]> CyberLeo.Net >> Repos - FreeBSD/FreeBSD.git/log
FreeBSD/FreeBSD.git
24 years agoTypo fixes.
Yoshinobu Inoue [Sun, 12 Mar 2000 19:50:12 +0000 (19:50 +0000)]
Typo fixes.

PR: docs/17333
Submitted by: Kazuo Horikawa <horikawa@psinet.com>

24 years agoMerge from NetBSD. Addition of inet_ntop() and inet_pton() description.
Yoshinobu Inoue [Sun, 12 Mar 2000 19:38:22 +0000 (19:38 +0000)]
Merge from NetBSD. Addition of inet_ntop() and inet_pton() description.

Specified by: Robert Muir <rmuir@looksharp.net>

Obtained from: NetBSD

24 years agoCosmetic fix. Re-order MLINKS for if_indextoname.3 and inet.3 as alphabetical
Yoshinobu Inoue [Sun, 12 Mar 2000 19:29:52 +0000 (19:29 +0000)]
Cosmetic fix. Re-order MLINKS for if_indextoname.3 and inet.3 as alphabetical
order.

24 years agoAdd USA_RESIDENT to release make variables; can't hurt.
Jordan K. Hubbard [Sun, 12 Mar 2000 19:19:31 +0000 (19:19 +0000)]
Add USA_RESIDENT to release make variables; can't hurt.

Submitted by: Tatsumi Hosokawa <hosokawa@itc.keio.ac.jp>

24 years agoCorrect MLINKS contents for rcmd.3, because it is obsolete due to
Yoshinobu Inoue [Sun, 12 Mar 2000 19:12:03 +0000 (19:12 +0000)]
Correct MLINKS contents for rcmd.3, because it is obsolete due to
recent changes to rcmd.3.

  links to iruserok_af.3, ruserok_af.3 are removed.
  link to iruserok_sa.3 is added.

24 years agoImport from KAME. Advanced API related function descriptions.
Yoshinobu Inoue [Sun, 12 Mar 2000 18:45:49 +0000 (18:45 +0000)]
Import from KAME. Advanced API related function descriptions.

Obtained from: KAME project

24 years agoImport from KAME. Mainly, IPv6 sysctl mib values related update.
Yoshinobu Inoue [Sun, 12 Mar 2000 18:09:58 +0000 (18:09 +0000)]
Import from KAME. Mainly, IPv6 sysctl mib values related update.

Obtained from: KAME project

24 years agoMerge from KAME. Basically man doc improvement and contents fix.
Yoshinobu Inoue [Sun, 12 Mar 2000 16:37:25 +0000 (16:37 +0000)]
Merge from KAME. Basically man doc improvement and contents fix.

Obtained from: KAME project

24 years agoImport ip6 and icmp6 man from KAME.
Yoshinobu Inoue [Sun, 12 Mar 2000 16:29:51 +0000 (16:29 +0000)]
Import ip6 and icmp6 man from KAME.

Obtained from: KAME project

24 years agoRemove period at the end of Xr macro sentences.
Yoshinobu Inoue [Sun, 12 Mar 2000 16:10:13 +0000 (16:10 +0000)]
Remove period at the end of Xr macro sentences.
(merge from KAME repository)

Obtained from: KAME project

24 years agoTry harder to make the lower 16 bits of fsids unique. The vfs type
Bruce Evans [Sun, 12 Mar 2000 14:23:21 +0000 (14:23 +0000)]
Try harder to make the lower 16 bits of fsids unique.  The vfs type
number was packed very wastefully, giving perfect non-uniqeness in
the lower 16 bits of fsids for filesystems with the same vfs type.
This made linux_stat() return perfectly non-unique (broken) 16-bit
st_dev's for nfs mount points, and effectively reduced mntid_base to
8 bits so that the vfs_getnewfsid() looped endlessly when there are
already 256 mounted filesystems with the required vfs type.

Approved by: jkh

24 years agoDont call ad_start unless the controller is idle.
Søren Schmidt [Sun, 12 Mar 2000 13:56:05 +0000 (13:56 +0000)]
Dont call ad_start unless the controller is idle.
This effectively nullified the usefullness of disksort().

24 years agoDisable RTPRIO and P1003.1B scheduling in ntpd. Only leave BSD_NICE
Peter Dufault [Sun, 12 Mar 2000 13:25:14 +0000 (13:25 +0000)]
Disable RTPRIO and P1003.1B scheduling in ntpd.  Only leave BSD_NICE
for a scheduling boost.  This is a conservative change that should
make no difference in practice and eliminate concerns about this being
the source of some SMP hangs.

