pckbcintr_hard and pckbcintr_soft, which, together, do the same as
pckbcintr. The first is called from the interrupt handler, the second
from a softinterrupt at arbitrary protection level (lower or equal to
IPL_TTY). They pass data via a small ringbuffer.
The new functions are intended for platforms that, due to
hardware/firmware restrictions are not able to make pckbcintr called at
IPL_TTY, like Krups javastations. Using the monolithic pckbcintr on
these plattforms leads to the input handlers for the slot data to be
run at elevated priority, causing various race conditions.
APs to run at "full speed" where before they ran at just 1 or 2Mb/s.
The AP will adapt the data rate for each client based on packet
losses and the received signal strength.
I have also enabled rate adaptation for STA mode and for IBSS mode,
but the hardware gives us less control over the data rate in those
modes.
denote a signed dBm Rx power, so rename them to _DBM_ and change
the signedness. Add new _DB_ flags for unsigned Rx power measured
from an arbitrary reference.
Try to synchronize net80211/ieee80211_radiotap.h with my tcpdump
sources (which public tcpdump has not seen, yet) and with FreeBSD.
* don't let cac_ccb_alloc() sleep since it regularly runs in an
interrupt context
* return EAGAIN instead EBUSY (or -1), which is what the upper ld
layer expects to get on transient resource shortages.
* ignore error from cac_ccb_start(), since a `fifo full' condition is
handled internally in the lower layer, i.e. the transaction started
from ldstart() has been queued at the lower layer and must be taken
off ld's disk queue.
in a different fashion. Individually, they have the same functionality,
but their layout is different. An example of such a chipset is
the Promise 203xx.
To be able to deal with this, transform the cmd and dma bus_space handles
into an array of handles, each seperately created with bus_space_subregion.
The code generated by using the extra indirection shouldn't change much,
since the extra indirection is negated by having the offset calculation
already done in bus_space_subregion. E.g.
bus_space_write_4(tag, handle, offset, value)
becomes
bus_space_write_4(tag, handles[offset], 0, value)
Reviewed by Manuel Bouyer. Tested on wdc_isa, wdc_pcmcia, viaide, piixide (i386)
and on cmdide (sparc64).
interrupt while SRST is asserted. Work around by blocking interrupts while
SRST is asserted, and clearing any pending interrupt before unblocking.
Fix kern/23529 from Michael Hertrick.
copyin() or copyout().
uvm_useracc() tells us whether the mapping permissions allow access to
the desired part of an address space, and many callers assume that
this is the same as knowing whether an attempt to access that part of
the address space will succeed. however, access to user space can
fail for reasons other than insufficient permission, most notably that
paging in any non-resident data can fail due to i/o errors. most of
the callers of uvm_useracc() make the above incorrect assumption. the
rest are all misguided optimizations, which optimize for the case
where an operation will fail. we'd rather optimize for operations
succeeding, in which case we should just attempt the access and handle
failures due to insufficient permissions the same way we handle i/o
errors. since there appear to be no good uses of uvm_useracc(), we'll
just remove it.
and mask ops since smc83c170 chips access them in 32bit width with proper
byteswap mechanism, so that all #if BYTE_ORDER in descriptors can be removed.
While here, do some slight optimizations in epic_start() and epic_intr().
Otherwise we would bump queue_freeze several times, but the kernel thread would
decrease it only one time.
Fix PR kern/23377 by Mattias Karlsson, and should also fix similar problem
reported by Mihai CHELARU and Bruce J.A. Nourish on current-users.
thorpej