sys/dev/pckbc/files.pckbc." This is true of sparc, so define __HAVE_NWSCONS,
and remove the code in pckbc_machdep_cnattach() that was conditional on its
not being defined and that is no longer needed (since with __HAVE_NWSCONS,
pckbc_cnattach() will call pckbd_cnattach() itself if necessary).
of bounds, then complain and recover by freeing all of the descriptors.
That will usually provoke additional complaints---see the next
paragraph.
If the firmware returns to us an RSS descriptor that is in-bounds
but out of service (id_node == NULL), then gripe and get out of
the interrupt handler. ***This will still leak RSS descriptors.***
I'm working on a better solution. Possibly I will just free all of
the descriptors.
bug report from Simon Burge: the firmware seems to lock up.
If rate-adaptation is disabled on a Lucent card, then do all the
rate-adaptation work *except* for writing the new transmit rate,
since I suspect that is what locks-up the firmware.
and writes of the EEPROM are too fast so invalid data is returned. So we
increase the DELAY()s. The right thing would be to check a ready bit
on the E^2 if such a thing exists. (checked by potr)
the latter is not a appropriate place to do so and it broke vfork.
- deactivate pmap before calling cpu_exit() to keep a balance of
pmap_activate/deactivate.
up attached at the same time, and so that they can both prevent the forthcoming
iomdkbc(4) attaching to the keyboard slot.
[ file missed in last commit ]
which is zero by default.
perform rbtree sanity checks only when it isn't zero
because the check is very heavy weight especially when
there're many entries.
in the case that there's no cached entries,
if kmem_map is already up, allocate a entry from it
so that we won't try to vm_map_lock recursively.
XXX assuming usage pattern of kmem_map.
* detect and store vc2 revision. This is (likely) the important factor in
determining the proper cursor offset, so add a bit of logic that does the
right thing on my revision 0 vc2. This will have to be determined
empirically, revision by revision.
* take a stab at detecting framebuffer depth. SGI does it quite differently,
but their algorithm doesn't work, so we'll do it along the same lines as
linux.
* implement newport_mmap() for use by X. The necessary xsrc commits have been
made (with the notable exception of the config/cf changes, which probably
shouldn't go in until 4.4.0 has been imported).
Therefore lkmdispatch(LKM_E_UNLOAD) must be called to free resources in
case the load function fails.
Reported by Christian Groessler on current-users.
Added software volume control functionality in vraiu_volume(), called from
vraiu_start(). Can be changed by the mixerctl 'inputs.dac' -- 0 is muted,
255 is the level that the driver previously outputted (quite loud on the
IBM Workpad Z50), and everything else is a smooth transition between the
two. Thanks to toddpw@ for lending me the hardware to test these changes.
via ioctl; drop the traditional device minor mapping for those in favour
of setting via lptctl(8)
introduce notion of 'control' device (minor bit 0x100 set); ATM this
device always skips the priming, which allows device open even with
disconnected printer
this also changes the default for interrupt use - polling is used
by default now
When creating a package element via "Store(0xXXX, Index(ArgX, 0xXXX))",
be sure to increment the refcount of the argument so it is not
prematurely deleted. This is a workaround and may appear in a different
form in ACPI-CA. This fixes battery evaluation on Thinkpads that was
broken by fixing the Dell battery state.
Submitted by: Luming Yu <luming.yu at intel.com>
audio device interface:
1) When attempting to match the appropriate mixer control, we weren't
checking the class label, but only the second level label, so for
devices that had both an "inputs.cd" and a "record.cd", for example,
we could never do the right thing except by chance. For this reason,
evidently, all the record masters were labeled (by the underlying
drivers) either "record.record" or "record.volume", to distinguish
from "outputs.master". We'll now accept "record.master", and document
that as the the new preferred way.
2) More importantly, the model was deficient. Selecting a port on many
chips completely disables most of the level controls, which doesn't play
nice with other applications which are trying to use the interface. Now,
selecting a port simply sets which mixer input control shall be changed,
setting state in the audio layer. In other words, the "mixerout" port
is really selected all the time, enabling the final stage mixer, and
setting "gain" sets the level of the appropriate input. It should be
possible for separate applications to each control the mic, dac, and cd
inputs at the same time using this interface, simply by reiterating their
port selections with each change, but applications that don't bother to
do that aren't any worse off than they were before.