Configuration scripts should never auto-configure P1003.1B
without a second test.  The behavior with respect to regular time
sharing, who can access it, etc., is not defined.

Approved by: jkh

24 years agoMerged from sys/isa/sio.c revision 1.291.
KATO Takenori [Sun, 12 Mar 2000 13:14:51 +0000 (13:14 +0000)]
Merged from sys/isa/sio.c revision 1.291.

24 years agocomment out extra .Pp (which is necessary by comment-out of .Bl-.El)
Jun-ichiro itojun Hagino [Sun, 12 Mar 2000 12:54:33 +0000 (12:54 +0000)]
comment out extra .Pp (which is necessary by comment-out of .Bl-.El)

24 years agofix partially commented out dangling sentence.
Jun-ichiro itojun Hagino [Sun, 12 Mar 2000 12:52:14 +0000 (12:52 +0000)]
fix partially commented out dangling sentence.

24 years agosdio: If sd I/O fails because of an I/O error, set the sd state
Greg Lehey [Sun, 12 Mar 2000 05:05:41 +0000 (05:05 +0000)]
sdio: If sd I/O fails because of an I/O error, set the sd state
correctly.

Reported-by: phk
Approved-by: jkh

24 years agoRemove PkgInteractive hack - it won't work like this.
Jordan K. Hubbard [Sun, 12 Mar 2000 03:57:26 +0000 (03:57 +0000)]
Remove PkgInteractive hack - it won't work like this.

24 years agoRecognize newer NVidia graphics cards. (GeForce and Quadro families)
Chris Piazza [Sun, 12 Mar 2000 01:24:15 +0000 (01:24 +0000)]
Recognize newer NVidia graphics cards. (GeForce and Quadro families)

Approved by: jkh

24 years agoImport stf(6to4 interface) man from KAME.
Yoshinobu Inoue [Sat, 11 Mar 2000 22:55:32 +0000 (22:55 +0000)]
Import stf(6to4 interface) man from KAME.

Obtained from: KAME project

24 years agoRemove the last vestiges of libRSAglue now that it's an empty stub.
Kris Kennaway [Sat, 11 Mar 2000 22:34:10 +0000 (22:34 +0000)]
Remove the last vestiges of libRSAglue now that it's an empty stub.
This should fix the buildworld problems some people were seeing.

Approved by: jkh

24 years agoDisable IPv4 over IPv4 tunnel on the 6to4 interface for better security.
Yoshinobu Inoue [Sat, 11 Mar 2000 22:11:57 +0000 (22:11 +0000)]
Disable IPv4 over IPv4 tunnel on the 6to4 interface for better security.

Approved by: jkh

24 years agoAdd missing return.
Yoshinobu Inoue [Sat, 11 Mar 2000 20:57:31 +0000 (20:57 +0000)]
Add missing return.

  "ndp" command should deletes only neighbor cache entries, but the
  program lacks necessary "return" after the neighbor cache
  entry check, so it might deletes non neighbor cache entries.
  (it seems that usually no problem happens.)

Approved by: jkh
Reviewed by: ume

24 years agoClear sin6_scope_id before passing routes into kernel, becaues kernel
Yoshinobu Inoue [Sat, 11 Mar 2000 20:52:01 +0000 (20:52 +0000)]
Clear sin6_scope_id before passing routes into kernel, becaues kernel
don't support routes with sin6_scope_id set.
Without this fix, routes with IPv6 scoped addr won't work when it is
assigned by "route" command.

Approved by: jkh

Reviewed by: ume

24 years agoForbid include of netinet6/ip6.h from user-land, and if included,
Yoshinobu Inoue [Sat, 11 Mar 2000 20:44:53 +0000 (20:44 +0000)]
Forbid include of netinet6/ip6.h from user-land, and if included,
print an error message which say, "include netinet/ip6.h".
This is postponed to apply to avoid tcpdump compile error.
Now apply this because tcpdump has been already fixed.

Approved by: jkh

Obtained from: KAME project

24 years agoCorrect socket ethernet card entry.
Warner Losh [Sat, 11 Mar 2000 20:23:34 +0000 (20:23 +0000)]
Correct socket ethernet card entry.
Remove ^M that snuck into the last commit by mistake.

