ĨV121420
220
301
143
171
145
149
317
135
195
389
451
148
160
166
164
197
172
148
293
152
207
392
292
176
314
146
151
398
484
165
188
514
127
541
523
222
359
171
174
117
148
320
246
147
250
490
242
211
203
307
160
240
126
446
223
279
193
220
115
203
222
502
129
432
161
227
145
354
362
211
223
252
305
252
252
200
151
209
693
140
219
152
136
167
119
125
155
136
458
271
578
286
228
411
136
370
255
183
219
148
126
143
132
126
154
204
232
213
219
418
134
262
167
211
207
163
205
225
145
135
396
148
287
240
156
197
119
117
132
209
117
112
385
112
153
297
314
192
181
191
251
191
135
333
135
116
118
370
273
105
166
151
149
173
256
198
112
113
177
196
272
225
196
107
169
272
119
124
116
204
182
551
120
135
116
119
148
213
239
106
105
848
163
887
223
139
1189
206
177
151
238
163
173
161
156
139
142
157
597
181
213
158
245
697
170
158
328
235
132
275
207
135
151
164
203
196
2430
1017
127
155
713

K 10
svn:author
V 3
des
K 8
svn:date
V 27
2003-10-23T08:27:16.000000Z
K 7
svn:log
V 208
Plug a memory leak in the PAM child process.  It is of no great consequence
as the process is short-lived, and the leak occurs very rarely and always
shortly before the process terminates.

MFC after:	3 days

END
K 10
svn:author
V 3
des
K 8
svn:date
V 27
2003-10-23T13:42:03.000000Z
K 7
svn:log
V 51
0x7FFFFFFFFF is >32 bits and needs an explicit LL.

END
K 10
svn:author
V 3
des
K 8
svn:date
V 27
2003-10-23T13:49:10.000000Z
K 7
svn:log
V 79
Clean up whitespace, remove "register" keyword, ANSIfy.
No functional changes.

END
K 10
svn:author
V 3
ume
K 8
svn:date
V 27
2003-10-23T13:50:01.000000Z
K 7
svn:log
V 53
preparation for RFC3493.  EAI_NODATA was deprecated.

END
K 10
svn:author
V 2
ru
K 8
svn:date
V 27
2003-10-23T13:53:19.000000Z
K 7
svn:log
V 58
Print the correct ICMP statistics for "no return routes".

END
K 10
svn:author
V 3
ume
K 8
svn:date
V 27
2003-10-23T13:55:36.000000Z
K 7
svn:log
V 224
EAI_ADDRFAMILY and EAI_NODATA was deprecated in RFC3493
(aka RFC2553bis).  Now, getaddrinfo(3) returns EAI_NONAME
instead of EAI_NODATA.  Our getaddrinfo(3) nor getnameinfo(3)
didn't use EAI_ADDRFAMILY.

Obtained from:	KAME

END
K 10
svn:author
V 3
ume
K 8
svn:date
V 27
2003-10-23T14:32:03.000000Z
K 7
svn:log
V 43
reorder functions to be in sync with KAME.

END
K 10
svn:author
V 3
hmp
K 8
svn:date
V 27
2003-10-23T14:32:23.000000Z
K 7
svn:log
V 102
Mdoc Janitor:

  * Add a forgotten .An macro for 'Jeroen Ruigrok van der Werven'.

Noticed by:	simon@

END
K 10
svn:author
V 3
imp
K 8
svn:date
V 27
2003-10-23T15:09:34.000000Z
K 7
svn:log
V 296
Merge from p4 (noticed these changes with DES' if_ethersubr.c changes caused
a minor conflict):
o Use ETHER_ADDR_LEN in preference to '6'.
o Remove two unnecessary (caddr_t) casts.  One of them causes problems in
  my tree where etherbroadcastaddr is const, and (caddr_t) casts the const
  away.

END
K 10
svn:author
V 7
trhodes
K 8
svn:date
V 27
2003-10-23T16:09:20.000000Z
K 7
svn:log
V 354
Fix several old bugs which got worse over time:

o WARNS should be WARNS?= (broke in rev 1.21).
o Includes should be sorted.
o Move "mntopt.h" out of the standard includes section.
o Rewrite usage() to match the manual page and make it < 80 characters.
o Remove extra .El call on line 187.  It is unused and causes mdoc(7) warnings.

Discussed with:	bde

END
K 10
svn:author
V 3
ume
K 8
svn:date
V 27
2003-10-23T16:11:46.000000Z
K 7
svn:log
V 56
oops, I forget to diable EAI_ADDRFAMILY and EAI_NODATA.

END
K 10
svn:author
V 6
brooks
K 8
svn:date
V 27
2003-10-23T16:29:20.000000Z
K 7
svn:log
V 65
Use IF_MAXUNIT instead of rolling our own identical TUN_MAXUNIT.

END
K 10
svn:author
V 3
imp
K 8
svn:date
V 27
2003-10-23T16:55:27.000000Z
K 7
svn:log
V 74
Const poison the crc calcuation routines so const data can be sent to
it.

END
K 10
svn:author
V 3
imp
K 8
svn:date
V 27
2003-10-23T16:57:38.000000Z
K 7
svn:log
V 72
Const poison crc routines (why these aren't centralized, I'm not sure).

END
K 10
svn:author
V 4
bmah
K 8
svn:date
V 27
2003-10-23T17:16:03.000000Z
K 7
svn:log
V 103
Bump version numbers in release documentation for upcoming
4.9-RELEASE.

Approved by:	 re (implicitly)

END
K 10
svn:author
V 3
ume
K 8
svn:date
V 27
2003-10-23T17:36:29.000000Z
K 7
svn:log
V 80
- style
- index() -> strchr()
- stop use of MAXHOSTNAMELEN

Obtained from:	KAME

END
K 10
svn:author
V 3
imp
K 8
svn:date
V 27
2003-10-23T17:47:55.000000Z
K 7
svn:log
V 56
Remove unnecessary (caddr_t) casts of if_broadcastaddr.

END
K 10
svn:author
V 3
jhb
K 8
svn:date
V 27
2003-10-23T17:48:30.000000Z
K 7
svn:log
V 200
Use a switch statement on the devid instead of if-else for determing which
code to use to see if the onboard video has been disabled or not.

Submitted by:	Keith Whitwell <keith@tungstengraphics.com>

END
K 10
svn:author
V 3
imp
K 8
svn:date
V 27
2003-10-23T17:48:53.000000Z
K 7
svn:log
V 60
Remove unnecessary casts to (caddr_t) for if_broadcastaddr.

END
K 10
svn:author
V 3
ume
K 8
svn:date
V 27
2003-10-23T17:54:17.000000Z
K 7
svn:log
V 114
workaround to have backward compatibility for EAI_NODATA.
it will be removed on 23 Apr 2004.

Submitted by:	terry

END
K 10
svn:author
V 3
jhb
K 8
svn:date
V 27
2003-10-23T18:08:56.000000Z
K 7
svn:log
V 299
Add simple support for AGP 3.0 including enabling 8x mode.  The simple
part of the support is that it still assumes one master and one target
where as AGP 3.0 actually supports multiple devices on the bus.

Submitted by:	Keith Whitwell <keith@tungstengraphics.com>
Sponsored by:	The Weather Channel

END
K 10
svn:author
V 7
wollman
K 8
svn:date
V 27
2003-10-23T18:17:36.000000Z
K 7
svn:log
V 195
Add appropriate const poisoning to the assert_*locked() family so that I can
call ASSERT_VOP_LOCKED(vp, __func__) without a diagnostic.

Inspired by:	the evil and rude OpenAFS cache manager code

END
K 10
svn:author
V 3
des
K 8
svn:date
V 27
2003-10-23T20:57:22.000000Z
K 7
svn:log
V 84
MFC: fix, improve & extend NFS support.

Submitted by:	hmp
Approved by:	re (murray)

END
K 10
svn:author
V 3
jhb
K 8
svn:date
V 27
2003-10-23T21:14:08.000000Z
K 7
svn:log
V 221
Move the P_COWINPROGRESS flag from being a per-process p_flag to being a
per-thread td_pflag which doesn't require any locks to read or write as it
is only read or written by curthread on itself.

Glanced at by:	mckusick

END
K 10
svn:author
V 3
jhb
K 8
svn:date
V 27
2003-10-23T21:20:34.000000Z
K 7
svn:log
V 54
Writes to p_flag in __setugid() no longer need Giant.

