]> CyberLeo.Net >> Repos - FreeBSD/FreeBSD.git/log
FreeBSD/FreeBSD.git
18 years agoSmall style cleanup.
Robert Watson [Sat, 9 Sep 2006 10:23:00 +0000 (10:23 +0000)]
Small style cleanup.

MFC after: 3 days

18 years agoAudit sysarch() operation argument.
Robert Watson [Sat, 9 Sep 2006 10:20:31 +0000 (10:20 +0000)]
Audit sysarch() operation argument.

MFC after: 3 days

18 years agoCatch up to xchat and samba package changes
Kris Kennaway [Sat, 9 Sep 2006 07:25:40 +0000 (07:25 +0000)]
Catch up to xchat and samba package changes

18 years agoAdd support for TSO. Thanks to Andre for adding support in the stack
Prafulla Deuskar [Sat, 9 Sep 2006 06:19:20 +0000 (06:19 +0000)]
Add support for TSO. Thanks to Andre for adding support in the stack
and Jack Vogel for driver changes.

Submitted by: Jack Vogel

18 years agoBump __FreeBSD_version for rev. 1.117 of libexec/rtld-elf/rtld.c.
Konstantin Belousov [Sat, 9 Sep 2006 04:41:40 +0000 (04:41 +0000)]
Bump __FreeBSD_version for rev. 1.117 of libexec/rtld-elf/rtld.c.

Requested by: jkim
Approved by: kan (mentor)

18 years agoAdd support to bge(4) to not break IPMI support when the driver attaches
Doug Ambrisko [Sat, 9 Sep 2006 03:36:57 +0000 (03:36 +0000)]
Add support to bge(4) to not break IPMI support when the driver attaches
to it.  Try to co-operate with the IPMI/ASF firmware accessing the PHY.
One we get link we don't mess with the PHY.  If we do then over time
the NIC will go off line.  It would be nice if we could tell if IPMI
was enabled on the chip but I can't figure out a reliable way to do
that.  The scheme I tried worked on a Dell PE850 but not on an HP machine.
So we assume any NIC that has ASF capability needs to deal with it.

The code was inspired by the support in Linux from kernel.org and Broadcom.
Broadcom did give me some info. but it is rather limited and is mostly
just what is in the Linux driver.  Thanks to the numerous people that
helped debug the many prior versions and that I didn't break other
bge(4) HW.

Reviewed by: several people
Tested by: even more

18 years agoThe following functions need not to be reimplemented, reuse 64bit
David Xu [Sat, 9 Sep 2006 01:22:13 +0000 (01:22 +0000)]
The following functions need not to be reimplemented, reuse 64bit
syscalls instead:
sigqueue, thr_set_name, thr_setscheduler, thr_getscheduler,
thr_setschedparam.

18 years agoFix style nits. No md5 changes in .o's. ;-)
Jung-uk Kim [Fri, 8 Sep 2006 21:46:01 +0000 (21:46 +0000)]
Fix style nits.  No md5 changes in .o's. ;-)

18 years agoStop cutting the tarball names to eight symbols. This makes
Ruslan Ermilov [Fri, 8 Sep 2006 19:54:37 +0000 (19:54 +0000)]
Stop cutting the tarball names to eight symbols.  This makes
it possible to e.g. distribute kernels with config names larger
than eight symbols, without the clash.  Previousy, LALALALA
and LALALALA-SMP would be the same tarball.  (I think this
comes from the old days where tarballs were put on the MS-DOS
formatted diskettes.)

MFC after: 3 days

18 years agoChange the way base allocation is done for internal malloc data
Jason Evans [Fri, 8 Sep 2006 17:52:15 +0000 (17:52 +0000)]
Change the way base allocation is done for internal malloc data
structures, in order to avoid the possibility of attempted recursive
lock acquisition for chunks_mtx.

Reported by: Slawa Olhovchenkov <slw@zxy.spb.ru>

18 years agoMF6: Attach if_bce.ko to the build.
Ruslan Ermilov [Fri, 8 Sep 2006 15:40:36 +0000 (15:40 +0000)]
MF6: Attach if_bce.ko to the build.

Noticed by: davidch

18 years agoTemporarily put ugly workaround in place to make sparc64 build happy.
Maksim Yevmenkin [Fri, 8 Sep 2006 15:05:38 +0000 (15:05 +0000)]
Temporarily put ugly workaround in place to make sparc64 build happy.
panther is down at the moment, so better fix will follow later.

18 years agoWhen looking up the symbol by dlsym, look it not only in the object
Konstantin Belousov [Fri, 8 Sep 2006 14:59:54 +0000 (14:59 +0000)]
When looking up the symbol by dlsym, look it not only in the object
given as dso handle, but also in the implicit dependencies of that dso.