24 years agoFix crashes on card eject for pccard modems. We check for NULL when
Warner Losh [Sat, 11 Mar 2000 20:22:09 +0000 (20:22 +0000)]
Fix crashes on card eject for pccard modems.  We check for NULL when
we get the com address.  If so, we go ahead and return.  Bruce thinks
there's a bug in the pccard layer that it terminates devices with
extreme prejustice rather than letting them deside for themselves when
to terminate (and he's likely right).  This fix doesn't change that,
but instead works around it by checking for NULL pointers at more
places than before.

The detach routine still calls functions at interrupt level that
aren't reentrant.  In theory this could cause a problem, but none
showed up in practice.  Future versions should correct this problem,
likely by making the detach process a thread/process at the pccard
level.  NEWCARD will do this, and the current pccard layer should
likely be modified to that as well, should it live long enough.

A few style nits of the same form that were in my original patch sent
off to bde were also fixed as part of this process.  Mostly use of
!ptr and return ENOPARENS.

This should prevent a crash on suspend with an active ppp link as
well, but that wasn't tested.

Reviewed by: bde
Approved by: jkh

24 years agoAdd a flag to the usage display.
Yoshinobu Inoue [Sat, 11 Mar 2000 20:14:08 +0000 (20:14 +0000)]
Add a flag to the usage display.

  There was a missing description for a new flags to netstat.
  I already added the fix to netstat man, but usage() change is
  also necessary.

Specified by: Ben Smithurst <ben@scientia.demon.co.uk>
Approved by: jkh

24 years agoPrevent multicast packets from being sent over stf0 interface,
Yoshinobu Inoue [Sat, 11 Mar 2000 20:10:39 +0000 (20:10 +0000)]
Prevent multicast packets from being sent over stf0 interface,
to avoid error messages printed on /var/log/messages.

  Recently added "stf" interface has IFF_MULTICAST bit set,
  because IPv6 don't work on non IFF_MULTICAST set interfaces
  in current implementation.
  But "stf" interface is multi-dest IPv6 over IPv4 tunnel, so
  can't treat multicast packet with no special configuration.

  On the other hand, tools like "rtadvd", "pim6dd", "pim6sd"
  send messages on IFF_MULTICAST set interfaces by default.
  So there will be many bogus error message on /var/log/messages,
  that multicast packets sent to "stf0" failed.

  So,

   -strip "stf0" from rtadvd sending interfaces in rc.network6
    ("rtadvd" is a daemon used only for local subnet, so it
     will never need to send its packets over "stf0" interfaces.)

   -Add default configuration files for "pim6dd" and "pim6sd",
    and disable "stf0" in those files.

Approved by: jkh

24 years agoReplace m_pkthdr.rcvif with oif when oif is not NULL, to count
Yoshinobu Inoue [Sat, 11 Mar 2000 20:03:22 +0000 (20:03 +0000)]
Replace m_pkthdr.rcvif with oif when oif is not NULL, to count
icmp6 error statistics based on sending interface.
This also prevent kernel panic when rcvif is not initialized after M_PKTHDR().
(The initialization issue also need to be fixed in the future.)

Approved by: jkh

Submitted by: k-sugyou@kame.net

24 years agoFix sockaddr_storage related macro definition, as ss_family member type change.
Yoshinobu Inoue [Sat, 11 Mar 2000 19:51:04 +0000 (19:51 +0000)]
Fix sockaddr_storage related macro definition, as ss_family member type change.
(Currently, no effect but for future portability)

Approved by: jkh

Reviewed by: bde

24 years agoNote FPU exceptions default change
Martin Cracauer [Sat, 11 Mar 2000 18:31:21 +0000 (18:31 +0000)]
Note FPU exceptions default change

24 years agoAdd "-l" option description.
Yoshinobu Inoue [Sat, 11 Mar 2000 14:33:13 +0000 (14:33 +0000)]
Add "-l" option description.

Noticed the necessity by Ben Smithurst <ben@scientia.demon.co.uk> message
on freebsd-current.

24 years agoRemove old hand crafted version of start_join_net for download.
Duncan Barclay [Sat, 11 Mar 2000 13:34:13 +0000 (13:34 +0000)]
Remove old hand crafted version of start_join_net for download.

24 years agoStart to use ray_start_join_net (my old code removed on next checkin)
Duncan Barclay [Sat, 11 Mar 2000 13:25:33 +0000 (13:25 +0000)]
Start to use ray_start_join_net (my old code removed on next checkin)
requires current and desired network parameters, so I have created a network parameter structure and used two on them in the softc. This gives a bit of room of expansion. Re-wrote the update check in _net to work with this.

Tidyed up softc due to above.

Removed startccs and startcmd as they were only used by NetBSDs startup timeout.

