Commit Graph

125983 Commits

Author SHA1 Message Date
Pawel Jakub Dawidek
de6f1c7c6c Not only a request from us can be passed to g_{mirror,raid3}_worker()
function, but also a request to us, in which case checking bio_cflags
is wrong, because the class above us is controling it, not we.

MFC after:	1 week
2006-08-09 09:41:53 +00:00
Alexander Leidinger
89cf3df166 Get rid of the rtermcap stuff (embedding termcaps into the program). This
is needed in sysinstall on boot, but not in sade on a fully working system.

After this change I noticed no change in behavior on a vty when TERM is not
set.
2006-08-09 09:12:01 +00:00
Xin LI
47dbd966d9 Bump WARNS level to 3.
Tested with:	make universe
2006-08-09 08:25:21 +00:00
Xin LI
88af688c34 Unbreak sparc64 build.
Reported by:	tinderbox
2006-08-09 08:24:46 +00:00
Xin LI
7b8fadf879 Fix PC98 build.
Spotted by:	make universe
2006-08-09 05:33:45 +00:00
Sam Leffler
fdffd7edcc disconnect sade until someone fixes tinderbox builds 2006-08-09 05:26:19 +00:00
Yoshihiro Takahashi
2d0fba0551 Add the sade to build-tools list. This fixes a tinderbox. 2006-08-09 02:27:31 +00:00
Brooks Davis
a36aa44a85 Add a new kernel environment variable "boot.netif.mtu" which is used to
set the MTU prior to mounting root via NFS.  This is required if the
server supports a higher than default MTU because the client will not
see the responses otherwise.

MFC after:	3 weeks
2006-08-09 01:56:17 +00:00
Warner Losh
9500ad1d7d Thomas Wintergerst reports that when this tsleep went away, certain
cards stopped working.  Specifically the AVM B1 PCMCIA Card no longer
detected.  Its CIS chain read back as all FF's.  Putting the delay
back solves those problems.  I've opted to put in a much shorter delay
because as far as I can tell, no delay is really needed here.  We'll
see how well this works in practice.
2006-08-09 00:05:54 +00:00
Marcel Moolenaar
d1a5c5275c Fix a phase-ordering bug: check the mediasize and sectorsize after
we obtained access. It is possible that GPT gets to taste a disk
first, which means the disk has not been opened before and it will
not get opened until after we checked the mediasize and sectorsize.
However, since the mediasize and sectorsize are determined at open
and that happens when access is optained, checking the mediasize
and sectorsize before obtaining access may result in GPT rejecting
the disk.
2006-08-08 21:33:26 +00:00
Giorgos Keramidas
49198c4297 Update usage & SYNOPSIS and clarify that input files are not removed.
Sort getopt option handling of -p too, while here.

The changes are adapted from a patch by Ruslan Ermilov, posted as
followup to docs/33852.

PR:             docs/33852
Submitted by:   Gary W. Swearingen <swear@blarg.net>
MFC after:      1 week
2006-08-08 21:25:22 +00:00
Olivier Houchard
49953e11d7 Rewrite ARM_USE_SMALL_ALLOC so that instead of the current behavior, it maps
whole the physical memory, cached, using 1MB section mappings. This reduces
the address space available for user processes a bit, but given the amount of
memory a typical arm machine has, it is not (yet) a big issue.
It then provides a uma_small_alloc() that works as it does for architectures
which have a direct mapping.
2006-08-08 20:59:38 +00:00
Ruslan Ermilov
d251852965 Add FreeBSD 6.2 to the list of known versions. 2006-08-08 19:47:10 +00:00
Robert Watson
ae476dd78f Add kqueue support to if_tun. Loosely based on if_tap changes.
Two almost identical patches based on the if_tap work were submitted
via GNATS; I started out with the patch in 100796 from David Gilbert,
but could have easily started with the patch from Vilmos Nebehaj which
I found only later.

MFC after:	1 week
PR:		93976, 100796
2006-08-08 19:22:25 +00:00
Sam Leffler
05680ab622 check return value of ath_tx_dmasetup
Noticed by:	yongari
2006-08-08 16:42:01 +00:00
Bruce M Simpson
a07fb7be3b Add Epson Stylus CX3650.
Update Dd.

Requested by:	brueffer
2006-08-08 15:42:31 +00:00
Xin LI
fd596b5014 Partitions are only meaningful for i386 and amd64 architectures, so disable
these code on other architectures.
2006-08-08 14:28:03 +00:00
Xin LI
c7d9f7f033 Make sade(8) WARNS=3 clean. 2006-08-08 13:45:46 +00:00
Konstantin Belousov
1565bf54af Lock the vnode around the call to VOP_GETATTR. Move the locked code
and vn_fullpath (that call malloc(..., M_WAITOK)) from under the
vm object lock, since sleep is not allowed while holding the mutex.

