]> CyberLeo.Net >> Repos - FreeBSD/FreeBSD.git/log
FreeBSD/FreeBSD.git
22 years agoFill in my MAINTAINER info for drm
Eric Anholt [Mon, 17 Jun 2002 08:58:29 +0000 (08:58 +0000)]
Fill in my MAINTAINER info for drm

Approved by: des

22 years agoKill superfluous blank line.
Maxim Sobolev [Mon, 17 Jun 2002 08:21:53 +0000 (08:21 +0000)]
Kill superfluous blank line.

22 years ago- Turn a DIAGNOSTIC check into a KASSERT(9).
Andrew R. Reiter [Mon, 17 Jun 2002 06:10:17 +0000 (06:10 +0000)]
- Turn a DIAGNOSTIC check into a KASSERT(9).

22 years ago- Turn a DIAGNOSTIC check into a KASSERT(9) macro.
Andrew R. Reiter [Mon, 17 Jun 2002 06:00:20 +0000 (06:00 +0000)]
- Turn a DIAGNOSTIC check into a KASSERT(9) macro.

22 years ago- Explicitly pass M_WAITOK in a few calls to uma_zalloc(9).
Andrew R. Reiter [Mon, 17 Jun 2002 05:57:37 +0000 (05:57 +0000)]
- Explicitly pass M_WAITOK in a few calls to uma_zalloc(9).

22 years agosppp needs slcompress.c nowadays.
John Hay [Mon, 17 Jun 2002 05:40:49 +0000 (05:40 +0000)]
sppp needs slcompress.c nowadays.

PR: 39369

22 years agoActually document pselect(3) so that Bruce can mention it in the release
Garrett Wollman [Mon, 17 Jun 2002 02:21:17 +0000 (02:21 +0000)]
Actually document pselect(3) so that Bruce can mention it in the release
notes. :-)

22 years agoMove dillon's time conversion functions to a new header <timeconv.h>.
Garrett Wollman [Mon, 17 Jun 2002 01:42:33 +0000 (01:42 +0000)]
Move dillon's time conversion functions to a new header <timeconv.h>.
Since they were never documented and have never appeared in a FreeBSD
release, no repo-copy of the header is done.  This removes namespace
pollution from <time.h>.

22 years agoClean up a bit of namespace pollution which crept in with the last rev.
Garrett Wollman [Mon, 17 Jun 2002 01:31:25 +0000 (01:31 +0000)]
Clean up a bit of namespace pollution which crept in with the last rev.

Submitted by: bde

22 years agoCompensate for dodgy Win98/WinME MSCHAPv2 responses later in the code
Brian Somers [Mon, 17 Jun 2002 01:12:38 +0000 (01:12 +0000)]
Compensate for dodgy Win98/WinME MSCHAPv2 responses later in the code
path... after we've talked to any RADIUS servers involved, so that we
haven't touched the data before it gets to the server.

Make it clearer in the code that this compensation is done by setting
a flag to a value of zero, a flag which rfc2759 says *MUST* be zero.

While we're here, don't bother passing the peer challenge into
radius_Authenticate().  It's already part of the key we're passing in
(this becomes obvious now that I've structured that data...).

This ``fix'' doesn't help to authenticate Win98/WinME users in my test
environment as ports/net/freeradius seems to ignore the flag
completely anyway, but it may help with other RADIUS servers.

22 years agoRemove 17 June as German national holiday.
Greg Lehey [Sun, 16 Jun 2002 22:59:21 +0000 (22:59 +0000)]
Remove 17 June as German national holiday.

22 years agoRemove <sys/select.h> to unbreak the build. I'm not sure if this affects
Dag-Erling Smørgrav [Sun, 16 Jun 2002 21:50:42 +0000 (21:50 +0000)]
Remove <sys/select.h> to unbreak the build.  I'm not sure if this affects
the functionality of the drm module, but hopefully it will tide us over
until Eric can take a look at it.

22 years agoFix two syntax errors and add declarations of statvfs() and fstatvfs().
Garrett Wollman [Sun, 16 Jun 2002 21:02:08 +0000 (21:02 +0000)]
Fix two syntax errors and add declarations of statvfs() and fstatvfs().

22 years agoTry handling lack of syscons (for diskless boot, or other) by
Doug Barton [Sun, 16 Jun 2002 20:59:16 +0000 (20:59 +0000)]
Try handling lack of syscons (for diskless boot, or other) by
testing for the presence of /dev/ttyv0.

PR: misc/39351
Submitted by: Dirk-Willem van Gulik <dirkx@covalent.net>

22 years agoRewrite the probe routine. This makes adding drives and quirks more easy.
Nick Hibma [Sun, 16 Jun 2002 20:53:35 +0000 (20:53 +0000)]
Rewrite the probe routine. This makes adding drives and quirks more easy.

While there change a few names to less confusing ones.

22 years ago o Remove GIANT_REQUIRED from vm_fault_user_wire().
Alan Cox [Sun, 16 Jun 2002 20:42:29 +0000 (20:42 +0000)]
 o Remove GIANT_REQUIRED from vm_fault_user_wire().
 o Move pmap_pageable() outside of Giant in vm_fault_unwire().
   (pmap_pageable() is a no-op on all supported architectures.)
 o Remove the acquisition and release of Giant from mlock().

22 years agoNow that we have a separate header file for sigset_t, use it and avoid
Garrett Wollman [Sun, 16 Jun 2002 18:40:16 +0000 (18:40 +0000)]
Now that we have a separate header file for sigset_t, use it and avoid
the full pollution of <signal.h>.

22 years agoUse <sys/_sigset.h> to get declaration of sigset_t, which has been moved
Garrett Wollman [Sun, 16 Jun 2002 18:35:24 +0000 (18:35 +0000)]
Use <sys/_sigset.h> to get declaration of sigset_t, which has been moved
to a separate header to facilitate its declaration in more than one
place.  Namespace issues not fixed.

22 years agoDelete everything except the sigset_t definitions (subsequent to
Garrett Wollman [Sun, 16 Jun 2002 18:33:59 +0000 (18:33 +0000)]
Delete everything except the sigset_t definitions (subsequent to
repo-copy).

22 years agoDon't try to build libssh if NO_OPENSSL is defined because NO_OPENSSL implies
Maxim Sobolev [Sun, 16 Jun 2002 14:24:51 +0000 (14:24 +0000)]
Don't try to build libssh if NO_OPENSSL is defined because NO_OPENSSL implies
NO_OPENSSH.

22 years agoOur strerror_r(3) returns int, not char *. This fixes bogus error messages
Maxim Sobolev [Sun, 16 Jun 2002 12:35:17 +0000 (12:35 +0000)]
Our strerror_r(3) returns int, not char *. This fixes bogus error messages
like "Cannot stat: (null)"

Reported by:    bde, kris

22 years agoRemoved a duplicate -ffreestanding. It's already set in bsd.kern.mk.
Maxime Henrion [Sun, 16 Jun 2002 10:42:05 +0000 (10:42 +0000)]
Removed a duplicate -ffreestanding.  It's already set in bsd.kern.mk.

Approved by: bde

22 years agoCorrect the handling of (for example) the N command by only zeroing the
Tim J. Robbins [Sun, 16 Jun 2002 08:44:39 +0000 (08:44 +0000)]
Correct the handling of (for example) the N command by only zeroing the
input space in mf_fgets() if we reach the end of all input files.

22 years agoRestore local bits lost in recent merge from NetBSD.
Matthew N. Dodd [Sun, 16 Jun 2002 08:29:35 +0000 (08:29 +0000)]
Restore local bits lost in recent merge from NetBSD.

22 years agoBotcho. I included in the last commit the
Matt Jacob [Sun, 16 Jun 2002 06:48:42 +0000 (06:48 +0000)]
Botcho. I included in the last commit the

device mpt # LSI-Logic MPT-Fusion

line. This driver isn't ready yet.

Submitted by: Scott Long

22 years agoWhen counting words, check the correct character variable to see whether it's
Tim J. Robbins [Sun, 16 Jun 2002 06:04:43 +0000 (06:04 +0000)]
When counting words, check the correct character variable to see whether it's
a space or not.

