Change all variables definitions (int foo) in sys/sys/*.h to variable
declarations (extern int foo), and compensate in the appropriate locations.

splassert(IPL_TTY) where comments tell us we should

First round of __P removal in sys

Move SET/CLR/ISSET macros to param.h. fgsch@ and millert@ ok

Increment reference count when copying session pointer pr#2112. ok deraadt@

Replace inclusion of <vm/foo.h> with the correct <uvm/bar.h> when necessary.
(Look ma, I might have broken the tree)

Setting the VTIME to a value larger than 214 lead to an integer overflow
in the calculation of the sleep timeout which could lead to a panic.

Instead of doing the very error-prone, and in more than

Setting the VTIME to a value larger than 214 lead to an integer overflow
in the calculation of the sleep timeout which could lead to a panic.

Instead of doing the very error-prone, and in more than one way incorrect,
sleep time calculation, allocate an external timeout and let it wake us.

Reported by: Raan Young <raan@graand-visions.com>

show more ...

Random cleanup (I don't like Ms in my tree).

Get rid of REAL_CLISTS. It was never implemented and the tentacles are
everywhhere.

tty stats sysctl, deraadt@ok

Casting magic to avoid warnings on alpha.

port kqueue changes from freebsd, plus all required openbsd glue.
okay deraadt@, millert@
from jlemon@freebsd.org:
extend kqueue down to the device layer, backwards compatible approach
suggested by p

port kqueue changes from freebsd, plus all required openbsd glue.
okay deraadt@, millert@
from jlemon@freebsd.org:
extend kqueue down to the device layer, backwards compatible approach
suggested by peter@freebsd.org

show more ...

Many drivers use a timeout on ttrstrt, make this a generic interface.

Changes to exit handling.

cpu_exit no longer frees the vmspace and u-area. This is now handled by a
separate kernel thread "reaper". This is to avoid sleeping locks in the
critical path of cpu_exit

Changes to exit handling.

cpu_exit no longer frees the vmspace and u-area. This is now handled by a
separate kernel thread "reaper". This is to avoid sleeping locks in the
critical path of cpu_exit where we're not allowed to sleep.

From NetBSD

show more ...

Iiieeek! This kludge was almost funny.

Probably something from the stone age. Kill it.

Use PAGE_SIZE instead of NBPG.

remove the "ifdef pmap_resident_count" hack and replace it with a macro.
change some &vm->vm_pmap to vm->vm_map.pmap

nasty indent

Please GCC 2.8's harsher view of good style

bad types; wileyc@sekiya.twics.co.jp

Count characters even when !OPOST and FLUSHO.
Don't output the \r for ONLCR if FLUSHO; mycroft

in ttywait(), double check if we need to sleep before sleeping

Updates for VFS Lite 2 + soft update.

When the input queues are flushed on a blocked tty, make sure to unblock
it, if it's currently in a blocked state due to input flow control. This
was typically seen with drivers implementing th t_hw

When the input queues are flushed on a blocked tty, make sure to unblock
it, if it's currently in a blocked state due to input flow control. This
was typically seen with drivers implementing th t_hwiflow function, and
starting of ppp (line discipline changes flushes the input queue).
Independently discovered and fixed slightly different than NetBSD PR#4227.

show more ...

back out vfs lite2 till after 2.2