On 2012/7/27 21:30, Bruce Evans wrote:
On Fri, 27 Jul 2012, Gleb Smirnoff wrote:
On Fri, Jul 27, 2012 at 10:32:55PM +1000, Bruce Evans wrote:
B> I just noticed that there is a technical problem -- the count is read
B> unlocked in the KASSERT. And since the comparision is for equality,
B> if yo
Author: davidxu
Date: Tue Jul 31 02:00:37 2012
New Revision: 238928
URL: http://svn.freebsd.org/changeset/base/238928
Log:
When a thread is blocked in direct write state, it only sets PIPE_DIRECTW
flag but not PIPE_WANTW, but FIFO pipe code does not understand this internal
state, when a FIF
Author: davidxu
Date: Tue Jul 31 05:48:35 2012
New Revision: 238936
URL: http://svn.freebsd.org/changeset/base/238936
Log:
I am comparing current pipe code with the one in 8.3-STABLE r236165,
I found 8.3 is a history BSD version using socket to implement FIFO
pipe, it uses per-file seqcount
On 2012/7/31 15:22, Giovanni Trematerra wrote:
On Tue, Jul 31, 2012 at 7:48 AM, David Xu wrote:
Author: davidxu
Date: Tue Jul 31 05:48:35 2012
New Revision: 238936
URL: http://svn.freebsd.org/changeset/base/238936
Log:
I am comparing current pipe code with the one in 8.3-STABLE r236165
On 2012/8/2 3:29, Bruce Evans wrote:
On Wed, 1 Aug 2012, Giovanni Trematerra wrote:
On Tue, Jul 31, 2012 at 10:21 AM, David Xu
wrote:
...
The old code broke some history semantic of FIFO pipe, you can try
the test
tool /usr/src/tools/regression/poll/pipepoll, try it before and
after my
On 2012/8/2 16:12, Bruce Evans wrote:
On Thu, 2 Aug 2012, David Xu wrote:
On 2012/8/2 3:29, Bruce Evans wrote:
On Wed, 1 Aug 2012, Giovanni Trematerra wrote:
...
[gianni@bombay] /usr/src/tools/regression/poll#./pipepoll
1..20
not ok 17 FIFO state 6a: expected POLLHUP; got POLLIN | POLLHUP
On 2012/8/2 22:17, Bruce Evans wrote:
On Thu, 2 Aug 2012, David Xu wrote:
On 2012/8/2 16:12, Bruce Evans wrote:
...
I made similar patches for sockets (to set POLLHUP on hangup (now in
-current) and to not set POLLIN on hangup unless there is still data
to be read). I started killing
Author: davidxu
Date: Mon May 21 03:06:31 2012
New Revision: 235714
URL: http://svn.freebsd.org/changeset/base/235714
Log:
Print key value, an index, otherwise we don't know which key is allocated.
Modified:
head/gnu/usr.bin/gdb/libgdb/fbsd-threads.c
Modified: head/gnu/usr.bin/gdb/libgdb/fbs
Author: davidxu
Date: Sun May 27 01:24:51 2012
New Revision: 236135
URL: http://svn.freebsd.org/changeset/base/236135
Log:
Return EBUSY for PTHREAD_MUTEX_ADAPTIVE_NP too when the mutex could not
be acquired.
PR: 168317
MFC after:3 days
Modified:
head/lib/libthr/thread/thr_mutex
Author: davidxu
Date: Wed May 30 01:52:53 2012
New Revision: 236275
URL: http://svn.freebsd.org/changeset/base/236275
Log:
MFC r236135:
Return EBUSY for PTHREAD_MUTEX_ADAPTIVE_NP too when the mutex could not
be acquired.
PR: 168317
Modified:
stable/9/lib/libthr/thread/thr_mutex.
Author: davidxu
Date: Wed May 30 01:54:14 2012
New Revision: 236276
URL: http://svn.freebsd.org/changeset/base/236276
Log:
MFC r236135:
Return EBUSY for PTHREAD_MUTEX_ADAPTIVE_NP too when the mutex could not
be acquired.
PR: 168317
Modified:
stable/8/lib/libthr/thread/thr_mutex.
