10464 Commits

Author SHA1 Message Date
davidxu
13fd270405 1. Always call user callback function in newly created thread, it seems
POSIX implies that the user callback function must be executed in clean
environment.
2. Use newly introduced pthread stubs in libc.
2006-03-07 08:28:07 +00:00
yar
7ba58b694e Add appropriate xrefs.
MFC after:	3 days
2006-03-06 13:15:12 +00:00
yar
f4b4b54f2b Since the whole login.access feature has moved to PAM,
login.access.5 will be installed from the respective PAM
module's src directory.

MFC after:	3 days
2006-03-06 12:31:25 +00:00
yar
4995e053c3 Sync with src/usr.bin/login/login.access.5.
src/usr.bin/login/login.access.5 should be removed from use
because the whole login.access feature has moved to this PAM
module.

MFC after:	3 days
2006-03-06 12:26:43 +00:00
deischen
6401ef60bd Only catch SIGINFO (for dumping thread states) when LIBPTHREAD_DEBUG
is defined in the environment.

Requested by:	jmg & a few others
2006-03-06 05:02:28 +00:00
deischen
e586a0042e Add some more pthread stubs so that librt can use them.
The thread jump table has been resorted, so you need to
keep libc, libpthread, and libthr in sync.

Submitted by:	xu
2006-03-05 18:10:28 +00:00
brueffer
6db6365c65 Remove a useless word.
PR:		94087
Submitted by:	Tadaaki Nagao <nagao@iij.ad.jp>
MFC after:	3 days
2006-03-05 10:32:15 +00:00
keramida
4465edc46a Use intmax_t' instead of plain int' for pid_t casts.
Useful tips from:	ru, bde
Approved by:		pjd
MFC after:		3 days
2006-03-04 15:20:28 +00:00
davidxu
8e36ba772a Use a thread pool to process notification if sigev_notify_attributes
is default and caller does not require dedicated thread. timer needs
a dedicated thread to maintain overrun count correctly in notification
context. mqueue and aio can use thread pool to do notification
concurrently, the thread pool has lifecycle control, some threads will
exit if they have idled for a while.
2006-03-04 00:18:19 +00:00
davidxu
869493f015 save sigev_node pointer. 2006-03-04 00:08:24 +00:00
jcamou
44dc71d01c Sync with actual code.
PR:		docs/87681
Noticed by:	Andreas Kohn <andreas@syndrom23.de>
Reviewed by:	brooks
Approved by:	trhodes (mentor)
MFC after:	3 days
2006-03-03 13:58:43 +00:00
phk
0ef226b65a Fix the %Q printf extension to behave as expected 2006-03-02 08:53:45 +00:00
davidxu
a8886a11b7 1. Fix a race in aio_return.
2. Save and restore syscall errno correctly.
3. Style fix.
2006-03-01 23:41:47 +00:00
davidxu
5d292c8fb4 Style fix. 2006-03-01 23:40:00 +00:00
davidxu
c468432d46 Fix a mutex lock/unlock mismatch. 2006-03-01 23:38:53 +00:00
ume
5e1521cce3 Handle the errors returned by res_querydomain() in same manner.
Reported by:	yar
Tested by:	yar, Rostislav Krasny <rosti.bsd__at__gmail.com>
MFC after:	1 week
2006-03-01 15:53:49 +00:00
davidxu
8ddf0913dd Forgot to revert to use weak symbols when I was debugging, fix it! 2006-03-01 08:50:36 +00:00
davidxu
c652436886 Add missing parameter mq_attr * for mq_open. 2006-03-01 08:06:09 +00:00
davidxu
22fb15cfe9 Bring in my initial version of POSIX realtime extension library.
Current the library implements mqueue, timer and aio with SIGEV_THREAD
notification supported.

Earlier version reviewed by: deischen
2006-03-01 06:37:34 +00:00
davidxu
f29e4c0c75 Disconnect mqueue from buildworld, as I will implement it in seperated
library.
2006-03-01 06:25:46 +00:00
deischen
b028e2897b Const'ify arguments to a couple of functions to fix breakage
with -O2.
2006-02-28 16:02:26 +00:00
delphij
3acbd90cd9 Don't do a time travel to 12006... 2006-02-28 07:36:25 +00:00
davidxu
6fdf2b7d81 Reimplement mutex_init to get rid of compile warning. 2006-02-28 06:06:19 +00:00
deischen
97af573880 Staticize a couple of functions.
Remove a few unused locks.

Remove locks from application namespace.
2006-02-27 22:10:59 +00:00
deischen
2a2559a103 Correct a comment.
Staticize two tables thare are not visible in <resolv.h>
and which are also local in Solaris' libresolv.

Remove two functions that are not referenced in libc nor
anywhere else I can find, not visible in <resolv.h> and
which are also local in Solaris libresolv.
2006-02-27 22:08:16 +00:00
wkoszek
1796fdb900 Fix typo in manual page reference.
Approved by:	cognet (mentor)
MFC after:	3 days
2006-02-26 23:01:11 +00:00
ume
a8acef0bfb Sync inet_net_pton() and inet_net_ntop() with latest BIND9's includes
an IPv6 support.

