]> CyberLeo.Net >> Repos - FreeBSD/FreeBSD.git/log
FreeBSD/FreeBSD.git
18 years agoAdd USB modes.
sos [Sun, 5 Mar 2006 21:32:38 +0000 (21:32 +0000)]
Add USB modes.

18 years agoDocument new chipsets (RT2561, RT2561S and RT2661).
damien [Sun, 5 Mar 2006 20:57:25 +0000 (20:57 +0000)]
Document new chipsets (RT2561, RT2561S and RT2661).

Obtained from: OpenBSD

18 years agoImport microcodes for RT2561, RT2561S and RT2661 chipsets.
damien [Sun, 5 Mar 2006 20:41:51 +0000 (20:41 +0000)]
Import microcodes for RT2561, RT2561S and RT2661 chipsets.
With permission from Ralink Technology.

18 years agoAdd support for the second (RT2561/RT2561S) and third (RT2661 MIMO XR)
damien [Sun, 5 Mar 2006 20:38:55 +0000 (20:38 +0000)]
Add support for the second (RT2561/RT2561S) and third (RT2661 MIMO XR)
generations of 802.11abg chipsets from Ralink Technology.
Get rid of the pccard front-end while I'm here since all adapters are
cardbus ones.

18 years agoAdd support for the second (RT2561/RT2561S) and third (RT2661 MIMO XR)
damien [Sun, 5 Mar 2006 20:36:56 +0000 (20:36 +0000)]
Add support for the second (RT2561/RT2561S) and third (RT2661 MIMO XR)
generations of 802.11abg chipsets from Ralink Technology.
Get rid of the pccard front-end while I'm here since all adapters are
cardbus ones.

Obtained from: OpenBSD

18 years agoUpdate atapi-fd to support direct devices such as disks.
sos [Sun, 5 Mar 2006 20:30:54 +0000 (20:30 +0000)]
Update atapi-fd to support direct devices such as disks.

18 years agoAdd some more pthread stubs so that librt can use them.
deischen [Sun, 5 Mar 2006 18:10:28 +0000 (18:10 +0000)]
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

18 years agoWhen we get an unknown ATAPI comand, try it even if it is not listed,
phk [Sun, 5 Mar 2006 17:52:32 +0000 (17:52 +0000)]
When we get an unknown ATAPI comand, try it even if it is not listed,
in all likelyhood just passing it through will work.

18 years agoAdd two new ATAPI commands.
sos [Sun, 5 Mar 2006 17:43:13 +0000 (17:43 +0000)]
Add two new ATAPI commands.

18 years agoGrammar nits & some mdoc fixes, i.e. use ``.Nm fec<RET>node'' consistently
keramida [Sun, 5 Mar 2006 15:56:22 +0000 (15:56 +0000)]
Grammar nits & some mdoc fixes, i.e. use ``.Nm fec<RET>node'' consistently
throughout the manpage text.

Reviewed by: wpaul

18 years agoRevert `proto ip' back to the previous behavior. The kernel side of
ume [Sun, 5 Mar 2006 15:55:46 +0000 (15:55 +0000)]
Revert `proto ip' back to the previous behavior.  The kernel side of
ipfw2 doesn't allow zero as protocol number.

MFC after: 3 days

18 years agoVarious textual improvements.
keramida [Sun, 5 Mar 2006 15:54:22 +0000 (15:54 +0000)]
Various textual improvements.

18 years ago- Correct grammos in comments and end them with full stops.
schweikh [Sun, 5 Mar 2006 15:53:07 +0000 (15:53 +0000)]
- Correct grammos in comments and end them with full stops.
- Use "if !" instead of empty true branches.
- Don't hardcode script name in usage message, use $0 instead.
- Cleanup some whitespace.

18 years agoEnhance loginfail: it will catch sshd, proftpd and su errors, as well as other programs
matteo [Sun, 5 Mar 2006 15:45:38 +0000 (15:45 +0000)]
Enhance loginfail: it will catch sshd, proftpd and su errors, as well as other programs

PR: conf/70973
Submitted by: Ryan Sommers" <ryans@gamersimpact.com>
Approved by: philip (mentor)
MFC after: 3 days

18 years agoBring the zh_CN.GB2312 translation for -HEAD up-to-date with the English
delphij [Sun, 5 Mar 2006 15:07:57 +0000 (15:07 +0000)]
Bring the zh_CN.GB2312 translation for -HEAD up-to-date with the English
Revisions.

