ronlieb wrote:
smallest example and used latest upstream llvm build from this morning
#include <immintrin.h>
void log_d_scalar(double a_input)
{
__m128d va, ve;
ve = _mm_getexp_sd(va, va);
}
clang++ -march=haswell t.cpp
t.cpp:8:10: error: always_inline function '_mm_getexp_sd' requires target
feature 'avx512f', but would be inlined into function 'log_d_scalar' that is
compiled without support for 'avx512f'
8 | ve = _mm_getexp_sd(va, va);
| ^
1 error generated.
https://github.com/llvm/llvm-project/pull/71318
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