Also, const-ify the read-only parameter objlist of symlook_list.

Reported by: "Simon 'corecode' Schubert" <corecode at fs ei tum de>
Approved by: kan (mentor)
X-MFC-After: 6.2

18 years agoDocument that bits are numbered from 1 in the %b format argument.
Ruslan Ermilov [Fri, 8 Sep 2006 14:05:03 +0000 (14:05 +0000)]
Document that bits are numbered from 1 in the %b format argument.

18 years agoRe-word the description of the "async" flag.
Daniel Gerzo [Fri, 8 Sep 2006 13:47:39 +0000 (13:47 +0000)]
Re-word the description of the "async" flag.

Suggested by: Milos Vyletel (mv@rulez.sk)
Reviewed by: ru
Approved by: keramida (mentor), trhodes (mentor)

18 years agoFix format character.
Pawel Jakub Dawidek [Fri, 8 Sep 2006 13:46:18 +0000 (13:46 +0000)]
Fix format character.

Reported by: andre

18 years agoFix octal representation of TSO4 and TSO6 bits in interface capabilities
Andre Oppermann [Fri, 8 Sep 2006 13:29:38 +0000 (13:29 +0000)]
Fix octal representation of TSO4 and TSO6 bits in interface capabilities
description.

18 years agoAdd a sysctl net.inet.tcp.nolocaltimewait that allows to suppress
Gleb Smirnoff [Fri, 8 Sep 2006 13:09:15 +0000 (13:09 +0000)]
Add a sysctl net.inet.tcp.nolocaltimewait that allows to suppress
creating a compress TIME WAIT states, if both connection endpoints
are local. Default is off.

18 years agoBump copyright year.
Pawel Jakub Dawidek [Fri, 8 Sep 2006 10:20:44 +0000 (10:20 +0000)]
Bump copyright year.

18 years agoUse __FBSDID in .c files.
Pawel Jakub Dawidek [Fri, 8 Sep 2006 10:19:24 +0000 (10:19 +0000)]
Use __FBSDID in .c files.

18 years agoIn "make universe", utilize simplified cross-build synopsis,
Ruslan Ermilov [Fri, 8 Sep 2006 10:13:15 +0000 (10:13 +0000)]
In "make universe", utilize simplified cross-build synopsis,
"make TARGET=<machine> build*", now that it actually works.

Prompted by: imp

18 years agoPrevious revision wasn't enough for "make TARGET=<machine> build*"
Ruslan Ermilov [Fri, 8 Sep 2006 10:09:02 +0000 (10:09 +0000)]
Previous revision wasn't enough for "make TARGET=<machine> build*"
to work.  Some build-tools that are part of cross-tools (cc_tools)
also need to know the proper values of TARGET_ARCH and TARGET, so
tell them.

18 years agoUpdate some of the bulgarian holidays since it no longer
Erwin Lansing [Fri, 8 Sep 2006 09:45:06 +0000 (09:45 +0000)]
Update some of the bulgarian holidays since it no longer
is a socialist state.

Submitted by: Shteryana Shopova <shteryana@FreeBSD.org>

18 years agoUse return value of _thr_umutex_lock instead of using zero.
David Xu [Fri, 8 Sep 2006 09:29:14 +0000 (09:29 +0000)]
Use return value of _thr_umutex_lock instead of using zero.

18 years ago- Split failure probability configuration into read failure probability and
Pawel Jakub Dawidek [Fri, 8 Sep 2006 09:21:21 +0000 (09:21 +0000)]
- Split failure probability configuration into read failure probability and
  write failure probability.
- Allow to specify an error number to return of failure.

MFC after: 3 days

18 years agoMinor comment fix.
Thomas Quinot [Fri, 8 Sep 2006 08:14:32 +0000 (08:14 +0000)]
Minor comment fix.

18 years agoSupport for PCI-Express 4Gb Cards.
Matt Jacob [Fri, 8 Sep 2006 05:27:04 +0000 (05:27 +0000)]
Support for PCI-Express 4Gb Cards.

18 years agoSo there is where that handbook paragraph came from. Kill it here too.
Tom Rhodes [Fri, 8 Sep 2006 04:56:21 +0000 (04:56 +0000)]
So there is where that handbook paragraph came from.  Kill it here too.
Remove a paragraph about over building security, it's a bit off.

Discussed with: des, FreeBSD-security

18 years agoMake 8139C+ work again which was broken since rev 1.68.
Pyun YongHyeon [Fri, 8 Sep 2006 00:58:02 +0000 (00:58 +0000)]
Make 8139C+ work again which was broken since rev 1.68.
Ever since rev 1.68 re(4) checks the validity of link in re_start.
But rlphy(4) got a garbled data due to a different bit layout used on
8139C+ and it couldn't report correct link state. To fix it, ignore
BMCR_LOOP and BMCR_ISO bits which have different meanings on 8139C+.
I think this also make dhclient(8) work on 8139C+.

