Skip to content

Conversation

@Jiajun-Ji
Copy link
Contributor

Add next-softmax.mlir, whichAdd next-softmax.mlir, using TOSA representation as the baseline.
Add next-softmax-manual.mlir, using vectorization optimization, with a width of 16 × f32 = 512 bits, for AVX512.
Add next-softmax-run, next-softmax-lower, and next-softmax-vec-manual-run to the Makefile.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant