ANNOUNCEMENT: Live Wireshark University & Allegro Packets online APAC Wireshark Training Session
April 17th, 2024 | 14:30-16:00 SGT (UTC+8) | Online

Wireshark-bugs: [Wireshark-bugs] [Bug 3917] tvb_new_real_data is prone to memory leak.

Date: Tue, 12 Jan 2010 13:31:42 -0800 (PST)
https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=3917

--- Comment #4 from Anders Broman <anders.broman@xxxxxxxxxxxx> 2010-01-12 13:31:41 PST ---
>> Committ revision 31499.
>> Leaves this unfixed, more elaboration on the possible leaks would help:
>> ./asn1/kerberos/packet-kerberos-template.c
>> may leak memory, not all code paths call tvb_free(encr_tvb).
>in decrypt_krb5_data(), tvb_free is not called at the end of the loop
Is this true in packet-kerberos.c too? packet-kerberos.c is a hand coded
dissector the asn1 one is incomplet and not yet "in service".

>> 
>> ./plugins/asn1/packet-asn1.c
>> leak and dissect a tvb buff in proto_reg_handoff() outside a TRY block.
>Not sure about the leak maybe I made a mistake.
I don't think this dissecor is used any more so low prio.

>> 
>> ./epan/dissectors/packet-ssl-utils.c
>> leak tvb structure memory.
>Tvbs allocated in ssl_add_record_info are never freed, only the tvb payload is.
>Not easy to fix.

This may be related to Bug
https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=3737 
I wonder if data can be added multiple times at redissection, but I lack time
to dig into the code :-(

-- 
Configure bugmail: https://bugs.wireshark.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.