END
K 10
svn:author
V 3
sam
K 8
svn:date
V 27
2003-10-23T21:41:00.000000Z
K 7
svn:log
V 59
check return result from rtalloc1 before invoking RTUNLOCK

END
K 10
svn:author
V 3
sam
K 8
svn:date
V 27
2003-10-24T00:09:18.000000Z
K 7
svn:log
V 305
o restructure initialization code so data structures are setup
  when loaded as a module
o cleanup data structures on module unload when no application has
  been started (i.e. kldload, kldunload w/o mrtd)
o remove extraneous unlocks immediately prior to destroying them

Supported by:	FreeBSD Foundation

END
K 10
svn:author
V 6
anholt
K 8
svn:date
V 27
2003-10-24T01:48:17.000000Z
K 7
svn:log
V 388
Update to latest from DRI CVS.  Primary new feature is mostly-complete smpng
locking, and the apparently unnecessary locking for -stable has been removed.
This may fix issues with missed interrupts since April, which manifested
themselves as slowdowns or hangs in radeon, in particular. Many cleanups also
took place.  In the shared code, there are improvements to r128 driver
stability.

END
K 10
svn:author
V 6
marcel
K 8
svn:date
V 27
2003-10-24T02:05:53.000000Z
K 7
svn:log
V 70
Remove two unused fields in the operand structure (o_read & o_write).

END
K 10
svn:author
V 6
marcel
K 8
svn:date
V 27
2003-10-24T02:33:26.000000Z
K 7
svn:log
V 93
Don't use fuword() or suword() unconditionally. They explicitly
disallow reading or writing.

END
K 10
svn:author
V 5
peter
K 8
svn:date
V 27
2003-10-24T02:50:39.000000Z
K 7
svn:log
V 419
Add __va_copy and make it always visible, in spite of the __ISO_C_VISIBLE
setting.  Make va_copy be an alias if __ISO_C_VISIBLE >= 1999.

Why?  more than a few ports have an autoconf that looks for __va_copy
because it is available on glibc.  It is critical that we use it if
at all possible on amd64.  It generally isn't a problem for i386 and its
ilk because autoconf driven code tends to fall back to an assignment.

END
K 10
svn:author
V 3
ume
K 8
svn:date
V 27
2003-10-24T03:49:38.000000Z
K 7
svn:log
V 35
oops, EAI_NONAME is not EAINONAME.

END
K 10
svn:author
V 4
arun
K 8
svn:date
V 27
2003-10-24T04:56:58.000000Z
K 7
svn:log
V 447
Use a TR of size 1 << IA64_ID_PAGE_SHIFT instead of 16M to avoid
overlapping TR/TC entries (which results in a machine check). Note
that we don't look at the size of the memory descriptor, because
it doesn't guarantee non-overlap.

With this change, a UP kernel could boot on a Intel Tiger4 machine
with the following options:

options         LOG2_ID_PAGE_SIZE=26		# 64M
options         LOG2_PAGE_SIZE=14               # 16K

Approved by: marcel

END
K 10
svn:author
V 5
silby
K 8
svn:date
V 27
2003-10-24T05:44:14.000000Z
K 7
svn:log
V 428
Reduce the number of tcp time_wait structs to maxsockets / 5; this ensures
that at most 20% of sockets can be in time_wait at one time, ensuring
that time_wait sockets do not starve real connections from inpcb
structures.

No implementation change is needed, jlemon already implemented a nice
LRU-ish algorithm for tcp_tw structure recycling.

This should reduce the need for sysadmins to lower the default msl on
busy servers.

END
K 10
svn:author
V 6
marcel
K 8
svn:date
V 27
2003-10-24T06:42:03.000000Z
K 7
svn:log
V 126
Implement db_disasm() by using the new disassembler. Temporarily
unimplement db_write_breakpoint() and db_clear_breakpoint().

END
K 10
svn:author
V 3
alc
K 8
svn:date
V 27
2003-10-24T06:43:04.000000Z
K 7
svn:log
V 266
 - Push down Giant from vm_pageout() to vm_pageout_scan(), freeing
   vm_pageout_page_stats() from Giant.
 - Modify vm_pager_put_pages() and vm_pager_page_unswapped() to expect the
   vm object to be locked on entry.  (All of the pager routines now expect
   this.)

END
K 10
svn:author
V 6
marcel
K 8
svn:date
V 27
2003-10-24T06:48:41.000000Z
K 7
svn:log
V 76
Remove unused file. db_disasm() has been implemented in db_interface.c
now.

END
K 10
svn:author
V 6
marcel
K 8
svn:date
V 27
2003-10-24T06:52:21.000000Z
K 7
svn:log
V 79
Remove ia64_pack_bundle() and ia64_unpack_bundle(). They are not
used anymore.

END
K 10
svn:author
V 3
ume
K 8
svn:date
V 27
2003-10-24T06:53:12.000000Z
K 7
svn:log
V 25
style.

Reported by:	bde

END
K 10
svn:author
V 6
marcel
K 8
svn:date
V 27
2003-10-24T06:53:43.000000Z
K 7
svn:log
V 53
Remove unused header. See also ia64/disasm/disasm.h.

END
K 10
svn:author
V 3
imp
K 8
svn:date
V 27
2003-10-24T07:20:13.000000Z
K 7
svn:log
V 227
Backswards is as backwards does: If we're MPSAFE, then we don't need
giant, which implies that we need to take out giant it we're NOT
MPSAFE.

# I can't believe the number of people that looked at this failed to
# detect this.

END
K 10
svn:author
V 5
harti
K 8
svn:date
V 27
2003-10-24T07:39:11.000000Z
K 7
svn:log
V 151
Netgraph part of the NgATM signalling AA layer. These nodes can
also be used as a general-purpose transport protocol above any
packet layer (IP, UDP).

END
K 10
svn:author
V 5
harti
K 8
svn:date
V 27
2003-10-24T07:42:08.000000Z
K 7
svn:log
V 53
Module build infrastructure for the NgATM SAA layer.

END
K 10
svn:author
V 8
simokawa
K 8
svn:date
V 27
2003-10-24T07:42:21.000000Z
K 7
svn:log
V 152
Don't check timeout just after booted.
Some transactions could be considered wrongly to be timeout
bacause interrupts are disabled during boot process.

END
K 10
svn:author
V 5
harti
K 8
svn:date
V 27
2003-10-24T10:01:36.000000Z
K 7
svn:log
V 395
Allow nghook to execute a program with the data socket connected to
stdin and stdout instead of relaying the data. Now it is possible
to say:

	nghook -e path: hook /usr/local/bin/foo arg1 arg2

and foo will have the hook to path: at file descriptors 0 and 1.

Add an option to specify control messages to be send to the node before
either executing the program or entering the data relay loop.

END
K 10
svn:author
V 5
harti
K 8
svn:date
V 27
2003-10-24T13:02:39.000000Z
K 7
svn:log
V 147
Mention the fact that our crontab is not fully POSIX.2 conform, because
it doesn't allow the dangerous variant of calling it without any
argument.

END
K 10
svn:author
V 8
simokawa
K 8
svn:date
V 27
2003-10-24T13:55:51.000000Z
K 7
svn:log
V 113
Fix for FW_ASYREQ.
- set send.pay_len correctly.
- copy response only if needed.
- remove unnecessary 'err = 0'.

END
K 10
svn:author
V 3
hmp
K 8
svn:date
V 27
2003-10-24T15:42:38.000000Z
K 7
svn:log
V 110
Mdoc Janitor:

  * Use uppercase for .Dt macro arguments.

  * Use 'manual' instead of 'man' for consistency.

END
K 10
svn:author
V 8
simokawa
K 8
svn:date
V 27
2003-10-24T15:44:10.000000Z
K 7
svn:log
V 209
Add dumb console driver and related bits.

dcons(4): very simple console and gdb port driver
dcons_crom(4): FireWire attachment
dconschat(8): User interface to dcons

Tested with: i386, i386-PAE, and sparc64.

END
K 10
svn:author
V 5
harti
K 8
svn:date
V 27
2003-10-24T16:44:27.000000Z
K 7
svn:log
V 66
Sysctl names should not contain dots. Convert them to underlines.

END
K 10
svn:author
V 3
ume
K 8
svn:date
V 27
2003-10-24T16:57:59.000000Z
K 7
svn:log
V 147
Since dp->dom_ifattach calls malloc() with M_WAITOK, we cannot
use mutex lock directly here.  Protect ifp->if_afdata instead.

