From mboxrd@z Thu Jan 1 00:00:00 1970 From: Kei Kebreau Subject: Re: [PATCH] gnu: Add fpc. (version 2) Date: Mon, 20 Mar 2017 11:02:04 -0400 Message-ID: <874lyo56vn.fsf@openmailbox.org> References: <87poj9xwzp.fsf@openmailbox.org> <87lgtxxial.fsf@openmailbox.org> <87efzpxfxh.fsf@openmailbox.org> <87a8adxct0.fsf@openmailbox.org> <20170314224527.7bee1241@scratchpost.org> Mime-Version: 1.0 Content-Type: multipart/signed; boundary="=-=-="; micalg=pgp-sha256; protocol="application/pgp-signature" Return-path: Received: from eggs.gnu.org ([2001:4830:134:3::10]:54923) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1cpyq3-0001JF-OF for guix-devel@gnu.org; Mon, 20 Mar 2017 11:03:17 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1cpypz-0003Qz-6D for guix-devel@gnu.org; Mon, 20 Mar 2017 11:03:15 -0400 Received: from lb1.openmailbox.org ([5.79.108.160]:39482 helo=mail.openmailbox.org) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_128_CBC_SHA1:16) (Exim 4.71) (envelope-from ) id 1cpypy-0003Qp-PB for guix-devel@gnu.org; Mon, 20 Mar 2017 11:03:11 -0400 In-Reply-To: <20170314224527.7bee1241@scratchpost.org> (Danny Milosavljevic's message of "Tue, 14 Mar 2017 22:45:27 +0100") List-Id: "Development of GNU Guix and the GNU System distribution." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: guix-devel-bounces+gcggd-guix-devel=m.gmane.org@gnu.org Sender: "Guix-devel" To: Danny Milosavljevic Cc: guix-devel@gnu.org --=-=-= Content-Type: text/plain Content-Transfer-Encoding: quoted-printable Danny Milosavljevic writes: > Hi, > >>+ (copy-file (assoc-ref inputs "fpc-binary") >>+ "fpc-bin.tar") >>+ (zero? (system* "tar" "xvf" "fpc-bin.tar"))))) > > Why the copy-file and then untar ? Can't it be untarred immediately from = where it is? If it is useful, maybe add a comment about the reason. > It turns out that this wasn't useful. I can't remember what I wanted to accomplish by copying the file first, so I've removed the unnecessary call. > Also, it might be easier to have a fpc-bootstrap package with the > bootstrap binaries that installs the binaries required to build fpc > normally and to make a fpc package depend on fpc-bootstrap as > native-inputs. What do you think? I myself don't have a strong > preference about it - but some other packages do it that way. > This is how I have fpc set up to build right now, unless I'm missing something. I've defined the bootstrap binaries for the i386 and x86_64 architectures, and the FPC package definition builds the source using the appropriate binaries depending on the architecture of the build target.=20 > About the fpc-reproducibility.patch , it might make sense to file bugs up= stream about it so they add it on their side. > I'll bring it up to them. > Also, these existing bootstrap compilers on sourceforge do not produce > bit reproducible executables, right? Should they also have the same > patch applied upstream in the future ? > The existing bootstrap compilers can produce bit reproducible executables when they are provided with a program without environment-dependent variables (time, hostname, etc.). As an example, the FPC source itself builds reproducibly using the bootstrap FPC binaries. > Overall LGTM! > Thanks for the review! > fpc supports armhf and usually so do we - although not here. We can add = armhf support in a future patch, though. You know, the FPC has released new binaries for its version 3.0.2 release. Now there are bootstrap binaries available for mips, arm-eabihf and powerpc as well. I can see about adding those and trying them out. --=-=-= Content-Type: application/pgp-signature; name="signature.asc" -----BEGIN PGP SIGNATURE----- iQIzBAEBCAAdFiEEg7ZwOtzKO2lLzi2m5qXuPBlGeg0FAljP7uwACgkQ5qXuPBlG eg2siQ/6A1FzZFdx/gWNqMr1k5EVyb6tNoV2sECNWnX/uLgWIeLKQv9Q5xr/rV+/ mr2CbWbKZfhC0EwkZrjfCxijdV26uyVErDF9qo626a171jlfHgZKPHUnswOIP4nF TqD4ybCEcwJ2dnqn7JEvkavF4t6mDNfxGzddI+8LcTLklDSlTH9h0v+NVGKpoe/m rwB3WKEMFRC+az8YrWjFA1JnBrjavmRciGhxKM/jHXem9D3tz4VfRyiNfAV5ig9t AD71c9hpvYo10oWiwbivKZw3EjXzoMC07sK8bY2FLPKhGL/N414yV7ckItmqZsL1 JZ2uoBntMLryG6zt736DGtF0pMcHRUW9RWLRwcbHjqpAl77C1kQJPtofN2nmHPoZ wFJtyT5+xH17aNpFNBzBeS/Zy/ZOMQdQVZmMvBhaRtbVNRcjVSNkkDdCxnxKddX5 4bEVN+C2aCVZGzYacE+kgJZ16iB6Wp9KbHvzmR5tJ9pUqkusJ0fFq1v38RC/WpKY dztpB2WF9rCN2MvLRpfzu4btqVJyv/sXkaixmQsseSXJiT0lvVNZ3msdwykr2TgK RFALrvv46i0XNsBE4P1FaH9qYGROOR2RNTt2uFrabKdjtVdSrC7FGj80egkqf+fH lNoqVXVpZ/PUJMSybw9HuRSQzJWXJCWb6swX9RU0GG1zbqSkIb4= =Euo0 -----END PGP SIGNATURE----- --=-=-=--