From patchwork Sun Jun 9 09:27:09 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: =?utf-8?q?R=C3=A9mi_Denis-Courmont?= X-Patchwork-Id: 49732 Delivered-To: ffmpegpatchwork2@gmail.com Received: by 2002:a59:c209:0:b0:460:55fa:d5ed with SMTP id d9csp1929063vqo; Sun, 9 Jun 2024 02:27:45 -0700 (PDT) X-Forwarded-Encrypted: i=2; AJvYcCXok7JRBBQU4mBZACNaG87Ctua5a2VdzKqnygjj6UDh5sLkVEcWDZdJ3xNqbB+WDR+9JWbz588xUVnYTVIn2VzLbVKRR1MenycTFQ== X-Google-Smtp-Source: AGHT+IE+kYFB0F10ZEkLJmpUSgeVqX/0OuFlCn2gMLBxZuAJJ41ltY2e9jtz8AELguML/Ri+OKet X-Received: by 2002:a17:907:84c:b0:a65:2b64:f84a with SMTP id a640c23a62f3a-a6c7650a2f4mr988295266b.25.1717925265608; Sun, 09 Jun 2024 02:27:45 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1717925265; cv=none; d=google.com; s=arc-20160816; b=NPUUZT385OmJdySln77HmletHEsxu3GYgQ0HkQwscYAkY6JuBrWlWSMc1h1H3iZT/2 aIjhO6ULzCHYH4elA3a8xjF8ZZqd3QaaMIoRHpOuAiveHxdjN8A5djbFG/gwPH125trB CtpYrJmUwbtcu6iKH4MEc/8KhSp9KRLK5CLO+1tbCmF442vmKOXTfXu2RY/RqBJ8fCnU ryZvYqSi7cwimfgD74AO/83vtC/8KWgaorNVYs13N0zmmUGENckG3VmYi/D9UUvJHlTh teYTlI7kCJMEGRYlq/K4wcODnb5zH7VhQLgvSnxjdMJFffXcX54j0vwkedR2t4fcUGgI 20aw== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=sender:errors-to:content-transfer-encoding:reply-to:list-subscribe :list-help:list-post:list-archive:list-unsubscribe:list-id :precedence:subject:mime-version:references:in-reply-to:message-id :date:to:from:delivered-to; bh=QqOBSpyrFCOBf6dfm5RIS9+PreJ+0P8cw5O9WLkjiXU=; fh=YOA8vD9MJZuwZ71F/05pj6KdCjf6jQRmzLS+CATXUQk=; b=pV8wVnB0XdvCWXfC2YDM/pb5QE4cWbDcR3uFrP/BstYsaK0guLvl5QEsyaNRptK9p8 VZOvwRwuDzHuK74jVUXaUEOJqJ4g8lgjCPr7HeQBHxoCXmNnyDpLr4JwzlqtlsUYkqrk LwJSdn2RLXh/lzJg8etYMP6BUDqlCaxxbj30sf6Ex0+QYIQLwinA8c7RNInxfCf4TWIa yrXQ5xwa1d8zPxKU9g3FvfSsDG9MzrFCHNybg/Y16C6F/KACrJwIgOMwfW8L/DYBFm1Q e8PD3+yjgJH9l8f2+6Bw1ndqvqoFZ05C1c6Gu9se2v0jHvEDBgt65+GuND9vSBAuzpUT zyTA==; dara=google.com ARC-Authentication-Results: i=1; mx.google.com; spf=pass (google.com: domain of ffmpeg-devel-bounces@ffmpeg.org designates 79.124.17.100 as permitted sender) smtp.mailfrom=ffmpeg-devel-bounces@ffmpeg.org Return-Path: Received: from ffbox0-bg.mplayerhq.hu (ffbox0-bg.ffmpeg.org. [79.124.17.100]) by mx.google.com with ESMTP id a640c23a62f3a-a6f05ad16c3si133071466b.57.2024.06.09.02.27.45; Sun, 09 Jun 2024 02:27:45 -0700 (PDT) Received-SPF: pass (google.com: domain of ffmpeg-devel-bounces@ffmpeg.org designates 79.124.17.100 as permitted sender) client-ip=79.124.17.100; Authentication-Results: mx.google.com; spf=pass (google.com: domain of ffmpeg-devel-bounces@ffmpeg.org designates 79.124.17.100 as permitted sender) smtp.mailfrom=ffmpeg-devel-bounces@ffmpeg.org Received: from [127.0.1.1] (localhost [127.0.0.1]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTP id 0A80868D7AA; Sun, 9 Jun 2024 12:27:21 +0300 (EEST) X-Original-To: ffmpeg-devel@ffmpeg.org Delivered-To: ffmpeg-devel@ffmpeg.org Received: from ursule.remlab.net (vps-a2bccee9.vps.ovh.net [51.75.19.47]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTP id D65CA68D77C for ; Sun, 9 Jun 2024 12:27:10 +0300 (EEST) Received: from basile.remlab.net (localhost [IPv6:::1]) by ursule.remlab.net (Postfix) with ESMTP id 7CA65C02F9 for ; Sun, 9 Jun 2024 12:27:10 +0300 (EEST) From: =?utf-8?q?R=C3=A9mi_Denis-Courmont?