Reported by: Gerrit Kuehn <gerrit AT pmp DOT uni-hannover DOT de>
Tested by: Gerrit Kuehn <gerrit AT pmp DOT uni-hannover DOT de>

18 years agoSigh. Fix name/path to SA-06:20.
Bruce A. Mah [Thu, 7 Sep 2006 23:49:26 +0000 (23:49 +0000)]
Sigh.  Fix name/path to SA-06:20.

18 years agoMFC noted: lukemftpd from NetBSD snapshot.
Bruce A. Mah [Thu, 7 Sep 2006 23:41:11 +0000 (23:41 +0000)]
MFC noted:  lukemftpd from NetBSD snapshot.

18 years agos/USBDEVNAME/device_get_nameunit/g
Maksim Yevmenkin [Thu, 7 Sep 2006 23:38:09 +0000 (23:38 +0000)]
s/USBDEVNAME/device_get_nameunit/g
s/USBBASEDEVICE/device_t/g

18 years agoCreate a 'ready' handler for each personality. The purpose of this handler
Matt Jacob [Thu, 7 Sep 2006 23:08:21 +0000 (23:08 +0000)]
Create a 'ready' handler for each personality. The purpose of this handler
is to able to be called after *all* attach and enable events are done.

We establish a SYSINIT hook to call this handler. The current usage for it
is to add scsi target resources *after* all enables are done. There seems
to be some dependencies between different halves of a dual-port with respect
to target mode.

Put in more meaningful event messages for some events- in particular
QUEUE FULL events so we can see what the queue depth was when the
IOC sent us this message.

MFC after: 1 week

18 years agoMerge from DRM upstream:
Eric Anholt [Thu, 7 Sep 2006 23:04:47 +0000 (23:04 +0000)]
Merge from DRM upstream:
- Add support for Intel 965 Express chipsets.
- Add support for R200 vertex programs, along with minor bugfixes.
- Add support for vblank synchronization to pipe B of Intel hardware
  (laptop screens).

18 years agoAdd bthidd(8) rc(8) script
Maksim Yevmenkin [Thu, 7 Sep 2006 22:25:08 +0000 (22:25 +0000)]
Add bthidd(8) rc(8) script

MFC after: 1 month

18 years agoMFp4: first cut at getting I2C transfers working (generically). I'm
Warner Losh [Thu, 7 Sep 2006 21:53:28 +0000 (21:53 +0000)]
MFp4: first cut at getting I2C transfers working (generically).  I'm
unsure if this driver correctly implements all the start/stop junk
right (but it did or didn't before I made this commit).

18 years agoMFp4: berndt pointed me at an errata that shows that the stat register
Warner Losh [Thu, 7 Sep 2006 21:50:01 +0000 (21:50 +0000)]
MFp4: berndt pointed me at an errata that shows that the stat register
offsets were originally documented incorrectly.  This fixes that.  It
shouldn't affect anything other than error stat reporting.

18 years agoUpdate bthidd(8) code and hook it up to the build.
Maksim Yevmenkin [Thu, 7 Sep 2006 21:47:49 +0000 (21:47 +0000)]
Update bthidd(8) code and hook it up to the build.
bthidd(8) now was integrated with vkbd(4) and supports
multiple keyboards via vkbd(4)/kbdmux(4).

The code was tested with Apple Bluetooth keyboard and
SE k700i cell phone (remote control feature).

MFC after: 1 month

18 years agoWhile experimenting with cross-building by specifying just TARGET,
Ruslan Ermilov [Thu, 7 Sep 2006 20:27:56 +0000 (20:27 +0000)]
While experimenting with cross-building by specifying just TARGET,
I found one bug.  Pass our idea of TARGET_ARCH and TARGET down to
XMAKE, the cross-tools make.  Previously it worked because usually
TARGET_ARCH was specified on the initial make's command line.
This should also allow us to simplify the "universe" target, which
I'm currently testing.

18 years agoFixed the -# option to work as documented in a manpage.
Ruslan Ermilov [Thu, 7 Sep 2006 19:48:52 +0000 (19:48 +0000)]
Fixed the -# option to work as documented in a manpage.

18 years agoAdded timestamping to buildworld, similar to buildkernel.
Ruslan Ermilov [Thu, 7 Sep 2006 19:36:16 +0000 (19:36 +0000)]
Added timestamping to buildworld, similar to buildkernel.

18 years agoChange the class from uint8_t to int8_t so people can filter on >0
Doug Ambrisko [Thu, 7 Sep 2006 19:32:05 +0000 (19:32 +0000)]
Change the class from uint8_t to int8_t so people can filter on >0
events.

