RepoMirrors
/
ffmpeg
mirror of https://git.ffmpeg.org/ffmpeg.git
1
0
Fork 0
Code Issues Projects Releases Wiki Activity

x86/vf_transpose: make ff_transpose_8x8_16_sse2 work on x86_32 

Reviewed-by: Paul B Mahol <onemda@gmail.com>
Signed-off-by: James Almer <jamrial@gmail.com>
This commit is contained in:
James Almer 2019-10-21 17:06:18 -03:00
parent 2a75006ddc
commit 738bc3e742
2 changed files with 6 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
libavfilter/x86/vf_transpose.asm
View File

@ -56,10 +56,7 @@ cglobal transpose_8x8_8, 4,5,8, src, src_linesize, dst, dst_linesize, linesize3
movq [dstq + linesize3q], m7 movq [dstq + linesize3q], m7
RET RET
%if ARCH_X86_64 cglobal transpose_8x8_16, 4,5,9, ARCH_X86_32 * 32, src, src_linesize, dst, dst_linesize, linesize3
INIT_XMM sse2
cglobal transpose_8x8_16, 4,5,9, src, src_linesize, dst, dst_linesize, linesize3
lea linesize3q, [src_linesizeq * 3] lea linesize3q, [src_linesizeq * 3]
movu m0, [srcq + src_linesizeq * 0] movu m0, [srcq + src_linesizeq * 0]
movu m1, [srcq + src_linesizeq * 1] movu m1, [srcq + src_linesizeq * 1]
@ -71,7 +68,11 @@ cglobal transpose_8x8_16, 4,5,9, src, src_linesize, dst, dst_linesize, linesize3
movu m6, [srcq + src_linesizeq * 2] movu m6, [srcq + src_linesizeq * 2]
movu m7, [srcq + linesize3q] movu m7, [srcq + linesize3q]
%if ARCH_X86_64
TRANSPOSE8x8W 0, 1, 2, 3, 4, 5, 6, 7, 8 TRANSPOSE8x8W 0, 1, 2, 3, 4, 5, 6, 7, 8
%else
TRANSPOSE8x8W 0, 1, 2, 3, 4, 5, 6, 7, [rsp], [rsp + 16]
%endif
lea linesize3q, [dst_linesizeq * 3] lea linesize3q, [dst_linesizeq * 3]
movu [dstq + dst_linesizeq * 0], m0 movu [dstq + dst_linesizeq * 0], m0
@ -84,5 +85,3 @@ cglobal transpose_8x8_16, 4,5,9, src, src_linesize, dst, dst_linesize, linesize3
movu [dstq + dst_linesizeq * 2], m6 movu [dstq + dst_linesizeq * 2], m6
movu [dstq + linesize3q], m7 movu [dstq + linesize3q], m7
RET RET
%endif

2
libavfilter/x86/vf_transpose_init.c
View File

@ -43,7 +43,7 @@ av_cold void ff_transpose_init_x86(TransVtable *v, int pixstep)
v->transpose_8x8 = ff_transpose_8x8_8_sse2; v->transpose_8x8 = ff_transpose_8x8_8_sse2;
} }
if (ARCH_X86_64 && EXTERNAL_SSE2(cpu_flags) && pixstep == 2) { if (EXTERNAL_SSE2(cpu_flags) && pixstep == 2) {
v->transpose_8x8 = ff_transpose_8x8_16_sse2; v->transpose_8x8 = ff_transpose_8x8_16_sse2;
} }
} }