On Sun, Apr 28, 2013 at 08:13:44AM -0400, Nathan Trapuzzano <[email protected]>
wrote:
> I've attached a log from running urxvtd under valgrind. It detected a
> lot of errors.
It's hard to see any errors in your valgrind output, I only found this (it
also seems you traced the children, and there is some binary corrtuption
in your log file).
==2766== Invalid free() / delete / delete[] / realloc()
==2766== at 0x4C2AD3C: free (in
/usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==2766== by 0x67C9D1F: tdestroy_recurse (in /usr/lib/libc-2.17.so)
==2766== by 0x683312B: free_mem (in /usr/lib/libc-2.17.so)
==2766== by 0x68332A1: __libc_freeres (in /usr/lib/libc-2.17.so)
==2766== by 0x4A246BC: _vgnU_freeres (in
/usr/lib/valgrind/vgpreload_core-amd64-linux.so)
==2766== by 0x433CEF: sig_handlers::sig_term(ev::sig&, int) (in
/usr/bin/urxvtd)
==2766== by 0x437E66: void ev::base<ev_signal,
ev::sig>::method_thunk<sig_handlers, &sig_handlers::sig_term>(ev_signal*, int)
(in /usr/bin/urxvtd)
==2766== by 0x458143: ev_invoke_pending() (in /usr/bin/urxvtd)
==2766== by 0x45887F: ev_run (in /usr/bin/urxvtd)
==2766== by 0x41E9A0: main (in /usr/bin/urxvtd)
==2766== Address 0xa450bc0 is 0 bytes inside a block of size 12 free'd
==2766== at 0x4C2AD3C: free (in
/usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==2766== by 0x436D54: stringvec::~stringvec() (in /usr/bin/urxvtd)
==2766== by 0x43322B: rxvt_term::~rxvt_term() (in /usr/bin/urxvtd)
==2766== by 0x4336BD: rxvt_term::destroy_cb(ev::idle&, int) (in
/usr/bin/urxvtd)
==2766== by 0x437C86: void ev::base<ev_idle,
ev::idle>::method_thunk<rxvt_term, &rxvt_term::destroy_cb>(ev_idle*, int) (in
/usr/bin/urxvtd)
==2766== by 0x458143: ev_invoke_pending() (in /usr/bin/urxvtd)
==2766== by 0x45887F: ev_run (in /usr/bin/urxvtd)
==2766== by 0x41E9A0: main (in /usr/bin/urxvtd)
Without line number info, it's hard to tell, but I would guess that
something frees env variables that don't belong to it, and that probably
wouldn't be urxvtd.
So, valgrind indeed gave a clue, but unfortunately, the same clue as we
have seen before. Somebody with the problem needs to isolate the buggy
component and then debug it and see where it causes the corruption.
> I'm also attaching my rxvt-unicode configure log. I usualy build with
> -O2, but rebuilt with `-O0 -g' for debugging this.
Something must be wrong with your setup then, as valgrind should dump line
number information by default. Did you switch it off or strip the binary?
--
The choice of a Deliantra, the free code+content MORPG
-----==- _GNU_ http://www.deliantra.net
----==-- _ generation
---==---(_)__ __ ____ __ Marc Lehmann
--==---/ / _ \/ // /\ \/ / [email protected]
-=====/_/_//_/\_,_/ /_/\_\
_______________________________________________
rxvt-unicode mailing list
[email protected]
http://lists.schmorp.de/cgi-bin/mailman/listinfo/rxvt-unicode