[x265] [PATCH] asm: chroma_hpp[64x64, 64x48, 64x32, 64x16] for i444 - improved 22990c->14176c, 17897c->10791c, 12050c->7186c, 5655c->3266c

[aasaipriya at multicorewareinc.com]

# HG changeset patch
# User Aasaipriya Chandran <aasaipriya at multicorewareinc.com>
# Date 1429246698 -19800
#      Fri Apr 17 10:28:18 2015 +0530
# Node ID d8ea45299885b4e00cec173569d06b5dd359a5e8
# Parent  b49bc22d93e527bc0279f5cf3149b7652f3cc871
asm: chroma_hpp[64x64, 64x48, 64x32, 64x16] for i444 - improved 22990c->14176c, 17897c->10791c, 12050c->7186c, 5655c->3266c

diff -r b49bc22d93e5 -r d8ea45299885 source/common/x86/asm-primitives.cpp
--- a/source/common/x86/asm-primitives.cpp	Thu Apr 16 15:28:19 2015 +0530
+++ b/source/common/x86/asm-primitives.cpp	Fri Apr 17 10:28:18 2015 +0530
@@ -2212,6 +2212,11 @@
         
         p.chroma[X265_CSP_I422].pu[CHROMA_422_2x8].filter_hpp = x265_interp_4tap_horiz_pp_2x8_avx2;
 
+        p.chroma[X265_CSP_I444].pu[LUMA_64x64].filter_hpp = x265_interp_4tap_horiz_pp_64x64_avx2;
+        p.chroma[X265_CSP_I444].pu[LUMA_64x32].filter_hpp = x265_interp_4tap_horiz_pp_64x32_avx2;
+        p.chroma[X265_CSP_I444].pu[LUMA_64x48].filter_hpp = x265_interp_4tap_horiz_pp_64x48_avx2;
+        p.chroma[X265_CSP_I444].pu[LUMA_64x16].filter_hpp = x265_interp_4tap_horiz_pp_64x16_avx2;
+
         //i444 chroma_hps
         p.chroma[X265_CSP_I444].pu[LUMA_64x32].filter_hps = x265_interp_4tap_horiz_ps_64x32_avx2;
         p.chroma[X265_CSP_I444].pu[LUMA_64x48].filter_hps = x265_interp_4tap_horiz_ps_64x48_avx2;
diff -r b49bc22d93e5 -r d8ea45299885 source/common/x86/ipfilter8.asm
--- a/source/common/x86/ipfilter8.asm	Thu Apr 16 15:28:19 2015 +0530
+++ b/source/common/x86/ipfilter8.asm	Fri Apr 17 10:28:18 2015 +0530
@@ -22098,3 +22098,93 @@
     pextrw            [r2 + r3 * 2], xm1,         6
     pextrw            [r2 + r4],     xm1,         7
     RET
+
+;-------------------------------------------------------------------------------------------------------------
+; void interp_4tap_horiz_pp_64xN(pixel *src, intptr_t srcStride, int16_t *dst, intptr_t dstStride, int coeffIdx
+;-------------------------------------------------------------------------------------------------------------
+%macro IPFILTER_CHROMA_PP_64xN_AVX2 1
+INIT_YMM avx2
+cglobal interp_4tap_horiz_pp_64x%1, 4,6,7
+    mov             r4d, r4m
+
+%ifdef PIC
+    lea               r5,           [tab_ChromaCoeff]
+    vpbroadcastd      m0,           [r5 + r4 * 4]
+%else
+    vpbroadcastd      m0,           [tab_ChromaCoeff + r4 * 4]
+%endif
+
+    mova              m1,           [interp4_horiz_shuf1]
+    vpbroadcastd      m2,           [pw_1]
+    mova              m6,           [pw_512]
+    ; register map
+    ; m0 - interpolate coeff
+    ; m1 - shuffle order table
+    ; m2 - constant word 1
+
+    dec               r0
+    mov               r4d,          %1
+
+.loop:
+    ; Row 0
+    vbroadcasti128    m3,           [r0]                        ; [x x x x x A 9 8 7 6 5 4 3 2 1 0]
+    pshufb            m3,           m1
+    pmaddubsw         m3,           m0
+    pmaddwd           m3,           m2
+    vbroadcasti128    m4,           [r0 + 4]
+    pshufb            m4,           m1
+    pmaddubsw         m4,           m0
+    pmaddwd           m4,           m2
+    packssdw          m3,           m4
+    pmulhrsw          m3,           m6
+
+    vbroadcasti128    m4,           [r0 + 16]
+    pshufb            m4,           m1
+    pmaddubsw         m4,           m0
+    pmaddwd           m4,           m2
+    vbroadcasti128    m5,           [r0 + 20]
+    pshufb            m5,           m1
+    pmaddubsw         m5,           m0
+    pmaddwd           m5,           m2
+    packssdw          m4,           m5
+    pmulhrsw          m4,           m6
+    packuswb          m3,           m4
+    vpermq            m3,           m3,      11011000b
+    movu              [r2],         m3
+
+    vbroadcasti128    m3,           [r0 + 32]                        ; [x x x x x A 9 8 7 6 5 4 3 2 1 0]
+    pshufb            m3,           m1
+    pmaddubsw         m3,           m0
+    pmaddwd           m3,           m2
+    vbroadcasti128    m4,           [r0 + 36]
+    pshufb            m4,           m1
+    pmaddubsw         m4,           m0
+    pmaddwd           m4,           m2
+    packssdw          m3,           m4
+    pmulhrsw          m3,           m6
+
+    vbroadcasti128    m4,           [r0 + 48]
+    pshufb            m4,           m1
+    pmaddubsw         m4,           m0
+    pmaddwd           m4,           m2
+    vbroadcasti128    m5,           [r0 + 52]
+    pshufb            m5,           m1
+    pmaddubsw         m5,           m0
+    pmaddwd           m5,           m2
+    packssdw          m4,           m5
+    pmulhrsw          m4,           m6
+    packuswb          m3,           m4
+    vpermq            m3,           m3,      11011000b
+    movu              [r2 + 32],         m3
+
+    add               r2,           r3
+    add               r0,           r1
+    dec               r4d
+    jnz               .loop
+    RET
+%endmacro
+
+IPFILTER_CHROMA_PP_64xN_AVX2  64
+IPFILTER_CHROMA_PP_64xN_AVX2  32
+IPFILTER_CHROMA_PP_64xN_AVX2  48
+IPFILTER_CHROMA_PP_64xN_AVX2  16