Being there, wrap VOP_GETATTR call with conditional Giant aquire.
Currently this is (almost) noop because pseudofs is Giant-locked.

Tested by:	kris
Approved by:	pjd (mentor)
MFC after:	2 weeks
2006-08-08 12:29:26 +00:00
Xin LI
7f557ac0a7 Typo (favour -> favor). 2006-08-08 08:51:37 +00:00
Xin LI
f5a6ba0f79 Shutdown curses when exit. While I am there, remove a stale
comment which does not apply to sade(8).
2006-08-08 07:51:58 +00:00
Gleb Smirnoff
eac2afced2 - Pass pointer to fd_set in ReadSockets.
- style(9) nits.

Submitted by:	ru
2006-08-08 07:04:40 +00:00
Alan Cox
ab83ac429d Reduce the scope of the page queues lock in vfs_busy_pages() now that
vm_page_sleep_if_busy() no longer requires the caller to hold the page
queues lock.
2006-08-08 06:00:49 +00:00
David Xu
065dbdc130 Axe unused member field. 2006-08-08 05:04:43 +00:00
David Xu
6b73f08519 Get number of CPUs and ignore spin count on single processor machine. 2006-08-08 04:42:41 +00:00
Alan Cox
079ba18aac Pass VM_PROT_EXECUTE to vm_fault() instead of VM_PROT_READ if the page
fault was caused by an instruction fetch.
2006-08-08 04:01:29 +00:00
Nate Lawson
7d2498889f Remove the global dock variable. Each dock device should be able to
function independently.  This change is not only load-tested since I don't
have hardware that supports acpi_dock.  Clean up comments and a name a
few constants.
2006-08-08 01:33:03 +00:00
Nate Lawson
ad3d78ead0 If a beep was enabled, turn it off 3 seconds after resume.
MFC after:	3 days
2006-08-08 01:30:54 +00:00
Kevin Lo
400e3077d8 Remove a bogus i = 0.
Approved by: cognet
2006-08-08 01:18:18 +00:00
Alexander Leidinger
41ae1d1e83 Connect sade to the build. 2006-08-07 23:37:51 +00:00
Alexander Leidinger
2e14ff0f41 Say welcome to 'sade', the SysAdmins Disk Editor. It's the fdisk and disklabel part
of sysinstall. So sysinstall may retire now, we have the important non-install part
of it covered.

ATM it doesn't understand GEOM stuff (like mirror, stripe, raid, ...), but patches
to change this and to clean it up internally are more than welcome.

Submitted by:	mami@nyitolap.hu
2006-08-07 23:35:49 +00:00
Alexander Leidinger
e6748c986c Forced commit to note repo copy from sysinstall.
Repo copy by:	markm
2006-08-07 23:20:23 +00:00
Alexander Leidinger
04d895e8a4 Use a low probe-priority to ensure that the emu10kx driver has a higher
priority than this one when both are available.
2006-08-07 23:00:36 +00:00
Alexander Leidinger
507ec6d5c5 "Fix typos in volume control DSP code."
Submitted by:	Yuriy Tsibizov <Yuriy.Tsibizov@gfk.ru>
2006-08-07 22:56:41 +00:00
Alexander Leidinger
0c115520f1 "Fixe playback of mono files on stereo outputs. In previous version
mono files are played only on left channel."

Submitted by:	Yuriy Tsibizov <Yuriy.Tsibizov@gfk.ru>
2006-08-07 22:55:21 +00:00
Alexander Leidinger
feaa7fe133 "Change type for flags bitmap to let use 1 as flag value without
overflow (to -1)."

Submitted by:	Yuriy Tsibizov <Yuriy.Tsibizov@gfk.ru>
2006-08-07 22:50:26 +00:00
Alexander Leidinger
e4c87b14fd "Workaround for sound lag in current snd_emu10kx driver. Real problem
is interaction between in-kernel sound buffer handling and hardware.
 With small buffer, there are times when both harwdare reads and
 kernel writes to the same buffer (it is only visible on slow machines, i
 think). I'm digging in channel.c and buffer.c to find a solution that
 allow use of large hardware buffers without sound lags - hardware can
 handle buffers up to 32Mb."

Submitted by:	Yuriy Tsibizov <Yuriy.Tsibizov@gfk.ru>
2006-08-07 22:44:01 +00:00
Poul-Henning Kamp
5b1deb3ce8 Return length from fwopen() file callback instead of zero.
The symptom is that syslog() fails to log anything but the "ident"
string if LOG_PERROR is specified to openlog(3) and the extensible
printf is in action.

