[x265] [PATCH 176 of 307] x86: AVX512 Fix error in chroma_vsp high bit depth
mythreyi at multicorewareinc.com
mythreyi at multicorewareinc.com
Sat Apr 7 04:32:54 CEST 2018
# HG changeset patch
# User Vignesh Vijayakumar<vignesh at multicorewareinc.com>
# Date 1511170651 -19800
# Mon Nov 20 15:07:31 2017 +0530
# Node ID 67e149415f9f8be0d5b7832fde9e02cc592bbf28
# Parent 8bbcc1bd3c1381e936695a6eff30a17cc2633b6f
x86: AVX512 Fix error in chroma_vsp high bit depth
This patch also optimises coeffIdx load into the register
diff -r 8bbcc1bd3c13 -r 67e149415f9f source/common/x86/ipfilter16.asm
--- a/source/common/x86/ipfilter16.asm Mon Nov 06 11:39:56 2017 +0530
+++ b/source/common/x86/ipfilter16.asm Mon Nov 20 15:07:31 2017 +0530
@@ -8589,15 +8589,19 @@
psrad m1, INTERP_SHIFT_SP
psrad m2, INTERP_SHIFT_SP
psrad m3, INTERP_SHIFT_SP
+
+ packssdw m0, m1
+ packssdw m2, m3
+ CLIPW2 m0, m2, m10, m11
%else
psrad m0, 6
psrad m1, 6
psrad m2, 6
psrad m3, 6
-%endif
-
packssdw m0, m1
packssdw m2, m3
+%endif
+
movu [r2], xm0
movu [r2 + r3], xm2
vextracti32x4 [r2 + 2 * r3], m0, 1
@@ -8613,36 +8617,37 @@
; void interp_4tap_vert(int16_t *src, intptr_t srcStride, int16_t *dst, intptr_t dstStride, int coeffIdx)
;-----------------------------------------------------------------------------------------------------------------
%macro CHROMA_VERT_S_8x8_AVX512 1
-%if ARCH_X86_64
INIT_ZMM avx512
-cglobal interp_4tap_vert_%1_8x8, 5, 11, 10
+cglobal interp_4tap_vert_%1_8x8, 5, 11, 12
add r1d, r1d
add r3d, r3d
sub r0, r1
shl r4d, 7
-
%ifdef PIC
lea r5, [tab_ChromaCoeffV_avx512]
- lea r5, [r5 + r4]
+ mova m8, [r5 + r4]
+ mova m9, [r5 + r4 + mmsize]
%else
lea r5, [tab_ChromaCoeffV_avx512 + r4]
-%endif
-
+ mova m8, [r5]
+ mova m9, [r5 + mmsize]
+%endif
%ifidn %1, sp
vbroadcasti32x4 m7, [INTERP_OFFSET_SP]
-%endif
- mova m8, [r5]
- mova m9, [r5 + mmsize]
+ pxor m10, m10
+ vbroadcasti32x8 m11, [pw_pixel_max]
+%endif
lea r10, [3 * r1]
lea r7, [3 * r3]
+
PROCESS_CHROMA_VERT_S_8x8_AVX512 %1
RET
-%endif
-%endmacro
-
-CHROMA_VERT_S_8x8_AVX512 ss
-CHROMA_VERT_S_8x8_AVX512 sp
-
+%endmacro
+
+%if ARCH_X86_64
+ CHROMA_VERT_S_8x8_AVX512 ss
+ CHROMA_VERT_S_8x8_AVX512 sp
+%endif
%macro FILTER_VER_S_CHROMA_8xN_AVX512 2
INIT_ZMM avx512
cglobal interp_4tap_vert_%1_8x%2, 5, 11, 10
@@ -8650,20 +8655,24 @@
add r3d, r3d
sub r0, r1
shl r4d, 7
-
%ifdef PIC
lea r5, [tab_ChromaCoeffV_avx512]
- lea r5, [r5 + r4]
+ mova m8, [r5 + r4]
+ mova m9, [r5 + r4 + mmsize]
%else
lea r5, [tab_ChromaCoeffV_avx512 + r4]
-%endif
+ mova m8, [r5]
+ mova m9, [r5 + mmsize]
+%endif
+
%ifidn %1, sp
vbroadcasti32x4 m7, [INTERP_OFFSET_SP]
-%endif
- mova m8, [r5]
- mova m9, [r5 + mmsize]
+ pxor m10, m10
+ vbroadcasti32x8 m11, [pw_pixel_max]
+%endif
lea r10, [3 * r1]
lea r7, [3 * r3]
+
%rep %2/8 - 1
PROCESS_CHROMA_VERT_S_8x8_AVX512 %1
lea r0, [r8 + 4 * r1]
@@ -8672,16 +8681,14 @@
PROCESS_CHROMA_VERT_S_8x8_AVX512 %1
RET
%endmacro
-
%if ARCH_X86_64
-FILTER_VER_S_CHROMA_8xN_AVX512 ss, 16
-FILTER_VER_S_CHROMA_8xN_AVX512 ss, 32
-FILTER_VER_S_CHROMA_8xN_AVX512 ss, 64
-FILTER_VER_S_CHROMA_8xN_AVX512 sp, 16
-FILTER_VER_S_CHROMA_8xN_AVX512 sp, 32
-FILTER_VER_S_CHROMA_8xN_AVX512 sp, 64
-%endif
-
+ FILTER_VER_S_CHROMA_8xN_AVX512 ss, 16
+ FILTER_VER_S_CHROMA_8xN_AVX512 ss, 32
+ FILTER_VER_S_CHROMA_8xN_AVX512 ss, 64
+ FILTER_VER_S_CHROMA_8xN_AVX512 sp, 16
+ FILTER_VER_S_CHROMA_8xN_AVX512 sp, 32
+ FILTER_VER_S_CHROMA_8xN_AVX512 sp, 64
+%endif
%macro PROCESS_CHROMA_VERT_S_16x4_AVX512 1
movu ym1, [r0]
lea r6, [r0 + 2 * r1]
@@ -8728,15 +8735,19 @@
psrad m1, INTERP_SHIFT_SP
psrad m2, INTERP_SHIFT_SP
psrad m3, INTERP_SHIFT_SP
+
+ packssdw m0, m1
+ packssdw m2, m3
+ CLIPW2 m0, m2, m10, m11
%else
psrad m0, 6
psrad m1, 6
psrad m2, 6
psrad m3, 6
-%endif
-
packssdw m0, m1
packssdw m2, m3
+%endif
+
movu [r2], ym0
movu [r2 + r3], ym2
vextracti32x8 [r2 + 2 * r3], m0, 1
@@ -8747,55 +8758,58 @@
; void interp_4tap_vert(int16_t *src, intptr_t srcStride, int16_t *dst, intptr_t dstStride, int coeffIdx)
;-----------------------------------------------------------------------------------------------------------------
%macro CHROMA_VERT_S_16x4_AVX512 1
-%if ARCH_X86_64
INIT_ZMM avx512
-cglobal interp_4tap_vert_%1_16x4, 5, 9, 10
+cglobal interp_4tap_vert_%1_16x4, 5, 9, 12
add r1d, r1d
add r3d, r3d
sub r0, r1
shl r4d, 7
-
%ifdef PIC
lea r5, [tab_ChromaCoeffV_avx512]
- lea r5, [r5 + r4]
+ mova m8, [r5 + r4]
+ mova m9, [r5 + r4 + mmsize]
%else
lea r5, [tab_ChromaCoeffV_avx512 + r4]
-%endif
-
mova m8, [r5]
mova m9, [r5 + mmsize]
+%endif
+
%ifidn %1, sp
vbroadcasti32x4 m7, [INTERP_OFFSET_SP]
+ pxor m10, m10
+ vbroadcasti32x8 m11, [pw_pixel_max]
%endif
lea r7, [3 * r3]
lea r8, [3 * r1]
PROCESS_CHROMA_VERT_S_16x4_AVX512 %1
RET
-%endif
-%endmacro
-
-CHROMA_VERT_S_16x4_AVX512 ss
-CHROMA_VERT_S_16x4_AVX512 sp
-
+%endmacro
+
+%if ARCH_X86_64
+ CHROMA_VERT_S_16x4_AVX512 ss
+ CHROMA_VERT_S_16x4_AVX512 sp
+%endif
%macro FILTER_VER_S_CHROMA_16xN_AVX512 2
INIT_ZMM avx512
-cglobal interp_4tap_vert_%1_16x%2, 5, 9, 10
+cglobal interp_4tap_vert_%1_16x%2, 5, 9, 12
add r1d, r1d
add r3d, r3d
sub r0, r1
shl r4d, 7
-
%ifdef PIC
lea r5, [tab_ChromaCoeffV_avx512]
- lea r5, [r5 + r4]
+ mova m8, [r5 + r4]
+ mova m9, [r5 + r4 + mmsize]
%else
lea r5, [tab_ChromaCoeffV_avx512 + r4]
-%endif
-
mova m8, [r5]
mova m9, [r5 + mmsize]
+%endif
+
%ifidn %1, sp
vbroadcasti32x4 m7, [INTERP_OFFSET_SP]
+ pxor m10, m10
+ vbroadcasti32x8 m11, [pw_pixel_max]
%endif
lea r7, [3 * r3]
lea r8, [3 * r1]
@@ -8911,6 +8925,13 @@
psrad m10, INTERP_SHIFT_SP
psrad m11, INTERP_SHIFT_SP
psrad m12, INTERP_SHIFT_SP
+
+ packssdw m0, m1
+ packssdw m2, m3
+ packssdw m9, m10
+ packssdw m11, m12
+ CLIPW2 m0, m2, m18, m19
+ CLIPW2 m9, m11, m18, m19
%else
psrad m0, 6
psrad m1, 6
@@ -8920,11 +8941,13 @@
psrad m10, 6
psrad m11, 6
psrad m12, 6
-%endif
+
packssdw m0, m1
packssdw m2, m3
packssdw m9, m10
packssdw m11, m12
+%endif
+
movu [r2], ym0
movu [r2 + r3], ym2
vextracti32x8 [r2 + 2 * r3], m0, 1
@@ -8984,18 +9007,25 @@
paddd m1, m7
paddd m2, m7
paddd m3, m7
+
psrad m0, INTERP_SHIFT_SP
psrad m1, INTERP_SHIFT_SP
psrad m2, INTERP_SHIFT_SP
psrad m3, INTERP_SHIFT_SP
+
+ packssdw m0, m1
+ packssdw m2, m3
+ CLIPW2 m0, m2, m18, m19
%else
psrad m0, 6
psrad m1, 6
psrad m2, 6
psrad m3, 6
-%endif
+
packssdw m0, m1
packssdw m2, m3
+%endif
+
movu [r2 + mmsize/2], xm0
movu [r2 + r3 + mmsize/2], xm2
vextracti32x4 [r2 + 2 * r3 + mmsize/2], m0, 1
@@ -9006,27 +9036,27 @@
vextracti32x4 [r2 + 2 * r3 + mmsize/2], m0, 3
vextracti32x4 [r2 + r7 + mmsize/2], m2, 3
%endmacro
-
%macro FILTER_VER_S_CHROMA_24xN_AVX512 2
INIT_ZMM avx512
-cglobal interp_4tap_vert_%1_24x%2, 5, 12, 18
+cglobal interp_4tap_vert_%1_24x%2, 5, 12, 20
add r1d, r1d
add r3d, r3d
sub r0, r1
shl r4d, 7
-
%ifdef PIC
lea r5, [tab_ChromaCoeffV_avx512]
- lea r5, [r5 + r4]
+ mova m16, [r5 + r4]
+ mova m17, [r5 + r4 + mmsize]
%else
lea r5, [tab_ChromaCoeffV_avx512 + r4]
-%endif
-
+ mova m16, [r5]
+ mova m17, [r5 + mmsize]
+%endif
%ifidn %1, sp
vbroadcasti32x4 m7, [INTERP_OFFSET_SP]
-%endif
- mova m16, [r5]
- mova m17, [r5 + mmsize]
+ pxor m18, m18
+ vbroadcasti32x8 m19, [pw_pixel_max]
+%endif
lea r10, [3 * r1]
lea r7, [3 * r3]
%rep %2/8 - 1
@@ -9037,113 +9067,162 @@
PROCESS_CHROMA_VERT_S_24x8_AVX512 %1
RET
%endmacro
-
%if ARCH_X86_64
- FILTER_VER_S_CHROMA_24xN_AVX512 ss,32
- FILTER_VER_S_CHROMA_24xN_AVX512 ss,64
- FILTER_VER_S_CHROMA_24xN_AVX512 sp,32
- FILTER_VER_S_CHROMA_24xN_AVX512 sp,64
-%endif
-
-%macro PROCESS_CHROMA_VERT_S_32x2_AVX512 1
+ FILTER_VER_S_CHROMA_24xN_AVX512 ss, 32
+ FILTER_VER_S_CHROMA_24xN_AVX512 ss, 64
+ FILTER_VER_S_CHROMA_24xN_AVX512 sp, 32
+ FILTER_VER_S_CHROMA_24xN_AVX512 sp, 64
+%endif
+
+%macro PROCESS_CHROMA_VERT_S_32x4_AVX512 1
movu m1, [r0]
+ lea r6, [r0 + 2 * r1]
+ movu m10, [r6]
movu m3, [r0 + r1]
+ movu m12, [r6 + r1]
punpcklwd m0, m1, m3
- pmaddwd m0, m9
+ punpcklwd m9, m10, m12
+ pmaddwd m0, m16
+ pmaddwd m9, m16
punpckhwd m1, m3
- pmaddwd m1, m9
-
+ punpckhwd m10, m12
+ pmaddwd m1, m16
+ pmaddwd m10, m16
movu m4, [r0 + 2 * r1]
+ movu m13, [r6 + 2 * r1]
punpcklwd m2, m3, m4
- pmaddwd m2, m9
+ punpcklwd m11, m12, m13
+ pmaddwd m2, m16
+ pmaddwd m11, m16
punpckhwd m3, m4
- pmaddwd m3, m9
-
- lea r0, [r0 + 2 * r1]
- movu m5, [r0 + r1]
+ punpckhwd m12, m13
+ pmaddwd m3, m16
+ pmaddwd m12, m16
+
+ movu m5, [r0 + r7]
+ movu m14, [r6 + r7]
punpcklwd m6, m4, m5
- pmaddwd m6, m10
+ punpcklwd m15, m13, m14
+ pmaddwd m6, m17
+ pmaddwd m15, m17
paddd m0, m6
+ paddd m9, m15
punpckhwd m4, m5
- pmaddwd m4, m10
+ punpckhwd m13, m14
+ pmaddwd m4, m17
+ pmaddwd m13, m17
paddd m1, m4
-
- movu m4, [r0 + 2 * r1]
+ paddd m10, m13
+
+ movu m4, [r0 + 4 * r1]
+ movu m13, [r6 + 4 * r1]
punpcklwd m6, m5, m4
- pmaddwd m6, m10
+ punpcklwd m15, m14, m13
+ pmaddwd m6, m17
+ pmaddwd m15, m17
paddd m2, m6
+ paddd m11, m15
punpckhwd m5, m4
- pmaddwd m5, m10
+ punpckhwd m14, m13
+ pmaddwd m5, m17
+ pmaddwd m14, m17
paddd m3, m5
-
+ paddd m12, m14
%ifidn %1,sp
paddd m0, m7
paddd m1, m7
paddd m2, m7
paddd m3, m7
+ paddd m9, m7
+ paddd m10, m7
+ paddd m11, m7
+ paddd m12, m7
+
psrad m0, INTERP_SHIFT_SP
psrad m1, INTERP_SHIFT_SP
psrad m2, INTERP_SHIFT_SP
psrad m3, INTERP_SHIFT_SP
+ psrad m9, INTERP_SHIFT_SP
+ psrad m10, INTERP_SHIFT_SP
+ psrad m11, INTERP_SHIFT_SP
+ psrad m12, INTERP_SHIFT_SP
+
+ packssdw m0, m1
+ packssdw m2, m3
+ packssdw m9, m10
+ packssdw m11, m12
+ CLIPW2 m0, m2, m18, m19
+ CLIPW2 m9, m11, m18, m19
%else
psrad m0, 6
psrad m1, 6
psrad m2, 6
psrad m3, 6
-%endif
+ psrad m9, 6
+ psrad m10, 6
+ psrad m11, 6
+ psrad m12, 6
+
packssdw m0, m1
packssdw m2, m3
+ packssdw m9, m10
+ packssdw m11, m12
+%endif
+
movu [r2], m0
movu [r2 + r3], m2
-%endmacro
-
+ movu [r2 + 2 * r3], m9
+ movu [r2 + r8], m11
+%endmacro
;-----------------------------------------------------------------------------------------------------------------
; void interp_4tap_vert(int16_t *src, intptr_t srcStride, int16_t *dst, intptr_t dstStride, int coeffIdx)
;-----------------------------------------------------------------------------------------------------------------
%macro FILTER_VER_S_CHROMA_32xN_AVX512 2
INIT_ZMM avx512
-cglobal interp_4tap_vert_%1_32x%2, 5, 7, 11
+cglobal interp_4tap_vert_%1_32x%2, 5, 9, 20
add r1d, r1d
add r3d, r3d
sub r0, r1
shl r4d, 7
-
%ifdef PIC
lea r5, [tab_ChromaCoeffV_avx512]
- lea r5, [r5 + r4]
+ mova m16, [r5 + r4]
+ mova m17, [r5 + r4 + mmsize]
%else
lea r5, [tab_ChromaCoeffV_avx512 + r4]
-%endif
-
+ mova m16, [r5]
+ mova m17, [r5 + mmsize]
+%endif
+ lea r7, [3 * r1]
+ lea r8, [3 * r3]
%ifidn %1, sp
vbroadcasti32x4 m7, [INTERP_OFFSET_SP]
-%endif
- mova m9, [r5]
- mova m10, [r5 + mmsize]
-
-%rep %2/2 - 1
- PROCESS_CHROMA_VERT_S_32x2_AVX512 %1
- lea r2, [r2 + 2 * r3]
+ pxor m18, m18
+ vbroadcasti32x8 m19, [pw_pixel_max]
+%endif
+
+%rep %2/4 - 1
+ PROCESS_CHROMA_VERT_S_32x4_AVX512 %1
+ lea r0, [r0 + 4 * r1]
+ lea r2, [r2 + 4 * r3]
%endrep
- PROCESS_CHROMA_VERT_S_32x2_AVX512 %1
- RET
-%endmacro
-
+ PROCESS_CHROMA_VERT_S_32x4_AVX512 %1
+ RET
+%endmacro
%if ARCH_X86_64
- FILTER_VER_S_CHROMA_32xN_AVX512 ss,8
- FILTER_VER_S_CHROMA_32xN_AVX512 ss,16
- FILTER_VER_S_CHROMA_32xN_AVX512 ss,24
- FILTER_VER_S_CHROMA_32xN_AVX512 ss,32
- FILTER_VER_S_CHROMA_32xN_AVX512 ss,48
- FILTER_VER_S_CHROMA_32xN_AVX512 ss,64
- FILTER_VER_S_CHROMA_32xN_AVX512 sp,8
- FILTER_VER_S_CHROMA_32xN_AVX512 sp,16
- FILTER_VER_S_CHROMA_32xN_AVX512 sp,24
- FILTER_VER_S_CHROMA_32xN_AVX512 sp,32
- FILTER_VER_S_CHROMA_32xN_AVX512 sp,48
- FILTER_VER_S_CHROMA_32xN_AVX512 sp,64
-%endif
-
+ FILTER_VER_S_CHROMA_32xN_AVX512 ss, 8
+ FILTER_VER_S_CHROMA_32xN_AVX512 ss, 16
+ FILTER_VER_S_CHROMA_32xN_AVX512 ss, 24
+ FILTER_VER_S_CHROMA_32xN_AVX512 ss, 32
+ FILTER_VER_S_CHROMA_32xN_AVX512 ss, 48
+ FILTER_VER_S_CHROMA_32xN_AVX512 ss, 64
+ FILTER_VER_S_CHROMA_32xN_AVX512 sp, 8
+ FILTER_VER_S_CHROMA_32xN_AVX512 sp, 16
+ FILTER_VER_S_CHROMA_32xN_AVX512 sp, 24
+ FILTER_VER_S_CHROMA_32xN_AVX512 sp, 32
+ FILTER_VER_S_CHROMA_32xN_AVX512 sp, 48
+ FILTER_VER_S_CHROMA_32xN_AVX512 sp, 64
+%endif
%macro PROCESS_CHROMA_VERT_S_48x4_AVX512 1
movu m1, [r0]
lea r6, [r0 + 2 * r1]
@@ -9218,6 +9297,13 @@
psrad m10, INTERP_SHIFT_SP
psrad m11, INTERP_SHIFT_SP
psrad m12, INTERP_SHIFT_SP
+
+ packssdw m0, m1
+ packssdw m2, m3
+ packssdw m9, m10
+ packssdw m11, m12
+ CLIPW2 m0, m2, m18, m19
+ CLIPW2 m9, m11, m18, m19
%else
psrad m0, 6
psrad m1, 6
@@ -9227,12 +9313,12 @@
psrad m10, 6
psrad m11, 6
psrad m12, 6
-%endif
-
packssdw m0, m1
packssdw m2, m3
packssdw m9, m10
packssdw m11, m12
+%endif
+
movu [r2], m0
movu [r2 + r3], m2
movu [r2 + 2 * r3], m9
@@ -9282,44 +9368,46 @@
psrad m1, INTERP_SHIFT_SP
psrad m2, INTERP_SHIFT_SP
psrad m3, INTERP_SHIFT_SP
+ packssdw m0, m1
+ packssdw m2, m3
+ CLIPW2 m0, m2, m18, m19
%else
psrad m0, 6
psrad m1, 6
psrad m2, 6
psrad m3, 6
-%endif
-
packssdw m0, m1
packssdw m2, m3
+%endif
+
movu [r2 + mmsize], ym0
movu [r2 + r3 + mmsize], ym2
vextracti32x8 [r2 + 2 * r3 + mmsize], m0, 1
vextracti32x8 [r2 + r8 + mmsize], m2, 1
%endmacro
-
%macro CHROMA_VERT_S_48x4_AVX512 1
-%if ARCH_X86_64
INIT_ZMM avx512
-cglobal interp_4tap_vert_%1_48x64, 5, 9, 18
+cglobal interp_4tap_vert_%1_48x64, 5, 9, 20
add r1d, r1d
add r3d, r3d
sub r0, r1
shl r4d, 7
%ifdef PIC
lea r5, [tab_ChromaCoeffV_avx512]
- lea r5, [r5 + r4]
+ mova m16, [r5 + r4]
+ mova m17, [r5 + r4 + mmsize]
%else
lea r5, [tab_ChromaCoeffV_avx512 + r4]
+ mova m16, [r5]
+ mova m17, [r5 + mmsize]
