Commit Graph

60705 Commits

Author SHA1 Message Date
John Birrell
47c14e2ae2 ELF compatibility definitions for OpenSolaris. These are covered
by Sun's CDDL and this file is only intended for inclusion where
_SOLARIS_C_SOURCE is defined (with the assumption that the code
being compiled is licensed under the CDDL too).
2006-10-04 07:19:39 +00:00
Martin Blapp
89ff1e4cb8 Back out part of rev. 1.149. While adding a workaround in ptcopen() to
avoid leaked ptys works fine, this opens a possible security hole.

Submitted by:	bde
MFC after:	3 days
2006-10-04 05:43:39 +00:00
Robert Watson
531147aa3e Regenerate. 2006-10-03 20:48:11 +00:00
Robert Watson
dfb041ca62 Change getpagesize() system call audit event to more clearly indicate
that we don't audit it.

MFC after:	3 days
Obtained from:	TrustedBSD Project
2006-10-03 20:48:03 +00:00
Robert Watson
888db9e177 Audit creat() system call (compat code), and change type for getpagesize(),
which isn't actually being audited anyway.

MFC after:	3 days
Obtained from:	TrustedBSD Project
2006-10-03 20:46:52 +00:00
Robert Watson
42edcb0058 Add BSM conversion switch entries for a number of system calls, many
administrative, to prevent console warnings and enable basic event
auditing (generally without arguments).

MFC after:	3 days
Obtained from:	TrustedBSD Project
2006-10-03 20:43:48 +00:00
Konstantin Belousov
30af71199e Fix the remaining race in the revs. 1.232, 1,233 that could occur during
unmount when mp structure is reused while waiting for coveredvp lock.
Introduce struct mount generation count, increment it on each reuse and
compare the generations before and after obtaining the coveredvp lock.

Reviewed by:	tegge, pjd
Approved by:	pjd (mentor)
MFC after:	2 weeks
2006-10-03 10:47:04 +00:00
Gleb Smirnoff
9a3fc40a26 Remove 3Com 985 deviceid that is really ti(4).
Submitted by:	erwin
2006-10-03 09:31:49 +00:00
Ian Dowse
dff926577b Fix compile in non-debug case. 2006-10-03 08:38:08 +00:00
Warner Losh
61db6a13b0 aic_pccard_products can be static.
Noticed by: cscope
2006-10-03 04:51:18 +00:00
John Birrell
678faae279 Solaris compatibility only: Be specific about the fact that
the inline function takes no arguments.
2006-10-03 04:01:30 +00:00
Ian Dowse
9c63c3f400 When changing the device address and max packet size in usbd_new_device(),
close and re-open the default pipe instead of relying on the host
controller driver to notice the changes. Remove the unreliable code
that attempted to update these fields while the pipe was active.
This fixes a case where the hardware could cache and continue to
use the old address, resulting in a "getting first desc failed"
error.

PR:		usb/103167
2006-10-03 01:13:26 +00:00
Ian Dowse
a55e16f4f7 Don't attempt to insert the transfer in an error case where it has
already been inserted.
2006-10-03 01:04:11 +00:00
Ariff Abdullah
3f630febba Backout remaining changes that make most AD1981B users unhappy.
The exact situation is understood, and proper solution will follow
sooner.
2006-10-02 20:46:34 +00:00
John Hay
ae0ddac700 Hopefully the last tweak in trying to make it possible to add ipv6 direct
host routes without side effects.

Submitted by:	JINMEI Tatuya
MFC after:	4 days
2006-10-02 19:15:10 +00:00
Poul-Henning Kamp
e5037a18a9 Use utc_offset() where applicable, and hide the internals of it
as static variables.
2006-10-02 18:23:37 +00:00
John Baldwin
948170b7d4 Trim trailing whitespace. 2006-10-02 18:16:29 +00:00
John Baldwin
278d119ae6 Update description of td_locks.
MFC after: 	3 days
Requested by:	pjd
2006-10-02 17:48:13 +00:00
Poul-Henning Kamp
60e88b878c Use calendrical calculations from subr_clock.c instead of home-rolled. 2006-10-02 16:32:36 +00:00
Ariff Abdullah
0f219a5803 - Streamline quirks management for a better future.
- Fix support for ASUS M5200ae (buggy BIOS)
- Fix few problems, reported by Coverity Prevent (TM).

