eessppeelllloo/ffmpeg
1
0
Fork 0
mirror of https://git.ffmpeg.org/ffmpeg.git synced 2025-12-18 21:10:10 +01:00
Code Issues Packages Projects Releases Wiki Activity

lavc/vp8dsp: avoid one multiplication on RISC-V

Browse Source 
Use shifts rather than multiply, and save one instruction.
This commit is contained in:
Rémi Denis-Courmont
2024-05-25 13:41:04 +03:00
parent 0aad5b9bf5
commit 4e56455d36
2 changed files with 29 additions and 28 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
libavcodec/riscv/vp8dsp_init.c
View File

@@ -84,6 +84,7 @@ av_cold void ff_vp78dsp_init_riscv(VP8DSPContext *c)
c->put_vp8_bilinear_pixels_tab[2][2][1] = ff_put_vp8_bilin4_hv_rvv; c->put_vp8_bilinear_pixels_tab[2][2][1] = ff_put_vp8_bilin4_hv_rvv;
c->put_vp8_bilinear_pixels_tab[2][2][2] = ff_put_vp8_bilin4_hv_rvv; c->put_vp8_bilinear_pixels_tab[2][2][2] = ff_put_vp8_bilin4_hv_rvv;
if (flags & AV_CPU_FLAG_RVB_ADDR) {
c->put_vp8_epel_pixels_tab[0][0][2] = ff_put_vp8_epel16_h6_rvv; c->put_vp8_epel_pixels_tab[0][0][2] = ff_put_vp8_epel16_h6_rvv;
c->put_vp8_epel_pixels_tab[1][0][2] = ff_put_vp8_epel8_h6_rvv; c->put_vp8_epel_pixels_tab[1][0][2] = ff_put_vp8_epel8_h6_rvv;
c->put_vp8_epel_pixels_tab[2][0][2] = ff_put_vp8_epel4_h6_rvv; c->put_vp8_epel_pixels_tab[2][0][2] = ff_put_vp8_epel4_h6_rvv;
@@ -112,6 +113,7 @@ av_cold void ff_vp78dsp_init_riscv(VP8DSPContext *c)
c->put_vp8_epel_pixels_tab[2][1][2] = ff_put_vp8_epel4_h6v4_rvv; c->put_vp8_epel_pixels_tab[2][1][2] = ff_put_vp8_epel4_h6v4_rvv;
#endif #endif
} }
}
#endif #endif
#endif #endif
} }

7
libavcodec/riscv/vp8dsp_rvv.S
View File

@@ -182,15 +182,14 @@ const subpel_filters
endconst endconst
.macro epel_filter size type regtype .macro epel_filter size type regtype
lla \regtype\()2, subpel_filters
.ifc \type,v .ifc \type,v
addi \regtype\()0, a6, -1 addi \regtype\()0, a6, -1
.else .else
addi \regtype\()0, a5, -1 addi \regtype\()0, a5, -1
.endif .endif
li \regtype\()1, 6 lla \regtype\()2, subpel_filters
mul \regtype\()0, \regtype\()0, \regtype\()1 sh1add \regtype\()0, \regtype\()0, \regtype\()0
add \regtype\()0, \regtype\()0, \regtype\()2 sh1add \regtype\()0, \regtype\()0, \regtype\()2
.irp n,1,2,3,4 .irp n,1,2,3,4
lb \regtype\n, \n(\regtype\()0) lb \regtype\n, \n(\regtype\()0)
.endr .endr