Obtained from: The FreeBSD Simplified Chinese Project CVS

18 years agoFix spelling error.
rwatson [Sun, 5 Mar 2006 13:30:37 +0000 (13:30 +0000)]
Fix spelling error.

Submitted by: remko

18 years agoRemove a useless word.
brueffer [Sun, 5 Mar 2006 10:32:15 +0000 (10:32 +0000)]
Remove a useless word.

PR: 94087
Submitted by: Tadaaki Nagao <nagao@iij.ad.jp>
MFC after: 3 days

18 years agoAdd another inverted amplifier sense quirk for Sony VAIO Type T.
ariff [Sun, 5 Mar 2006 02:26:51 +0000 (02:26 +0000)]
Add another inverted amplifier sense quirk for Sony VAIO Type T.

PR: kern/94056
Submitted by: Yoshihiko Sarumaru <mistral at imasy.or.jp>
MFC after: 3 days

18 years agoturn some WARNs for unknown events into INFOs
mjacob [Sat, 4 Mar 2006 21:46:34 +0000 (21:46 +0000)]
turn some WARNs for unknown events into INFOs

18 years agoCreate a mac_bsdextended_check_vp function that takes a cred, a
dwmalone [Sat, 4 Mar 2006 20:47:19 +0000 (20:47 +0000)]
Create a mac_bsdextended_check_vp function that takes a cred, a
vnode and a mode and checks if a given access mode is permitted.
This centralises the mac_bsdextended_enabled check and the GETATTR
calls and makes the implementation of the mac policy methods simple.

This should make it easier for us to match vnodes on more complex
attributes than just uid and gid in the future, but for now there
should be no functional change.

Approved/Reviewed by: rwatson, trhodes
MFC after: 1 month

18 years agoWe need to check if file system size is equal to provider's size, because
pjd [Sat, 4 Mar 2006 19:41:54 +0000 (19:41 +0000)]
We need to check if file system size is equal to provider's size, because
sysinstall(8) still bogusly puts first partition at offset 0 instead of 16,
so glabel/ufs will find file system on slice instead of partition.

Before sysinstall is fixed, we must keep this code, which means that we
wont't be able to detect UFS file systems created with 'newfs -s ...'.

PS. bsdlabel(8) creates partitions properly.

MFC after: 3 days

18 years agoDelete file from FreeBSD vendor branch of OpenBSM that has been removed
rwatson [Sat, 4 Mar 2006 17:14:22 +0000 (17:14 +0000)]
Delete file from FreeBSD vendor branch of OpenBSM that has been removed
from the TrustedBSD OpenBSM distribution.

Obtained from: TrustedBSD Project

18 years agoCount drops when the first of two pipe mallocs fails.
rwatson [Sat, 4 Mar 2006 17:09:17 +0000 (17:09 +0000)]
Count drops when the first of two pipe mallocs fails.

Obtained from: TrustedBSD Project

18 years agoUpdate src/sys/security/audit for OpenBSM 1.0 alpha 5:
rwatson [Sat, 4 Mar 2006 17:00:55 +0000 (17:00 +0000)]
Update src/sys/security/audit for OpenBSM 1.0 alpha 5:

- Include audit_internal.h to get definition of internal audit record
  structures, as it's no longer in audit.h.  Forward declare au_record
  in audit_private.h as not all audit_private.h consumers care about
  it.

- Remove __APPLE__ compatibility bits that are subsumed by configure
  for user space.

- Don't expose in6_addr internals (non-portable, but also cleaner
  looking).

- Avoid nested include of audit.h in audit_private.h.

Obtained from: TrustedBSD Project

18 years agoUpdate src/sys/bsm for OpenBSM 1.0 alpha 5 changes:
rwatson [Sat, 4 Mar 2006 16:54:21 +0000 (16:54 +0000)]
Update src/sys/bsm for OpenBSM 1.0 alpha 5 changes:

- Add new comments.
- Move private data structures from public audit.h to audit_internal.h to
  avoid exposing queue.h macros to undesiring consumers.

