diff mbox series

[FFmpeg-devel,2/3] lavc/h264pred: R-V V pred16x16_horizontal_8

Message ID CAEa-L+u9+hriEEJEcVenUWx_bY6_ERAmy+CVWHKj9xpk0FNnHQ@mail.gmail.com
State New
Headers show
Series [FFmpeg-devel,1/3] lavc/h264pred: R-V V pred16x16_vertical_8 | 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 Jan. 16, 2024, 4:15 p.m. UTC
diff mbox series

Patch

From 806f84ea5557c4652e48451decc4c679c9485472 Mon Sep 17 00:00:00 2001
From: sunyuechi <sunyuechi@iscas.ac.cn>
Date: Tue, 16 Jan 2024 23:56:33 +0800
Subject: [PATCH 2/3] lavc/h264pred: R-V V pred16x16_horizontal_8

C908
pred16x16_horizontal_8_c: 3.0
pred16x16_horizontal_8_rvv_i32: 2.5
---
 libavcodec/riscv/h264pred_init.c |  2 ++
 libavcodec/riscv/h264pred_rvv.S  | 15 +++++++++++++++
 2 files changed, 17 insertions(+)

diff --git a/libavcodec/riscv/h264pred_init.c b/libavcodec/riscv/h264pred_init.c
index 179bd8c8a5..8665bc729e 100644
--- a/libavcodec/riscv/h264pred_init.c
+++ b/libavcodec/riscv/h264pred_init.c
@@ -25,6 +25,7 @@ 
 #include "libavcodec/h264pred.h"
 
 void ff_pred16x16_vertical_8_rvv(uint8_t *src, ptrdiff_t stride);
+void ff_pred16x16_horizontal_8_rvv(uint8_t *src, ptrdiff_t stride);
 
 av_cold void ff_h264_pred_init_riscv(H264PredContext *h, int codec_id,
                                    const int bit_depth,
@@ -36,6 +37,7 @@  av_cold void ff_h264_pred_init_riscv(H264PredContext *h, int codec_id,
 
             if (flags & AV_CPU_FLAG_RVV_I32) {
                 h->pred16x16[VERT_PRED8x8] = ff_pred16x16_vertical_8_rvv;
+                h->pred16x16[HOR_PRED8x8] = ff_pred16x16_horizontal_8_rvv;
             }
         #endif
     }
diff --git a/libavcodec/riscv/h264pred_rvv.S b/libavcodec/riscv/h264pred_rvv.S
index 0e48662922..ba1e9045e1 100644
--- a/libavcodec/riscv/h264pred_rvv.S
+++ b/libavcodec/riscv/h264pred_rvv.S
@@ -33,3 +33,18 @@  func ff_pred16x16_vertical_8_rvv, zve32x
 
         ret
 endfunc
+
+func ff_pred16x16_horizontal_8_rvv, zve32x
+        li        t0, 16
+1:
+        lbu       t1, -1(a0)
+        vsetivli  zero, 16, e8, m1, ta, ma
+        vmv.v.x   v0, t1
+        vsetivli  zero, 4, e8, mf4, ta, ma
+        addi      t0, t0, -1
+        vse32.v   v0, (a0)
+        add       a0, a0, a1
+        bnez      t0, 1b
+
+        ret
+endfunc
-- 
2.43.0