Reported by:	grehan

END
K 10
svn:author
V 3
ume
K 8
svn:date
V 27
2003-10-24T18:00:17.000000Z
K 7
svn:log
V 34
show maxmtu.

Obtained from:	KAME

END
K 10
svn:author
V 3
ume
K 8
svn:date
V 27
2003-10-24T18:26:30.000000Z
K 7
svn:log
V 353
Switch Advanced Sockets API for IPv6 from RFC2292 to RFC3542
(aka RFC2292bis).  Though I believe this commit doesn't break
backward compatibility againt existing binaries, it breaks
backward compatibility of API.
Now, the applications which use Advanced Sockets API such as
telnet, ping6, mld6query and traceroute6 use RFC3542 API.

Obtained from:	KAME

END
K 10
svn:author
V 3
phk
K 8
svn:date
V 27
2003-10-24T18:36:49.000000Z
K 7
svn:log
V 130
When grabbing vnodes to service NFS requests, make sure to call
vn_start_write() early to avoid snapshot deadlocks.

By:	mckusick

END
K 10
svn:author
V 3
ume
K 8
svn:date
V 27
2003-10-24T18:43:24.000000Z
K 7
svn:log
V 186
oops, revert previous change to getaddrinfo.c.  This is not related
to RFC3493.  The previous change was related to RFC3484 (Default
Address Selection for IPv6), and it will come later.

END
K 10
svn:author
V 3
phk
K 8
svn:date
V 27
2003-10-24T18:46:23.000000Z
K 7
svn:log
V 100
Fix two old/new consumer confusions.

Submitted by:    Pawel Jakub Dawidek <nick@garage.freebsd.pl>

END
K 10
svn:author
V 3
phk
K 8
svn:date
V 27
2003-10-24T18:47:31.000000Z
K 7
svn:log
V 127
Close the right consumers if we run into trouble opening them all.

Submitted by:	Pawel Jakub Dawidek <nick@garage.freebsd.pl>

END
K 10
svn:author
V 3
ume
K 8
svn:date
V 27
2003-10-24T19:51:49.000000Z
K 7
svn:log
V 23
correct tab and order.

END
K 10
svn:author
V 3
ume
K 8
svn:date
V 27
2003-10-24T20:37:05.000000Z
K 7
svn:log
V 110
remove the ip6r0_addr and ip6r0_slmap members from ip6_rthdr0{}
according to rfc2292bis.

Obtained from:	KAME

END
K 10
svn:author
V 3
jhb
K 8
svn:date
V 27
2003-10-24T20:55:15.000000Z
K 7
svn:log
V 129
- Renumber the skerberos4 dist to fill in the gaps left when krb4 was axed.
- Update the DIST_CRYPTO_ALL value to match reality.

END
K 10
svn:author
V 3
jhb
K 8
svn:date
V 27
2003-10-24T21:01:31.000000Z
K 7
svn:log
V 409
- Fail to probe if acpi0 probed ok as this driver basically tries to probe
  the ACPI timer and we shouldn't do that if ACPI is already around to do
  that for us.
- Set a description and tweak the order of checks in the probe function
  to more closely match other PCI drivers.

This should probably be moved to sys/dev/piix/piix.c at some point and
turned on for all i386 kernels rather than just SMP ones.

END
K 10
svn:author
V 3
jhb
K 8
svn:date
V 27
2003-10-24T21:02:26.000000Z
K 7
svn:log
V 37
A few whitespace and comment tweaks.

END
K 10
svn:author
V 3
jhb
K 8
svn:date
V 27
2003-10-24T21:05:30.000000Z
K 7
svn:log
V 339
- Add a DDB command 'show intrcnt' to show the non-zero interrupt counts.
- Add a DDB function to dump the contents of an ithread and optionally
  details about each handler in that ithread.  This function can be used
  by MD code to implement DDB commands that display information about
  interrupt sources and their registered handlers.

END
K 10
svn:author
V 3
jmg
K 8
svn:date
V 27
2003-10-24T21:07:53.000000Z
K 7
svn:log
V 69
don't allow reading from files that haven't been open'd for reading.

END
K 10
svn:author
V 8
kensmith
K 8
svn:date
V 27
2003-10-24T21:20:26.000000Z
K 7
svn:log
V 129
	- fix description of what processes SIGCONT can be sent to

PR:		docs/58413
Reviewed by:	rwatson
Approved by:	blackend (mentor)

END
K 10
svn:author
V 3
phk
K 8
svn:date
V 27
2003-10-24T21:23:47.000000Z
K 7
svn:log
V 53
Use 'k' as suffix for Kilo

Pointed out by:	several.

END
K 10
svn:author
V 6
anholt
K 8
svn:date
V 27
2003-10-24T21:45:21.000000Z
K 7
svn:log
V 258
Don't try to use dev->dma_lock unless dma is initialized (dev->dma != NULL)
in bufs_info sysctl handler.  dev->dma and dev->dma_lock existence are
protected by DRM_LOCK().  Fixes panic on sysctl hw.dri when the device is
uninitialied (when you aren't in X).

END
K 10
svn:author
V 3
imp
K 8
svn:date
V 27
2003-10-24T22:02:29.000000Z
K 7
svn:log
V 269
Parse the ! lines that will soon be coming from the kernel.  These are
a generalized notification mechanism for subsystems wishing to report
events.

Revieded by: njl

# The kernel side seems like it might be causing panics for me, but should
# be forthcoming shortly.

END
K 10
svn:author
V 8
ambrisko
K 8
svn:date
V 27
2003-10-24T22:34:56.000000Z
K 7
svn:log
V 113
Add support for another multiple serial port card based on OX16PCI954 device
id 0x950a.

MFC:	after 4.9 release.

END
K 10
svn:author
V 3
imp
K 8
svn:date
V 27
2003-10-24T22:41:54.000000Z
K 7
svn:log
V 130
Convenience functions to generate notifications from the kernel.  The ACPI
code will start using these shortly.

Reviewed by: njl

END
K 10
svn:author
V 7
rwatson
K 8
svn:date
V 27
2003-10-25T03:50:44.000000Z
K 7
svn:log
V 155
Sort type declarations together.
Remove an excess carriage return.

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

END
K 10
svn:author
V 3
imp
K 8
svn:date
V 27
2003-10-25T04:05:33.000000Z
K 7
svn:log
V 212
Start to minimize diffs between vx and ep.  These latter is based on a
more advanced version of the chips supported by the former.  Matt Dodd
and I are working towards merging them, and this a step on that path.

END
K 10
svn:author
V 3
imp
K 8
svn:date
V 27
2003-10-25T04:09:49.000000Z
K 7
svn:log
V 159
Whole grab-bag of changes:
o Make the driver MPSAFE
o Some changes due to diff reduction effort with vx.
o Removed some obsolete junk.

Reviewed by: sam, modd

END
K 10
svn:author
V 3
njl
K 8
svn:date
V 27
2003-10-25T05:03:25.000000Z
K 7
svn:log
V 159
Add devctl(4) notify support to ACPI.  Various subsystems now notify
userland whenever events occur.  See the example in devd.conf below
to see how to use it.

END
K 10
svn:author
V 5
peter
K 8
svn:date
V 27
2003-10-25T05:14:38.000000Z
K 7
svn:log
V 105
GC workaround code for detecting pentium4's and disabling PSE and PG_G.
It's been ifdef'ed out for ages.

END
K 10
svn:author
V 3
alc
K 8
svn:date
V 27
2003-10-25T05:21:16.000000Z
K 7
svn:log
V 59
 - Call vnode_pager_input_old() with the vm object locked.

END
K 10
svn:author
V 3
ume
K 8
svn:date
V 27
2003-10-25T06:51:23.000000Z
K 7
svn:log
V 116
according to RFC3542 10.5, the 5th argment of inet6_opt_next()
is not size_t but socklen_t.

Reported by:	tinderbox

END
K 10
svn:author
V 3
des
K 8
svn:date
V 27
2003-10-25T09:32:18.000000Z
K 7
svn:log
V 600
 - fabsl.c should be named s_fabsl.c for consistency with libmsun's
   documented naming scheme (unfortunately the documentation isn't in the
   tree as far as I can tell); no repocopy is required as there is no
   history to preserve.

 - replace simple and almost-correct implementation with slightly hackish
   but definitely correct implementation (tested on i386, alpha, sparc64)
   which requires pulling in fpmath.h and the MD _fpmath.h from libc.

