Showing the changes to libavcodec (1 of 1 file). Show all
Unified · Split
| @@ -23,7 +23,6 @@ · 22 unmodified lines | |||
| 23 | func ff_put_pixels_clamped_rvv, zve32x | 23 | func ff_put_pixels_clamped_rvv, zve32x |
| 24 | vsetivli zero, 8, e16, m1, ta, ma | 24 | vsetivli zero, 8, e16, m1, ta, ma |
| 25 | vlseg8e16.v v24, (a0) | 25 | vlseg8e16.v v24, (a0) |
| 26 | 1: | ||
| 27 | /* RVV only has signed-signed and unsigned-unsigned clipping. | 26 | /* RVV only has signed-signed and unsigned-unsigned clipping. |
| 28 | * We need two steps for signed-to-unsigned clipping. */ | 27 | * We need two steps for signed-to-unsigned clipping. */ |
| 29 | vsetvli t0, zero, e16, m8, ta, ma | 28 | vsetvli t0, zero, e16, m8, ta, ma |
| @@ -54,17 +53,18 @@ func ff_put_signed_pixels_clamped_rvv, zve64x · 24 unmodified lines | |||
| 54 | ret | 53 | ret |
| 55 | endfunc | 54 | endfunc |
| 56 | 55 | ||
| 57 | func ff_add_pixels_clamped_rvv, zve32x | 56 | func ff_add_pixels_clamped_rvv, zve64x |
| 58 | vsetivli zero, 8, e8, mf2, ta, ma | 57 | vsetivli zero, 8, e8, mf2, ta, ma |
| 59 | vlseg8e16.v v24, (a0) | 58 | li t0, 8 * 8 |
| 60 | vlsseg8e8.v v16, (a1), a2 | 59 | vlse64.v v16, (a1), a2 |
| 61 | vwaddu.wv v24, v24, v16 | 60 | vsetvli zero, t0, e8, m4, ta, ma |
| 62 | vwaddu.wv v25, v25, v17 | 61 | vle16.v v24, (a0) |
| 63 | vwaddu.wv v26, v26, v18 | 62 | vwaddu.wv v24, v24, v16 |
| 64 | vwaddu.wv v27, v27, v19 | 63 | vsetvli zero, zero, e16, m8, ta, ma |
| 65 | vwaddu.wv v28, v28, v20 | 64 | vmax.vx v24, v24, zero |
| 66 | vwaddu.wv v29, v29, v21 | 65 | vsetvli zero, zero, e8, m4, ta, ma |
| 67 | vwaddu.wv v30, v30, v22 | 66 | vnclipu.wi v16, v24, 0 |
| 68 | vwaddu.wv v31, v31, v23 | 67 | vsetivli zero, 8, e8, mf2, ta, ma |
| 69 | j 1b | 68 | vsse64.v v16, (a1), a2 |
| 69 | ret | ||
| 70 | endfunc | 70 | endfunc |