]> CyberLeo.Net >> Repos - FreeBSD/FreeBSD.git/log
FreeBSD/FreeBSD.git
23 years agoMove bindery authentication ncps to ncp_ncp.c file. ncp_login.c will stay
Boris Popov [Thu, 22 Mar 2001 10:38:16 +0000 (10:38 +0000)]
Move bindery authentication ncps to ncp_ncp.c file. ncp_login.c will stay
empty for a while.

23 years agoProperly set flags on the broken connection.
Boris Popov [Thu, 22 Mar 2001 10:29:39 +0000 (10:29 +0000)]
Properly set flags on the broken connection.

23 years agoEnhancement to 1.99 -> 1.100.
Ruslan Ermilov [Thu, 22 Mar 2001 09:54:25 +0000 (09:54 +0000)]
Enhancement to 1.99 -> 1.100.
Avoid EPERM from ln(1) in a different (proper) way.

23 years agoo Actually extract version of interface and store it along with the name.
Boris Popov [Thu, 22 Mar 2001 08:58:45 +0000 (08:58 +0000)]
o Actually extract version of interface and store it along with the name.

o Add new parameter to the modlist_lookup() function to perform lookups
  with strict version matching.

o Collapse duplicate code to function(s).

23 years agoSlightly reorganize code in the linker_load_dependancies() function to make
Boris Popov [Thu, 22 Mar 2001 07:55:33 +0000 (07:55 +0000)]
Slightly reorganize code in the linker_load_dependancies() function to make
codepath more straightforward.

23 years agoDon't create legacy wd* devices by default. Let the user do this if
Jordan K. Hubbard [Thu, 22 Mar 2001 07:30:02 +0000 (07:30 +0000)]
Don't create legacy wd* devices by default.  Let the user do this if
they so desire it.

Suggested by: rwatson

23 years agoRemove support for old way of handling module dependencies.
Boris Popov [Thu, 22 Mar 2001 07:14:42 +0000 (07:14 +0000)]
Remove support for old way of handling module dependencies.

Approved by: peter

23 years agoUse orthogonal www.FreeBSD.org syntax.
Jordan K. Hubbard [Thu, 22 Mar 2001 07:01:19 +0000 (07:01 +0000)]
Use orthogonal www.FreeBSD.org syntax.

Whined about by: rwatson

23 years agoAdd dependancy on libmchain module.
Boris Popov [Thu, 22 Mar 2001 06:51:53 +0000 (06:51 +0000)]
Add dependancy on libmchain module.

Spotted by: Andrzej Tobola <san@iem.pw.edu.pl>

23 years agoFirst step towards making loadable modules independent of having
Warner Losh [Thu, 22 Mar 2001 06:00:07 +0000 (06:00 +0000)]
First step towards making loadable modules independent of having
pccard in the kernel for those drivers with pccard attachments.  This
makes the compat layer a little larger by introducing some inlines,
but should almost make it possible to have independent attachments.
The pccard_match function are the only one left, which I will take
care of shortly.

23 years agoSync to last changes.
Warner Losh [Thu, 22 Mar 2001 05:52:42 +0000 (05:52 +0000)]
Sync to last changes.

23 years agoAdd zoom air 4000.
Warner Losh [Thu, 22 Mar 2001 05:51:58 +0000 (05:51 +0000)]
Add zoom air 4000.

23 years agoAxe a few __P() while I'm in the neighborhood.
Warner Losh [Thu, 22 Mar 2001 05:49:18 +0000 (05:49 +0000)]
Axe a few __P() while I'm in the neighborhood.

23 years agoStyle(9) changes:
Warner Losh [Thu, 22 Mar 2001 05:41:20 +0000 (05:41 +0000)]
Style(9) changes:
put parens around return (statements);
Use ANSI funcitons only.

23 years agoInclude docs with "default X dist" so XF86Config won't complain about it.
Jordan K. Hubbard [Thu, 22 Mar 2001 05:36:54 +0000 (05:36 +0000)]
Include docs with "default X dist" so XF86Config won't complain about it.

23 years agoDon't install Kerberos stuff if user has selected minimal or "user"
Jordan K. Hubbard [Thu, 22 Mar 2001 05:33:15 +0000 (05:33 +0000)]
Don't install Kerberos stuff if user has selected minimal or "user"
modes.

