141599 Commits

Author SHA1 Message Date
kib
d5ddd71e3f When doing rfork(0), i.e. separating curproc VM from any other user of
the same vmspace, decrement the reference count of the shared LDT instead
of a newly-made copy. Code factually removed LDT from the process that
did rfork(0).

Introduce user_ldt_deref() function that does decrement of refcount for
the struct proc_ldt, and call it in the rfork(0) case on the shared LDT.

Reviewed by:	jhb
MFC after:	1 week
2008-09-12 09:53:29 +00:00
kib
0b37248817 The user_ldt_alloc() function shall return with dt_lock locked.
The user_ldt_free() function shall return with dt_lock unlocked.
Error handling code in both functions do not handle this, fix it by
doing necessary lock/unlock.

While there, fix minor style nits.

MFC after:	1 week
2008-09-12 09:51:11 +00:00
kib
85102b5b18 Remove warning about static LDT segment allocation. Applications
continue using it after ~7 years since warning was introduced, and there
is no reason to discourage them.

MFC after:	1 week
2008-09-12 09:49:01 +00:00
kientzle
f86ac1c574 Portability: Don't use symbolic constants when the values are more
standard than the names are.  Remove some trailing whitespace.
2008-09-12 05:33:00 +00:00
kientzle
ccdf5b420c Portability: Not everyone is lucky enough to have ftruncate() 2008-09-12 04:08:11 +00:00
kientzle
96223dee11 Portability: Support platforms that lack (struct stat).st_blksize or SSIZE_MAX
Of course, FreeBSD has both.
2008-09-12 04:03:34 +00:00
kientzle
93721eec26 Windows compatibility: The stub replacement for compression_program
(which always returns an error when invoked) needs its prototype.
2008-09-12 03:53:54 +00:00
takawata
4ffcdcfe35 Connect ufoma(4) to build and,
disable handsfree interface that is not yet ported to newtty.
2008-09-12 03:45:08 +00:00
danger
34902adea1 Sweep this man page a bit:
- new sentence = new line
- use .Dq macro to quote words
- some minor rewording

MFC after:	2 weeks
2008-09-11 22:11:41 +00:00
jhb
1d4a561a28 Add a 'hw.pci.mcfg' tunable. It can be set to 0 to disable memory-mapped
PCI config access.
2008-09-11 21:42:11 +00:00
cognet
5b5d402e31 Bandaid: disable interrupts to make sure intr_enabled and the IER register
are in sync. I'm not sure why it is needed, and why it wouldn't be on other
arm platforms, but it prevents a lockup under heavy I/O.
2008-09-11 20:43:38 +00:00
cognet
a815eb98d3 Remove the unused field "pc_prvspace" from the MD fields for the struct
pcpu. There's not even a thing such as a "struct pcup".
While I'm there, remove a comment that makes no sense for arm.

Spotted out by:	Mark Tinguely
2008-09-11 20:39:46 +00:00
roberto
2834eba255 Makefile.inc already defines OPENSSL if crypto is available/wanted.
PR:		bin/127296
Submitted by:	oliver
MFC after:	3 days
2008-09-11 20:32:06 +00:00
ed
8cc770d042 Fix two small typo's in comments in the nullfs vnops code.
Submitted by:	Jille Timmermans <jille quis cx>
2008-09-11 20:15:34 +00:00
jhb
8470014e06 Update the comments above the 0xcf9 register reset attempt to match the
code.  We only attempt a single reset using this method (a "hard" reset),
and we use two writes to ensure there is a 0 -> 1 transition in bit 2 to
force a reset.

MFC after:	1 week
2008-09-11 18:33:57 +00:00
rdivacky
74e5140a73 The ERESTART to EINTR conversion is already done in
kern_select so there is no need to repeat it in
linux_select().

Submitted by:	Dmitry Chagin <dchagin@>
MFC after:	1 week
Approved by:	kib (mentor)
2008-09-11 15:28:28 +00:00
raj
9c3efb4d4b ARM nexus style(9) improvements. 2008-09-11 12:39:54 +00:00
raj
df986188e9 ARM interrupts improvements.
- Fix nexus_setup_intr() abuse of setting up multiple IRQs in one go. Calling
  arm_setup_irqhandler() in loop is bogus, as there's just one cookie given
  from the caller and it is overwritten in each iteration so that only the
  last handler's cookie value prevails.

