https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=4176
--- Comment #6 from Guy Harris <guy@xxxxxxxxxxxx> 2010-01-28 19:46:07 PST ---
None of the libraries in
These files contain "freeaddrinfo":
WS_30513\k5sprt32.dll 10/06/2009 00:19 32,768 1.6.3.16
WS_30666\k5sprt32.dll 10/06/2009 00:19 32,768 1.6.3.16
WS_30513\krb5_32.dll 10/06/2009 00:19 720,896 1.6.3.16
WS_30666\krb5_32.dll 10/06/2009 00:19 720,896 1.6.3.16
WS_30513\libwireshark.dll 11/10/2009 19:47 27,926,528 1.3.1.30513
WS_30666\libwireshark.dll 22/10/2009 20:17 28,043,776 1.3.1.30666
WS_30666\libgio-2.0-0.dll 07/10/2009 15:11 670,569
should be Wireshark-specific - they should be used by TShark as well.
If the Windows run-time linker resolves symbols at call time rather than at
library-load time, it could be that, for whatever reason, TShark somehow never
ends up calling freeaddrinfo().
Presumably
> The pairs of k*.dll are identical according to FC /B.
doesn't apply to libwireshark.dll, unless the fc command is incapable of
correctly handling what I presume are files of different sizes; k5sprt32.dll
and krb5_32.dll have the same size and mod time, so they probably have the same
contents - and we might not be calling any of the routines therein that use
freeaddrinfo(), if there are any.
At least from a quick look at the libgio source, it *does* - at least in GLib
2.22.4 - use freeaddrinfo().
--
Configure bugmail: https://bugs.wireshark.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.