 - try not to make a mess of the Makefile in the process.

 - enterprising minds are encouraged to implement more C99 long double
   functions.

END
K 10
svn:author
V 3
ume
K 8
svn:date
V 27
2003-10-25T09:37:10.000000Z
K 7
svn:log
V 48
correct namespace pollution.

Submitted by:	bde

END
K 10
svn:author
V 3
ume
K 8
svn:date
V 27
2003-10-25T10:57:08.000000Z
K 7
svn:log
V 126
revert following unwanted changes:
  - __packed to __attribute__((__packed__)
  -  uintN_t back to u_intN_t

Reported by:	bde

END
K 10
svn:author
V 3
phk
K 8
svn:date
V 27
2003-10-25T12:56:51.000000Z
K 7
svn:log
V 60
Style changes.  Inching closer to convergence with OpenBSD.

END
K 10
svn:author
V 8
simokawa
K 8
svn:date
V 27
2003-10-25T12:59:22.000000Z
K 7
svn:log
V 39
dcons_crom.c needs dcons and firewire.

END
K 10
svn:author
V 3
des
K 8
svn:date
V 27
2003-10-25T13:45:11.000000Z
K 7
svn:log
V 75
Document fabsl(3).

Submitted by:	Stefan Farfeleder <stefan@fafoe.narf.at>

END
K 10
svn:author
V 8
simokawa
K 8
svn:date
V 27
2003-10-25T14:51:36.000000Z
K 7
svn:log
V 22
remove debug message.

END
K 10
svn:author
V 8
simokawa
K 8
svn:date
V 27
2003-10-25T14:53:53.000000Z
K 7
svn:log
V 28
update for conserver-8.0.4.

END
K 10
svn:author
V 8
simokawa
K 8
svn:date
V 27
2003-10-25T15:04:49.000000Z
K 7
svn:log
V 58
Move validity check of 'xfer->fc != NULL' to right place.

END
K 10
svn:author
V 8
simokawa
K 8
svn:date
V 27
2003-10-25T15:05:59.000000Z
K 7
svn:log
V 39
Fix fwmem_strategy() race in 4-stable.

END
K 10
svn:author
V 7
rwatson
K 8
svn:date
V 27
2003-10-25T15:28:20.000000Z
K 7
svn:log
V 361
Make MAC_EXTERNALIZE() and MAC_INTERNALIZE() simply take the object
type, rather than "object_label" as the first argument.  This reduces
complexity a little for the consumer, and also makes it easier for
use to rename the underlying entry points in struct mac_policy_obj.

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

END
K 10
svn:author
V 7
rwatson
K 8
svn:date
V 27
2003-10-25T16:10:41.000000Z
K 7
svn:log
V 174
Allow MAC policies to block/revoke kern_alq write access to a file.

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

END
K 10
svn:author
V 7
rwatson
K 8
svn:date
V 27
2003-10-25T16:14:09.000000Z
K 7
svn:log
V 481
When generate a core dump, use advisory locking in an advisory way:
if we do acquire an advisory lock, great!  We'll release it later.
However, if we fail to acquire a lock, we perform the coredump
anyway.  This problem became particularly visible with NFS after
the introduction of rpc.lockd: if the lock manager isn't running,
then locking calls will fail, aborting the core dump (resulting in
a zero-byte dump file).

Reported by:	Yogeshwar Shenoy <ynshenoy@alumni.cs.ucsb.edu>

END
K 10
svn:author
V 7
rwatson
K 8
svn:date
V 27
2003-10-25T16:43:50.000000Z
K 7
svn:log
V 189
Check (locked) before performing an advisory unlock following a failure
of vn_start_write().  Otherwise, we may inconsistently attempt to release
the advisory lock.

Pointed out by:	teggej

END
K 10
svn:author
V 3
alc
K 8
svn:date
V 27
2003-10-25T18:33:04.000000Z
K 7
svn:log
V 135
 - Align a comment within struct vm_page.
 - Annotate the vm_page's valid field as synchronized by the containing
   vm object's lock.

END
K 10
svn:author
V 5
peter
K 8
svn:date
V 27
2003-10-25T18:51:41.000000Z
K 7
svn:log
V 316
For the SMP case, flush the TLB at the beginning of the page zero/copy
routines.  Otherwise we run into trouble with speculative tlb preloads
on SMP systems.  This effectively defeats Jeff's revision 1.438
optimization (for his pentium4-M laptop) in the SMP case.  It breaks
other systems, particularly athlon-MP's.

END
K 10
svn:author
V 3
des
K 8
svn:date
V 27
2003-10-25T19:53:28.000000Z
K 7
svn:log
V 44
Better safe than clever.

Submitted by:	das

END
K 10
svn:author
V 3
imp
K 8
svn:date
V 27
2003-10-25T19:56:19.000000Z
K 7
svn:log
V 277
Convert to bus_space.
Make the pccard attachment work with NEWCARD
Start locking of the driver, but only the macros are defined right now

Tested on: Megahertz CC10BT/2
# (These cards are very popular on ebay these days, and run < $10 including
#  shipping from some sellers).

END
K 10
svn:author
V 3
imp
K 8
svn:date
V 27
2003-10-25T21:28:40.000000Z
K 7
svn:log
V 162
Minor style nits suggested by sam and mdodd:

o give an argument to EP_BUSY_WAIT
o use foo_locked rather than foo_body
o Add locking assertions for extra safety.

END
K 10
svn:author
V 3
jmg
K 8
svn:date
V 27
2003-10-25T22:03:10.000000Z
K 7
svn:log
V 91
fix spelling of ATTACH

Submitted by:	Johny Mattsson
MFC after:	2 week
(do to code freeze)

END
K 10
svn:author
V 3
alc
K 8
svn:date
V 27
2003-10-25T23:42:17.000000Z
K 7
svn:log
V 126
 - Add some of the required vm object locking, including assertions where
   the vm object lock is required and already held.

END
K 10
svn:author
V 3
phk
K 8
svn:date
V 27
2003-10-25T23:47:33.000000Z
K 7
svn:log
V 56
Consistently cast to (u_char *) when filling with junk.

END
K 10
svn:author
V 4
grog
K 8
svn:date
V 27
2003-10-26T00:03:49.000000Z
K 7
svn:log
V 33
Add end and start dates for DST.

END
K 10
svn:author
V 7
trhodes
K 8
svn:date
V 27
2003-10-26T00:35:05.000000Z
K 7
svn:log
V 47
style.Makefile: Add a '?' before '=' in WARNS.

END
K 10
svn:author
V 3
imp
K 8
svn:date
V 27
2003-10-26T00:51:40.000000Z
K 7
svn:log
V 40
Const poison string accessor functions.

END
K 10
svn:author
V 4
grog
K 8
svn:date
V 27
2003-10-26T01:18:08.000000Z
K 7
svn:log
V 33
Add coup d'état in Chile, 1973.

END
K 10
svn:author
V 6
alfred
K 8
svn:date
V 27
2003-10-26T02:19:00.000000Z
K 7
svn:log
V 59
constify the second args to timevaladd() and timevalsub().

END
K 10
svn:author
V 6
alfred
K 8
svn:date
V 27
2003-10-26T02:38:34.000000Z
K 7
svn:log
V 108
constify bintime_add, bintime_sub, bintime2timespec, timespec2bintime,
bintime2timeval and timeval2bintime.

END
K 10
svn:author
V 5
peter
K 8
svn:date
V 27
2003-10-26T03:01:24.000000Z
K 7
svn:log
V 137
Fix a warning about mismatched pointers.  A pointer to "void *" is not the
same as a pointer to "char *".  Tell the compiler this is ok.

END
K 10
svn:author
V 5
peter
K 8
svn:date
V 27
2003-10-26T03:12:47.000000Z
K 7
svn:log
V 118
When we pass a string as auxillary data (type long), be sure to convert
it to a suitable type for the initialization.

END
K 10
svn:author
V 5
peter
K 8
svn:date
V 27
2003-10-26T03:15:54.000000Z
K 7
svn:log
V 124
Disable the log() builtin (the math function), because it will not do much
logging of information like the program expects.

END
K 10
svn:author
V 5
peter
K 8
svn:date
V 27
2003-10-26T03:34:13.000000Z
K 7
svn:log
V 323
Move the -Wtraditional warning from the global cpp flags to when the -p
(portability warnings) switch is used.  Add -Wno-system-headers after it so
that we dont get 500 screenfulls of warnings about #elif in /usr/include.
I'm not entirely happy with this.  Maybe cdefs.h shouldn't use #elif and
instead nest #else clauses?

