From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from list by lists.gnu.org with archive (Exim 4.71) id 1SPMK7-0001jk-2j for mharc-grub-devel@gnu.org; Tue, 01 May 2012 19:17:35 -0400 Received: from eggs.gnu.org ([208.118.235.92]:36338) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1SPMK3-0001j9-Rz for grub-devel@gnu.org; Tue, 01 May 2012 19:17:33 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1SPMK1-0005ZR-PE for grub-devel@gnu.org; Tue, 01 May 2012 19:17:31 -0400 Received: from mail-wi0-f169.google.com ([209.85.212.169]:36056) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1SPMK1-0005ZM-DD for grub-devel@gnu.org; Tue, 01 May 2012 19:17:29 -0400 Received: by wibhm17 with SMTP id hm17so3397738wib.0 for ; Tue, 01 May 2012 16:17:27 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=message-id:date:from:user-agent:mime-version:to:subject:references :in-reply-to:x-enigmail-version:content-type; bh=VJcG3nah7Akz/+K3W/FsG3hlA/k5g1jSt1hxPCrHAYQ=; b=IxMpmCfjOppyI8XOE1IQS1tDqX6f5gkdw6wUPBFXIiHU6h/dYWHHjQCEXi+weKyDVW FGfKQl0clRR6POSoWULLqzi8fWF+QfnlET6VmpJr+TvvLj0RgNKhcSYMNM5E0psnkake aaZkcYWjeV3EB9c4bOPDS9+xxXOoKKwfKPjYn5EIQGtj/OTYvy4E8ggO0/Jo/D+UU4pi AxIa+/sjDt6ecIBJN4LQ0lM0B3GNkBEFSwnNv0YJS8dqJNdWTv/fkpcrsVfePGZjGZ9S oYkXBHLz+IwBlWfpHmMvufj7mP2bXIrGfqsOYrGZm45DlexzEGwHrJlhMQ8U3vUsmpkx 3nZA== Received: by 10.216.139.194 with SMTP id c44mr16145345wej.112.1335914247352; Tue, 01 May 2012 16:17:27 -0700 (PDT) Received: from debian.x201.phnet (35-233.197-178.cust.bluewin.ch. [178.197.233.35]) by mx.google.com with ESMTPS id ff9sm40038368wib.2.2012.05.01.16.17.25 (version=TLSv1/SSLv3 cipher=OTHER); Tue, 01 May 2012 16:17:26 -0700 (PDT) Message-ID: <4FA06F02.7070007@gmail.com> Date: Wed, 02 May 2012 01:17:22 +0200 From: =?UTF-8?B?VmxhZGltaXIgJ8+GLWNvZGVyL3BoY29kZXInIFNlcmJpbmVua28=?= User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:10.0.3) Gecko/20120329 Icedove/10.0.3 MIME-Version: 1.0 To: grub-devel@gnu.org Subject: Re: [PATCH] Increase flexibility of kernel naming, allow non-versioned kernels. References: <4F9CC7A0.3090406@gentoo.org> <4F9D6C1C.6040202@gmail.com> <4F9D9F6A.7080803@gentoo.org> In-Reply-To: <4F9D9F6A.7080803@gentoo.org> X-Enigmail-Version: 1.4.1 Content-Type: multipart/signed; micalg=pgp-sha512; protocol="application/pgp-signature"; boundary="------------enigB283BD590562AE2D3AB5E12D" X-detected-operating-system: by eggs.gnu.org: Genre and OS details not recognized. X-Received-From: 209.85.212.169 X-BeenThere: grub-devel@gnu.org X-Mailman-Version: 2.1.14 Precedence: list Reply-To: The development of GNU GRUB List-Id: The development of GNU GRUB List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 01 May 2012 23:17:33 -0000 This is an OpenPGP/MIME signed message (RFC 2440 and 3156) --------------enigB283BD590562AE2D3AB5E12D Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable On 29.04.2012 22:07, Mike Gilbert wrote: > On 04/29/2012 12:28 PM, Vladimir '=CF=86-coder/phcoder' Serbinenko wrot= e: >> On 29.04.2012 06:46, Mike Gilbert wrote: >>> I am relaying a patch by Robin Johnson, one of the core infrastructur= e >>> staff at Gentoo Linux. In the process of building some Gentoo servers= >>> utilizing GRUB 2, he has made some changes to 10_linux that should ma= ke >>> grub-mkconfig "out of the box" for additional users. >>> >>> I did a little work to clean up the indentation and wrote a proper >>> changelog. Credit should go to him. >>> >>> Here is his description: >>> >>> Increase flexibility of kernel naming, allow non-versioned kernels. >>> The block that tried to find the kernels was getting unweidly long, a= s >>> was not >>> easily customizable by users or distributors. Refactor to introduce a= new >>> variable, GRUB_KERNEL_GLOB, that allows complete control over the nam= ing >>> used >>> to search for kernel binaries. >> It was well structured and its length is solely due to distros using >> multiple names. We can't make it shorter and this patch surely didn't.