Skip to content

cmake: require YMM asm for inline AVX probe #2385

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

Open
cary-ilm wants to merge 1 commit into
base: main
Choose a base branch
from
+11 −1
Open

cmake: require YMM asm for inline AVX probe #2385

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
12 changes: 11 additions & 1 deletion cmake/CMakeLists.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -37,7 +37,17 @@ int main()
int edx = 0;
__asm__(
\"xgetbv ;\"
\"vzeroupper \"
//
// This feature probe is specifically used to decide whether we can
// compile the DWA fast path that contains GCC-style inline asm using
// AVX/YMM instructions. Some toolchains (notably older Apple 'as'
// invoked by GCC on legacy macOS) can assemble xgetbv/vzeroupper but
// still reject YMM/VEX instructions with errors like:
// no such instruction: `vmovaps (%rsi), %ymm0'
// So include a representative YMM instruction in the try-compile.
//
\"vmovaps %%ymm0, %%ymm0 ;\"
\"vzeroupper \"
: \"=a\"(eax), \"=d\"(edx) : \"c\"(n) : );
#else
# error No SSE support enabled to query AVX support
Expand Down
Loading