-
Notifications
You must be signed in to change notification settings - Fork 290
Avx512f avx512vl #1010
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Avx512f avx512vl #1010
Conversation
…64,pd: mm256,mm; cvtepu8_epi32,epi64: mm256,mm
r? @Amanieu (rust-highfive has picked a reviewer for you, use r? to override) |
Same issue, osX github CI ---- core_arch::x86::abm::assert__lzcnt_u32_lzcnt stdout ---- |
…oundss_i64,si64,u64; mm_cvtroundu64_ss,sd
It seems i686-unknown-linux-gnu github CI has some problems? |
stdarch-test bugs? Compiling stdarch-test v0.1.0 (/checkout/crates/stdarch-test) |
|
Check against the Clang headers to see which intrinsics are wrapped in |
Thanks. Before we put "epi64" and "pd" tests on x86_64. Should we move those back to x86? |
The tests for intrinsics defined in x86 should be in x86, and the tests for intrinsics defined in x86_64 should be in x86_64. That way x86_64 runs tests in both x86 and x86_64. |
Considering how large these files are already, maybe you should split the tests off into a separate file. |
Yes, but all test will be around 20000 lines. |
I guess it doesn't really make a difference. |
cvtepi8_epi64: mm256,mm; cvtepu16_epi32,epi64: mm256,mm; cvtepu32_epi32_epi64, mm256,mm
cvtepu8_epi32,epi64: mm256,mm
cvtpd_epi32,epu32: mm256,mm
cvtpd_ps: mm256,mm
cvtps_epi32,epu32,ph: mm256,mm;
cvtsepi32_epi8,epi16: mm256,mm
cvtsepi32_mask_storeu_epi16,epi8: mm256,mm
cvtsepi64_epi16,epi32: mm256,mm
cvtsepi64_mask_storeu_epi8,epi16,epi32: mm256,mm
cvtusepi32_epi8,epi16: mm256,mm
cvtusepi32_mask_storeu_epi8,epi16: mm256, mm
cvtusepi64_epi8,epi16,epi32: mm256,mm
cvtusepi64_mask_storeu_epi8,epi16,epi32: mm256,mm
cvtsi512_si32
cvttpd_epi32,epu32: mm256,mm; cvttps_epi32,epu32: mm256,mm
mm_cvt_roundi64_ss,sd