diff mbox series

[FFmpeg-devel,3/3] lavc/vp8dsp: R-V V vp8_idct_dc_add4uv

Message ID CAEa-L+sHTMbZUA5jz60_uUXKYquLevh8rov8Z2HKgwbPAD3FwA@mail.gmail.com
State New
Headers show
Series [FFmpeg-devel,1/3] lavc/vp8dsp: R-V V vp8_idct_dc_add | expand

Checks

Context Check Description
andriy/configure_x86 warning Failed to apply patch
yinshiyou/configure_loongarch64 warning Failed to apply patch

Commit Message

flow gg Feb. 2, 2024, 5:04 a.m. UTC

Comments

flow gg Feb. 2, 2024, 8 a.m. UTC | #1
add check rvb in this reply

flow gg <hlefthleft@gmail.com> 于2024年2月2日周五 13:04写道:

>
>
diff mbox series

Patch

From 71df566c93a0b7538405cd9bb2da6da155084283 Mon Sep 17 00:00:00 2001
From: sunyuechi <sunyuechi@iscas.ac.cn>
Date: Fri, 2 Feb 2024 12:50:50 +0800
Subject: [PATCH 3/3] lavc/vp8dsp: R-V V vp8_idct_dc_add4uv

c908:
vp8_idct_dc_add4uv_c: 387.7
vp8_idct_dc_add4uv_rvv_i32: 134.5
---
 libavcodec/riscv/vp8dsp_init.c |  2 ++
 libavcodec/riscv/vp8dsp_rvv.S  | 13 +++++++++++++
 2 files changed, 15 insertions(+)

diff --git a/libavcodec/riscv/vp8dsp_init.c b/libavcodec/riscv/vp8dsp_init.c
index 6615d3d440..58f0afa6f9 100644
--- a/libavcodec/riscv/vp8dsp_init.c
+++ b/libavcodec/riscv/vp8dsp_init.c
@@ -27,6 +27,7 @@ 
 
 void ff_vp8_idct_dc_add_rvv(uint8_t *dst, int16_t block[16], ptrdiff_t stride);
 void ff_vp8_idct_dc_add4y_rvv(uint8_t *dst, int16_t block[4][16], ptrdiff_t stride);
+void ff_vp8_idct_dc_add4uv_rvv(uint8_t *dst, int16_t block[4][16], ptrdiff_t stride);
 
 av_cold void ff_vp8dsp_init_riscv(VP8DSPContext *c)
 {
@@ -36,6 +37,7 @@  av_cold void ff_vp8dsp_init_riscv(VP8DSPContext *c)
     if (flags & AV_CPU_FLAG_RVV_I32 && ff_get_rv_vlenb() >= 16) {
         c->vp8_idct_dc_add = ff_vp8_idct_dc_add_rvv;
         c->vp8_idct_dc_add4y = ff_vp8_idct_dc_add4y_rvv;
+        c->vp8_idct_dc_add4uv = ff_vp8_idct_dc_add4uv_rvv;
     }
 #endif
 }
diff --git a/libavcodec/riscv/vp8dsp_rvv.S b/libavcodec/riscv/vp8dsp_rvv.S
index 87c369fb16..eb93111055 100644
--- a/libavcodec/riscv/vp8dsp_rvv.S
+++ b/libavcodec/riscv/vp8dsp_rvv.S
@@ -58,3 +58,16 @@  func ff_vp8_idct_dc_add4y_rvv, zve32x
 
         ret
 endfunc
+
+func ff_vp8_idct_dc_add4uv_rvv, zve32x
+        vsetivli      zero, 4, e8, mf4, ta, ma
+        vp8_idct_dc_addy
+        vp8_idct_dc_add
+        addi          a0, a0, -4
+        sh2add        a0, a2, a0
+        addi          a1, a1, 32
+        vp8_idct_dc_addy
+        vp8_idct_dc_add
+
+        ret
+endfunc
-- 
2.43.0