Hi Johannes,
On 4/25/23 12:14 PM, James Prestwood wrote:
Hi Johannes,
On 4/25/23 11:10 AM, Johannes Berg wrote:
Hi,
I can't tell with the information you gave - depends on the waiting
period for a response frame you ask for, I guess? With HW ROC (if
you're
using iwlwifi) we cannot extend the previous period, see
ieee80211_coalesce_hw_started_roc().
This is purely virtual at the moment, but in my case its looking like it
cannot be extended either since the CMD_FRAME just queues a separate
request.
Hm. Not sure then? I guess then you'd be falling into the max duration
or so?
if (!local->ops->remain_on_channel) {
/* If there's no hardware remain-on-channel, and
* doing so won't push us over the maximum r-o-c
* we allow, then we can just add the new one to
* the list and mark it as having started now.
* If it would push over the limit, don't try to
* combine with other started ones (that haven't
* been running as long) but potentially sort it
* with others that had the same fate.
*/
unsigned long now = jiffies;
u32 elapsed = jiffies_to_msecs(now -
tmp->start_time);
struct wiphy *wiphy = local->hw.wiphy;
u32 max_roc =
wiphy->max_remain_on_channel_duration;
if (elapsed + roc->duration > max_roc) {
combine_started = false;
continue;
}
Or maybe that logic here is broken somewhat ...
I guess my assumption was if ROC is currently active then a CMD_FRAME
(on the same channel) should just go out immediately, not be queued, and
not even bother checking the duration. (and in this case I'm not even
setting a duration for CMD_FRAME).
Does the OFFCHANNEL_TX_OK flag have any bearing on this? I found that I
have to set it even if ROC is ongoing, maybe thats another topic of
discussion.
Not sure I understand this part. ROC is fine mostly for the "wait for
some frame and send a response", but not so much suited for "send a
frame and wait for a response" part. So 3-way-handshakes are iffy with
it...
Yeah, and it actually has worked great for the entire DPP procedure
using the same channel (presence -> auth request -> auth response ->
auth confirm) assuming both sides respond in a timely fashion.
The comes when changing the channel after the auth request. The auth
request gets queued separately, which then delays the ROC and we
can't/shouldn't send anything until ROC starts. The only strange thing
is we actually receive the auth response on the new channel before the
ROC for that new channel even starts. Its like the hardware and driver
aren't quite in sync.
Did you say hwsim? That'd be weird. Though in hwsim I think you have an
additional quirk - it never really *leaves* the original channel it's
connected on, it kind of sticks around on *both* which isn't real but
some kind of simplification there. We might want to fix that eventually.
But not sure it's connected already in this case?
Ah ok, thats probably whats happening then. And it is connected in this
case. We're connected to a BSS then initiate DPP to configure some
unprovisioned device. Start the protocol with the channel it used for
presence announcements, then tell the peer to transition to our
currently connected channel (by setting the frequency in the auth request).
So when you asked about being connected it got me thinking, and I went
to check and saw I was doing something very stupid. When we switch to
the new channel its the connected channel, so doing ROC is pointless
(surprised it actually allowed it). And this avoids the issue in
question entirely since there is no delay waiting for another ROC, just
cancel the last request and wait for a frame on the connected channel.
So thats my bad :) The behavior in question is still weird IMO (not
sending CMD_FRAME right away), but in any case it at least works now.
Thanks for the help getting my brain to work :)
But anyways I think its best to use ROC for presence (waiting for
announcements) but then use CMD_FRAME for the rest of the protocol.
Right, that's pretty much the intent for this kind of thing. Similar in
P2P where we designed all this, really.
Ok cool, thanks for clarifying
johannes