23 years agoHopefully fix some of the bugs in passing credentials over UNIX domai...
Alfred Perlstein [Thu, 22 Mar 2001 04:31:30 +0000 (04:31 +0000)]
Hopefully fix some of the bugs in passing credentials over UNIX                 domain sockets.

Make struct cmessage visible from socket.h (about 4 places were
defining it for themselves which wasn't good)

Make __rpc_get_local_uid() useable and give it prototype that's
visible.

Fix some issues with printing out usernames from rpcbind and keyserv.

23 years agoo Remove BINMODE=0550 -- no need to restrict execution rights as the
Robert Watson [Thu, 22 Mar 2001 04:27:46 +0000 (04:27 +0000)]
o Remove BINMODE=0550 -- no need to restrict execution rights as the
  authorization check is on /dev/{,k}mem.
o Update man page to reflect requirement that gid kmem privileges must
  be held, not root.  (submitted by: peter)

23 years agoo Strip BINMODE from Makefile. Installing watch with 0500 provides no
Robert Watson [Thu, 22 Mar 2001 04:10:51 +0000 (04:10 +0000)]
o Strip BINMODE from Makefile.  Installing watch with 0500 provides no
  added security benefit, as it doesn't provide extra privilege.
  Protection against inappropriate snooping is done by protecting
  /dev/snp* properly.

23 years agoo Slap some "_"'s in front of variable names relating to extattr functions,
Robert Watson [Thu, 22 Mar 2001 04:05:40 +0000 (04:05 +0000)]
o Slap some "_"'s in front of variable names relating to extattr functions,
  so as not to pollute application namespace.

Submitted by: bde

23 years agoadd/correct a bunch of codec ids
Cameron Grant [Thu, 22 Mar 2001 02:41:24 +0000 (02:41 +0000)]
add/correct a bunch of codec ids

Obtained from: linux, alsa, various datasheets

23 years agoNew release notes: OpenSSH resistance to traffic analysis, make(1)
Bruce A. Mah [Thu, 22 Mar 2001 02:40:50 +0000 (02:40 +0000)]
New release notes:  OpenSSH resistance to traffic analysis, make(1)
bugs fixed, fsck_ffs in background on mounted filesystems, TI-RPC,
portmap(8)->rpcbind(8), IPv6 support in NFS, lockd, rc(8) dependency
handling.

Synch up hardware devices with etc/defaults/pccard.conf 1.98.2.19
(yes, from RELENG_4).  This change needs to be propagated to RELNOTESng.

23 years ago- XF86.patch is no longer needed as it is all checked into the XFree86
John Baldwin [Thu, 22 Mar 2001 01:47:02 +0000 (01:47 +0000)]
- XF86.patch is no longer needed as it is all checked into the XFree86
  port conditional on the BUILD_XDIST variable.
- Define BUILD_XDIST=yes when compiling XFree86.

23 years agoProperly deal with one char weekdays
Andrey A. Chernov [Thu, 22 Mar 2001 00:38:46 +0000 (00:38 +0000)]
Properly deal with one char weekdays

23 years agoUppercase first month letter
Andrey A. Chernov [Thu, 22 Mar 2001 00:25:26 +0000 (00:25 +0000)]
Uppercase first month letter
Terminate weekday string

23 years agoThis adds /etc/netconfig to the list of files that get copied to
Alfred Perlstein [Wed, 21 Mar 2001 23:53:40 +0000 (23:53 +0000)]
This adds /etc/netconfig to the list of files that get copied to
/stand/etc on the mfsroot, and hence to /etc on the install target
disk for use during the install. It also links netconfig and services
into /etc on the mfsroot so that they are available to a fixit
shell.

Submitted by: iedowse

23 years agoypserv has intimate knowledge of the server transport handle which is
Bill Paul [Wed, 21 Mar 2001 23:47:18 +0000 (23:47 +0000)]
ypserv has intimate knowledge of the server transport handle which is
needed to make the asynchronous DNS lookup mechanism work. (It needs to
be able to get/set the transaction ID in the trasport handle so it can
deliver a delayed UDP response when a reply is received from a DNS
server.) With TI-TPC, the transport handle has changed slightly (what
used to be an int is now a size_t) so we need to account for this.

