Hi Eike,
On Tue, Aug 30, 2011 at 07:42:54PM +0200, Eike Rathke wrote:
> On Tuesday, 2011-08-30 17:49:06 +0200, Francois Tigeot wrote:
>
> > > Francois, did you adapt solenv/bin/linkoo as well?
> > Good catch, I didn't.
> >
> > What does this file do ? I did a complete build before commiting the
Hi Francois,
On Tuesday, 2011-08-30 17:49:06 +0200, Francois Tigeot wrote:
> > Francois, did you adapt solenv/bin/linkoo as well?
>
> Good catch, I didn't.
>
> What does this file do ? I did a complete build before commiting the patch
> and there was no obvious problem.
When doing make dev-i
On Tue, Aug 30, 2011 at 12:01 PM, Francois Tigeot wrote:
> On Tue, Aug 30, 2011 at 11:58:08AM -0500, Norbert Thiebaud wrote:
>> On Tue, Aug 30, 2011 at 11:48 AM, Francois Tigeot
>> wrote:
>> > On Tue, Aug 30, 2011 at 11:21:18AM -0500, Norbert Thiebaud wrote:
>> >> On Tue, Aug 30, 2011 at 10:43 A
On Tue, Aug 30, 2011 at 11:58:08AM -0500, Norbert Thiebaud wrote:
> On Tue, Aug 30, 2011 at 11:48 AM, Francois Tigeot
> wrote:
> > On Tue, Aug 30, 2011 at 11:21:18AM -0500, Norbert Thiebaud wrote:
> >> On Tue, Aug 30, 2011 at 10:43 AM, Francois Tigeot
> >> wrote:
> >> > On Tue, Aug 30, 2011 at
On Tue, Aug 30, 2011 at 11:48 AM, Francois Tigeot wrote:
> On Tue, Aug 30, 2011 at 11:21:18AM -0500, Norbert Thiebaud wrote:
>> On Tue, Aug 30, 2011 at 10:43 AM, Francois Tigeot
>> wrote:
>> > On Tue, Aug 30, 2011 at 10:06:57AM -0500, Norbert Thiebaud wrote:
>> >> On Tue, Aug 30, 2011 at 1:35 AM
On Tue, Aug 30, 2011 at 11:21:18AM -0500, Norbert Thiebaud wrote:
> On Tue, Aug 30, 2011 at 10:43 AM, Francois Tigeot
> wrote:
> > On Tue, Aug 30, 2011 at 10:06:57AM -0500, Norbert Thiebaud wrote:
> >> On Tue, Aug 30, 2011 at 1:35 AM, Francois Tigeot
> >> wrote:
>
> I don't know what you packa
On Tue, Aug 30, 2011 at 10:43 AM, Francois Tigeot wrote:
> On Tue, Aug 30, 2011 at 10:06:57AM -0500, Norbert Thiebaud wrote:
>> On Tue, Aug 30, 2011 at 1:35 AM, Francois Tigeot
>> wrote:
>> >
>> > The build system uses so many flags and tricks it can't run as-is under
>> > the pkgsrc framework,
On Tue, Aug 30, 2011 at 05:42:42PM +0200, Eike Rathke wrote:
>
> Which reminds me there may be other places that rely on the $UPD path
> particle.
>
> Francois, did you adapt solenv/bin/linkoo as well?
Good catch, I didn't.
What does this file do ? I did a complete build before commiting the pa
On Tue, Aug 30, 2011 at 10:06:57AM -0500, Norbert Thiebaud wrote:
> On Tue, Aug 30, 2011 at 1:35 AM, Francois Tigeot wrote:
> >
> > The build system uses so many flags and tricks it can't run as-is under
> > the pkgsrc framework, many libraries are not found at link time.
> > Some platforms don't
Hi,
On Tuesday, 2011-08-30 16:39:01 +0200, Francois Tigeot wrote:
> > So, I'm fine with removing 350.
>
> I'll push the commit then :-)
As it may help others as well: I just did
cd solver ; ln -s 350/$INPATH .
because I probably will forget to mv solver/350/$INPATH solver/
after the next gi
Hi Kohei,
On Tuesday, 2011-08-30 10:20:05 -0400, Kohei Yoshida wrote:
> You can try git-new-workdir
>
> http://kohei.us/2010/11/16/working-with-a-branch-using-git-new-workdir/
Way cool, thanks! I'll try that one day.
Eike
--
PGP/OpenPGP/GnuPG encrypted mail preferred in all private commun
On Tue, Aug 30, 2011 at 1:35 AM, Francois Tigeot wrote:
>
> The build system uses so many flags and tricks it can't run as-is under
> the pkgsrc framework, many libraries are not found at link time.
> Some platforms don't understand the -rlink-path and related options either.
>
> Using LD_LIBRARY_
On Tue, Aug 30, 2011 at 03:42:22PM +0200, Eike Rathke wrote:
>
> On Tuesday, 2011-08-30 14:59:23 +0200, Stephan Bergmann wrote:
>
> > >>> Can't we at least get rid of the '350' subdirectory ?