Noticed by: bde

22 years agoIf the HBA is already 'touched', still set maxluns. Othewise for
Matt Jacob [Sun, 16 Jun 2002 05:18:22 +0000 (05:18 +0000)]
If the HBA is already 'touched', still set maxluns. Othewise for
CAM_QUIRK_HILUN devices we loop thru 32bits of lun. Oops.

Switch to using USEC_DELAY rather than USEC_SLEEP at isp_reset time.

Try to paper around a defect in clients that don't correctly registers
themeselves with the fabric nameserver.

Minor updates for Mirapoint support- they still use code that is not
HANDLE_LOOPSTATE_IN_OUTER_LAYERS, and, surprise surprise, this old
stuff had some bugs in it.

Clean up some target mode stuff.

MFC after: 1 week

22 years agoAdd support for ISP_FC_GETHINFO, which returns current connection
Matt Jacob [Sun, 16 Jun 2002 05:08:02 +0000 (05:08 +0000)]
Add support for ISP_FC_GETHINFO, which returns current connection
topology, speed, loopid, WWPN/WWNN, etc.

Beef up target mode. Add isp_handle_platform_notify_scsi and
isp_handle_platform_notify_fc platform handlers to handle immediate
notifies (isp_handle_platform_notify_scsi is still stubbed out).

In implementation of isp_handle_platform_notify_fc, for IN_ABORT_TASK,
peel off a pending XPT_IMMED_NOTIFY and call xpt_done on it and hope
that somebody upstream is listening.

Make sure on final CTIO2s that we set residual correctly. These are
absolutely crucial. Make sure we set relative offset for each CTIO2
based upon bytes we've already xferred. This is what the private
adjunct datat to the original ATIO is. Note state of command so
we can figure out where to find it if we get an ABORT from the firmware.

Make sure we *always* set CAM_TAG_ACTION_VALID for ATIO2s. Make sure
we keep track of the original lun.

If se sent status (or we're otherwise done with the command), don't
forget to free the adjunct structure.

22 years agoExtend private adjunct to ATIO to have both tag lun, and extended state
Matt Jacob [Sun, 16 Jun 2002 05:02:25 +0000 (05:02 +0000)]
Extend private adjunct to ATIO to have both tag lun, and extended state
(so we can, when things get lost, find out who currently is processing
on behalf of this open exchange. Invariably, when things are lost and
wedged, it's CAM).

Keep an atio resource counter locally.

MFC after: 1 week

22 years agoForce commit (last CVS comment was wrong).
Matt Jacob [Sun, 16 Jun 2002 05:00:20 +0000 (05:00 +0000)]
Force commit (last CVS comment was wrong).

Go back to *not* fully evaluating loop/fabric state if our role
is ISP_ROLE_NONE.

22 years agoAdd ISP_FC_GETHINFO ioctl.
Matt Jacob [Sun, 16 Jun 2002 04:59:30 +0000 (04:59 +0000)]
Add ISP_FC_GETHINFO ioctl.

MFC after: 1 week

22 years agoSet all 23XX cards as 'touched' (we have trouble, unpredictably, about
Matt Jacob [Sun, 16 Jun 2002 04:58:00 +0000 (04:58 +0000)]
Set all 23XX cards as 'touched' (we have trouble, unpredictably, about
running ABOUT FIRMWARE with some that were started by BIOS downloads).

Redo CTIO2 dma mapping- use continuation segments instead of multiple
CTIO2s. Thanks to Veritas for sponsoring this work (in a different
context).

MFC after: 1 week

22 years agoChange isp_target_async to a function returning an integer.
Matt Jacob [Sun, 16 Jun 2002 04:56:07 +0000 (04:56 +0000)]
Change isp_target_async to a function returning an integer.

Roll most immediate notifies into something the platform has to handle.

22 years agoSet default command count to 0xfe. This tells the f/w essentially
Matt Jacob [Sun, 16 Jun 2002 04:54:46 +0000 (04:54 +0000)]
Set default command count to 0xfe. This tells the f/w essentially
to *not* do flow control based upon resource counts for the firmware.
Increase default immediate notify count to 16.

