[FFmpeg-devel] [PATCH 2/2] lavc/vvc_mc: R-V V dmvr
uk7b at foxmail.com
uk7b at foxmail.com
Sat Sep 28 19:47:12 EEST 2024
From: sunyuechi <sunyuechi at iscas.ac.cn>
k230 banana_f3
dmvr_8_12x20_c: 619.3 ( 1.00x) 624.1 ( 1.00x)
dmvr_8_12x20_rvv_i32: 128.6 ( 4.82x) 103.4 ( 6.04x)
dmvr_8_20x12_c: 610.0 ( 1.00x) 665.6 ( 1.00x)
dmvr_8_20x12_rvv_i32: 137.6 ( 4.44x) 92.9 ( 7.17x)
dmvr_8_20x20_c: 1008.0 ( 1.00x) 1082.7 ( 1.00x)
dmvr_8_20x20_rvv_i32: 221.1 ( 4.56x) 155.4 ( 6.97x)
dmvr_h_8_12x20_c: 2008.0 ( 1.00x) 2009.7 ( 1.00x)
dmvr_h_8_12x20_rvv_i32: 239.6 ( 8.38x) 186.7 (10.77x)
dmvr_h_8_20x12_c: 1989.5 ( 1.00x) 2009.4 ( 1.00x)
dmvr_h_8_20x12_rvv_i32: 230.3 ( 8.64x) 155.4 (12.93x)
dmvr_h_8_20x20_c: 3304.1 ( 1.00x) 3342.9 ( 1.00x)
dmvr_h_8_20x20_rvv_i32: 378.3 ( 8.73x) 248.9 (13.43x)
dmvr_hv_8_12x20_c: 3609.8 ( 1.00x) 3603.4 ( 1.00x)
dmvr_hv_8_12x20_rvv_i32: 369.1 ( 9.78x) 322.1 (11.19x)
dmvr_hv_8_20x12_c: 3628.3 ( 1.00x) 3624.2 ( 1.00x)
dmvr_hv_8_20x12_rvv_i32: 322.8 (11.24x) 238.7 (15.19x)
dmvr_hv_8_20x20_c: 5933.8 ( 1.00x) 5936.6 ( 1.00x)
dmvr_hv_8_20x20_rvv_i32: 526.5 (11.27x) 374.1 (15.87x)
dmvr_v_8_12x20_c: 2156.3 ( 1.00x) 2155.4 ( 1.00x)
dmvr_v_8_12x20_rvv_i32: 239.6 ( 9.00x) 176.2 (12.24x)
dmvr_v_8_20x12_c: 2137.6 ( 1.00x) 2165.9 ( 1.00x)
dmvr_v_8_20x12_rvv_i32: 230.3 ( 9.28x) 155.2 (13.96x)
dmvr_v_8_20x20_c: 4183.8 ( 1.00x) 3592.9 ( 1.00x)
dmvr_v_8_20x20_rvv_i32: 369.3 (11.33x) 249.2 (14.42x)
---
libavcodec/riscv/vvc/vvc_mc_rvv.S | 120 +++++++++++++++++++++++++++++
libavcodec/riscv/vvc/vvcdsp_init.c | 22 ++++++
2 files changed, 142 insertions(+)
diff --git a/libavcodec/riscv/vvc/vvc_mc_rvv.S b/libavcodec/riscv/vvc/vvc_mc_rvv.S
index 18532616d9..2c634af48f 100644
--- a/libavcodec/riscv/vvc/vvc_mc_rvv.S
+++ b/libavcodec/riscv/vvc/vvc_mc_rvv.S
@@ -285,3 +285,123 @@ endfunc
func_w_avg 128
func_w_avg 256
#endif
+
+func dmvr zve32x, zbb, zba
+ lpad 0
+ li t0, 4
+1:
+ add t1, a1, a2
+ addi t4, a0, 128*2
+ vle8.v v0, (a1)
+ vle8.v v4, (t1)
+ addi a3, a3, -2
+ vwmulu.vx v16, v0, t0
+ vwmulu.vx v20, v4, t0
+ vse16.v v16, (a0)
+ vse16.v v20, (t4)
+ sh1add a1, a2, a1
+ add a0, a0, 128*2*2
+ bnez a3, 1b
+ ret
+endfunc
+
+.macro dmvr_h_v mn, type, w, vlen
+dmvr_\type\vlen\w:
+ lla t4, ff_vvc_inter_luma_dmvr_filters
+ sh1add t4, \mn, t4
+ lbu t5, (t4)
+ lbu t6, 1(t4)
+1:
+ vsetvlstatic8 \w, \vlen
+.ifc \type,h
+ addi t0, a1, 1
+ addi t1, a1, 2
+.else
+ add t0, a1, a2
+ add t1, t0, a2
+.endif
+ vle8.v v0, (a1)
+ vle8.v v4, (t0)
+ vle8.v v8, (t1)
+ addi a3, a3, -2
+ addi t2, a0, 128*2
+ vwmulu.vx v12, v0, t5
+ vwmulu.vx v24, v4, t5
+ vwmaccu.vx v12, t6, v4
+ vwmaccu.vx v24, t6, v8
+ vsetvlstatic16 \w, \vlen
+ vssrl.vi v12, v12, 2
+ vssrl.vi v24, v24, 2
+ vse16.v v12, (a0)
+ vse16.v v24, (t2)
+ add a0, a0, 128*4
+ sh1add a1, a2, a1
+ bnez a3, 1b
+ ret
+.endm
+
+.macro dmvr_load_h dst, filter0, filter1, w, vlen
+ vsetvlstatic8 \w, \vlen
+ addi a6, a1, 1
+ vle8.v \dst, (a1)
+ vle8.v v2, (a6)
+ vwmulu.vx v4, \dst, \filter0
