changeset 9579:2cac9b74c41c

asm-avx2: intra_pred_ang8_31
author Praveen Tiwari <praveen@multicorewareinc.com>
date Thu, 26 Feb 2015 12:46:54 +0530
parents 4257bfeb9b87
children f19bc989e8bb
files source/common/x86/asm-primitives.cpp source/common/x86/intrapred.h source/common/x86/intrapred8.asm
diffstat 3 files changed, 43 insertions(+-), 1 deletions(-) [+]
line wrap: on
line diff
--- a/source/common/x86/asm-primitives.cpp	Wed Feb 25 20:17:05 2015 +0530
+++ b/source/common/x86/asm-primitives.cpp	Thu Feb 26 12:46:54 2015 +0530
@@ -1843,6 +1843,7 @@ void setupAssemblyPrimitives(EncoderPrim
         p.cu[BLOCK_8x8].intra_pred[33] = x265_intra_pred_ang8_33_avx2;
         p.cu[BLOCK_8x8].intra_pred[4] = x265_intra_pred_ang8_4_avx2;
         p.cu[BLOCK_8x8].intra_pred[32] = x265_intra_pred_ang8_32_avx2;
+        p.cu[BLOCK_8x8].intra_pred[31] = x265_intra_pred_ang8_31_avx2;
     }
 }
 #endif // if HIGH_BIT_DEPTH
--- a/source/common/x86/intrapred.h	Wed Feb 25 20:17:05 2015 +0530
+++ b/source/common/x86/intrapred.h	Thu Feb 26 12:46:54 2015 +0530
@@ -162,6 +162,7 @@ void x265_intra_pred_ang8_3_avx2(pixel* 
 void x265_intra_pred_ang8_33_avx2(pixel* dst, intptr_t dstStride, const pixel* srcPix, int dirMode, int bFilter);
 void x265_intra_pred_ang8_4_avx2(pixel* dst, intptr_t dstStride, const pixel* srcPix, int dirMode, int bFilter);
 void x265_intra_pred_ang8_32_avx2(pixel* dst, intptr_t dstStride, const pixel* srcPix, int dirMode, int bFilter);
+void x265_intra_pred_ang8_31_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);
--- a/source/common/x86/intrapred8.asm	Wed Feb 25 20:17:05 2015 +0530
+++ b/source/common/x86/intrapred8.asm	Thu Feb 26 12:46:54 2015 +0530
@@ -71,6 +71,12 @@ c_ang8_9_30:          db 23, 9, 23, 9, 2
 c_ang8_src5_13_6_14:  db 4, 5, 5, 6, 6, 7, 7, 8, 8, 9, 9, 10, 10, 11, 11, 12, 5, 6, 6, 7, 7, 8, 8, 9, 9, 10, 10, 11, 11, 12, 12, 13
 c_ang8_19_8:          db 13, 19, 13, 19, 13, 19, 13, 19, 13, 19, 13, 19, 13, 19, 13, 19, 24, 8, 24, 8, 24, 8, 24, 8, 24, 8, 24, 8, 24, 8, 24, 8
 
+c_ang8_17_2:          db 15, 17, 15, 17, 15, 17, 15, 17, 15, 17, 15, 17, 15, 17, 15, 17, 30, 2, 30, 2, 30, 2, 30, 2, 30, 2, 30, 2, 30, 2, 30, 2
+c_ang8_19_4:          db 13, 19, 13, 19, 13, 19, 13, 19, 13, 19, 13, 19, 13, 19, 13, 19, 28, 4, 28, 4, 28, 4, 28, 4, 28, 4, 28, 4, 28, 4, 28, 4
+c_ang8_21_6:          db 11, 21, 11, 21, 11, 21, 11, 21, 11, 21, 11, 21, 11, 21, 11, 21, 26, 6, 26, 6, 26, 6, 26, 6, 26, 6, 26, 6, 26, 6, 26, 6
+c_ang8_23_8:          db 9, 23, 9, 23, 9, 23, 9, 23, 9, 23, 9, 23, 9, 23, 9, 23, 24, 8, 24, 8, 24, 8, 24, 8, 24, 8, 24, 8, 24, 8, 24, 8,
+c_ang8_src4_12_5_13:  db 3, 4, 4, 5, 5, 6, 6, 7, 7, 8, 8, 9, 9, 10, 10, 11, 4, 5, 5, 6, 6, 7, 7, 8, 8, 9, 9, 10, 10, 11, 11, 12
+
 ;; (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
@@ -32205,7 +32211,6 @@ cglobal intra_pred_ang8_33, 3,4,5
     movhps            [r0 + r3], xm2
     RET
 
-
 INIT_YMM avx2
 cglobal intra_pred_ang8_4, 3,4,5
     movu              m3, [pw_1024]
@@ -32285,3 +32290,38 @@ cglobal intra_pred_ang8_32, 3,4,5
     movhps            [r0 + 2 * r1], xm4
     movhps            [r0 + r3], xm2
     RET
+
+INIT_YMM avx2
+cglobal intra_pred_ang8_31, 3, 4, 5
+    movu              m3, [pw_1024]
+    vbroadcasti128    m0, [r2 + 1]
+
+    pshufb            m1, m0, [c_ang8_src1_9_2_10]
+    pshufb            m2, m0, [c_ang8_src2_10_3_11]
+    pshufb            m4, m0, [c_ang8_src3_11_4_12]
+    pshufb            m0,     [c_ang8_src4_12_5_13]
+
+    pmaddubsw         m1, [c_ang8_17_2]
+    pmulhrsw          m1, m3
+    pmaddubsw         m2, [c_ang8_19_4]
+    pmulhrsw          m2, m3
+    pmaddubsw         m4, [c_ang8_21_6]
+    pmulhrsw          m4, m3
+    pmaddubsw         m0, [c_ang8_23_8]
+    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