Justin Ruggles <justin.rugg...@gmail.com> writes:
> +static av_always_inline av_const int ff_ctz_c(unsigned int v)
> +{
> + int c;
> +
> + if (v & 0x1)
> + return 0;
> +
> + c = 1;
> + if (!(v & 0xffff)) {
> + v >>= 16;
> + c += 16;
> + }
> + if (!(v & 0xff)) {
> + v >>= 8;
> + c += 8;
> + }
> + if (!(v & 0xf)) {
> + v >>= 4;
> + c += 4;
> + }
> + if (!(v & 0x3)) {
> + v >>= 2;
> + c += 2;
> + }
> + c -= v & 0x1;
> +
> + return c;
> +}
This doesn't work for an input of zero. Is that intentional?
Furthermore, on CPUs with no CTZ intruction, but with CLZ, av_log2(v & -v)
is faster. GCC has been seen to use this trick in the __builtin_ctz()
function though I don't know if it always does it.
--
Måns Rullgård
m...@mansr.com
_______________________________________________
libav-devel mailing list
libav-devel@libav.org
https://lists.libav.org/mailman/listinfo/libav-devel