On 2012/6/12 6:05, Pawel Jakub Dawidek wrote:
Author: pjd
Date: Mon Jun 11 22:05:26 2012
New Revision: 236935
URL: http://svn.freebsd.org/changeset/base/236935
Log:
fdgrowtable() no longer drops the filedesc lock so it is enough to
retry finding free file descriptor only once after fdgrowt
and the only caller always passes 0 as low. Given
that, how about the following:
http://people.freebsd.org/~mjg/patches/fd_last_used-cleanup.patch
Looks good too.
Updated in similar manner:
http://people.freebsd.org/~mjg/patches/fd_last_used-cleanup2.patch
I ha
Author: davidxu
Date: Thu Mar 29 02:46:43 2012
New Revision: 233642
URL: http://svn.freebsd.org/changeset/base/233642
Log:
Reduce code size by creating common timed sleeping function.
Modified:
head/sys/kern/kern_umtx.c
Modified: head/sys/kern/kern_umtx.c
Author: davidxu
Date: Fri Mar 30 05:40:26 2012
New Revision: 233690
URL: http://svn.freebsd.org/changeset/base/233690
Log:
Merge umtxq_sleep and umtxq_nanosleep into a single function by using
an abs_timeout structure which describes timeout info.
Modified:
head/sys/kern/kern_umtx.c
Modifi
Author: davidxu
Date: Fri Mar 30 05:49:32 2012
New Revision: 233691
URL: http://svn.freebsd.org/changeset/base/233691
Log:
Remove trailing space.
Modified:
head/sys/kern/kern_umtx.c
Modified: head/sys/kern/kern_umtx.c
==
Author: davidxu
Date: Fri Mar 30 09:03:53 2012
New Revision: 233693
URL: http://svn.freebsd.org/changeset/base/233693
Log:
Fix COMPAT_FREEBSD32 build.
Submitted by: Andreas Tobler < andreast at fgznet dot ch >
Modified:
head/sys/kern/kern_umtx.c
Modified: head/sys/kern/kern_umtx.c
=
Author: davidxu
Date: Fri Mar 30 12:57:14 2012
New Revision: 233700
URL: http://svn.freebsd.org/changeset/base/233700
Log:
Remove trailing semicolon, it is a typo.
Modified:
head/sys/kern/kern_umtx.c
Modified: head/sys/kern/kern_umtx.c
Author: davidxu
Date: Sat Mar 31 06:48:41 2012
New Revision: 233729
URL: http://svn.freebsd.org/changeset/base/233729
Log:
Remove stale comments.
Modified:
head/sys/kern/kern_umtx.c
Modified: head/sys/kern/kern_umtx.c
==
Author: davidxu
Date: Thu Apr 5 02:24:08 2012
New Revision: 233912
URL: http://svn.freebsd.org/changeset/base/233912
Log:
umtx operation UMTX_OP_MUTEX_WAKE has a side-effect that it accesses
a mutex after a thread has unlocked it, it event writes data to the mutex
memory to clear contention
Author: davidxu
Date: Thu Apr 5 03:05:02 2012
New Revision: 233913
URL: http://svn.freebsd.org/changeset/base/233913
Log:
In sem_post, the field _has_waiters is no longer used, because some
application destroys semaphore after sem_wait returns. Just enter
kernel to wake up sleeping threads,
Author: davidxu
Date: Tue Apr 17 09:02:55 2012
New Revision: 234371
URL: http://svn.freebsd.org/changeset/base/234371
Log:
MFC r233912:
mtx operation UMTX_OP_MUTEX_WAKE has a side-effect that it accesses
a mutex after a thread has unlocked it, it event writes data to the mutex
memory to
Author: davidxu
Date: Tue Apr 17 09:09:14 2012
New Revision: 234372
URL: http://svn.freebsd.org/changeset/base/234372
Log:
Merge 233103, 233912 from head:
233103:
Some software think a mutex can be destroyed after it owned it, for
example, it uses a serialization point like following:
Author: davidxu
Date: Tue Apr 17 09:18:06 2012
New Revision: 234373
URL: http://svn.freebsd.org/changeset/base/234373
Log:
Merge 233103 from head:
Some software think a mutex can be destroyed after it owned it, for
example, it uses a serialization point like following:
pthread_mut
Author: davidxu
Date: Thu May 3 03:05:18 2012
New Revision: 234937
URL: http://svn.freebsd.org/changeset/base/234937
Log:
Merge 233103, 233912 from head:
233103:
Some software think a mutex can be destroyed after it owned it, for
example, it uses a serialization point like following:
Author: davidxu
Date: Thu May 3 09:17:31 2012
New Revision: 234947
URL: http://svn.freebsd.org/changeset/base/234947
Log:
MFp4:
Enqueue thread in LIFO, this can cause starvation, but it gives better
performance. Use _thr_queuefifo to control the frequency of FIFO vs LIFO,
you can use envi
Author: davidxu
Date: Sat May 5 23:51:24 2012
New Revision: 235068
URL: http://svn.freebsd.org/changeset/base/235068
Log:
Fix mis-merged line, move SC_LOOKUP() call to
upper level.
