From patchwork Mon May 13 16:59:25 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: uk7b@foxmail.com X-Patchwork-Id: 48864 Delivered-To: ffmpegpatchwork2@gmail.com Received: by 2002:a05:6a21:3a48:b0:1af:fc2d:ff5a with SMTP id zu8csp466273pzb; Mon, 13 May 2024 10:01:15 -0700 (PDT) X-Forwarded-Encrypted: i=2; AJvYcCU6/M6DadMdQuubOuFHvLpsbXXPVQ4qCryCFo8l2GcoH/DWSnbktHLqi3tbkkPVrmOWys7PwF6F8Cxzc5EO4RflDZywiwIHaafnbA== X-Google-Smtp-Source: AGHT+IFCJIXzUZ322W6PdDRWueQHi3VdjdnjsGpZiUhjsFmLVnORVhu1hWuFQxyF5jcoFSS/a4zG X-Received: by 2002:a17:906:1406:b0:a5a:7a4e:7e80 with SMTP id a640c23a62f3a-a5a7a4e7effmr82807766b.72.1715619675108; Mon, 13 May 2024 10:01:15 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1715619675; cv=none; d=google.com; s=arc-20160816; b=Vo1eg05xIrtGDGebs+azqQg3iW8iiyZbeXyYlzgGchie6LtnQ1YnlPSXd28dwUKCT8 jMrOcJlMRY/ZedXBWuBaDRiAEUs29G5x0UN1cz4iPWkxNjtuFp5bq86dPnucUMrVAQ/s 2G++IPk0xgErGlSchBxERCA95d964JZCZB5HQr3rf4rBKlBwL9gZdTVCEL13I8WCcDF9 qhtqcmBAXhxd8/fBZcxBong4LXeXpdlz3kVZgEMdIYikKw48qooyoyM/DiwsFg005OJO Nu7/PjiXAnDccU5XGaaqtYRPRFkJuL24H+QivM7LW9cJHqTKX2mu8ZFPwGJV/RH7Uv7Y +26Q== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=sender:errors-to:content-transfer-encoding:cc:reply-to :list-subscribe:list-help:list-post:list-archive:list-unsubscribe :list-id:precedence:subject:mime-version:references:in-reply-to:date :to:from:message-id:dkim-signature:delivered-to; bh=P5o6BQ02ZnB4f8yDn3tENV4D6pakcsbRTxJgmDOSqJI=; fh=D0bFwGkf4X22/D/bfeDVrXKIx7S6kcXsNzy10j8ORbQ=; b=il+yF/QclkXimdNrPltyKnu3r4TVrKqcjLd2CSyMG5uafI2FDho3PXpG/3PuApQOw/ 8WIgr6srE7ZMxTcZSpDD9yu+7PXcQFKt4hh9Obc8BqpAOKQlqZSSosYP/FI6DeFaXl4X dcqy2/S+WkwtzJ0SoF68SyBrBSMZlg72fGLAd8GKGzSASNjELsgbygJ1zId3zMHPloN9 +Hj4F20dz2xqJjmyH1NMqlMka5s8UKwCg1J+Pjmgi7luQiOgIe+ZtufvaB9xNBUcFJ5v YI7ECBKss51nE+xMJA6NHAue0KmCDvzWZU813tepb+BgkM/RJbgZ+zoq52fM9jcCMqmV K8Qg==; dara=google.com ARC-Authentication-Results: i=1; mx.google.com; dkim=neutral (body hash did not verify) header.i=@foxmail.com header.s=s201512 header.b=L3qycfVi; 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; dmarc=fail (p=NONE sp=NONE dis=NONE) header.from=foxmail.com Return-Path: Received: from ffbox0-bg.mplayerhq.hu (ffbox0-bg.ffmpeg.org. [79.124.17.100]) by mx.google.com with ESMTP id a640c23a62f3a-a5a17c2d660si530524466b.1040.2024.05.13.10.01.14; Mon, 13 May 2024 10:01:15 -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; dkim=neutral (body hash did not verify) header.i=@foxmail.com header.s=s201512 header.b=L3qycfVi; 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; dmarc=fail (p=NONE sp=NONE dis=NONE) header.from=foxmail.com Received: from [127.0.1.1] (localhost [127.0.0.1]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTP id 2013A68D6F6; Mon, 13 May 2024 20:00:04 +0300 (EEST) X-Original-To: ffmpeg-devel@ffmpeg.org Delivered-To: ffmpeg-devel@ffmpeg.org Received: from out203-205-251-27.mail.qq.com (out203-205-251-27.mail.qq.com [203.205.251.27]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTPS id 0997668D68E for ; Mon, 13 May 2024 19:59:52 +0300 (EEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=foxmail.com; s=s201512; t=1715619583; bh=L1gMSaWNd94/n3lU2iNGhyMXkQy2/Tomn529g9H6IMU=; h=From:To:Cc:Subject:Date:In-Reply-To:References; b=L3qycfViyhf2jSYEZzKi6dKBqqQZv/+RBw5hH3lSd+lIXNm6d35AtoiixhOWkQRU6 7RLPeAynKSA2XNHfS45SFOnxnWbEqBCAPHREUzU+0aNkVzK+LlokL6zMQk0fCGVo/m JUYNwAvUlPFTzjKgTH6RGPMmIWqfvwy09gT+V8Mc= Received: from localhost.localdomain ([42.56.223.122]) by newxmesmtplogicsvrszb9-0.qq.com (NewEsmtp) with SMTP id EE3A4485; Tue, 14 May 2024 00:59:35 +0800 X-QQ-mid: xmsmtpt1715619582tfw2lbvkr Message-ID: X-QQ-XMAILINFO: Nz4vUwhpFJ9NMj0nGmf7IM5ZmcvIE/TLufuvsMQi5+miHA+xAHCT0PUbWj1X8o WySugUkl8hZzi/11Z91qGudl052Cp9LAJzGjxOJ8piInDLiIWa0VPOUnGmMg0zPf/JBAAoXircQB 0ICo01A4ecN60RT1MnOAfaGcAWtsicGXA2Ok9uzPZXF+WBG7y1P+1z7uIHAAH9Z6NtCGC07l8c/M 0v3/XaQ3Rnha+IdLW/EHKrL1YuKMExiYbdlwhj289kSZ8puTFhuieVw32Kw/lgLaEvG4bfeG0zf6 5PUuXFLUAu+xVCBgDSlP0zMld95H+es+g5Jz0K+dLjmrmWEEI4wQZtnEOkvmHcaP1PlUoI2mGZZ0 ssTRm8IKSjEGMgYPnA0jFYkbX3673+c8h3tzVT/moYEqnURDX1ZTQmgiSuuq6P8JJ7o3f+99lLBM budRx3+BuK4XQkr37C3X4aXhjNqwyXmjqiZA/Pr4RZYjBxy1hXgVBLRq2a8pF56NQzcUkgbmtS/W 4ZwlQJoyGi/mez1dOhQQZJF/8PdBy3xlytZ8WVkT8sUqj+re9B6TRYIyiv54aw07a65HTgvBwLdJ bCUadmFi6vUuEbxi1ItbneGQbWwfBhaukmltYqGbHh95zf0pNvVcxLB/Ylg7wn94MMb87h1Ek+zT eSK+hm85UyJ/KuxFLFgBjuTh4LUrhYOMYk+Hk1ggg1MCK2iXWG0Qh/Pia2yVD0wPDjrmk1Er8pGh yBmomjtkJMXVieTpbHD33l4zcnt9dSTPUk3WwrbdSSOKHcH0HLJC6FrF18SUYfE02hkwTeXByapU k0Lr3mFfQdC1H7Jy6nVv/vLylrd+f4pWGid384SPWY8uBmeeFqCfMcBj2JVbmsXHr3TSZsCqdrKG oSmsfxHC08XKwYo4ZToZruoHbzct8fvFazqAhu+b7SAqBzIjn6Jeo= X-QQ-XMRINFO: MSVp+SPm3vtS1Vd6Y4Mggwc= From: uk7b@foxmail.com To: ffmpeg-devel@ffmpeg.org Date: Tue, 14 May 2024 00:59:25 +0800 X-OQ-MSGID: <20240513165926.1467967-8-uk7b@foxmail.com> X-Mailer: git-send-email 2.45.0 In-Reply-To: <20240513165926.1467967-1-uk7b@foxmail.com> References: <20240513165926.1467967-1-uk7b@foxmail.com> MIME-Version: 1.0 Subject: [FFmpeg-devel] [PATCH v3 8/9] lavc/vp9dsp: R-V V mc bilin hv 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 Cc: sunyuechi Errors-To: ffmpeg-devel-bounces@ffmpeg.org Sender: "ffmpeg-devel" X-TUID: ubo/3QgBpEOE From: sunyuechi C908: vp9_avg_bilin_4hv_8bpp_c: 11.0 vp9_avg_bilin_4hv_8bpp_rvv_i64: 3.7 vp9_avg_bilin_8hv_8bpp_c: 38.7 vp9_avg_bilin_8hv_8bpp_rvv_i64: 7.2 vp9_avg_bilin_16hv_8bpp_c: 147.0 vp9_avg_bilin_16hv_8bpp_rvv_i64: 14.2 vp9_avg_bilin_32hv_8bpp_c: 574.5 vp9_avg_bilin_32hv_8bpp_rvv_i64: 42.7 vp9_avg_bilin_64hv_8bpp_c: 2311.5 vp9_avg_bilin_64hv_8bpp_rvv_i64: 201.7 vp9_put_bilin_4hv_8bpp_c: 