lineage-rockchip/android_external_ffmpeg
1
0
Fork 0
Code Activity Actions

aarch64/vvc: Add put_epel_hv

Browse source 
On Apple M1:

put_chroma_hv_8_4x4_c:                                   1.7 ( 1.00x)
put_chroma_hv_8_4x4_neon:                                0.2 ( 7.67x)
put_chroma_hv_8_8x8_c:                                   5.5 ( 1.00x)
put_chroma_hv_8_8x8_neon:                                0.5 (11.53x)
put_chroma_hv_8_16x16_c:                                18.5 ( 1.00x)
put_chroma_hv_8_16x16_neon:                              1.5 (12.53x)
put_chroma_hv_8_32x32_c:                                72.5 ( 1.00x)
put_chroma_hv_8_32x32_neon:                              4.7 (15.34x)
put_chroma_hv_8_64x64_c:                               274.0 ( 1.00x)
put_chroma_hv_8_64x64_neon:                             18.5 (14.83x)
put_chroma_hv_8_128x128_c:                            1058.7 ( 1.00x)
put_chroma_hv_8_128x128_neon:                           75.2 (14.07x)

On Android Pixel 8 Pro:

put_chroma_hv_8_4x4_c:                                   1.2 ( 1.00x)
put_chroma_hv_8_4x4_neon:                                0.0 ( 0.00x)
put_chroma_hv_8_4x4_i8mm:                                0.2 ( 5.00x)
put_chroma_hv_8_8x8_c:                                   4.0 ( 1.00x)
put_chroma_hv_8_8x8_neon:                                0.5 ( 8.00x)
put_chroma_hv_8_8x8_i8mm:                                0.5 ( 8.00x)
put_chroma_hv_8_16x16_c:                                15.2 ( 1.00x)
put_chroma_hv_8_16x16_neon:                              2.5 ( 6.10x)
put_chroma_hv_8_16x16_i8mm:                              2.2 ( 6.78x)
put_chroma_hv_8_32x32_c:                                61.0 ( 1.00x)
put_chroma_hv_8_32x32_neon:                              9.8 ( 6.26x)
put_chroma_hv_8_32x32_i8mm:                              8.5 ( 7.18x)
put_chroma_hv_8_64x64_c:                               229.5 ( 1.00x)
put_chroma_hv_8_64x64_neon:                             38.5 ( 5.96x)
put_chroma_hv_8_64x64_i8mm:                             34.0 ( 6.75x)
put_chroma_hv_8_128x128_c:                             919.8 ( 1.00x)
put_chroma_hv_8_128x128_neon:                          154.5 ( 5.95x)
put_chroma_hv_8_128x128_i8mm:                          140.0 ( 6.57x)
This commit is contained in:
Zhao Zhili 2024-09-11 17:13:38 +08:00 • committed by Nuo Mi
parent 0dcf204e5d
commit 1be5a2374f
3 changed files with 147 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

14
libavcodec/aarch64/vvc/dsp_init.c
View file

@ -84,6 +84,13 @@ void ff_vvc_dsp_init_aarch64(VVCDSPContext *const c, const int bd)
c->inter.put[1][5][0][1] =
c->inter.put[1][6][0][1] = ff_vvc_put_epel_h32_8_neon;
c->inter.put[1][1][1][1] = ff_vvc_put_epel_hv4_8_neon;
c->inter.put[1][2][1][1] = ff_vvc_put_epel_hv8_8_neon;
c->inter.put[1][3][1][1] = ff_vvc_put_epel_hv16_8_neon;
c->inter.put[1][4][1][1] = ff_vvc_put_epel_hv32_8_neon;
c->inter.put[1][5][1][1] = ff_vvc_put_epel_hv64_8_neon;
c->inter.put[1][6][1][1] = ff_vvc_put_epel_hv128_8_neon;
c->inter.put_uni[0][1][0][0] = ff_vvc_put_pel_uni_pixels4_8_neon;
c->inter.put_uni[0][2][0][0] = ff_vvc_put_pel_uni_pixels8_8_neon;
c->inter.put_uni[0][3][0][0] = ff_vvc_put_pel_uni_pixels16_8_neon;
@ -134,6 +141,13 @@ void ff_vvc_dsp_init_aarch64(VVCDSPContext *const c, const int bd)
c->inter.put[1][4][0][1] = ff_vvc_put_epel_h32_8_neon_i8mm;
c->inter.put[1][5][0][1] = ff_vvc_put_epel_h64_8_neon_i8mm;
c->inter.put[1][6][0][1] = ff_vvc_put_epel_h128_8_neon_i8mm;
c->inter.put[1][1][1][1] = ff_vvc_put_epel_hv4_8_neon_i8mm;
c->inter.put[1][2][1][1] = ff_vvc_put_epel_hv8_8_neon_i8mm;
c->inter.put[1][3][1][1] = ff_vvc_put_epel_hv16_8_neon_i8mm;
c->inter.put[1][4][1][1] = ff_vvc_put_epel_hv32_8_neon_i8mm;
c->inter.put[1][5][1][1] = ff_vvc_put_epel_hv64_8_neon_i8mm;
c->inter.put[1][6][1][1] = ff_vvc_put_epel_hv128_8_neon_i8mm;
}
} else if (bd == 10) {
c->alf.filter[LUMA] = alf_filter_luma_10_neon;