We have no mac80211 drivers that generate beacons themselves
and would require this, and we also have no way to support such
drivers currently. Hence, it seems we should remove this to
decrease the confusion. Since the generic hostapd code actually
requires a positive result though we cannot just remove the
implementation. The same information is put into the beacon,
however.
Signed-off-by: Johannes Berg <[email protected]>
---
hostapd/driver_devicescape.c | 14 --------------
1 file changed, 14 deletions(-)
--- hostap.orig/hostapd/driver_devicescape.c 2007-09-26 14:45:22.000000000 +0200
+++ hostap/hostapd/driver_devicescape.c 2007-09-26 14:47:17.000000000 +0200
@@ -695,20 +695,6 @@ static int i802_sta_set_flags(void *priv
static int i802_set_generic_elem(const char *ifname, void *priv,
const u8 *elem, size_t elem_len)
{
- struct i802_driver_data *drv = priv;
- struct iwreq iwr;
-
- memset(&iwr, 0, sizeof(iwr));
-
- os_strlcpy(iwr.ifr_name, drv->hapd->conf->iface, IFNAMSIZ);
- iwr.u.data.length = elem_len;
- iwr.u.data.pointer = (void*)elem;
-
- if (ioctl(drv->ioctl_sock, SIOCSIWGENIE, &iwr) < 0) {
- perror("Failed to set generic info element");
- return -1;
- }
-
return 0;
}
On Tue, 2007-10-02 at 20:01 -0700, Jouni Malinen wrote:
> hostapd_set_generic_elem() wrapper for this driver wrapper function
> should return 0 if the wrapper is not implemented (function pointer is
> NULL).. Did I miss something or shouldn't we just remove
> i802_set_generic_elem() function completely instead of leaving the empty
> "return 0" version in?
Oh no you're right, somehow I was under the impression that the wrapper
function returned an error if it wasn't implemented. Want a new patch?
johannes
On Wed, Oct 03, 2007 at 08:47:34AM +0200, Johannes Berg wrote:
> On Tue, 2007-10-02 at 20:01 -0700, Jouni Malinen wrote:
> > hostapd_set_generic_elem() wrapper for this driver wrapper function
> > should return 0 if the wrapper is not implemented (function pointer is
> > NULL).. Did I miss something or shouldn't we just remove
> > i802_set_generic_elem() function completely instead of leaving the empty
> > "return 0" version in?
>
> Oh no you're right, somehow I was under the impression that the wrapper
> function returned an error if it wasn't implemented. Want a new patch?
No need for a new patch; I applied your patch with a modification of
jusr removing the function completely.
--
Jouni Malinen PGP id EFC895FA
On Wed, Sep 26, 2007 at 02:54:18PM +0200, Johannes Berg wrote:
> We have no mac80211 drivers that generate beacons themselves
> and would require this, and we also have no way to support such
> drivers currently. Hence, it seems we should remove this to
> decrease the confusion. Since the generic hostapd code actually
> requires a positive result though we cannot just remove the
> implementation. The same information is put into the beacon,
> however.
hostapd_set_generic_elem() wrapper for this driver wrapper function
should return 0 if the wrapper is not implemented (function pointer is
NULL).. Did I miss something or shouldn't we just remove
i802_set_generic_elem() function completely instead of leaving the empty
"return 0" version in?
--
Jouni Malinen PGP id EFC895FA