[PATCH] ath10k: dont call csa_finish more than once

Michal Kazior michal.kazior at tieto.com
Mon Mar 24 02:37:33 EDT 2014

On 21 March 2014 16:07, Johannes Berg <johannes at sipsolutions.net> wrote:
> On Fri, 2014-03-21 at 15:58 +0100, Michal Kazior wrote:
>> > Didn't we have this discussion on the hwsim patch before?
>> Oh. I'm not aware of the discussion. Probably missed it.
> Or I got it wrong :)
>> > I'd argue that if all drivers must track this, then the mac80211 API is
>> > defective.
>> I can make this a mac80211 patch then?
> Yeah I guess. What was the reason for this though? Here it's multi-vif?
> Before I think we were talking about races?

Before multi-vif if ieee80211_csa_finalize() failed and left
csa_active==true driver could end up poking the csa_finalize_work()
over and over again I guess?

For multi-vif csa using reservation there are 2 reasons:
 a) channel switches across interfaces may not be tightly
synchronized. I did tests where 2 long (cs_count=100) channel switches
were a few seconds apart. In that case driver poked csa_finalize work
quite a lot resulting in a mess.
 b) my current approach has the following flow:

1. reserve_chanctx()
2. (some time later) driver calls csa_finish, and csa_finalize calls
use_reserved_chanctx() because reserved_chanctx is set
3. (possibly some time later, for multi-vif reservations) csa_finalize
is called again but this time without reserved_chanctx, so it verifies
if bss_conf.chandef is csa_chandef (after successful reservation it
should be).

This means if driver would poke csa_finalize() more than once then
use_reserved_chanctx() would be called again and again as a result.


More information about the ath10k mailing list