Modified:
head/lib/libthr/thread/thr_sleepq.c
Modified: head/lib/libthr/thread/thr_sleepq.c
Author: davidxu
Date: Thu May 10 09:30:37 2012
New Revision: 235218
URL: http://svn.freebsd.org/changeset/base/235218
Log:
Create a common function lookup() to search a chan, this eliminates
redundant SC_LOOKUP() calling.
Modified:
head/lib/libthr/thread/thr_sleepq.c
Modified: head/lib/lib
Author: davidxu
Date: Tue Jan 22 03:23:14 2013
New Revision: 245761
URL: http://svnweb.freebsd.org/changeset/base/245761
Log:
Make "-D skip" option work with FIFO by opening file in non-blocking mode.
Reviewed by: jhb
Tested by:delphij
Modified:
head/gnu/usr.bin/grep/grep.c
Modif
Author: davidxu
Date: Sat Feb 16 06:07:07 2013
New Revision: 246872
URL: http://svnweb.freebsd.org/changeset/base/246872
Log:
Simplify code by using flag O_EXLOCK.
PR: kern/175674
Modified:
head/lib/libc/gen/sem_new.c
Modified: head/lib/libc/gen/sem_new.c
===
Author: davidxu
Date: Sun Feb 17 02:52:42 2013
New Revision: 246894
URL: http://svnweb.freebsd.org/changeset/base/246894
Log:
Make more code be protected by internal mutex, and now it is fork-safe, in
error case, the file exclusive lock is now released as soon as possible,
in previous code,
Author: davidxu
Date: Sat Aug 11 00:06:56 2012
New Revision: 239187
URL: http://svn.freebsd.org/changeset/base/239187
Log:
tvtohz will print out an error message if a negative value is given
to it, avoid this problem by detecting timeout earlier.
Reported by: pho
Modified:
head/sys/ker
Author: davidxu
Date: Sat Aug 11 23:17:02 2012
New Revision: 239200
URL: http://svn.freebsd.org/changeset/base/239200
Log:
MFp4:
Further decreases unexpected context switches by defering mutex wakeup
until internal sleep queue lock is released.
Modified:
head/lib/libthr/thread/thr_cond.c
Author: davidxu
Date: Sat Aug 11 23:48:39 2012
New Revision: 239202
URL: http://svn.freebsd.org/changeset/base/239202
Log:
Some style fixes inspired by @bde.
Modified:
head/sys/kern/kern_umtx.c
Modified: head/sys/kern/kern_umtx.c
==
Author: davidxu
Date: Sun Aug 12 00:56:56 2012
New Revision: 239206
URL: http://svn.freebsd.org/changeset/base/239206
Log:
Do defered mutex wakeup once.
Modified:
head/lib/libthr/thread/thr_cond.c
Modified: head/lib/libthr/thread/thr_cond.c
===
+1,39 @@
+/*
+ * Copyright (c) 2012 David Xu .
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
Author: davidxu
Date: Fri Aug 17 02:47:16 2012
New Revision: 239349
URL: http://svn.freebsd.org/changeset/base/239349
Log:
regen.