18 years agoNew release notes: SA-06:19.openssl, SA-06:20.bind.
Bruce A. Mah [Thu, 7 Sep 2006 18:58:19 +0000 (18:58 +0000)]
New release notes:  SA-06:19.openssl, SA-06:20.bind.

Modified release notes:  BIND 9.3.2-P1.

MFCs noted:  GCC 3.4.6.

18 years agoAll multicast listeners on a port should get one copy of the packet. This
John Hay [Thu, 7 Sep 2006 18:44:54 +0000 (18:44 +0000)]
All multicast listeners on a port should get one copy of the packet. This
was broken during the locking changes.

18 years agoChange the event log dump on initial boot to use get_event versus
Doug Ambrisko [Thu, 7 Sep 2006 18:40:49 +0000 (18:40 +0000)]
Change the event log dump on initial boot to use get_event versus
AEN.  This makes the boot messages cleaner.  I now know how this
structure works so I can implement it versus guessing.  Remove the
not ready type code since it is ready now.

I added the time stamp/locale/class so people can parse messages better.
Create a sysctl so that we can set the locale/class level.

18 years agoPrepare for upcoming bthidd(8) update. Install vkbd(4) header into dev/vkbd.
Maksim Yevmenkin [Thu, 7 Sep 2006 18:24:24 +0000 (18:24 +0000)]
Prepare for upcoming bthidd(8) update. Install vkbd(4) header into dev/vkbd.

MFC after: 1 month

18 years agoChase revision 1.10 in usage string and a comment.
Ceri Davies [Thu, 7 Sep 2006 17:10:26 +0000 (17:10 +0000)]
Chase revision 1.10 in usage string and a comment.

18 years ago- Rewrite logic to guess TARGET_ARCH/TARGET to be more readable,
Ruslan Ermilov [Thu, 7 Sep 2006 16:41:30 +0000 (16:41 +0000)]
- Rewrite logic to guess TARGET_ARCH/TARGET to be more readable,
  but without changes in behavior.

- Add sanity checking for TARGET_ARCH/TARGET.

18 years agoUse a single constant to define the sizes of the physmap[], phys_avail[],
John Baldwin [Thu, 7 Sep 2006 15:03:02 +0000 (15:03 +0000)]
Use a single constant to define the sizes of the physmap[], phys_avail[],
and dump_avail[] arrays so they are in sync (previously it was possible
to store more entries in the physmap[] then we could store in phys_avail[],
which was pointless).  While I'm here, bump up the length of these tables
to hold 30 entries on amd64 and 16 on i386.  This allows machines with
fairly fragmented memory maps to boot ok (at least one machine would
not boot FreeBSD/i386 but would boot FreeBSD/amd64 because amd64 allowed
for more fragments).

MFC after: 3 days

18 years agoBack when we had T/TCP support, we used to apply different
Ruslan Ermilov [Thu, 7 Sep 2006 13:06:00 +0000 (13:06 +0000)]
Back when we had T/TCP support, we used to apply different
timeouts for TCP and T/TCP connections in the TIME_WAIT
state, and we had two separate timed wait queues for them.
Now that is has gone, the timeout is always 2*MSL again,
and there is no reason to keep two queues (the first was
unused anyway!).

Also, reimplement the remaining queue using a TAILQ (it
was technically impossible before, with two queues).

18 years agoSecond step of TSO (TCP segmentation offload) support in our network stack.
Andre Oppermann [Thu, 7 Sep 2006 12:53:01 +0000 (12:53 +0000)]
Second step of TSO (TCP segmentation offload) support in our network stack.

TSO is only used if we are in a pure bulk sending state.  The presence of
TCP-MD5, SACK retransmits, SACK advertizements, IPSEC and IP options prevent
using TSO.  With TSO the TCP header is the same (except for the sequence number)
for all generated packets.  This makes it impossible to transmit any options
which vary per generated segment or packet.

The length of TSO bursts is limited to TCP_MAXWIN.

The sysctl net.inet.tcp.tso globally controls the use of TSO and is enabled.

TSO enabled sends originating from tcp_output() have the CSUM_TCP and CSUM_TSO
flags set, m_pkthdr.csum_data filled with the header pseudo-checksum and
m_pkthdr.tso_segsz set to the segment size (net payload size, not counting
IP+TCP headers or TCP options).

IPv6 currently lacks a pseudo-header checksum function and thus doesn't support
TSO yet.

Tested by: Jack Vogel <jfvogel-at-gmail.com>
Sponsored by: TCP/IP Optimization Fundraise 2005

18 years agoRemove a microoptimization for i386 that was a micropessimization for amd64.
Ruslan Ermilov [Thu, 7 Sep 2006 09:49:08 +0000 (09:49 +0000)]
Remove a microoptimization for i386 that was a micropessimization for amd64.