23 years agoAdd padding for some short names to make their length the same to not
Andrey A. Chernov [Wed, 21 Mar 2001 23:24:41 +0000 (23:24 +0000)]
Add padding for some short names to make their length the same to not
break colums

23 years agoo Remove an unnecessary debugging printf from ufs_extattr_lookup(),
Robert Watson [Wed, 21 Mar 2001 23:00:39 +0000 (23:00 +0000)]
o Remove an unnecessary debugging printf from ufs_extattr_lookup(),
  which resulted in the output of warning messages at boot if
  UFS_EXTATTR_AUTOSTART was enabled but ".attribute" and possible
  sub-directories weren't in a mounted MFS or UFS file systems.

Pointed out by: dcs
Obtained from: TrustedBSD Project

23 years agoHelp standalone builds by getting libutil.h from src/lib/libutil
Peter Wemm [Wed, 21 Mar 2001 22:54:59 +0000 (22:54 +0000)]
Help standalone builds by getting libutil.h from src/lib/libutil

23 years agoRemove fixed length restrictions and padding
Andrey A. Chernov [Wed, 21 Mar 2001 22:54:44 +0000 (22:54 +0000)]
Remove fixed length restrictions and padding

23 years agoFind <libkvm.h> in the source tree. This helps standalone builds.
Peter Wemm [Wed, 21 Mar 2001 22:52:48 +0000 (22:52 +0000)]
Find <libkvm.h> in the source tree. This helps standalone builds.

23 years agoBack out rev 1.4. The advertised 64K limit on transfers only applies when
Scott Long [Wed, 21 Mar 2001 21:53:09 +0000 (21:53 +0000)]
Back out rev 1.4.  The advertised 64K limit on transfers only applies when
using 64bit S/G entries.  With this reverted, we are seeing >92MB/sec reads
and >42MB/sec writes on a RAID-5 container.

23 years agoFix width for printing time
Andrey A. Chernov [Wed, 21 Mar 2001 21:49:33 +0000 (21:49 +0000)]
Fix width for printing time

23 years agoDon't attempt to parse %c
Andrey A. Chernov [Wed, 21 Mar 2001 21:17:54 +0000 (21:17 +0000)]
Don't attempt to parse %c

23 years agoUntil this is *truly* unused (sorry Steve) and replaced by something
Jordan K. Hubbard [Wed, 21 Mar 2001 21:16:18 +0000 (21:16 +0000)]
Until this is *truly* unused (sorry Steve) and replaced by something
better, put it back.

23 years agoDon't attempt to parse %c
Andrey A. Chernov [Wed, 21 Mar 2001 20:47:43 +0000 (20:47 +0000)]
Don't attempt to parse %c

23 years agoDon't attempt to parse %c
Andrey A. Chernov [Wed, 21 Mar 2001 20:32:20 +0000 (20:32 +0000)]
Don't attempt to parse %c

23 years agoInclude headers to unbreak world.
Alfred Perlstein [Wed, 21 Mar 2001 20:30:35 +0000 (20:30 +0000)]
Include headers to unbreak world.

Submitted by: Ollivier Robert <roberto@eurocontrol.fr>

23 years agoDon't attempt to parse %c
Andrey A. Chernov [Wed, 21 Mar 2001 20:26:51 +0000 (20:26 +0000)]
Don't attempt to parse %c

23 years agoAlways disable paging when exiting back to real mode after receiving a
John Baldwin [Wed, 21 Mar 2001 20:08:07 +0000 (20:08 +0000)]
Always disable paging when exiting back to real mode after receiving a
fatal trap.  Also, reload the GDT register to point to BTX's GDT before
playing around with the segment registers to return to real mode.  This is
helpful if the kernel causes a fatal exception before it has setup its own
IDT and fault handlers.  For example, if one happens to break mtx_init().
Without these changes BTX would recursively page fault (if paging was not
disabled) or triple fault and reset the CPU (without the GDT reload)
instead of providing a potentially useful register dump.

Reviewed by: rnordier

23 years agoCosmetique, use %R instead of %H:%M
Andrey A. Chernov [Wed, 21 Mar 2001 19:19:57 +0000 (19:19 +0000)]
Cosmetique, use %R instead of %H:%M

23 years agoReorder mergemaster entries to be adjacent. Reorder FFS-related
Bruce A. Mah [Wed, 21 Mar 2001 19:16:51 +0000 (19:16 +0000)]
Reorder mergemaster entries to be adjacent.  Reorder FFS-related
entries to be adjacent.

