linux.git/net/mac80211/ht.c, branch v5.0

ieee80211: bump IEEE80211_MAX_AMPDU_BUF to support HE

2018-06-18T20:39:39+00:00

Bump the IEEE80211_MAX_AMPDU_BUF size to 0x100 for HE support
and - for now - use IEEE80211_MAX_AMPDU_BUF_HT everywhere.

This is derived from my internal patch, parts of which Luca
had sent upstream.

Signed-off-by: Luca Coelho 
Signed-off-by: Johannes Berg

mac80211: fix TX aggregation stop race

2018-05-18T09:14:36+00:00

The mac80211 tear down code is not waiting for the driver call back.
This can bring down the the TX path (TID) till the user manually
reconnects. (Observed with iwldvm and enabled TX aggregation.)

The race can be prevented when the ampdu_mlme worker handles the tear
down.

The race:
 * ieee80211_sta_tear_down_BA_sessions calls
   ___ieee80211_stop_tx_ba_session for all TIDs,

 * then cancels the ampdu_mlme worker

 * and cleanups the TIDs the driver already has called back for.

 * ieee80211_stop_tx_ba_cb will never be called for a TID if the callback
   came after the the check in ieee80211_sta_tear_down_BA_sessions.

Signed-off-by: Alexander Wetzel 
[johannes: "enabled" -> "blocked" and invert logic, simplify init]
Signed-off-by: Johannes Berg

mac80211: Use proper smps_mode enum in sta opmode event

2018-03-29T08:19:55+00:00

SMPS_MODE change value notified via nl80211 contains mac80211
specific value(ieee80211_smps_mode) and user space application
will not know those values. This patch add support to map
the mac80211 enum value to nl80211_smps_mode which will be
understood by the userspace application.

Signed-off-by: Tamizh chelvam 
Signed-off-by: Johannes Berg

Merge tag 'mac80211-next-for-davem-2018-01-04' of git://git.kernel.org/pub/scm/linux/kernel/git/jberg/mac80211-next

2018-01-04T19:33:29+00:00

Johannes Berg says:

====================
We have things all over the place, no point listing them.

One thing is notable: I applied two patches and later
reverted them - we'll get back to that once all the driver
situation is sorted out.
====================

Signed-off-by: David S. Miller

mac80211: mark expected switch fall-throughs

2017-12-11T11:16:04+00:00

In preparation to enabling -Wimplicit-fallthrough, mark switch cases
where we are expecting to fall through.

Notice that in some cases I replaced "fall through on else" and
"otherwise fall through" comments with just a "fall through" comment,
which is what GCC is expecting to find.

Signed-off-by: Gustavo A. R. Silva 
Signed-off-by: Johannes Berg

mac80211: fix locking in ieee80211_sta_tear_down_BA_sessions

2017-12-11T09:50:00+00:00

Due to overlap between
commit 1281103770e9 ("mac80211: Simplify locking in ieee80211_sta_tear_down_BA_sessions()")
and the way that Luca modified
commit 72e2c3438ba3 ("mac80211: tear down RX aggregations first")
when sending it upstream from Intel's internal tree, we get
the following warning:

WARNING: CPU: 0 PID: 5472 at net/mac80211/agg-tx.c:315 ___ieee80211_stop_tx_ba_session+0x158/0x1f0

since there's no appropriate locking around the call to
___ieee80211_stop_tx_ba_session; Sara's original just had
a call to the locked __ieee80211_stop_tx_ba_session (one
less underscore) but it looks like Luca modified both of
the calls when fixing it up for upstream, leading to the
problem at hand.

Move the locking appropriately to fix this problem.

Reported-by: Kalle Valo 
Reported-by: Pavel Machek 
Tested-by: Pavel Machek 
Signed-off-by: Johannes Berg

mac80211: tear down RX aggregations first

2017-11-27T10:24:59+00:00

When doing HW restart we tear down aggregations.
Since at this point we are not TX'ing any aggregation, while
the peer is still sending RX aggregation over the air, it will
make sense to tear down the RX aggregations first.

Signed-off-by: Sara Sharon 
Signed-off-by: Luca Coelho 
Signed-off-by: Johannes Berg

mac80211: Simplify locking in ieee80211_sta_tear_down_BA_sessions()

2017-09-21T09:42:01+00:00

Simplify the locking in  ieee80211_sta_tear_down_BA_sessions() and
lock sta->ampdu_mlme.mtx over the entire function instead of
locking/unlocking it for each TID etc.

Signed-off-by: Ilan Peer 
Signed-off-by: Johannes Berg

mac80211: fix deadlock in driver-managed RX BA session start

2017-09-06T13:22:02+00:00

When an RX BA session is started by the driver, and it has to tell
mac80211 about it, the corresponding bit in tid_rx_manage_offl gets
set and the BA session work is scheduled. Upon testing this bit, it
will call __ieee80211_start_rx_ba_session(), thus deadlocking as it
already holds the ampdu_mlme.mtx, which that acquires again.

Fix this by adding ___ieee80211_start_rx_ba_session(), a version of
the function that requires the mutex already held.

Cc: stable@vger.kernel.org
Fixes: 699cb58c8a52 ("mac80211: manage RX BA session offload without SKB queue")
Reported-by: Matteo Croce 
Signed-off-by: Johannes Berg

mac80211: Complete ampdu work schedule during session tear down

2017-09-06T13:22:02+00:00

Commit 7a7c0a6438b8 ("mac80211: fix TX aggregation start/stop callback race")
added a cancellation of the ampdu work after the loop that stopped the
Tx and Rx BA sessions. However, in some cases, e.g., during HW reconfig,
the low level driver might call mac80211 APIs to complete the stopping
of the BA sessions, which would queue the ampdu work to handle the actual
completion. This work needs to be performed as otherwise mac80211 data
structures would not be properly synced.

Fix this by checking if BA session STOP_CB bit is set after the BA session
cancellation and properly clean the session.

Signed-off-by: Ilan Peer 
[Johannes: the work isn't flushed because that could do other things we
 don't want, and the locking situation isn't clear]
Signed-off-by: Johannes Berg