24 years agoMake inetd compilable without INET6.
Yoshinobu Inoue [Sat, 11 Mar 2000 11:28:08 +0000 (11:28 +0000)]
Make inetd compilable without INET6.

Approved by: jkh

Submitted by: jhb

24 years agoPrevent sockstat print out some empty entries related to IPv6.
Yoshinobu Inoue [Sat, 11 Mar 2000 11:25:00 +0000 (11:25 +0000)]
Prevent sockstat print out some empty entries related to IPv6.

  Sockstat use netstat tcp/udp socket print, and fstat tcp/udp
  socket print, but it just specify all of IPPROTO_IP and
  IPPROTO_IPV6 sockets for fstat.
  Now IPv6 socket also use raw socket, but only netstat don't print it,
  so now they are printed as empty entries in sockstat output.

Approved by: jkh

24 years ago This is null commit to add correct comment for the last fix to this file.
Yoshinobu Inoue [Sat, 11 Mar 2000 11:20:52 +0000 (11:20 +0000)]
  This is null commit to add correct comment for the last fix to this file.
  (Last commit message is not related to this file. This file is mistakenly
  committed with last commit of other files.)

Prevent possible kernel panic at tcp_input().

   When some IPv4 apps do setsockopt() to control message receipt,
   then kernel will panic at control message creation processing.
   (Currently, only bgp daemon in KAME package known to cause the problem.)

Submitted by: jinmei@kame.net

24 years agoIPv6 6to4 support.
Yoshinobu Inoue [Sat, 11 Mar 2000 11:17:24 +0000 (11:17 +0000)]
IPv6 6to4 support.

   Now most big problem of IPv6 is getting IPv6 address
   assignment.
   6to4 solve the problem. 6to4 addr is defined like below,

          2002: 4byte v4 addr : 2byte SLA ID : 8byte interface ID

   The most important point of the address format is that an IPv4 addr
   is embeded in it. So any user who has IPv4 addr can get IPv6 address
   block with 2byte subnet space. Also, the IPv4 addr is used for
   semi-automatic IPv6 over IPv4 tunneling.

   With 6to4, getting IPv6 addr become dramatically easy.
   The attached patch enable 6to4 extension, and confirmed to work,
   between "Richard Seaman, Jr." <dick@tar.com> and me.

Approved by: jkh

Reviewed by: itojun

24 years agoOne more patch for the atkbd driver. It will make sure that the
Kazutaka YOKOTA [Sat, 11 Mar 2000 07:44:10 +0000 (07:44 +0000)]
One more patch for the atkbd driver.  It will make sure that the
keyboard port and interrupt is enabled and the driver is attached even
when the keyboard itself is not present when the system is booting.
(This has been the behavior through out 2.X and 3.X, but is somehow
broken in 4.0.)

# I certainly don't recommend people to `hot-plug' the AT keyboard,
# because the interface isn't designed for hot-plugging and such act
# will often break the keyboard controller.  But, so many people want to
# do that anyway...

Approved by: jkh

24 years agoTeach the 'dc' driver how to pick up settings left over by the
Mike Smith [Sat, 11 Mar 2000 05:20:56 +0000 (05:20 +0000)]
Teach the 'dc' driver how to pick up settings left over by the
SRM on alpha systems.  This is an expedient if not entirely
elegant solution to the problem.

Submitted by: gallatin
Approved by: jkh

24 years agoThe function arpintr() incorrectly checks m->m_len to detect incomplete
Robert Watson [Sat, 11 Mar 2000 00:24:29 +0000 (00:24 +0000)]
The function arpintr() incorrectly checks m->m_len to detect incomplete
ARP packets. This can incorrectly reject complete frames since the frame
could be stored in more than one mbuf.

The following patches fix the length comparisson, and add several
diagnostic log messages to the interrupt handler for out-of-the-norm ARP
packets. This should make ARP problems easier to detect, diagnose and
fix.

Submitted by: C. Stephen Gunn <csg@waterspout.com>
Approved by: jkh
Reviewed by: rwatson

24 years agoAdd COM1 SA MC220
Warner Losh [Fri, 10 Mar 2000 23:17:51 +0000 (23:17 +0000)]
Add COM1 SA MC220

Submitted by: Michael Reifenberger <root@nihil.plaut.de>

24 years agoFix fd leak in libdisk's Disk_Names()
Warner Losh [Fri, 10 Mar 2000 22:54:04 +0000 (22:54 +0000)]
Fix fd leak in libdisk's Disk_Names()

24 years agoCosmetics..
Søren Schmidt [Fri, 10 Mar 2000 22:10:34 +0000 (22:10 +0000)]
Cosmetics..