Obtained from: TrustedBSD Project

18 years agoUpdate FreeBSD import instructions for OpenBSM based on the new autoconf
rwatson [Sat, 4 Mar 2006 16:50:04 +0000 (16:50 +0000)]
Update FreeBSD import instructions for OpenBSM based on the new autoconf
and automake OpenBSM world order.

Obtained from: TrustedBSD Project

18 years agoAdd generated config.h for FreeBSD.
rwatson [Sat, 4 Mar 2006 16:49:24 +0000 (16:49 +0000)]
Add generated config.h for FreeBSD.

Obtained from: TrustedBSD Project

18 years agoRemove files deleted in the OpenBSM distribution from the OpenBSM
rwatson [Sat, 4 Mar 2006 16:48:36 +0000 (16:48 +0000)]
Remove files deleted in the OpenBSM distribution from the OpenBSM
vendor branch.

Obtained from: TrustedBSD Project

18 years agoThis commit was generated by cvs2svn to compensate for changes in r156283,
rwatson [Sat, 4 Mar 2006 16:45:52 +0000 (16:45 +0000)]
This commit was generated by cvs2svn to compensate for changes in r156283,
which included commits to RCS files with non-trunk default branches.

18 years agoVendor branch import of TrustedBSD OpenBSM 1.0 alpha 5:
rwatson [Sat, 4 Mar 2006 16:45:52 +0000 (16:45 +0000)]
Vendor branch import of TrustedBSD OpenBSM 1.0 alpha 5:

- Update install notes to indicate /etc files are to be installed manually.
- On systems without LOG_SECURITY, use LOG_AUTH.
- Convert to autoconf/automake in order to move to a more portable (not
  BSD-specific) build infrastructure, and more easy conditional building of
  components.  Currently, the primary feature loss is that automake does
  not have native support for manual symlinks.  This will be addressed in a
  future OpenBSM release.
- Add compat/queue.h, to be used on systems dated BSD queue macro libraries
  (as found on Linux).
- Rename CHANGELOG to HISTORY, as our change log doesn't follow some of the
  existing conventions for a CHANGELOG.
- Some private data structures moved from audit.h to audit_internal.h to
  prevent inappropriate use by applications and name space pollution.
- Improved detection and use of endian macros using autoconf.
- Avoid non-portable use of struct in6_addr, which is largely opaque.
- Avoid leaking BSD kernel socket related token code to user space in
  bsm_token.c.
- Teach System V IPC calls to look for Linux naming variations for certain
  struct ipc_perm fields.
- Test for audit system calls, and if not present, don't build
  bsm_wrappers.c, bsm_notify.c, audit(8), and auditd(8), which rely on
  those system calls.
- au_close() is not implemented on systems that don't have audit system
  calls, but au_close_buffer() is.
- Work around missing BSDisms in bsm_wrapper.c.
- Fix nested includes so including libbsm.h in an application on Linux
  picks up the necessary definitions.

Obtained from: TrustedBSD Project

18 years agoReduce number of spaces for full name by four, and reduce padding
rwatson [Sat, 4 Mar 2006 16:13:16 +0000 (16:13 +0000)]
Reduce number of spaces for full name by four, and reduce padding
after tty entry by one space in order to provide extra spaces for
the tty entry.  As a result, full pts names are now visible (up
to 999 pts's anyway):

Before:
Login            Name                 TTY  Idle  Login  Time   Office  Phone
robert           Robert Watson       *v0   3:55  Fri    02:54
robert           Robert Watson        p0     19  Sat    11:01
robert           Robert Watson        pts        Sat    14:55

After:

Login            Name             TTY      Idle  Login  Time   Office  Phone
robert           Robert Watson   *v0       5:08  Fri    02:54
robert           Robert Watson    p0          8  Sat    11:01
robert           Robert Watson    pts/5          Sat    14:55

MFC after: 1 week

18 years agocrypto.ko depends on zlib.
wkoszek [Sat, 4 Mar 2006 15:50:46 +0000 (15:50 +0000)]
crypto.ko depends on zlib.

Submitted by: Ben Kelly <bkelly at vadev.org>
Approved by: rwatson
Point hat to: me
MFC after: 1 day

18 years agoUse `intmax_t' instead of plain `int' for pid_t casts.
keramida [Sat, 4 Mar 2006 15:20:28 +0000 (15:20 +0000)]
Use `intmax_t' instead of plain `int' for pid_t casts.

Useful tips from: ru, bde
Approved by: pjd
MFC after: 3 days

18 years agoAdd OpenBSM and src/sys/security/audit to the maintainer list.
rwatson [Sat, 4 Mar 2006 14:24:32 +0000 (14:24 +0000)]
Add OpenBSM and src/sys/security/audit to the maintainer list.

18 years agoFix evil examples - first partition should start at offset 16.
pjd [Sat, 4 Mar 2006 12:09:50 +0000 (12:09 +0000)]
Fix evil examples - first partition should start at offset 16.

MFC after: 3 days

18 years agofixed a bug that an MLD report is not advertised when group-specific MLD query is...
suz [Sat, 4 Mar 2006 09:17:11 +0000 (09:17 +0000)]
fixed a bug that an MLD report is not advertised when group-specific MLD query is received.

PR: kern/93526
Obtained from: KAME
MFC after: 1 day

18 years agoMFi386: revision 1.1220.
nyan [Sat, 4 Mar 2006 07:31:58 +0000 (07:31 +0000)]
MFi386: revision 1.1220.

18 years agoAdd missing cast.
phk [Sat, 4 Mar 2006 06:07:26 +0000 (06:07 +0000)]
Add missing cast.

18 years agoMore detailed logging if timestepwarnings are enabled.
phk [Sat, 4 Mar 2006 06:06:43 +0000 (06:06 +0000)]
More detailed logging if timestepwarnings are enabled.

18 years agoInstead of just hinting at available octets, list some.
trhodes [Sat, 4 Mar 2006 02:38:40 +0000 (02:38 +0000)]
Instead of just hinting at available octets, list some.
Mention that the setting of securelevel may affect one's ability to alter flags.
Xref security.7.
Bump doc date.

18 years agoUse a thread pool to process notification if sigev_notify_attributes
davidxu [Sat, 4 Mar 2006 00:18:19 +0000 (00:18 +0000)]
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.

18 years agouse strlcpy in cvtstatfs and copy_statfs instead of bcopy to ensure
ps [Sat, 4 Mar 2006 00:09:09 +0000 (00:09 +0000)]
use strlcpy in cvtstatfs and copy_statfs instead of bcopy to ensure
the copied strings are properly terminated.

bzero the statfs32 struct in copy_statfs.

18 years agosave sigev_node pointer.
davidxu [Sat, 4 Mar 2006 00:08:24 +0000 (00:08 +0000)]
save sigev_node pointer.

18 years agoFix bug in malloc_uninit():
ps [Fri, 3 Mar 2006 22:36:52 +0000 (22:36 +0000)]
Fix bug in malloc_uninit():
Releasing items from the mt_zone can not be done by a simple
uma_zfree() call since mt_zone is allocated with the UMA_ZONE_MALLOC
flag. Use uma_zfree_arg instead and supply the slab.

This bug caused panics in low memory situations on unloading kernel
modules containing MALLOC_DEFINE(..) statements.

Submitted by: ups

18 years agoTake the functionality contained in the former "options TDFX_LINUX"
yar [Fri, 3 Mar 2006 21:37:38 +0000 (21:37 +0000)]
Take the functionality contained in the former "options TDFX_LINUX"
into a separate module.  Accordingly, convert the option into a device
named similarly.

Note for MFC: Perhaps the option should stay in RELENG_6 for POLA reasons.

Suggested by: scottl
Reviewed by: cokane
MFC after: 5 days

18 years agoPrint "clear" and "dump" only once.
wkoszek [Fri, 3 Mar 2006 21:35:57 +0000 (21:35 +0000)]
Print "clear" and "dump" only once.

Reviewed by: pjd
Approved by: cognet (mentor)
MFC after: 3 days

18 years agoDon't build csup if NO_CRYPT or NO_OPENSSL is defined, since we depend
mux [Fri, 3 Mar 2006 21:12:05 +0000 (21:12 +0000)]
Don't build csup if NO_CRYPT or NO_OPENSSL is defined, since we depend
on those.

