Hajimu UMEMOTO [Tue, 7 Mar 2006 19:13:00 +0000 (19:13 +0000)]
Remove trailing dot from the servername in the serverlist to be able
to use portsnap from behind a http proxy. Some HTTP proxy don't like
trailing dot in the servername.
Yaroslav Tykhiy [Tue, 7 Mar 2006 09:53:11 +0000 (09:53 +0000)]
Two years ago, in rev. 1.12, a typo slipped in etc/isdn/Makefile:
a control variable was misspelled as `SCRIPTS_NAME_isdntel.sh'
(should've been `SCRIPTSNAME_isdntel.sh') when the Makefile was
converted to using <bsd.prog.mk>. Consequently, `isdntel.sh' has
been installed as `isdntel' since then. Nevertheless, nobody has
been unhappy with the change. So it's time to make it official.
Approved by: hm
Pointy hat to: ru
MFC after: 3 days
David Xu [Tue, 7 Mar 2006 08:28:07 +0000 (08:28 +0000)]
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.
Tim Kientzle [Tue, 7 Mar 2006 06:05:38 +0000 (06:05 +0000)]
Change the output of --version to match GNU "gnits" standards.
(I'm not using GPL, but I still think there are good
ideas in the GNU projects. ;-) Among other things,
this should make it easier for clients of bsdtar to
recognize it automatically:
bsdtar --version | grep bsdtar
Add an override (DONTSTRIP) so that one can easily install debugable
binaries when one cannot easily edit a package's Makefiles and/or
configure scripts.
Revert parts of 1.51 and add a missing \& after "i.e." that is not the
end of a sentence. Ruslan notes that:
* The part about hexadecimal representation was intentional -- node ID
is parsed as the ng_parse_hint32_type, and is represented
(input/output) as a hexadecimal number
* "This value" was more correct, as the alternative name is
"[<value>]:" where <value> is hexadecimal value of the node ID.
* "ID based name" (which is "[<hexid>]:") was correct, and what's now is
incorrect -- node ID (number) cannot be equivalent to a name.
Yaroslav Tykhiy [Mon, 6 Mar 2006 12:54:03 +0000 (12:54 +0000)]
login.access.5 and login_access.c are no longer used
in usr.bin/login because the login.access feature has
moved to PAM completely.
Their counterparts in lib/libpam/modules/pam_login_access
have been found to be in sync with, and even in better shape
than, login.access.5 and login_access.c here.
Therefore cvs rm login.access.5 and login_access.c from
usr.bin/login so that nobody will waste their time on fixing
or developing the files here.
By default (for security reasons) syslogd(8) doesn't create log files
when they don't exist, but sometimes its quite useful (eg. we use
non-standard log files and memory backed /var/, which is populated on
boot).
Add -C option which tells syslogd(8) to create log files if they don't
exist.
Add new 'setkeyboard' method to the /etc/rc.d/syscons. It accepts the
keyboard device name (i.e. /dev/kbd0). This method will do nothing is
kbdmux(4) is the current active keyboard, otherwise it will switch
active keyboard as requested.
Modify ukbd(4) entries in the /etc/devd.conf to use /etc/rc.d/syscons
and new 'setkeyboard' method.
No comments from: freebsd-current@
MFC after: 1 day
Andrew Thompson [Mon, 6 Mar 2006 02:28:41 +0000 (02:28 +0000)]
If we miss the LINK_UP event from the network interface then the bridge port
will remain in the disabled state until another link event happens in the
future (if at all). Add a timer to periodically check the interface state and
recover.
Reported by: Nik Lam <freebsdnik j2d.lam.net.au>
MFC after: 3 days
Yaroslav Tykhiy [Sun, 5 Mar 2006 22:52:17 +0000 (22:52 +0000)]
Retire NETSMBCRYPTO as a kernel option and make its functionality
enabled by default in NETSMB and smbfs.ko.
With the most of modern SMB providers requiring encryption by
default, there is little sense left in keeping the crypto part
of NETSMB optional at the build time.
This will also return smbfs.ko to its former properties users
are rather accustomed to.
Discussed with: freebsd-stable, re (scottl)
Not objected by: bp, tjr (silence)
MFC after: 5 days
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.
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.
Daniel Eischen [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.
- 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.
David Malone [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.
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 ...'.
Robert Watson [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.
Robert Watson [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.
Robert Watson [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
Tom Rhodes [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.
David Xu [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.
Paul Saab [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.
Yaroslav Tykhiy [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
Maxime Henrion [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.