Fedora Core 1 Test Update: mc-4.6.0-8.3

[Leonard den Ottolander]

Hello Jakub,

> Actually, I found another bug in that patch (and that the original code is
> very crappy).

Tell me about it ;-) . But it is a good indication of where and how to
solve the issue.

The fact that 8.3 still segfaulted on me made me assume you were using
the original patch. As it was getting quite late (so it has on your end
I see :) I didn't look any closer yet.

> It always calloced the arrays to make sure they are in the
> end NULL terminated, of course realloc without any memset would not keep the
> rest cleared and so if you're unlucky it would segfault.  Triggered with
> explicit memset (, 0xaa, ).  Rewritten so that it terminates both the
> keywords and contexts explicitely and no longer misuses calloc.

Can you make the patch available anywhere before the push?

> Bill, mc-4.6.0-8.4 is in updates-candidate, can you please repush? Thanks.

And please be so kind to announce 8.1 and push it to the main tree in
the mean time. I am a bit surprised (to use an understatement) a
security patch lies around in testing for over 10 days but never got
announced. Not everybody is using testing on a regular basis, and even I
missed this update as it was not announced. And I was monitoring, as
announcements for RHEL and RH9 have already been made.

> Now I'm really away...

Thanks and hope you have a good night.

Bye,
Leonard.

-- 
mount -t life -o ro /dev/dna /genetic/research