10.0 vp9_put_bilin_4hv_8bpp_rvv_i64: 3.2 vp9_put_bilin_8hv_8bpp_c: 35.2 vp9_put_bilin_8hv_8bpp_rvv_i64: 6.5 vp9_put_bilin_16hv_8bpp_c: 133.7 vp9_put_bilin_16hv_8bpp_rvv_i64: 13.0 vp9_put_bilin_32hv_8bpp_c: 538.2 vp9_put_bilin_32hv_8bpp_rvv_i64: 39.7 vp9_put_bilin_64hv_8bpp_c: 2114.0 vp9_put_bilin_64hv_8bpp_rvv_i64: 153.7 --- libavcodec/riscv/vp9_mc_rvv.S | 34 ++++++++++++++++++++++++++++++++++ libavcodec/riscv/vp9dsp_init.c | 10 ++++++++++ 2 files changed, 44 insertions(+) diff --git a/libavcodec/riscv/vp9_mc_rvv.S b/libavcodec/riscv/vp9_mc_rvv.S index c633809675..22ae194367 100644 --- a/libavcodec/riscv/vp9_mc_rvv.S +++ b/libavcodec/riscv/vp9_mc_rvv.S @@ -104,6 +104,39 @@ func ff_\op\()_bilin_\len\()\type\()_rvv, zve32x endfunc .endm +.macro bilin_hv len op +func ff_\op\()_bilin_\len\()hv_rvv, zve32x +.ifc \op,avg + csrwi vxrm, 0 +.endif + vsetvlstatic8 \len t0 64 + neg t1, a5 + neg t2, a6 + li t4, 8 + bilin_load v24, \len, put, h, a5 + add a2, a2, a3 +1: + addi a4, a4, -1 + bilin_load v4, \len, put, h, a5 + vwmulu.vx v16, v4, a6 + vwmaccsu.vx v16, t2, v24 + vwadd.wx v16, v16, t4 + vnsra.wi v16, v16, 4 + vadd.vv v0, v16, v24 +.ifc \op,avg + vle8.v v16, (a0) + vaaddu.vv v0, v0, v16 +.endif + vse8.v v0, (a0) + vmv.v.v v24, v4 + add a2, a2, a3 + add a0, a0, a1 + bnez a4, 1b + + ret +endfunc +.endm + const subpel_filters_regular .byte 0, 0, 0, 128, 0, 0, 0, 0 .byte 0, 1, -5, 126, 8, -3, 1, 0 @@ -334,6 +367,7 @@ endfunc .irp op put avg bilin_h_v \len \op h a5 bilin_h_v \len \op v a6 + bilin_hv \len \op .irp name regular sharp smooth .irp type h v epel \len \op \name \type 128 diff --git a/libavcodec/riscv/vp9dsp_init.c b/libavcodec/riscv/vp9dsp_init.c index c78d22a7f3..f3e9302a73 100644 --- a/libavcodec/riscv/vp9dsp_init.c +++ b/libavcodec/riscv/vp9dsp_init.c @@ -104,6 +104,16 @@ static av_cold void vp9dsp_mc_init_riscv(VP9DSPContext *dsp, int bpp) dsp->mc[4][FILTER_BILINEAR ][0][1][0] = ff_put_bilin_4h_rvv; dsp->mc[4][FILTER_BILINEAR ][1][0][1] = ff_avg_bilin_4v_rvv; dsp->mc[4][FILTER_BILINEAR ][1][1][0] = ff_avg_bilin_4h_rvv; + dsp->mc[0][FILTER_BILINEAR ][0][1][1] = ff_put_bilin_64hv_rvv; + dsp->mc[0][FILTER_BILINEAR ][1][1][1] = ff_avg_bilin_64hv_rvv; + dsp->mc[1][FILTER_BILINEAR ][0][1][1] = ff_put_bilin_32hv_rvv; + dsp->mc[1][FILTER_BILINEAR ][1][1][1] = ff_avg_bilin_32hv_rvv; + dsp->mc[2][FILTER_BILINEAR ][0][1][1] = ff_put_bilin_16hv_rvv; + dsp->mc[2][FILTER_BILINEAR ][1][1][1] = ff_avg_bilin_16hv_rvv; + dsp->mc[3][FILTER_BILINEAR ][0][1][1] = ff_put_bilin_8hv_rvv; + dsp->mc[3][FILTER_BILINEAR ][1][1][1] = ff_avg_bilin_8hv_rvv; + dsp->mc[4][FILTER_BILINEAR ][0][1][1] = ff_put_bilin_4hv_rvv; + dsp->mc[4][FILTER_BILINEAR ][1][1][1] = ff_avg_bilin_4hv_rvv; if (flags & AV_CPU_FLAG_RVB_ADDR) { init_subpel2(0, 0, 1, v, put, 128);