This increases the group multiplier as per T-Head C910 benchmarks: alac_append_extra_bits_mono_c: 803.0 alac_append_extra_bits_stereo_c: 1604.2 alac_decorrelate_stereo_c: 1077.5 LMUL=1 alac_append_extra_bits_mono_rvv_i32: 418.2 alac_append_extra_bits_stereo_rvv_i32: 693.2 alac_decorrelate_stereo_rvv_i32: 673.5 LMUL=2 alac_append_extra_bits_mono_rvv_i32: 382.2 alac_append_extra_bits_stereo_rvv_i32: 648.2 alac_decorrelate_stereo_rvv_i32: 542.7 LMUL=4 alac_append_extra_bits_mono_rvv_i32: 241.5 alac_append_extra_bits_stereo_rvv_i32: 512.7 alac_decorrelate_stereo_rvv_i32: 364.2 LMUL=8 alac_append_extra_bits_mono_rvv_i32: 239.7 alac_append_extra_bits_stereo_rvv_i32: 497.2 alac_decorrelate_stereo_rvv_i32: 426.7
Unified · Split
| @@ -25,7 +25,7 @@ func ff_alac_decorrelate_stereo_rvv, zve32x · 24 unmodified lines | |||
| 25 | ld a4, 8(a0) | 25 | ld a4, 8(a0) |
| 26 | ld a0, 0(a0) | 26 | ld a0, 0(a0) |
| 27 | 1: | 27 | 1: |
| 28 | vsetvli t0, a1, e32, m1, ta, ma | 28 | vsetvli t0, a1, e32, m4, ta, ma |
| 29 | vle32.v v24, (a4) | 29 | vle32.v v24, (a4) |
| 30 | sub a1, a1, t0 | 30 | sub a1, a1, t0 |
| 31 | vle32.v v16, (a0) | 31 | vle32.v v16, (a0) |
| @@ -47,7 +47,7 @@ func ff_alac_append_extra_bits_mono_rvv, zve32x · 15 unmodified lines | |||
| 47 | ld a0, (a0) | 47 | ld a0, (a0) |
| 48 | ld a1, (a1) | 48 | ld a1, (a1) |
| 49 | 1: | 49 | 1: |
| 50 | vsetvli t0, a4, e32, m1, ta, ma | 50 | vsetvli t0, a4, e32, m8, ta, ma |
| 51 | vle32.v v16, (a0) | 51 | vle32.v v16, (a0) |
| 52 | sub a4, a4, t0 | 52 | sub a4, a4, t0 |
| 53 | vle32.v v24, (a1) | 53 | vle32.v v24, (a1) |
| @@ -67,7 +67,7 @@ func ff_alac_append_extra_bits_stereo_rvv, zve32x · 13 unmodified lines | |||
| 67 | ld a7, 8(a1) | 67 | ld a7, 8(a1) |
| 68 | ld a1, (a1) | 68 | ld a1, (a1) |
| 69 | 1: | 69 | 1: |
| 70 | vsetvli t0, a4, e32, m1, ta, ma | 70 | vsetvli t0, a4, e32, m8, ta, ma |
| 71 | vle32.v v16, (a0) | 71 | vle32.v v16, (a0) |
| 72 | sub a4, a4, t0 | 72 | sub a4, a4, t0 |
| 73 | vle32.v v0, (a6) | 73 | vle32.v v0, (a6) |