From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 3D05AC433EF for ; Wed, 6 Apr 2022 20:19:58 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S234569AbiDFUVv (ORCPT ); Wed, 6 Apr 2022 16:21:51 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:36386 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S232729AbiDFUVJ (ORCPT ); Wed, 6 Apr 2022 16:21:09 -0400 Received: from mail-qt1-x834.google.com (mail-qt1-x834.google.com [IPv6:2607:f8b0:4864:20::834]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 50B3C515AF for ; Wed, 6 Apr 2022 10:50:05 -0700 (PDT) Received: by mail-qt1-x834.google.com with SMTP id s7so5664821qtk.6 for ; Wed, 06 Apr 2022 10:50:05 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=ndufresne-ca.20210112.gappssmtp.com; s=20210112; h=message-id:subject:from:to:cc:date:in-reply-to:references :content-transfer-encoding:user-agent:mime-version; bh=nbgYNk5Esw76+oSTzku5Iq0RBseYj50R251OC+s1Zko=; b=oE77XjhZJdBZs73vBmWDSWOcpU+0g2TAF7dqKJsm0FrsMjk1RH5VR1BlQ5zyc+Ew0E YAyZ7/3GSXX52HLi4w0ZLcWVx8WEfp2H70UxOLZhb4x9Tw8uCMDrbM5d5FHn6SM8+gwO oJK9Mz7nWH9ol3Rrrfkwe3OikbspOO3ypvUVHlstZXQHmen/ZBRY15meUrtAA7c3OsiH QTmrE/4Q7sxCk3fFXjAd0HZnALSONcplj2n9nhPVNVpiCO2IvqkTQM6QBRqJh5D4ZlDw eJ5RkVM+Vz9/V7oOKSfMoNmKqTRPPtOn7EnIojoHf3XYwhNUGCPXlmEsvVhrbnUTWod8 nu/g== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:message-id:subject:from:to:cc:date:in-reply-to :references:content-transfer-encoding:user-agent:mime-version; bh=nbgYNk5Esw76+oSTzku5Iq0RBseYj50R251OC+s1Zko=; b=QRgHKWEq1yqW8eMDbFB3qkJUMKka9cXxNLtOnx84NBcNhgBUfnp+0pLaiDyVZPnSwf sgUiUBWUHYhsulEVT9IV/Q2mBI23JFotUy8ucYAH4n5yFv3Y3NBdXEAMfJtHLwEp0IOf SFsabpSHSEy7WL+DpgckS7Pjhrd2LiLM9yMYZXmRctxSg0syfp5KEodDq00fTGhDcgGL jeJwnljjjcUZZ3ewozavn+KLFOmAbMPmp0SaDOfg70NWBS1DMB9a46zn3QOefcxxyuhP 5ITvRI9wwJz/jsY4UknN5HbL9EHUJVsRGct81NBbm/iwbrR/k5AB48x/LV/UoefLwWA6 3ngg== X-Gm-Message-State: AOAM530nfwGKwxoD5OMn2xvzBiRmMZcRK6brqK7usRSHIHcED5DoirD8 a3LBYPwnomgaF9w3Sp4ACrCGGg== X-Google-Smtp-Source: ABdhPJylSRQCASHDcc/QNJp7o7BwkgXxThlNUItggDkKtsHMdcE9vZQxW5pxT61gwxRLUSzl83zB0g== X-Received: by 2002:ac8:5a8f:0:b0:2e1:df21:d86f with SMTP id c15-20020ac85a8f000000b002e1df21d86fmr8769717qtc.450.1649267404445; Wed, 06 Apr 2022 10:50:04 -0700 (PDT) Received: from nicolas-tpx395.localdomain (173-246-12-168.qc.cable.ebox.net. [173.246.12.168]) by smtp.gmail.com with ESMTPSA id 78-20020a370551000000b0067e3a58c090sm10057930qkf.82.2022.04.06.10.50.03 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Wed, 06 Apr 2022 10:50:04 -0700 (PDT) Message-ID: Subject: Re: [RFC PATCH 0/8] media: hantro: Add 10-bit support From: Nicolas Dufresne To: Jernej =?UTF-8?Q?=C5=A0krabec?= , ezequiel@vanguardiasur.com.ar, p.zabel@pengutronix.de, Benjamin Gaignard Cc: mchehab@kernel.org, hverkuil-cisco@xs4all.nl, gregkh@linuxfoundation.org, wens@csie.org, samuel@sholland.org, linux-media@vger.kernel.org, linux-kernel@vger.kernel.org, linux-rockchip@lists.infradead.org, linux-staging@lists.linux.dev, linux-arm-kernel@lists.infradead.org, linux-sunxi@lists.linux.dev Date: Wed, 06 Apr 2022 13:50:02 -0400 In-Reply-To: <2820101.e9J7NaK4W3@jernej-laptop> References: <20220227144926.3006585-1-jernej.skrabec@gmail.com> <4386971.LvFx2qVVIh@jernej-laptop> <2820101.e9J7NaK4W3@jernej-laptop> Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable User-Agent: Evolution 3.44.0 (3.44.0-1.fc36) MIME-Version: 1.0 Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Le mercredi 06 avril 2022 =C3=A0 19:21 +0200, Jernej =C5=A0krabec a =C3=A9c= rit=C2=A0: > Dne sreda, 06. april 2022 ob 08:54:07 CEST je Benjamin Gaignard napisal(a= ): > > Le 05/04/2022 =C3=A0 20:40, Jernej =C5=A0krabec a =C3=A9crit : > > > Hi Benjamin! > > >=20 > > > Dne torek, 05. april 2022 ob 18:07:41 CEST je Benjamin Gaignard=20 > napisal(a): > > > > Le 27/02/2022 =C3=A0 15:49, Jernej Skrabec a =C3=A9crit : > > > > > First two patches add 10-bit formats to UAPI, third extends filte= ring > > > > > mechanism, fourth fixes incorrect assumption, fifth moves registe= r > > > > > configuration code to proper place, sixth and seventh enable 10-b= it > > > > > VP9 decoding on Allwinner H6 and last increases core frequency on > > > > > Allwinner H6. > > > > >=20 > > > > > I'm sending this as RFC to get some comments: > > > > > 1. format definitions - are fourcc's ok? are comments/description= s ok? > > > > > 2. is extended filtering mechanism ok? > > > > >=20 > > > > > I would also like if these patches are tested on some more HW. > > > > > Additionally, can someone test tiled P010? > > > > >=20 > > > > > Please take a look. > > > >=20 > > > > Hi Jernej, > > > >=20 > > > > I have create a branch to test this series with VP9 and HEVC: > > > > https://gitlab.collabora.com/benjamin.gaignard/for-upstream/-/tree/= 10bit_ > > > > imx 8m Feel free to pick what I may need in it. > > > >=20 > > > > That doesn't improve fluster scores, I think more dev are still nee= ded in > > > > GST before getting something fully functional. > > > > Anyway I able to select P010 pixel format if the input is a 10bit > > > > bitstream.>=20 > > > What kind of improvements do you expect? Actually, this series is des= igned > > > to change nothing for platforms, where 10-bit format is not added int= o > > > the list of supported formats. I think reasons are quite obvious. Fir= st, > > > not every device may support 10-bit output. Second, as you might alre= ady > > > figured it out, registers in this series are set only for legacy core= s. I > > > have no idea, what needs to be done for newer ones, since I don't hav= e > > > them. Anyway, I tested this with fluster and only one additional test > > > passes, because it is the only one for 10-bit YUV420. > >=20 > > In this series you will find that I have added the registers for the ne= w > > cores, fix hevc to be able to use 10-bit, and enable that in IMX8M. >=20 > Your changes seems reasonable, but at this point I wouldn't bother with= =20 > fluster. Instead, try to test with one specific bitstream or even a sampl= e video=20 > file. I just tested with one random 10-bit VP9 video that I found when wo= rking=20 > on this series. That way you avoid any corner cases which sometimes plaqu= e=20 > fluster testing (reference bitstreams smaller than min. supported size).= =20 > Anyway, re-check vendor lib if there is any other place to adjust somethi= ng=20 > for 10-bit. Just so we don't forget, there is a handful of 10bit tests that Daniel Alme= ida omitted when he added tests to fluster (though only 1 is 420). I will try a= nd fix that later on. There is otherwise 5G worth of 10bit tests available. In fluster we decided to go for the same subset libvpx uses, otherwise no one = would ever want to download these tests. https://storage.googleapis.com/downloads.webmproject.org/vp9/decoder-test-s= treams/Profile_2_10bit. About the "min supported", G2 VP9 scores is 157/303 here (in comparision rk= vdec is 225, and MTK VCODEC 275). At this failure level this has no longer anyth= ing to do with the size of the render. There is likely couple of bugs hidden in= the driver for the corner cases tested by the suite. Also, to illustrate that t= he size isn't the only variable in the failures, we have a vp90-2-02-size- 64x34.webm that pass (the driver pretends that 64x64 is the minimum). I did= n't look at G2 output very closely, but on RKVDEC, in similar failures we have perfect keyframe, and a single corrupted tile on the following decode. My b= elief is that there is bugs in the drivers to be found and fixed. In absence of v= endor support, or working reference it will be difficult / near impossible to fix= , but I'm documenting this so we stop thinking this is just "not supported". cheers, Nicolas