From patchwork Thu Apr 11 11:58:44 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Diego Felix de Souza via ffmpeg-devel X-Patchwork-Id: 48009 Delivered-To: ffmpegpatchwork2@gmail.com Received: by 2002:a05:6a20:c90a:b0:1a7:a0dc:8de5 with SMTP id gx10csp235801pzb; Thu, 11 Apr 2024 04:59:58 -0700 (PDT) X-Forwarded-Encrypted: i=2; AJvYcCVkwUMQy/rvkrSbuGd0aM/2o14VqsqfsAJeWj9hiM7mYBULp3aa0UYrsubdGlCllSnsvp9Dsgb361PyaVZ2CGo2dvfLKCSuHxbULw== X-Google-Smtp-Source: AGHT+IG87xC8l8IleQnRvhmL6akG/2Xp2Wj8/OROkJfZqPP6AjtOPzstsNb4Iwimza/tL5BgNNCS X-Received: by 2002:a17:906:1519:b0:a52:292:31eb with SMTP id b25-20020a170906151900b00a52029231ebmr3476878ejd.4.1712836798086; Thu, 11 Apr 2024 04:59:58 -0700 (PDT) Return-Path: Received: from ffbox0-bg.mplayerhq.hu (ffbox0-bg.ffmpeg.org. [79.124.17.100]) by mx.google.com with ESMTP id o11-20020a1709061b0b00b00a4e5b3738fesi683141ejg.896.2024.04.11.04.59.56; Thu, 11 Apr 2024 04:59:58 -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; arc=fail (body hash mismatch); 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 57F2568D164; Thu, 11 Apr 2024 14:59:53 +0300 (EEST) X-Original-To: ffmpeg-devel@ffmpeg.org Delivered-To: ffmpeg-devel@ffmpeg.org Received: from NAM10-BN7-obe.outbound.protection.outlook.com (mail-bn7nam10on2060.outbound.protection.outlook.com [40.107.92.60]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTPS id B29CD68CD68 for ; Thu, 11 Apr 2024 14:59:45 +0300 (EEST) ARC-Seal: i=1; a=rsa-sha256; s=arcselector9901; d=microsoft.com; cv=none; b=Uf8x4V+0El1YI9+HwMISfExDcJkZZJz/5BlRR/AOLClO8GECDMIvYGjRqE9g8jNbiiiHp8RdSnaBdhEbW4Ph5XVtbf/eRlM1JHX2KASTV+KxPvAjSuNLwpS/AhwobUPzdBljqkS5Cm+4TBAmtKiNQOyEDcpDcALhTkcB5oGqyWxKQBWRfH3oDOf24MsxUMOYlRyQHuB/je2YttV8fldSognZ7VhaPj5dLQbjhY+GWx6CPOpLV1LMPu6MTrv3zXDXDhI8BfXJ9as8E6yeYbsHVbwpaStXMpFBJqVD0VN9TvMFxSzUwvWIkqG1b29yKhakeKxgrZwvKTqXTml8ADCnOg== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=microsoft.com; s=arcselector9901; h=From:Date:Subject:Message-ID:Content-Type:MIME-Version:X-MS-Exchange-AntiSpam-MessageData-ChunkCount:X-MS-Exchange-AntiSpam-MessageData-0:X-MS-Exchange-AntiSpam-MessageData-1; bh=UK2Dtm3t7utsYA8Atmn7kE34OPlYamc5iMwuVxjEvHU=; b=EScQDko/h3uNalLVq1pUTLrVEStSb9GXv9Oqk4z7JTu/AeoHDtX2xrlC7pKmDYBgpvFzPudtPQHiM8jAmFDkRAmZjs5Dg74B5UhfmtE2OTUcb7KltkUq8whXTp/B/ygKpAUqcTf0RsHezoiaLYwwcXDI2ZCG2RXg+xY++wLWLUIdvdyrkkyUwSg5YVz5MOxSWFHbZlKO/iVx5DfUEBcfBLnAxTqfP5Z182zQW9gP3RvqexA6xyxuGwjKkJo5UJD0ZqdKg76em3jMk0n1LrtpH4/FhqstDIQwAOWzKd1Hx60sQLlZzQMx5R2Wk8yQjw95M+tQwfFARw0qWUq2bZ93zg== ARC-Authentication-Results: i=1; mx.microsoft.com 1; spf=pass (sender ip is 216.228.117.160) smtp.rcpttodomain=ffmpeg.org smtp.mailfrom=nvidia.com; dmarc=pass (p=reject sp=reject pct=100) action=none header.from=nvidia.com; dkim=none (message not signed); arc=none (0) Received: from SJ0PR13CA0011.namprd13.prod.outlook.com (2603:10b6:a03:2c0::16) by SN7PR12MB7978.namprd12.prod.outlook.com (2603:10b6:806:34b::9) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.20.7409.55; Thu, 11 Apr 2024 11:59:40 +0000 Received: from MWH0EPF000989E8.namprd02.prod.outlook.com (2603:10b6:a03:2c0:cafe::e) by SJ0PR13CA0011.outlook.office365.com (2603:10b6:a03:2c0::16) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.20.7495.7 via Frontend Transport; Thu, 11 Apr 2024 11:59:40 +0000 X-MS-Exchange-Authentication-Results: spf=pass (sender IP is 216.228.117.160) smtp.mailfrom=nvidia.com; dkim=none (message not signed) header.d=none;dmarc=pass action=none header.from=nvidia.com; Received-SPF: Pass (protection.outlook.com: domain of nvidia.com designates 216.228.117.160 as permitted sender) receiver=protection.outlook.com; client-ip=216.228.117.160; helo=mail.nvidia.com; pr=C Received: from mail.nvidia.com (216.228.117.160) by MWH0EPF000989E8.mail.protection.outlook.com (10.167.241.135) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.20.7452.22 via Frontend Transport; Thu, 11 Apr 2024 11:59:39 +0000 Received: from rnnvmail201.nvidia.com (10.129.68.8) by mail.nvidia.com (10.129.200.66) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.2.986.41; Thu, 11 Apr 2024 04:59:20 -0700 Received: from nvidia.com (10.126.231.35) by rnnvmail201.nvidia.com (10.129.68.8) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.2.1258.12; Thu, 11 Apr 2024 04:59:19 -0700 To: Date: Thu, 11 Apr 2024 11:58:44 +0000 Message-ID: <20240411115844.290887-1-ddesouza@nvidia.com> X-Mailer: git-send-email 2.34.1 MIME-Version: 1.0 X-Originating-IP: [10.126.231.35] X-ClientProxiedBy: rnnvmail203.nvidia.com (10.129.68.9) To rnnvmail201.nvidia.com (10.129.68.8) X-EOPAttributedMessage: 0 X-MS-PublicTrafficType: Email X-MS-TrafficTypeDiagnostic: MWH0EPF000989E8:EE_|SN7PR12MB7978:EE_ X-MS-Office365-Filtering-Correlation-Id: 120a238b-0cb4-42ea-0ca6-08dc5a1eddce X-MS-Exchange-SenderADCheck: 1 X-MS-Exchange-AntiSpam-Relay: 0 X-Microsoft-Antispam: BCL:0; X-Microsoft-Antispam-Message-Info: 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 X-Forefront-Antispam-Report: CIP:216.228.117.160; CTRY:US; LANG:en; SCL:1; SRV:; IPV:NLI; SFV:NSPM; H:mail.nvidia.com; PTR:dc6edge1.nvidia.com; CAT:NONE; SFS:(13230031)(36860700004)(1800799015)(82310400014)(376005); DIR:OUT; SFP:1101; X-OriginatorOrg: Nvidia.com X-MS-Exchange-CrossTenant-OriginalArrivalTime: 11 Apr 2024 11:59:39.5148 (UTC) X-MS-Exchange-CrossTenant-Network-Message-Id: 120a238b-0cb4-42ea-0ca6-08dc5a1eddce X-MS-Exchange-CrossTenant-Id: 43083d15-7273-40c1-b7db-39efd9ccc17a X-MS-Exchange-CrossTenant-OriginalAttributedTenantConnectingIp: TenantId=43083d15-7273-40c1-b7db-39efd9ccc17a; Ip=[216.228.117.160]; Helo=[mail.nvidia.com] X-MS-Exchange-CrossTenant-AuthSource: MWH0EPF000989E8.namprd02.prod.outlook.com X-MS-Exchange-CrossTenant-AuthAs: Anonymous X-MS-Exchange-CrossTenant-FromEntityHeader: HybridOnPrem X-MS-Exchange-Transport-CrossTenantHeadersStamped: SN7PR12MB7978 Subject: [FFmpeg-devel] [PATCH] Multi NVENC Split Frame Encoding in HEVC and AV1 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: , X-Patchwork-Original-From: Diego Felix de Souza via ffmpeg-devel From: Diego Felix de Souza via ffmpeg-devel Reply-To: FFmpeg development discussions and patches Cc: ddesouza@nvidia.com