No new text.

23 years agoDon't attempt to parse %c
Andrey A. Chernov [Wed, 21 Mar 2001 19:16:41 +0000 (19:16 +0000)]
Don't attempt to parse %c

23 years agoDon't attempt to parse %c
Andrey A. Chernov [Wed, 21 Mar 2001 19:08:01 +0000 (19:08 +0000)]
Don't attempt to parse %c

23 years agoNew release notes: ahc(4) bugfixes, FNV NFS hash, syscons(4) keyboard
Bruce A. Mah [Wed, 21 Mar 2001 19:04:31 +0000 (19:04 +0000)]
New release notes:  ahc(4) bugfixes, FNV NFS hash, syscons(4) keyboard
paste, glob(3) limits, ping(8) -m.

23 years agoDon't attempt to parse %c
Andrey A. Chernov [Wed, 21 Mar 2001 18:43:49 +0000 (18:43 +0000)]
Don't attempt to parse %c

23 years agoDon't attempt to parse %c, use nl_langinfo instead
Andrey A. Chernov [Wed, 21 Mar 2001 15:51:50 +0000 (15:51 +0000)]
Don't attempt to parse %c, use nl_langinfo instead

23 years agoDon't attempt to parse %c, use nl_langinfo instead
Andrey A. Chernov [Wed, 21 Mar 2001 15:41:02 +0000 (15:41 +0000)]
Don't attempt to parse %c, use nl_langinfo instead

23 years agoRestore old locale properly
Andrey A. Chernov [Wed, 21 Mar 2001 15:39:17 +0000 (15:39 +0000)]
Restore old locale properly

23 years agoSilence minor cc warning
Andrey A. Chernov [Wed, 21 Mar 2001 15:14:47 +0000 (15:14 +0000)]
Silence minor cc warning

23 years agoDon't attempt to parse %c output, use nl_langinfo instead
Andrey A. Chernov [Wed, 21 Mar 2001 15:13:50 +0000 (15:13 +0000)]
Don't attempt to parse %c output, use nl_langinfo instead

23 years agoThe size of the disk can't be gotten reliably with the read capacity
Søren Schmidt [Wed, 21 Mar 2001 14:59:38 +0000 (14:59 +0000)]
The size of the disk can't be gotten reliably with the read capacity
command, so use the info from the TOC instead.

23 years agoGet rid of non-standard %E[Ff] formats, userland apps already fixed
Andrey A. Chernov [Wed, 21 Mar 2001 14:52:12 +0000 (14:52 +0000)]
Get rid of non-standard %E[Ff] formats, userland apps already fixed

23 years agoUse nl_langinfo instead of %Ef
Andrey A. Chernov [Wed, 21 Mar 2001 14:32:02 +0000 (14:32 +0000)]
Use nl_langinfo instead of %Ef

23 years agoParanthesise macros. This fixes an unlikely bug and silences lint.
Mark Murray [Wed, 21 Mar 2001 14:31:51 +0000 (14:31 +0000)]
Paranthesise macros. This fixes an unlikely bug and silences lint.

23 years agoUse nl_langinfo instead of %EF
Andrey A. Chernov [Wed, 21 Mar 2001 14:24:05 +0000 (14:24 +0000)]
Use nl_langinfo instead of %EF

23 years agoOops, back out prev. change - POSIX require %y in d_fmt
Andrey A. Chernov [Wed, 21 Mar 2001 14:12:37 +0000 (14:12 +0000)]
Oops, back out prev. change - POSIX require %y in d_fmt
Cosmetique - use exact POSIX string for %c

23 years agoChange type of channel speed variable from "int" to "u_int32_t" to
Orion Hodson [Wed, 21 Mar 2001 14:10:51 +0000 (14:10 +0000)]
Change type of channel speed variable from "int" to "u_int32_t" to
avoid overflow when scaling playback rate.  Fixes 44.1k playback from
being 48k (I am deaf to this difference...).

