RepoMirrors/ffmpeg
1
0
Fork 0
mirror of https://git.ffmpeg.org/ffmpeg.git synced 2024-12-28 10:22:10 +00:00
Code Issues Projects Releases Wiki Activity

lavc/vc1dsp: fuse multiply-adds in R-V V inv_trans_4

Browse Source 
T-Head C908 (cycles):            before   after
vc1dsp.vc1_inv_trans_4x4_rvv_i32: 128.0   120.0
vc1dsp.vc1_inv_trans_4x8_rvv_i32: 244.0   240.0
vc1dsp.vc1_inv_trans_8x4_rvv_i32: 239.2   235.2
This commit is contained in:
Rémi Denis-Courmont 2024-06-30 14:15:33 +03:00
parent d69e522523
commit 78e1565f84
1 changed files with 4 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
libavcodec/riscv/vc1dsp_rvv.S
View File

@ -194,14 +194,12 @@ func ff_vc1_inv_trans_4_rvv, zve32x
li t4, 22 li t4, 22
vmul.vx v10, v2, t3 vmul.vx v10, v2, t3
li t2, 10 li t2, 10
vmul.vx v14, v1, t4 vmul.vx v26, v1, t4
vmul.vx v27, v3, t4
vadd.vv v24, v8, v10 # t1 vadd.vv v24, v8, v10 # t1
vsub.vv v25, v8, v10 # t2 vsub.vv v25, v8, v10 # t2
vmul.vx v16, v3, t2 vmacc.vx v26, t2, v3 # t3
vmul.vx v18, v3, t4 vnmsac.vx v27, t2, v1 # t4
vmul.vx v20, v1, t2
vadd.vv v26, v14, v16 # t3
vsub.vv v27, v18, v20 # t4
vwadd.vv v8, v24, v26 vwadd.vv v8, v24, v26
vwsub.vv v10, v25, v27 vwsub.vv v10, v25, v27
vwadd.vv v12, v25, v27 vwadd.vv v12, v25, v27