Use WMMA when possible #26
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
Member
Author
|
Locally tests pass, CI is broken. |
Contributor
|
Wow!! Terrific! Excellent work🙏 Would BFloat16 require JuliaGPU/CUDA.jl#1425? |
Member
Author
Most likely, I've tried it without those and it gave a bunch of errors. |
Contributor
|
JuliaGPU/CUDA.jl#3009 could be promising. Note that it requires Float32 accumulation. Comparing ONIONop (no WMMA) to NNop master (with WMMA) with some buffer and type conversion to work around the accumulation: 
|
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Outdated suggestions cannot be applied.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.
For now CUDA.jl only. AMDGPU.jl will come next.
FP16: