virtualenv/FFmpeg
1
0
Fork 0
mirror of https://github.com/FFmpeg/FFmpeg.git synced 2024-12-23 12:43:46 +02:00
Code Issues Releases Activity

lavc/audiodsp: RISC-V V scalarproduct_int16

Browse Source
This commit is contained in:
Rémi Denis-Courmont 2022-09-26 17:52:42 +03:00 committed by Lynne
parent f127a5d29d
commit f41ae62f39
2 changed files with 23 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
libavcodec/riscv/audiodsp_init.c
View File

@ -26,6 +26,7 @@
void ff_vector_clipf_rvf(float *dst, const float *src, int len, float min, float max); void ff_vector_clipf_rvf(float *dst, const float *src, int len, float min, float max);
int32_t ff_scalarproduct_int16_rvv(const int16_t *v1, const int16_t *v2, int len);
void ff_vector_clip_int32_rvv(int32_t *dst, const int32_t *src, int32_t min, void ff_vector_clip_int32_rvv(int32_t *dst, const int32_t *src, int32_t min,
int32_t max, unsigned int len); int32_t max, unsigned int len);
void ff_vector_clipf_rvv(float *dst, const float *src, int len, float min, float max); void ff_vector_clipf_rvv(float *dst, const float *src, int len, float min, float max);
@ -37,8 +38,10 @@ av_cold void ff_audiodsp_init_riscv(AudioDSPContext *c)
if (flags & AV_CPU_FLAG_RVF) if (flags & AV_CPU_FLAG_RVF)
c->vector_clipf = ff_vector_clipf_rvf; c->vector_clipf = ff_vector_clipf_rvf;
#if HAVE_RVV #if HAVE_RVV
if (flags & AV_CPU_FLAG_RVV_I32) if (flags & AV_CPU_FLAG_RVV_I32) {
c->scalarproduct_int16 = ff_scalarproduct_int16_rvv;
c->vector_clip_int32 = ff_vector_clip_int32_rvv; c->vector_clip_int32 = ff_vector_clip_int32_rvv;
}
if (flags & AV_CPU_FLAG_RVV_F32) if (flags & AV_CPU_FLAG_RVV_F32)
c->vector_clipf = ff_vector_clipf_rvv; c->vector_clipf = ff_vector_clipf_rvv;
#endif #endif

19
libavcodec/riscv/audiodsp_rvv.S
View File

@ -20,6 +20,25 @@
#include "libavutil/riscv/asm.S" #include "libavutil/riscv/asm.S"
func ff_scalarproduct_int16_rvv, zve32x
vsetvli zero, zero, e16, m1, ta, ma
vmv.s.x v8, zero
1:
vsetvli t0, a2, e16, m1, ta, ma
vle16.v v16, (a0)
sub a2, a2, t0
vle16.v v24, (a1)
sh1add a0, t0, a0
vwmul.vv v0, v16, v24
sh1add a1, t0, a1
vsetvli zero, t0, e32, m2, ta, ma
vredsum.vs v8, v0, v8
bnez a2, 1b
vmv.x.s a0, v8
ret
endfunc
func ff_vector_clip_int32_rvv, zve32x func ff_vector_clip_int32_rvv, zve32x
1: 1:
vsetvli t0, a4, e32, m1, ta, ma vsetvli t0, a4, e32, m1, ta, ma