Reported by: marius

18 years agoCast the pointer to void * before casting it back to struct type * in
mux [Fri, 3 Mar 2006 18:54:33 +0000 (18:54 +0000)]
Cast the pointer to void * before casting it back to struct type * in
STAILQ_LAST.  This quiets a warning from GCC about increased required
alignment for the cast.

Idea from:      cognet

18 years ago- use a more common style to print memory sizes
netchild [Fri, 3 Mar 2006 18:54:05 +0000 (18:54 +0000)]
- use a more common style to print memory sizes
- add some more cache sizes (2nd and 3rd level) [1]

Submitted by: HATANOU Tomomi <hatanou@infolab.ne.jp> [1]
PR: 91328 [1]

18 years agoThis commit was generated by cvs2svn to compensate for changes in r156251,
mux [Fri, 3 Mar 2006 18:45:12 +0000 (18:45 +0000)]
This commit was generated by cvs2svn to compensate for changes in r156251,
which included commits to RCS files with non-trunk default branches.

18 years agoImport today's csup sources to get the warning fix for queue.h. We
mux [Fri, 3 Mar 2006 18:45:12 +0000 (18:45 +0000)]
Import today's csup sources to get the warning fix for queue.h.  We
probably want to do something similar in sys/queue.h; it's bad to have
STAILQ_LAST() generate GCC warnings.

18 years agoFix the decaying MODULES_WITH_WORLD again by adding a missing opt_*.h to SRCS.
yar [Fri, 3 Mar 2006 18:42:09 +0000 (18:42 +0000)]
Fix the decaying MODULES_WITH_WORLD again by adding a missing opt_*.h to SRCS.

18 years agoMove XBOX option to options. While it is only valid on i386,
imp [Fri, 3 Mar 2006 18:09:37 +0000 (18:09 +0000)]
Move XBOX option to options.  While it is only valid on i386,
syscons_isa is shared with other machines.

18 years agoUnbreak byte counters when network interfaces are in monitor mode by
csjp [Fri, 3 Mar 2006 17:21:08 +0000 (17:21 +0000)]
Unbreak byte counters when network interfaces are in monitor mode by
re-organizing the monitor return logic. We perform interface monitoring
checks after we have determined if the CRC is still on the packet, if
it is, m_adj() is called which will adjust the packet length. This
ensures that we are not including CRC lengths in the byte counters for
each packet.

Discussed with: andre, glebius

18 years agoConnect csup the build!
mux [Fri, 3 Mar 2006 14:55:43 +0000 (14:55 +0000)]
Connect csup the build!

18 years ago- Don't put the .h in SRCS.
mux [Fri, 3 Mar 2006 14:54:47 +0000 (14:54 +0000)]
- Don't put the .h in SRCS.
- Use LDADD += -lpthread and DPADD += ${LIBPTHREAD} instead of -pthread.
- s/-I${.CURDIR}/-I./ to fix make obj.

Submitted by: ru

18 years agoCommitted the xbox syscons(8)-able console driver.
rink [Fri, 3 Mar 2006 14:52:57 +0000 (14:52 +0000)]
Committed the xbox syscons(8)-able console driver.

Reviewed by:    arch@ (no comments)
Approved by:    imp (mentor)

18 years agoSync with actual code.
jcamou [Fri, 3 Mar 2006 13:58:43 +0000 (13:58 +0000)]
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

18 years agoCorrectly switch IFCAP_VLAN_HWTAGGING on and off.
glebius [Fri, 3 Mar 2006 12:32:51 +0000 (12:32 +0000)]
Correctly switch IFCAP_VLAN_HWTAGGING on and off.

Reported & tested by: Peter Blok <pblok bsd4all.org>

18 years ago- Do not leak read lock in IP_FW_TABLE_GETSIZE case of ipfw_ctl().
glebius [Fri, 3 Mar 2006 12:10:59 +0000 (12:10 +0000)]
- Do not leak read lock in IP_FW_TABLE_GETSIZE case of ipfw_ctl().
- Acquire read (not write) lock in case of IP_FW_TABLE_LIST.

