search

DLTcollab / sse2neon

A translator from Intel SSE intrinsics to Arm/Aarch64 NEON implementation
MIT License
1.3k stars 208 forks source link

Implement `_mm_sha1` intrinsics #558

Closed Cuda-Chen closed 1 year ago

Cuda-Chen commented 1 year ago

Though _mm_sha1 intrinsics do not belong to SSE, I consider I would like to implement as SHA-1 is a widely used has function.

jserv commented 1 year ago

I would say "No" at first glance. At present, the SSE intrinsics coverage reaches 99%, and I am moving forward the extended AVX2NEON, which would reuse the infrastructure of SSE2NEON. The SHA1 series depend on wider memory access, and I would stick to AVX translation first.

Cuda-Chen commented 1 year ago

Got it. Then I will close this issue for not confusing users.