[POWER] Vectorize BF16 GEMV #4944
Merged
Conversation
This file contains 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
… to enable using BF16 GEMV for one dimension matrices. Updated unit test to support inc_x != 1 or inc_y for GEMV.
…LAS into vectorizeBF16GEMV
|
All these failures either are not related to the PR or it took too long to compile. |
|
Right, some of the jobs on Azure tend to time out when they get scheduled on ancient hardware. And I'm still trying to figure out locally why the mingw32 job is failing. |
Merged
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Outdated suggestions cannot be applied.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.
Vectorize BF16 GEMV - Up to 26X faster. Versions for VSX and MMA.
Add GEMM_GEMV_FORWARD_BF16 compilation flag to allow BF16 GEMV forwarding (like F32).
Updated BF16 to F32 unit test to allow for non inc_x/inc_y equal to 1 for GEMV.