/* most likely the symlink is relative, so change our
directory to the dir of the symlink */
- rc = chdir(dirname(outName));
+ rc = chdir(dirname(strdupa(outName)));
do {
buf = alloca(len + 1);
rc = readlink(basename(outName), buf, len);
--
1.7.10.4
_______________________________________________
Anaconda-devel-list mailing list
Anaconda-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/anaconda-devel-list
/* most likely the symlink is relative, so change our
directory to the dir of the symlink */
- rc = chdir(dirname(outName));
+ rc = chdir(dirname(strdupa(outName)));
do {
buf = alloca(len + 1);
rc = readlink(basename(outName), buf, len);
Peter,
BTW, I know there has been a record number of releases, but if a 8.15
could be released soon with this fix it'd be great.
This is currently affecting our ability to test RHEL 5 systems.
Thanks a lot!
CR.
_______________________________________________
Anaconda-devel-list mailing list
Anaconda-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/anaconda-devel-list
> Because dirname() may modify the path. See:
>
> http://www.kernel.org/doc/man-pages/online/pages/man3/dirname.3.html
>
> This fixes a segfault found while manually testing grubby on RHEL 5.
>
> This reverts commit 54ff9d923f1dd90dde81e0e90e9448fdc52044c4.
Ok, you are right in this, it looks like I had read the dirname()
manpage without paying attention to that detail.
This introduces back a memory leak -- which I know it is not a huge
issue provided that Grubby is not a long-running process. Nevertheless,
going through the code and fixing those would be desirable in the long
term (IMHO).
Regards,
--
Adrian Perez <aperez@igalia.com> - Sent from my toaster
Igalia - Free Software Engineering
_______________________________________________
Anaconda-devel-list mailing list
Anaconda-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/anaconda-devel-list
On 06/25/2012 10:12 AM, Adrian Perez de Castro wrote:
Ok, you are right in this, it looks like I had read the dirname()
manpage without paying attention to that detail.
This introduces back a memory leak -- which I know it is not a huge
issue provided that Grubby is not a long-running process. Nevertheless,
going through the code and fixing those would be desirable in the long
term (IMHO).
Yeah, you're right. I've pushed a fix upstream, though I kicked off an
8.15 build before you noticed this, so the fix will be in whenever 8.16
happens. It's a very minor problem, so I'm not too worried about getting
a fixed build done immediately.
--
Peter
_______________________________________________
Anaconda-devel-list mailing list
Anaconda-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/anaconda-devel-list