24 years ago-e -> -I change.
Warner Losh [Fri, 10 Mar 2000 22:03:00 +0000 (22:03 +0000)]
-e -> -I change.
-s for sumary

Approved by: jkh

24 years agoCorrect man page title.
Archie Cobbs [Fri, 10 Mar 2000 20:38:24 +0000 (20:38 +0000)]
Correct man page title.

24 years agoDo the right thing with USA_RESIDENT
Jordan K. Hubbard [Fri, 10 Mar 2000 20:34:22 +0000 (20:34 +0000)]
Do the right thing with USA_RESIDENT

24 years agos/chose/choose/ in all the relevant places.
Jordan K. Hubbard [Fri, 10 Mar 2000 19:53:53 +0000 (19:53 +0000)]
s/chose/choose/ in all the relevant places.

Noticed by: John Reynolds <jjreynold@home.com>

24 years agoAdded notes about:
Tatsumi Hosokawa [Fri, 10 Mar 2000 18:59:03 +0000 (18:59 +0000)]
Added notes about:
(1) ed and ep PC-card drivers.
(2) sn driver and Megahertz X-Jack Ethernet
(3) Melco Airconnect under wi driver

24 years agoChange the default FPU control word so that exceptions for new
Martin Cracauer [Fri, 10 Mar 2000 17:56:33 +0000 (17:56 +0000)]
Change the default FPU control word so that exceptions for new
processes are now masked until set by fpsetmask(3).

Submitted by: bde
Approved by: jkh, bde

24 years agoFix whitepace in the arguments of two Xr macros.
Sheldon Hearn [Fri, 10 Mar 2000 14:46:30 +0000 (14:46 +0000)]
Fix whitepace in the arguments of two Xr macros.

24 years ago- typos
Nik Clayton [Fri, 10 Mar 2000 11:48:49 +0000 (11:48 +0000)]
- typos
- Add double spaces following full stops to improve typeset output
- mdoc-ification.  (Though I'm uncertain whether option values and
  contents should be .Dq or something else).
- Fix a missed /etc/ssh change
- Expand wording on RandomSeed and behaviour when X11 isn't forwarded.
- Change examples to literal mode.
- Trim trailing whitespace

PR: docs/17292
Submitted by: Peter Jeremy <peter.jeremy@alcatel.com.au>

24 years agoMerged from sys/isa/sio.c revisions 1.289 and 1.290.
KATO Takenori [Fri, 10 Mar 2000 10:34:36 +0000 (10:34 +0000)]
Merged from sys/isa/sio.c revisions 1.289 and 1.290.

24 years agoshmat: If VM_PROT_READ_IS_EXEC is defined and prot includes VM_PROT_READ,
Alan Cox [Fri, 10 Mar 2000 09:11:24 +0000 (09:11 +0000)]
shmat: If VM_PROT_READ_IS_EXEC is defined and prot includes VM_PROT_READ,
VM_PROT_EXECUTE must be added to prot before calling vm_map_find.

Without this change, an mprotect on a shmat'ed region fails (when
it shouldn't).  This bug was reported Feb 28 by Brooks Davis
<brooks@one-eyed-alien.net> on -hackers.

Reviewed by: bde
Approved by: jkh

24 years agoDont call make_dev until we are sure the device is there, this prevents
Søren Schmidt [Fri, 10 Mar 2000 08:09:53 +0000 (08:09 +0000)]
Dont call make_dev until we are sure the device is there, this prevents
panic'ing when the device is tried accessed later.

24 years agoAdd support for PnP ISA cards, and some laptop PnP support:
Warner Losh [Fri, 10 Mar 2000 05:43:29 +0000 (05:43 +0000)]
Add support for PnP ISA cards, and some laptop PnP support:

o break out some of the probe routine the allocation of resources
  into an attach routine
o Recognize PnP ids
o Allocate IRQ per card rather than per system
o Better polling reporting
o Remove unneeded include files in slot.h
o store a pseudo unit number on each device we find.
o Pass a unit number to interrupt/timeout routine and use it for polling
  the hardware.

Tested on: My VAIO and with the Linksys pccard reader.
Approved by: jkh

24 years agoCorrect PnP ids for pcic-like cards.
Warner Losh [Fri, 10 Mar 2000 05:26:32 +0000 (05:26 +0000)]
Correct PnP ids for pcic-like cards.