For unclear, likely quaint historical reasons, syslog uses fwopen()
on a stack buffer, rather than using the more straightforward
and faster snprintf().

Along the way, fflush(3) is called, and since the callback writer
function returns zero instead of the length "written", __SERR
naturally gets set on the filedescriptor.

The extensible printf, in difference from the normal printf refuses
to output anything to an __SERR marked filedescriptor, and thus
the actual syslog message is supressed.

MFC:	after 2 weeks
2006-08-07 20:12:18 +00:00
Pawel Jakub Dawidek
7fd26f86d1 Add missing #. 2006-08-07 20:09:09 +00:00
John Baldwin
0cdba3b7e5 When a user uses a hint to specify the IRQ for a link device, accept IRQs
that aren't listed as valid in the link device's set of possible IRQs.
This allows the hints to be used to work around broken BIOSes that don't
specify the correct ste of possible IRQs.  A warning is issued in the
dmesg in this case to be consistent with the $PIR handling code.

MFC after:	1 week
2006-08-07 19:52:30 +00:00
Giorgos Keramidas
bfbb8643cf Fix some minor nits in the README file of wesside
Reviewed by:   sam, Andrea Bittau <a.bittau@cs.ucl.ac.uk>
2006-08-07 17:08:05 +00:00
Yaroslav Tykhiy
ff7e281d6d - Improve and extend mark-up.
- Don't use full path in .Nm (we just don't do that).
- Correct some frivolous and poorly rendering language,
  such as using possessive case for .Nm or .Fl .
- Use the same capitalization for "user ID" as in setuid(2) and getuid(2).
- Bring SEE ALSO in accord with the text.

MFC after:	5 days
2006-08-07 16:05:55 +00:00
Gleb Smirnoff
b9124a7354 Add line edit and history support to ngctl(8) via editline(3).
Details:
- The main thread runs editline(3) functions, that can block.
- A separate thread is launched to monitor netgraph sockets.
- The access to the descriptors is protected by a mutex. At
  runtime the monitoring thread owns the mutex. When the main
  thread reads a command from el_gets() it asks the monitoring
  thread to release a mutex and sleep until the main thread
  processes the command.

This makes ngctl(8) depend on libedit, and libpthread. Thus, the
new functionality isn't compiled in if release is being built
with -DRELEASE_CRUNCH.

PR:		bin/87352
Reviewed by:	ru, Nuno Antunes <nuno.antunes gmail.com>
2006-08-07 14:17:05 +00:00
Gleb Smirnoff
42787b76b6 Recognize the 5750 C2.
Obtained from:	OpenBSD (brad)
2006-08-07 12:51:50 +00:00
Robert Watson
e4445a031f Move definition of UNIX domain socket protosw and domain entries from
uipc_proto.c to uipc_usrreq.c, making localdomain static.  Remove
uipc_proto.c as it's no longer used.  With this change, UNIX domain
sockets are entirely encapsulated in uipc_usrreq.c.
2006-08-07 12:02:43 +00:00
Giorgos Keramidas
6d35e42f0c Spell "determine" correctly.
Reviewed by:	jb
2006-08-07 10:33:07 +00:00
Xin LI
49fe7301af Update NetBSD and OpenBSD SCM tags to match the reality. Note that
NetBSD revisions 1.8 and 1.9 are not actually applied to our code base
because we have solved the problem differently, therefore, these changes
can be safely skipped.
2006-08-07 10:29:18 +00:00
Xin LI
7e87aff9d5 Merge all applicable NetBSD and OpenBSD improvements over this manpage
till NetBSD mailwrapper.8,v 1.11 and OpenBSD mailwrapper.8,v 1.8:

 - Separate exit status out from diagnostics section.
 - Fix typos.

Obtained from:	NetBSD, OpenBSD
MFC After:	Along with mailwrapper(8) updates.
2006-08-07 10:22:08 +00:00
Gleb Smirnoff
2e87c3cc4d - Use log(9) instead of printf(9).
- Print node ID, where possible.
- Prepend log messages with function name, or at least with "ng_pppoe".

Reviewed by:	julian
Tested by:	Joao Barros <joao.barros gmail.com>
2006-08-07 08:05:10 +00:00
Yaroslav Tykhiy
03335ffbe9 Fix things I misadvised to Nate Lawson:
- "Ns No" is redundant, so reduce it to just "Ns";
- use Li to mark up CPU and system power states because here they are
  just strings to be typed literally (add more such mark-up as well);
- the best practice of typesetting tells us to use an en-dash to
  indicate a closed range; fortunately, groff_char(7) provides such a symbol.

In addition, consistently use a comma between independent clauses
coordinated by a conjunction.

Pointed out by:	ru (mark-up part)
MFC after:	3 days
2006-08-07 07:28:47 +00:00