Am 20.09.19 um 12:03 schrieb Jouni Malinen:
On Sun, Sep 15, 2019 at 10:08:29PM +0200, Alexander Wetzel wrote:
wpa_tdls_set_key() did set key_id to -1 as a signal to handle the key
install a bit different than for other pairwise keys.
Since we cleaned up the key install logic with a previous patch this is
no longer needed and can be removed.
Signed-off-by: Alexander Wetzel <alexander@xxxxxxxxxxxxxx>
---
The workaround cleaned up here is not limited to nl80211, so I put it
into a separate patch. It just finalizes the nl80211 driver key install
cleanup without breaking anything in between.
This key_id == -1 case is used by an old out-of-tree driver wrapper to
make TDLS work. In other words, this commit would break that. Not that
I'm too worried about out-of-tree code, but I'd like to understand what
exactly this patch is trying to achieve. Is this just cleanup and as
such, could it be dropped without breaking anything in this patch
series?
Yes, this is basically only a cleanup.
When we want to drop the patch it it won't have any real consequences
for the rest of the series.
It only would get in the way when we decide to also use unicast keyid 1
for TDLS, too. But the standard has no guidance on that and TDLS isn't
caring about rekeys much... (I looked into that a few hours and then
basically dropped any plans to use Extended Key ID also for TDLS. We
would first to have define something like "Extended-TDLS" and when
rekeys are not an issue in TDLS why bother?)
Alexaner
_______________________________________________
Hostap mailing list
Hostap@xxxxxxxxxxxxxxxxxxx
http://lists.infradead.org/mailman/listinfo/hostap
