https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108881
Martin Liška <marxin at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |crazylht at gmail dot com,
| |marxin at gcc dot gnu.org
--- Comment #1 from Martin Liška <marxin at gcc dot gnu.org> ---
Test-case:
$ cat pr108881.i
typedef float __m256 __attribute__((__vector_size__(32)));
__attribute__((__vector_size__(16 * sizeof(short)))) short res2;
__m256 x3, x4;
avx512bf16_test() { res2 = __builtin_ia32_cvtne2ps2bf16_v16hi(x3, x4); }
It has changed since r13-3565-g6913cad2a38bc406:
gcc pr108881.c -c -mavx512bf16
pr108881.c: In function ‘avx512bf16_test’:
pr108881.c:5:18: warning: implicit declaration of function
‘__builtin_ia32_cvtne2ps2bf16_v16hi’; did you mean
‘__builtin_ia32_cvtne2ps2bf16_v16bf’? [-Wimplicit-function-declaration]
5 | res2 = __builtin_ia32_cvtne2ps2bf16_v16hi (x3, x4);
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| __builtin_ia32_cvtne2ps2bf16_v16bf
pr108881.c:5:18: error: incompatible types when assigning to type ‘__m256bh’
from type ‘int’
