| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
| |
so it isn't considered again
|
|
|
|
| |
can easily happen if the bss is quiet
|
|
|
|
| |
created due to rx'd frames had an extra reference
|
|
|
|
| |
as onoe rate control invokes this once a second
|
|
|
|
|
|
| |
node may be orphaned
Submitted by: Tai-hwa Liang
|
|
|
|
|
|
|
|
| |
happen on the first management frame received from a neighbor; we assume
any merge candidate will send more frames and those should be processed
with a suitable table entry.
Stepped on by: Tai-hwa Liang
|
|
|
|
|
|
|
|
| |
Hold a lock on the table instead of futzing with reference counts which
was potentially dangerous except drivers were quiescent while we did this
so the table contents never changed. Disable the hack logic for removing
scan candidates with multiple association failures; it's never done the
right thing and will be fixed correctly with background scanning goes in.
|
|
|
|
| |
discarded when not scanning
|
|
|
|
|
| |
not we're going to process the frame; this makes the counters reflect frames
actually processes instead of received (discarded frames were already counted)
|
|
|
|
|
|
| |
table and not the scan table
Noticed by: Tai-hwa Liang
|
|
|
|
| |
it's time to join the bss we can't get an error
|
|
|
|
|
| |
candidate on 5Ghz channel to candidate on 2Ghz channel only when the
rssi are comparable (wasn't considering rssi).
|
|
|
|
| |
diverging too much to make tracking these files worthwhile
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
o ic_inact_auth is a bad name, it's the inactivity threshold
for being associated but not authorized; use it that way
o reset ni_inact when switching inactivity thresholds to
minimize the race against the timer (don't want to lock
for this stuff)
o change the inactivity probe threshold from a one-shot to
cover a range: when below this threshold but not expired
send a probe each inactivity interval; should probably
guard against the interval being turned way down as this
could cause us to spam the net with probes
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
we're at it:
o WPA/802.11i has a unicast key and a group key; in station mode
everything is sent with the unicast key--we were consulting the
destination mac address and incorrectly using the group key
o (perpetuate fallback use of the default tx key to maintain
compatibility with the way wpa_supplicant works)
o correct EAPOL encryption logic to check unicast key instead
of assuming other state implies this
o move QoS encapsulation up to before enmic work so TKIP has the
information required to calculate the pseudo-header
o do not do QoS-encapsulation of EAPOL frames as some ap's do the
wrong thing with such frames (may need to revisit this if ap's
start dropping non-QoS frames from stations assoc'd with QoS)
o move ieee80211_mbuf_adjust closer to its caller
|
|
|
|
| |
users that want it can explicitly enable it
|
| |
|
| |
|
| |
|
| |
|
|
|
|
| |
can happen on an ibss merge
|
| |
|
| |
|
|
|
|
|
| |
when associated; this corrects a problem where doing a scan while associated
caused the associated ap to sometimes not be included in the scan set
|
|
|
|
|
| |
is invalid so we can distinguish this from needing a parameter update;
fixes dynamic update of wme parameters
|
| |
|
|
|
|
| |
capture doesn't see it
|
|
|
|
|
| |
flags for testing
o debug msg consistency fixups
|
| |
|
| |
|
|
|
|
| |
for an ieee80211 header taking into account padding requirements
|
|
|
|
|
| |
when IEEE80211_F_DATAPAD is set (e.g. ath); must use ieee80211_hdrspace
instead of ieee80211_hdrsize
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
| |
should push software state to the hardware (was ERESTART which caused the
system call to be retried)
Submitted by: Tor Egge
|
| |
|
|
|
|
|
|
| |
here but it includes completed 802.11g, WPA, 802.11i, 802.1x, WME/WMM,
AP-side power-save, crypto plugin framework, authenticator plugin framework,
and access control plugin frameowrk.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
device drivers to declare that the ifp->if_start() method implemented
by the driver requires Giant in order to operate correctly.
Add a 'struct task' to 'struct ifnet' that can be used to execute a
deferred ifp->if_start() in the event that if_start needs to be called
in a Giant-free environment. To do this, introduce if_start(), a
wrapper function for ifp->if_start(). If the interface can run MPSAFE,
it directly dispatches into the interface start routine. If it can't
run MPSAFE, we're running with debug.mpsafenet != 0, and Giant isn't
currently held, the task is queued to execute in a swi holding Giant
via if_start_deferred().
Modify if_handoff() to use if_start() instead of direct dispatch.
Modify 802.11 to use if_start() instead of direct dispatch.
This is intended to provide increased compatibility for non-MPSAFE
network device drivers in the presence of Giant-free operation via
asynchronous dispatch. However, this commit does not mark any network
interfaces as IFF_NEEDSGIANT.
|
|
|
|
|
|
|
|
|
|
|
|
| |
your (network) modules as well as any userland that might make sense of
sizeof(struct ifnet).
This does not change the queueing yet. These changes will follow in a
seperate commit. Same with the driver changes, which need case by case
evaluation.
__FreeBSD_version bump will follow.
Tested-by: (i386)LINT
|
|
|
|
| |
one on death-row in <sys/kernel.h>
|
|
|
|
|
|
|
| |
o The ieee80211_media_status() function updates the ifi_link_state field
and calls rt_ifmsg() to notify listeners on the routing socket.
Approved by: sam
|
|
|
|
|
|
| |
include FCS; if you want to provide it you can put it in the header.
Submitted by: David Young
|