END
K 10
svn:author
V 5
peter
K 8
svn:date
V 27
2003-10-26T03:43:35.000000Z
K 7
svn:log
V 40
clnt_call takes args of type xdrproc_t.

END
K 10
svn:author
V 5
peter
K 8
svn:date
V 27
2003-10-26T03:51:47.000000Z
K 7
svn:log
V 167
(mostly) Clean up some const warnings here.  The code takes some liberties
because it is the originator of various const strings and knows that they
came from malloc.

END
K 10
svn:author
V 5
peter
K 8
svn:date
V 27
2003-10-26T03:55:58.000000Z
K 7
svn:log
V 73
Pacify gcc about casting pointers to integers (for the lowest few bits).

END
K 10
svn:author
V 5
peter
K 8
svn:date
V 27
2003-10-26T04:04:12.000000Z
K 7
svn:log
V 116
Preserve the constness of the value argument passed to env_setenv() as
it gets passed through the filter functions.

END
K 10
svn:author
V 5
peter
K 8
svn:date
V 27
2003-10-26T04:10:50.000000Z
K 7
svn:log
V 112
The third arg to strncmp() is size_t, not int.  This causes a warning on
systems where size_t is long, not int.

END
K 10
svn:author
V 5
peter
K 8
svn:date
V 27
2003-10-26T04:12:02.000000Z
K 7
svn:log
V 69
Pointers dont fit in an int on 64 bit platforms.  Fix a gcc warning.

END
K 10
svn:author
V 5
peter
K 8
svn:date
V 27
2003-10-26T04:20:19.000000Z
K 7
svn:log
V 110
Since I'm having so much fun with const, beat my head against the wall
some more.  Make this compile cleanly.

END
K 10
svn:author
V 5
peter
K 8
svn:date
V 27
2003-10-26T04:26:52.000000Z
K 7
svn:log
V 130
Fix some warnings by defining has_sa_sigaction to 1.  When the configure
script was run (years and years ago), we didn't have it.

END
K 10
svn:author
V 5
peter
K 8
svn:date
V 27
2003-10-26T04:30:05.000000Z
K 7
svn:log
V 51
Pacify gcc warning with a Douglas Adams reference.

END
K 10
svn:author
V 5
peter
K 8
svn:date
V 27
2003-10-26T04:32:53.000000Z
K 7
svn:log
V 41
Tidy up some xdrproc_t related warnings.

END
K 10
svn:author
V 5
peter
K 8
svn:date
V 27
2003-10-26T04:36:47.000000Z
K 7
svn:log
V 301
Give wider types to sscanf to fix two warnings (u_short cannot be > 0xffff)
and to make sure that we catch oversized arguments rather than silently
truncate them.  I dont know if sscanf will reject an integer if it will
not fit in the short return variable or not, but this way it should be
detected.

END
K 10
svn:author
V 5
peter
K 8
svn:date
V 27
2003-10-26T04:37:57.000000Z
K 7
svn:log
V 54
64 bit fixes. in_addr_t is an uint32_t, not a u_long.

END
K 10
svn:author
V 5
peter
K 8
svn:date
V 27
2003-10-26T04:43:02.000000Z
K 7
svn:log
V 192
Fix gcc warnings.  If NAME_MAX is 255, and d_namlen is a uint8_t, then
d_namlen can never be > NAME_MAX.  Stop gcc worrying about this by
using a preprocessor test to see if NAME_MAX changes.

END
K 10
svn:author
V 5
peter
K 8
svn:date
V 27
2003-10-26T04:45:08.000000Z
K 7
svn:log
V 145
Fix a 64 bit warning.  Have set_T_dev_t() take a pointer to a size_t rather
than a pointer to an int, since that is what it really wants anyway.

END
K 10
svn:author
V 5
peter
K 8
svn:date
V 27
2003-10-26T04:47:31.000000Z
K 7
svn:log
V 62
Fix some 64 bit warnings.  You can't fit a pointer in an int.

END
K 10
svn:author
V 5
peter
K 8
svn:date
V 27
2003-10-26T04:49:58.000000Z
K 7
svn:log
V 102
The math function logf() probably isn't doing us much good for logging
stuff.  Add -fno-builtin-logf.

END
K 10
svn:author
V 5
peter
K 8
svn:date
V 27
2003-10-26T04:57:32.000000Z
K 7
svn:log
V 25
More xdrproc_t warnings.

END
K 10
svn:author
V 5
peter
K 8
svn:date
V 27
2003-10-26T04:58:44.000000Z
K 7
svn:log
V 23
More xdrproc_t issues.

END
K 10
svn:author
V 5
peter
K 8
svn:date
V 27
2003-10-26T05:03:11.000000Z
K 7
svn:log
V 38
Fix warning about cgetent() argument.

END
K 10
svn:author
V 5
peter
K 8
svn:date
V 27
2003-10-26T05:05:48.000000Z
K 7
svn:log
V 114
Fix a warning that showed up on 64 bit systems.  It was actually a real
bug that had been swept under the carpet.

END
K 10
svn:author
V 5
peter
K 8
svn:date
V 27
2003-10-26T05:27:14.000000Z
K 7
svn:log
V 23
Prototype yp_maplist()

END
K 10
svn:author
V 5
peter
K 8
svn:date
V 27
2003-10-26T05:28:01.000000Z
K 7
svn:log
V 18
Stop gcc warnings

END
K 10
svn:author
V 5
peter
K 8
svn:date
V 27
2003-10-26T05:30:40.000000Z
K 7
svn:log
V 290
Fix up warnings.  There were some biggies here.  There seems to be two
different sets of definitions in /usr/include/rpcsvc, mostly compatable,
but with different names.  Will the real one please stand up?
In order to get prototypes for yp_maplist, we had to use the 'other'
naming system.

END
K 10
svn:author
V 5
peter
K 8
svn:date
V 27
2003-10-26T05:33:59.000000Z
K 7
svn:log
V 18
Make 64 bit safe.

END
K 10
svn:author
V 3
imp
K 8
svn:date
V 27
2003-10-26T05:34:01.000000Z
K 7
svn:log
V 61
Style changes:
o kill register
o minor function name tweaks.

END
K 10
svn:author
V 5
peter
K 8
svn:date
V 27
2003-10-26T05:51:51.000000Z
K 7
svn:log
V 202
Make this compile cleanly.  Switch to using rpcgen to generate
code that can be compiled on an ansi-C compiler, so that it puts the
xdrproc_t casts in.  This also implies a slight function name change.

END
K 10
svn:author
V 5
peter
K 8
svn:date
V 27
2003-10-26T05:54:07.000000Z
K 7
svn:log
V 219
Use -fno-builtin-log so gcc doesn't get ideas about using a math function
to log data.  Clean up an unused variable that was hidden by the WARNS?=2
being commented out.  Uncomment it now that it compiles cleanly again.

END
K 10
svn:author
V 5
peter
K 8
svn:date
V 27
2003-10-26T05:58:21.000000Z
K 7
svn:log
V 98
Use -fno-builtin-log.  Fix svc_sendreply() missing required casts.
It is now WARNS=2 clean again.

END
K 10
svn:author
V 5
peter
K 8
svn:date
V 27
2003-10-26T06:01:06.000000Z
K 7
svn:log
V 87
Fix a 64 bit time_t bogon that I missed from before.  lastlog.ll_time is
not a time_t.

END
K 10
svn:author
V 5
peter
K 8
svn:date
V 27
2003-10-26T06:10:44.000000Z
K 7
svn:log
V 97
Make this compile cleanly.  It passes WARNS=2, but I haven't checked
it is so on more platforms.

END
K 10
svn:author
V 3
ume
K 8
svn:date
V 27
2003-10-26T06:11:31.000000Z
K 7
svn:log
V 158
Since semantic of IPV6_PKTINFO was changed in RFC3542, we need to
use IPV6_RECVPKTINFO instead.

Reported by:	someone (I had removed the mail wrongly, sorry)

END
K 10
svn:author
V 5
peter
K 8
svn:date
V 27
2003-10-26T06:14:10.000000Z
K 7
svn:log
V 97
Deal with xdrproc_t casts.  Make these compile cleanly with WARNS=2
(but I haven't turned it on)

END
K 10
svn:author
V 3
ume
K 8
svn:date
V 27
2003-10-26T06:21:16.000000Z
K 7
svn:log
V 43
one more IPV6_PKTINFO -> IPV6_RECVPKTINFO.