- Proper intr masking/unmasking handling: the IRQ source is masked at PIC level
  only after the last handler has been removed from the list.

Reviewed by:	cognet, imp, sam, stass
Obtained from:	Grzegorz Bernacki gjb ! semihalf dot com
2008-09-11 12:36:13 +00:00
raj
30165fb1e1 IXP425: split handling of the two QMGR interrupts so they are separately
managed. Adjust ixpqmgr_{attach,detach} to comply with device_* interface.

Reviewed by:	cognet, imp, sam, stass
Tested by:	cognet
2008-09-11 12:17:21 +00:00
kib
8fbe72145e Document dev_clone eventhandler interface 2008-09-11 11:43:47 +00:00
kib
26d6e1da5d Add a first draft of the cdevpriv(9) documentation 2008-09-11 11:41:33 +00:00
kib
d397045a7c Document make_dev_credf(9), destroy_dev_sched(9) and destroy_dev_drain(9)
functions.

Reviewed by:	brueffer
2008-09-11 11:38:48 +00:00
kib
e3f791c62f Update insmntque(9) manpage, bringing it close to reality.
Reviewed by:	brueffer
2008-09-11 11:36:21 +00:00
adrian
0d6a982e10 Fix the device name spacing.
The old logic padded the device name out but assumed the unit number was one digit
long; this fails for things like SATA devices which (for me) begin at ad10.

Assemble the full device name in a temporary buffer and then calcluate padding
based on that string.
2008-09-11 09:55:54 +00:00
marius
1c6ab85afc Revert r177108 and restore r60506 for sparc64 as long as libstand
isn't fixed to only open the network device once and not do a open
and close dance on every file access; the firmwares of newer sparc64
machines perform an auto-negotiation with every open which in turn
causes netbooting to take horribly long if we open and close the
device over and over again.
2008-09-10 20:53:22 +00:00
marius
56df8196be Add drivers for the power management devices found on Fireplane/
Safari- and JBus-based machines. Currently the main purpose of
these drivers is debugging of the resource allocation on nexus(4)
and the register content of these devices though.
2008-09-10 20:22:27 +00:00
csjp
1fa65beb80 Make sure the TTY has not disappeared out from under us before calling
ttydevsw_outwakeup().  This should fix panics which occur after remote
login sessions timeout during moderate TTY activity.  An example of
where this might occur is where a pending write to the terminal is
occurring while sshd(8) is shutting down the TTY after a TCP timeout.

Submitted by:	ed
2008-09-10 20:12:10 +00:00
marius
5f92bd8a23 Work around Cheetah+ erratum 34 (USIII+ erratum #10) by relocating
the locked entry in it16 slot 0, which typically is occupied by the
PROM, and manually entering locked entries in slots != 0.

Thanks to Hubert Feyrer for donating the Blade 2000 this change was
developed on.
2008-09-10 20:07:08 +00:00
bz
83b4eaaa16 mld_timerresid() returns ms so instead of doing the maths in usec
and then dividing down to ms, do the maths in ms.

Obtained from:	NetBSD mld6.c rev. 1.47
MFC after:	2 months
2008-09-10 19:42:13 +00:00
jhb
af0471aaec Teach WITNESS about the interlocks used with lockmgr. This removes a bunch
of spurious witness warnings since lockmgr grew witness support.  Before
this, every time you passed an interlock to a lockmgr lock WITNESS treated
it as a LOR.

Reviewed by:	attilio
2008-09-10 19:13:30 +00:00
jhb
bc107b3828 Disable the inline assembly crc32 routine and use the C version instead.
The assembly version is reported to be broken on 5.x+.

PR:		kern/100425
Submitted by:	Rashid N. Achilov  shelton www.granch.ru
MFC after:	1 week
2008-09-10 18:42:19 +00:00
jhb
70e9d3b331 Resurrect the sbni(4) driver. Someone finally tested the MPSAFE patches and
the driver worked ok with them.

Tested by:	friends of yar
2008-09-10 18:36:58 +00:00
sbruno
0d2f69ae42 Beginning of overhaul of fwcontrol:
-  Documentation of send_phy_config()
-  cleanup of malloc's() and added error checking throughout
-  new capability to iterate over multiple firewire buses
-  update usage() display
-  cleanup command line parsing to allow out of order switches
-  cleanup command line parsing to allow multiple switches per invocation
-  cleanup grammar of man page a bit
-  add some ranges to the man page to indicate what values are valid

Since fwcontrol's code is the same across 6/7/head this can be
applied to all branches after the MFC period.