CID:		246991, 246676, 246675, 246674, 246477
Found by:	Coverity Prevent (TM)
2006-10-02 16:30:04 +00:00
Poul-Henning Kamp
add846a989 Use Calendrical Calculations from subr_clock.c instead of home copy&pasted. 2006-10-02 16:21:16 +00:00
Poul-Henning Kamp
e4c9547050 Use calendaric calculation support from subr_clock.c instead of home-rolled.
Eventually, this RTC should probably use subr_rtc.c as well
2006-10-02 16:18:40 +00:00
Poul-Henning Kamp
f97c1c4bf7 Introduce utc_offset() to capture a calculation currently done all over the
place.
2006-10-02 16:17:23 +00:00
Poul-Henning Kamp
8e52f5465d remove orphaned sysctl_machdep_adjkerntz() 2006-10-02 16:08:20 +00:00
Poul-Henning Kamp
94d67e0fb8 Move tz_minuteswest and tz_dsttime to subr_clock.c 2006-10-02 16:06:26 +00:00
Ruslan Ermilov
1997f96b85 Ignore the UKBD_DFLT_KEYMAP option when compiling as a module.
The parallel LINT build sometimes broke if kernel-depend wasn't
fast enough in generating ukbdmap.h.  If someone thinks this
option would still be useful for the module, a proper fix is
to add the code generating ukbdmap.h into modules/ukbd/Makefile
and backing this change out.
2006-10-02 16:02:41 +00:00
Poul-Henning Kamp
b69f71eb29 Second part of a little cleanup in the calendar/timezone/RTC handling.
Split subr_clock.c in two parts (by repo-copy):
   subr_clock.c contains generic RTC and calendaric stuff. etc.
   subr_rtc.c contains the newbus'ified RTC interface.

Centralize the machdep.{adjkerntz,disable_rtc_set,wall_cmos_clock}
sysctls and associated variables into subr_clock.c.  They are
not machine dependent and we have generic code that relies on being
present so they are not even optional.
2006-10-02 15:42:02 +00:00
Ariff Abdullah
4f53344b73 Disable ad1981_patch for now until the real solution can be found
and makes everybody happy.
2006-10-02 15:32:12 +00:00
Ariff Abdullah
aab23ae39a Make this buildable on IA64.
Reported by:	phk
2006-10-02 15:26:37 +00:00
Poul-Henning Kamp
f645b0b51c First part of a little cleanup in the calendar/timezone/RTC handling.
Move relevant variables to <sys/clock.h> and fix #includes as necessary.

Use libkern's much more time- & spamce-efficient BCD routines.
2006-10-02 12:59:59 +00:00
Robert Watson
3c1b7e8b4d Trim some no longer XXX comments.
Remove some commented out debugging printfs.

MFC after:	3  days
Obtained from:	TrustedBSD Project
2006-10-02 11:32:23 +00:00
George V. Neville-Neil
90ce6fa1c8 Turn off automatic link local address if ipv6_enable is not set to YES
in rc.conf

Reviewed by:    KAME core team, cperciva
MFC after:      3 days
2006-10-02 10:13:30 +00:00
Konstantin Belousov
45ea8737bf Correct the comment: numvnodes is decreased on vdestroying the vnode.
OKed by:	tegge
Approved by:	pjd (mentor)
MFC after:	1 week
2006-10-02 07:25:58 +00:00
Robert Watson
5d8ea5963c Audit path argument when changing audit trails.
Call NDFREE(), which while not currently strictly necessary, isn't a
bad idea.

MFC after:	3 days
Obtained from:	TrustedBSD Project
2006-10-02 06:56:10 +00:00
Tor Egge
ad4276811a Correct check for when IO_SYNC should be set for filesystem
not using softupdates when truncating a directory to zero length.

Discussed with:	bde
2006-10-02 02:08:31 +00:00
Tor Egge
04aa807cb6 If the buffer lock has waiters after the buffer has changed identity then
getnewbuf() needs to drop the buffer in order to wake waiters that might
sleep on the buffer in the context of the old identity.
2006-10-02 02:06:27 +00:00
Matt Jacob
73cf209ffb Put a bit of hysteresis into both BUSY SCSI status returns
and CAM_RESRC_UNAVAIL returns. Delay a tunable amount for
either between retries.