Change isp_target_async to a function returning an integer.

22 years agoAdd MBOX_DRIVER_HEARTBEAT/MBOX_FW_HEARTBEAT/FC4_FC_SVC defines.
Matt Jacob [Sun, 16 Jun 2002 04:53:26 +0000 (04:53 +0000)]
Add MBOX_DRIVER_HEARTBEAT/MBOX_FW_HEARTBEAT/FC4_FC_SVC defines.

MFC after: 1 week

22 years agoRoll minor version. Add ISPASYNC_FW_RESTARTED async event. Add
Matt Jacob [Sun, 16 Jun 2002 04:52:53 +0000 (04:52 +0000)]
Roll minor version. Add ISPASYNC_FW_RESTARTED async event. Add
DEFAULT_FRAMESIZE && DEFAULT_EXEC_THROTTLE references.

MFC after: 1 week

22 years agoWhen trying to emulate a cross-device move operation with cp and rm, add
Tim J. Robbins [Sun, 16 Jun 2002 04:06:58 +0000 (04:06 +0000)]
When trying to emulate a cross-device move operation with cp and rm, add
a "--" argument after the options so filenames with leading dashes are
handled correctly.

PR: 39318
Submitted by: Serge van den Boom <svdb@stack.nl>
MFC after: 1 week

22 years agoJust rename two generic-queue init routines from 'init_*' to '*_gi'
Garance A Drosehn [Sun, 16 Jun 2002 01:51:37 +0000 (01:51 +0000)]
Just rename two generic-queue init routines from 'init_*' to '*_gi'
("gi" short for 'generic init'...).

MFC after: 10 days

22 years agoAdd a nearly complete rewrite of the lpc command 'down'. The only user-
Garance A Drosehn [Sun, 16 Jun 2002 01:43:29 +0000 (01:43 +0000)]
Add a nearly complete rewrite of the lpc command 'down'.  The only user-
visible change should be that more than one queue can now be specified,
if one uses the '-msg' parameter to separate the list of queues from the
status message to set.

The previous implementation of 'down' remains available as the command
'xdown', available for instant fallback if there seems to be anything
wrong with the new one.  If no one reports a problem after a few weeks,
then a later update will remove 'xdown'.

Reviewed by: freebsd-print@bostonradio.org
MFC after: 10 days

22 years agoMake the description of the 'down' command a little more readable.
Garance A Drosehn [Sat, 15 Jun 2002 23:47:40 +0000 (23:47 +0000)]
Make the description of the 'down' command a little more readable.

MFC after: 10 days

22 years agoAdd pselect(3) to the build. Need to figure out the most appropriate
Garrett Wollman [Sat, 15 Jun 2002 23:42:59 +0000 (23:42 +0000)]
Add pselect(3) to the build.  Need to figure out the most appropriate
way to document this interface.

22 years agoAdd some (but not all) of the things POSIX expects to be declared in
Garrett Wollman [Sat, 15 Jun 2002 23:39:10 +0000 (23:39 +0000)]
Add some (but not all) of the things POSIX expects to be declared in
<sys/select.h>.

22 years agoFix visibility macros. Declare fsblkcnt_t and fsfilcnt_t (for statvfs())
Garrett Wollman [Sat, 15 Jun 2002 23:38:43 +0000 (23:38 +0000)]
Fix visibility macros.  Declare fsblkcnt_t and fsfilcnt_t (for statvfs())
per POSIX.

22 years agoFix visibility issues; use <sys/timespec.h>.
Garrett Wollman [Sat, 15 Jun 2002 23:37:33 +0000 (23:37 +0000)]
Fix visibility issues; use <sys/timespec.h>.

22 years agoPartially fix namespace visibility issues by using new visibility macros.
Garrett Wollman [Sat, 15 Jun 2002 23:35:37 +0000 (23:35 +0000)]
Partially fix namespace visibility issues by using new visibility macros.
Some issues still remain, and will require research in old POSIX standards
if we care to get them right.

22 years agoReorganize the way that arguments are processed in lpc's generic-queue
Garance A Drosehn [Sat, 15 Jun 2002 23:31:27 +0000 (23:31 +0000)]
Reorganize the way that arguments are processed in lpc's generic-queue
commands, to make things a little cleaner (mainly for a later update).

