[0.6.0-UT] Fix Test Shard #519
Open
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
gulsumgudukbay
requested changes
Sep 10, 2025
| def test_shard_map(self): | ||
| if jtu.is_device_rocm: | ||
| self.skipTest("Skip on ROCm: tests/ffi_test.py::FfiTest::test_shard_map") | ||
| # if jtu.is_device_rocm: |
| mesh = jtu.create_mesh((len(jax.devices()),), ("i",)) | ||
| x = self.rng().randn(8, 4, 5).astype(np.float32) | ||
| n = len(jax.devices()) | ||
| x = x[:(x.shape[0] // n) * n] |
There was a problem hiding this comment.
This trims the array, however trimming the array could hinder the mathematical correction of this test. Instead of trimming we can also pad the array like I did in #509 to prevent information loss. Could you please explain how this is mathematically correct and we are not losing info?
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
This fix addresses a failure on the MI200 GPU, where the test case was unable to run due to the input array size not being divisible by the number of available GPUs, which is 7 for MI200. The solution trims the input array to ensure its size is divisible by the number of GPUs, allowing for proper distribution across the devices.