= > Right. I think the intent is not to make it "shorter", just slightly > easier to manage. Well it didn't. So I'll prefer these tossing around to be kept out and I'll skip discussing them in the rest of this mail. As for the goal of moving this part out of the script altogether and making it distro-specific config it's an impossible goal since upstream version has to be functional on common distros unless distros go out of their way to make it difficult. >> I don't consider that being able to customise the kernel naming is of >> any advantage. Saying otherwise is like saying "let's make location of= >> /etc/passwd customizable". Some objects just have to be named as given= =2E > I do not follow that analogy. In contrast to /etc/passwd, the only > program on the system that actually needs to know the path to the kerne= l > is the boot loader. So long as the boot loader can find it, you can put= > the kernel anywhere, and name it anything. Even if it was though, it still creates problem for no benefit. Also Super GRUB Disk and bootinfoscript both need to find the kernels and initrds. >> I don't see any advantage to renaming kernels, especially given that y= ou >> can change versionstring as you see fit. > I agree that there is no "advantage" to these various naming schemes; i= t > is just something that people have done over the years due to the lack > of a standard back in the day. > > We have many users on Gentoo who compile their kernel and then manually= > copy the resulting bzImage to /boot with whatever name they please. Thi= s > patch is an attempt GRUB compatible with that reality. > > I am respectfully asking you to merge it to save myself some work in > re-basing a disto-specific patch, or my users from having to change > their behavior due to a somewhat rigid boot loader configuration file > generator. I fail to see how "configure your uncommon naming scheme into a config file" is less annoying than "stick to a common naming scheme". Both require informing users and both require user action. Sticking to standard naming scheme has obvious benefits of standardising the filenames which is especially interesting for tools like os-prober, bootinfoscript, Super GRUB Disk. >>> Add 'bzImage' to the list of default names to support more distributi= on >>> naming >>> variants. >>> >>> Adjust the default set of globs to look for unversioned kernels befor= e >>> versioned kernels, to find symlinked kernel names. >> Additionally empty version would probably result in some quirks not >> addressed in this patch. > Could you expand on this? I do not have the knowledge you do in this > area, so I cannot predict all possible problems. Off the top of my head: it's used for title, id, initrd name and sorting. All those would have to be adjusted. > Thanks. > > > > _______________________________________________ > Grub-devel mailing list > Grub-devel@gnu.org > https://lists.gnu.org/mailman/listinfo/grub-devel --=20 Regards Vladimir '=CF=86-coder/phcoder' Serbinenko --------------enigB283BD590562AE2D3AB5E12D Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.12 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iF4EAREKAAYFAk+gbwIACgkQNak7dOguQglFJAEAmqYQJ7dyXrutEvsEzf4KKOS8 Gtz4T6a6SxT6EYpni0IA/0t3s/TjLfeMTheFWWIzri/IRfNILM5/xHEkCcayNfWH =tjWK -----END PGP SIGNATURE----- --------------enigB283BD590562AE2D3AB5E12D--