Commit Graph

140641 Commits

Author SHA1 Message Date
rrs
f33a4be603 Fix build breakage - kthread_exit() in 8 now has no arguments
MFC after:	1 week
2008-07-29 09:30:50 +00:00
rrs
e10e4b9d41 - Out with some printfs.
- Fix a initialization of last_tsn_used
- Fix handling of mapped IPv4 addresses
Obtained from:	Michael Tuexen and I :-)
MFC after:	1 week
2008-07-29 09:06:35 +00:00
yongari
ec0068765f style(9) - space after keywords, don't indent case. 2008-07-29 09:02:00 +00:00
yongari
3a922a1bb6 Cache PCI vendor/device ids to avoid unnecessary PCI configuration
space access in device probe.
Also nuke referencing softc in device probe.
2008-07-29 08:49:36 +00:00
yongari
9c99165a6f s/printf/device_printf/g
Don't hard code function name in device_printf() and use __func__.
While I'm here nuke bfe_unit in softc as it's not needed anymore.
2008-07-29 08:32:29 +00:00
roberto
ed14bd1bcf Macbook [Pro] keyboards in AZERTY don't need shift to get to '~'.
MFC after:	3 days
2008-07-29 08:07:27 +00:00
kmacy
b51643853a Factor sockbuf, sockopt, and sockstate out of socketvar.h in to separate headers.
Reviewed by:	rwatson
MFC after:	3 days
2008-07-29 07:45:05 +00:00
bmah
ac76a1b45b Modified release notes: Add one more name to credits on KSE item.
Submitted by:	jhb@
Reviewed by:	brooks@
2008-07-29 04:23:18 +00:00
yongari
e8a74e3558 Add missing jme(4), msk(4), nfe(4), re(4) and stge(4) in NOTES and
ensure that LINT builds include these devices.

Reported by:	Peter Jeremy
2008-07-29 01:15:11 +00:00
emax
ab370dee42 Simplify ubt_isoc_in_complete2(). Also should fix off by 1 bug.
MFC after:	3 months
2008-07-29 00:17:53 +00:00
mav
a5ebe70ed0 Don't use memcpy() to copy several bytes.
Store IDs is host order. It is not so important to bloat code for it.
Combine m_adj() and M_PREPEND() into single M_PREPEND().
2008-07-28 22:22:38 +00:00
jhb
793ef12f8c Re-enable em(4) and igb(4) in NOTES.
PR:		conf/112081
2008-07-28 22:16:58 +00:00
des
504f86aa14 In some cases (such as LC_COLLATE), nb_NO and nn_NO are identical, and it
makes sense to have them both link to no_NO.

In other cases (such as LC_TIME), they differ, and the correct solution
is to have no_NO link to nb_NO, rather than the other way around.o

MFC after:	2 weeks
2008-07-28 21:50:21 +00:00
jhb
11be6dec9b Remove a stale reference to sys/dev/ixgbe/tcp_lro.c. 2008-07-28 21:47:04 +00:00
des
975daa69ac Add collation sequence for no_NO. The same sequence applies to nb_NO
and nn_NO, which are symlinked to no_NO.

The patch in the PR contained a number of errors apparently based on
(sometimes incorrect) pronunciation; for instance, v and w are
distinct letters and should be collated in that order, even if they
are pronounced the same, while <u:> should be collated with u, even
though it is often mispronounced as y.  For lack of a solid reference,
I have taken sv_SE and simply changed the last three letters of the
alphabet.

PR:		conf/51920
MFC after:	2 weeks
2008-07-28 21:33:34 +00:00
jhb
9cf4eed4fa Fix a few bugs with the _gettemp() routine which implements mkstemp(),
mkstemps(), and mkdtemp().
- Add proper range checking for the 'slen' parameter passed to mkstemps().
- Try all possible permutations of a template if a collision is encountered.
  Previously, once a single template character reached 'z', it would not wrap
  around to '0' and keep going until it encountered the original starting
  letter.  In the edge case that the randomly generated starting name used
  all 'z' characters, only that single name would be tried before giving up.