In collaboration with: ru

18 years agoDocument the monitor mode and explain how it can be used.
thompsa [Fri, 3 Mar 2006 09:32:25 +0000 (09:32 +0000)]
Document the monitor mode and explain how it can be used.

18 years agoSince we are using random ethernet addresses for the bridge, it is possible
thompsa [Fri, 3 Mar 2006 09:12:21 +0000 (09:12 +0000)]
Since we are using random ethernet addresses for the bridge, it is possible
that we might have address collisions, so make sure that this hardware address
isn't already in use on another bridge.

Submitted by: csjp
MFC after: 1 month

18 years agoDon't truncate f_mntfromname & f_mntonname to 16 characters when
ps [Fri, 3 Mar 2006 07:20:54 +0000 (07:20 +0000)]
Don't truncate f_mntfromname & f_mntonname to 16 characters when
translating statfs into ostatfs.  This allows 4.x binaries making
statfs calls to work on 6.x.

18 years agomicro fix from justin
mjacob [Fri, 3 Mar 2006 07:04:43 +0000 (07:04 +0000)]
micro fix from justin

18 years agoSlightly re-worked bpf(4) code associated with bridging: if we have a
csjp [Fri, 3 Mar 2006 05:58:18 +0000 (05:58 +0000)]
Slightly re-worked bpf(4) code associated with bridging: if we have a
destination interface as a member of our bridge or this is a unicast packet,
push it through the bpf(4) machinery.

For broadcast or multicast packets, don't bother with the bpf(4) because it will
be re-injected into ether_input. We do this before we pass the packets through
the pfil(9) framework, as it is possible that pfil(9) will drop the packet or
possibly modify it, making it very difficult to debug firewall issues on the
bridge.

Further, implemented IFF_MONITOR for bridge interfaces. This does much the same
thing that it does for regular network interfaces: it pushes the packet to any
bpf(4) peers and then returns. This bypasses all of the bridge machinery,
saving mutex acquisitions, list traversals, and other operations performed by
the bridging code.

This change to the bridging code is useful in situations where individuals use a
bridge to multiplex RX/TX signals from two interfaces, as is required by some
network taps for de-multiplexing links and transmitting the RX/TX signals
out through two separate interfaces. This behaviour is quite common for network
taps monitoring links, especially for certain manufacturers.

Reviewed by: thompsa
MFC after: 1 month
Sponsored by: Seccuris Labs

18 years agoiir works on PAE now.
scottl [Fri, 3 Mar 2006 04:30:18 +0000 (04:30 +0000)]
iir works on PAE now.

18 years agoAdd make glue for building csup. Not yet connected to the build.
mux [Fri, 3 Mar 2006 04:28:13 +0000 (04:28 +0000)]
Add make glue for building csup.  Not yet connected to the build.

18 years agoInitial import of csup.
mux [Fri, 3 Mar 2006 04:11:29 +0000 (04:11 +0000)]
Initial import of csup.

18 years agoThis commit was generated by cvs2svn to compensate for changes in r156230,
mux [Fri, 3 Mar 2006 04:11:29 +0000 (04:11 +0000)]
This commit was generated by cvs2svn to compensate for changes in r156230,
which included commits to RCS files with non-trunk default branches.

18 years agoWhen there are no mount options, an implicit "rw" should be printed in
keramida [Fri, 3 Mar 2006 02:46:15 +0000 (02:46 +0000)]
When there are no mount options, an implicit "rw" should be printed in
the output of ``mount -p''.

Approved by: rodrigc

18 years ago- Print message about cpufreq and timecounter TSC
mnag [Fri, 3 Mar 2006 02:06:04 +0000 (02:06 +0000)]
- Print message about cpufreq and timecounter TSC

Approved by: njl
MFC after: 1 day

18 years ago- Add entry about using timecounter TSC and cpu frequency changing.
mnag [Fri, 3 Mar 2006 02:04:52 +0000 (02:04 +0000)]
- Add entry about using timecounter TSC and cpu frequency changing.

