Skip to content

Fix AVX512DQ. #1146

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
serge-sans-paille merged 1 commit into from
Jul 21, 2025
+11 −7
Merged

Fix AVX512DQ. #1146

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
14 changes: 7 additions & 7 deletions include/xsimd/arch/xsimd_avx512dq.hpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,8 +9,8 @@
* The full license is in the file LICENSE, distributed with this software. *
****************************************************************************/

#ifndef XSIMD_AVX512_DQHPP
#define XSIMD_AVX512_D_HPP
#ifndef XSIMD_AVX512DQ_HPP
#define XSIMD_AVX512DQ_HPP

#include "../types/xsimd_avx512dq_register.hpp"

Expand Down Expand Up @@ -47,12 +47,12 @@ namespace xsimd

// bitwise_not
template <class A>
XSIMD_INLINE batch<float, A> bitwise_not(batch<float, A> const& self, requires_arch<avx512f>) noexcept
XSIMD_INLINE batch<float, A> bitwise_not(batch<float, A> const& self, requires_arch<avx512dq>) noexcept
{
return _mm512_xor_ps(self, _mm512_castsi512_ps(_mm512_set1_epi32(-1)));
}
template <class A>
XSIMD_INLINE batch<double, A> bitwise_not(batch<double, A> const& self, requires_arch<avx512f>) noexcept
XSIMD_INLINE batch<double, A> bitwise_not(batch<double, A> const& self, requires_arch<avx512dq>) noexcept
{
return _mm512_xor_pd(self, _mm512_castsi512_pd(_mm512_set1_epi32(-1)));
}
Expand Down Expand Up @@ -96,7 +96,7 @@ namespace xsimd
// tmp1 = [a0..8, b0..8]
// tmp2 = [a8..f, b8..f]
#define XSIMD_AVX512_HADDP_STEP1(I, a, b) \
batch<float, avx512f> res##I; \
batch<float, avx512dq> res##I; \
{ \
auto tmp1 = _mm512_shuffle_f32x4(a, b, _MM_SHUFFLE(1, 0, 1, 0)); \
auto tmp2 = _mm512_shuffle_f32x4(a, b, _MM_SHUFFLE(3, 2, 3, 2)); \
Expand Down Expand Up @@ -180,7 +180,7 @@ namespace xsimd

// reduce_add
template <class A>
XSIMD_INLINE float reduce_add(batch<float, A> const& rhs, requires_arch<avx512f>) noexcept
XSIMD_INLINE float reduce_add(batch<float, A> const& rhs, requires_arch<avx512dq>) noexcept
{
__m256 tmp1 = _mm512_extractf32x8_ps(rhs, 1);
__m256 tmp2 = _mm512_extractf32x8_ps(rhs, 0);
Expand All @@ -192,7 +192,7 @@ namespace xsimd
namespace detail
{
template <class A>
XSIMD_INLINE batch<double, A> fast_cast(batch<int64_t, A> const& x, batch<double, A> const&, requires_arch<avx512dq>) noexcept
XSIMD_INLINE batch<double, A> fast_cast(batch<int64_t, A> const& self, batch<double, A> const&, requires_arch<avx512dq>) noexcept
{
return _mm512_cvtepi64_pd(self);
}
Expand Down
4 changes: 4 additions & 0 deletions include/xsimd/arch/xsimd_isa.hpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -72,6 +72,10 @@
#include "./xsimd_avx512f.hpp"
#endif

#if XSIMD_WITH_AVX512DQ
Copy link
Contributor Author

@degasus degasus Jul 21, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

FYI: XSIMD_WITH_AVX512CD is missing as well. However it does not define anything, but might yield the same issue again.

#include "./xsimd_avx512dq.hpp"
#endif

#if XSIMD_WITH_AVX512BW
#include "./xsimd_avx512bw.hpp"
#endif
Expand Down