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 64E8C4A137 for ; Fri, 22 Mar 2024 13:08:17 +0000 (UTC) Received: from [127.0.1.1] (localhost [127.0.0.1]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTP id 9265D68D583; Fri, 22 Mar 2024 15:08:14 +0200 (EET) Received: from haasn.dev (haasn.dev [78.46.187.166]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTP id 8E4BF68D512 for ; Fri, 22 Mar 2024 15:08:07 +0200 (EET) DKIM-Signature: v=1; a=rsa-sha256; c=simple/simple; d=haasn.xyz; s=mail; t=1711112887; bh=F4rwHyGIJyl2Vlp/ofzsbBun2r15PKso6T/Tjwrjtts=; h=Date:From:To:Cc:Subject:In-Reply-To:References:From; b=IOsZqSbE5GAt6q0rM8vsBvbzyHW2ZU0fDhkbaCjbLOQKB/QJWSoBfHZ2UMNIV83f8 e3jbczEGoqJnrXankcUdD///onwZS2WltWBDNvQiokHtE5YF4DngLu7RpbBcwlxpMx FfZuhWOYmhgvFA8xt0aZLOhR7MfxJzzuFO9aQeeU= Received: from haasn.dev (unknown [10.30.0.2]) by haasn.dev (Postfix) with ESMTP id 50D65418F4; Fri, 22 Mar 2024 14:08:07 +0100 (CET) Date: Fri, 22 Mar 2024 14:08:07 +0100 Message-ID: <20240322140807.GB34745@haasn.xyz> From: Niklas Haas To: FFmpeg development discussions and patches In-Reply-To: <171110047375.7287.3672479821689972140@lain.khirnov.net> References: <20240319191642.95217-1-ffmpeg@haasn.xyz> <171101621793.7287.2859730890037554804@lain.khirnov.net> <20240321131132.GB15980@haasn.xyz> <171110047375.7287.3672479821689972140@lain.khirnov.net> MIME-Version: 1.0 Content-Disposition: inline Subject: Re: [FFmpeg-devel] [PATCH 1/4] fftools/ffmpeg_enc: strip DOVI config record for 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: , Reply-To: FFmpeg development discussions and patches Cc: Niklas Haas 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: On Fri, 22 Mar 2024 10:41:13 +0100 Anton Khirnov wrote: > Quoting Niklas Haas (2024-03-21 13:11:32) > > On Thu, 21 Mar 2024 11:16:57 +0100 Anton Khirnov wrote: > > > Quoting Niklas Haas (2024-03-19 20:16:39) > > > > From: Niklas Haas > > > > > > > > AV1 streams don't use configuration records, so delete them when > > > > encoding to AV1. Ideally this would be, as the comment suggests, handled > > > > at the frame-level (and stripped by the av1 encoder), but given the > > > > status quo of copying the packet-level data here directly, we should > > > > definitely make an effort to strip it. > > > > --- > > > > fftools/ffmpeg_enc.c | 25 ++++++++++++++----------- > > > > 1 file changed, 14 insertions(+), 11 deletions(-) > > > > > > I'm very much not a fan of having codec-specific code in ffmpeg CLI. It > > > implies that every single caller must now be aware of this > > > (undocumented?) interaction of this specific side data with this > > > specific codec ID. > > > > Note: This is an existing bug, not introduced by this series. This > > series just makes it obvious. The status quo is that, beacuse of this > > logic in ffmpeg_enc.c, we incorrectly forward dolby vision configuration > > records when transcoding to AV1. > > I know pretty much nothing about dolby vision, so could you please > explain why precisely is this incorrect? And at what point in the > transcoding chain does the side data become invalid? Dolby Vision basically consists of two separate pieces of metadata: 1. The (per-stream) configuration struct, AV_PKT_DATA_DOVI_CONF 2. The per-frame structs (RPUs), AV_FRAME_DATA_DOVI_METADATA (ditto AV_FRAME_DATA_DOVI_RPU_BUFFER, which is the same) A valid HEVC dolby vision file should contain both - the configuration struct tells the decoder that hey, this file is dolby vision (and what profile to expect, whether there's an enhancement layer, etc.). The RPUs contain the actual DV-specific details of how each frame is encoded. A valid AV1 dolby vision file, on the other hand, only uses the per-frame RPUs, it does not have a configuration struct at all. The current logic in ffmpeg_enc.c copies over all stream-level metadata, including the DOVI_CONF struct, to the output file. This generates a stream which is *marked* as being Dolby Vision, but in which none of the frames actually contain DV RPUs. This *probably* violates some spec somewhere, and at the very least is not desirable behavior. (And for AV1, the configuration struct's presence is definitely a no-go) Basically, we want to handle all of these scenarios: 1. When transcoding DV HEVC (profile 8) to DV AV1 (profile 10), we need to strip the configuration struct somewhere 2. When transcoding DV HEVC (profile 8) to HEVC, we need to strip the configuration struct IFF we're also stripping the per-frame RPUs (e.g. as a result of filtering). 3. When transcoding DV AV1 (profile 10) to HEVC, we need to *synthesize* a configuration struct containing the correct values. I think the best way forward for now is: 1. Always strip the dovi configuration record when transcoding 2. Have the encoder generate (and attach to avctx.coded_side_data) the correct configuration record. I will write a patch for #2. > > Or, indeed, when transcoding to *any* format - since current FFmpeg also > > does not propagate dolby vision RPUs, we generate broken files pretty > > much always when transcoding dolby vision. So we definitely need to > > strip the metadata from the stream muxer *somewhere*. Where else comes > > to mind? > > > > This also gets into another topic I wanted to touch on, which is that > > the presence of dynamic dolby vision metadata currently hinders the > > ability of libavfilter to treat the video primaries/gamma as > > a negotiable colorspace property (the way it is done currently for YUV > > matrix/range). This is because when interpreted as such, DV metadata > > fundamentally changes the colorspace of the incoming video stream. > > Ideally we would like some way to negotiate DV metadata on the > > query_formats() level. > > > > Ideally, we'd want something like AVCOL_SPC_DOLBYVISION, but we can't > > easily introduce that without breaking ISO/IEC 23091 compatibility.. > > In principle it could be yet another negotiated field, could it not? You > just added a bunch of those recently, what's another one? Adding more fields to this negotiation process is a very obnoxious and tedious process, with lots of boilerplate for each new field added. Maybe we can come up with some better mechanism first? _______________________________________________ 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".