This came up because the MPT IOC was returning "IOC out of
resources" for some user and this caused a CAM_RESRC_UNAVAIL
return. Putting a bit of delay between retries helped them
out.

There was some discussion that an async event should be used
to clear CAM_RESRC_UNAVAIL. That's probably a better notion
eventually.

Reviewed by:	scsi@freebsd.org (ade, scott)
MFC after:	1 week
2006-10-01 20:23:50 +00:00
Ariff Abdullah
f545d39274 Add module loading option for Intel High Definition Audio Controller
- snd_hda(4)
2006-10-01 14:58:30 +00:00
Alexander Leidinger
f06e3ebc56 snd_ak452x is replaced by snd_spicds and snd_envy24ht is new. 2006-10-01 14:57:34 +00:00
Ariff Abdullah
4b8939a1a5 Add notes and option for Intel High Definition Audio Controller
- snd_hda(4)
2006-10-01 14:56:10 +00:00
Ariff Abdullah
9f52a325cd Various cleanups. Revert DMA alignmnet back to 2 since the original
cause of DMA pointer train wreck has been solved.

Reminded by:	silby
MFC after:	1 month
2006-10-01 13:30:30 +00:00
Ariff Abdullah
62c0702891 Unbreak build caused by recent envy24/spicds commit. 2006-10-01 11:34:54 +00:00
Ariff Abdullah
2c72e284ca Connect snd_hda(4) to build process...
Add support for Intel High Definition Audio Controller.

This driver make a special guarantee that "playback" works
on majority hardwares with minimal or without specific vendor
quirk.

This driver is a product of collaborative effort made by:

  Stephane E. Potvin <sepotvin@videotron.ca>
       Andrea Bittau <a.bittau@cs.ucl.ac.uk>
       Wesley Morgan <morganw@chemikals.org>
      Daniel Eischen <deischen@FreeBSD.org>
     Maxime Guillaud <bsd-ports@mguillaud.net>
      Ariff Abdullah <ariff@FreeBSD.org>

....and various people from freebsd-multimedia@FreeBSD.org

Refer to snd_hda(4) for features and issues.

Welcome To HDA.

Sponsored by:	Defenxis Sdn. Bhd.
2006-10-01 11:18:56 +00:00
Ariff Abdullah
1f0387f732 Add support for Intel High Definition Audio Controller.
This driver make a special guarantee that "playback" works
on majority hardwares with minimal or without specific vendor
quirk.

This driver is a product of collaborative effort made by:

  Stephane E. Potvin <sepotvin@videotron.ca>
       Andrea Bittau <a.bittau@cs.ucl.ac.uk>
       Wesley Morgan <morganw@chemikals.org>
      Daniel Eischen <deischen@FreeBSD.org>
     Maxime Guillaud <bsd-ports@mguillaud.net>
      Ariff Abdullah <ariff@FreeBSD.org>

....and various people from freebsd-multimedia@FreeBSD.org

Refer to snd_hda(4) for features and issues.

Welcome To HDA.

Sponsored by:	Defenxis Sdn. Bhd.
2006-10-01 11:13:00 +00:00
Simon L. B. Nielsen
1b43c389b6 Bump __FreeBSD_version for OpenSSL 0.9.8d import. 2006-10-01 08:26:41 +00:00
Andrew Thompson
f26fc2ad3d Remove licence clauses 3 & 4
OKed by:	Jason L. Wright
2006-10-01 03:48:32 +00:00
Martin Blapp
570d6457d1 Readd rev. 1.145 because of vfs bugs and races near revoke(). Until they
are fixed we can't free any slaves. Add a workaround to not to leak ptys
by number.
2006-09-30 22:51:05 +00:00
John Hay
584b68e792 A better fix is to check if it is a host route.
Submitted by:	ume
MFC after:	5 days
2006-09-30 20:25:33 +00:00
Alexander Leidinger
5cc415a9eb Respect style.Makefile(5).
Cluebat waving by:	ru
2006-09-30 18:12:33 +00:00
Alexander Leidinger
5fb623adda Add snd_envy24ht and remove the snd_ak4* module. 2006-09-30 18:07:26 +00:00