END
K 10
svn:author
V 3
alc
K 8
svn:date
V 27
2003-10-26T06:29:26.000000Z
K 7
svn:log
V 240
 - Simplify vm_object_collapse()'s collapse case, reducing the number
   of lock acquires and releases performed.
 - Move an assertion from vm_object_collapse() to vm_object_zdtor()
   because it applies to all cases of object destruction.

END
K 10
svn:author
V 3
imp
K 8
svn:date
V 27
2003-10-26T06:31:46.000000Z
K 7
svn:log
V 43
Add ID for olicom GOCARD Ethernet (OC2220)

END
K 10
svn:author
V 3
imp
K 8
svn:date
V 27
2003-10-26T06:32:50.000000Z
K 7
svn:log
V 24
Sync to pccarddevs 1.74

END
K 10
svn:author
V 3
imp
K 8
svn:date
V 27
2003-10-26T06:34:22.000000Z
K 7
svn:log
V 26
add newcard support to ex

END
K 10
svn:author
V 3
bde
K 8
svn:date
V 27
2003-10-26T06:40:37.000000Z
K 7
svn:log
V 277
Fix previous commit.  glob_filename() returns "(char **) -1" for certain
errors, not necessarily a pointer such that (intptr_t)pointer is -1.

Also fix the style bug that the cast was not followed by a space.
This style of this file is now perfectly non-KNF for this cast too.

END
K 10
svn:author
V 8
simokawa
K 8
svn:date
V 27
2003-10-26T07:42:32.000000Z
K 7
svn:log
V 175
Revert previous change.

We have to put dcons_crom at first after 'optional' and
dcons and firewire are necessary and not optional for dcons_crom anyway.

Reported by: brooks

END
K 10
svn:author
V 3
ume
K 8
svn:date
V 27
2003-10-26T10:30:17.000000Z
K 7
svn:log
V 13
fix comment.

END
K 10
svn:author
V 3
ume
K 8
svn:date
V 27
2003-10-26T10:49:18.000000Z
K 7
svn:log
V 74
use uint32_t instead of u_int32_t for newly introduced
struct definition.

END
K 10
svn:author
V 3
ume
K 8
svn:date
V 27
2003-10-26T11:11:13.000000Z
K 7
svn:log
V 59
revert previous change.  we don't need to include types.h.

END
K 10
svn:author
V 3
des
K 8
svn:date
V 27
2003-10-26T11:20:47.000000Z
K 7
svn:log
V 57
MFC: (1.14) plug a memory leak in the PAM child process.

END
K 10
svn:author
V 3
ume
K 8
svn:date
V 27
2003-10-26T12:00:27.000000Z
K 7
svn:log
V 81
- errx() doesn't need `\n'.
- use %u for unsigned variable.

Obtained from:	KAME

END
K 10
svn:author
V 8
kensmith
K 8
svn:date
V 27
2003-10-26T14:21:30.000000Z
K 7
svn:log
V 158
	- Add note that loading modules only works if the system
	  securelevel is less than 1.

PR:		docs/58546
Reviewed by:	rwatson
Approved by:	blackend (mentor)

END
K 10
svn:author
V 3
ume
K 8
svn:date
V 27
2003-10-26T14:29:04.000000Z
K 7
svn:log
V 105
use official # for IFT_STF
(are there any backward compat issue?  i don't think so)

Obtained from:	KAME

END
K 10
svn:author
V 3
ume
K 8
svn:date
V 27
2003-10-26T15:06:06.000000Z
K 7
svn:log
V 20
drop unused fields.

END
K 10
svn:author
V 3
ume
K 8
svn:date
V 27
2003-10-26T15:15:36.000000Z
K 7
svn:log
V 21
drop unused defines.

END
K 10
svn:author
V 5
peter
K 8
svn:date
V 27
2003-10-26T16:35:11.000000Z
K 7
svn:log
V 83
Argh. Forward declare some structs rather than increase include file
dependencies.

END
K 10
svn:author
V 3
ume
K 8
svn:date
V 27
2003-10-26T18:17:01.000000Z
K 7
svn:log
V 103
re-add wrongly disappered IPV6_CHECKSUM stuff by introducing
ip6_raw_ctloutput().

Obtained from:	KAME

END
K 10
svn:author
V 4
fjoe
K 8
svn:date
V 27
2003-10-26T18:34:23.000000Z
K 7
svn:log
V 178
Fix setting PIO mode timings on AcerLabs Aladdin IV/V chipsets
(add missing 'break' statements).

Approved by:	sos, re
Reported by:	Juan Manuel Sanchez <juanmasf@mi.madritel.es>

END
K 10
svn:author
V 4
emax
K 8
svn:date
V 27
2003-10-26T19:09:12.000000Z
K 7
svn:log
V 131
Connect /etc/bluetooth and /usr/share/examples/netgraph/bluetooth to the build.

Reviewed by: imp, simon
Approved by: imp (mentor)

END
K 10
svn:author
V 4
fjoe
K 8
svn:date
V 27
2003-10-26T19:27:16.000000Z
K 7
svn:log
V 102
MFS: rev. 1.35.2.35:
Fix setting PIO mode timings on AcerLabs Aladdin IV/V chipsets.

Approved by:	re

END
K 10
svn:author
V 3
des
K 8
svn:date
V 27
2003-10-26T19:42:55.000000Z
K 7
svn:log
V 15
Fix Xr markup.

END
K 10
svn:author
V 3
alc
K 8
svn:date
V 27
2003-10-26T19:55:35.000000Z
K 7
svn:log
V 77
 - Introduce and use a mutex synchronizing access to the swblock hash table.

END
K 10
svn:author
V 3
imp
K 8
svn:date
V 27
2003-10-26T20:09:18.000000Z
K 7
svn:log
V 179
Minor tweak to Olicom 2220.  Also add the OC-2232 while I'm here.
Looks like 'special magic' is needed to access the ethernet part of
that combo card, so I've not added it to ex.

END
K 10
svn:author
V 3
imp
K 8
svn:date
V 27
2003-10-26T20:09:57.000000Z
K 7
svn:log
V 27
Sync to 1.75 of pccarddevs

END
K 10
svn:author
V 3
imp
K 8
svn:date
V 27
2003-10-26T22:22:34.000000Z
K 7
svn:log
V 32
Add Olicom GoCard 28.8k version

END
K 10
svn:author
V 3
imp
K 8
svn:date
V 27
2003-10-26T22:24:01.000000Z
K 7
svn:log
V 24
Sync to pccarddevs 1.76

END
K 10
svn:author
V 3
imp
K 8
svn:date
V 27
2003-10-26T22:28:20.000000Z
K 7
svn:log
V 111
Pass sc to GO_WINDOW macro.
Use SET_IRQ for pccard case, rather than exapnding it inline.

Requested by: mdodd

END
K 10
svn:author
V 3
imp
K 8
svn:date
V 27
2003-10-26T22:35:44.000000Z
K 7
svn:log
V 90
Make sn MPSAFE.  This has survived a couple of make buildworlds with
my Megahertz XJ10BT.

END
K 10
svn:author
V 3
das
K 8
svn:date
V 27
2003-10-27T01:28:07.000000Z
K 7
svn:log
V 458
Fix a problem where libm compiled under 5.X would depend on features
that are only in libc.so.5.  This broke some 4.X applications linked
to libm and run under 5.X.

Background:
In C99, isinf() and isnan() cannot be implemented as regular
functions.  We use macros that call libc functions in 5.X, but for
libm-internal use, we need to use the old versions until the next
time libm's major version number is bumped.

Submitted by:	bde
Reported by:	imp, kris

END
K 10
svn:author
V 6
murray
K 8
svn:date
V 27
2003-10-27T02:18:05.000000Z
K 7
svn:log
V 25
Identify as 4.9-RELEASE.

END
K 10
svn:author
V 6
murray
K 8
svn:date
V 27
2003-10-27T02:20:56.000000Z
K 7
svn:log
V 40
Bump __FreeBSD_version for FreeBSD 4.9.

END
K 10
svn:author
V 6
murray
K 8
svn:date
V 27
2003-10-27T02:22:50.000000Z
K 7
svn:log
V 21
Bump for 4.9-STABLE.

END
K 10
svn:author
V 6
murray
K 8
svn:date
V 27
2003-10-27T02:24:47.000000Z
K 7
svn:log
V 24
Identify as 4.9-STABLE.