= To: ffmpeg-devel@ffmpeg.org Date: Sun, 9 Jun 2024 12:27:09 +0300 Message-ID: <20240609092709.1356010-4-remi@remlab.net> X-Mailer: git-send-email 2.45.1 In-Reply-To: <20240609092709.1356010-1-remi@remlab.net> References: <20240609092709.1356010-1-remi@remlab.net> MIME-Version: 1.0 Subject: [FFmpeg-devel] [PATCH 4/4] lavc/h263dsp: R-V V dct_unquantize 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 Errors-To: ffmpeg-devel-bounces@ffmpeg.org Sender: "ffmpeg-devel" X-TUID: IiAQtUsjxZ6M T-Head C908: h263dsp.dct_unquantize_inter_c: 5.7 h263dsp.dct_unquantize_inter_rvv_i32: 2.7 h263dsp.dct_unquantize_intra_c: 5.7 h263dsp.dct_unquantize_intra_rvv_i32: 3.0 SpacemiT X60: h263dsp.dct_unquantize_inter_c: 5.0 h263dsp.dct_unquantize_inter_rvv_i32: 1.2 h263dsp.dct_unquantize_intra_c: 5.0 h263dsp.dct_unquantize_intra_rvv_i32: 1.5 --- libavcodec/riscv/h263dsp_init.c | 12 +++++++++--- libavcodec/riscv/h263dsp_rvv.S | 22 ++++++++++++++++++++++ 2 files changed, 31 insertions(+), 3 deletions(-) diff --git a/libavcodec/riscv/h263dsp_init.c b/libavcodec/riscv/h263dsp_init.c index 21b536366c..5d73fde865 100644 --- a/libavcodec/riscv/h263dsp_init.c +++ b/libavcodec/riscv/h263dsp_init.c @@ -25,6 +25,7 @@ #include "libavutil/riscv/cpu.h" #include "libavcodec/h263dsp.h" +void ff_h263_dct_unquantize_rvv(int16_t *, size_t start, size_t end, int, int); void ff_h263_h_loop_filter_rvv(uint8_t *src, int stride, int q); void ff_h263_v_loop_filter_rvv(uint8_t *src, int stride, int q); @@ -33,9 +34,14 @@ av_cold void ff_h263dsp_init_riscv(H263DSPContext *c) #if HAVE_RVV int flags = av_get_cpu_flags(); - if ((flags & AV_CPU_FLAG_RVV_I32) && ff_rv_vlen_least(128)) { - c->h263_h_loop_filter = ff_h263_h_loop_filter_rvv; - c->h263_v_loop_filter = ff_h263_v_loop_filter_rvv; + if (flags & AV_CPU_FLAG_RVV_I32) { + if (flags & AV_CPU_FLAG_RVB_ADDR) + c->h263_dct_unquantize = ff_h263_dct_unquantize_rvv; + + if (ff_rv_vlen_least(128)) { + c->h263_h_loop_filter = ff_h263_h_loop_filter_rvv; + c->h263_v_loop_filter = ff_h263_v_loop_filter_rvv; + } } #endif } diff --git a/libavcodec/riscv/h263dsp_rvv.S b/libavcodec/riscv/h263dsp_rvv.S index 97503d527c..319c51a0bb 100644 --- a/libavcodec/riscv/h263dsp_rvv.S +++ b/libavcodec/riscv/h263dsp_rvv.S @@ -20,6 +20,28 @@ #include "libavutil/riscv/asm.S" +func ff_h263_dct_unquantize_rvv, zve32x + sub a2, a2, a1 + sh1add a0, a1, a0 + addi a2, a2, 1 +1: + vsetvli t0, a2, e16, m4, ta, mu + vle16.v v8, (a0) + sub a2, a2, t0 + vmv.v.x v24, a4 + vmslt.vi v0, v8, 0 + vneg.v v24, v24, v0.t + vmsne.vi v0, v8, 0 + vwmul.vx v16, v8, a3 + vwadd.wv v16, v16, v24, v0.t + vncvt.x.x.w v8, v16 + vse16.v v8, (a0) + sh1add a0, t0, a0 + bnez a2, 1b + + ret +endfunc + .option push .option norelax func ff_h263_h_loop_filter_rvv, zve32x