PR:		kern/93740
Submitted by:	Rudolf Cejka <cejkar__at__fit.vutbr.cz>
Obtained from:	BIND9
MFC after:	1 week
2006-02-26 05:44:21 +00:00
jkoshy
62ed249b1e Add an alias 'unhalted-cycles' denoting cycles where the CPU is
not in a halt or sleep state.
2006-02-25 05:34:00 +00:00
deischen
2c3357a0a6 Fix a race condition introduced when redzones were added. Use an
atomic operation to return and adjust the stack.

Submitted by:	luoqi
2006-02-24 22:03:10 +00:00
ume
d095cafa05 - Just query 'as is', if there is a trailing dot in the name.
- Don't query 'as is' twice.

PR:		bin/62139
Reported by:	Rostislav Krasny <rosti.bsd__at__gmail.com>
Tested by:	Rostislav Krasny <rosti.bsd__at__gmail.com>
Obtained from:	BIND9 (with some modification)
MFC after:	1 week
2006-02-24 16:53:42 +00:00
deischen
4dd473489e Eliminate a race condition in timed waits (cv, mutex, and sleeps).
MFC Candidate.

PR:	93592
2006-02-23 21:34:08 +00:00
yar
d3fcb2cf25 Tell the truth about a) how to get the first fileid, and
b) what return values from kldstat(2) can be expected.

Bump .Dd.

MFC after:	3 days
2006-02-22 15:11:12 +00:00
davidxu
bb43155dbb Disable POSIX SIGEV_THREAD notification support, soon we will have a common
library which can be reused both for libthr and libpthread.
2006-02-22 02:52:22 +00:00
davidxu
91fda076fb Read cycle count before sending signal, this fixes a userland thread
suspension bug.

MFC after: 3 days
2006-02-20 09:02:40 +00:00
des
748c235cfd Add utility functions for checking if a given kernel module is loaded,
and loading it.
2006-02-18 11:25:28 +00:00
brd
ee6a29f334 - Massage wording to make it easier to understand and fix some grammar.
Submitted by:	sbahra at gwu dot edu
Reviewed by:	ru@
Approved by:	ceri@
MFC after:	3 days
2006-02-16 09:56:31 +00:00
deischen
63bc390672 Don't forget to initialize a tailq before using it.
MFC candidate
Noticed by:	luoqi
2006-02-16 01:33:36 +00:00
davidxu
70ea4e1323 Rework last change of pthread_once, create a function _thr_once_init to
reinitialize its internal locks.
2006-02-15 23:05:03 +00:00
davidxu
44a8ce5006 After fork(), reinitialize internal locks for pthread_once(). 2006-02-15 13:41:02 +00:00
gnn
a3483ac2d7 Clean up some descriptions and remove ambiguities in the language.
Add explanations to the examples.

MFC after:	1 week
2006-02-14 13:02:00 +00:00
kientzle
68f2d4ef56 Minor fixes to the code that generates an internal ustar filename
for Pax extended attribute entries.
2006-02-14 04:05:03 +00:00
kientzle
ad0af0f704 Extract device number information from SVR4 CPIO archives.
Without this, you cannot properly restore device node entries
from such archives.

Thanks to: Steve 'dillo Okay for reporting this oversight.
2006-02-14 04:00:14 +00:00
kan
b1ec908186 Remove extra slash from pty slave device name returned by ptsname. 2006-02-13 00:04:04 +00:00
rwatson
cea51dc68c Update copyright for 2006.
MFC after:	3 days
2006-02-11 19:21:39 +00:00
rwatson
f2e35917b9 The uma_zone data structure defines the size of its uz_cpu[] array as 1,
but then sizes the containing data structure at run-time to make room
for per-cpu cache data.  Modify libmemstat to separately allocate a
buffer to hold per-cpu cache data, sized based on the run-time mp_maxid
variable when using libkvm to access UMA data.  This avoids reading
invalid cache data from beyond the end of the uma_zone data structure
on the stack, which can result in invalid statistics and/or reads from
invalid kernel addresses.

Foot target practice by:	ps
MFC after:			3 days
2006-02-11 19:19:29 +00:00
rwatson
c1cbeeaeb1 When reporting an error reading from UMA per-cpu cache pointers using KVM,
return a KVM error rather than an out of memory error, so that the caller
reports the KVM error state.  This replaces a misleading error message
with a more accurate although equally confusing one.

MFC after:	3 days
2006-02-11 18:55:03 +00:00
rwatson
35f263ed80 Read all_cpus variable out of kmem, and validate CPUs against the all_cpus
cpu mask before looking at the cache entries for the CPU.  For systems
with sparse CPU id arrays, this skips otherwise uninitialized cache
structures.

MFC after:	3 days
2006-02-11 18:44:37 +00:00
rwatson
244cb93d72 Correct a typo in the extraction of zone information from UMA using kmem:
bytes = allocated - freed, not bytes = allocated = freed.

MFC after:	3 days
2006-02-11 16:54:00 +00:00
davidxu
b9945d52e2 Use ps_linfo to retrieve LWP info, current it is used to retrieve
signal mask and pending signals.
2006-02-07 02:55:34 +00:00
davidxu
6cfb2f2fce Use ps_linfo to retrieve LWP info. 2006-02-07 02:51:25 +00:00