From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from ffbox0-bg.mplayerhq.hu (ffbox0-bg.ffmpeg.org [79.124.17.100]) by master.gitmailbox.com (Postfix) with ESMTP id 6747D40333 for ; Tue, 18 Jul 2023 18:53:21 +0000 (UTC) Received: from [127.0.1.1] (localhost [127.0.0.1]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTP id 7C6FD68C53B; Tue, 18 Jul 2023 21:53:02 +0300 (EEST) Received: from ursule.remlab.net (vps-a2bccee9.vps.ovh.net [51.75.19.47]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTP id C8F0368C2EC for ; Tue, 18 Jul 2023 21:52:54 +0300 (EEST) Received: from basile.remlab.net (localhost [IPv6:::1]) by ursule.remlab.net (Postfix) with ESMTP id 1F6A1C00AD for ; Tue, 18 Jul 2023 21:52:54 +0300 (EEST) From: =?UTF-8?q?R=C3=A9mi=20Denis-Courmont?= To: ffmpeg-devel@ffmpeg.org Date: Tue, 18 Jul 2023 21:52:52 +0300 Message-Id: <20230718185253.18927-2-remi@remlab.net> X-Mailer: git-send-email 2.40.1 MIME-Version: 1.0 Subject: [FFmpeg-devel] [PATCH 2/3] swscale/rgb2rgb2: rework RISC-V V shuffle_bytes_{0321, 2103} X-BeenThere: ffmpeg-devel@ffmpeg.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: FFmpeg development discussions and patches List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Reply-To: FFmpeg development discussions and patches Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Errors-To: ffmpeg-devel-bounces@ffmpeg.org Sender: "ffmpeg-devel" Archived-At: List-Archive: List-Post: This avoids strided loads. Before: shuffle_bytes_0321_rvv_i32: 307.7 shuffle_bytes_2103_rvv_i32: 308.7 After: shuffle_bytes_0321_rvv_i32: 59.7 shuffle_bytes_2103_rvv_i32: 61.5 --- libswscale/riscv/rgb2rgb_rvv.S | 45 +++++++++++++++------------------- 1 file changed, 20 insertions(+), 25 deletions(-) diff --git a/libswscale/riscv/rgb2rgb_rvv.S b/libswscale/riscv/rgb2rgb_rvv.S index 8f30d760cb..366f80c1c3 100644 --- a/libswscale/riscv/rgb2rgb_rvv.S +++ b/libswscale/riscv/rgb2rgb_rvv.S @@ -21,38 +21,33 @@ #include "libavutil/riscv/asm.S" func ff_shuffle_bytes_0321_rvv, zve32x - addi t1, a0, 3 - addi t2, a0, 2 - addi t3, a0, 1 + li t1, 0x00ff00ff + j 1f +endfunc + +func ff_shuffle_bytes_2103_rvv, zve32x + li t1, ~0x00ff00ff 1: + not t2, t1 srai a2, a2, 2 - li t4, 4 2: - vsetvli t0, a2, e8, m1, ta, ma - sub a2, a2, t0 - vlse8.v v8, (a0), t4 - sh2add a0, t0, a0 - vlse8.v v9, (t1), t4 - sh2add t1, t0, t1 - vlse8.v v10, (t2), t4 - sh2add t2, t0, t2 - vlse8.v v11, (t3), t4 - sh2add t3, t0, t3 - vsseg4e8.v v8, (a1) - sh2add a1, t0, a1 - bnez a2, 2b + vsetvli t0, a2, e32, m8, ta, ma + vle32.v v8, (a0) + sub a2, a2, t0 + vand.vx v16, v8, t2 + sh2add a0, t0, a0 + vand.vx v8, v8, t1 + vsrl.vi v24, v16, 16 + vsll.vi v16, v16, 16 + vor.vv v8, v8, v24 + vor.vv v8, v16, v8 + vse32.v v8, (a1) + sh2add a1, t0, a1 + bnez a2, 2b ret endfunc -func ff_shuffle_bytes_2103_rvv, zve32x - addi t1, a0, 1 - addi t2, a0, 0 - addi t3, a0, 3 - addi a0, a0, 2 - j 1b -endfunc - func ff_shuffle_bytes_1230_rvv, zve32x addi t1, a0, 2 addi t2, a0, 3 -- 2.40.1 _______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-devel To unsubscribe, visit link above, or email ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".