HIP: Remove GCN from list of devices that avoid MMQ #11831
Merged
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Outdated suggestions cannot be applied.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.
My intuition was that if the rocblas path is faster on CDNA with mfma disabled, its likely also faster on GCN gpus, as these are very similar.
After testing and discussion with @cb88 on Vega10 (GFX900) and Vega20 (GFX906) it however turns out to not be the case.
Further complicating things, the change in intended code path did not take effect until d6d24cd masking the fact that this is a pessimization at first.