Fwd: [PATCH] Add eapol_sm_notify_eap_config function to eapol_supp_sm.c.

[David Smith (☕)]

It looks like this message from me got misposted, so resending original.

On Thu, Jan 15, 2009 at 4:14 AM, Jouni Malinen <j at w1.fi> wrote:
>
> On Wed, Jan 14, 2009 at 11:37:02AM +0900, David Smith (?) wrote:
> > This function is for reconfiguring the EAPOL state machine when the EAP
> > configuration is changed, either by reconfiguring smartcard modules over
> > DBus or a reload of the config file.
> >
> > Also, modify ctrl_iface_dbus_handlers.c and wpa_supplicant.c to use new
> > interface.
>
> Hmm.. Is this change really needed for DBus case?
> wpas_dbus_iface_set_smartcard_modules() is already calling
> eapol_sm_deinit() followed by wpa_supplicant_init_eapol() which will in
> practice deinit and init EAP peer which seems to be exactly what the
> proposed new eapol_sm_notify_eap_config() function is doing. Not that
> I'm saying that this is the best way of doing the update (thanks for
> pointing out that place for me ;-).

Sorry I should have provided more backstory. The original problem that
started this was I was seeing some segfaults and when I debugged it,
it was eap_notify_lower_layer_success() when eapol_get_bool() would
dereference the eapol_cb in the eap_sm (I can produce a coredump
and/or stack trace if you're interested).