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 EDEDB4720F for ; Sat, 30 Sep 2023 19:04:46 +0000 (UTC) Received: from [127.0.1.1] (localhost [127.0.0.1]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTP id E518968CD12; Sat, 30 Sep 2023 22:04:43 +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 633C368CC17 for ; Sat, 30 Sep 2023 22:04:37 +0300 (EEST) Received: from basile.remlab.net (localhost [IPv6:::1]) by ursule.remlab.net (Postfix) with ESMTP id 01E54C006B for ; Sat, 30 Sep 2023 22:04:36 +0300 (EEST) From: =?UTF-8?q?R=C3=A9mi=20Denis-Courmont?= To: ffmpeg-devel@ffmpeg.org Date: Sat, 30 Sep 2023 22:04:36 +0300 Message-ID: <20230930190436.16518-1-remi@remlab.net> X-Mailer: git-send-email 2.42.0 MIME-Version: 1.0 Subject: [FFmpeg-devel] [PATCH] lavu/float_dsp: avoid reg-stride in R-V V fmul_window 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: --- libavutil/riscv/float_dsp_rvv.S | 45 ++++++++++++++++++--------------- 1 file changed, 25 insertions(+), 20 deletions(-) diff --git a/libavutil/riscv/float_dsp_rvv.S b/libavutil/riscv/float_dsp_rvv.S index 7e9e84d526..91b70bf148 100644 --- a/libavutil/riscv/float_dsp_rvv.S +++ b/libavutil/riscv/float_dsp_rvv.S @@ -75,32 +75,37 @@ endfunc func ff_vector_fmul_window_rvv, zve32f // a0: dst, a1: src0, a2: src1, a3: window, a4: length - addi t0, a4, -1 - add t1, t0, a4 - sh2add a2, t0, a2 - sh2add t0, t1, a0 - sh2add t3, t1, a3 - li t1, -4 // byte stride + vsetvli t0, zero, e16, m4, ta, ma + sh2add a2, a4, a2 + vid.v v0 + sh3add t3, a4, a3 + vadd.vi v0, v0, 1 + sh3add t0, a4, a0 1: - vsetvli t2, a4, e32, m4, ta, ma - vle32.v v16, (a1) + vsetvli t2, a4, e16, m2, ta, ma slli t4, t2, 2 - vlse32.v v20, (a2), t1 + vrsub.vx v2, v0, t2 + sub t3, t3, t4 + vsetvli zero, zero, e32, m4, ta, ma + sub a2, a2, t4 + vle32.v v8, (t3) + sub t0, t0, t4 + vle32.v v4, (a2) sub a4, a4, t2 - vle32.v v24, (a3) + vrgatherei16.vv v28, v8, v2 + vle32.v v16, (a1) add a1, a1, t4 - vlse32.v v28, (t3), t1 - sub a2, a2, t4 - vfmul.vv v0, v16, v28 + vrgatherei16.vv v20, v4, v2 + vle32.v v24, (a3) add a3, a3, t4 - vfmul.vv v8, v16, v24 - sub t3, t3, t4 - vfnmsac.vv v0, v20, v24 - vfmacc.vv v8, v20, v28 - vse32.v v0, (a0) + vfmul.vv v12, v16, v28 + vfmul.vv v16, v16, v24 + vfnmsac.vv v12, v20, v24 + vfmacc.vv v16, v20, v28 + vrgatherei16.vv v8, v16, v2 + vse32.v v12, (a0) add a0, a0, t4 - vsse32.v v8, (t0), t1 - sub t0, t0, t4 + vse32.v v8, (t0) bnez a4, 1b ret -- 2.42.0 _______________________________________________ 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".