| Age | Commit message (Collapse) | Author |
|---|
|
sufficient exclusion on alpha, leading to self-locking panics.
|
|
by aligning the stack gap. Fixes gettext port build (and probably many others).
ok miod@, tedu@, tested by sthen@
|
|
machines with more than 4GB of memory. Make sysctl(8) print values obtained
using these sysctls for hw.physmem and hw.usermem.
ok krw@, miod@
|
|
|
|
in rev 1.141. From Christian Ehrhardt and Pedro Martelletto.
OK hshoexer@ miod@
|
|
subtly different from CIRCLEQ, it is possible, when emptying the whole
timeout chain, to end up with CIRCQ_EMPTY being false, and bad things
happen. Back to the drawing board...
|
|
the flag argument to the m_*get* functions. Since PR_LIMITFAIL
can cause us to return NULL if hit the limits.
ok claudio@
|
|
ten minutes (same behaviour as nmea(4)). If the cable is pulled, the sensor
no longer keeps the OK status forever. Tested with a PEX511 as MSTS sender.
|
|
radio clocks can transmit over serial ports. This is implemented as a
line discipline similar to nmea(4) and provides a timedelta sensor.
See http://www.meinberg.de/english/specs/timestr.htm for details on the
MSTS format.
ldattach(8) is extended to support the "msts" line discipline and two stopbits
(which some radio clocks, e.g. the C51 use). Do a "make includes" before your
next system build.
Initially from Maurice Janssen based on nmea(4). "go for it" deraadt
|
|
|
|
returning EINVAL, you'll get ENOSYS. No serious code has used this system
call in at least fifteen years.
The libc stub will be removed at the next major crank time.
ok henning@ deraadt@ krw@ toby@
|
|
|
|
|
|
|
|
clocks with drifts larger than 500ppm to be corrected.
|
|
always fails". Introduce vop_generic_lookup(), which is a trivial lookup
routine that always fails and use that instead, zap the redundant copies.
ok toby@, tedu@, art@
|
|
ok tedu@ kettenis@
|
|
there in NetBSD and FreeBSD trees.
Tested by thib@ who found that it shaved 18min wall clock time of coping a 20G file.
Been in snaps for some time
"looks ok" markus@ "makes sense" blambert@ ok claudio@ thib@
|
|
|
|
remove PNORELOCK flag, as PNORELOCK is used for msleep
ok art@ thib@
|
|
creation of threads which may have caused multile syswq threads to start.
ok dlg
|
|
|
|
they where used to keep "offsets" of arguments too the VOP's
for use by bypass routines, these have been unused now for
quite some time;
Tested on sparc64/amd64 (shaves about 1k of GENERIC).
ok beck@
|
|
machines where the stack grows up. Fixes systrace(1) on hppa.
ok deraadt@
|
|
|
|
|
|
remove pool_cache code. it was barely used, and quite complex. it's
silly to have both a "fast" and "faster" allocation interface. provide
a ctor/dtor interface, and convert the few cache users to use it. no
caching at this time.
use mutexes to protect pools. they should be initialized with pool_setipl
if the pool may be used in an interrupt context, without existing spl
protection.
ok art deraadt thib
|
|
Since mutexes mess around with spl levels, and the sched-lock isn't a
mutex, we need to make sure to fix the IPL when msleep does the locking.
ok art.
|
|
parameter. The mutex is unlocked just before sleep and relocked after
unless P_NORELOCK is in flags, in which case it is left unlocked.
ok art@.
|
|
mfs. OK deraadt@
|
|
|
|
accurate but more expensive nanotime(), the loss of precision shouldn't matter.
Inspired from the other *BSD which did a similar change.
ok tedu@ millert@ henning@ deraadt@
|
|
bad idea to have functions that react unpredictably.
|
|
|
|
|
|
|
|
|
|
syscalls as NOLOCK and MPSAFE. The flags have slightly different semantics:
NOLOCK - the syscall doesn't grab any locks whatsoever.
MPSAFE - the syscall deals with its own locking.
What this means in practice is that NOLOCK syscalls can always be done
without the biglock. The MPSAFE syscalls can be done without the biglock
on CPUs that don't handle interrupts that require biglock (to preserve
lock ordering).
deraadt@ ok
|
|
this reduces kernel size quite a bit. ok claudio
|
|
sys/dev/pci/pciide.c from naddy@
|
|
suggested by otto.
|
|
ten minutes (or if a serial GPS is disconnect and we get no NMEA sentences
at all).
ok otto
|
|
code. At this moment all architectures get the copy of the old code
except i386 which gets a new shiny implementation that doesn't spin
at splhigh (doh!) and doesn't try to grab the biglock when releasing
the biglock (double doh!).
Shaves 10% of system time during kernel compile and might solve a few
bugs as a bonus.
Other architectures coming shortly.
miod@ deraadt@ ok
|
|
|
|
combined, this lets us use crypto before the thread is running
and therefore cryptoraid can attach nice and early.
ok/testing deraadt mbalmer marco
|
|
"previous" device still being there, so if we detach a device, we really
have to start from scratch again (like the comment on this code actually
says we should). Simplifies the code as a bonus.
ok dlg@, deraadt@
|
|
match the comment. OK deraadt@
|
|
|
|
|
|
ok miod@ krw@
|
