[FFmpeg-devel] [PATCH 4/6] avcodec/x86/hevc_mc: add qpel_h4_8_avx512icl

[jianhua.wu at intel.com]

From: Wu Jianhua <jianhua.wu at intel.com>

ff_hevc_put_hevc_qpel_h4_8_sse4       993694
ff_hevc_put_hevc_qpel_h4_8_avx512icl  686647

Signed-off-by: Wu Jianhua <jianhua.wu at intel.com>
---
 libavcodec/x86/hevc_mc.asm    | 12 ++++++++++--
 libavcodec/x86/hevcdsp.h      |  1 +
 libavcodec/x86/hevcdsp_init.c |  1 +
 3 files changed, 12 insertions(+), 2 deletions(-)

diff --git a/libavcodec/x86/hevc_mc.asm b/libavcodec/x86/hevc_mc.asm
index 8c128f5202..25880b8858 100644
--- a/libavcodec/x86/hevc_mc.asm
+++ b/libavcodec/x86/hevc_mc.asm
@@ -87,6 +87,7 @@ QPEL_TABLE 12, 4, w, sse4
 QPEL_TABLE  8,16, b, avx2
 QPEL_TABLE 10, 8, w, avx2
 
+QPEL_TABLE  4, 1, b, avx512icl_h
 QPEL_TABLE  8, 1, b, avx512icl_h
 QPEL_TABLE  8, 1, d, avx512icl_v
 QPEL_TABLE 16, 1, b, avx512icl_h
@@ -1734,7 +1735,7 @@ HEVC_PUT_HEVC_QPEL_HV 16, 10
 
 ; required: m0-m5
 ; %1: dst register index
-; %2: name for src
+; %2: src
 %macro QPEL_H_LOAD_COMPUTE 2
     pxor            m%1, m%1
 %if mmsize == 64
@@ -1754,9 +1755,13 @@ cglobal hevc_put_hevc_qpel_h%1_%2, 5, 6, 8, dst, src, srcstride, height, mx, tmp
     QPEL_LOAD_SHUF   2, 3
 .loop:
     QPEL_H_LOAD_COMPUTE   6, srcq - 3
-%if %1 == 8
+%if %1 < 16
     vpmovdw             xm6, m6
+%if %1 == 4
+    movq             [dstq], xm6
+%else
     movu             [dstq], xm6
+%endif
 %else
     vpmovdw             ym6, m6
     movu             [dstq], ym6
@@ -1829,6 +1834,9 @@ cglobal hevc_put_hevc_qpel_hv%1_%2, 6, 7, 8, dst, src, srcstride, height, mx, my
 %if ARCH_X86_64
 %if HAVE_AVX512ICL_EXTERNAL
 
+INIT_XMM avx512icl
+HEVC_PUT_HEVC_QPEL_AVX512ICL 4, 8
+
 INIT_YMM avx512icl
 HEVC_PUT_HEVC_QPEL_AVX512ICL 8, 8
 HEVC_PUT_HEVC_QPEL_HV_AVX512ICL 8, 8
diff --git a/libavcodec/x86/hevcdsp.h b/libavcodec/x86/hevcdsp.h
index 6e3fc01ad0..51ffdc9628 100644
--- a/libavcodec/x86/hevcdsp.h
+++ b/libavcodec/x86/hevcdsp.h
@@ -233,6 +233,7 @@ WEIGHTING_PROTOTYPES(8, sse4);
 WEIGHTING_PROTOTYPES(10, sse4);
 WEIGHTING_PROTOTYPES(12, sse4);
 
+void ff_hevc_put_hevc_qpel_h4_8_avx512icl(int16_t *dst, uint8_t *_src, ptrdiff_t _srcstride, int height, intptr_t mx, intptr_t my, int width);
 void ff_hevc_put_hevc_qpel_h8_8_avx512icl(int16_t *dst, uint8_t *_src, ptrdiff_t _srcstride, int height, intptr_t mx, intptr_t my, int width);
 void ff_hevc_put_hevc_qpel_h16_8_avx512icl(int16_t *dst, uint8_t *_src, ptrdiff_t _srcstride, int height, intptr_t mx, intptr_t my, int width);
 void ff_hevc_put_hevc_qpel_hv8_8_avx512icl(int16_t *dst, uint8_t *_src, ptrdiff_t _srcstride, int height, intptr_t mx, intptr_t my, int width);
diff --git a/libavcodec/x86/hevcdsp_init.c b/libavcodec/x86/hevcdsp_init.c
index 4023faa654..be1484d06e 100644
--- a/libavcodec/x86/hevcdsp_init.c
+++ b/libavcodec/x86/hevcdsp_init.c
@@ -879,6 +879,7 @@ void ff_hevc_dsp_init_x86(HEVCDSPContext *c, const int bit_depth)
             c->add_residual[3] = ff_hevc_add_residual_32_8_avx2;
         }
         if (EXTERNAL_AVX512ICL(cpu_flags)) {
+            c->put_hevc_qpel[1][0][1] = ff_hevc_put_hevc_qpel_h4_8_avx512icl;
             c->put_hevc_qpel[3][0][1] = ff_hevc_put_hevc_qpel_h8_8_avx512icl;
             c->put_hevc_qpel[5][0][1] = ff_hevc_put_hevc_qpel_h16_8_avx512icl;
             c->put_hevc_qpel[3][1][1] = ff_hevc_put_hevc_qpel_hv8_8_avx512icl;
-- 
2.17.1