Commit 6ecc3fd6 authored by Mickaël Raulet's avatar Mickaël Raulet Committed by Michael Niedermayer

x86/hevc_mc: use aligned loads

Signed-off-by: 's avatarMichael Niedermayer <michaelni@gmx.at>
parent 383fddee
...@@ -771,7 +771,7 @@ typedef struct HEVCLocalContext { ...@@ -771,7 +771,7 @@ typedef struct HEVCLocalContext {
DECLARE_ALIGNED(32, uint8_t, edge_emu_buffer)[(MAX_PB_SIZE + 7) * EDGE_EMU_BUFFER_STRIDE * 2]; DECLARE_ALIGNED(32, uint8_t, edge_emu_buffer)[(MAX_PB_SIZE + 7) * EDGE_EMU_BUFFER_STRIDE * 2];
/* The extended size between the new edge emu buffer is abused by SAO */ /* The extended size between the new edge emu buffer is abused by SAO */
DECLARE_ALIGNED(32, uint8_t, edge_emu_buffer2)[(MAX_PB_SIZE + 7) * EDGE_EMU_BUFFER_STRIDE * 2]; DECLARE_ALIGNED(32, uint8_t, edge_emu_buffer2)[(MAX_PB_SIZE + 7) * EDGE_EMU_BUFFER_STRIDE * 2];
DECLARE_ALIGNED(16, int16_t, tmp [MAX_PB_SIZE * MAX_PB_SIZE]); DECLARE_ALIGNED(32, int16_t, tmp [MAX_PB_SIZE * MAX_PB_SIZE]);
int ct_depth; int ct_depth;
CodingUnit cu; CodingUnit cu;
......
...@@ -103,14 +103,14 @@ QPEL_TABLE 10, 8, w, avx2 ...@@ -103,14 +103,14 @@ QPEL_TABLE 10, 8, w, avx2
%endif ;avx %endif ;avx
%elif %1 <= 16 %elif %1 <= 16
%if cpuflag(avx2) %if cpuflag(avx2)
movu %3, [%2] mova %3, [%2]
%else %else
movdqa %3, [%2] ; load data from source2 movdqa %3, [%2] ; load data from source2
movdqa %4, [%2+16] ; load data from source2 movdqa %4, [%2+16] ; load data from source2
%endif ; avx %endif ; avx
%else ; %1 = 32 %else ; %1 = 32
movu %3, [%2] mova %3, [%2]
movu %4, [%2+32] mova %4, [%2+32]
%endif %endif
%endmacro %endmacro
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment