[x265] [PATCH] asm: intra_pred_ang8_24 8bpp, improved 206.33c -> 177.70c over SSE version

chen chenm003 at 163.com
Wed Mar 11 01:39:14 CET 2015


 

At 2015-03-10 15:15:48,praveen at multicorewareinc.com wrote:
># HG changeset patch
># User Praveen Tiwari <praveen at multicorewareinc.com>
># Date 1425971668 -19800
># Node ID f79f44d2e55fa9ddb88955a5ffa136efa550ca8f
># Parent  810995b991eba3f7dcd9014db3b58a6b07723be3
>asm: intra_pred_ang8_24 8bpp, improved 206.33c -> 177.70c over SSE version
>
>intra_ang_8x8[24]       11.25x   177.70          1999.00
>
>diff -r 810995b991eb -r f79f44d2e55f source/common/x86/asm-primitives.cpp
>--- a/source/common/x86/asm-primitives.cpp	Tue Mar 10 11:27:29 2015 +0530
>+++ b/source/common/x86/asm-primitives.cpp	Tue Mar 10 12:44:28 2015 +0530
>@@ -1497,6 +1497,7 @@
>         p.cu[BLOCK_8x8].intra_pred[27] = x265_intra_pred_ang8_27_avx2;
>         p.cu[BLOCK_8x8].intra_pred[25] = x265_intra_pred_ang8_25_avx2;
>         p.cu[BLOCK_8x8].intra_pred[11] = x265_intra_pred_ang8_11_avx2;
>+        p.cu[BLOCK_8x8].intra_pred[24] = x265_intra_pred_ang8_24_avx2;
> 
>         // copy_sp primitives
>         p.cu[BLOCK_16x16].copy_sp = x265_blockcopy_sp_16x16_avx2;
>diff -r 810995b991eb -r f79f44d2e55f source/common/x86/intrapred.h
>--- a/source/common/x86/intrapred.h	Tue Mar 10 11:27:29 2015 +0530
>+++ b/source/common/x86/intrapred.h	Tue Mar 10 12:44:28 2015 +0530
>@@ -180,6 +180,7 @@
> void x265_intra_pred_ang8_27_avx2(pixel* dst, intptr_t dstStride, const pixel* srcPix, int dirMode, int bFilter);
> void x265_intra_pred_ang8_25_avx2(pixel* dst, intptr_t dstStride, const pixel* srcPix, int dirMode, int bFilter);
> void x265_intra_pred_ang8_11_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_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 810995b991eb -r f79f44d2e55f source/common/x86/intrapred8.asm
>--- a/source/common/x86/intrapred8.asm	Tue Mar 10 11:27:29 2015 +0530
>+++ b/source/common/x86/intrapred8.asm	Tue Mar 10 12:44:28 2015 +0530
>@@ -108,6 +108,11 @@
>                          10, 22, 10, 22, 10, 22, 10, 22, 10, 22, 10, 22, 10, 22, 10, 22, 12, 20, 12, 20, 12, 20, 12, 20, 12, 20, 12, 20, 12, 20, 12, 20, \
>                          14, 18, 14, 18, 14, 18, 14, 18, 14, 18, 14, 18, 14, 18, 14, 18, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16
> 
>+c_ang8_mode_24:       db 5, 27, 5, 27, 5, 27, 5, 27, 5, 27, 5, 27, 5, 27, 5, 27, 10, 22, 10, 22, 10, 22, 10, 22, 10, 22, 10, 22, 10, 22, 10, 22, \
>+                         15, 17, 15, 17, 15, 17, 15, 17, 15, 17, 15, 17, 15, 17, 15, 17, 20, 12, 20, 12, 20, 12, 20, 12, 20, 12, 20, 12, 20, 12, 20, 12, \

we'd better a new 'db' in every line.
other is fine
 
>+                         25, 7, 25, 7, 25, 7, 25, 7, 25, 7, 25, 7, 25, 7, 25, 7, 30, 2, 30, 2, 30, 2, 30, 2, 30, 2, 30, 2, 30, 2, 30, 2, \
>+                         3, 29, 3, 29, 3, 29, 3, 29, 3, 29, 3, 29, 3, 29, 3, 29, 8, 24, 8, 24, 8, 24, 8, 24, 8, 24, 8, 24, 8, 24, 8, 24
>+
> ;; (blkSize - 1 - x)
> pw_planar4_0:         dw 3,  2,  1,  0,  3,  2,  1,  0
> pw_planar4_1:         dw 3,  3,  3,  3,  3,  3,  3,  3
>@@ -10361,3 +10366,40 @@
>     movq              [r0 + 2 * r1], xm2
>     movhps            [r0 + r3], xm2
>     RET
>+
>+INIT_YMM avx2
>+cglobal intra_pred_ang8_24, 3, 5, 5
>+    mova              m3, [pw_1024]
>+    vbroadcasti128    m0, [r2]
>+
>+    pshufb            m0, [intra_pred_shuff_0_8]
>+
>+    lea               r4, [c_ang8_mode_24]
>+    pmaddubsw         m1, m0, [r4]
>+    pmulhrsw          m1, m3
>+    pmaddubsw         m2, m0, [r4 + mmsize]
>+    pmulhrsw          m2, m3
>+    pmaddubsw         m4, m0, [r4 + 2 * mmsize]
>+    pmulhrsw          m4, m3
>+    pslldq            xm0, 2
>+    pinsrb            xm0, [r2 + 16 + 6], 0
>+    pinsrb            xm0, [r2 + 0], 1
>+    vinserti128       m0, m0, xm0, 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/dc6af26e/attachment-0001.html>


More information about the x265-devel mailing list