Approved by: njl
MFC after: 1 day

18 years agoFix endian bugs accessing ioctl arguments that are passed by value.
emax [Fri, 3 Mar 2006 00:46:28 +0000 (00:46 +0000)]
Fix endian bugs accessing ioctl arguments that are passed by value.

Pointed out by: marius
MFC after: 1 day

18 years agoEliminate a deadlock when creating snapshots. Blocking vn_start_write() must
tegge [Thu, 2 Mar 2006 22:13:28 +0000 (22:13 +0000)]
Eliminate a deadlock when creating snapshots.  Blocking vn_start_write() must
be called without any vnode locks held.  Remove calls to vn_start_write() and
vn_finished_write() in vnode_pager_putpages() and add these calls before the
vnode lock is obtained to most of the callers that don't already have them.

18 years agoHold extra reference to vm object while cleaning pages.
tegge [Thu, 2 Mar 2006 21:38:38 +0000 (21:38 +0000)]
Hold extra reference to vm object while cleaning pages.

18 years agoDon't try to show marker nodes.
tegge [Thu, 2 Mar 2006 21:31:15 +0000 (21:31 +0000)]
Don't try to show marker nodes.

18 years agoDocument which version of 802.1D STP we implement.
thompsa [Thu, 2 Mar 2006 20:50:33 +0000 (20:50 +0000)]
Document which version of 802.1D STP we implement.

18 years agoFix types of time_second and time_uptime variables.
ru [Thu, 2 Mar 2006 19:47:41 +0000 (19:47 +0000)]
Fix types of time_second and time_uptime variables.

18 years agoThe altq(4) code already handles SMP so clarify what ALTQ_NOPCC is needed for.
thompsa [Thu, 2 Mar 2006 19:45:59 +0000 (19:45 +0000)]
The altq(4) code already handles SMP so clarify what ALTQ_NOPCC is needed for.

18 years agoUpdate URL at cisco.com.
glebius [Thu, 2 Mar 2006 15:11:05 +0000 (15:11 +0000)]
Update URL at cisco.com.

18 years agoActually bump the date to the commit-date for the last commit.
brueffer [Thu, 2 Mar 2006 14:55:07 +0000 (14:55 +0000)]
Actually bump the date to the commit-date for the last commit.

18 years agoDocument the graid3(8), gstripe(8) and gconcat(8) periodic scripts.
brueffer [Thu, 2 Mar 2006 14:48:17 +0000 (14:48 +0000)]
Document the graid3(8), gstripe(8) and gconcat(8) periodic scripts.

18 years agoAdd the graid3(8), gstripe(8) and gconcat(8) status scripts, default is "off".
brueffer [Thu, 2 Mar 2006 14:46:00 +0000 (14:46 +0000)]
Add the graid3(8), gstripe(8) and gconcat(8) status scripts, default is "off".

Approved by: rwatson (mentor)

18 years agoAdd periodic scripts that check the status of graid3(8), gstripe(8)
brueffer [Thu, 2 Mar 2006 14:44:19 +0000 (14:44 +0000)]
Add periodic scripts that check the status of graid3(8), gstripe(8)
and gconcat(8) devices, respectively.

Approved by: rwatson (mentor)

18 years agoo Fix an URL to the "The BSD Family Tree" article.
maxim [Thu, 2 Mar 2006 14:38:45 +0000 (14:38 +0000)]
o Fix an URL to the "The BSD Family Tree" article.

Spotted by: Marcin Jessa

18 years agoAdd signal set sq_kill to sigqueue structure, the member saves all
davidxu [Thu, 2 Mar 2006 14:06:40 +0000 (14:06 +0000)]
Add signal set sq_kill to sigqueue structure, the member saves all
signals sent by kill() syscall, without this, a signal sent by
sigqueue() can cause a signal sent by kill() to be lost.

18 years agoUse BUS_DMA_COHERENT.
cognet [Thu, 2 Mar 2006 14:06:38 +0000 (14:06 +0000)]
Use BUS_DMA_COHERENT.

18 years agoUpdate copyright years and remove third clause.
des [Thu, 2 Mar 2006 14:00:40 +0000 (14:00 +0000)]
Update copyright years and remove third clause.

18 years agoRead site.rc after default.rc.
des [Thu, 2 Mar 2006 13:58:04 +0000 (13:58 +0000)]
Read site.rc after default.rc.

