Hi, On Sat, Jul 14, 2012 at 9:29 PM, Justin Ruggles <justin.rugg...@gmail.com> wrote: > +%else ; sse > + mova xmm0, [srcq ] > + mova xmm1, [srcq+src1q] > + mova xmm2, [srcq+src2q] > + mova xmm3, [srcq+src3q] > + mova xmm4, [srcq+src4q] > + mova xmm5, [srcq+src5q] > + mulps xmm0, xmm6 > + mulps xmm1, xmm6 > + mulps xmm2, xmm6 > + mulps xmm3, xmm6 > + mulps xmm4, xmm6 > + mulps xmm5, xmm6 > + cvtps2pi m0, xmm0 > + cvtps2pi m1, xmm1 > + cvtps2pi m2, xmm2 > + cvtps2pi m3, xmm3 > + cvtps2pi m4, xmm4 > + cvtps2pi m5, xmm5 > + packssdw m0, m3 ; m0 = 0, 6, 3, 9 > + packssdw m1, m4 ; m1 = 1, 7, 4, 10 > + packssdw m2, m5 ; m2 = 2, 8, 5, 11 > + ; unpack words > + pshufw m3, m0, q1032 ; m3 = 3, 9, 0, 6 > + punpcklwd m0, m1 ; m0 = 0, 1, 6, 7 > + punpckhwd m1, m2 ; m1 = 4, 5, 10, 11 > + punpcklwd m2, m3 ; m2 = 2, 3, 8, 9 > + ; unpack dwords > + pshufw m3, m0, q1032 ; m3 = 6, 7, 0, 1 > + punpckldq m0, m2 ; m0 = 0, 1, 2, 3 (final) > + punpckhdq m2, m1 ; m2 = 8, 9, 10, 11 (final) > + punpckldq m1, m3 ; m1 = 4, 5, 6, 7 (final) > + mova [dstq ], m0 > + mova [dstq+ 8], m1 > + mova [dstq+16], m2
I'd agree with Loren that the use of "sse" as a function name, but having "mX" refer to mmx registers, is somewhat confusing. I guess it's OK since it's obvious from the code what is intended, just wanted to note that it's confusing. Ronald _______________________________________________ libav-devel mailing list libav-devel@libav.org https://lists.libav.org/mailman/listinfo/libav-devel