Approved by: jkh

24 years agoMinor tweaks to Xircom entries. With this I can now attach to the
Warner Losh [Fri, 10 Mar 2000 00:43:49 +0000 (00:43 +0000)]
Minor tweaks to Xircom entries.  With this I can now attach to the
modem side of my Xircom CreditCard Ethernet+Modem 33.6.  Eliminate bad
Xircom entry.  Complete IBM Push/Pop Modem(14.4 -> Push/Pop
Modem(14.4K).

Approved in concept by: jkh

24 years agoFix typo.
Greg Lehey [Thu, 9 Mar 2000 22:53:13 +0000 (22:53 +0000)]
Fix typo.

24 years agoAdd in IPV4 NIS support.
Jonathan Lemon [Thu, 9 Mar 2000 22:52:30 +0000 (22:52 +0000)]
Add in IPV4 NIS support.

PR: 17290 (but not the same patch)
Approved by: jkh

24 years agoUse the ccs update command mechanism for dowloading parameters on startup
Duncan Barclay [Thu, 9 Mar 2000 21:34:47 +0000 (21:34 +0000)]
Use the ccs update command mechanism for dowloading parameters on startup

24 years agoFix a problem which caused i/o resource allocation conflict of this
Hellmuth Michaelis [Thu, 9 Mar 2000 20:57:47 +0000 (20:57 +0000)]
Fix a problem which caused i/o resource allocation conflict of this
particular ISDN card with the 2nd IDE controller and/or other cards
depending on configuration.

The amount of i/o ports allocated are reduced from 0x40 per range to
0x20 per range.

Submitted by: Andrew Gordon <arg@arg1.demon.co.uk>
Approved by: jkh

24 years agoDon't use the old raw name for disk devices.
David E. O'Brien [Thu, 9 Mar 2000 19:35:32 +0000 (19:35 +0000)]
Don't use the old raw name for disk devices.

24 years agoIntroduce ethernet bridge support for if_dc
Robert Watson [Thu, 9 Mar 2000 19:28:19 +0000 (19:28 +0000)]
Introduce ethernet bridge support for if_dc

Approved by: jkh

24 years ago In the 'found' case for ufs_lookup() the underlying bp's data was
Matthew Dillon [Thu, 9 Mar 2000 18:54:59 +0000 (18:54 +0000)]
In the 'found' case for ufs_lookup() the underlying bp's data was
    being accessed after the bp had been releaed.  A simple move of the
    brelse() solves the problem.

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

24 years agoFix various unsigned vs signed errors that caused problems with uids
Paul Richards [Thu, 9 Mar 2000 18:11:16 +0000 (18:11 +0000)]
Fix various unsigned vs signed errors that caused problems with uids
and gids bigger than 16 bits. Added checks for uids and gids that are
bigger than 32 bits.

Approved by: jkh (partly, this fix is bigger than I first intended)

24 years agoFix bug in linux_wait4 and linux_waitpid where garbage in the status
Marcel Moolenaar [Thu, 9 Mar 2000 17:52:01 +0000 (17:52 +0000)]
Fix bug in linux_wait4 and linux_waitpid where garbage in the status
argument could panic the kernel.

Submitted by: Ian Dowse <iedowse@maths.tcd.ie>
Prompted by: jkh, gallatin
Approved by: prompters

24 years agoMore grammer, wording, and mdoc fixes.
Yoshinobu Inoue [Thu, 9 Mar 2000 16:41:27 +0000 (16:41 +0000)]
More grammer, wording, and mdoc fixes.

Submitted by: bde
Reviewed by: sheldonh

24 years agoAdd vendor/device ID for Corega FEther USB-TX.
Masahide MAEKAWA [Thu, 9 Mar 2000 16:35:21 +0000 (16:35 +0000)]
Add vendor/device ID for Corega FEther USB-TX.

Submitted by: iwasaki
Approved by: jkh

24 years agoAdd Compaq `ida' driver to GENERIC, update it's LINT entry.
Jonathan Lemon [Thu, 9 Mar 2000 16:32:56 +0000 (16:32 +0000)]
Add Compaq `ida' driver to GENERIC, update it's LINT entry.

Approved by: jordan

24 years agoRegen.
Masahide MAEKAWA [Thu, 9 Mar 2000 16:29:49 +0000 (16:29 +0000)]
Regen.

24 years agoAdd Corega FEther USB-TX.
Masahide MAEKAWA [Thu, 9 Mar 2000 16:28:58 +0000 (16:28 +0000)]
Add Corega FEther USB-TX.