23 years agoReplace %y with %Y
Andrey A. Chernov [Wed, 21 Mar 2001 13:55:09 +0000 (13:55 +0000)]
Replace %y with %Y

23 years agoUse %c for expire/change dates
Andrey A. Chernov [Wed, 21 Mar 2001 13:46:09 +0000 (13:46 +0000)]
Use %c for expire/change dates

23 years agoUse nl_langinfo instead of %Ef
Andrey A. Chernov [Wed, 21 Mar 2001 13:41:03 +0000 (13:41 +0000)]
Use nl_langinfo instead of %Ef

23 years agoUse nl_langinfo instead of %Ef
Andrey A. Chernov [Wed, 21 Mar 2001 13:33:09 +0000 (13:33 +0000)]
Use nl_langinfo instead of %Ef

23 years agoUse nl_langinfo instead of %Ef
Andrey A. Chernov [Wed, 21 Mar 2001 13:21:20 +0000 (13:21 +0000)]
Use nl_langinfo instead of %Ef

23 years agoChange ordering of SPDIF register pokes. SPDIF enable needs to be the
Orion Hodson [Wed, 21 Mar 2001 12:51:37 +0000 (12:51 +0000)]
Change ordering of SPDIF register pokes.  SPDIF enable needs to be the
last poke in sequence.  Enabling SPDIF was coercing output rate to
48K, not good for 44.1K tracks.

23 years agoSet the device names as early as possible.
Søren Schmidt [Wed, 21 Mar 2001 11:49:07 +0000 (11:49 +0000)]
Set the device names as early as possible.

23 years agoHandle the case where the last piece of a RAID0 (striped) disk is
Søren Schmidt [Wed, 21 Mar 2001 11:48:14 +0000 (11:48 +0000)]
Handle the case where the last piece of a RAID0 (striped) disk is
not of interleave size.

23 years agoCleanup the diskerr messages a bit.
Søren Schmidt [Wed, 21 Mar 2001 11:46:15 +0000 (11:46 +0000)]
Cleanup the diskerr messages a bit.

23 years agoDo not change/get mode on a nonexisting device.
Søren Schmidt [Wed, 21 Mar 2001 11:44:58 +0000 (11:44 +0000)]
Do not change/get mode on a nonexisting device.

23 years agoProperly fix the world breakage.
Ruslan Ermilov [Wed, 21 Mar 2001 10:56:29 +0000 (10:56 +0000)]
Properly fix the world breakage.

Unpollute the Makefile.inc's abuse of SRCS variable, and move
the generation of header files into include/ subdirecory.

Reviewed by: buildworld

23 years agoAdditions to run checks on live filesystems. This change will not
Kirk McKusick [Wed, 21 Mar 2001 09:48:03 +0000 (09:48 +0000)]
Additions to run checks on live filesystems. This change will not
affect current systems until fsck is modified to use these new
facilities. To try out this change, set the fsck passno to zero
in /etc/fstab to cause the filesystem to be mounted without running
fsck, then run `fsck_ffs -p -B <filesystem>' after the system has
been brought up multiuser to run a background cleanup on <filesystem>.
Note that the <filesystem> in question must have soft updates enabled.

23 years agoAdd libssl and libcrypto compat4x libraries since the major number
Paul Saab [Wed, 21 Mar 2001 08:49:02 +0000 (08:49 +0000)]
Add libssl and libcrypto compat4x libraries since the major number
changed.  These were taken from the 4.2-RELEASE dist on ftp.freebsd.org.
This will be MFC'd shortly as it is required in RELENG_4 to maintain
compatability with binaries linked against these libraries.

23 years agoPentium II's do not support SSE, that came in with the PIII
Kris Kennaway [Wed, 21 Mar 2001 08:29:39 +0000 (08:29 +0000)]
Pentium II's do not support SSE, that came in with the PIII

Submitted by: sf

23 years agoOnly flush rules that have a rule number above that set by a new
Paul Richards [Wed, 21 Mar 2001 08:19:31 +0000 (08:19 +0000)]
Only flush rules that have a rule number above that set by a new
sysctl, net.inet.ip.fw.permanent_rules.

This allows you to install rules that are persistent across flushes,
which is very useful if you want a default set of rules that
maintains your access to remote machines while you're reconfiguring
the other rules.

Reviewed by: Mark Murray <markm@FreeBSD.org>

23 years agoAdd kernel support for running fsck on active filesystems.
Kirk McKusick [Wed, 21 Mar 2001 04:09:01 +0000 (04:09 +0000)]
Add kernel support for running fsck on active filesystems.

