https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=4056
--- Comment #6 from Guy Harris <guy@xxxxxxxxxxxx> 2010-02-09 01:22:00 PST ---
> In my understanding the problem is with cfile.pd and wtap_seek_read() in packet_list_store.c:packet_list_dissect_and_cache_record(), it clobbers the data source for the first tab bytes view.
>
> Either 1) we use a different buffer or 2) we don't update the byte view if it's already displayed.
At least for that problem, I've checked in a fix that does 1) - I just use an
on-the-stack pseudo_header structure and pd array. It appears to fix the
version of the problem that I reproduced with the instructions in the bug.
However, as noted, there are other issues as well. Perhaps what we need is
support for multiple memory pools, with epan_init() allocating a memory pool
and perhaps having the pinfo structure point to it, have the dissectors
allocate from that pool rather than a single ephemeral memory pool, and have
ep_dissect_free destroy the pool.
--
Configure bugmail: https://bugs.wireshark.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.