Modified:
head/sys/compat/freebsd32/freebsd32_proto.h
head/sys/compat/freebsd32/freebsd32_syscall.h
head/sys/compat/freebsd32/freebsd32_syscalls.c
head/sys/c
Author: davidxu
Date: Tue Aug 21 09:17:13 2012
New Revision: 239485
URL: http://svn.freebsd.org/changeset/base/239485
Log:
Fix prototype. Also the function should return error code instead of
-1 on error.
Modified:
head/lib/libc/gen/clock_getcpuclockid.c
Modified: head/lib/libc/gen/clock_g
uclockid.3
==
--- /dev/null 00:00:00 1970 (empty, because file is newly added)
+++ head/lib/libc/gen/clock_getcpuclockid.3 Tue Aug 21 09:18:28 2012
(r239486)
@@ -0,0 +1,95 @@
+.\" Copyright (c) 2012 David Xu
+.\"
Author: davidxu
Date: Thu Aug 23 05:15:15 2012
New Revision: 239609
URL: http://svn.freebsd.org/changeset/base/239609
Log:
Eliminate redundant code, _thr_spinlock_init() has already been called
in init_private(), don't call it again in fork() wrapper.
Modified:
head/lib/libthr/thread/thr_fo
Author: davidxu
Date: Mon Aug 27 03:09:39 2012
New Revision: 239718
URL: http://svn.freebsd.org/changeset/base/239718
Log:
In suspend_common(), don't wait for a thread which is in creation, because
pthread_suspend_all_np() may have already suspended its parent thread.
Add locking code in pth
Author: davidxu
Date: Mon Sep 10 05:00:29 2012
New Revision: 240295
URL: http://svn.freebsd.org/changeset/base/240295
Log:
Add missing prototype for clock_getcpuclockid.
Modified:
head/include/time.h
Modified: head/include/time.h
==
Author: davidxu
Date: Mon Sep 10 05:09:39 2012
New Revision: 240296
URL: http://svn.freebsd.org/changeset/base/240296
Log:
Process CPU-Time Clocks option is supported, define _POSIX_CPUTIME.
Modified:
head/include/unistd.h
Modified: head/include/unistd.h
=
Author: davidxu
Date: Mon Sep 10 05:12:45 2012
New Revision: 240297
URL: http://svn.freebsd.org/changeset/base/240297
Log:
POSIX requires sigevent to be visible after mqueue.h is included.
Modified:
head/include/mqueue.h
Modified: head/include/mqueue.h
===
Author: davidxu
Date: Mon Mar 9 02:34:02 2009
New Revision: 189549
URL: http://svn.freebsd.org/changeset/base/189549
Log:
Don't reference non-existent __fcntl_compat if WITHOUT_SYSCALL_COMPAT is
defined.
Submitted by: Pawel Worach "pawel dot worach at gmail dot com"
Modified:
head/lib/
Author: davidxu
Date: Mon Mar 9 05:54:43 2009
New Revision: 189553
URL: http://svn.freebsd.org/changeset/base/189553
Log:
Don't ignore other fcntl functions, directly call __sys_fcntl if
WITHOUT_SYSCALL_COMPAT is not defined.
Reviewed by: deischen
Modified:
head/lib/libthr/thread/thr
Author: davidxu
Date: Fri Mar 13 06:06:20 2009
New Revision: 189756
URL: http://svn.freebsd.org/changeset/base/189756
Log:
1) Check NULL pointer before calling umtx_pi_adjust_locked(), this avoids
a PANIC.
2) Rework locking for POSIX priority-mutex, this fixes a
race where a thread m
Author: davidxu
Date: Wed Nov 18 01:35:36 2009
New Revision: 199465
URL: http://svn.freebsd.org/changeset/base/199465
Log:
Fix compiler warnings.
Modified:
head/lib/librt/sigev_thread.c
Modified: head/lib/librt/sigev_thread.c
==
M. Warner Losh wrote:
In message: <200911180135.nai1zaie051...@svn.freebsd.org>
David Xu writes:
: Author: davidxu
: Date: Wed Nov 18 01:35:36 2009
: New Revision: 199465
: URL: http://svn.freebsd.org/changeset/base/199465
:
: Log:
: Fix compiler warnings.
