Dont use 256 bit code paths on ARM
ARM only has 128 bit registers. simde simulates 256 bit operations using them, which is fairly pointless for us.
This commit is contained in:
parent
23a4012aeb
commit
76c6630084
1 changed files with 4 additions and 2 deletions
|
|
@ -124,7 +124,8 @@ init_simd(void *x) {
|
|||
#ifdef __APPLE__
|
||||
#ifdef __arm64__
|
||||
// simde takes care of NEON on Apple Silicon
|
||||
has_sse4_2 = true; has_avx2 = true;
|
||||
// ARM has only 128 bit registers.
|
||||
has_sse4_2 = true; has_avx2 = false;
|
||||
#else
|
||||
do_check();
|
||||
// On GitHub actions there are some weird macOS machines which report avx2 not available but sse4.2 is available and then
|
||||
|
|
@ -138,7 +139,8 @@ init_simd(void *x) {
|
|||
#ifdef __aarch64__
|
||||
// no idea how to probe ARM cpu for NEON support. This file uses pretty
|
||||
// basic AVX2 and SSE4.2 intrinsics, so hopefully they work on ARM
|
||||
has_sse4_2 = true; has_avx2 = true;
|
||||
// ARM has only 128 bit registers.
|
||||
has_sse4_2 = true; has_avx2 = false;
|
||||
#else
|
||||
do_check();
|
||||
#endif
|
||||
|
|
|
|||
Loading…
Reference in a new issue