]> CyberLeo.Net >> Repos - FreeBSD/FreeBSD.git/log
FreeBSD/FreeBSD.git
21 years ago- scopeid is u_int32_t
ume [Fri, 25 Oct 2002 17:07:02 +0000 (17:07 +0000)]
- scopeid is u_int32_t
- strtoul pedant.  pointed out by deraadt

Obtained from: KAME
MFC after: 1 week

21 years agoExplicitly list architectures supporting sysinstall
phk [Fri, 25 Oct 2002 17:04:49 +0000 (17:04 +0000)]
Explicitly list architectures supporting sysinstall

21 years agoLoose the g_dev_clone() noise.
phk [Fri, 25 Oct 2002 17:00:15 +0000 (17:00 +0000)]
Loose the g_dev_clone() noise.

21 years ago- kill strcpy
ume [Fri, 25 Oct 2002 16:24:28 +0000 (16:24 +0000)]
- kill strcpy
- port range check need to be done before htons.  from deraadt
- %d/%u audit
- correct bad practice in the code - it uses two changing variables
  to manage buffer (buf and buflen).  we eliminate buflen and use
  fixed point (ep) as the ending pointer.
- use snprintf, not sprintf
- pass correct name into q.name.  from lukem@netbsd
- sync comment

Obtained from: KAME
MFC after: 1 week

21 years agoGreatly improve readability of trap() by using a table to convert between
jake [Fri, 25 Oct 2002 16:00:47 +0000 (16:00 +0000)]
Greatly improve readability of trap() by using a table to convert between
trap types and signals to send.  Rearrange KASSERTs to better handle faults
early before curthread is setup, or in the case that it gets corrupted or
set to 0.

21 years agoAdd newly-added sripts to FILES.
thomas [Fri, 25 Oct 2002 15:23:26 +0000 (15:23 +0000)]
Add newly-added sripts to FILES.

Reviewed by: roberto

21 years agoAdd a new /etc/periodic/security script to check for packets
thomas [Fri, 25 Oct 2002 15:16:54 +0000 (15:16 +0000)]
Add a new /etc/periodic/security script to check for packets
rejected by ipfilter (510.ipfdenied), and a corresponding periodic.conf
knob (daily_status_security_ipfdenied_enable).

Reviewed by: roberto
Approved by: re@

21 years agoFactor out code across various /etc/periodic/security scripts into a
thomas [Fri, 25 Oct 2002 15:14:16 +0000 (15:14 +0000)]
Factor out code across various /etc/periodic/security scripts into a
separate file, /etc/periodic/security/security.functions.

Reviewed by: roberto (mentor)
Approved by: re@

21 years agoMFS: Update sysinstall text for the number and size or ports in the ports
bmah [Fri, 25 Oct 2002 14:52:27 +0000 (14:52 +0000)]
MFS:  Update sysinstall text for the number and size or ports in the ports
collection.

src/release/sysinstall/dist.c 1.175.2.29 -> 1.175.2.30

21 years agoadd manpages for FireWire.
simokawa [Fri, 25 Oct 2002 14:34:24 +0000 (14:34 +0000)]
add manpages for FireWire.

21 years agoMake the example "assertion failure" message match reality.
tjr [Fri, 25 Oct 2002 14:22:20 +0000 (14:22 +0000)]
Make the example "assertion failure" message match reality.

21 years agoUse an internal buffer for the result when the first argument is NULL.
tjr [Fri, 25 Oct 2002 13:24:45 +0000 (13:24 +0000)]
Use an internal buffer for the result when the first argument is NULL.

21 years agosuspend thread only when it can be interrupted.
davidxu [Fri, 25 Oct 2002 13:12:36 +0000 (13:12 +0000)]
suspend thread only when it can be interrupted.

21 years agolet thread_schedule_upcall() handle idle kse.
davidxu [Fri, 25 Oct 2002 12:50:31 +0000 (12:50 +0000)]
let thread_schedule_upcall() handle idle kse.