Reviewed by: freebsd-print@bostonradio.org
MFC after: 10 days

22 years agoAdd a new command to 'lpc' called 'setstatus', which would be used to
Garance A Drosehn [Sat, 15 Jun 2002 22:51:58 +0000 (22:51 +0000)]
Add a new command to 'lpc' called 'setstatus', which would be used to
change the status message of a print queue.  This includes some minor
changes to the upstat() routine, so that error messages are not printed
while seteuid(priv-user).

Reviewed by: freebsd-audit and freebsd-print@bostonradio.org
MFC after: 10 days

22 years agonmount'ify unionfs further by using separate options instead
Maxime Henrion [Sat, 15 Jun 2002 22:48:14 +0000 (22:48 +0000)]
nmount'ify unionfs further by using separate options instead
of passing a flags mount options.  This removes the include of
sys/fs/unionfs/union.h in mount_unionfs as it should be.

Reviewed by: phk

22 years agoConvert UDF to nmount.
Maxime Henrion [Sat, 15 Jun 2002 22:40:13 +0000 (22:40 +0000)]
Convert UDF to nmount.

Reviewed by: scottl

22 years agomdcreate_vnode() isn't correctly clearing things out of the linked
Poul-Henning Kamp [Sat, 15 Jun 2002 19:18:43 +0000 (19:18 +0000)]
mdcreate_vnode() isn't correctly clearing things out of the linked
list if the file is of 0 size or mdsetcred() fails.

Submitted by: Martin Faxer <gmh003532@brfmasthugget.se>

22 years ago o Remove GIANT_REQUIRED from useracc() and vsunlock(). Neither
Alan Cox [Sat, 15 Jun 2002 19:10:19 +0000 (19:10 +0000)]
 o Remove GIANT_REQUIRED from useracc() and vsunlock().  Neither
   vm_map_check_protection() nor vm_map_unwire() expect Giant
   to be held.

22 years agoDo not issue deprecated ioctl.
Hajimu UMEMOTO [Sat, 15 Jun 2002 18:56:32 +0000 (18:56 +0000)]
Do not issue deprecated ioctl.

Obtained from: KAME
MFC after: 2 weeks

22 years agoMake sendmail really not start when sendmail_enable="NONE"
Gordon Tetlow [Sat, 15 Jun 2002 18:39:05 +0000 (18:39 +0000)]
Make sendmail really not start when sendmail_enable="NONE"

Submitted by: Dennis Kristensen <snicki@snicki.dk>
Reviewed by: Mike Makonnen <makonnen@pacbell.net>

22 years agoDon't install obsolete gifconfig(8). Please use ifconfig(8) instead.
Hajimu UMEMOTO [Sat, 15 Jun 2002 18:04:51 +0000 (18:04 +0000)]
Don't install obsolete gifconfig(8).  Please use ifconfig(8) instead.

22 years agoDon't install obsolete prefix(8). We now have `ifconfig eui64'.
Hajimu UMEMOTO [Sat, 15 Jun 2002 15:41:44 +0000 (15:41 +0000)]
Don't install obsolete prefix(8).  We now have `ifconfig eui64'.

22 years agoAdd eui64 option which fills interface index (lowermost 64bit of
Hajimu UMEMOTO [Sat, 15 Jun 2002 15:35:24 +0000 (15:35 +0000)]
Add eui64 option which fills interface index (lowermost 64bit of
an IPv6 address) automatically.
This should obsolete prefix(8).

Obtained from: NetBSD
MFC after: 2 weeks

22 years agoRemove local prototype for main().
Juli Mallett [Sat, 15 Jun 2002 11:28:09 +0000 (11:28 +0000)]
Remove local prototype for main().

22 years agoProtoize. Remove un-needed cast to char in switch of getopt(3)'s return value.
Juli Mallett [Sat, 15 Jun 2002 11:26:25 +0000 (11:26 +0000)]
Protoize.  Remove un-needed cast to char in switch of getopt(3)'s return value.
FBSDID.

