Skip to content

Implement missing AVX512 intrinsics #1600

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.

Sign up for GitHub

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

Merged
merged 7 commits into from
Jul 6, 2024
Merged

Implement missing AVX512 intrinsics #1600

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
4fc8598
Fix the stream intrinsics
sayantn Jun 29, 2024
fdc73f0
Implemented missing gather-scatters
sayantn Jun 30, 2024
5a4089c
Implemented VEX versions
sayantn Jul 1, 2024
2f14492
Implemented the missing AVX512BF16 intrinsics
sayantn Jul 1, 2024
0f1f4d1
Implemented some missing functions
sayantn Jul 2, 2024
520033a
Merge branch 'master' into completionist
sayantn Jul 2, 2024
5dc20a2
Added a `bf16` type
sayantn Jul 5, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
161 changes: 0 additions & 161 deletions crates/core_arch/missing-x86.md
View file
Open in desktop

Large diffs are not rendered by default.

12 changes: 6 additions & 6 deletions crates/core_arch/src/x86/avx.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1738,8 +1738,8 @@ pub unsafe fn _mm256_lddqu_si256(mem_addr: *const __m256i) -> __m256i {
#[stable(feature = "simd_x86", since = "1.27.0")]
pub unsafe fn _mm256_stream_si256(mem_addr: *mut __m256i, a: __m256i) {
crate::arch::asm!(
"vmovntdq [{mem_addr}], {a}",
mem_addr = in(reg) mem_addr,
vps!("vmovntdq", ",{a}"),
p = in(reg) mem_addr,
a = in(ymm_reg) a,
options(nostack, preserves_flags),
);
Expand All @@ -1766,8 +1766,8 @@ pub unsafe fn _mm256_stream_si256(mem_addr: *mut __m256i, a: __m256i) {
#[allow(clippy::cast_ptr_alignment)]
pub unsafe fn _mm256_stream_pd(mem_addr: *mut f64, a: __m256d) {
crate::arch::asm!(
"vmovntpd [{mem_addr}], {a}",
mem_addr = in(reg) mem_addr,
vps!("vmovntpd", ",{a}"),
p = in(reg) mem_addr,
a = in(ymm_reg) a,
options(nostack, preserves_flags),
);
Expand Down Expand Up @@ -1795,8 +1795,8 @@ pub unsafe fn _mm256_stream_pd(mem_addr: *mut f64, a: __m256d) {
#[allow(clippy::cast_ptr_alignment)]
pub unsafe fn _mm256_stream_ps(mem_addr: *mut f32, a: __m256) {
crate::arch::asm!(
"vmovntps [{mem_addr}], {a}",
mem_addr = in(reg) mem_addr,
vps!("vmovntps", ",{a}"),
p = in(reg) mem_addr,
a = in(ymm_reg) a,
options(nostack, preserves_flags),
);
Expand Down
4 changes: 2 additions & 2 deletions crates/core_arch/src/x86/avx2.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3149,9 +3149,9 @@ pub unsafe fn _mm256_srlv_epi64(a: __m256i, count: __m256i) -> __m256i {
pub unsafe fn _mm256_stream_load_si256(mem_addr: *const __m256i) -> __m256i {
let dst: __m256i;
crate::arch::asm!(
"vmovntdqa {a}, [{mem_addr}]",
vpl!("vmovntdqa {a}"),
a = out(ymm_reg) dst,
mem_addr = in(reg) mem_addr,
p = in(reg) mem_addr,
options(pure, readonly, nostack, preserves_flags),
);
dst
Expand Down
Loading