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 aws-us-west-2-korg-lkml-1.web.codeaurora.org (localhost.localdomain [127.0.0.1]) by smtp.lore.kernel.org (Postfix) with ESMTP id CB2BAC38142 for ; Fri, 27 Jan 2023 18:03:41 +0000 (UTC) Received: from mailout4.zoneedit.com (mailout4.zoneedit.com [64.68.198.64]) by mx.groups.io with SMTP id smtpd.web10.109307.1674842611724070196 for ; Fri, 27 Jan 2023 10:03:32 -0800 Authentication-Results: mx.groups.io; dkim=missing; spf=none, err=permanent DNS error (domain: denix.org, ip: 64.68.198.64, mailfrom: denis@denix.org) Received: from localhost (localhost [127.0.0.1]) by mailout4.zoneedit.com (Postfix) with ESMTP id EBFD240BD6; Fri, 27 Jan 2023 18:03:30 +0000 (UTC) Received: from mailout4.zoneedit.com ([127.0.0.1]) by localhost (zmo14-pco.easydns.vpn [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Qn870_c5uf6a; Fri, 27 Jan 2023 18:03:30 +0000 (UTC) Received: from mail.denix.org (pool-100-15-88-116.washdc.fios.verizon.net [100.15.88.116]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by mailout4.zoneedit.com (Postfix) with ESMTPSA id AD96F406B7; Fri, 27 Jan 2023 18:03:24 +0000 (UTC) Received: by mail.denix.org (Postfix, from userid 1000) id 3517D163712; Fri, 27 Jan 2023 13:03:04 -0500 (EST) Date: Fri, 27 Jan 2023 13:03:04 -0500 From: Denys Dmytriyenko To: Ryan Eatmon Cc: Andrew Davis , rs@ti.com, denys@ti.com, detheridge@ti.com, meta-arago@lists.yoctoproject.org, k-bhargav@ti.com Subject: Re: [meta-arago] [PATCH] all: Remove excessive switches based on gpu Message-ID: <20230127180304.GG22689@denix.org> References: <20230124231733.2219206-1-rs@ti.com> <20230126205801.GE22689@denix.org> <46352935-21d8-680c-776b-580bdc333ce6@ti.com> <0c318fc9-17b6-7a33-714a-64acb12b48de@ti.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <0c318fc9-17b6-7a33-714a-64acb12b48de@ti.com> User-Agent: Mutt/1.5.20 (2009-06-14) List-Id: X-Webhook-Received: from li982-79.members.linode.com [45.33.32.79] by aws-us-west-2-korg-lkml-1.web.codeaurora.org with HTTPS for ; Fri, 27 Jan 2023 18:03:41 -0000 X-Groupsio-URL: https://lists.yoctoproject.org/g/meta-arago/message/14192 On Fri, Jan 27, 2023 at 10:52:34AM -0600, Ryan Eatmon wrote: > > > On 1/27/2023 9:52, Andrew Davis wrote: > >On 1/26/23 6:51 PM, Randolph Sapp via lists.yoctoproject.org wrote: > >>On Thu, Jan 26 2023 at 03:58:01 PM -0500, Denys Dmytriyenko > >> wrote: > >>>I'm realizing now that simply depending on "opengl" > >>>DISTRO_FEATURES won't be > >>>enough, unfortunately. > >>> > >>>The issue is that DISTRO_FEATURES are set by the distro globally for all > >>>machines, whether with accelerated graphics, display w/o acceleration or > >>>completely headless. E.g. when building for J7200, "opengl" > >>>will be set, but > >>>none of the graphics packages should be built or installed > >>>into the image. > >>> > >>>So, some sort of check for MACHINE_FEATURES or PREFERRED_PROVIDER (less > >>>elegant) would still be needed in some places... > >> > >>Arguably that's a package group / distro target issue. I believe > >>when a user selects the default image with all features enabled > >>for J721E and J721S2 the package selection should be equivalent > >>(or at least as equivalent as possible with the stuff going on > >>now in meta-ti) to align with the distro features selected. I > >>believe the solution to this is to add a headless target/image > >>that removes Qt, GTK, etc. from the selected package groups and > >>to suggest the use of that instead if users want a truly > >>headless distro. > >> > >> > > > >Agree here, I believe this was the solution we came up with in > >our call the other day. > > > >I'd go as far as suggesting we drop the "default" image and try to match > >Ubuntu naming here. Have 3 images: Desktop, Server, and IoT. (IoT is > >a rebranded tisdk-thinlinux-image). User selects the one that matches > >their usecase, no need to force them into one bucket automatically based > >on their hardware. > > > >Andrew > > > That level of renaming will likely require a broader audience. I > agree that it would be clearer naming to follow what other > distributions do and not invent our own names. Agree. As I mentioned on the call yesterday, last round of image (re-)naming took quite a while and lots of effort to get everyone on the product side aligned and agree... Moreover, images is just part of the problem - some thought needs to be put into how the devkit is produced (and or named) and the top-level SDK product bundle, which combines all images, devkit and docs/manifests. -- Denys