Commit Graph

122922 Commits

Author SHA1 Message Date
Hartmut Brandt
0108135cd6 Subtracting two pointers produces a ptrdiff_t not a size_t so use
the %td to print this instead of %zu or %d.
2006-03-21 15:00:10 +00:00
Hartmut Brandt
326570abe3 This driver has been MPSAFE from the beginning so declare the interrupt
as such.

Reminded by:	rwatson@
2006-03-21 14:56:42 +00:00
Hartmut Brandt
5b4d2450cc This driver has been MPSAFE from the beginning, so declare the interrupt
as such.

Reminded by:	rwatson@
2006-03-21 14:54:42 +00:00
Gleb Smirnoff
93a69f5703 No direct call to carp_ifdetach() anymore. It is called by
event handler.

PR:		kern/82908
Submitted by:	Dan Lukes <dan obluda.cz>
2006-03-21 14:31:18 +00:00
Gleb Smirnoff
0fa0801895 o Introduce carp_multicast_cleanup(), which removes and frees
multicast addresses from carp interface. [1]
o Rewrite carpdetach(), so that it does the following things: [1]
  - Stops callouts.
  - Decrements carp_suppress_preempt, if needed.
  - Downs interface and sets CARP state to INIT.
  - Calls carp_multicast_cleanup().
  - Detaches softc from carp_if and if we are the last frees
    the carp_if.
o Use new carpdetach() in carp_clone_destroy().
o In carp_ifdetach() acquire the carp_if lock and cleanup all
  interfaces hanging on carp_if. [1]
o Make carp_ifdetach() static and use EVENT(9) to call it
  from if_detach(). [2]
o In carp_setrun() exit if the softc doesn't have a valid pointer
  to parent. [1]

Obtained from:	OpenBSD [1]
Submitted by:	Dan Lukes <dan obluda.cz> [2]
PR:		kern/82908 [2]
2006-03-21 14:29:48 +00:00
Xin LI
f1fa5b4f51 When we are doing initialization against q, use its own size, not
the size of q2.  This should be a no-op because q and q2 are of
the same type.

Submitted by:	Alexey Dobriyan <adobriyan gmail com>
2006-03-21 12:39:29 +00:00
Pyun YongHyeon
5dbde3ff57 Backout rev. 1.46. It caused Rx checksum offload breakage on little
endian systems.

Reported by:	joerg
Tested by:	joerg
2006-03-21 12:21:51 +00:00
Ruslan Ermilov
a50767eb92 Unbreak WITHOUT_LIBPHREAD/WITHOUT_LIBC_R option support, depending
on platform.
2006-03-21 11:00:54 +00:00
Ruslan Ermilov
c60fab8c03 Document new world older for world/kernel build options. 2006-03-21 10:10:05 +00:00
David Xu
bf1a322061 Rethink it a bit, if there is a STOP flag, don't bother to resume other
threads.
2006-03-21 10:05:15 +00:00
Ruslan Ermilov
862aa05f64 Remove mentions of old compat options from here; equivalents are
now provided in src.conf(5).
2006-03-21 09:49:05 +00:00
Ruslan Ermilov
fa07cb9cd3 Install the autogenerated src.conf(5) manpage. 2006-03-21 09:10:35 +00:00
Ruslan Ermilov
41c8ddc37f Fix generation of '$FreeBSD$'. 2006-03-21 09:08:45 +00:00
Ruslan Ermilov
8644ea9e0a Slightly better description. 2006-03-21 09:06:24 +00:00
Ruslan Ermilov
c5922fbf80 A tool to generate the src.conf(5) manpage. 2006-03-21 09:05:42 +00:00
Ruslan Ermilov
b81594171c Folded WITH_HESIOD_SUPPORT into WITH_HESIOD. 2006-03-21 09:00:51 +00:00
Ruslan Ermilov
d55557650a Desupport the *_COLLECT2 no-op. 2006-03-21 08:59:36 +00:00
David Xu
568b4ebbcc Because JOB control has higher priority than single threading in
thread_suspend_check(), call thread_stopped() to report SIGCHLD
if there is JOB control in progress.
2006-03-21 08:41:15 +00:00
Dag-Erling Smørgrav
e2ed8189e9 Open the lock file write-only. 2006-03-21 08:08:58 +00:00
Ruslan Ermilov
5990647d42 Prepare to autogenerate the src.conf(5) manpage. 2006-03-21 07:50:50 +00:00
Ruslan Ermilov
6f1a7c180c Remove an outdated README. 2006-03-21 07:26:51 +00:00
David Xu
39d3e6198d Remove stale KSE code.
Reviewed by: alc
2006-03-21 06:46:27 +00:00
Ariff Abdullah
3fdb3676ba Apply more thorough fixes while dealing with device opening and closing:
- Determine open direction using 'flags', not 'mode'. This bug exist since
  past 4 years.
- Don't allow opening the same device twice, be it in a same or different
  direction.
- O_RDWR is allowed, provided that it is done by a single open (for example
  by mixer(8)) and the underlying hardware support true full-duplex operation.
- Do various paranoid checking in case other process/thread trying to hijack
  the same device twice (or more).

