>+    pxor        m15, m15           ; zero

This appears to never be used.  Remove it, then put your most commonly
used constant in m15 instead.

>+idct_put_fn sse2, 16

The second argument is unnecessary because it is always the same.

>+    pmaddwd     m2,  m0, [w4_plus_w6]

Since this function has a vast number of memory arguments, you may
want to try using this technique to reduce code size and improve
instruction fetch/decode speed:
https://github.com/DarkShikari/x264-devel/commit/361b672b3cbe92f0a37a7385529253f554904ac2#L3R2601

Jason
_______________________________________________
libav-devel mailing list
[email protected]
https://lists.libav.org/mailman/listinfo/libav-devel

Reply via email to