22 years agoRemove <sys/types> where we use <sys/param> already.
Juli Mallett [Sat, 15 Jun 2002 11:03:28 +0000 (11:03 +0000)]
Remove <sys/types> where we use <sys/param> already.

protoize.

22 years agoRemove obsolescent -tabstop option from Synopsis (it was already removed
Tim J. Robbins [Sat, 15 Jun 2002 10:52:20 +0000 (10:52 +0000)]
Remove obsolescent -tabstop option from Synopsis (it was already removed
from the usage() message). Document what the -t option does in a more
standard place. Add a standard Diagnostics section.

22 years agoAllow <blank>s to be used to separate tab stop positions with the -t
Tim J. Robbins [Sat, 15 Jun 2002 10:16:39 +0000 (10:16 +0000)]
Allow <blank>s to be used to separate tab stop positions with the -t
argument, not just ASCII space characters and commas. Don't count
non-printing characters when determining column position.

22 years agoUpdate for lukemftp-1.6-beta2.
Mike Heffner [Sat, 15 Jun 2002 09:42:17 +0000 (09:42 +0000)]
Update for lukemftp-1.6-beta2.

22 years agoThis commit was generated by cvs2svn to compensate for changes in r98247,
Mike Heffner [Sat, 15 Jun 2002 09:40:37 +0000 (09:40 +0000)]
This commit was generated by cvs2svn to compensate for changes in r98247,
which included commits to RCS files with non-trunk default branches.

22 years agoImport lukemftp-1.6-beta2.
Mike Heffner [Sat, 15 Jun 2002 09:40:37 +0000 (09:40 +0000)]
Import lukemftp-1.6-beta2.

22 years agoSort sections.
Tim J. Robbins [Sat, 15 Jun 2002 08:41:12 +0000 (08:41 +0000)]
Sort sections.

22 years agofstat() returns information about the target of any symbolic link that
Tim J. Robbins [Sat, 15 Jun 2002 08:31:19 +0000 (08:31 +0000)]
fstat() returns information about the target of any symbolic link that
was opened, not the link itself. Remove dead code.

22 years agoA better prinflike fix...
Brian Somers [Sat, 15 Jun 2002 08:03:59 +0000 (08:03 +0000)]
A better prinflike fix...

22 years agoRemove whitespace at the end of lines.
Brian Somers [Sat, 15 Jun 2002 08:03:30 +0000 (08:03 +0000)]
Remove whitespace at the end of lines.

22 years agoImprove parsing of character and equivalence classes:
Tim J. Robbins [Sat, 15 Jun 2002 07:38:27 +0000 (07:38 +0000)]
Improve parsing of character and equivalence classes:
[:*] and [=*] are parsed as `infinitely many repetitions of :' (or *)
instead of literal characters (SUSv3)

22 years agoo Move more information from BUGS into SECURITY CONSIDERATIONS and
Chris Costello [Sat, 15 Jun 2002 06:00:56 +0000 (06:00 +0000)]
o Move more information from BUGS into SECURITY CONSIDERATIONS and
  condense the redundant bits.
o Provide an example for using snprintf over sprintf.  This may be
  supplemented with an asprintf() example soon.

Sponsored by: DARPA, NAI Labs

22 years ago o Remove the acquisition and release of Giant from munlock().
Alan Cox [Sat, 15 Jun 2002 05:05:04 +0000 (05:05 +0000)]
 o Remove the acquisition and release of Giant from munlock().

Reviewed by: tegge

22 years agoFix a printflike format error
Brian Somers [Sat, 15 Jun 2002 01:36:36 +0000 (01:36 +0000)]
Fix a printflike format error

22 years agoRemove a forgotten diagnostic
Brian Somers [Sat, 15 Jun 2002 01:35:03 +0000 (01:35 +0000)]
Remove a forgotten diagnostic

22 years agoAdd missing verb.
Giorgos Keramidas [Fri, 14 Jun 2002 22:15:22 +0000 (22:15 +0000)]
Add missing verb.

22 years agoNote that sio is required.
Matt Jacob [Fri, 14 Jun 2002 20:20:13 +0000 (20:20 +0000)]
Note that sio is required.

