From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tobias Geerinckx-Rice Subject: Re: jfs, jfsutils? Date: Fri, 13 Dec 2019 17:07:48 +0100 Message-ID: <87lfrgi4yz.fsf@nckx> References: <882e8091-aaa7-0705-306d-d4bf2824e050@gmail.com> <87y2viifel.fsf@nckx> <87v9qmialu.fsf@nckx> <4454c84b-fa97-df27-d5d5-7e7a8643f8ef@gmail.com> Mime-Version: 1.0 Content-Type: multipart/signed; boundary="=-=-="; micalg=pgp-sha512; protocol="application/pgp-signature" Return-path: Received: from eggs.gnu.org ([2001:470:142:3::10]:60876) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1ifnTx-0005Ie-Ab for help-guix@gnu.org; Fri, 13 Dec 2019 11:07:58 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1ifnTu-0006QK-AM for help-guix@gnu.org; Fri, 13 Dec 2019 11:07:56 -0500 Received: from tobias.gr ([2001:470:7405::1]:42754) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1ifnTt-0006LB-Q7 for help-guix@gnu.org; Fri, 13 Dec 2019 11:07:54 -0500 In-reply-to: <4454c84b-fa97-df27-d5d5-7e7a8643f8ef@gmail.com> List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: help-guix-bounces+gcggh-help-guix=m.gmane.org@gnu.org Sender: "Help-Guix" To: Michael Zucchi Cc: help-guix@gnu.org --=-=-= Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: quoted-printable Michael, Michael Zucchi =E5=86=99=E9=81=93=EF=BC=9A > Ok so it seems you're half done already, but I've been plodding=20 > away=20 > (and didn't check my email on my old computer) and have both=20 > building. That's great! Shame about the duplicated effort, but such things=20 happen. > I'm not really sure where to put it or how to compile each part=20 > (it just file-systems.scm if possible. linux.scm is already a bloated=20 grab-bag. For now, you can test each package at a time by adding the=20 variable (e.g =E2=80=98jfs_fsck/static=E2=80=99) as the last line of your c= urrent=20 .scm file. This will make the file =E2=80=98evaluate to jfs_fsck/static=E2= =80=99,=20 so you can: $ guix build -f thefile.scm You'll have to change that last line every time you want to change=20 the package being built. You'll need a Guix git checkout to get much further. If you=20 haven't yet done so, read =E2=80=98Building from Git=E2=80=99 in the manual= and=20 bootstrap yourself some Guix. Then you can copy your 3 package expressions to file-systems.scm,=20 keeping it in alphabetical order or at least not making things=20 worse. After making some small adjustments like adding a license:=20 prefix, you can build your packages with: ~/guix$ ./pre-inst-env guix build jfsutils > build jfsutils, not jfs_fsck/static) and I used the slackware=20 > patches, ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ Typo that should be the other way 'round? Are you sure that LDFLAGS=3D-static is needed when using=20 static-package? Here's my aborted static version: ++(define-public jfsutils/static ++ (static-package ++ (package ++ (inherit jfsutils) ++ (name "jfsutils-static") ++ (inputs ++ `(("util-linux" ,util-linux) ++ ("util-linux:static" ,util-linux "static")))))) I got my patches from Debian, but they do the same job. There=20 were others[0] but they weren't vital or relevant to Guix. > but attached is what i have at the moment. I used jfs_fsck for=20 > the=20 > static fsck because that's the default name but it could be=20 > fsck.jfs. > > I started with a definitions of e2fsprogs and it's related=20 > functions=20 > from a linux.scm i found in /gnu/store and based it on that. > > I've a number of built package 'attempts' building up - should I=20 > just=20 > run guix gc to clear them out or will that delete stuff i'm just=20 > going=20 > to need again? It will by default remove the =E2=80=98build=E2=80=99 dependencies, yes. I advise against GC'ing unless you desperately need space, and it=20 won't come from a handful of relatively tiny stale JFSutils. Guix=20 also deduplicates the store by default. At file level, even. > (I presume attachments work ok on this list.) Sure. ;; COPYING is gpl3, the headers are gpl2+? That's interesting. Where'd you find gpl2+? Kind regards, T G-R [0]: https://sources.debian.org/patches/jfsutils/1.1.15-4/ --=-=-= Content-Type: application/pgp-signature; name="signature.asc" -----BEGIN PGP SIGNATURE----- iQIzBAEBCgAdFiEEfo+u0AlEeO9y5k0W2Imw8BjFSTwFAl3zt1QACgkQ2Imw8BjF STyVNxAArCyVao/2JBUYJcvn9eZ9/x/JnMaOYYXk/zRpO7Jd0yX2FLVj0KIpl5a5 6UAWtvQydCxECpeovcXKoQ7J/sdDho/lmXQZjcu5dkC3LqShqZNIwjfRpDJlmY/J dtSp9aORI5XiDFJutNcvtSNE5+Fd8tUUS+n/1KW/nmj2SIXBRglt4FuFvz0lg5n5 vf4QKpMm21gUApA0muMkvENgz0AFGqBlSB3wKtqfNO5c9ZPWKXSRlbVQUIA56nlU rlNmU32s08XzwNEZf8VQosq3jx+PMicxln87CC83A1hWHF4cuJ5vayk7AMoUkuQu SV7t/xiCmtUXMpT9uGwc+raKQS1h8JZ5AJtzPivrHzwR/XyayuPKxXAh5tGijKmz 2+BrspqvsswJiRjqzXNxU052BsAn2ArMpLWOSvdB/AqQ4mM+o/Y8XHJWe7uLGzS8 OkV0GOTHSK8Sh4FaSc9SnaU2xpAkGSLiF96e1S15KdfMvEHtgF4kAgxhU3TYL2Uo vh+Dt3WX+FfYgFzn7d35JQAv4VPsf2unShbjuXQZ33gNxQOpOTB+CBPf5D8mO/c0 y/nC0PkJI0I69RuULIX5sZiKt/HzTDIP+KAGM4MB6K4ViOcrQslTN4K14rBvB0BG nfB+41gKoP9NHRPYm+wbbYCoCOB5PZf8QGRiFcrq/uGcUElojQg= =m2Ko -----END PGP SIGNATURE----- --=-=-=--