On 2012-01-12 19:01:27 +0100, Christophe Gisquet wrote: > 2012/1/12 Måns Rullgård <[email protected]>: > > Janne Grunau <[email protected]> writes: > > > >> 30-50% faster than the C implementation, 0.5% overall speedup on > >> bourne.rmvb. > > Probably this patch can stand on its own, but my patch series improves > those idct-only functions to perform idct and add at the same time. > > If those patches are finally accepted, it therefore may be unneeded.
I wasn't paying enough attention but why have you submitted a patch for dc only inverse transform with x86 asm if you're going to remove it a couple days later? Janne _______________________________________________ libav-devel mailing list [email protected] https://lists.libav.org/mailman/listinfo/libav-devel
