| Age | Commit message (Collapse) | Author |
|---|
|
|
|
originally pointed out by Mike Silbersack on the fbsd version of the iwi
driver.
|
|
|
|
to have problems with that. noticed by Andrey Matveev.
|
|
for the R17 register).
|
|
from IEEE80211_MODE_AUTO if ic_fixed_rate != -1. it is no longer the case.
this implements a temporary workaround. the whole rate selection should be
rewritten.
|
|
these particular cases. avoid extra calls to letoh32().
|
|
|
|
hostap mode.
|
|
|
|
|
|
It turned out that priority ring base address must be set before atim ring.
This was not documented anywhere. Stupid hardware!
|
|
we disable RF auto-tuning.
|
|
document registers in ralreg.h based on the spec.
do some cleaning in the code while i'm here.
|
|
|
|
RT2526 and RT5222. based on ural code.
|
|
|
|
ic_fixed_rate is an index inside ic->ic_sup_rates[ic->ic_curmode], not inside
any ni->ni_rates. for the moment, i implement a simple workaround until it
get fixed.
|
|
|
|
|
|
extend.
|
|
|
|
fix beacon expired interrupt handler in hostap mode.
|
|
|
|
functions. this makes RTS protection mode and 802.11g work.
|
|
|
|
of tx descriptors.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
make sure we are not enabling TSF synchronization and rate adaptation in
monitor mode.
|
|
initialize power mode in ral_init().
|
|
it seems to override critical default values (like sensitivity).
|
|
fix debug messages level
simplify e2prom code
|
|
|
|
|
|
|
|
fix endianness issues.
pointed out by "barryg" (Barry Dexter A. Gonzaga).
|
|
pointed out by jsg@
|
|
in the tx ring (CTS/RTS frame + data frame).
|
|
|
|
|
|
|
|
|
