apache / tvm

Open deep learning compiler stack for cpu, gpu and specialized accelerators
https://tvm.apache.org/
Apache License 2.0
11.42k stars 3.4k forks source link

[FP8][Codegen] Add make_fp8 vector constructors #17065

Closed vinx13 closed 1 month ago

vinx13 commented 1 month ago

This allows vectorized fp8 loading.

Co-authored-by: Chris Sullivan [csullivan@octo.ai](mailto:csullivan@octo.ai)

cc @yongwww @csullivan @tqchen