mirror of
https://git.ffmpeg.org/ffmpeg.git
synced 2025-12-08 06:09:50 +00:00
lavc/h264dsp: R-V V high-depth add_pixels8
T-Head C908 (cycles); h264_add_pixels8_9bpp_c: 270.5 h264_add_pixels8_9bpp_rvv_i32: 164.2 h264_add_pixels8_10bpp_c: 270.5 h264_add_pixels8_10bpp_rvv_i32: 164.2 h264_add_pixels8_12bpp_c: 270.5 h264_add_pixels8_12bpp_rvv_i32: 164.2 h264_add_pixels8_14bpp_c: 270.5 h264_add_pixels8_14bpp_rvv_i32: 164.2
This commit is contained in:
parent
7744c08240
commit
3002310b70
2 changed files with 24 additions and 0 deletions
|
|
@ -87,3 +87,25 @@ func ff_h264_add_pixels8_8_rvv, zve64x
|
||||||
vsse64.v v8, (a0), a2
|
vsse64.v v8, (a0), a2
|
||||||
ret
|
ret
|
||||||
endfunc
|
endfunc
|
||||||
|
|
||||||
|
func ff_h264_add_pixels8_16_rvv, zve32x
|
||||||
|
li t0, 8
|
||||||
|
vsetivli zero, 8, e16, m1, ta, ma
|
||||||
|
1:
|
||||||
|
vle32.v v16, (a1)
|
||||||
|
addi t0, t0, -1
|
||||||
|
vle16.v v8, (a0)
|
||||||
|
.equ offset, 0
|
||||||
|
.rept 256 / __riscv_xlen
|
||||||
|
sx zero, offset(a1)
|
||||||
|
.equ offset, offset + (__riscv_xlen / 8)
|
||||||
|
.endr
|
||||||
|
vncvt.x.x.w v24, v16
|
||||||
|
addi a1, a1, 8 * 4
|
||||||
|
vadd.vv v8, v8, v24
|
||||||
|
vse16.v v8, (a0)
|
||||||
|
add a0, a0, a2
|
||||||
|
bnez t0, 1b
|
||||||
|
|
||||||
|
ret
|
||||||
|
endfunc
|
||||||
|
|
|
||||||
|
|
@ -63,6 +63,7 @@ void ff_h264_idct8_add_14_rvv(uint8_t *dst, int16_t *block, int stride);
|
||||||
|
|
||||||
void ff_h264_add_pixels8_8_rvv(uint8_t *dst, int16_t *block, int stride);
|
void ff_h264_add_pixels8_8_rvv(uint8_t *dst, int16_t *block, int stride);
|
||||||
void ff_h264_add_pixels4_8_rvv(uint8_t *dst, int16_t *block, int stride);
|
void ff_h264_add_pixels4_8_rvv(uint8_t *dst, int16_t *block, int stride);
|
||||||
|
void ff_h264_add_pixels8_16_rvv(uint8_t *dst, int16_t *block, int stride);
|
||||||
void ff_h264_add_pixels4_16_rvv(uint8_t *dst, int16_t *block, int stride);
|
void ff_h264_add_pixels4_16_rvv(uint8_t *dst, int16_t *block, int stride);
|
||||||
|
|
||||||
extern int ff_startcode_find_candidate_rvb(const uint8_t *, int);
|
extern int ff_startcode_find_candidate_rvb(const uint8_t *, int);
|
||||||
|
|
@ -126,6 +127,7 @@ av_cold void ff_h264dsp_init_riscv(H264DSPContext *dsp, const int bit_depth,
|
||||||
dsp->h264_idct8_add = ff_h264_idct8_add_14_rvv;
|
dsp->h264_idct8_add = ff_h264_idct8_add_14_rvv;
|
||||||
}
|
}
|
||||||
if (bit_depth > 8 && zvl128b) {
|
if (bit_depth > 8 && zvl128b) {
|
||||||
|
dsp->h264_add_pixels8_clear = ff_h264_add_pixels8_16_rvv;
|
||||||
if (flags & AV_CPU_FLAG_RVV_I64)
|
if (flags & AV_CPU_FLAG_RVV_I64)
|
||||||
dsp->h264_add_pixels4_clear = ff_h264_add_pixels4_16_rvv;
|
dsp->h264_add_pixels4_clear = ff_h264_add_pixels4_16_rvv;
|
||||||
}
|
}
|
||||||
|
|
|
||||||
Loading…
Add table
Add a link
Reference in a new issue