diff options
| author | James Almer <jamrial@gmail.com> | 2015-02-06 13:17:19 -0300 |
|---|---|---|
| committer | James Almer <jamrial@gmail.com> | 2015-02-06 15:06:15 -0300 |
| commit | aea29a891ff1b5cca104656bc1eb8ab0e55fbb16 (patch) | |
| tree | 9bd4eea8791c6e1c1210d2666c3419c585881fcf | |
| parent | bcb09251152551bb4328b8724b692b0c4f274487 (diff) | |
| download | ffmpeg-aea29a891ff1b5cca104656bc1eb8ab0e55fbb16.tar.gz | |
x86/hevc_sao: fix loading of RIP address
pb_eo must be handled as a rip relative address for MSVC64, so an
intermediate register is needed. Should fix link failures.
Suggested by Hendrik Leppkes and Christophe Gisquet.
Tested-By: Hendrik Leppkes <h.leppkes@gmail.com>
Reviewed-by: Christophe Gisquet <christophe.gisquet@gmail.com>
Signed-off-by: James Almer <jamrial@gmail.com>
| -rw-r--r-- | libavcodec/x86/hevc_sao.asm | 42 |
1 files changed, 9 insertions, 33 deletions
diff --git a/libavcodec/x86/hevc_sao.asm b/libavcodec/x86/hevc_sao.asm index 605896704b..f4eca0c065 100644 --- a/libavcodec/x86/hevc_sao.asm +++ b/libavcodec/x86/hevc_sao.asm @@ -293,24 +293,14 @@ HEVC_SAO_BAND_FILTER_16 12, 64, 2 ;void ff_hevc_sao_edge_filter_<width>_8_<opt>(uint8_t *_dst, uint8_t *_src, ptrdiff_t stride_dst, int16_t *sao_offset_val, ; int eo, int width, int height); %macro HEVC_SAO_EDGE_FILTER_8 2-3 +%if ARCH_X86_64 +cglobal hevc_sao_edge_filter_%1_8, 4, 9, 8, dst, src, dststride, offset, eo, a_stride, b_stride, height, tmp +%define tmp2q heightq %if WIN64 -cglobal hevc_sao_edge_filter_%1_8, 4, 8, 8, dst, src, dststride, offset, a_stride, b_stride, height, tmp -%define eoq heightq movsxd eoq, dword r4m - movsx a_strideq, byte [pb_eo+eoq*4+1] - movsx b_strideq, byte [pb_eo+eoq*4+3] - imul a_strideq, EDGE_SRCSTRIDE - imul b_strideq, EDGE_SRCSTRIDE - movsx tmpq, byte [pb_eo+eoq*4] - add a_strideq, tmpq - movsx tmpq, byte [pb_eo+eoq*4+2] - add b_strideq, tmpq - mov heightd, r6m - -%elif ARCH_X86_64 -cglobal hevc_sao_edge_filter_%1_8, 5, 9, 8, dst, src, dststride, offset, eo, a_stride, b_stride, height, tmp -%define tmp2q heightq +%else movsxd eoq, eod +%endif lea tmp2q, [pb_eo] movsx a_strideq, byte [tmp2q+eoq*4+1] movsx b_strideq, byte [tmp2q+eoq*4+3] @@ -439,26 +429,13 @@ INIT_YMM cpuname ;void ff_hevc_sao_edge_filter_<width>_<depth>_<opt>(uint8_t *_dst, uint8_t *_src, ptrdiff_t stride_dst, int16_t *sao_offset_val, ; int eo, int width, int height); %macro HEVC_SAO_EDGE_FILTER_16 3 +cglobal hevc_sao_edge_filter_%2_%1, 4, 9, 16, dst, src, dststride, offset, eo, a_stride, b_stride, height, tmp +%define tmp2q heightq %if WIN64 -cglobal hevc_sao_edge_filter_%2_%1, 4, 8, 16, dst, src, dststride, offset, a_stride, b_stride, height, tmp -%define eoq heightq movsxd eoq, dword r4m - movsx a_strideq, byte [pb_eo+eoq*4+1] - movsx b_strideq, byte [pb_eo+eoq*4+3] - imul a_strideq, EDGE_SRCSTRIDE>>1 - imul b_strideq, EDGE_SRCSTRIDE>>1 - movsx tmpq, byte [pb_eo+eoq*4] - add a_strideq, tmpq - movsx tmpq, byte [pb_eo+eoq*4+2] - add b_strideq, tmpq - mov heightd, r6m - add a_strideq, a_strideq - add b_strideq, b_strideq - -%else ; UNIX64 -cglobal hevc_sao_edge_filter_%2_%1, 5, 9, 16, dst, src, dststride, offset, eo, a_stride, b_stride, height, tmp -%define tmp2q heightq +%else movsxd eoq, eod +%endif lea tmp2q, [pb_eo] movsx a_strideq, byte [tmp2q+eoq*4+1] movsx b_strideq, byte [tmp2q+eoq*4+3] @@ -471,7 +448,6 @@ cglobal hevc_sao_edge_filter_%2_%1, 5, 9, 16, dst, src, dststride, offset, eo, a mov heightd, r6m add a_strideq, a_strideq add b_strideq, b_strideq -%endif ; ARCH %if cpuflag(avx2) SPLATW m8, [offsetq+2] |