Re: Couple of crashes with 21.5.8 and kkcc
>>>> "viteno" == Norbert Koch
<nk(a)viteno.net> writes:
viteno> sperber(a)informatik.uni-tuebingen.de (Michael Sperber [Mr. Preprocessor])
writes:
viteno> Hi!
> Hm. David's recent fix which I just checked in might help
with the
> problem. If it fails on some specific Lisp object, as in your
> xemacs.bt1, could you try pobj from gdb on that object so we have a
> better idea what part of the invariants are failing?
viteno> Unfortunately, the patch hasn't helped. After having applied them and
viteno> recompilation, I've started a new session in the debugger. It crashed
viteno> right at the start. I've traced the gdb session. The output is
viteno> attached to this message.
viteno> Do you need information from higher stack levels?
No, this is fine. All crash reports I've seen so far are either stack
overflows (good; I know how to make them go away eventually) or GCPRO
bugs.
These crashes belong to the latter category: line 4360 in alloc.c is
the loop marking GCPRO'd objects. It seems kkcc is merely exposing
these bugs which are probably somewhere else in the code. Unless I
find a way of making these reproducible for me, they'll be very hard
to find, unfortunately.
--
Cheers =8-} Mike
Friede, Völkerverständigung und überhaupt blabla