Skip to content

<complex>: use hardware FMA on Clang too #5721

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 3 commits into from
Sep 22, 2025
Merged

<complex>: use hardware FMA on Clang too #5721

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
92339b3
`<complex>`: use hardware FMA on Clang too
AlexGuteniev Sep 15, 2025
0938a22
Merge remote-tracking branch 'upstream/main' into its-there-if-you-ask
AlexGuteniev Sep 15, 2025
363c239
__underscores__
AlexGuteniev Sep 15, 2025
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
8 changes: 5 additions & 3 deletions stl/inc/complex
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -21,8 +21,6 @@
// https://learn.microsoft.com/en-us/cpp/build/arm64-windows-abi-conventions?view=msvc-170#base-requirements
// Both floating-point and NEON support are presumed to be present in hardware.
#define _FMP_USING_STD_FMA
#elif defined(__clang__) // ^^^ defined(_M_ARM64) || defined(_M_ARM64EC) ^^^
// TRANSITION, not using x86/x64 FMA intrinsics for Clang yet
#elif defined(_M_IX86) || defined(_M_X64)
#define _FMP_USING_X86_X64_INTRINSICS
#include <__msvc_bit_utils.hpp>
Expand Down Expand Up @@ -167,7 +165,11 @@ namespace _Float_multi_prec {
}

#ifdef _FMP_USING_X86_X64_INTRINSICS
_NODISCARD inline double _Sqr_error_x86_x64_fma(const double _Xval, const double _Prod0) noexcept {
#ifdef __clang__
[[__gnu__::__target__("fma")]]
#endif // defined(__clang__)
_NODISCARD inline double
_Sqr_error_x86_x64_fma(const double _Xval, const double _Prod0) noexcept {
const __m128d _Mx = _mm_set_sd(_Xval);
const __m128d _Mprod0 = _mm_set_sd(_Prod0);
const __m128d _Mresult = _mm_fmsub_sd(_Mx, _Mx, _Mprod0);
Expand Down