It turns out there where several issues. All dealing with preferences.
1 - UAT_END_FIELD not included in the uat_field_t structure. (Did not break
Linux, but windows was not so kind.)
2 - In my call to prefs_register_protocol( my_proto, callback_fn) if
callback_fn was NULL, wireshark/tshark would close correctly. If not NULL
then wireshark and tshark would remain in memory when closed. Again windows
only. Linux was OK.
I have not determined why when the callback function was included, the system
failed to close. The callback only updated some memory tables with the new
info from the uat_new structure.
Any ideas on where to look?
Thanks as always.
Alex Lindberg
--- On Tue, 8/2/11, Alex Lindberg <[email protected]> wrote:
From: Alex Lindberg <[email protected]>
Subject: [Wireshark-dev] Freeing memory when quitting Wireshark
To: [email protected]
Date: Tuesday, August 2, 2011, 2:21 PM
When I quit a custom build of Wireshark (Win32), although the application
disappears from the desktop, it remains in memory.
I use g_regex for a number of dissectors. In previous versions I was using the
PCRE library and did not have this issue.
When Wireshark quits, is there a "un-register" routing that is called in the
dissector? I am suspecting that I am not releasing all my memory from the
compiled g_regex_new functions.
Thanks as always.
Alex Lindberg
-----Inline Attachment Follows-----
___________________________________________________________________________
Sent via: Wireshark-dev mailing list <[email protected]>
Archives: http://www.wireshark.org/lists/wireshark-dev
Unsubscribe: https://wireshark.org/mailman/options/wireshark-dev
mailto:[email protected]?subject=unsubscribe
___________________________________________________________________________
Sent via: Wireshark-dev mailing list <[email protected]>
Archives: http://www.wireshark.org/lists/wireshark-dev
Unsubscribe: https://wireshark.org/mailman/options/wireshark-dev
mailto:[email protected]?subject=unsubscribe
