Date: Sat, 2 Sep 2006 23:26:48 +0200
Thomas Girard <[EMAIL PROTECTED]> wrote:
> Hello,
Hello Thomas,
>
> I can't reproduce this problem on my i386 box. But according to the
> report, the bug can only be seen on an amd64. Having a look at the
> buildd log[1] reveals that gcc complains a log about "cast to pointer
> from integer of different size".
>
> Having a look at the source code shows that it indeed relies on pointers
> being 32 bits, e.g. manedit/editordnd.c EditorDNDParseCmd function reads
> pointers assuming they are guint32.
>
> The attached patch tries to correct those warnings. I have asked
> someone with an AMD64 box to try to reproduce the bug with and whithout
> this patch, but he was not able to reproduce it in *both* cases.
>
> Can you please try it?
Done. I just applied the attached patch to a clean 0.6.1-2 source tree
and compiled it in a clean SID chroot. It crashes in the same way.
--- 8< ---
[EMAIL PROTECTED]: ~ $ rm .maneditrc
[EMAIL PROTECTED]: ~ $ manedit
[EMAIL PROTECTED]: ~ $ manedit
ManEdit triggered a segmentation fault (1 times)
ManEdit triggered a segmentation fault (2 times)
ManEdit triggered a segmentation fault (3 times)
ManEdit attempting immediate process exit().
--- 8< ---
I'm available for all the required debug operations. In [1] you can
temporally fetch the patched binary file.
Tag this bug as 'unreproducible' and decrease its severity, if you
consider it appropriate.
>
> Thanks,
You're welcome.
[1]
http://zoidberg.criptonita.com/~nacho/debian/debug/manedit_0.6.1-2_amd64.deb
--
bye,
- Nacho
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
