search

Lokathor / safe_arch

Exposes arch-specific intrinsics as safe function (via cfg).
https://docs.rs/safe_arch
Apache License 2.0
47 stars 8 forks source link

Missing 256-bit vpermps ? #104

Open HadrienG2 opened 1 year ago

HadrienG2 commented 1 year ago

So, I was looking through the shuffles multiple time, and I did not find an equivalent of the _mm256_permutevar8x32_ps Intel intrinsic, which maps into the vpermps hardware instruction.

According to your naming convention, I think it should be called shuffle_av_f32_all_m256.

Lokathor commented 1 year ago

Not sure how I missed this one, but sure I'd take a PR for this.

HadrienG2 commented 1 year ago

Ah, after looking through the source code, it's actually there, but incorrectly named shuffleavi32_all_m256.

So another candidate for naming convention cleanup.