| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
|
|
|
| |
o mark rx frames including FCS in the payload with the
IEEE80211_RADIOTAP_F_FCS flag
o remove hack to copy 802.11 headers with padding out of line; instead mark
the frames with IEEE80211_RADIOTAP_F_DATAPAD and require applications to
do the work
o split precalculated radiotap flags into tx+rx now that they can be different
Note the full usefulness of these changes depends on updates to applications
that process radiotap data.
|
|
|
|
|
|
|
|
|
|
| |
o don't reclaim any previous beacon state in ath_beacon_alloc; do it
explicitly in ath_newstate
o reference count the node held in the beacon frame state block
o process ibss merge more intelligently; let the state machine do the
right thing instead of explicitly setting the new bssi id
o explicitly stop tx dma before doing beacon setup to handle the ibss
merge case
|
| |
|
|
|
|
| |
o replace the private macro to convert MS->TU with the common one
|
|
|
|
|
| |
allocation scheme introduced a race condition during device state
transitions
|
|
|
|
|
|
|
| |
o increase the max per-frame tx descriptor count and the number of tx
buffers for forthcoming fast frame support
o correct the max scatter/gather count; it cannot be larger than the
max(tx,rx,beacon) descriptor counts
|
| |
|
| |
|
|
|
|
| |
suboptimal but needed for fast frames which won't fit in a single cluster
|
|
|
|
| |
they make the led's flash unnecessarily in adhoc mode
|
| |
|
| |
|
|
|
|
|
|
|
|
|
| |
(fix imported from madwifi by Takanori Watanabe)
o eliminate save/restore of pci registers handled by the system
o eliminate duplicate zero of the softc (noted by njl)
o consolidate common code
MFC after: 1 week
|
| |
|
|
|
|
| |
Submitted by: Tai-hwa Liang
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
o catch one place where we were not using ath_chan_change to
switch channels; this fixes a problem where the channel
settings were not being correctly reported in captured packets
o return unique channel identification in the channel flags;
ethereal gets confused if you return merged flags (e.g. ofdm,
cck, and 2Ghz) (this is workaround and should be removed if
we can ever cleanup radiotap consumers)
o correct short/long preamble flag state for rx and treat tx
the same--use a new hwflags array that gives us the data
based on the h/w rate index/cookie
o add gross hack to handle radiotap capture of frames that
come in with hardware padding; should be replaced by a
flag in the radiotap header and more smarts in the apps
that decode radiotap data
|
|
|
|
|
| |
the index of the first entry on a mic error so we're
consistent with parts that don't have split keycache
|
|
|
|
|
|
|
|
|
|
|
| |
o lintval is in ms; must convert to TU's for passing to the hal
o roundup to calculate nexttbtt (should look at current tsf and pull the
calculated nextbtt forward but this'll do for now)
o don't or- in HAL_BEACON_RESET_TSF when doing station timer setup; this
is not needed and messes up the sleep timer calcs, though it's unclear
if it mattered as the hal masks these values before use
Submitted by: Thorsten von Eicken
|
|
|
|
| |
there are none since we're only called when the ref count goes to zero
|
| |
|
|
|
|
|
|
|
|
|
| |
pointers to an integer via uintptr_t.
Fix an apparent bug that caused a compile failure.
ieee80211_iterate_nodes() takes ic->ic_sta as its first argument on the
onoe module. It had just 'ic' here in the same context, which was a
mismatched argument.
|
|
|
|
|
|
|
|
| |
of a sizeof, need to use %z to get the correct type on all our platforms.
Also, convert integers<->pointers via uintptr_t.
(I think Sam's instructions were for me to commit this. If I
misunderstood, then I apologize in advance.)
|
| |
|
| |
|
|
|
|
|
| |
MPSAFE only debug_mpsafenet is 1 so callbacks to send management
frames hold Giant; this is another bandaid on the path to removing Giant.
|
|
|
|
| |
satisfy the assertion in if_start.
|
|
|
|
|
|
|
|
|
|
| |
an(4), ath(4), hme(4), ndis(4), vr(4) and wi(4)
Please help testing: http://people.freebsd.org/~mlaier/ALTQ_driver/
Tested by: Vaidas Damosevicius (an, ath, wi)
Roman Divacky (vr)
Submitted by: yongari (hme)
|
|
|
|
| |
Noticed by: phk
|
|
|
|
|
|
| |
the 802.11 layer does useful work
Obtained from: madwifi
|
|
|
|
|
|
| |
the 802.11 layer does useful work
Obtained from: madwifi
|
|
|
|
|
|
| |
was too much delay
Obtained from: madwifi
|
| |
|
|
|
|
| |
Obtained from: madwifi
|
|
|
|
|
|
| |
work done in ath_intr when the irq is shared
Obtained from: madwifi
|
|
|
|
| |
Obtained from: madwifi
|
| |
|
|
|
|
|
|
| |
o force little-endian byte order for header
o pad header to 32-bit boundary to guard against applications that assume
packet data alignment
|
|
|
|
| |
(ieee80211_ifattach() calls ether_ifattach().)
|
|
|
|
|
| |
Submitted by: Mark Santcroos <marks@ripe.net>
Reviewed by: imp, dfr, bde
|
|
|
|
| |
"I'll cope" by: sam
|
|
|
|
|
| |
This fixes a problem when operating as an AP where clients would get
stuck in the node table because the reference count never went to zero.
|
|
|
|
|
| |
so don't just stop trying to send a beacon frame or we'll be more likely
to lose sync. This only seems to happen on some older chips.
|
|
|
|
| |
resetting the hardware is sufficient, no need to reset the 802.11 fsm
|
| |
|
| |
|
|
|
|
| |
ah_debug is now treated as a bit vector
|
|
|
|
|
|
|
|
|
| |
o move tx taps from ath_start to ath_tx_start so lots more
state is available to tap
o add tx flags
o add tx rate
o add tx power (constant for the moment)
o add tx antenna state
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
a new bpf_mtap2 routine that does the right thing for an mbuf
and a variable-length chunk of data that should be prepended.
o while we're sweeping the drivers, use u_int32_t uniformly when
when prepending the address family (several places were assuming
sizeof(int) was 4)
o return M_ASSERTVALID to BPF_MTAP* now that all stack-allocated
mbufs have been eliminated; this may better be moved to the bpf
routines
Reviewed by: arch@ and several others
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
o fix race condition when processing rx descriptors: because we use
a self-linked descriptor at the end of the rx descriptor list to
avoid rx overruns (which can easily happen for 5212 parts that enable
PHY errors) we must carefully check that a descriptor is "done" by
looking ahead to the next descriptor before believing the done bit
in the current descriptor (this is all handled in the HAL since the
rx descriptor format is chip-specific so we need to pass in two
additional parameters--the physical address of the current descriptor
and the virtual address of the next descriptor in the list)
o check copyout return status for SIOCGATHSTATS ioctl
Approved by: re (scottl)
|
|
|
|
|
|
|
|
| |
this problem put these lines back in. While they should be
unnecessary, they appear to be sometimes necessary.
Reviewed in concept: dfr
Approved by: re (scottl@)
|