Errors-To: ffmpeg-devel-bounces@ffmpeg.org Sender: "ffmpeg-devel" X-TUID: dyel9npYZPAB From: Diego Felix de Souza When Split frame encoding is enabled, each input frame is partitioned into horizontal strips which are encoded independently and simultaneously by separate NVENCs, usually resulting in increased encoding speed compared to single NVENC encoding. --- libavcodec/nvenc.c | 16 ++++++++++++++++ libavcodec/nvenc.h | 2 ++ libavcodec/nvenc_av1.c | 8 ++++++++ libavcodec/nvenc_hevc.c | 8 ++++++++ 4 files changed, 34 insertions(+) -- 2.34.1 ----------------------------------------------------------------------------------- NVIDIA GmbH Wuerselen Amtsgericht Aachen HRB 8361 Managing Directors: Rebecca Peters, Donald Robertson, Janet Hall, Ludwig von Reiche ----------------------------------------------------------------------------------- This email message is for the sole use of the intended recipient(s) and may contain confidential information. Any unauthorized review, use, disclosure or distribution is prohibited. If you are not the intended recipient, please contact the sender by reply email and destroy all copies of the original message. ----------------------------------------------------------------------------------- diff --git a/libavcodec/nvenc.c b/libavcodec/nvenc.c index b6c5ed3e6b..f4d0d21715 100644 --- a/libavcodec/nvenc.c +++ b/libavcodec/nvenc.c @@ -1696,6 +1696,22 @@ FF_ENABLE_DEPRECATION_WARNINGS if (ctx->weighted_pred == 1) ctx->init_encode_params.enableWeightedPrediction = 1; +#ifdef NVENC_HAVE_SPLIT_FRAME_ENCODING + if (avctx->codec->id != AV_CODEC_ID_H264 ) + ctx->init_encode_params.splitEncodeMode = ctx->split_encode_mode; + + if ((ctx->split_encode_mode != NV_ENC_SPLIT_DISABLE_MODE) && + ((ctx->weighted_pred == 1) && (avctx->codec->id == AV_CODEC_ID_HEVC ))) { + av_log(avctx, AV_LOG_WARNING, "Split encoding is not " + "supported if any of the following features: weighted prediction, " + "alpha layer encoding, subframe mode, output into video memory " + "buffer, picture timing/buffering period SEI message insertion " + "with DX12 interface are enabled in case of HEVC. For AV1, split " + "encoding is not supported when output into video memory buffer " + "is enabled.\n"); + } +#endif + if (ctx->bluray_compat) { ctx->aud = 1; ctx->dpb_size = FFMIN(FFMAX(avctx->refs, 0), 6); diff --git a/libavcodec/nvenc.h b/libavcodec/nvenc.h index 85ecaf1b5f..09de00badc 100644 --- a/libavcodec/nvenc.h +++ b/libavcodec/nvenc.h @@ -81,6 +81,7 @@ typedef void ID3D11Device; // SDK 12.1 compile time feature checks #if NVENCAPI_CHECK_VERSION(12, 1) #define NVENC_NO_DEPRECATED_RC +#define NVENC_HAVE_SPLIT_FRAME_ENCODING #endif // SDK 12.2 compile time feature checks @@ -280,6 +281,7 @@ typedef struct NvencContext int tf_level; int lookahead_level; int unidir_b; + int split_encode_mode; } NvencContext; int ff_nvenc_encode_init(AVCodecContext *avctx); diff --git a/libavcodec/nvenc_av1.c b/libavcodec/nvenc_av1.c index d37ee07bff..45dc3c26e0 100644 --- a/libavcodec/nvenc_av1.c +++ b/libavcodec/nvenc_av1.c @@ -157,6 +157,14 @@ static