On Sun, Jul 22, 2012 at 1:16 PM, Ronald S. Bultje <[email protected]> wrote:
> From: "Ronald S. Bultje" <[email protected]>
>
> This completes the conversion of h264dsp to yasm; note that h264 also
> uses some dsputil functions, most notably qpel. Performance-wise, the
> yasm-version is ~10 cycles faster (182->172) on x86-64, and ~8 cycles
> faster (201->193) on x86-32.
> ---
> libavcodec/x86/h264_deblock.asm | 168
> +++++++++++++++++++++++++++++++++++++++
> libavcodec/x86/h264dsp_mmx.c | 162 ++-----------------------------------
> 2 files changed, 175 insertions(+), 155 deletions(-)
> + xor b_idxq, b_idxq ; for (b_idx = 0; b_idx < edges; b_idx += step)
b_idxd
> + cglobal h264_loop_filter_strength, 9, 9, 0, bs, nnz, ref, mv, bidir, edges,
Spurious error message on x86_32, due to claiming to use 9 gprs. Which
would have been fatal if yasm supported user-defined fatal errors. Since
ifdeffing the cglobal would be inconvenient, fix the error message instead:
--- a/libavutil/x86/x86inc.asm
+++ b/libavutil/x86/x86inc.asm
@@ -446,6 +446,9 @@ DECLARE_ARG 7, 8, 9, 10, 11, 12, 13, 14
%macro PROLOGUE 2-4+ ; #args, #regs, #xmm_regs, arg_names...
%assign num_args %1
%assign regs_used %2
+ %if num_args > 7
+ %assign num_args 7
+ %endif
%if regs_used > 7
%assign regs_used 7
%endif
--
--Loren Merritt
_______________________________________________
libav-devel mailing list
[email protected]
https://lists.libav.org/mailman/listinfo/libav-devel