18 years agoFix the markup.
Ruslan Ermilov [Thu, 7 Sep 2006 08:31:59 +0000 (08:31 +0000)]
Fix the markup.

Prompted by: imp

18 years agomake "make TARGET=foo" work correctly. Before, it would fail to set
Warner Losh [Thu, 7 Sep 2006 07:37:16 +0000 (07:37 +0000)]
make "make TARGET=foo" work correctly.  Before, it would fail to set
TARGET_ARCH correctly.  Now it does, even for pc98.  We should suggest
TARGET=foo in preference to TARGET_ARCH because the former is
unambiguous and the latter isn't, so update the docs.

This means that a long standing gripe I've had with this comes to a
close.  I can build pc98 w/o specify both things.  make TARGET=arm
works (rather than trying to build a arm:amd64 image and dying badly
in the attempt).

If you specify only TARGET_ARCH, then you get the old behavior.

# we can likely simplify the UNIVERSE target now to use this, but I'm not
# up for breaking that tonight :-).

# We should consider adding some kind of sanity check for TARGET_ARCH
# and TARGET.

18 years agomore usb fallout changes
Matt Jacob [Thu, 7 Sep 2006 06:18:34 +0000 (06:18 +0000)]
more usb fallout changes

18 years agoStatic -> static.
Matt Jacob [Thu, 7 Sep 2006 06:00:03 +0000 (06:00 +0000)]
Static -> static.

18 years agoCatch up with USB changes, device_ptr_t was removed, we need device_t here.
Tom Rhodes [Thu, 7 Sep 2006 05:34:04 +0000 (05:34 +0000)]
Catch up with USB changes, device_ptr_t was removed, we need device_t here.

18 years agoCatch up to USB changes.
Scott Long [Thu, 7 Sep 2006 05:21:52 +0000 (05:21 +0000)]
Catch up to USB changes.

18 years agoRemove old debugging code from the interrupt handler.
Scott Long [Thu, 7 Sep 2006 05:05:41 +0000 (05:05 +0000)]
Remove old debugging code from the interrupt handler.

18 years agoMore removing compatibility macros.
Warner Losh [Thu, 7 Sep 2006 00:06:42 +0000 (00:06 +0000)]
More removing compatibility macros.

md5 still the same.

"Dave, stop.  I feel my mind slipping away." -- hal

18 years agoOnly include FreeBSD defines. This file is slated to go to the happy
Warner Losh [Wed, 6 Sep 2006 23:49:36 +0000 (23:49 +0000)]
Only include FreeBSD defines.  This file is slated to go to the happy
hunting ground in the sky.

Also, remove the #defines OBE.

18 years agos/Static/static/g
Warner Losh [Wed, 6 Sep 2006 23:44:25 +0000 (23:44 +0000)]
s/Static/static/g
s/device_ptr_t/device_t/g

No md5 changes in the .o's

# Note to the md5 tracking club: $FreeBSD$ changes md5 after every commit
# so you need to checkout -kk to get $FreeBSD$ instead of the actual value
# of the keyword.

18 years agoMore antideclarification.
Warner Losh [Wed, 6 Sep 2006 23:29:53 +0000 (23:29 +0000)]
More antideclarification.

md5 still the same after all these years (and after -g was removed).

18 years agoReserve a precious 16bit gap in the mbuf pkthdr struct for ethernet 802.1pq
Andre Oppermann [Wed, 6 Sep 2006 22:33:49 +0000 (22:33 +0000)]
Reserve a precious 16bit gap in the mbuf pkthdr struct for ethernet 802.1pq
vlan tags.

18 years agoMake TSO (TCP segmentation offload) capabilities visible and accessible with
Andre Oppermann [Wed, 6 Sep 2006 22:07:14 +0000 (22:07 +0000)]
Make TSO (TCP segmentation offload) capabilities visible and accessible with
'ifconfig em0 tso' and 'ifconfig em0 -tso'.  TSO for IPv4 and IPv6 is always
enabled or disabled together.  The driver may enable only one if it doesn't
support both.

Document 'tso' and '-tso' in the ifconfig(8) man pages.

Sponsored by: TCP/IP Optimization Fundraise 2005

18 years agoUnbreak in the case when device apic is compiled into non-SMP kernel.
Maxim Sobolev [Wed, 6 Sep 2006 22:05:34 +0000 (22:05 +0000)]
Unbreak in the case when device apic is compiled into non-SMP kernel.