18 years agoIgnore generated files.
des [Thu, 2 Mar 2006 13:54:01 +0000 (13:54 +0000)]
Ignore generated files.

18 years agoAdd nested include of <wchar.h>
phk [Thu, 2 Mar 2006 10:01:52 +0000 (10:01 +0000)]
Add nested include of <wchar.h>

18 years agoFix the %Q printf extension to behave as expected
phk [Thu, 2 Mar 2006 08:53:45 +0000 (08:53 +0000)]
Fix the %Q printf extension to behave as expected

18 years ago - Acquire lk in softdep_slowdown so that it's owned when we call
jeff [Thu, 2 Mar 2006 08:52:53 +0000 (08:52 +0000)]
 - Acquire lk in softdep_slowdown so that it's owned when we call
   softdep_speedup().
 - Assert that lk is held in softdep_speedup() rather than acquiring it.
   This avoids a potential lock recursion.

18 years agoSuffer a little bit of math every 16 second and tighten calibration of
phk [Thu, 2 Mar 2006 08:09:46 +0000 (08:09 +0000)]
Suffer a little bit of math every 16 second and tighten calibration of
cpu_ticks to the low side of PPM.

18 years ago - Move softdep from using a global worklist to per-mount worklists. This
jeff [Thu, 2 Mar 2006 05:50:23 +0000 (05:50 +0000)]
 - Move softdep from using a global worklist to per-mount worklists.  This
   has many positive effects including improved smp locking, reducing
   interdependencies between mounts that can lead to deadlocks, etc.
 - Add the softdep worklist and various counters to the ufsmnt structure.
 - Add a mount pointer to the workitem and remove mount pointers from the
   various structures derived from the workitem as they are now redundant.
 - Remove the poor-man's semaphore protecting softdep_process_worklist and
   softdep_flushworklist.  Several threads may now process the list
   simultaneously.
 - Add softdep_waitidle() to block the thread until all pending
   dependencies being operated on by other threads have been flushed.
 - Use softdep_waitidle() in unmount and snapshots to block either
   operation until the fs is stable.
 - Remove softdep worklist processing from the syncer and move it into the
   softdep_flush() thread.  This thread processes all softdep mounts
   once each second and when it is called via the new softdep_speedup()
   when there is a resource shortage.  This removes the softdep hook
   from the kernel and various hacks in header files to support it.

Reviewed by/Discussed with: tegge, truckman, mckusick
Tested by: kris

18 years ago - Lock Giant if needed around the call to vnode_create_vobject(). This is
jeff [Thu, 2 Mar 2006 05:37:44 +0000 (05:37 +0000)]
 - Lock Giant if needed around the call to vnode_create_vobject().  This is
   only important if devfs is not mpsafe.

Sponsored by: Isilon Systems, Inc.
Found by: kris

18 years agoDo not use the TSC where its known to be broken, this will cause the queue
thompsa [Thu, 2 Mar 2006 00:51:39 +0000 (00:51 +0000)]
Do not use the TSC where its known to be broken, this will cause the queue
speeds to perform below the desired bitrate and throughput will be erratic.

This makes queueing work on the Geode SC1100, K5 model 0 and IDT WinChip C6
processors.

MFC after: 3 days

18 years agoUse 8 * sizeof(int) instead of hardcoding 32, for the unlikely case this
cognet [Thu, 2 Mar 2006 00:37:40 +0000 (00:37 +0000)]
Use 8 * sizeof(int) instead of hardcoding 32, for the unlikely case this
code ever get used on a plateform where sizeof(int) != 4.

Suggested by: jmg

18 years agoNote that BUS_DMA_COHERENT currently only affects sparc64 and arm.
cognet [Wed, 1 Mar 2006 23:56:18 +0000 (23:56 +0000)]
Note that BUS_DMA_COHERENT currently only affects sparc64 and arm.

Suggested by: scottl

18 years ago1. Fix a race in aio_return.
davidxu [Wed, 1 Mar 2006 23:41:47 +0000 (23:41 +0000)]
1. Fix a race in aio_return.
2. Save and restore syscall errno correctly.
3. Style fix.