Reviewed by:	Dieter freebsd@sopwith.solgatos.com
Approved by:	mentor Scott scottl@samsco.org
MFC after:	60 days
2008-09-10 18:09:52 +00:00
jhb
ddef310111 Some K8 chipsets don't expose all of the PCI devices on bus 0 via PCIe
memory-mapped config access.  Add a workaround for these systems by
checking the first function of each slot on bus 0 using both the
memory-mapped config access and the older type 1 I/O port config access.
If we find a slot that is only visible via the type 1 I/O port config
access, we flag that slot.  Future PCI config transactions to flagged
slots on bus 0 use type 1 I/O port config access rather than memory mapped
config access.
2008-09-10 18:06:08 +00:00
jhb
fd768740df Various whitespace fixes. 2008-09-10 17:59:21 +00:00
jhb
069152e433 Add a proper detach method to the est(4) driver using cpufreq_unregister().
MFC after:	1 week
2008-09-10 17:41:41 +00:00
obrien
5bde7108b7 If arm.h is going to define WCHAR_TYPE, don't assume WCHAR_TYPE_SIZE
doesn't already have a definition.

Reported by:	imp
2008-09-10 15:47:31 +00:00
jasone
f4190c00f8 Use PAGE_{SIZE,MASK,SHIFT} from machine/param.h rather than hard-coding
page size and using sysconf(3).

Suggested by:	marcel
2008-09-10 14:27:34 +00:00
trasz
9303940ffe Remove VSVTX, VSGID and VSUID. This should be a no-op,
as VSVTX == S_ISVTX, VSGID == S_ISGID and VSUID == S_ISUID.

Approved by:	rwatson (mentor)
2008-09-10 13:16:41 +00:00
bms
11fabd36c2 Add a missing break statement; IFDATA_LINKSPECIFIC would fall through
to IFDATA_DRIVERNAME otherwise.

Reviewed by:	brooks
MFC after:	1 week
2008-09-10 13:08:37 +00:00
maxim
2473c0ee1e o Correct a comment: a test file size is a four pages not three. 2008-09-10 09:32:25 +00:00
kmacy
e3882f3608 Get initial bootstrap of APs working under xen.
Note that the APs still blow up in sched_throw().

MFC after:	1 month
2008-09-10 07:11:08 +00:00
gonzo
977bb41a58 Fix path in all includes: remove /mips32 subdirectory 2008-09-10 03:49:08 +00:00
weongyo
120d55b5a1 rename flags and add a ZYD_FLAG_DETACHING flag to indicate we're
detaching that when the USB is pulled out forcibly during the driver is
running background scan, a page fault can be occurred even if we called
usb_rem_task() when detaching.  It looks like a kind of races.
2008-09-10 03:40:51 +00:00
kmacy
49e85d35c3 - Fix regression with GETMEM
- Remove gratuitous bswap macros
- check for rev 3 with t3b

Obtained from:	Chelsio Inc.
MFC after:	3 days
2008-09-10 01:10:17 +00:00
rpaulo
fe0d90253c Allow building k8temp on i386.
MFC after:	1 week
2008-09-09 22:37:29 +00:00
rpaulo
7e277dfef0 Build k8temp on i386. 2008-09-09 22:32:30 +00:00
remko
2697f0bc73 Add support for the CyberPower CP1500AVRLCD.
PR:		126845
Submitted by:	gavin
Approved by:	imp (mentor, implicit)
MFC after:	1 week
2008-09-09 19:59:51 +00:00
bms
67bb59274c Add support to rc.initdiskless for /conf/T/M/remount_subdir.
This allows the location of the configuration data to be relocated
within the filesystem containing it. A nullfs mount is used in order
to achieve this.

Obtained from:	XORP, Inc.
2008-09-09 18:40:50 +00:00
jasone
c30fff5419 Move call to _malloc_thread_cleanup() so that if this is the last thread,
the call never happens.  This is necessary because malloc may be used
during exit handler processing.

Submitted by:	davidxu
2008-09-09 17:14:32 +00:00