const AVOption options[] = { { "1", "", 0, AV_OPT_TYPE_CONST, { .i64 = NV_ENC_LOOKAHEAD_LEVEL_1 }, 0, 0, VE, .unit = "lookahead_level" }, { "2", "", 0, AV_OPT_TYPE_CONST, { .i64 = NV_ENC_LOOKAHEAD_LEVEL_2 }, 0, 0, VE, .unit = "lookahead_level" }, { "3", "", 0, AV_OPT_TYPE_CONST, { .i64 = NV_ENC_LOOKAHEAD_LEVEL_3 }, 0, 0, VE, .unit = "lookahead_level" }, +#endif +#ifdef NVENC_HAVE_SPLIT_FRAME_ENCODING + { "split_encode_mode", "Specifies the split encoding mode", OFFSET(split_encode_mode), AV_OPT_TYPE_INT, { .i64 = NV_ENC_SPLIT_DISABLE_MODE }, 0, NV_ENC_SPLIT_DISABLE_MODE, VE, .unit = "split_encode_mode" }, + { "disabled", "Disabled for all configurations", 0, AV_OPT_TYPE_CONST, { .i64 = NV_ENC_SPLIT_DISABLE_MODE }, 0, 0, VE, .unit = "split_encode_mode" }, + { "auto", "Enabled or disabled depending on the preset and tuning info", 0, AV_OPT_TYPE_CONST, { .i64 = NV_ENC_SPLIT_AUTO_MODE }, 0, 0, VE, .unit = "split_encode_mode" }, + { "forced", "Enabled with number of horizontal strips selected by the driver", 0, AV_OPT_TYPE_CONST, { .i64 = NV_ENC_SPLIT_AUTO_FORCED_MODE }, 0, 0, VE, .unit = "split_encode_mode" }, + { "2", "Enabled with number of horizontal strips forced to 2 when number of NVENCs > 1", 0, AV_OPT_TYPE_CONST, { .i64 = NV_ENC_SPLIT_TWO_FORCED_MODE }, 0, 0, VE, .unit = "split_encode_mode" }, + { "3", "Enabled with number of horizontal strips forced to 3 when number of NVENCs > 2", 0, AV_OPT_TYPE_CONST, { .i64 = NV_ENC_SPLIT_THREE_FORCED_MODE }, 0, 0, VE, .unit = "split_encode_mode" }, #endif { NULL } }; diff --git a/libavcodec/nvenc_hevc.c b/libavcodec/nvenc_hevc.c index bd8b6153f3..1f5e56ecd0 100644 --- a/libavcodec/nvenc_hevc.c +++ b/libavcodec/nvenc_hevc.c @@ -216,6 +216,14 @@ static const AVOption options[] = { #endif #ifdef NVENC_HAVE_UNIDIR_B { "unidir_b", "Enable use of unidirectional B-Frames.", OFFSET(unidir_b), AV_OPT_TYPE_BOOL, { .i64 = 0 }, 0, 1, VE }, +#endif +#ifdef NVENC_HAVE_SPLIT_FRAME_ENCODING + { "split_encode_mode", "Specifies the split encoding mode", OFFSET(split_encode_mode), AV_OPT_TYPE_INT, { .i64 = NV_ENC_SPLIT_DISABLE_MODE }, 0, NV_ENC_SPLIT_DISABLE_MODE, VE, .unit = "split_encode_mode" }, + { "disabled", "Disabled for all configurations", 0, AV_OPT_TYPE_CONST, { .i64 = NV_ENC_SPLIT_DISABLE_MODE }, 0, 0, VE, .unit = "split_encode_mode" }, + { "auto", "Enabled or disabled depending on the preset and tuning info", 0, AV_OPT_TYPE_CONST, { .i64 = NV_ENC_SPLIT_AUTO_MODE }, 0, 0, VE, .unit = "split_encode_mode" }, + { "forced", "Enabled with number of horizontal strips selected by the driver", 0, AV_OPT_TYPE_CONST, { .i64 = NV_ENC_SPLIT_AUTO_FORCED_MODE }, 0, 0, VE, .unit = "split_encode_mode" }, + { "2", "Enabled with number of horizontal strips forced to 2 when number of NVENCs > 1", 0, AV_OPT_TYPE_CONST, { .i64 = NV_ENC_SPLIT_TWO_FORCED_MODE }, 0, 0, VE, .unit = "split_encode_mode" }, + { "3", "Enabled with number of horizontal strips forced to 3 when number of NVENCs > 2", 0, AV_OPT_TYPE_CONST, { .i64 = NV_ENC_SPLIT_THREE_FORCED_MODE }, 0, 0, VE, .unit = "split_encode_mode" }, #endif { NULL } };