PR:		standards/66531
Submitted by:	Jim Luther
Obtained from:	Apple
MFC after:	1 week
2008-07-28 21:18:59 +00:00
jhb
bed722e078 When choosing a CPU for a thread in a cpuset, prefer the last CPU that the
thread ran on if there are no other CPUs in the set with a shorter per-CPU
runqueue.
2008-07-28 20:39:21 +00:00
mav
c8ae327077 Move inpcb lock higher to protect some nonbinding fields reading.
It fixes nothing at this time, but decided to be more correct.
2008-07-28 19:32:18 +00:00
jhb
3bc77fdd97 Don't attempt authentication at all if it has been disabled via '-a off'.
This works around a bug in HP-UX's telnet client and also gives a much
saner user experience when using FreeBSD's telnet client.

PR:		bin/19405
Submitted by:	Joel Ray Holveck   joelh of gnu.org
MFC after:	1 month
2008-07-28 18:58:16 +00:00
jhb
421b41fe8c Really fix this. 2008-07-28 18:33:43 +00:00
pjd
642dbd51b0 Properly check if td_name is empty and if it is, print process name,
instead of empty thread name.

Reviewed by:	jhb
2008-07-28 18:10:26 +00:00
antoine
41ca50f84b Add files to remove when MK_GNU_CPIO = no 2008-07-28 18:00:45 +00:00
antoine
62eb6bc8c9 Remove a file that is not used and does not exist. 2008-07-28 17:56:37 +00:00
antoine
2288482a5b Remove an empty directory that is already in ObsoleteFiles.inc from
mtree/BSD.usr.dist
2008-07-28 17:42:37 +00:00
jhb
68f0af82de Implement support for cpusets in the 4BSD scheduler.
- When a cpuset is applied to a thread, walk the cpuset to see if it is a
  "full" cpuset (includes all available CPUs).  If not, set a new
  TDS_AFFINITY flag to indicate that this thread can't run on all CPUs.
  When inheriting a cpuset from another thread during thread creation, the
  new thread also inherits this flag.  It is in a new ts_flags field in
  td_sched rather than using one of the TDF_SCHEDx flags because fork()
  clears td_flags after invoking sched_fork().
- When placing a thread on a runqueue via sched_add(), if the thread is not
  pinned or bound but has the TDS_AFFINITY flag set, then invoke a new
  routine (sched_pickcpu()) to pick a CPU for the thread to run on next.
  sched_pickcpu() walks the cpuset and picks the CPU with the shortest
  per-CPU runqueue length.  Note that the reason for the TDS_AFFINITY flag
  is to avoid having to walk the cpuset and examine runq lengths in the
  common case.
- To avoid walking the per-CPU runqueues in sched_pickcpu(), add an array
  of counters to hold the length of the per-CPU runqueues and update them
  when adding and removing threads to per-CPU runqueues.

MFC after:	2 weeks
2008-07-28 17:25:24 +00:00
obrien
ffa6157788 Allow kernel config files to include files other than those in the CWD,
using unquoted paths.
2008-07-28 17:11:57 +00:00
imp
0c7c769464 Export the hardware type (as number and name), the secondary firmware
revision and (on Prism cards) the primary firmware revision via
sysctl.  Move the printing of this information under bootverbose,
since it is relatively easy to get to it now.
2008-07-28 17:00:37 +00:00
jhb
69cc3c8c8a Various and sundry style and whitespace fixes. 2008-07-28 15:52:02 +00:00
edwin
969134af63 Fix text in the comment why we check for ELF32_R_TYPE
Approved by:	bde@
MFC after:	2 days2 days
2008-07-28 12:49:16 +00:00
ed
0307110dbc Fix the email address formats in some posix_spawn(3) manpages.
It seems I made a small bug when writing some of the posix_spawn(3)
manpages. Remove the redundant "Ed Schouten", which broke the AUTHORS
section.