21 years agoUpdate the Standards section for the addition of the C99 va_copy macro.
tjr [Fri, 25 Oct 2002 12:22:14 +0000 (12:22 +0000)]
Update the Standards section for the addition of the C99 va_copy macro.

21 years agoGoodbye and good riddance!
phk [Fri, 25 Oct 2002 11:48:56 +0000 (11:48 +0000)]
Goodbye and good riddance!

21 years agoConfirmed kill: "Bogon #1" is dead, killed by DEVFS.
phk [Fri, 25 Oct 2002 11:47:58 +0000 (11:47 +0000)]
Confirmed kill: "Bogon #1" is dead, killed by DEVFS.

21 years agoComment out more MAKEDEV bogosity ("BOGON #1" as Jordan put it :-)
phk [Fri, 25 Oct 2002 09:54:27 +0000 (09:54 +0000)]
Comment out more MAKEDEV bogosity ("BOGON #1" as Jordan put it :-)

21 years agoPreprocess with tbl(1) only where needed.
ru [Fri, 25 Oct 2002 09:14:09 +0000 (09:14 +0000)]
Preprocess with tbl(1) only where needed.

21 years agoDon't preprocess with tbl(1).
ru [Fri, 25 Oct 2002 09:05:10 +0000 (09:05 +0000)]
Don't preprocess with tbl(1).

21 years agoKill the remnants of BOG.
ru [Fri, 25 Oct 2002 09:03:09 +0000 (09:03 +0000)]
Kill the remnants of BOG.

21 years agoPreprocess with tbl(1).
ru [Fri, 25 Oct 2002 08:55:47 +0000 (08:55 +0000)]
Preprocess with tbl(1).

21 years agoDon't preprocess with tbl(1).
ru [Fri, 25 Oct 2002 08:55:24 +0000 (08:55 +0000)]
Don't preprocess with tbl(1).

21 years agoDon't preprocess with tbl(1).
ru [Fri, 25 Oct 2002 08:40:24 +0000 (08:40 +0000)]
Don't preprocess with tbl(1).

21 years agoDisable the kernacc() check in mtx_validate() until such time that kernacc
phk [Fri, 25 Oct 2002 08:40:20 +0000 (08:40 +0000)]
Disable the kernacc() check in mtx_validate() until such time that kernacc
does not require Giant.

This means that we may miss panics on a class of mutex programming bugs,
but only if running with a Chernobyl setting of debug-flags.

Spotted by: Pete Carah <pete@ns.altadena.net>

21 years agoSimplify.
ru [Fri, 25 Oct 2002 08:39:25 +0000 (08:39 +0000)]
Simplify.

21 years agoDon't preprocess with tbl(1).
ru [Fri, 25 Oct 2002 08:10:40 +0000 (08:10 +0000)]
Don't preprocess with tbl(1).

21 years agoPreprocess with tbl(1).
ru [Fri, 25 Oct 2002 08:06:58 +0000 (08:06 +0000)]
Preprocess with tbl(1).

21 years agoIn vrele() we can actually have a VCHR with v_rdev == NULL if we
phk [Fri, 25 Oct 2002 07:58:25 +0000 (07:58 +0000)]
In vrele() we can actually have a VCHR with v_rdev == NULL if we
came from the bottom of addaliasu().  Don't panic.

21 years agoAdjust the size passed to readlink so that the null termination
dougb [Fri, 25 Oct 2002 07:26:36 +0000 (07:26 +0000)]
Adjust the size passed to readlink so that the null termination
falls within the range of the path variable.

Cribbed from the latest NetBSD source.

Obtained from: provos@NetBSD.org

21 years agofix style-o
julian [Fri, 25 Oct 2002 07:17:07 +0000 (07:17 +0000)]
fix style-o

21 years agoMore work on the interaction between suspending and sleeping threads.
julian [Fri, 25 Oct 2002 07:11:12 +0000 (07:11 +0000)]
More work on the interaction between suspending and sleeping threads.
Also clean up some code used with 'single-threading'.

Reviewed by: davidxu

21 years agoUse the correct values for LDBL_*. Libc doesn't completely support
imp [Fri, 25 Oct 2002 07:02:52 +0000 (07:02 +0000)]
Use the correct values for LDBL_*.  Libc doesn't completely support
long doubles at the moment (printf truncates them to doubles).
However, long doubles to appear to work to the ranges listed in this
commit on both -stable (4.5) and -current.  There may be some slight
rounding issues with long doubles, but that's an orthogonal issue to
these constants.

I've had this in my local tree for 3 months, and in my company's local
tree for 15 months with no ill effects.

Obtained from: NetBSD
Not likely to like it: bde

21 years agoThe ORIENTLOCK macro is no longer needed since all functions use
tjr [Fri, 25 Oct 2002 07:01:56 +0000 (07:01 +0000)]
The ORIENTLOCK macro is no longer needed since all functions use
FLOCKFILE/FUNLOCKFILE explicitly.

21 years agoMinor cleanups.
jake [Fri, 25 Oct 2002 06:26:34 +0000 (06:26 +0000)]
Minor cleanups.
- use fields in sysent instead of PS_STRINGS
- set TSTATE_PRIV in frame0.tf_tstate for what its worth

21 years agoHide the test in <target> under '.if make(<target>)' so as to not get any
jmallett [Fri, 25 Oct 2002 06:17:44 +0000 (06:17 +0000)]
Hide the test in <target> under '.if make(<target>)' so as to not get any
errors/warnings related to crud in said test block.

21 years agoAdd the functionality to create an (empty) GPT from scratch. The
marcel [Fri, 25 Oct 2002 05:23:08 +0000 (05:23 +0000)]
Add the functionality to create an (empty) GPT from scratch. The
code is directly copied from migrate.c. The intend is to express
migrate in terms of create and add. The functionality to add
partitions is not yet there.

21 years agoNew release note: New cd(4) ioctls, burncd(8) -s max.
bmah [Fri, 25 Oct 2002 03:50:41 +0000 (03:50 +0000)]
New release note:  New cd(4) ioctls, burncd(8) -s max.

Modified release note:  cdcontrol(1) speed max.

While here, note some belated MFCs:  burncd(8) -d, burncd(8) VCD/SVCD
support.

Mostly submitted by: nate

21 years agoThe FTP connection caching needs a better interface -- connections are
njl [Fri, 25 Oct 2002 01:17:32 +0000 (01:17 +0000)]
The FTP connection caching needs a better interface -- connections are
closed through _fetch_close() which is the only one who knows the connection
REALLY was closed (since ref -> 0).  However, FTP keeps its own local
cached_connection and checks if it is valid by comparing it to NULL.  This
is bogus since it may have been freed elsewhere by _fetch_close().

This change checks if we are closing the cached_connection and the ref is 1
(soon to be 0).  If so, set cached_connection to NULL so we don't
accidentally reuse it.  The REAL fix should be to move connection caching
to the common.c level (_fetch_* functions) and NULL the cache(s) in
_fetch_close().  Then all layers could benefit from caching.

21 years agoWithin ufs, the ffs_sync and ffs_fsync functions did not always
mckusick [Fri, 25 Oct 2002 00:20:37 +0000 (00:20 +0000)]
Within ufs, the ffs_sync and ffs_fsync functions did not always
check for and/or report I/O errors. The result is that a VFS_SYNC
or VOP_FSYNC called with MNT_WAIT could loop infinitely on ufs in
the presence of a hard error writing a disk sector or in a filesystem
full condition. This patch ensures that I/O errors will always be
checked and returned.  This patch also ensures that every call to
VFS_SYNC or VOP_FSYNC with MNT_WAIT set checks for and takes
appropriate action when an error is returned.

Sponsored by:   DARPA & NAI Labs.

21 years agofix typo.
davidxu [Fri, 25 Oct 2002 00:13:46 +0000 (00:13 +0000)]
fix typo.

21 years agoExtract out KSE specific code from machine specific code
julian [Thu, 24 Oct 2002 23:09:48 +0000 (23:09 +0000)]
Extract out KSE specific code from machine specific code
so that there is ony one copy of it. Fix that one copy
so that KSEs with no mailbox in a KSE program are not a cause
of page faults (this can legitmatly happen).

Submitted by: (parts) davidxu

21 years agoOops, forgot to commit this file. This is part of the fix
mux [Thu, 24 Oct 2002 22:32:13 +0000 (22:32 +0000)]
Oops, forgot to commit this file.  This is part of the fix
for ipfw2 panics on sparc64.

21 years agoRedo reference vectors to include the mediasize and sectorsize.
phk [Thu, 24 Oct 2002 21:35:57 +0000 (21:35 +0000)]
Redo reference vectors to include the mediasize and sectorsize.

21 years agoUse a better test to prevent tasting geom.ctl so we don't screw the
phk [Thu, 24 Oct 2002 21:32:49 +0000 (21:32 +0000)]
Use a better test to prevent tasting geom.ctl so we don't screw the
regression tests.

21 years agoResync with kernel.
phk [Thu, 24 Oct 2002 21:14:05 +0000 (21:14 +0000)]
Resync with kernel.

21 years agoFix problem with my ability to tell the difference between 'r' and 'l'... We
jmallett [Thu, 24 Oct 2002 20:37:58 +0000 (20:37 +0000)]
Fix problem with my ability to tell the difference between 'r' and 'l'... We
want to check var[1], not val[1].

Submitted by: Mark Valentine <mark@thuvia.demon.co.uk>
Pointed out by: sam
Pointy hat to: jmallett

21 years agoFix the spechash lock order reversal by keeping an updated sum
phk [Thu, 24 Oct 2002 19:38:56 +0000 (19:38 +0000)]
Fix the spechash lock order reversal by keeping an updated sum
of v_usecount in the dev_t which vcount() can return without
locking any vnodes.

Seen by: jhb

21 years agoMake sure GEOM has stopped rattling the disks before we try to mount
phk [Thu, 24 Oct 2002 19:26:08 +0000 (19:26 +0000)]
Make sure GEOM has stopped rattling the disks before we try to mount
the root filesystem, this may be implicated in the PC98 issue.

21 years agoDon't taste the first provider, it's /dev/geom.ctl and it's not going
phk [Thu, 24 Oct 2002 19:20:28 +0000 (19:20 +0000)]
Don't taste the first provider, it's /dev/geom.ctl and it's not going
to taste like anything we like anyway.

21 years agoOops, I missed a few changes in 'device acpica' -> 'device acpi' change.
jhb [Thu, 24 Oct 2002 19:17:06 +0000 (19:17 +0000)]
Oops, I missed a few changes in 'device acpica' -> 'device acpi' change.

Submitted by: Hiten Pandya <hiten@angelica.unixdaemons.com>

21 years agoRename 'device acpica' to 'device acpi'.
jhb [Thu, 24 Oct 2002 19:05:04 +0000 (19:05 +0000)]
Rename 'device acpica' to 'device acpi'.

Approved by: msmith, iwasaki

21 years agoDocument the (alpha only) "out of memory" and "PT_LOAD: too few segments"
gallatin [Thu, 24 Oct 2002 18:41:02 +0000 (18:41 +0000)]
Document the (alpha only) "out of memory" and "PT_LOAD: too few segments"
problems with the new groff and mention possible workarounds for people
buildworlding to current.

Reviewed by: ru (a slighly older version)

21 years agoFix ipfw2 panics on 64-bit platforms.
mux [Thu, 24 Oct 2002 18:04:44 +0000 (18:04 +0000)]
Fix ipfw2 panics on 64-bit platforms.

Quoting luigi:

In order to make the userland code fully 64-bit clean it may
be necessary to commit other changes that may or may not cause
a minor change in the ABI.

Reviewed by: luigi

21 years agosrc and dst address were erroneously swapped in SRC_SET and DST_SET
luigi [Thu, 24 Oct 2002 18:01:53 +0000 (18:01 +0000)]
src and dst address were erroneously swapped in SRC_SET and DST_SET
commands.  Use the correct one. Also affects ipfw2 in -stable.

21 years agoFix a bug in MAKEDEV -- when creating /dev/usb, the code was using
luigi [Thu, 24 Oct 2002 17:59:58 +0000 (17:59 +0000)]
Fix a bug in MAKEDEV -- when creating /dev/usb, the code was using
whatever random value was already in $unit. It happened to work
because the variable is often unset, but this is not always the
case e.g. when you call "MAKEDEV sio2 usb ..."

This affects -stable as well.

Reviewed by: n_hibma
Silence from: re
MFC after: 3 days

21 years agoDon't try to be cute and save a call/return by implementing a degenerate
phk [Thu, 24 Oct 2002 17:55:49 +0000 (17:55 +0000)]
Don't try to be cute and save a call/return by implementing a degenerate
vrele() inline.

21 years agoanother int * to size_t * change for getbsize() to make it compile
gallatin [Thu, 24 Oct 2002 17:43:56 +0000 (17:43 +0000)]
another int * to size_t *  change for getbsize() to make it compile
on LP64 hosts

21 years agogetbsize() now takes a pointer to a size_t as its first arg. A pointer
gallatin [Thu, 24 Oct 2002 17:27:32 +0000 (17:27 +0000)]
getbsize() now takes a pointer to a size_t as its first arg.  A pointer
to an int no longer works on an LP64 platform...

21 years agoReport the MONITOR bit.
phk [Thu, 24 Oct 2002 17:23:52 +0000 (17:23 +0000)]
Report the MONITOR bit.

21 years agoProvide kernel options for the various MAC policy modules so that
rwatson [Thu, 24 Oct 2002 17:21:40 +0000 (17:21 +0000)]
Provide kernel options for the various MAC policy modules so that
they may be statically linked into the kernel.  Note that statically
linked modules, unlike dynamically linked modules, get INVARIANTS,
so if there are INVARIANTS failures, you'll bump into them rather
than not.  Add the options to NOTES.

Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories

21 years agodevice rc no longer requires a count since jhb newbussified the rc
rwatson [Thu, 24 Oct 2002 16:46:53 +0000 (16:46 +0000)]
device rc no longer requires a count since jhb newbussified the rc
device driver.

21 years agoHave ftpd specify the LOGIN_SETMAC flag to setlogincontext() so that
rwatson [Thu, 24 Oct 2002 16:19:52 +0000 (16:19 +0000)]
Have ftpd specify the LOGIN_SETMAC flag to setlogincontext() so that
MAC labels are set if MAC is enabled and configured for the user
logging in.

Note that lukemftpd is not considered a supported application when
MAC is enabled, as it does not use the standard system interfaces for
managing user contexts; if lukemftpd is used with labeled MAC policies,
it will not properly give up privileges when switching to the user
account.

Approved by: re
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories

21 years ago# WARNING: lukemftpd does not support PAM, MAC, per-class nologin files,
rwatson [Thu, 24 Oct 2002 15:46:10 +0000 (15:46 +0000)]
# WARNING: lukemftpd does not support PAM, MAC, per-class nologin files,
# or any login.conf resource limits or features; use it only if this is
# appropriate for your environment.  If you require these features, use
# the regular FreeBSD ftpd below.

Discourage users from using lukemftpd if they rely any of these standard
FreeBSD features that are fully supported by our native ftpd.  There
may be other features that are not yet supported that I have not yet
discovered.

21 years agoInstall mac_partition include files.
rwatson [Thu, 24 Oct 2002 15:25:37 +0000 (15:25 +0000)]
Install mac_partition include files.

Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories

21 years agorespect TDF_SINTR, also for SINGLE_NO_EXIT threading mode, if a thread
davidxu [Thu, 24 Oct 2002 14:43:48 +0000 (14:43 +0000)]
respect TDF_SINTR, also for SINGLE_NO_EXIT threading mode, if a thread
was already suspended, do nothing.

21 years ago#ifdef out assignToPartition on non x86 arches to unbreak the world
gallatin [Thu, 24 Oct 2002 13:35:52 +0000 (13:35 +0000)]
#ifdef out assignToPartition on non x86 arches to unbreak the world
on alpha, sparc64 and ia64

21 years agoEliminate warnings.
ru [Thu, 24 Oct 2002 13:31:34 +0000 (13:31 +0000)]
Eliminate warnings.

21 years agoPreprocess with tbl(1), eliminate warnings.
ru [Thu, 24 Oct 2002 13:21:29 +0000 (13:21 +0000)]
Preprocess with tbl(1), eliminate warnings.

21 years agoRestored sigaction's name in its prototype.
bde [Thu, 24 Oct 2002 13:03:46 +0000 (13:03 +0000)]
Restored sigaction's name in its prototype.

21 years agoFixed a warning.
ru [Thu, 24 Oct 2002 12:59:10 +0000 (12:59 +0000)]
Fixed a warning.

21 years agoWhen expanding a specific [1-char] variable, only expand said specific
jmallett [Thu, 24 Oct 2002 12:57:42 +0000 (12:57 +0000)]
When expanding a specific [1-char] variable, only expand said specific
[1-char] variable.  Don't just automatically expand something which
starts with that character.

Obtained from: OpenBSD [3 years ago!]

21 years ago* Modernize aio(4), providing instructions for static and dynamic kernel
sheldonh [Thu, 24 Oct 2002 12:57:05 +0000 (12:57 +0000)]
* Modernize aio(4), providing instructions for static and dynamic kernel
  linking.

* Fix disorder in the SEE ALSO sections of aio_*(2).

* Remove unnecessary cross-references from the SEE ALSO sections of
  aio_*(2); config(8), kldload(8) and kldunload(8) are cross-referenced
  from aio(4).

* Remove the KERNEL OPTIONS sections from aio_*(2), now that these
  pages cross-reference aio(4), which contains suitable kernel linking
  reference material.

21 years agoFixed references and a bunch of other warnings.
ru [Thu, 24 Oct 2002 12:56:20 +0000 (12:56 +0000)]
Fixed references and a bunch of other warnings.

21 years agoAdd cross-references to the aio(4) manual page.
des [Thu, 24 Oct 2002 12:22:57 +0000 (12:22 +0000)]
Add cross-references to the aio(4) manual page.

Submitted by: Craig Rodrigues <rodrigc@attbi.com>

21 years agoAdd a simple man page for the aio facility.
des [Thu, 24 Oct 2002 12:22:19 +0000 (12:22 +0000)]
Add a simple man page for the aio facility.

21 years agoA version of SGR disabling stuff that is believed to be safe.
ru [Thu, 24 Oct 2002 11:05:58 +0000 (11:05 +0000)]
A version of SGR disabling stuff that is believed to be safe.

Submitted by: Werner LEMBERG <wl@gnu.org>

21 years agoAdd new loader tunable hw.acpi.ec.event_driven so that we control
iwasaki [Thu, 24 Oct 2002 10:56:33 +0000 (10:56 +0000)]
Add new loader tunable hw.acpi.ec.event_driven so that we control
EcWaitEventIntr()'s behavior.
Some EC controllers require this for working properly (including
sleep/wakeup).

21 years agoFix kernel build on sparc64 in the IPDIVERT case.
mux [Thu, 24 Oct 2002 09:58:50 +0000 (09:58 +0000)]
Fix kernel build on sparc64 in the IPDIVERT case.

21 years agodon't forget to remove kse from idle queue.
davidxu [Thu, 24 Oct 2002 09:16:46 +0000 (09:16 +0000)]
don't forget to remove kse from idle queue.

21 years agoMove thread related code from kern_proc.c to kern_thread.c.
julian [Thu, 24 Oct 2002 08:46:34 +0000 (08:46 +0000)]
Move thread related code from kern_proc.c to kern_thread.c.
Add code to free KSEs and KSEGRPs on exit.
Sort KSE prototypes in proc.h.
Add the missing kse_exit() syscall.

ksetest now does not leak KSEs and KSEGRPS.

Submitted by: (parts) davidxu

21 years ago - Now that uma_zalloc_internal is not the fast path don't be so fussy about
jeff [Thu, 24 Oct 2002 07:59:03 +0000 (07:59 +0000)]
 - Now that uma_zalloc_internal is not the fast path don't be so fussy about
   extra function calls.  Refactor uma_zalloc_internal into seperate functions
   for finding the most appropriate slab, filling buckets, allocating single
   items, and pulling items off of slabs.  This makes the code significantly
   cleaner.
 - This also fixes the "Returning an empty bucket." panic that a few people
   have seen.

Tested On: alpha, x86

21 years agoo Fix a size calculation based on a 8KB page, while under EFI
marcel [Thu, 24 Oct 2002 07:53:12 +0000 (07:53 +0000)]
o  Fix a size calculation based on a 8KB page, while under EFI
   pages are 4KB.
o  As a second order fix, don't assume we have enough space
   after the bootinfo block left in a page to hold the memory
   map.
o  A third order fix as that we removed the assumption that a
   bootinfo block fits in a single 8KB page.

PR: ia64/39415
submitted by: Espen Skoglund <esk@ira.uka.de>

21 years agoRemove unneeded variables.
nyan [Thu, 24 Oct 2002 07:35:30 +0000 (07:35 +0000)]
Remove unneeded variables.

21 years agoThe rc driver is not needed for pc98.
nyan [Thu, 24 Oct 2002 07:18:24 +0000 (07:18 +0000)]
The rc driver is not needed for pc98.

21 years agoBuild the following modules on ia64 as well:
marcel [Thu, 24 Oct 2002 07:16:24 +0000 (07:16 +0000)]
Build the following modules on ia64 as well:
ar, asr, bktr, cardbus, cbb, ep, exca, hea, hfa, ipfilter, lomac,
netgraph, pccard, rc, sound, sppp, vinum.

21 years ago - Move the destructor calls so that they are not called with the zone lock
jeff [Thu, 24 Oct 2002 06:17:30 +0000 (06:17 +0000)]
 - Move the destructor calls so that they are not called with the zone lock
   held.  This avoids a lock order reversal when destroying zones.
   Unfortunately, this also means that the free checks are not done before
   the destructor is called.

Reported by: phk

21 years agoUnbreak by merging the change in r1.51 of src/libexec/ftpd/Makefile.
kuriyama [Thu, 24 Oct 2002 04:55:25 +0000 (04:55 +0000)]
Unbreak by merging the change in r1.51 of src/libexec/ftpd/Makefile.

21 years agoDuplicate the variable name in the v->name field, as otherwise it points to
jmallett [Thu, 24 Oct 2002 04:10:55 +0000 (04:10 +0000)]
Duplicate the variable name in the v->name field, as otherwise it points to
data that will be modified.  And do the appropriate thing now and free the
v->name buffer along with other relinquished memory.

XXX There is duplication here of destroying a Var, which is probably bogus,
and probably missed in a few places.

21 years agoRemove gcc-specific optimization/debugging CFLAGS
kris [Thu, 24 Oct 2002 03:56:16 +0000 (03:56 +0000)]
Remove gcc-specific optimization/debugging CFLAGS

21 years agoReplace wcsstr() with an implementation based on strstr(), which is far
tjr [Thu, 24 Oct 2002 02:53:45 +0000 (02:53 +0000)]
Replace wcsstr() with an implementation based on strstr(), which is far
more efficient. The problem with the previous implementation was that it
calculated the length of the first argument ("big") with wcslen() when
it was not necessary.

21 years agoRestore Berkeley SCCS id.
tjr [Thu, 24 Oct 2002 02:48:45 +0000 (02:48 +0000)]
Restore Berkeley SCCS id.

21 years agoCommit of Makefile missed in earlier pass.
rwatson [Thu, 24 Oct 2002 02:04:03 +0000 (02:04 +0000)]
Commit of Makefile missed in earlier pass.

Submitted by: kuriyama

21 years agoUnbreak the automatic remapping of an INADDR_ANY destination address
iedowse [Thu, 24 Oct 2002 02:02:34 +0000 (02:02 +0000)]
Unbreak the automatic remapping of an INADDR_ANY destination address
to the primary local IP address when doing a TCP connect(). The
tcp_connect() code was relying on in_pcbconnect (actually in_pcbladdr)
modifying the passed-in sockaddr, and I failed to notice this in
the recent change that added in_pcbconnect_setup(). As a result,
tcp_connect() was ending up using the unmodified sockaddr address
instead of the munged version.

There are two cases to handle: if in_pcbconnect_setup() succeeds,
then the PCB has already been updated with the correct destination
address as we pass it pointers to inp_faddr and inp_fport directly.
If in_pcbconnect_setup() fails due to an existing but dead connection,
then copy the destination address from the old connection.

21 years agoMinor tweaks to the acpi.4 man page: xref acpiconf.8, and one of the
rwatson [Thu, 24 Oct 2002 01:44:50 +0000 (01:44 +0000)]
Minor tweaks to the acpi.4 man page: xref acpiconf.8, and one of the
bugs (device naming) isn't present [anymore].

Submitted by: Hiten Pandya <hiten@angelica.unixdaemons.com>

21 years agoRemove the Standards section again until we get these functions sorted
tjr [Thu, 24 Oct 2002 01:24:26 +0000 (01:24 +0000)]
Remove the Standards section again until we get these functions sorted
out. This will probably have to wait until after 5.0-R.

21 years agoPlace mac_prepare() with the other mac_prepare*() functions.
chris [Thu, 24 Oct 2002 01:16:56 +0000 (01:16 +0000)]
Place mac_prepare() with the other mac_prepare*() functions.

21 years agoAttempt improved use of fts results: use the correct path to the
rwatson [Thu, 24 Oct 2002 01:01:53 +0000 (01:01 +0000)]
Attempt improved use of fts results: use the correct path to the
object to retrieve label information on, rather than directly
consuming the fts-provided paths (none of which are quite right).
This is based on the similar readlink() code, and may contain
the same bugs.

Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories

21 years agomac_free() no longer accepts a void * parameter; only mac_t's are supposed
chris [Thu, 24 Oct 2002 01:01:29 +0000 (01:01 +0000)]
mac_free() no longer accepts a void * parameter; only mac_t's are supposed
to be passed.  Point this out in a warning notice, which will eventually
go away, sometime between now and -RELEASE.

Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs

21 years agoTeach "ls -Z" to use the policy-agnostic MAC label interfaces rather
rwatson [Thu, 24 Oct 2002 00:07:30 +0000 (00:07 +0000)]
Teach "ls -Z" to use the policy-agnostic MAC label interfaces rather
than the LOMAC-specific interfaces for listing MAC labels.  This permits
ls to view MAC labels in a manner similar to getfmac, when ls is used
with the -l argument.  Next generation LOMAC will use the MAC Framework
so should "just" work with this and other policies.  Not the prettiest
code in the world, but then, neither is ls(1).

Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories