From mboxrd@z Thu Jan 1 00:00:00 1970 From: =?UTF-8?Q?G=C3=A1bor_Boskovits?= Subject: Re: gcc-ddc Date: Sat, 2 Dec 2017 15:53:25 +0100 Message-ID: References: <871skskduj.fsf@gnu.org> <87lgj0wqbd.fsf@elephly.net> <87mv3dv59h.fsf@elephly.net> <877eu9f56j.fsf@gnu.org> <87mv31dw43.fsf@gnu.org> Mime-Version: 1.0 Content-Type: multipart/alternative; boundary="089e082658b0c78b61055f5ca57d" Return-path: Received: from eggs.gnu.org ([2001:4830:134:3::10]:49704) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1eL9AY-00010n-LI for guix-devel@gnu.org; Sat, 02 Dec 2017 09:53:32 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1eL9AX-000448-Di for guix-devel@gnu.org; Sat, 02 Dec 2017 09:53:30 -0500 In-Reply-To: <87mv31dw43.fsf@gnu.org> 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: Jan Nieuwenhuizen Cc: Guix-devel --089e082658b0c78b61055f5ca57d Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Ok, I will do that. I can eliminate differences in the wrappers easily, then I will look into the libtool patch. It seems, that the la files are causing the checksum difference in cc1 and cc1plus, because if I remove libraries from the checksum, then they agree. I also checked the same things when I compile with clang. The differences are the same, so we can do this. The rest is just "cosmetics". 2017-12-02 15:48 GMT+01:00 Jan Nieuwenhuizen : > G=C3=A1bor Boskovits writes: > > > Aside from these libtool files we can now say, that this ddc project > succeeded. > > Wait... The libtool's .la files are now the only files that show any > difference, even when gcc is compiled into it's own prefix? That's > amazing!!! > > > I've contacted the libtool developers if we can extend the wrapper > approach to the .la files. > > Great! > > > It seems, that in some older version of libtool those were just sourced > as shell script, but > > I don't know if now they do something more fancy with it or not... > > Anyways, if it's just shell script, then the environment variable > approach can also work out there. > > The only problem seems, that I should do the substitution before > checksumming the compiler. > > I think I can inject something into the makefile, or use a patched > vesion of libtool. > > > > A patched libtool could be a better option, so other ddc projects can > use it. > > Indeed. > > > I guess I can do something like add an environment variable > GUIX_INSTALL_DIRECTORY, or something like that... > > What's different about GUIX_INSTALL_DIRECTORY than the usual: PREFIX? > > > Any maybe name this version libtool-for-ddc. > > It should be noted in the package documentation, that this package is > not recommended for general use. > > Are you proposing to patch libtool? If so, even if patching it for > [gcc-]dcc only would be a great hack for now. Please go forward with > that idea at least until libtool devs help us with a full solution. > > janneke > > -- > Jan Nieuwenhuizen | GNU LilyPond http://lilypond.org > Freelance IT http://JoyofSource.com | Avatar=C2=AE http://AvatarAcademy.c= om > --089e082658b0c78b61055f5ca57d Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable
Ok, I will do that.

I can eliminate dif= ferences in the wrappers easily, then I will look into the libtool patch.
It seems, that the la files are causing the checksum difference in= cc1 and cc1plus, because if I remove libraries from the checksum, then the= y agree.
I also checked the same things when I compile with clang= . The differences are the same, so we can do this. The rest is just "c= osmetics".


2017-12-02 15:48 GMT+01:00 Jan Nieuwenhuizen <janne= ke@gnu.org>:
G=C3=A1bor Boskovits writes:

> Aside from these libtool files we can now say, that this ddc project s= ucceeded.

Wait... The libtool's .la files are now the only files that show= any
difference, even when gcc is compiled into it's own prefix?=C2=A0 That&= #39;s amazing!!!

> I've contacted the libtool developers if we can extend the wrapper= approach to the .la files.

Great!

> It seems, that in some older version of libtool those were just source= d as shell script, but
> I don't know if now they do something more fancy with it or not...=
> Anyways, if it's just shell script, then the environment variable = approach can also work out there.
> The only problem seems, that I should do the substitution before check= summing the compiler.
> I think I can inject something into the makefile, or use a patched ves= ion of libtool.
>
> A patched libtool could be a better option, so other ddc projects can = use it.

Indeed.

> I guess I can do something like add an environment variable GUIX_INSTA= LL_DIRECTORY, or something like that...

What's different about GUIX_INSTALL_DIRECTORY than the usual: PR= EFIX?

> Any maybe name this version libtool-for-ddc.
> It should be noted in the package documentation, that this package is = not recommended for general use.

Are you proposing to patch libtool?=C2=A0 If so, even if patching it= for
[gcc-]dcc only would be a great hack for now.=C2=A0 Please go forward with<= br> that idea at least until libtool devs help us with a full solution.

janneke

--
Jan Nieuwenhuizen <janneke@gnu.org> | GNU LilyPond http://lilypond.org
Freelance IT http://JoyofSource.com | Avatar=C2=AE http://AvatarAcademy.c= om

--089e082658b0c78b61055f5ca57d--