Approved by:	philip (mentor, implicit)
2008-07-28 09:36:56 +00:00
ed
010ac0ae05 Fix a small typo in the procstat(1) manpage: messsage queue.
Approved by:	philip (mentor)
MFC after:	3 days
2008-07-28 08:01:24 +00:00
mav
ac42a1aad0 Some style and assertion fixes to the previous commits hinted by rwatson.
There is no functional changes.
2008-07-28 06:57:28 +00:00
alc
88e89bb2e1 Correct an off-by-one error in the previous change to pmap_change_attr().
Change the nearby comment to mention the recursive map.
2008-07-28 05:41:35 +00:00
alc
93ecf09bb7 Don't allow pmap_change_attr() to be applied to the recursive mapping. 2008-07-28 04:59:48 +00:00
alc
4dbcd0e70f Don't allow pmap_change_attr() to be applied to the recursive mapping. 2008-07-28 04:13:49 +00:00
alc
8c160a5c01 Add a check for 1GB page mappings to pmap_change_attr() so that it fails
gracefully.  (On K10 family processors the direct map is implemented using
1GB page mappings.)
2008-07-28 03:58:49 +00:00
yongari
d499ae8522 Fix buffer discard index.
While I'm here dicard all buffers if errored frame is part of
multi-segmented frames.

Pointed out by:	sephe
Reviewd by:	sephe
MFC after:	3 days
2008-07-28 02:37:15 +00:00
davidxu
bcd09d8c85 Add manual pages for posix_spawn() functions.
PR:	standards/122051
2008-07-28 02:22:19 +00:00
kmacy
70741e0245 - track maximum wait time
- resize columns based on actual observed numerical values

MFC after:	3 days
2008-07-27 21:45:20 +00:00
mav
b6bbe78d99 According to in_pcb.h protocol binding information has double locking.
It allows access it while list travercing holding only global pcbinfo lock.
2008-07-27 20:48:22 +00:00
mav
8a791bfa67 According to in_pcb.h protocol binding information has double locking.
It allows access it while list travercing holding only global pcbinfo lock.
2008-07-27 20:30:34 +00:00
alc
369f479043 Style fixes to several function definitions. 2008-07-27 18:18:50 +00:00
alc
4455d56f31 Enhance pmap_change_attr(). Use pmap_demote_pde() to demote a 2MB page
mapping to 4KB page mappings when the specified attribute change only
applies to a portion of the 2MB page.  Previously, in such cases,
pmap_change_attr() gave up and returned an error.

Submitted by:	Magesh Dhasayyan
2008-07-27 17:32:36 +00:00
pjd
3f1807709d Assert for exclusive vnode lock in vinactive(), vrecycle() and vgonel()
functions.

Reviewed by:	kib
2008-07-27 11:48:15 +00:00
pjd
4dd19696a7 - Move vp test for beeing NULL under IGNORE_LOCK().
- Check if panicstr isn't set, if it is ignore the lock. This helps to avoid
  confusion, because lockmgr is a no-op when panicstr isn't NULL, so
  asserting anything at this point doesn't make sense and can just race with
  other panic.

Discussed with:	kib
2008-07-27 11:46:42 +00:00
julian
4c1af4639a Add the ability to add new addresses for interfacesto just one FIB
(Other more specific related options will follow)
This allows one to set multiple p2p links to the same place
and select which to use by having each in different FIBS.
2008-07-27 01:29:28 +00:00
sam
1fc2ef0208 don't deauth a station because it sends a ps-poll w/ a bogus aid in it;
turns out some devices do this and since we otherwise validate the station
is associated and don't use the aid for anything being lenient here allows
them to function

Submitted by:	Chris Zimmermann
MFC after:	2 weeks
2008-07-26 23:50:27 +00:00
mav
8a94b46be4 Increase UDBHASHSIZE from 16 to 128 items.
Previous value was chosen 10 years ago and not very effective now.
This change gives several percents speedup on 1000 L2TP mpd links.
2008-07-26 23:07:34 +00:00
mav
0023432ccb According to in_pcb.h protocol binding information has double locking.
It allows access it while list travercing holding only global pcbinfo lock.
This relaxed locking noticably increses receive socket lookup performance.
2008-07-26 21:12:00 +00:00
darrenr
c85943e33b 2020447 IPFilter's NAT can undo name server random port selection
(fix output port range, was a random number in [0,max-min]
	 (byteswapped on litle endian), instead of [min,max])

Submitted by:	darrenr
2008-07-26 19:46:00 +00:00