While looking through the diff between Lilliput/JDK17 and upstream JDK17, I noticed one change in library_call.cpp that should not be there, it probably slipped through from debugging.
void clear_upper_avx() {
#ifdef X86
- if (UseAVX >= 2) {
+ if (false && UseAVX >= 2) {
C->set_clear_upper_avx(true);
}
#endif
}
This prevents emitting vzeroupper instructions on x86_64. I believe this will have performance impact when transitioning between AVX and SSE code. We should revert this.
void clear_upper_avx() {
#ifdef X86
- if (UseAVX >= 2) {
+ if (false && UseAVX >= 2) {
C->set_clear_upper_avx(true);
}
#endif
}
This prevents emitting vzeroupper instructions on x86_64. I believe this will have performance impact when transitioning between AVX and SSE code. We should revert this.