d), 4, ATTACH
- // EXPECTED: 1 0
- // CHECK: 0 1
- printf("%d %d\n", &d == mapped_ptr,
- (uintptr_t)&d == (uintptr_t)mapped_ptr - offset);
+ // EXPECTED: 1 0
+ // CHECK-NEXT: 0 1
+ intptr_t offset_device = (intptr_t)mapped_ptr - (intptr_t)&d;
+ printf("%d %d\n", &d == mapped_ptr, offset == offset_device);
+ printf("%lu %ld\n", offset, offset_device);
}
}
}
|
kparzysz
reviewed
Oct 21, 2025
offload/test/mapping/use_device_addr/target_data_use_device_addr_class_member_ref_with_map.cpp
Outdated
Show resolved
Hide resolved
This comment was marked as outdated.
This comment was marked as outdated.
…urrent-bad-behavior-from-uda-tests
This comment was marked as outdated.
This comment was marked as outdated.
|
Maybe I messed up the rebuild while in a hurry. I'm doing a fresh one now, so should know soon. |
|
Yeah, that was a fluke. Now it's failing with: |
|
That's interesting. The difference is that somehow on ARM, in addition to having the same difference as offset, the pointers have a difference of 0xffff in the top few bytes. Could you try one last time with HEAD commit? I would like to see if using Thanks for your help! |
|
|
Thanks for helping with the experiments! I relaxed the check for now. If the issues continue affecting the test even after the codegen fix, we might have to debug further. Thanks again for your help and patience :) |
abhinavgaba
deleted the
remove-checks-for-current-bad-behavior-from-uda-tests
branch
|
Thanks, the latest version seems to work now. I'll do a full rebuild to make sure. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
The failure was reported here:
#164039 (comment)
The test was checking for the "bad" behavior so as to keep track of it, but there seem to be some issues with the pointer arithmetic specific to aarch64.
The update for now is to not check for the "bad" behavior fully.
We may need to debug further if similar issues are encountered eventually once the codegen has been fixed.