END
K 10
svn:author
V 3
njl
K 8
svn:date
V 27
2003-10-27T02:42:08.000000Z
K 7
svn:log
V 56
Fix style problems with new options.

Requested by:	bde

END
K 10
svn:author
V 3
kan
K 8
svn:date
V 27
2003-10-27T04:16:32.000000Z
K 7
svn:log
V 120
Delay if_lo module intialization until domain list has been
completely populated. This prevents a system crash on boot.

END
K 10
svn:author
V 4
bmah
K 8
svn:date
V 27
2003-10-27T05:37:00.000000Z
K 7
svn:log
V 145
The next release of FreeBSD will most likely be 4.9.1, according to
current RE plans.

Approved by:	re (implicitly)
Requested by:	scottl, murray

END
K 10
svn:author
V 3
imp
K 8
svn:date
V 27
2003-10-27T05:37:01.000000Z
K 7
svn:log
V 14
Correct CEM33

END
K 10
svn:author
V 3
imp
K 8
svn:date
V 27
2003-10-27T05:37:34.000000Z
K 7
svn:log
V 13
sync to 1.77

END
K 10
svn:author
V 6
marcel
K 8
svn:date
V 27
2003-10-27T05:45:35.000000Z
K 7
svn:log
V 752
Add support for userland to access I/O port space. This is primarily
added for XFree86. There are 2 reasons for doing this with sysarch():
1. The memory mapped I/O space is not at a fixed physical address. An
   application has to use some interface to get the base address. It
   gets worse if the machine has multiple memory mapped I/O spaces.
2. Access to the memory mapped I/O space needs to happen through a
   translation that is flagged as uncachable. There's no interface
   that allows a process to do uncached memory I/O, other than though
   /dev/mem (possibly).

So, until we either disallow direct access to I/O or bus space from
userland or have a better way of doing this, sysarch() has the least
negative impact on existing interfaces.

END
K 10
svn:author
V 3
alc
K 8
svn:date
V 27
2003-10-27T05:58:15.000000Z
K 7
svn:log
V 71
 - Complete the synchronization of accesses to the swblock hash table.

END
K 10
svn:author
V 3
ken
K 8
svn:date
V 27
2003-10-27T06:15:55.000000Z
K 7
svn:log
V 794
In camperiphdone(), make sure we check for fatal errors and bail out
instead of retrying them blindly.

This should fix some of the problems people have been having with cdrom
drives taking a long time to probe.  This should also eliminate the need
for the initial TUR in cdsize().

cam_periph.c:	Don't keep retrying if the error we get back is a fatal
		error.  This should help us detect the transition from
		"Logical unit not ready, cause not reportable" to "Medium
		not present" in the "TUR many" handler.  (The TUR many
		handler gets triggered for Logical unit not ready, cause
		not reportable errors.)

scsi_cd.c:	Remove the initial test unit ready in cdsize().  Hopefully
		it isn't necessary after the above change.

Submitted by:	gibbs (mostly)
Tested by:	peter
MFC After:	2 weeks

END
K 10
svn:author
V 3
njl
K 8
svn:date
V 27
2003-10-27T06:26:51.000000Z
K 7
svn:log
V 130
Call the VESA reset BIOS vector on the resume path.  This may help displays
after resume.  I have not found it to break anything.

END
K 10
svn:author
V 3
imp
K 8
svn:date
V 27
2003-10-27T06:41:40.000000Z
K 7
svn:log
V 47
const char ** needs to be passed, not char **.

END
K 10
svn:author
V 4
jeff
K 8
svn:date
V 27
2003-10-27T06:47:05.000000Z
K 7
svn:log
V 1094
 - Use a better algorithm in sched_pctcpu_update()

Contributed by:	Thomaswuerfl@gmx.de

 - In sched_prio(), adjust the run queue for threads which may need to move
   to the current queue due to priority propagation .
 - In sched_switch(), fix style bug introduced when the KSE support went in.
   Columns are 80 chars wide, not 90.
 - In sched_switch(), Fix the comparison in the idle case and explicitly
   re-initialize the runq in the not propagated case.
 - Remove dead code in sched_clock().
 - In sched_clock(), If we're an IDLE class td set NEEDRESCHED so that threads
   that have become runnable will get a chance to.
 - In sched_runnable(), if we're not the IDLETD, we should not consider
   curthread when examining the load.  This mimics the 4BSD behavior of
   returning 0 when the only runnable thread is running.
 - In sched_userret(), remove the code for setting NEEDRESCHED entirely.
   This is not necessary and is not implemented in 4BSD.
 - Use the correct comparison in sched_add() when checking to see if an idle
   prio task has had it's priority temporarily elevated.

END
K 10
svn:author
V 6
marcel
K 8
svn:date
V 27
2003-10-27T06:50:57.000000Z
K 7
svn:log
V 110
Fix truss so that it doesn't abort/exit when a syscall has been given
a NULL-pointer for a sockaddr argument.

END
K 10
svn:author
V 3
ume
K 8
svn:date
V 27
2003-10-27T07:15:22.000000Z
K 7
svn:log
V 85
M_DONTWAIT was passed into malloc().

Submitted by:	Ian Dowse <iedowse@maths.tcd.ie>

END
K 10
svn:author
V 4
jeff
K 8
svn:date
V 27
2003-10-27T07:15:47.000000Z
K 7
svn:log
V 58
 - Don't set td_priority directly here, use sched_prio().

END
K 10
svn:author
V 6
murray
K 8
svn:date
V 27
2003-10-27T07:56:27.000000Z
K 7
svn:log
V 142
Add proper entries to release table so that pkg_add -r on a 4.9 system
will fetch 4.9 packages, instead of falling back on 4-stable packages.

END
K 10
svn:author
V 6
murray
K 8
svn:date
V 27
2003-10-27T07:57:32.000000Z
K 7
svn:log
V 68
MFS: 4.9-RELEASE should fetch the release packages with pkg_add -r.

END
K 10
svn:author
V 7
cvs2svn
K 8
svn:date
V 27
2003-10-27T07:57:33.000000Z
K 7
svn:log
V 77
This commit was manufactured by cvs2svn to create tag
'RELENG_4_9_0_RELEASE'.
END
K 10
svn:author
V 6
jkoshy
K 8
svn:date
V 27
2003-10-27T09:19:07.000000Z
K 7
svn:log
V 66
Correct a typo.

Submitted by:	"T.M. Sommers" <tms2@mail.ptd.net>

END
K 10
svn:author
V 5
harti
K 8
svn:date
V 27
2003-10-27T11:19:08.000000Z
K 7
svn:log
V 62
Allow building the NgATM SAAL layer directly into the kernel.

END
K 10
svn:author
V 5
harti
K 8
svn:date
V 27
2003-10-27T11:43:20.000000Z
K 7
svn:log
V 45
Man pages for the NgATM SAAL protocol nodes.

END
K 10
svn:author
V 5
harti
K 8
svn:date
V 27
2003-10-27T13:51:28.000000Z
K 7
svn:log
V 48
Add a definition for libngatm.

Reviewed by:	ru

END
K 10
svn:author
V 4
emax
K 8
svn:date
V 27
2003-10-27T16:10:05.000000Z
K 7
svn:log
V 64
Create missing /usr/share/examples/netgraph/bluetooth directory

END
K 10
svn:author
V 5
harti
K 8
svn:date
V 27
2003-10-27T16:21:59.000000Z
K 7
svn:log
V 502
When we cannot allocate an external buffer (bacause we've hit
the maximum number of pages for buffers) return -1 instead of 0.
This fixes a panic under conditions when many mbufs are needed.

Update the head pointer of the receive buffer pool queue even when
we could not supply a buffer to the chip. Otherwise the chip will
not re-interrupt us for another try. A better strategy would probably
be to remember this condition and to supply buffers without an interrupt
as soon as buffers get available.

END
K 10
svn:author
V 6
scottl
K 8
svn:date
V 27
2003-10-27T16:39:49.000000Z
K 7
svn:log
V 86
Directly call the 'boot' word instead of indirectly evaluating it.

Submitted by: dcs

END
K 10
svn:author
V 4
bmah
K 8
svn:date
V 27
2003-10-27T18:49:51.000000Z
K 7
svn:log
V 119
Trim release notes and errata for a post-4.9 world, bump version
numbers for 4.9-STABLE.