Reported by: jhay
MFC after: 2 weeks

18 years agoUse sysctl_handle_long() instead of duplicating it's logic for
John Baldwin [Wed, 6 Sep 2006 21:59:36 +0000 (21:59 +0000)]
Use sysctl_handle_long() instead of duplicating it's logic for
kern.ipc.maxsockbuf so that this sysctl works for 32-bit binaries running
on amd64 via compat/freebsd32.

MFC after: 3 days

18 years agoRemove the portibility layer goo.
Warner Losh [Wed, 6 Sep 2006 21:53:28 +0000 (21:53 +0000)]
Remove the portibility layer goo.
Produces the same .o as before.

md5 by: /sbin/md5

18 years agoFirst step of TSO (TCP segmentation offload) support in our network stack.
Andre Oppermann [Wed, 6 Sep 2006 21:51:59 +0000 (21:51 +0000)]
First step of TSO (TCP segmentation offload) support in our network stack.

 o add IFCAP_TSO[46] for drivers to announce this capability for IPv4 and IPv6
 o add CSUM_TSO flag to mbuf pkthdr csum_flags field
 o add tso_segsz field to mbuf pkthdr
 o enhance ip_output() packet length check to allow for large TSO packets
 o extend tcp_maxmtu[46]() with a flag pointer to pass interface capabilities
 o adjust all callers of tcp_maxmtu[46]() accordingly

Discussed on: -current, -net
Sponsored by: TCP/IP Optimization Fundraise 2005

18 years agoMostly re-write man page to reflect current state of this interface.
Warner Losh [Wed, 6 Sep 2006 21:43:55 +0000 (21:43 +0000)]
Mostly re-write man page to reflect current state of this interface.
Warn of races.

# Doc police: Please review

18 years agoThis commit was generated by cvs2svn to compensate for changes in r162079,
Doug Barton [Wed, 6 Sep 2006 21:27:11 +0000 (21:27 +0000)]
This commit was generated by cvs2svn to compensate for changes in r162079,
which included commits to RCS files with non-trunk default branches.

18 years agoVendor import of BIND 9.3.2-P1, which addresses the following security
Doug Barton [Wed, 6 Sep 2006 21:27:11 +0000 (21:27 +0000)]
Vendor import of BIND 9.3.2-P1, which addresses the following security
vulnerabilities:

http://www.niscc.gov.uk/niscc/docs/re-20060905-00590.pdf?lang=en
2066.  [security]      Handle SIG queries gracefully. [RT #16300]

http://www.kb.cert.org/vuls/id/697164
1941.  [bug]           ncache_adderesult() should set eresult even if no
                       rdataset is passed to it. [RT #15642]

All users of BIND 9 are encouraged to upgrade to this version.

18 years agojhb points out that these mallocs don't need to be checked because
Warner Losh [Wed, 6 Sep 2006 20:22:43 +0000 (20:22 +0000)]
jhb points out that these mallocs don't need to be checked because
of M_WAITOK.

18 years agoWhile convenient, avoid using alloca() for reasons specified in
Ruslan Ermilov [Wed, 6 Sep 2006 20:15:43 +0000 (20:15 +0000)]
While convenient, avoid using alloca() for reasons specified in
the BUGS section of the alloca(3) manpage.  In particular, when
the number of TCP sockets is several tens of thousand, trying to
"sysctl -a" would SIGSEGV on the net.inet.tcp.pcblist entry (it
would exceed the stacksize ulimit, in an undetectable manner).

Reported by: Igor Sysoev

18 years agoMFp4: check the return value of malloc and report an error when invalid.
Warner Losh [Wed, 6 Sep 2006 19:46:02 +0000 (19:46 +0000)]
MFp4: check the return value of malloc and report an error when invalid.

18 years agoCheck inp_flags instead of inp_vflag for INP_ONESBCAST flag.
Andre Oppermann [Wed, 6 Sep 2006 19:04:36 +0000 (19:04 +0000)]
Check inp_flags instead of inp_vflag for INP_ONESBCAST flag.

PR: kern/99558
Tested by: Andrey V. Elsukov <bu7cher-at-yandex.ru>
Sponsored by: TCP/IP Optimization Fundraise 2005
MFC after: 3 days

18 years agoImprove description of if_capabilities, if_capenable and ifi_hwassist.
Andre Oppermann [Wed, 6 Sep 2006 18:06:04 +0000 (18:06 +0000)]
Improve description of if_capabilities, if_capenable and ifi_hwassist.

Sponsored by: TCP/IP Optimization Fundraise 2005

18 years agoFix stateful filtering of loopback IPv6 traffic to an address not configured
Max Laier [Wed, 6 Sep 2006 17:19:45 +0000 (17:19 +0000)]
Fix stateful filtering of loopback IPv6 traffic to an address not configured
on lo0.  While here fix a comment.