Submitted by: iwasaki
Approved by: jkh

24 years agoChange default of rtadvd to "YES".
Yoshinobu Inoue [Thu, 9 Mar 2000 15:19:58 +0000 (15:19 +0000)]
Change default of rtadvd to "YES".
Also add IPv6 example for "ifconfig_ifname_alias".

Suggested by: bmah@CA.Sandia.GOV

24 years agoEnable SCM_RIGHTS on alpha. Allocate necessary buffer as conversion between
Yoshinobu Inoue [Thu, 9 Mar 2000 15:15:27 +0000 (15:15 +0000)]
Enable SCM_RIGHTS on alpha. Allocate necessary buffer as conversion between
int and struct file *.

Approved by: jkh

Submitted by: brian
Reviewed by: bde, brian, peter

24 years agoImplement Linux BLKGETSIZE ioctl, and open the door to implementing
Nick Sayer [Thu, 9 Mar 2000 15:14:14 +0000 (15:14 +0000)]
Implement Linux BLKGETSIZE ioctl, and open the door to implementing
other BLK.* ioctls should the desire arize.

Approved by: jkh (via dufault)

24 years agoFix addr length argument value passed to sendto().
Yoshinobu Inoue [Thu, 9 Mar 2000 15:07:38 +0000 (15:07 +0000)]
Fix addr length argument value passed to sendto().
Some inetd internal udp servers didn't worked with problem.
Also fix recvfrom() "fromlen" arg type from int * to socklen_t *.

Approved by: jkh

Submitted by: bde

24 years agoFix pointer addition bug (3 -> 4).
Jonathan Lemon [Thu, 9 Mar 2000 15:05:20 +0000 (15:05 +0000)]
Fix pointer addition bug (3 -> 4).

Submitted by:  david.w.james@bt.com

24 years agoInitialize mbuf pointer at getting ipsec policy.
Yoshinobu Inoue [Thu, 9 Mar 2000 14:57:16 +0000 (14:57 +0000)]
Initialize mbuf pointer at getting ipsec policy.
Without this, kernel will panic at getsockopt() of IPSEC_POLICY.
Also make compilable libipsec/test-policy.c which tries getsockopt() of
IPSEC_POLICY.

Approved by: jkh

Submitted by: sakane@kame.net

24 years agoMake LOGIN_CAP work properly.
Mark Murray [Thu, 9 Mar 2000 14:54:00 +0000 (14:54 +0000)]
Make LOGIN_CAP work properly.

Submitted by: ache

24 years agoMake LOGIN_CAP work properly.
Mark Murray [Thu, 9 Mar 2000 14:52:31 +0000 (14:52 +0000)]
Make LOGIN_CAP work properly.

24 years agoAdded several socket close() which were mistakenly forgotten.
Yoshinobu Inoue [Thu, 9 Mar 2000 14:51:05 +0000 (14:51 +0000)]
Added several socket close() which were mistakenly forgotten.

Reported by: Hidetoshi Shimokawa <simokawa@sat.t.u-tokyo.ac.jp>

24 years agoUpdate icmp node info query message bit order of query types,
Yoshinobu Inoue [Thu, 9 Mar 2000 14:47:21 +0000 (14:47 +0000)]
Update icmp node info query message bit order of query types,
according to draft-ietf-ipngwg-icmp-name-lookups-04 to 05 change.
This is necessary before 4.0, because,
  -This change is non backword compatible
  -Other KAME derived platforms applied 05
  -Author of the draft said he never do backword imcompatible changes
   again.

Approved by: jkh

Obtained from: KAME project

24 years agoFixed a null pointer panic for dumpon(8) on a nonexistent device whose
Bruce Evans [Thu, 9 Mar 2000 12:40:41 +0000 (12:40 +0000)]
Fixed a null pointer panic for dumpon(8) on a nonexistent device whose
driver uses the new disk layer.

Reviewed by: phk
Approved by: jkh

24 years agoMerge conflicts.
David E. O'Brien [Thu, 9 Mar 2000 10:11:08 +0000 (10:11 +0000)]
Merge conflicts.

24 years agoMark up the optional -c option with the Op macro.
Sheldon Hearn [Thu, 9 Mar 2000 09:27:40 +0000 (09:27 +0000)]
Mark up the optional -c option with the Op macro.

PR: 17271
Submitted by: David Malone <dwmalone@maths.tcd.ie>

24 years agoThis commit was generated by cvs2svn to compensate for changes in r57844,
David E. O'Brien [Thu, 9 Mar 2000 09:21:46 +0000 (09:21 +0000)]
This commit was generated by cvs2svn to compensate for changes in r57844,
which included commits to RCS files with non-trunk default branches.

