[x265] [PATCH] asm: filter_vsp, filter_vss for 2x4 in avx2

Divya Manivannan divya at multicorewareinc.com
Wed Apr 29 06:50:35 CEST 2015


# HG changeset patch
# User Divya Manivannan <divya at multicorewareinc.com>
# Date 1430281993 -19800
#      Wed Apr 29 10:03:13 2015 +0530
# Node ID 861c8a143802759b1c960289058bb3165481f6cd
# Parent  c4d9ee2cef03ef74f5623784d514ffcdf725bec4
asm: filter_vsp, filter_vss for 2x4 in avx2

diff -r c4d9ee2cef03 -r 861c8a143802 source/common/ipfilter.cpp
--- a/source/common/ipfilter.cpp	Tue Apr 28 14:34:45 2015 -0500
+++ b/source/common/ipfilter.cpp	Wed Apr 29 10:03:13 2015 +0530
@@ -463,6 +463,7 @@
 
     CHROMA_422(4, 8);
     CHROMA_422(4, 4);
+    CHROMA_422(2, 4);
     CHROMA_422(2, 8);
     CHROMA_422(8,  16);
     CHROMA_422(8,  8);
diff -r c4d9ee2cef03 -r 861c8a143802 source/common/x86/asm-primitives.cpp
--- a/source/common/x86/asm-primitives.cpp	Tue Apr 28 14:34:45 2015 -0500
+++ b/source/common/x86/asm-primitives.cpp	Wed Apr 29 10:03:13 2015 +0530
@@ -2291,6 +2291,7 @@
         p.chroma[X265_CSP_I422].pu[CHROMA_422_16x24].filter_vss = x265_interp_4tap_vert_ss_16x24_avx2;
         p.chroma[X265_CSP_I422].pu[CHROMA_422_12x32].filter_vss = x265_interp_4tap_vert_ss_12x32_avx2;
         p.chroma[X265_CSP_I422].pu[CHROMA_422_4x32].filter_vss = x265_interp_4tap_vert_ss_4x32_avx2;
+        p.chroma[X265_CSP_I422].pu[CHROMA_422_2x4].filter_vss = x265_interp_4tap_vert_ss_2x4_avx2;
 
         //i444 for chroma_vss
         p.chroma[X265_CSP_I444].pu[LUMA_4x4].filter_vss = x265_interp_4tap_vert_ss_4x4_avx2;
@@ -2481,6 +2482,7 @@
         p.chroma[X265_CSP_I422].pu[CHROMA_422_16x24].filter_vsp = x265_interp_4tap_vert_sp_16x24_avx2;
         p.chroma[X265_CSP_I422].pu[CHROMA_422_12x32].filter_vsp = x265_interp_4tap_vert_sp_12x32_avx2;
         p.chroma[X265_CSP_I422].pu[CHROMA_422_4x32].filter_vsp = x265_interp_4tap_vert_sp_4x32_avx2;
+        p.chroma[X265_CSP_I422].pu[CHROMA_422_2x4].filter_vsp = x265_interp_4tap_vert_sp_2x4_avx2;
 
         //i444 for chroma_vsp
         p.chroma[X265_CSP_I444].pu[LUMA_4x4].filter_vsp = x265_interp_4tap_vert_sp_4x4_avx2;
diff -r c4d9ee2cef03 -r 861c8a143802 source/common/x86/ipfilter8.h
--- a/source/common/x86/ipfilter8.h	Tue Apr 28 14:34:45 2015 -0500
+++ b/source/common/x86/ipfilter8.h	Wed Apr 29 10:03:13 2015 +0530
@@ -739,6 +739,8 @@
 CHROMA_422_P2S_FILTERS_SSE4(_sse4);
 CHROMA_422_P2S_FILTERS_SSSE3(_ssse3);
 CHROMA_422_P2S_FILTERS_AVX2(_avx2);
+void x265_interp_4tap_vert_ss_2x4_avx2(const int16_t* src, intptr_t srcStride, int16_t* dst, intptr_t dstStride, int coeffIdx);
+void x265_interp_4tap_vert_sp_2x4_avx2(const int16_t* src, intptr_t srcStride, pixel* dst, intptr_t dstStride, int coeffIdx);
 
 CHROMA_444_FILTERS(_sse4);
 CHROMA_444_SP_FILTERS(_sse4);


More information about the x265-devel mailing list