MFC after:	5 days
2006-03-21 06:35:48 +00:00
Ariff Abdullah
02dbda9d17 Recover (?) support for AD1815 based ISA soundcards.
PR:		kern/94388
Submitted by:	Krzysztof Kotlenga <piernik at gmail dot com>
MFC after:	3 days
2006-03-21 03:47:25 +00:00
Tor Egge
41d7199b9d Remove unused leaked debug function prototype. 2006-03-21 01:04:24 +00:00
Giorgos Keramidas
f92e18f4fc Add descriptions for the sysctls:
net.inet.icmp.drop_redirect
    net.inet.icmp.log_redirect
    net.inet.icmp.icmplim
    net.inet.icmp.icmplim_output

Approved & text by:	andre
2006-03-20 21:44:12 +00:00
Warner Losh
33924ad466 Minor style(9) and KNF elimination as I prepare to fix a bug. 2006-03-20 21:28:46 +00:00
Alexander Leidinger
f50143e803 regen 2006-03-20 19:48:02 +00:00
Alexander Leidinger
61da9d97fb Fix tinderbox on alpha.
Tested by:	cross-compile
2006-03-20 19:46:56 +00:00
John Baldwin
aef8cd01ed Drop some unneeded casts since we program the kernel in C rather than C++. 2006-03-20 19:39:08 +00:00
Alexander Leidinger
79d8404261 regen: fix of linuxolator with testing in a cross-build 2006-03-20 18:54:29 +00:00
Alexander Leidinger
3a192a2050 Fix the linuxolator on amd64 (cross-build). 2006-03-20 18:53:26 +00:00
Florent Thoumie
4442f896de - Replace awk parts with shell.
Approved by:	brooks
MFC after:	2 weeks
2006-03-20 18:00:14 +00:00
Wojciech A. Koszek
bc02a9764c Keep references sorted alphabetically.
Spotted by:	brueffer
Approved by:	cognet (mentor)
2006-03-20 17:10:09 +00:00
Wojciech A. Koszek
2ecd560bcc There is no need to pass NULL to the pam_error() as the last argument.
Remove it.

Reviewed by:	des
Approved by:	cognet (mentor)
2006-03-20 16:56:08 +00:00
Wojciech A. Koszek
84044ca360 Add reference to conscontrol(8).
Reviewed by:	ru
Approved by:	cognet (mentor)
2006-03-20 16:53:00 +00:00
Joseph Koshy
d8acba2a3e Update the pmc(3) manual page's date string and freshen the year in the (c) line for
pmc.3 and libpmc.c.
2006-03-20 15:28:59 +00:00
Ruslan Ermilov
cea557add0 Extend coverage of the MK_IPX build option to the following:
- <netipx> headers [1]
- IPX library (libipx)
- IPX support in ifconfig(8)
- IPXrouted(8)
- new MK_NCP option

New MK_NCP build option controls:

- <netncp> and <fs/nwfs> headers
- NCP library (libncp)
- ncplist(1) and ncplogin(1)
- mount_nwfs(8)
- ncp and nwfs kernel modules

User knobs: WITHOUT_IPX, WITHOUT_IPX_SUPPORT, WITHOUT_NCP.

[1] <netsmb/netbios.h> unconditionally uses <netipx> headers
so they are still installed.  This needs to be dealt with.
2006-03-20 14:24:58 +00:00
Ruslan Ermilov
a3aed80d1e Document the requirement to run rpc.lockd(8) on both client
and server when installing onto an NFS file system.
2006-03-20 12:27:08 +00:00
Jason Evans
c07ee180bc Separate completely full runs from runs that are merely almost full, so
that no linear searching is necessary if we resort to allocating from a
run that is known to be mostly full.  There are pathological edge cases
that could have caused severely degraded performance, and this change
fixes that.
2006-03-20 04:05:05 +00:00
David Xu
6f79c82641 Set default contention scope to system. 2006-03-20 03:14:14 +00:00
Christian S.J. Peron
2ed4894a26 Restore fd optimization with a few minor tweaks, to quote tegge:
"fdinit() fails to initialize newfdp->fd_fd.fd_lastfile to -1.  This breaks
fdcopy() which will incorrectly set newfdp->fd_freefile to 1 if no files are
open and the last file descriptor marked as unused for fdp was 0.  This later
causes descriptor 0 to be unavailable in newfdp when the optimization is
enabled.

When the last file descriptor previously marked as used is nonzero and marked
as unused, fdunused() incorrectly sets fdp->fd_lastfile to fd - 1 due to
fd_last_used() returning (size - 1).  This hides the problem that breaks the
optimization."

This allows us to keep the optimization, while un-breaking it.

This is a RELENG_6 candidate.

PR:		kern/87208
MFC after:	1 week
Submitted by:	tegge
2006-03-20 00:13:47 +00:00
Tor Egge
700118c72f Allow compilation when not using softupdates. 2006-03-19 22:16:44 +00:00
Tor Egge
7de3839d0d Let snapshots make a copy of old contents for all buffers taking part in a
cluster instead of just the first buffer.

Delay buf_start() calls until snapshots have a copy of old content.

PR:		kern/93942
2006-03-19 21:43:36 +00:00
Tor Egge
30b3a49fab Add kludge to avoid deadlock when unlinking snapshot. 2006-03-19 21:29:20 +00:00
Tor Egge
95e7a3c3ac Reduce probability of unmount failing after having unmounted snapshots. 2006-03-19 21:09:19 +00:00
Tor Egge
8c86028f11 Ensure that vnode for directory isn't reclaimed before ffs_snapshot() has
completed expunging unlinked files.  It could come back at another memory
location causing a lock order reversal.
2006-03-19 21:05:10 +00:00
Tor Egge
87f0769a57 Call vn_start_write() before locking vnode. 2006-03-19 20:45:06 +00:00
Tor Egge
d50ef66d03 Don't call vn_finished_write() if vn_start_write() failed. 2006-03-19 20:43:07 +00:00
Jeff Roberson
e44270a781 - Correct an assert in vop_rename_pre. fdvp may be locked if it is either
the target directory or file.  This case should fail in the filesystem
   anyway and perhaps kern_rename() should catch it.

Sponsored by:	Isilon Systems, Inc.
2006-03-19 20:14:46 +00:00