The user is expected to set the master output with another application,
dedicated to that task. Though it is now meaningful to select "no port"
with the audio device interface, to manipulate the master output, there's
no particular reason for an audio device consumer to do that. (I added
the capability in order to restore the initial state of the audio device,
for testing purposes. It might also be useful to users of broken binary-
only applications.)
Observe that the mixer device interface (and so, "mixerctl") still
retains all capabilities, including the ability to set the actual input
port on the chip, overriding the level controls. No change is being made
to the mixer device interface. The mixer device simply presents all the
controls on the chip, with no attempt at abstraction, so there are no
bugs there.
The upshot is, that applications that have been trying to use the audio
device interface to change the volume, such as mplayer, now "just work".
I've tested these changes extensively with "eso" and "eap" since first
proposing them on tech-kern last January, and somewhat with "esm" and a
few others. This closes both PR kern/10221, and PR kern/17159.
and uncontrolled growth.
The key fix is from Dan Carasone, who noticed that buf_canfree() was
counting in _bytes_ but freeing in _buffers_, which caused the instant
drop to lowater observed by some users.
We now control the rate of growth; the probability of getting a new
allocation is inversely proportional to the current size of the
cache. This idea is from a long-ago conversation with Kirk McKusick
and, if memory serves, was used for the file-system cache in some
other BSD variant at some point in history.
With growth and shrinkage more or less dealt with, we return the
default maximum cache size to 15%. The default _minimum_ cache size
is raised from 1/16 of the maximum cache size to 1/8, since 1/16 was
chosen when the maximum size was 30% of memory.
Finally, after observing the behaviour of the pagedaemon and the
buffer cache drainer under pathological workloads (e.g. a benchmark
that steps through 75% of available memory backwards) I have moved
the call to buf_drain() to the beginning of the pagedaemon from the
end; if the pagedaemon bogs down, it still won't get run as often
as it should, but at least this way it will see the state of the
free count and free target _before_ the scan step does its thing.
the recently introduced files.ppbus.
Also, let the mvme68k port use it's own lpt implementation for now,
so define USE_ALT_LPT option in std.memv68k.
This makes the mvme68k kernels configure and build again.
obmitted resulting in a compilation failure in the autobuilds.
It effectively removes the now redundant reference to "beep.h" and removes
an associated prototype that wasn't used.
- for in-kernel maps, disable map entry merging so that
unmap operations won't block. (workaround for PR/24039)
- for in-kernel maps, allocate kva for vm_map_entry from
the map itsself and eliminate MAX_KMAPENT and
uvm_map_entry_kmem_pool.
PR#23470, with minor updates by me. This is only the syscall support
from that PR, for now.
Changes: port over fix from FreeBSD for multicast address generation.
Changed bcopy to memcpy. For now, #ifdef notyet the portions of
kern_uuid.c that are meant to be used by (currently nonexistent) other
things in the kernel. Added syscall to COMPAT_FREEBSD as well, though
that's currently not useful, as any program new enough to use this call
also uses other syscalls we don't (yet) emulate.
RAM a chance to boot. Discussed with soren@, tested on customer's board.
NB: While acceptable as a temporary workaround since most (all?) Cobalt
hardware comes with no less than 16Megs on board, changing a kernel load
address or making a boot loader relocatable could be a better solution.
tripping over this getting too large, and suffering other performance
problems due to the lack of good backpressure shrinking the bufcache
when other memory is required. Again, this tunable should be
revisited when the backpressure mechanism has been improved.
sysctl vm.bufcache can be used to manually tune those rare machines
that might need more than this.
See comments in rev 1.106 for more detail.
unknown problem with dcache_inv_range by using a wbinv for now
(similarly for ARM10).
When setting the ARM9 system control register, use the computed
cpuctrlmask value (not 0xffffffff) so that the clocking-mode bits are
not reset to FastBus mode (which isn't very fast).
* Add resolution and depth variables to softc (for use in X driver, although
depth detection is currently problematic),
* Store more information about various chip revisions,
* Implement ioctls to help the X driver detect device presence and geometry.