:
: Mo
M. Warner Losh wrote:
In message: <200911180135.nai1zaie051...@svn.freebsd.org>
David Xu writes:
: Author: davidxu
: Date: Wed Nov 18 01:35:36 2009
: New Revision: 199465
: URL: http://svn.freebsd.org/changeset/base/199465
:
: Log:
: Fix compiler warnings.
:
: Mo
Author: davidxu
Date: Wed Nov 18 05:09:03 2009
New Revision: 199469
URL: http://svn.freebsd.org/changeset/base/199469
Log:
link libpthread because the librt really needs it to fully function.
Modified:
head/lib/librt/Makefile
Modified: head/lib/librt/Makefile
Author: davidxu
Date: Wed Nov 18 06:10:50 2009
New Revision: 199475
URL: http://svn.freebsd.org/changeset/base/199475
Log:
Make following functions be cancellation points:
mq_receive
mq_send
mq_timereceive
mq_timedsend
Modified:
head/lib/librt/mq.c
Modified: head/lib/librt/mq.c
=
M. Warner Losh wrote:
In message: <200911180509.nai593kq058...@svn.freebsd.org>
David Xu writes:
: Author: davidxu
: Date: Wed Nov 18 05:09:03 2009
: New Revision: 199469
: URL: http://svn.freebsd.org/changeset/base/199469
:
: Log:
: link libpthread because the librt
Author: davidxu
Date: Tue Jul 20 02:23:12 2010
New Revision: 210274
URL: http://svn.freebsd.org/changeset/base/210274
Log:
Fix function name in error messages.
Modified:
head/sys/kern/kern_sig.c
Modified: head/sys/kern/kern_sig.c
==
Author: davidxu
Date: Tue Aug 17 02:50:12 2010
New Revision: 211409
URL: http://svn.freebsd.org/changeset/base/211409
Log:
Tweak code a bit to be POSIX compatible, when a cancellation request
is acted upon, or when a thread calls pthread_exit(), the thread first
disables cancellation by sett
Author: davidxu
Date: Fri Aug 20 04:15:05 2010
New Revision: 211522
URL: http://svn.freebsd.org/changeset/base/211522
Log:
According to specification, function fcntl() is a cancellation point only
when cmd argument is F_SETLKW.
Modified:
head/lib/libthr/thread/thr_syscalls.c
Modified: head
Author: davidxu
Date: Fri Aug 20 04:28:30 2010
New Revision: 211523
URL: http://svn.freebsd.org/changeset/base/211523
Log:
If thread set a TDP_WAKEUP for itself, clears the flag and returns EINTR
immediately, this is used for implementing reliable pthread cancellation.
Modified:
head/sys/ke
Author: davidxu
Date: Fri Aug 20 05:15:39 2010
New Revision: 211524
URL: http://svn.freebsd.org/changeset/base/211524
Log:
In current implementation, thread cancellation is done in signal handler,
which does not know what is the state of interrupted system call, for
example, open() system
Author: davidxu
Date: Fri Aug 20 13:42:48 2010
New Revision: 211526
URL: http://svn.freebsd.org/changeset/base/211526
Log:
Reduce redundant code.
Submitted by: kib
Modified:
head/lib/libthr/thread/thr_sig.c
Modified: head/lib/libthr/thread/thr_sig.c
=
Author: davidxu
Date: Fri Aug 20 23:51:34 2010
New Revision: 211534
URL: http://svn.freebsd.org/changeset/base/211534
Log:
make sure thread lock is locked.
Modified:
head/sys/kern/subr_sleepqueue.c
Modified: head/sys/kern/subr_sleepqueue.c
Author: davidxu
Date: Tue Aug 24 07:22:24 2010
New Revision: 211732
URL: http://svn.freebsd.org/changeset/base/211732
Log:
- According to specification, SI_USER code should only be generated by
standard kill(). On other systems, SI_LWP is generated by lwp_kill().
This will allow conformi
Author: davidxu
Date: Tue Aug 24 07:29:55 2010
New Revision: 211733
URL: http://svn.freebsd.org/changeset/base/211733
Log:
Optimize thr_suspend, if timeout is zero, don't call msleep, just
return immediately.
