diff mbox series

[FFmpeg-devel,15/31] lavu/floatdsp: RISC-V V butterflies_float

Message ID 20220926145251.56351-15-remi@remlab.net
State Accepted
Commit 47ce9735cc8362ea6b3a508dd660233af28d77ba
Headers show
Series initial RISC-V CPU extensions | 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 Sept. 26, 2022, 2:52 p.m. UTC 
From: Rémi Denis-Courmont <remi@remlab.net>

---
 libavutil/riscv/float_dsp_init.c |  2 ++
 libavutil/riscv/float_dsp_rvv.S  | 18 ++++++++++++++++++
 2 files changed, 20 insertions(+)
diff mbox series

Patch

diff --git a/libavutil/riscv/float_dsp_init.c b/libavutil/riscv/float_dsp_init.c
index 8982436647..a1cd180cdc 100644
--- a/libavutil/riscv/float_dsp_init.c
+++ b/libavutil/riscv/float_dsp_init.c
@@ -33,6 +33,7 @@  void ff_vector_fmul_scalar_rvv(float *dst, const float *src, float mul,
                                 int len);
 void ff_vector_fmul_add_rvv(float *dst, const float *src0, const float *src1,
                              const float *src2, int len);
+void ff_butterflies_float_rvv(float *v1, float *v2, int len);
 
 void ff_vector_dmul_rvv(double *dst, const double *src0, const double *src1,
                          int len);
@@ -51,6 +52,7 @@  av_cold void ff_float_dsp_init_riscv(AVFloatDSPContext *fdsp)
         fdsp->vector_fmac_scalar = ff_vector_fmac_scalar_rvv;
         fdsp->vector_fmul_scalar = ff_vector_fmul_scalar_rvv;
         fdsp->vector_fmul_add = ff_vector_fmul_add_rvv;
+        fdsp->butterflies_float = ff_butterflies_float_rvv;
     }
 
     if (flags & AV_CPU_FLAG_RVV_F64) {
diff --git a/libavutil/riscv/float_dsp_rvv.S b/libavutil/riscv/float_dsp_rvv.S
index db62402878..a721c44667 100644
--- a/libavutil/riscv/float_dsp_rvv.S
+++ b/libavutil/riscv/float_dsp_rvv.S
@@ -93,6 +93,24 @@  func ff_vector_fmul_add_rvv, zve32f
         ret
 endfunc
 
+// (a0) = (a0) + (a1), (a1) = (a0) - (a1) [0..a2-1]
+func ff_butterflies_float_rvv, zve32f
+1:
+        vsetvli  t0, a2, e32, m1, ta, ma
+        vle32.v  v16, (a0)
+        sub      a2, a2, t0
+        vle32.v  v24, (a1)
+        vfadd.vv v0, v16, v24
+        vfsub.vv v8, v16, v24
+        vse32.v  v0, (a0)
+        sh2add   a0, t0, a0
+        vse32.v  v8, (a1)
+        sh2add   a1, t0, a1
+        bnez     a2, 1b
+
+        ret
+endfunc
+
 // (a0) = (a1) * (a2) [0..a3-1]
 func ff_vector_dmul_rvv, zve64d
 1: