122825 Commits

Author SHA1 Message Date
tegge
69041c93c0 Block secondary writes while expunging active unlinked files.
Fix detection of active unlinked files by checking VI_OWEINACT and
VI_DOINGINACT in addition to v_usecount.

Defer inactive handling for unlinked files if the file system is mostly
suspended (secondary writes being blocked).

Perform deferred inactive handling after the file system is resumed.
2006-03-11 01:08:37 +00:00
philip
dffb931651 Fix multicast support for cs89x0 chips. Just setting the RX_MULTCAST_ACCEPT
flag isn't enough - the filter needs to be set up too, or no multicast frames
are accepted.

Sponsored by:	Philips Industrial Applications (indirectly)
MFC after:	3 days
2006-03-10 23:50:53 +00:00
jkim
87f2ab82b2 Micro-optimize invalid UUID check. 2006-03-10 23:33:27 +00:00
damien
799fc2f615 o acknowledge interrupts early in the interrupt handler
o stop processing interrupts after a firmware fatal error or a radio kill
o clarify the possible values for the 'antenna' sysctl.
o by default, let the firmware do antenna diversity.
  the  firmware will periodically switch to another antenna to evaluate the
  signal quality.
2006-03-10 20:35:09 +00:00
damien
fb0aee8aca make use of the firmware(9) subsystem.
use intel's firmware version 3.0 layout.
2006-03-10 19:59:09 +00:00
sos
905b5752ff Add atausb module 2006-03-10 19:15:20 +00:00
sos
176317a071 Add "device atausb" 2006-03-10 19:10:05 +00:00
sos
5419ca4c93 Module that adds support for using USB storage devices through the
ATA framework. Mainly written to be able to use USB Flash keys.
This is work in progress so use with care :)
Doesn't need CAM and cannot coexist with umass.c
2006-03-10 19:08:52 +00:00
damien
25007ede44 fix the locking. 2006-03-10 18:55:30 +00:00
imp
410473acf0 Add note about 'audit' group.
Add note about the '_dhcp' user.

While one is expected to run mergemaster -p before installworld, make
a note of the points where this will actively fail due to the addition of
users/groups.
2006-03-10 18:40:31 +00:00
trhodes
eeffa50069 Use .Cm for all key words, not just the ones I added. More in line with
mdoc(7) according to ru.

Requested by:	ru
2006-03-10 12:37:19 +00:00
sos
c8d0fc841f write the right unit # on verbose output. 2006-03-10 12:20:53 +00:00
nyan
3b59efd81b Disable an SMP detection on pc98. 2006-03-10 12:18:41 +00:00
ru
ba6a23db6e Hopefully fix all nearby style bugs that Bruce has mentioned. 2006-03-10 11:07:39 +00:00
ru
47517466d5 - Fix variable assignment to be portable.
- "Line up" continuations.

Submitted by:	bde
2006-03-10 11:00:11 +00:00
davidxu
40ba85224c Add entries for new pthread stubs. 2006-03-10 09:58:18 +00:00
davidxu
45ada7f2f2 Block all signals in helper threads except those should not be blocked. 2006-03-10 08:01:23 +00:00
deischen
90b66ded94 Remove including of libc sources. All the required functions
are exported by libc with prototypes in our standard headers.
I guess at one time this was necessary, but not any longer.
2006-03-10 07:58:02 +00:00
pjd
f0925fcaf9 When inserting a new component md_provsize metadata field wasn't set, which
means that old problem was triggered (when two providers end at the same
offset, eg. ad0 and ad0s1 and the wrong was is picked up by gmirror/graid3).

Reported by:	Michal Suszko <dry@dry.pl>
MFC after:	3 days
2006-03-10 07:41:31 +00:00
sam
b1c2c5a706 fix switching between agressive and non-agressive wmm modes
Obtained from:	atheros
MFC after:	3 days
2006-03-10 06:18:03 +00:00
davidxu
651e183fe9 It is not necessary to read %gs twice. 2006-03-10 05:55:26 +00:00
davidxu
b36ad9d674 Fix stack offset to allow gcc's stack aligment code to work correctly.
MFC after: 3 days
2006-03-10 02:54:45 +00:00
tegge
bce91540c2 Remove unneeded (and broken) usage of MNT_REF()/MNT_REL(). 2006-03-10 02:31:12 +00:00
cognet
b82f657662 MFp4: Forget the asm inlined version of in_cksum_hdr(). It doesn't work if
the pointer is unaligned, and it just doesn't worth it.
2006-03-09 23:33:59 +00:00
jkim
9dcc540f35 - Implement serial numbers, UUID, and asset tag (turned off by default).
Use 'BOOT_SENSITIVE_INFO=YES' variable to turn them on.
- Use 'uint*_t' instead of 'u_int*_t', correct compilation warnings, and
update copyright while I am here.
2006-03-09 22:49:44 +00:00
jkim
7e4fc3b8f9 Implement printf 'X' conversion for both libstand and kernel. 2006-03-09 22:37:34 +00:00
sam
83bdb1ecb3 missed a case that needs WITH_SMP
Noticed by:	kensmith
2006-03-09 21:54:32 +00:00
gad
00873ac107 Rever the previous changes. It turns out that it perfectly correct
for a makefile to set 'NO_MAN=' when the makefile is for a program
that will not create a man page.

Based on reaction from:		ru bde
2006-03-09 19:52:44 +00:00
ru
c7c425a278 Fix this to what I've actually submitted. 2006-03-09 17:55:08 +00:00
yar
6dff0c5ef2 Revise the names of modules in the recent note on tdfx_linux. 2006-03-09 17:50:01 +00:00
sam
68d84579df handle KERNELS as before
Submitted by:	ru, andrew
2006-03-09 17:11:17 +00:00
jhb
329536bd48 Flip the switch and don't route interrupts to hyperthreads in a HT system.
In at least one benchmark this showed around a 20% performance increase.
If other workloads do benefit from having hyperthreads service interrupts,
we can always make this a loader tunable.

MFC after:	3 days
Tested by:	ps
2006-03-09 16:38:52 +00:00
imp
4e16f81076 Don't dereference an uninitialized pointer. 2006-03-09 16:28:24 +00:00
imp
7570cde46b Revert last change, per ru@'s objection. I misunderstood consensus 2006-03-09 16:24:58 +00:00
yar
4a02e1707a Avoid pulling in the whole <net/pfvar.h> by opaquely declaring
the structs pflog_packet() takes pointers to.

Approved by:	mlaier
MFC after:	3 days
2006-03-09 15:54:01 +00:00
yar
7260b91842 The pf' and pflog' sources do not depend on DEV_PF or DEV_PFLOG,
which is normal for own files of a device driver.

DEV_FOO should be used if an unrelated kernel file needs to know of
the `foo' driver's static presence.  Obviously, module source files
should never use DEV_*.
2006-03-09 15:48:45 +00:00
yar
ca115e8e21 Revert to setting vlan and vlandev parametes synchronously, as soon
as both have been read from the command line.  Still use the callback,
but this time only to verify that both vlan and vlandev have been
found on the command line.

This should allow for control over the relative order of processing
parameters, which is needed to satisfy some caveats of the if_vlan
driver.  E.g., MTU cannot be changed on a vlan interface until it's
attached to its parent.

PR:             bin/94028
Reviewed by:    ru
MFC after:      3 days
2006-03-09 14:58:09 +00:00
phk
057f1af5e2 Add -F option to not fail on already installed packages.
Submitted by:   Eugene Grosbein <eugen@grosbein.pp.ru>
2006-03-09 14:49:19 +00:00
phk
5a08101277 In addition to -b (suppress builds) add -k (suppress kernel build) and
-w (suppress world).
2006-03-09 14:43:07 +00:00
ru
4f01c7cf85 Don't acquire a lock before calling vlan_unconfig().
This fixes a panic when doing "ifconfig ... -vlandev".

OK'ed by:	glebius
2006-03-09 14:42:51 +00:00
keramida
cbce9d083e After revision 1.4 of `src/usr.sbin/bsnmpd/bsnmpd/Makefile' the
gensnmptree utility is required at "make depend" time, because it
is used to generate `oid.h' for bsnmpd.

This means that all versions of HEAD before 2004/01/23 16:22:49 need
gensnmptree as a bootstrap util when it's not already installed as
part of the base system.  The first __FreeBSD_version that we can
assume this can work is 700014 which happened after the gensnmptree
change (in -r 1.263 of `src/sys/sys/param.h', at 2006/02/17 14:14:15).

For __FreeBSD_version values before 700014 add gensnmptree to the
bootstrap tools, to allow upgrades from versions of FreeBSD before
that date.

Approved by:	ru, harti
2006-03-09 12:55:02 +00:00
sos
05ddef3112 Only use the channel in the request given to ata_completed if its
still present.
2006-03-09 10:25:35 +00:00
sos
e55404d344 typo. 2006-03-09 10:24:03 +00:00
phk
d27836ecbb Oops, forgot newline. 2006-03-09 09:44:10 +00:00
phk
15ba3472e1 Add slop to "backwards" cpu accounting messages, 3 usec or 1% whichever
triggers.

This should eliminate all the trivial messages which result from minor
increases in cpu_tick frequency.

Machines which don't du cpu clock fiddling shouldn't issue "backwards"
messages now.

Laptops and other machines where the initial estimate of cputicks may be
waaaay off will still issue warnings.
2006-03-09 09:33:17 +00:00
phk
64b4ff0373 silence cpu_tick calibration and notice only (under bootverbose)
when the frequency increases.
2006-03-09 09:30:33 +00:00
phk
b7db2706e8 Ignore kenv strings which overflow the room we have, rather than pretend
we have room for them.
2006-03-09 09:29:41 +00:00
sos
2ba86f7da2 Get rid of all the "long long"/"maxint" casting around in printf's.
On all our platforms intmax == int64_t so simply using %j to print
int64_t's is safe all over, and doesn't pullute the code.
2006-03-09 08:34:44 +00:00
davidxu
a84244a660 Remove _STOPEVENT call, it is already called in issignal, simplify
code for SIGKILL signal.
2006-03-09 08:31:51 +00:00
ume
b2cbc78f68 ns_name_skip was not prefixed by `__'.
Obtained from:	BIND9
2006-03-09 03:47:11 +00:00