On Sun, 6 Feb 2022, Samir Ribić via Tinycc-devel wrote:
A bit faster than VMOVUPS is VMOVAPS, but the numbers must be at addresses divisible by 64.
They are the same speed. VMOVUPS will go slowly if you give it an unaligned address, while VMOVAPS will fault; but they do exactly the same thing given an aligned address. That said, if you do intend to use aligned addresses for performance, you should probably use the 'aligned' instruction variants where you can, to check your work.
-E
_______________________________________________ Tinycc-devel mailing list Tinycc-devel@nongnu.org https://lists.nongnu.org/mailman/listinfo/tinycc-devel