+ vwmaccu.vx v4, \filter1, v2
+ vsetvlstatic16 \w, \vlen
+ vssrl.vi \dst, v4, 2
+.endm
+
+.macro dmvr_hv w, vlen
+dmvr_hv\vlen\w:
+ lla t0, ff_vvc_inter_luma_dmvr_filters
+ sh1add t1, a4, t0
+ sh1add t2, a5, t0
+ lbu t3, (t1) // filter[mx][0]
+ lbu t4, 1(t1) // filter[mx][1]
+ lbu t5, (t2) // filter[my][0]
+ lbu t6, 1(t2) // filter[my][1]
+ dmvr_load_h v12, t3, t4, \w, \vlen
+ add a1, a1, a2
+1:
+ vmul.vx v28, v12, t5
+ addi a3, a3, -1
+ dmvr_load_h v12, t3, t4, \w, \vlen
+ vmacc.vx v28, t6, v12
+ vssrl.vi v28, v28, 4
+ vse16.v v28, (a0)
+ add a1, a1, a2
+ addi a0, a0, 128*2
+ bnez a3, 1b
+ ret
+.endm
+
+.macro func_dmvr vlen, name
+func ff_vvc_\name\()_8_rvv_\vlen\(), zve32x, zbb, zba
+ lpad 0
+ li t0, 20
+ beq a6, t0, DMVR\name\vlen\()20
+ .irp w,12,20
+DMVR\name\vlen\w:
+ .ifc \name, dmvr
+ vsetvlstatic8 \w, \vlen
+ j \name
+ .else
+ csrwi vxrm, 0
+ j \name\()\vlen\w
+ .endif
+ .endr
+endfunc
+.endm
+
+
+.irp vlen,256,128
+.irp w,12,20
+dmvr_h_v a4, h, \w, \vlen
+dmvr_h_v a5, v, \w, \vlen
+dmvr_hv \w, \vlen
+.endr
+func_dmvr \vlen, dmvr
+func_dmvr \vlen, dmvr_h
+func_dmvr \vlen, dmvr_v
+func_dmvr \vlen, dmvr_hv
+.endr
diff --git a/libavcodec/riscv/vvc/vvcdsp_init.c b/libavcodec/riscv/vvc/vvcdsp_init.c
index ac1e7dda7d..7df3ce58db 100644
--- a/libavcodec/riscv/vvc/vvcdsp_init.c
+++ b/libavcodec/riscv/vvc/vvcdsp_init.c
@@ -37,6 +37,26 @@ void bf(ff_vvc_w_avg, bd, opt)(uint8_t *dst, ptrdiff_t dst_stride,
AVG_PROTOTYPES(8, rvv_128)
AVG_PROTOTYPES(8, rvv_256)
+#define DMVR_PROTOTYPES(bd, opt) \
+void ff_vvc_dmvr_##bd##_##opt(int16_t *dst, const uint8_t *src, ptrdiff_t src_stride, \
+ int height, intptr_t mx, intptr_t my, int width); \
+void ff_vvc_dmvr_h_##bd##_##opt(int16_t *dst, const uint8_t *src, ptrdiff_t src_stride, \
+ int height, intptr_t mx, intptr_t my, int width); \
+void ff_vvc_dmvr_v_##bd##_##opt(int16_t *dst, const uint8_t *src, ptrdiff_t src_stride, \
+ int height, intptr_t mx, intptr_t my, int width); \
+void ff_vvc_dmvr_hv_##bd##_##opt(int16_t *dst, const uint8_t *src, ptrdiff_t src_stride, \
+ int height, intptr_t mx, intptr_t my, int width); \
+
+DMVR_PROTOTYPES(8, rvv_128)
+DMVR_PROTOTYPES(8, rvv_256)
+
+#define DMVR_INIT(bd, opt) do { \
+ c->inter.dmvr[0][0] = ff_vvc_dmvr_##bd##_##opt; \
+ c->inter.dmvr[0][1] = ff_vvc_dmvr_h_##bd##_##opt; \
+ c->inter.dmvr[1][0] = ff_vvc_dmvr_v_##bd##_##opt; \
+ c->inter.dmvr[1][1] = ff_vvc_dmvr_hv_##bd##_##opt; \
+} while (0)
+
void ff_vvc_dsp_init_riscv(VVCDSPContext *const c, const int bd)
{
#if HAVE_RVV
@@ -51,6 +71,7 @@ void ff_vvc_dsp_init_riscv(VVCDSPContext *const c, const int bd)
# if (__riscv_xlen == 64)
c->inter.w_avg = ff_vvc_w_avg_8_rvv_256;
# endif
+ DMVR_INIT(8, rvv_256);
break;
default:
break;
@@ -63,6 +84,7 @@ void ff_vvc_dsp_init_riscv(VVCDSPContext *const c, const int bd)
# if (__riscv_xlen == 64)
c->inter.w_avg = ff_vvc_w_avg_8_rvv_128;
# endif
+ DMVR_INIT(8, rvv_128);
break;
default:
break;
--
2.46.2
More information about the ffmpeg-devel
mailing list