%endif
lea r7, [3 * r1]
lea r8, [3 * r3]
-
%ifidn %1, sp
vbroadcasti32x4 m7, [INTERP_OFFSET_SP]
-%endif
- mova m16, [r5]
- mova m17, [r5 + mmsize]
-
+ pxor m18, m18
+ vbroadcasti32x8 m19, [pw_pixel_max]
+%endif
%rep 15
PROCESS_CHROMA_VERT_S_48x4_AVX512 %1
lea r0, [r0 + 4 * r1]
@@ -9327,12 +9415,12 @@
%endrep
PROCESS_CHROMA_VERT_S_48x4_AVX512 %1
RET
-%endif
-%endmacro
-
-CHROMA_VERT_S_48x4_AVX512 sp
-CHROMA_VERT_S_48x4_AVX512 ss
-
+%endmacro
+
+%if ARCH_X86_64
+ CHROMA_VERT_S_48x4_AVX512 sp
+ CHROMA_VERT_S_48x4_AVX512 ss
+%endif
%macro PROCESS_CHROMA_VERT_S_64x2_AVX512 1
movu m1, [r0]
movu m3, [r0 + r1]
@@ -9347,13 +9435,11 @@
pmaddwd m8, m15
punpckhwd m9, m11
pmaddwd m9, m15
-
movu m4, [r0 + 2 * r1]
punpcklwd m2, m3, m4
- pmaddwd m2, [r5]
+ pmaddwd m2, m15
punpckhwd m3, m4
- pmaddwd m3, [r5]
-
+ pmaddwd m3, m15
movu m12, [r0 + 2 * r1 + mmsize]
punpcklwd m10, m11, m12
pmaddwd m10, m15
@@ -9411,6 +9497,13 @@
psrad m9, INTERP_SHIFT_SP
psrad m10, INTERP_SHIFT_SP
psrad m11, INTERP_SHIFT_SP
+
+ packssdw m0, m1
+ packssdw m2, m3
+ packssdw m8, m9
+ packssdw m10, m11
+ CLIPW2 m0, m2, m17, m18
+ CLIPW2 m8, m10, m17, m18
%else
psrad m0, 6
psrad m1, 6
@@ -9420,11 +9513,13 @@
psrad m9, 6
psrad m10, 6
psrad m11, 6
-%endif
+
packssdw m0, m1
packssdw m2, m3
packssdw m8, m9
packssdw m10, m11
+%endif
+
movu [r2], m0
movu [r2 + r3], m2
movu [r2 + mmsize], m8
@@ -9435,25 +9530,25 @@
;-----------------------------------------------------------------------------------------------------------------
%macro FILTER_VER_S_CHROMA_64xN_AVX512 2
INIT_ZMM avx512
-cglobal interp_4tap_vert_%1_64x%2, 5, 7, 17
+cglobal interp_4tap_vert_%1_64x%2, 5, 7, 19
add r1d, r1d
add r3d, r3d
sub r0, r1
shl r4d, 7
-
%ifdef PIC
lea r5, [tab_ChromaCoeffV_avx512]
- lea r5, [r5 + r4]
+ mova m15, [r5 + r4]
+ mova m16, [r5 + r4 + mmsize]
%else
lea r5, [tab_ChromaCoeffV_avx512 + r4]
-%endif
-
+ mova m15, [r5]
+ mova m16, [r5 + mmsize]
+%endif
%ifidn %1, sp
vbroadcasti32x4 m7, [INTERP_OFFSET_SP]
-%endif
- mova m15, [r5]
- mova m16, [r5 + mmsize]
-
+ pxor m17, m17
+ vbroadcasti32x8 m18, [pw_pixel_max]
+%endif
%rep %2/2 - 1
PROCESS_CHROMA_VERT_S_64x2_AVX512 %1
lea r2, [r2 + 2 * r3]
More information about the x265-devel
mailing list