24 years agoBring in bug fixes from the GCC anoncvs server's "gcc-2_95-branch"
David E. O'Brien [Thu, 9 Mar 2000 09:21:46 +0000 (09:21 +0000)]
Bring in bug fixes from the GCC anoncvs server's "gcc-2_95-branch"
branch as of March 7th, 2000.

24 years agoAdd blurb about perl modules needing recompiling.
Warner Losh [Thu, 9 Mar 2000 07:34:46 +0000 (07:34 +0000)]
Add blurb about perl modules needing recompiling.

24 years agoAdd comment suggesting that people send new entries to me. I've been
Warner Losh [Thu, 9 Mar 2000 07:30:12 +0000 (07:30 +0000)]
Add comment suggesting that people send new entries to me.  I've been
coordinating things for a while.

24 years agoBuildworld fixes for NO_OPENSSH and NO_OPENSSL
Kris Kennaway [Thu, 9 Mar 2000 06:29:05 +0000 (06:29 +0000)]
Buildworld fixes for NO_OPENSSH and NO_OPENSSL

Approved by: jkh

24 years agoDon't forget to check for unsupported features when updating. It was
Bruce Evans [Thu, 9 Mar 2000 05:21:10 +0000 (05:21 +0000)]
Don't forget to check for unsupported features when updating.  It was
possible to defeat the check for rw incompatibilty by mounting ro and
updating to rw.

Approved by: jkh

24 years agoAdd the logical ID of Avance Logic ALS100. The vendor ID is also
Seigo Tanimura [Thu, 9 Mar 2000 02:13:21 +0000 (02:13 +0000)]
Add the logical ID of Avance Logic ALS100. The vendor ID is also
checked, which was introduced in 1.18.

PR: kern/16487
Submitted by: Oliver Fromme <oliver.fromme@heim3.tu-clausthal.de>
Approved by: jkh

24 years agoRemove any mention of PCI support relating to userconfig.
Mike Smith [Thu, 9 Mar 2000 01:29:42 +0000 (01:29 +0000)]
Remove any mention of PCI support relating to userconfig.

Remove the comment about Mylex RAID controllers, as this has been fixed.

Approved by: jkh (partly)

24 years agoRemove the PCI device class and all the PCI-only devices. With the newbus
Mike Smith [Thu, 9 Mar 2000 01:27:55 +0000 (01:27 +0000)]
Remove the PCI device class and all the PCI-only devices.  With the newbus
conversion this no longer works, and userconfig is not expected to ever
again be used for these devices.

Approved by: jkh

24 years agoMore new/corrected devices:
Warner Losh [Thu, 9 Mar 2000 00:35:38 +0000 (00:35 +0000)]
More new/corrected devices:
o Correct Entry for Intel EtherExpress (jkoshy)
o Add Option International 56K modem (markm)
o Socket LP-E low power WinCE ethernet card (imp)

Correction approved by: jkh
Adding devices approved in principle by: jkh

24 years agoMega additions of NetBSD code - most of the NetBSD code is now in here, if
Duncan Barclay [Wed, 8 Mar 2000 23:28:06 +0000 (23:28 +0000)]
Mega additions of NetBSD code - most of the NetBSD code is now in here, if
not actually used yet. This created a lot of prototypes.

I moved code around so that the functions are in the same order as NetBSD - this helps diff a lot.

Things that are missing are the TX routine and ifmedia stuff and ioctls

Not all of the stuff is enabled - we are missing calls in ccs_done and ioctl.

Promiscious mode is working.

Am almost ready to use the NetBSD start up routine - essentially all I need
to do is not use a seperate timer to call download_timo.

Other misc. things:
callout_stop is a newish feature for cancelling a timer without
argument checking it.

tried to add a sysctl knob but it doesn't work in the module

enabled infrastructure code to call netbsd bits.

24 years agoCosmetic fix: "id" -> "idad" for devstat.
Jonathan Lemon [Wed, 8 Mar 2000 21:10:15 +0000 (21:10 +0000)]
Cosmetic fix: "id" -> "idad" for devstat.

Missed in prior (approved) commit by: jlemon

24 years agoAdd Compaq Arrays
Jonathan Lemon [Wed, 8 Mar 2000 18:14:19 +0000 (18:14 +0000)]
Add Compaq Arrays

24 years agoCorrect device naming for IDA disk nodes (idad)
Jonathan Lemon [Wed, 8 Mar 2000 18:14:02 +0000 (18:14 +0000)]
Correct device naming for IDA disk nodes (idad)