Approved by:	re (implicitly)

END
K 10
svn:author
V 6
jesper
K 8
svn:date
V 27
2003-10-27T20:14:53.000000Z
K 7
svn:log
V 63
Add status checking of ATA raid to the daily periodic scripts.

END
K 10
svn:author
V 3
jhb
K 8
svn:date
V 27
2003-10-27T22:15:02.000000Z
K 7
svn:log
V 152
Fix pmap_unmapdev() to call pmap_kremove() instead of implementing it
directly so that it more closely mirrors pmap_mapdev() which calls
pmap_kenter().

END
K 10
svn:author
V 6
marcel
K 8
svn:date
V 27
2003-10-27T22:54:34.000000Z
K 7
svn:log
V 601
The previous commit removed both clause 3 and clause 4 from the UCB
license. Only clause 3 has been revoked. Restore the fourth clause
as clause 3.

Pointed out by: das@

Remove my name as a copyright holder since I don't use a BSD license
compatible or comparable to the UCB license. I choose not to add a
complete second license for my work for aesthetic reasons, nor to
replace the UCB license on grounds of rewriting more than 90% of the
source files. The rewrite can also be seen as an enhancement and since
the files were practically empty, it's rather trivial to have changed
90% of the files.

END
K 10
svn:author
V 5
peter
K 8
svn:date
V 27
2003-10-28T01:06:37.000000Z
K 7
svn:log
V 76
Oops. Remove some rather noisy debug printfs that slipped in there
somehow.

END
K 10
svn:author
V 8
simokawa
K 8
svn:date
V 27
2003-10-28T01:31:50.000000Z
K 7
svn:log
V 61
I wrongly committed an older version.
Update references etc.

END
K 10
svn:author
V 4
jeff
K 8
svn:date
V 27
2003-10-28T03:28:48.000000Z
K 7
svn:log
V 234
 - Only change the run queue in sched_prio() if the kse is non null.  threads
   can be in the TD_ON_RUNQ state and not have an associated kse.
 - Remove the PRI_IDLE special case from sched_clock(), it was not actually
   necessary.

END
K 10
svn:author
V 3
jmg
K 8
svn:date
V 27
2003-10-28T05:02:03.000000Z
K 7
svn:log
V 142
take mux's fix to the next level, scan the names and make the field larger
as necessary... on sparcs where:
atapci0 vec1996
is a bit too long

END
K 10
svn:author
V 3
tjr
K 8
svn:date
V 27
2003-10-28T05:22:19.000000Z
K 7
svn:log
V 40
Catch up with removal of /usr/bin/svr4.

END
K 10
svn:author
V 3
sam
K 8
svn:date
V 27
2003-10-28T05:47:40.000000Z
K 7
svn:log
V 182
speedup stream socket recv handling by tracking the tail of
the mbuf chain instead of walking the list for each append

Submitted by:	ps/jayanth
Obtained from:	netbsd (jason thorpe)

END
K 10
svn:author
V 3
ume
K 8
svn:date
V 27
2003-10-28T16:16:04.000000Z
K 7
svn:log
V 114
mib name was changed by fixing a spelling.
net.key.prefered_oldsa -> net.key.preferred_oldsa

Obtained from:	KAME

END
K 10
svn:author
V 3
ume
K 8
svn:date
V 27
2003-10-28T16:29:26.000000Z
K 7
svn:log
V 43
cleanup use of m_tag.

Obtained from:	KAME

END
K 10
svn:author
V 3
ume
K 8
svn:date
V 27
2003-10-28T16:45:29.000000Z
K 7
svn:log
V 59
make sure to accept only IPv6 packet.

Obtained from:	KAME

END
K 10
svn:author
V 6
scottl
K 8
svn:date
V 27
2003-10-28T17:18:42.000000Z
K 7
svn:log
V 69
Directly call the 'reboot' word instead of indirectly evaluating it.

END
K 10
svn:author
V 4
bmah
K 8
svn:date
V 27
2003-10-28T18:15:15.000000Z
K 7
svn:log
V 109
First errata against 4.9-RELEASE:  Change in HyperThreading support.

While here, add trademark boilerplate.

END
K 10
svn:author
V 3
sos
K 8
svn:date
V 27
2003-10-28T19:01:48.000000Z
K 7
svn:log
V 103
This should allow us to boot with DMA enabled on unknown PCI ATA
chipsets, well at least newer ones...

END
K 10
svn:author
V 6
marcel
K 8
svn:date
V 27
2003-10-28T19:38:26.000000Z
K 7
svn:log
V 2333
When switching the RSE to use the kernel stack as backing store, keep
the RNAT bit index constant. The net effect of this is that there's
no discontinuity WRT NaT collections which greatly simplifies certain
operations. The cost of this is that there can be up to 504 bytes of
unused stack between the true base of the kernel stack and the start
of the RSE backing store. The cost of adjusting the backing store
pointer to keep the RNAT bit index constant, for each kernel entry,
is negligible.

The primary reasons for this change are:
1. Asynchronuous contexts in KSE processes have the disadvantage of
   having to copy the dirty registers from the kernel stack onto the
   user stack. The implementation we had so far copied the registers
   one at a time without calculating NaT collection values. A process
   that used speculation would not work. Now that the RNAT bit index
   is constant, we can block-copy the registers from the kernel stack
   to the user stack without having to worry about NaT collections.
   They will be in the right place on the user stack.
2. The ndirty field in the trapframe is now also usable in userland.
   This was previously not the case because ndirty also includes the
   space occupied by NaT collections. The value could be off by 8,
   depending on the discontinuity. Now that the RNAT bit index is
   contants, we have exactly the same number of NaT collection points
   on the kernel stack as we would have had on the user stack if we
   didn't switch backing stores.
3. Debuggers and other applications that use ptrace(2) can now copy
   the dirty registers from the kernel stack (using ptrace(2)) and
   copy them whereever they want them (onto the user stack of the
   inferior as might be the case for gdb) without having to worry
   about NaT collections in the same way the kernel doesn't have to
   worry about them.

There's a second order effect caused by the randomization of the
base of the backing store, for it depends on the number of dirty
registers the processor happened to have at the time of entry into
the kernel. The second order effect is that the RSE will have a
better cache utilization as compared to having the backing store
always aligned at page boundaries. This has not been measured and
may be in practice only minimally beneficial, if at all measurable.

END
K 10
svn:author
V 3
jhb
K 8
svn:date
V 27
2003-10-28T20:46:23.000000Z
K 7
svn:log
V 924
According to the submitter, POSIX mandates that all interval timers are
reset in a child process after a fork().  Currently, however, only the
real timer is cleared while the virtual and profiling timers are inherited.

The realtimer is cleared because it lives directly in struct proc in
p_realtimer.  It is in the zero'd section of struct proc.  The other timers
live in the p_timer[] array in struct pstats.  These timers are copied on
fork() rather than zero'd.  The fix is to move p_timer[] to the zero'd
part of struct pstats so that they are zero'd instead of copied on fork().

Note: Since at least FreeBSD 2.0 (and possibly earlier) we've had storage
for two real interval timers.  Now that the uarea is less important,
perhaps we could move all of p_timer[] over to struct proc and drop the
p_realtimer special case to fix that.

PR:		kern/58647
Reported by:	Dan Nelson <dnelson@allantgroup.com>
MFC after:	1 week

END
K 10
svn:author
V 6
brooks
K 8
svn:date
V 27
2003-10-28T20:58:02.000000Z
K 7
svn:log
V 32
Use VLANNAME instead of "vlan".

END
K 10
svn:author
V 3
sos
K 8
svn:date
V 27
2003-10-28T21:08:14.000000Z
K 7
svn:log
V 63
Cleanup the interrupt code that deals with the busmaster bits.

END
K 10
svn:author
V 5
peter
K 8
svn:date
V 27
2003-10-28T22:36:54.000000Z
K 7
svn:log
V 618
Don peril sensitive (ie: bikeshed sensitive) sunglasses and quietly
send strhash(3) off to sleep with the fishes.  Nothing in our tree uses it.
It has no documentation.  It is nonstandard and in spite of the filename
strhash.c and strhash.h, it lives in application namespace by providing
compulsory global symbols hash_create()/hash_destroy()/hash_search()/
hash_traverse()/hash_purge()/hash_stats() regardless of whether you
#include <strhash.h> or not.  If it turns out that there is a huge
application for this after all, I can repocopy it somewhere safer and
we can revive it elsewhere.  But please, not in libc!

END