[x265] [PATCH] asm: intra_pred_ang8_22

[chen]

code is right, same problem on constant table

At 2015-03-10 20:17:18,praveen at multicorewareinc.com wrote:
># HG changeset patch
># User Praveen Tiwari <praveen at multicorewareinc.com>
># Date 1425989753 -19800
># Node ID 62dcfcba14e59543f92752c8145c612e13af472f
># Parent  fc805c02d0f28815e86e65c2f6d0bfea69db0f7b
>asm: intra_pred_ang8_22
>
>diff -r fc805c02d0f2 -r 62dcfcba14e5 source/common/x86/asm-primitives.cpp
>--- a/source/common/x86/asm-primitives.cpp	Tue Mar 10 14:01:26 2015 +0530
>+++ b/source/common/x86/asm-primitives.cpp	Tue Mar 10 17:45:53 2015 +0530
>@@ -1499,6 +1499,7 @@
>         p.cu[BLOCK_8x8].intra_pred[11] = x265_intra_pred_ang8_11_avx2;
>         p.cu[BLOCK_8x8].intra_pred[12] = x265_intra_pred_ang8_12_avx2;
>         p.cu[BLOCK_8x8].intra_pred[24] = x265_intra_pred_ang8_24_avx2;
>+        p.cu[BLOCK_8x8].intra_pred[22] = x265_intra_pred_ang8_22_avx2;
> 
>         // copy_sp primitives
>         p.cu[BLOCK_16x16].copy_sp = x265_blockcopy_sp_16x16_avx2;
>diff -r fc805c02d0f2 -r 62dcfcba14e5 source/common/x86/intrapred.h
>--- a/source/common/x86/intrapred.h	Tue Mar 10 14:01:26 2015 +0530
>+++ b/source/common/x86/intrapred.h	Tue Mar 10 17:45:53 2015 +0530
>@@ -182,6 +182,7 @@
> void x265_intra_pred_ang8_11_avx2(pixel* dst, intptr_t dstStride, const pixel* srcPix, int dirMode, int bFilter);
> void x265_intra_pred_ang8_12_avx2(pixel* dst, intptr_t dstStride, const pixel* srcPix, int dirMode, int bFilter);
> void x265_intra_pred_ang8_24_avx2(pixel* dst, intptr_t dstStride, const pixel* srcPix, int dirMode, int bFilter);
>+void x265_intra_pred_ang8_22_avx2(pixel* dst, intptr_t dstStride, const pixel* srcPix, int dirMode, int bFilter);
> void x265_all_angs_pred_4x4_sse4(pixel *dest, pixel *refPix, pixel *filtPix, int bLuma);
> void x265_all_angs_pred_8x8_sse4(pixel *dest, pixel *refPix, pixel *filtPix, int bLuma);
> void x265_all_angs_pred_16x16_sse4(pixel *dest, pixel *refPix, pixel *filtPix, int bLuma);
>diff -r fc805c02d0f2 -r 62dcfcba14e5 source/common/x86/intrapred8.asm
>--- a/source/common/x86/intrapred8.asm	Tue Mar 10 14:01:26 2015 +0530
>+++ b/source/common/x86/intrapred8.asm	Tue Mar 10 17:45:53 2015 +0530
>@@ -132,6 +132,11 @@
> %assign x x+1
> %endrep
> 
>+c_ang8_mode_22:       db 13, 19, 13, 19, 13, 19, 13, 19, 13, 19, 13, 19, 13, 19, 13, 19, 26, 6, 26, 6, 26, 6, 26, 6, 26, 6, 26, 6, 26, 6, 26, 6, \
>+                         7, 25, 7, 25, 7, 25, 7, 25, 7, 25, 7, 25, 7, 25, 7, 25, 20, 12, 20, 12, 20, 12, 20, 12, 20, 12, 20, 12, 20, 12, 20, 12,\
>+                         1, 31, 1, 31, 1, 31, 1, 31, 1, 31, 1, 31, 1, 31, 1, 31, 14, 18, 14, 18, 14, 18, 14, 18, 14, 18, 14, 18, 14, 18, 14, 18, \
>+                         27, 5, 27, 5, 27, 5, 27, 5, 27, 5, 27, 5, 27, 5, 27, 5, 8, 24, 8, 24, 8, 24, 8, 24, 8, 24, 8, 24, 8, 24, 8, 24
>+
> SECTION .text
> 
> cextern pw_2
>@@ -10451,3 +10456,58 @@
>     movq              [r0 + 2 * r1], xm2
>     movhps            [r0 + r3], xm2
>     RET
>+
>+INIT_YMM avx2
>+cglobal intra_pred_ang8_22, 3, 5, 6
>+    mova              m3, [pw_1024]
>+    vbroadcasti128    m0, [r2]
>+    pshufb            m0, [intra_pred_shuff_0_8]
>+    lea               r4, [c_ang8_mode_22]
>+
>+    ; Row [0, 1] 
>+    pmaddubsw         m1, m0, [r4]
>+    pmulhrsw          m1, m3
>+
>+    ; Row [2, 3]
>+    vextracti128      xm5, m0, 0
>+    pslldq            xm5, 2
>+    pinsrb            xm5, [r2 + 16 + 2], 0
>+    pinsrb            xm5, [r2 + 0], 1
>+    vinserti128       m0, m5, xm5, 1
>+    pmaddubsw         m2, m0, [r4 + mmsize]
>+    pmulhrsw          m2, m3
>+
>+    ; Row [4, 5]
>+    vextracti128      xm5, m0, 0
>+    pslldq            xm5, 2
>+    pinsrb            xm5, [r2 + 16 + 5], 0
>+    pinsrb            xm5, [r2 + 16 + 2], 1
>+    vinserti128       m0, m5, xm5, 1
>+    pmaddubsw         m4, m0, [r4 + 2 * mmsize]
>+    pmulhrsw          m4, m3
>+
>+    ; Row [6, 7]
>+    vextracti128      xm5, m0, 0
>+    pslldq            xm5, 2
>+    pinsrb            xm5, [r2 + 16 + 7], 0
>+    pinsrb            xm5, [r2 + 16 + 5], 1
>+    vinserti128       m0, m0, xm5, 1
>+    pmaddubsw         m0, [r4 + 3 * mmsize]
>+    pmulhrsw          m0, m3
>+
>+    packuswb          m1, m2
>+    packuswb          m4, m0
>+
>+    lea               r3, [3 * r1]
>+    movq              [r0], xm1
>+    vextracti128      xm2, m1, 1
>+    movq              [r0 + r1], xm2
>+    movhps            [r0 + 2 * r1], xm1
>+    movhps            [r0 + r3], xm2
>+    lea               r0, [r0 + 4 * r1]
>+    movq              [r0], xm4
>+    vextracti128      xm2, m4, 1
>+    movq              [r0 + r1], xm2
>+    movhps            [r0 + 2 * r1], xm4
>+    movhps            [r0 + r3], xm2
>+    RET
>_______________________________________________
>x265-devel mailing list
>x265-devel at videolan.org
>https://mailman.videolan.org/listinfo/x265-devel
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/x265-devel/attachments/20150311/746132f9/attachment-0001.html>