PR: kern/102647
Reported by: Frank Steinborn
Submitted by: suz (earlier version)
MFC after: 3 days

18 years agoFix the socket option IP_ONESBCAST by giving it its own case in ip_output()
Andre Oppermann [Wed, 6 Sep 2006 17:12:10 +0000 (17:12 +0000)]
Fix the socket option IP_ONESBCAST by giving it its own case in ip_output()
and skip over the normal IP processing.

Add a supporting function ifa_ifwithbroadaddr() to verify and validate the
supplied subnet broadcast address.

PR: kern/99558
Tested by: Andrey V. Elsukov <bu7cher-at-yandex.ru>
Sponsored by: TCP/IP Optimization Fundraise 2005
MFC after: 3 days

18 years agoRemove leading zeros
Brian Somers [Wed, 6 Sep 2006 15:58:59 +0000 (15:58 +0000)]
Remove leading zeros

Suggested by: mdoc police (ru)

18 years agoRemove call to fdfree() for the AIO daemons to prevent kernel panics
Mark Peek [Wed, 6 Sep 2006 15:11:20 +0000 (15:11 +0000)]
Remove call to fdfree() for the AIO daemons to prevent kernel panics
with linprocfs. This call is not needed since file descriptor sharing
was removed in v1.125.

Reviewed by: alc, davidxu, ambrisko
MFC after: 3 days

18 years agoRefine previous revision to allow acpi_wakecode.h to be safely built
Ruslan Ermilov [Wed, 6 Sep 2006 14:23:40 +0000 (14:23 +0000)]
Refine previous revision to allow acpi_wakecode.h to be safely built
from both the acpi module build directory and a kernel build directory.
The latter didn't work when one attempted to build a kernel which had
"device acpi" with the "make kernel-toolchain buildkernel" command
because a cross-compiler couldn't find anything in the standard system
include path (it's empty in the kernel-toolchain case).

Fix this by passing a better root path to kernel headers (src/sys)
which works for both cases, kernel and module (-I@ only worked for
module).

Also, while here, pass -nostdinc (and a different spelling for icc) --
it's a feature that the kernel source tree is self-contained, and this
change enforces this.

Reported by: glebius

18 years agoo Backout rev. 1.125 of in_pcb.c. It appeared to behave extremely
Gleb Smirnoff [Wed, 6 Sep 2006 13:56:35 +0000 (13:56 +0000)]
o Backout rev. 1.125 of in_pcb.c. It appeared to behave extremely
  bad under high load. For example with 40k sockets and 25k tcptw
  entries, connect() syscall can run for seconds. Debugging showed
  that it iterates the cycle millions times and purges thousands of
  tcptw entries at a time.
  Besides practical unusability this change is architecturally
  wrong. First, in_pcblookup_local() is used in connect() and bind()
  syscalls. No stale entries purging shouldn't be done here. Second,
  it is a layering violation.
o Return back the tcptw purging cycle to tcp_timer_2msl_tw(),
  that was removed in rev. 1.78 by rwatson. The commit log of this
  revision tells nothing about the reason cycle was removed. Now
  we need this cycle, since major cleaner of stale tcptw structures
  is removed.
o Disable probably necessary, but now unused
  tcp_twrecycleable() function.

Reviewed by: ru

18 years agoRemove __DATE__ so that compiling the same source produces the same binary
Brian Somers [Wed, 6 Sep 2006 06:33:39 +0000 (06:33 +0000)]
Remove __DATE__ so that compiling the same source produces the same binary
(for non-static binaries at least).

18 years agoIf the peer REJects our MRU request and that request is for a value
Brian Somers [Wed, 6 Sep 2006 06:23:55 +0000 (06:23 +0000)]
If the peer REJects our MRU request and that request is for a value
less than the current MTU, set our mtu to the value requested.

18 years agoReplace internal usage of struct umtx with umutex which can supports
David Xu [Wed, 6 Sep 2006 04:04:10 +0000 (04:04 +0000)]
Replace internal usage of struct umtx with umutex which can supports
real-time if we want, no functionality is changed.

18 years agoThe poison pill of death: adding a target mode reply handler and target
Matt Jacob [Tue, 5 Sep 2006 23:53:07 +0000 (23:53 +0000)]
The poison pill of death: adding a target mode reply handler and target
resources to a non-FC card killed us dead. Sorry for the breakage since
last July 12.

18 years agoFix problems with destroy and forcible destroy functionality:
Pawel Jakub Dawidek [Tue, 5 Sep 2006 21:56:00 +0000 (21:56 +0000)]
Fix problems with destroy and forcible destroy functionality:
- hold/release device in start/done routines, this will probably slow
  down things a bit, but previous code was racy;
- only release device if g_gate_destroy() failed - if it succeeded device
  is dead and there is nothing to release;
- various other changes which makes forcible destruction reliable.

