-
|
I wonder when I set "CLANG=1" to use CLANG as my backend. Can the code atomatically leverage SIMD in intel CPU? If not, where I can support this? Thank you! |
Beta Was this translation helpful? Give feedback.
Answered by
nimlgen
Nov 6, 2024
Replies: 1 comment
-
|
No SSE/AVX intrinsics generated for now. Depends on llvm's auto-vectorization (which is not really good in our cases). Try with DEBUG>=6 to see asm and maybe different compilers (easy to switch in ops_clang.py) |
Beta Was this translation helpful? Give feedback.
0 replies
Answer selected by
chosen-ox
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
No SSE/AVX intrinsics generated for now. Depends on llvm's auto-vectorization (which is not really good in our cases). Try with DEBUG>=6 to see asm and maybe different compilers (easy to switch in ops_clang.py)