> > >>
> > >> Yes please.
> > >
> > > The 350 should be 340 (isn't it?) for 3-4 branches, so different
Hi Eike,
On Tue, 2011-08-30 at 15:42 +0200, Eike Rathke wrote:
> True, but instead of getting rid of solver/350 I'd rather introduce
> a module/$INPATH/350 ... on the other hand, having two cloned repos
> and
> build trees for different branches is a better approach. Just was
> thinking in terms o
On Aug 30, 2011, at 3:42 PM, Eike Rathke wrote:
> On Tuesday, 2011-08-30 14:59:23 +0200, Stephan Bergmann wrote:
> Can't we at least get rid of the '350' subdirectory ?
Yes please.
>>>
>>> The 350 should be 340 (isn't it?) for 3-4 branches, so different output
>>> hierarchies when
Hi Stephan,
On Tuesday, 2011-08-30 14:59:23 +0200, Stephan Bergmann wrote:
> >>> Can't we at least get rid of the '350' subdirectory ?
> >>
> >> Yes please.
> >
> > The 350 should be 340 (isn't it?) for 3-4 branches, so different output
> > hierarchies when switching branches is nice to have i
Sent from my iPhone
On 30.08.2011, at 14:40, Eike Rathke wrote:
> Hi Tor,
>
> On Tuesday, 2011-08-30 00:15:50 -0600, Tor Lillqvist wrote:
>
>>> Can't we at least get rid of the '350' subdirectory ?
>>
>> Yes please.
>
> The 350 should be 340 (isn't it?) for 3-4 branches, so different outp
Hi,
On Tuesday, 2011-08-30 00:15:50 -0600, Tor Lillqvist wrote:
> For instance, when cross-compiling to PowerPC Linux from x86_64 Linux
> (hypothetical example, but as such potentially useful), idlc would be
> in solver/350/unxlngx6.pro/bin, and the PowerPC binaries being
> produced in solver/350
Hi Tor,
On Tuesday, 2011-08-30 00:15:50 -0600, Tor Lillqvist wrote:
> > Can't we at least get rid of the '350' subdirectory ?
>
> Yes please.
The 350 should be 340 (isn't it?) for 3-4 branches, so different output
hierarchies when switching branches is nice to have in order to not
spoil alread
On Tue, Aug 30, 2011 at 01:29:36AM -0600, Tor Lillqvist wrote:
> > Then we should only need two build directories:
> > - one for the host binaries
> > - one for the target binaries
>
> Bzzt, you are misunderstanding the GNU autofoo cross-compilation
> terminology;) (A *very* common trap to fa
> Then we should only need two build directories:
> - one for the host binaries
> - one for the target binaries
Bzzt, you are misunderstanding the GNU autofoo cross-compilation terminology;)
(A *very* common trap to fall in...)
In general, in GNU autofoo cross-compilation terminology, there
On Tue, Aug 30, 2011 at 12:15:50AM -0600, Tor Lillqvist wrote:
> > I have no idea how cross-compilation is currently implemented, but can't
> > that
> > be done with the same build directory name for all platforms ?
>
> Not unless the way it is done currently is changed. New ideas, and patches t
On Tue, Aug 30, 2011 at 01:07:03AM -0500, Norbert Thiebaud wrote:
> On Tue, Aug 30, 2011 at 12:51 AM, Francois Tigeot
> wrote:
> > On Mon, Aug 29, 2011 at 12:17:33AM -0600, Tor Lillqvist wrote:
> >> > I'd like to consolidate the different OUTPATH values to a single one or
> >> > two
> >> > at mo
> I have no idea how cross-compilation is currently implemented, but can't that
> be done with the same build directory name for all platforms ?
Not unless the way it is done currently is changed. New ideas, and patches to
implement them welcome. But don't just break it.
The tools that we build
On Tue, Aug 30, 2011 at 12:51 AM, Francois Tigeot wrote:
> On Mon, Aug 29, 2011 at 12:17:33AM -0600, Tor Lillqvist wrote:
>> > I'd like to consolidate the different OUTPATH values to a single one or two
>> > at most (unx and wnt). Thoughts ?
>>
>> That will break cross-compilation as currently imp
On Mon, Aug 29, 2011 at 12:17:33AM -0600, Tor Lillqvist wrote:
> > I'd like to consolidate the different OUTPATH values to a single one or two
> > at most (unx and wnt). Thoughts ?
>
> That will break cross-compilation as currently implemented. It isn't entirely
> unthinkable that somebody might
> I'd like to consolidate the different OUTPATH values to a single one or two
> at most (unx and wnt). Thoughts ?
That will break cross-compilation as currently implemented. It isn't entirely
unthinkable that somebody might at some point want to cross-compile from some
(well, *the*) mainstream L
27 matches
Mail list logo
