Wireshark-bugs: [Wireshark-bugs] [Bug 8456] Buildbot crash output: fuzz-2013-03-10-3098.pcap
Date Index · Thread Index · Other Months · All Mailing Lists
Date Prev · Date Next · Thread Prev · Thread Next
From: bugzilla-daemon@xxxxxxxxxxxxx
Date: Sun, 10 Mar 2013 17:41:57 +0000

Comment # 5 on bug 8456 from 
(In reply to comment #4)
> (In reply to comment #3)
> > (In reply to comment #2)
> > > Jakub, CCing you since I suspect the latter is a side-effect of when you
> > > reduced the scope of ep_ memory?
> > 
> > Nah, we never supported ep_ memory in col_set_str() [see r30012 and bug
> > #3984].
> > 
> > export WIRESHARK_EP_VERIFY_POINTERS=
> > $ ./tshark -r /tmp/fuzz-2013-03-10-3098.pcap > /dev/null
> > 
> > [cut]
> > col_set_str(pinfo->cinfo, COL_INFO, 
> >      val_to_str(pdu_type, rtr_pdu_type_vals, "Unknown %d"));
> >      ^^^^^^^^^^
> >      unsafe, when value not in value_string array it will return ep_ memory.

For references this was broken in r47482 (val_to_str_const changed to
val_to_str).

> I believe this could be fixed by having val_to_str use pinfo-scoped wmem
> memory? It would fix tshark at least, not sure if Wireshark requires an even
> larger scope?

Looking at packet-rpki-rtr.c code setting COL_INFO is done in while() loop, so
probably col_append_sep_str() is much better than col_set_str().
[and it'll fix this bug]

If you want to know my opinion about bug #3984, I think we should always make a
copy of buffer passed in col_set_str() and forget about it.
[But I never profile it].

You are receiving this mail because:
  • You are watching all bug changes.