23 years agoClear the fs_clean flag only when the FS_UNCLEAN flag is not set
Kirk McKusick [Wed, 21 Mar 2001 04:05:20 +0000 (04:05 +0000)]
Clear the fs_clean flag only when the FS_UNCLEAN flag is not set
(as is done in unmount).

Remove a snapshot inode from the superblock list when its last
name goes away rather than when its last reference goes away.
That way it will be properly reclaimed by fsck after a crash
rather than reenabled when the filesystem is mounted.

23 years agoAdd SRCS= mk-mdmap.c and SRCS= wire-test.c to fix the world
Michael C . Wu [Wed, 21 Mar 2001 04:04:34 +0000 (04:04 +0000)]
Add SRCS= mk-mdmap.c and SRCS= wire-test.c to fix the world
breakage.  amd has its own SRCS= that prevents bsd.*.mk to function correctly.

Submitted by: bmah
Reviewed by: peter

23 years agoReport the correct inode number when panicing with freeing free inode.
Kirk McKusick [Wed, 21 Mar 2001 04:01:02 +0000 (04:01 +0000)]
Report the correct inode number when panicing with freeing free inode.
Report the correct block number when panicing with freeing free block.

23 years agoFor parallel SCSI, let us now do status with the final CTIO. For the 1080,
Matt Jacob [Wed, 21 Mar 2001 00:49:37 +0000 (00:49 +0000)]
For parallel SCSI, let us now do status with the final CTIO. For the 1080,
I was hanging after sending a xfer CTIO and a status CTIO for a non-discon
INQUIRY- the xfer CTIO was returned as completed OK, but the status CTIO
was dropped on the floor. All the fields looked good. I don't know why
it got dropped. But allowing status to go back with data xfer seemed to
work. I also noticed that with a non-disconnecting command that the
firmware handle in the ATIO is zero- this leads me to believe that the
f/w really can only handle one CTIO at a time in the discon case, and
it had no idea what to do with the second (status) CTIO.

23 years agoCheck CT2_SENDSTATUS/CT_SENDSTATUS against cto->ct_flags, not
Matt Jacob [Wed, 21 Mar 2001 00:46:44 +0000 (00:46 +0000)]
Check CT2_SENDSTATUS/CT_SENDSTATUS against cto->ct_flags, not
CAM_SEND_STATUS. Set a timeout of 2 seconds per CTIO. Make sure
that the 'real' tag value is being checked against- not the
one that also carries the firmware handle.

23 years agoNew release notes: fxp(4) needs "device miibus", pkg_create(1)
Bruce A. Mah [Tue, 20 Mar 2001 23:41:44 +0000 (23:41 +0000)]
New release notes:  fxp(4) needs "device miibus", pkg_create(1)
dependency order (submitted by sobomax), "lprm -" works for remote
printers, mergemaster(8) updates, ftpd(8) updates.

Also make sure ESS Maestro-3/Allegro is mentioned in appropriate
hardware lists (submitted by scottl).

23 years agoDocument situation around Adaptec SCSI cards more accurately. Esp.
Wilko Bulte [Tue, 20 Mar 2001 22:57:07 +0000 (22:57 +0000)]
Document situation around Adaptec SCSI cards more accurately. Esp.
for 164SX machines.

23 years agos/portmap/rpcbind
Alfred Perlstein [Tue, 20 Mar 2001 21:02:39 +0000 (21:02 +0000)]
s/portmap/rpcbind
Pointed out by: Hajimu UMEMOTO <ume@mahoroba.org>

23 years agowhitespace cleanup
Alfred Perlstein [Tue, 20 Mar 2001 20:28:09 +0000 (20:28 +0000)]
whitespace cleanup

23 years agoList libfetch.
Dag-Erling Smørgrav [Tue, 20 Mar 2001 20:21:08 +0000 (20:21 +0000)]
List libfetch.

Submitted by: bde

23 years agoUpdate copyright info, and make some slight cosmetic changes.
Coleman Kane [Tue, 20 Mar 2001 19:34:22 +0000 (19:34 +0000)]
Update copyright info, and make some slight cosmetic changes.

23 years agoChange Voodoo Graphics device ID to 0x0001121a so it gets probed.
Coleman Kane [Tue, 20 Mar 2001 19:30:46 +0000 (19:30 +0000)]
Change Voodoo Graphics device ID to 0x0001121a so it gets probed.

