diff mbox series

[FFmpeg-devel,2/3] lavc/sbrdsp: R-V V sum_square

Message ID 20231029202559.95350-2-remi@remlab.net
State Accepted
Commit b0aba7dd0c1ee34941370997fd20049150f04a36
Headers show
Series [FFmpeg-devel,1/3] lavc/sbrdsp: R-V V sum64x5 | expand

Checks

Context Check Description
yinshiyou/make_loongarch64 success Make finished
yinshiyou/make_fate_loongarch64 success Make fate finished
andriy/make_x86 success Make finished
andriy/make_fate_x86 success Make fate finished

Commit Message

Rémi Denis-Courmont Oct. 29, 2023, 8:25 p.m. UTC 
sum_square_c:       803.5
sum_square_rvv_f32: 283.2
---
 libavcodec/riscv/sbrdsp_init.c |  2 ++
 libavcodec/riscv/sbrdsp_rvv.S  | 19 +++++++++++++++++++
 2 files changed, 21 insertions(+)
diff mbox series

Patch

diff --git a/libavcodec/riscv/sbrdsp_init.c b/libavcodec/riscv/sbrdsp_init.c
index 837f24e1e0..e0e62278b0 100644
--- a/libavcodec/riscv/sbrdsp_init.c
+++ b/libavcodec/riscv/sbrdsp_init.c
@@ -24,6 +24,7 @@ 
 #include "libavcodec/sbrdsp.h"
 
 void ff_sbr_sum64x5_rvv(float *z);
+float ff_sbr_sum_square_rvv(float (*x)[2], int n);
 
 av_cold void ff_sbrdsp_init_riscv(SBRDSPContext *c)
 {
@@ -32,6 +33,7 @@  av_cold void ff_sbrdsp_init_riscv(SBRDSPContext *c)
 
     if ((flags & AV_CPU_FLAG_RVV_F32) && (flags & AV_CPU_FLAG_RVB_ADDR)) {
         c->sum64x5 = ff_sbr_sum64x5_rvv;
+        c->sum_square = ff_sbr_sum_square_rvv;
     }
 #endif
 }
diff --git a/libavcodec/riscv/sbrdsp_rvv.S b/libavcodec/riscv/sbrdsp_rvv.S
index e1d548b41b..4684630953 100644
--- a/libavcodec/riscv/sbrdsp_rvv.S
+++ b/libavcodec/riscv/sbrdsp_rvv.S
@@ -48,3 +48,22 @@  func ff_sbr_sum64x5_rvv, zve32f
 
         ret
 endfunc
+
+func ff_sbr_sum_square_rvv, zve32f
+        vsetvli t0, zero, e32, m8, ta, ma
+        slli    a1, a1, 1
+        vmv.v.x v8, zero
+        vmv.s.x v0, zero
+1:
+        vsetvli t0, a1, e32, m8, tu, ma
+        vle32.v v16, (a0)
+        sub     a1, a1, t0
+        vfmacc.vv v8, v16, v16
+        sh2add  a0, t0, a0
+        bnez    a1, 1b
+
+        vfredusum.vs v0, v8, v0
+        vfmv.f.s fa0, v0
+NOHWF   fmv.x.w  a0, fa0
+        ret
+endfunc