# HG changeset patch
# User Praveen Tiwari <prav...@multicorewareinc.com>
# Date 1425990471 -19800
# Node ID c5215ffa12fee77b89c98888edf9faa369bacd41
# Parent 62dcfcba14e59543f92752c8145c612e13af472f
asm: intra_pred_ang8_14
diff -r 62dcfcba14e5 -r c5215ffa12fe source/common/x86/asm-primitives.cpp
--- a/source/common/x86/asm-primitives.cpp Tue Mar 10 17:45:53 2015 +0530
+++ b/source/common/x86/asm-primitives.cpp Tue Mar 10 17:57:51 2015 +0530
@@ -1500,6 +1500,7 @@
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;
+ p.cu[BLOCK_8x8].intra_pred[14] = x265_intra_pred_ang8_14_avx2;
// copy_sp primitives
p.cu[BLOCK_16x16].copy_sp = x265_blockcopy_sp_16x16_avx2;
diff -r 62dcfcba14e5 -r c5215ffa12fe source/common/x86/intrapred.h
--- a/source/common/x86/intrapred.h Tue Mar 10 17:45:53 2015 +0530
+++ b/source/common/x86/intrapred.h Tue Mar 10 17:57:51 2015 +0530
@@ -183,6 +183,7 @@
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_intra_pred_ang8_14_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 62dcfcba14e5 -r c5215ffa12fe source/common/x86/intrapred8.asm
--- a/source/common/x86/intrapred8.asm Tue Mar 10 17:45:53 2015 +0530
+++ b/source/common/x86/intrapred8.asm Tue Mar 10 17:57:51 2015 +0530
@@ -10511,3 +10511,70 @@
movhps [r0 + 2 * r1], xm4
movhps [r0 + r3], xm2
RET
+
+INIT_YMM avx2
+cglobal intra_pred_ang8_14, 3, 5, 6
+ mova m3, [pw_1024]
+ movu xm1, [r2 + 16]
+ pinsrb xm1, [r2], 0
+ pshufb xm1, [intra_pred_shuff_0_8]
+ vinserti128 m0, m1, xm1, 1
+
+ ; Row [0, 1]
+ lea r4, [c_ang8_mode_22]
+ pmaddubsw m1, m0, [r4]
+ pmulhrsw m1, m3
+
+ ; Row [2, 3]
+ vextracti128 xm5, m0, 0
+ pslldq xm5, 2
+ pinsrb xm5, [r2 + 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 + 5], 0
+ pinsrb xm5, [r2 + 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 + 7], 0
+ pinsrb xm5, [r2 + 5], 1
+ vinserti128 m0, m0, xm5, 1
+ pmaddubsw m0, [r4 + 3 * mmsize]
+ pmulhrsw m0, m3
+
+ packuswb m1, m2
+ packuswb m4, m0
+
+ vperm2i128 m2, m1, m4, 00100000b
+ vperm2i128 m1, m1, m4, 00110001b
+ punpcklbw m4, m2, m1
+ punpckhbw m2, m1
+ punpcklwd m1, m4, m2
+ punpckhwd m4, m2
+ mova m0, [trans8_shuf]
+ vpermd m1, m0, m1
+ vpermd m4, m0, m4
+
+ lea r3, [3 * r1]
+ movq [r0], xm1
+ movhps [r0 + r1], xm1
+ vextracti128 xm2, m1, 1
+ movq [r0 + 2 * r1], xm2
+ movhps [r0 + r3], xm2
+ lea r0, [r0 + 4 * r1]
+ movq [r0], xm4
+ movhps [r0 + r1], xm4
+ vextracti128 xm2, m4, 1
+ movq [r0 + 2 * r1], xm2
+ movhps [r0 + r3], xm2
+ RET
_______________________________________________
x265-devel mailing list
x265-devel@videolan.org
https://mailman.videolan.org/listinfo/x265-devel
