lavc/takdsp: R-V V decorrelate_sr

C908:
decorrelate_sr_c: 95.5
decorrelate_sr_rvv_i32: 28.2

Signed-off-by: Rémi Denis-Courmont <remi@remlab.net>

libavcodec/riscv/takdsp_init.c

@@ -26,6 +26,7 @@
 #include "libavcodec/takdsp.h"
 
 void ff_decorrelate_ls_rvv(int32_t *p1, int32_t *p2, int length);
+void ff_decorrelate_sr_rvv(int32_t *p1, int32_t *p2, int length);
 
 av_cold void ff_takdsp_init_riscv(TAKDSPContext *dsp)
 {
@@ -34,6 +35,7 @@ av_cold void ff_takdsp_init_riscv(TAKDSPContext *dsp)
 
     if ((flags & AV_CPU_FLAG_RVV_I32) && (flags & AV_CPU_FLAG_RVB_ADDR)) {
         dsp->decorrelate_ls = ff_decorrelate_ls_rvv;
+        dsp->decorrelate_sr = ff_decorrelate_sr_rvv;
     }
 #endif
 }

libavcodec/riscv/takdsp_rvv.S

@@ -33,3 +33,17 @@ func ff_decorrelate_ls_rvv, zve32x
         bnez a2, 1b
         ret
 endfunc
+
+func ff_decorrelate_sr_rvv, zve32x
+1:
+        vsetvli  t0, a2, e32, m8, ta, ma
+        vle32.v  v0, (a0)
+        sub      a2, a2, t0
+        vle32.v  v8, (a1)
+        sh2add   a1, t0, a1
+        vsub.vv  v16, v8, v0
+        vse32.v  v16, (a0)
+        sh2add   a0, t0, a0
+        bnez     a2, 1b
+        ret
+endfunc