On Mon, Feb 06, 2006 at 11:06:14PM -0800, Nathaniel Gray wrote:
>
> Here's the compiler error:
> gcc -r -keep_private_externs -nostdlib -o .libs/io-wmf.so-master.o
> io-wmf.lo && gcc -bundle -o .libs/io-wmf.so .libs/io-wmf.so-
> master.o -L/usr/local/lib -L/usr/X11R6/lib -L/sw/lib -L/Volumes
Hi folks,
This evening I downloaded fink 0.8.0, added unstable/main to the
Trees list, then ran "sudo fink install gimp2-dev"
An hour later I got
Failed: compiling libwmf-0.2.8.2-5 failed
Here's the compiler error:
gcc -r -keep_private_externs -nostdlib -o .libs/io-wmf.so-master.o
io-wmf.
On Tue, Feb 07, 2006 at 12:57:19AM -0500, Jack Howarth wrote:
> Dan,
> Argh. I had replaced the copy of the 6.0.2 pdflib info/patch with
> ones for 5.0.3. However that was the problem. Fink index doesn't
> recognize that you have regressed the version of the info file that
> way. I had to get
Dan,
Argh. I had replaced the copy of the 6.0.2 pdflib info/patch with
ones for 5.0.3. However that was the problem. Fink index doesn't
recognize that you have regressed the version of the info file that
way. I had to get rid of the pdflib.info in my local directory, do
a fink index and then I
On Tue, Feb 07, 2006 at 12:39:20AM -0500, Jack Howarth wrote:
> Has anyone run into this before? I was building some test pdflib packages
> for 6.0.2 in my local/main/finkinfo directory and then decide to deinstall
> them and delete the pdflib.info and pdflib.patch that was used to create them.
Has anyone run into this before? I was building some test pdflib packages
for 6.0.2 in my local/main/finkinfo directory and then decide to deinstall
them and delete the pdflib.info and pdflib.patch that was used to create them.
However despite using all the options for 'fink cleanup' everytime
I noticed that the version number for gromacs-mpi in 10.4 had been
bumped up to 3.3-1003 because it requires other modified packages in that
branch. The problem is that gromacs itself also needs to be bumped up
in version to 3.3-1003 as well (it is still at 3.3-3). This is because
the two packa
Alex,
Okay. So I should have...
Depends: gmp-shlibs (>= 4.1.3-11), cctools (>= 576-1) | odcctools (>=
576-20050327), %N-shlibs (= %e:%v-%r), libiconv
However, the current gcc4 packaging appears to be broken since it doesn't
have any restriction on the version of gcc4-shlibs that is installed
On 2/6/06, Jack Howarth <[EMAIL PROTECTED]> wrote:
> Martin,
>I think I have pinpointed the problem with my gcc4.info package.
> The change...
>
> -Depends: gmp-shlibs (>= 4.1.4-1), libmpfr1-shlibs, cctools (>= 576-1) |
> odcctools (>= 576-20050327), %N-shlibs, libiconv
> +Depends: gmp-shlibs
Alex,
I think we're talking past each other here. Doesn't 'when the 10.4 tree
is ready' imply that this means when all the packages in the 10.4 tree build?
I am trying to pass along fixes for my package to build which would seem to
be helpful rather than harmful to progressing towards the goal
Martin,
I think I have pinpointed the problem with my gcc4.info package.
The change...
-Depends: gmp-shlibs (>= 4.1.4-1), libmpfr1-shlibs, cctools (>= 576-1) |
odcctools (>= 576-20050327), %N-shlibs, libiconv
+Depends: gmp-shlibs (>= 4.1.3-11), cctools (>= 576-1) | odcctools (>=
576-20050327)
On 2/6/06, Jack Howarth <[EMAIL PROTECTED]> wrote:
> Alex,
>The changes I have to sparky-py (a 3.111-1002 version) are simply
> to allow it to build in the 10.4 tree. These include...
>
> 1) the GCC variable was changed from 3.3 to 4.0
> 2) I eliminated the changes to the 'CXX= g+
Alex,
The changes I have to sparky-py (a 3.111-1002 version) are simply
to allow it to build in the 10.4 tree. These include...
1) the GCC variable was changed from 3.3 to 4.0
2) I eliminated the changes to the 'CXX= g++' section of
the patch to c++/Makefile
since the GCC
On Sat, Feb 04, 2006 at 10:43:32PM -0500, Jack Howarth wrote:
>Is anyone else seeing this sort of problem under current 10.4-transitional
> in the cvs? I typically build gcc4 from the gcc4.1 branch every few weeks
> with the info file at the end of this message. However tonight I am getting
> f
On 2/6/06, Jack Howarth <[EMAIL PROTECTED]> wrote:
> While adjusting my sparky-py packaging to build under the new 10.4 tree
> last night, I noticed I had optimized the compile with a '-mtune=970'. What
> is the recommended approach for handling this under 10.4 with the new
> Architecture list.
While adjusting my sparky-py packaging to build under the new 10.4 tree
last night, I noticed I had optimized the compile with a '-mtune=970'. What
is the recommended approach for handling this under 10.4 with the new
Architecture list. It would be nice if the Architecture list could be used
to
Martin,
I have the fink 0.24.11-21 package installed with the 10.4-transitional
unstable tree in use. The only thing I can think of is that something went
astray when I initially tried to convert to the 10.4 branch but I don't think
that is the case as I kept backups of the /sw directory befor
17 matches
Mail list logo