MFC after: 3 days

18 years agoCoverity: initialize some variables before potential use.
Matt Jacob [Tue, 5 Sep 2006 20:28:28 +0000 (20:28 +0000)]
Coverity: initialize some variables before potential use.

18 years ago- Move descriptions of BOOT_COMCONSOLE_PORT, BOOT_COMCONSOLE_SPEED,
Ruslan Ermilov [Tue, 5 Sep 2006 20:00:07 +0000 (20:00 +0000)]
- Move descriptions of BOOT_COMCONSOLE_PORT, BOOT_COMCONSOLE_SPEED,
  and LOADER_TFTP_SUPPORT options into the world section since boot
  blocks are built as part of the world.

- Document BOOT_PXELDR_ALWAYS_SERIAL and BOOT_PXELDR_PROBE_KEYBOARD
  options of pxeboot(8).

MFC after: 3 days

18 years ago- Include <sys/reboot.h> to get the RB_* defines.
Ruslan Ermilov [Tue, 5 Sep 2006 19:28:03 +0000 (19:28 +0000)]
- Include <sys/reboot.h> to get the RB_* defines.

- Make the PROBE_KEYBOARD option better resemble the -P option in
  boot2, i.e., if keyboard isn't present then boot with both
  RB_SERIAL and RB_MULTIPLE set.

Reviewed by: jhb

18 years agoUse net.inet6.ip6.redirect / ip6_sendredirects as part of the decision
John Hay [Tue, 5 Sep 2006 19:20:42 +0000 (19:20 +0000)]
Use net.inet6.ip6.redirect / ip6_sendredirects as part of the decision
to generate icmp6 redirects. Now it is possible to switch redirects off.

MFC after: 1 week

18 years agoThe FreeBSD by default "disables" hyper-threading cores, by not scheduling
Maxim Sobolev [Tue, 5 Sep 2006 17:15:24 +0000 (17:15 +0000)]
The FreeBSD by default "disables" hyper-threading cores, by not scheduling
any threads to them. However, it still counts those cores as "active but
permanently idle" when calculating system-wide CPUs statistics. It is
incorrect, since it skews statistics quite a bit and creates real problems
for certain types of applications (monitoring applications for example),
by making them believe that the system does have enough idle CPU resources,
while in fact it does not.

Correct the problem by not calling performance counting routines on "disabled"
cores. The cleaner solution would be to just disable APIC timer interrupts on
those cores completely, but ENOTIME here and it is not clear if the
additional complexity really worth minor performance gain.

Reviewed by: ssouhlal
Sponsored by: Sippy Software, Inc.
MFC after: 2 weeks

18 years agoInclude agp_i810.c in amd64 AGP builds to get support for the Intel 915 Express
Eric Anholt [Tue, 5 Sep 2006 16:55:13 +0000 (16:55 +0000)]
Include agp_i810.c in amd64 AGP builds to get support for the Intel 915 Express
chipsets.

PR: kern/93676
Submitted by: Jan Blaha <Jan.Blaha@unet.cz>
MFC after: 1 week

18 years agoalloca() cannot check if the allocation is valid; mention the consequences.
Ruslan Ermilov [Tue, 5 Sep 2006 16:30:11 +0000 (16:30 +0000)]
alloca() cannot check if the allocation is valid; mention the consequences.

Obtained from: OpenBSD

18 years agoGC dead code. If we want to stay polite to the foreign compilers,
Ruslan Ermilov [Tue, 5 Sep 2006 16:21:26 +0000 (16:21 +0000)]
GC dead code.  If we want to stay polite to the foreign compilers,
we can find another way to issue an #error, but using a preprocessed
assembler for that purpose and clobbering libc.a with an empty .o
just for the sake of #error reporting is way too much of a burden.

18 years agoNew release notes: libpcap 0.9.4, tcpdump 3.9.4.
Bruce A. Mah [Tue, 5 Sep 2006 14:55:31 +0000 (14:55 +0000)]
New release notes:  libpcap 0.9.4, tcpdump 3.9.4.

18 years agoSame as pthread_setschedparam, use sizeof(struct sched_param) instead.
David Xu [Tue, 5 Sep 2006 14:39:06 +0000 (14:39 +0000)]
Same as pthread_setschedparam, use sizeof(struct sched_param) instead.

18 years agoPass correct parameter size.
David Xu [Tue, 5 Sep 2006 14:37:22 +0000 (14:37 +0000)]
Pass correct parameter size.

18 years agoFinally fix rev. 1.256
Gleb Smirnoff [Tue, 5 Sep 2006 14:00:59 +0000 (14:00 +0000)]
Finally fix rev. 1.256

Pointy hat to: glebius