Modified:
head/sys/kern/kern_thr.c
Modified: head/sys/kern/kern_thr.c
===
Author: davidxu
Date: Tue Aug 24 08:09:30 2010
New Revision: 211735
URL: http://svn.freebsd.org/changeset/base/211735
Log:
Bump __FreeBSD_version for revision 211732.
Noticed by: thompa
Modified:
head/sys/sys/param.h
Modified: head/sys/sys/param.h
===
Author: davidxu
Date: Tue Aug 24 09:57:06 2010
New Revision: 211737
URL: http://svn.freebsd.org/changeset/base/211737
Log:
Add wrapper for setcontext() and swapcontext(), the wrappers
unblock SIGCANCEL which is needed by thread cancellation.
Modified:
head/lib/libthr/pthread.map
head/lib/
Author: davidxu
Date: Wed Aug 25 03:14:32 2010
New Revision: 211794
URL: http://svn.freebsd.org/changeset/base/211794
Log:
If a thread is removed from umtxq while sleeping, reset error code
to zero, this gives userland a better indication that a thread needn't
to be cancelled.
Modified:
h
Author: davidxu
Date: Thu Aug 26 07:09:48 2010
New Revision: 211833
URL: http://svn.freebsd.org/changeset/base/211833
Log:
Decrease rdlock count only when thread unlocked a reader lock.
MFC after:3 days
Modified:
head/lib/libthr/thread/thr_rtld.c
Modified: head/lib/libthr/thread/thr
Author: davidxu
Date: Thu Aug 26 09:04:27 2010
New Revision: 211836
URL: http://svn.freebsd.org/changeset/base/211836
Log:
eliminate unused code.
Modified:
head/lib/libthr/thread/thr_exit.c
head/lib/libthr/thread/thr_private.h
Modified: head/lib/libthr/thread/thr_exit.c
===
Author: davidxu
Date: Fri Aug 27 03:23:07 2010
New Revision: 211859
URL: http://svn.freebsd.org/changeset/base/211859
Log:
clear lock to zero state if it is destroyed.
Modified:
head/lib/libthr/thread/thr_rtld.c
Modified: head/lib/libthr/thread/thr_rtld.c
Author: davidxu
Date: Fri Aug 27 05:20:22 2010
New Revision: 211860
URL: http://svn.freebsd.org/changeset/base/211860
Log:
Unregister thread specific data destructor when a corresponding dso
is unloaded.
Modified:
head/lib/libthr/thread/thr_fork.c
head/lib/libthr/thread/thr_private.h
he
Author: davidxu
Date: Sun Aug 29 08:15:57 2010
New Revision: 211960
URL: http://svn.freebsd.org/changeset/base/211960
Log:
MFC r211833: Decrease rdlock count only when thread unlocked
a reader lock.
Modified:
stable/8/lib/libthr/thread/thr_rtld.c
Modified: stable/8/lib/libthr/thread/thr_rt
Author: davidxu
Date: Sun Aug 29 08:19:07 2010
New Revision: 211961
URL: http://svn.freebsd.org/changeset/base/211961
Log:
Record merge info for r211960.
-This line, and those below, will be ignored--
> Description of fields to fill in above: 76 columns --|
> PR:
condition. I have not read the code, maybe
this is not a problem.
Regards,
David Xu
___
svn-src-all@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"
Author: davidxu
Date: Tue Aug 31 07:15:50 2010
New Revision: 212047
URL: http://svn.freebsd.org/changeset/base/212047
Log:
If a process is being debugged, skips job control caused by SIGSTOP/SIGCONT
signals, because it is managed by debugger, however a normal signal sent to
a interruptibly s
Jilles Tjoelker wrote:
On Tue, Aug 31, 2010 at 07:15:50AM +, David Xu wrote:
Author: davidxu
Date: Tue Aug 31 07:15:50 2010
New Revision: 212047
URL: http://svn.freebsd.org/changeset/base/212047
Log:
If a process is being debugged, skips job control caused by SIGSTOP/SIGCONT
signals
Author: davidxu
Date: Wed Sep 1 01:26:07 2010
New Revision: 212075
URL: http://svn.freebsd.org/changeset/base/212075
Log:
rescure comments from RELENG_4.
Modified:
head/sys/kern/kern_sig.c
Modified: head/sys/kern/kern_sig.c
===
Author: davidxu
Date: Wed Sep 1 02:18:33 2010
New Revision: 212076
URL: http://svn.freebsd.org/changeset/base/212076
Log:
Add signal handler wrapper, the reason to add it becauses there are
some cases we want to improve:
1) if a thread signal got a signal while in cancellation point,
Author: davidxu
Date: Wed Sep 1 03:11:21 2010
New Revision: 212077
URL: http://svn.freebsd.org/changeset/base/212077
Log:
Change atfork lock from mutex to rwlock, also make mutexes used by malloc()
module private type, when private type mutex is locked/unlocked, thread
critical region is en
Author: davidxu
Date: Wed Sep 1 03:55:10 2010
New Revision: 212078
URL: http://svn.freebsd.org/changeset/base/212078
Log:
pthread_atfork should acquire writer lock and protect the code
with critical region.
Modified:
head/lib/libthr/thread/thr_fork.c
Modified: head/lib/libthr/thread/thr_f
Author: davidxu
Date: Wed Sep 1 07:09:46 2010
New Revision: 212083
URL: http://svn.freebsd.org/changeset/base/212083
Log:
In function __pthread_cxa_finalize(), also make code for removing
atfork handler be async-signal safe.
Modified:
head/lib/libthr/thread/thr_fork.c
Modified: head/lib/l
Author: davidxu
Date: Wed Sep 1 13:22:55 2010
New Revision: 212095
URL: http://svn.freebsd.org/changeset/base/212095
Log:
Remove incorrect comments, also make sure signal is
disabled when unregistering sigaction.
Modified:
head/lib/libthr/thread/thr_sig.c
Modified: head/lib/libthr/thread/
Kostik Belousov wrote:
On Wed, Sep 01, 2010 at 02:18:33AM +, David Xu wrote:
Author: davidxu
Date: Wed Sep 1 02:18:33 2010
New Revision: 212076
URL: http://svn.freebsd.org/changeset/base/212076
Log:
Add signal handler wrapper, the reason to add it becauses there are
some cases we want
Kostik Belousov wrote:
the tf_err may not be equal to ksi_addr! This may need to be fixed.
The change was introduced by
r151316 | davidxu | 2005-10-14
/* Old FreeBSD-style arguments. */
- sf.sf_siginfo = code;
- sf.sf_addr = regs->tf_err;
+ sf.sf_siginfo = ksi->ksi_code;
+
Kostik Belousov wrote:
On Thu, Sep 02, 2010 at 04:34:58PM +, David Xu wrote:
Kostik Belousov wrote:
On Wed, Sep 01, 2010 at 02:18:33AM +, David Xu wrote:
Author: davidxu
Date: Wed Sep 1 02:18:33 2010
New Revision: 212076
URL: http://svn.freebsd.org/changeset/base/212076
Log:
Add
John Baldwin wrote:
On Thursday, September 02, 2010 6:22:18 am Kostik Belousov wrote:
On Thu, Sep 02, 2010 at 05:35:39PM +, David Xu wrote:
Kostik Belousov wrote:
the tf_err may not be equal to ksi_addr! This may need to be fixed.
The change was introduced by
John Baldwin wrote:
On Thursday, September 02, 2010 7:36:08 am David Xu wrote:
John Baldwin wrote:
On Thursday, September 02, 2010 6:22:18 am Kostik Belousov wrote:
On Thu, Sep 02, 2010 at 05:35:39PM +, David Xu wrote:
Kostik Belousov wrote
Author: davidxu
Date: Mon Sep 6 03:00:54 2010
New Revision: 212245
URL: http://svn.freebsd.org/changeset/base/212245
Log:
Fix off-by-one error in function _thr_sigact_unload, also disable the
function, it seems some gnome application tends to crash if we
unregister sigaction automatically.
Author: davidxu
Date: Wed Sep 8 01:55:03 2010
New Revision: 212311
URL: http://svn.freebsd.org/changeset/base/212311
Log:
MFC: r212047
If a process is being debugged, skips job control caused by SIGSTOP/SIGCONT
signals, because it is managed by debugger, however a normal signal sent to
Author: davidxu
Date: Wed Sep 8 02:18:20 2010
New Revision: 212312
URL: http://svn.freebsd.org/changeset/base/212312
Log:
To avoid possible race condition, SIGCANCEL is always sent except the
thread is dead.
Modified:
head/lib/libthr/thread/thr_cancel.c
Modified: head/lib/libthr/thread/th
Author: davidxu
Date: Fri Sep 10 01:47:37 2010
New Revision: 212405
URL: http://svn.freebsd.org/changeset/base/212405
Log:
Because POSIX does not allow EINTR to be returned from sigwait(),
add a wrapper for it in libc and rework the code in libthr, the
system call still can return EINTR, we
Author: davidxu
Date: Sat Sep 11 08:15:18 2010
New Revision: 212458
URL: http://svn.freebsd.org/changeset/base/212458
Log:
kill() does queue signal now.
Modified:
head/share/man/man3/siginfo.3
Modified: head/share/man/man3/siginfo.3
===
Author: davidxu
Date: Sat Sep 11 09:26:36 2010
New Revision: 212459
URL: http://svn.freebsd.org/changeset/base/212459
Log:
Make a link for pthread_timedjoin_np.
Modified:
head/share/man/man3/Makefile
Modified: head/share/man/man3/Makefile
=
Author: davidxu
Date: Mon Sep 13 07:03:01 2010
New Revision: 212536
URL: http://svn.freebsd.org/changeset/base/212536
Log:
Convert thread list lock from mutex to rwlock.
Modified:
head/lib/libthr/thread/thr_affinity.c
head/lib/libthr/thread/thr_attr.c
head/lib/libthr/thread/thr_cancel.c
m_wake /* UMTX_OP_SEM_WAKE */
};
int
Added: head/sys/sys/_umtx.h
==
--- /dev/null 00:00:00 1970 (empty, because file is newly added)
+++ head/sys/sys/_umtx.hMon Jan 4 05:27:49 2010(r201472)
@@ -0,0 +1,66 @@
(r201546)
@@ -0,0 +1,64 @@
+/*
+ * Copyright (c) 2010 David Xu
+ *
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code
Author: davidxu
Date: Tue Jan 5 03:39:31 2010
New Revision: 201547
URL: http://svn.freebsd.org/changeset/base/201547
Log:
Don't check has_waiters twice, inline some small functions.
performance result on my machine:
mutex Elapsed: 902115 us; per iteration: 90 ns.
semaphore
Alexander Kabaev wrote:
Hi,
some quick questions:
FB10_COMPAT_PRIVATE is not used anywhere, why did it get committed?
That aside, use of SYM_DEFAULT is pretty much always wrong.
Yes, this needs to be cleaned.
Also, does libthr really need to export sem_ symbols in 1.2 namespace?
I do not th
Author: davidxu
Date: Tue Jan 5 05:44:52 2010
New Revision: 201553
URL: http://svn.freebsd.org/changeset/base/201553
Log:
Remove unused macros.
Modified:
head/lib/libc/gen/sem.c
Modified: head/lib/libc/gen/sem.c
==
Author: davidxu
Date: Tue Jan 5 05:47:18 2010
New Revision: 201555
URL: http://svn.freebsd.org/changeset/base/201555
Log:
forgot to remove SYM_FBP10. ;-)
Modified:
head/lib/libc/gen/sem.c
Modified: head/lib/libc/gen/sem.c
Alexander Kabaev wrote:
On Tue, 05 Jan 2010 13:39:06 +0800
David Xu wrote:
Alexander Kabaev wrote:
Hi,
some quick questions:
FB10_COMPAT_PRIVATE is not used anywhere, why did it get committed?
That aside, use of SYM_DEFAULT is pretty much always wrong.
Yes, this needs to be cleaned
Author: davidxu
Date: Tue Jan 5 06:21:29 2010
New Revision: 201559
URL: http://svn.freebsd.org/changeset/base/201559
Log:
Remove extra new semaphore stubs, because libc already has them, and
ld can find the newest version which is default.
Poked by: kan@
Deleted:
head/lib/libthr/threa
201 - 300 of 335 matches
Mail list logo