23 years agoSet the default manual section for usr.sbin/ to 8.
Ruslan Ermilov [Tue, 20 Mar 2001 18:17:26 +0000 (18:17 +0000)]
Set the default manual section for usr.sbin/ to 8.

23 years agoSet the default manual section for sbin/ to 8.
Ruslan Ermilov [Tue, 20 Mar 2001 18:13:31 +0000 (18:13 +0000)]
Set the default manual section for sbin/ to 8.

23 years agoPlace the call to pccard_insert_beep() in the inserted() timeout
Ian Dowse [Tue, 20 Mar 2001 18:10:19 +0000 (18:10 +0000)]
Place the call to pccard_insert_beep() in the inserted() timeout
routine instead of pccard_event(). This avoids spurious extra calls
to pccard_insert_beep() at insert or remove time which could occur
due to noise on the card-present lines.

Clean up some code in pccard_beep.c; we were depending on the order
of evaluation of function arguments, which is undefined in C. Also,
use `0' rather than `NULL' for integer values.

Reviewed by: sanpei, imp

23 years agoSet the default manual section for libexec/ to 8.
Ruslan Ermilov [Tue, 20 Mar 2001 18:10:13 +0000 (18:10 +0000)]
Set the default manual section for libexec/ to 8.

23 years agoSet the default manual section for games/ to 6.
Ruslan Ermilov [Tue, 20 Mar 2001 18:08:43 +0000 (18:08 +0000)]
Set the default manual section for games/ to 6.

23 years agoIf the MANSECT variable is specified, it will override the default
Ruslan Ermilov [Tue, 20 Mar 2001 18:06:09 +0000 (18:06 +0000)]
If the MANSECT variable is specified, it will override the default
manual section.  If, for example, MANSECT is set to 8, the default
MAN1=${PROG}.1 feature of bsd.prog.mk becomes MAN8=${PROG}.8.
Useful for games, libexec, sbin and usr.sbin subtrees.

Reviewed by: bde

23 years agoReboot(8) normally waits 5 seconds after sending SIGTERMs to all
Ian Dowse [Tue, 20 Mar 2001 17:22:48 +0000 (17:22 +0000)]
Reboot(8) normally waits 5 seconds after sending SIGTERMs to all
processes and then sends SIGKILLs. If a lot of processes are swapped
out, this delay may not be long enough, so processes such as an X
server may be killed before they have had time to clean up properly.

Make this delay more dynamic by waiting up to 60 seconds for swap
page-in activity to end. While I'm here, ANSIfy and remove a
`register' specifier.

23 years agoCorrect the top memory stats display: convert page counts to kB.
Thomas Moestl [Tue, 20 Mar 2001 16:02:16 +0000 (16:02 +0000)]
Correct the top memory stats display: convert page counts to kB.

Submitted by: Andrea Campi <andrea@webcom.it>
Approved by: rwatson
Pointy hat to: myself

23 years agoMake the pseudo-driver for "/dev/fd/*" handle fd's larger than 255.
Poul-Henning Kamp [Tue, 20 Mar 2001 13:26:13 +0000 (13:26 +0000)]
Make the pseudo-driver for "/dev/fd/*" handle fd's larger than 255.

PR: 25936

23 years agoAdd a KASSERT on unit2minor() so that we catch it if people try to pass
Poul-Henning Kamp [Tue, 20 Mar 2001 13:24:24 +0000 (13:24 +0000)]
Add a KASSERT on unit2minor() so that we catch it if people try to pass
us unit numbers which doesn't fit in 24 bits.

23 years agoFix a reference to the "vn" driver in a warning message.
Poul-Henning Kamp [Tue, 20 Mar 2001 12:31:53 +0000 (12:31 +0000)]
Fix a reference to the "vn" driver in a warning message.

23 years agomdoc(7) police: fixes to the previous revision:
Ruslan Ermilov [Tue, 20 Mar 2001 10:57:25 +0000 (10:57 +0000)]
mdoc(7) police: fixes to the previous revision:

- fixed bad formatting
- avoid using German
- removed hard sentence break

23 years agoFixed some typos.
Ruslan Ermilov [Tue, 20 Mar 2001 10:47:21 +0000 (10:47 +0000)]
Fixed some typos.