PR: 39305
Submitted by: Michael Hogsett <hogsett@csl.sri.com>
MFC after: 1 week

22 years agoChange vfs_copyopt() so that the length argument passed to it
Maxime Henrion [Fri, 14 Jun 2002 20:04:21 +0000 (20:04 +0000)]
Change vfs_copyopt() so that the length argument passed to it
must be the exact same size as the mount option.  This makes
vfs_copyopt() much more useful.

22 years agoImplement the <sys/statvfs.h> header. Related changes to <sys/types.h>
Garrett Wollman [Fri, 14 Jun 2002 19:37:06 +0000 (19:37 +0000)]
Implement the <sys/statvfs.h> header.  Related changes to <sys/types.h>
are still awaiting a worldstone.  Functions and their declarations to
come later.

22 years ago- Chainsaw the storage pool code. This was being used by a bunch of code
Andrew R. Reiter [Fri, 14 Jun 2002 19:31:07 +0000 (19:31 +0000)]
- Chainsaw the storage pool code.  This was being used by a bunch of code
  within the HARP atm stack and the hea and hfa device drivers, but since
  all of these systems were changed to use UMA zones, there is no use for
  the api any longer.

22 years ago o Use vm_map_wire() and vm_map_unwire() in place of vm_map_pageable() and
Alan Cox [Fri, 14 Jun 2002 18:21:01 +0000 (18:21 +0000)]
 o Use vm_map_wire() and vm_map_unwire() in place of vm_map_pageable() and
   vm_map_user_pageable().
 o Remove vm_map_pageable() and vm_map_user_pageable().
 o Remove vm_map_clear_recursive() and vm_map_set_recursive().  (They were
   only used by vm_map_pageable() and vm_map_user_pageable().)

Reviewed by: tegge

22 years ago- Turn the hea and hfa HARP storage pools into UMA zones and insert
Andrew R. Reiter [Fri, 14 Jun 2002 16:59:38 +0000 (16:59 +0000)]
- Turn the hea and hfa HARP storage pools into UMA zones and insert
  the necesary uma_zcreate() and uma_zdestroy calls into module loading
  handler and the device attach handling.
- Change the related HARP netatm code to use UMA zone functions when
  dealing with the zones that were formerly the ATM interface (hea, hfa)
  storage pools.
- Have atm_physif_freenifs() now get passed an uma_zone_t so that we can
  properly free the allocated NIF's back to their zone.

This should be the last commit to remove any code that makes use of the
netatm storage pool api.  I will be removing the api code within the near
future.

Reviewed by: mdodd

22 years agoMove the #include and #define's to the top of the file.
Tim J. Robbins [Fri, 14 Jun 2002 15:56:52 +0000 (15:56 +0000)]
Move the #include and #define's to the top of the file.

22 years agoBump the size of the equivalence set to NCHARS; this file was left out
Tim J. Robbins [Fri, 14 Jun 2002 15:53:38 +0000 (15:53 +0000)]
Bump the size of the equivalence set to NCHARS; this file was left out
of a previous commit implementing equivalence classes.

22 years agoAdd an entry for the US Robotics 2415.
Nick Sayer [Fri, 14 Jun 2002 15:32:01 +0000 (15:32 +0000)]
Add an entry for the US Robotics 2415.

PR: i386/38879
Submitted by: jordi_yc@lycos.es
MFC after: 1 week

22 years agoUse the Cm macro for the clear, init, reset and longname operands.
Tim J. Robbins [Fri, 14 Jun 2002 15:16:03 +0000 (15:16 +0000)]
Use the Cm macro for the clear, init, reset and longname operands.

22 years agoAdjust the exit status to conform to SUSv3 and document the fact it takes
Tim J. Robbins [Fri, 14 Jun 2002 15:08:05 +0000 (15:08 +0000)]
Adjust the exit status to conform to SUSv3 and document the fact it takes
multiple `argument' operands.

22 years agoSort sections. Avoid using "The -? option" at the start of option descriptions.
Tim J. Robbins [Fri, 14 Jun 2002 10:11:41 +0000 (10:11 +0000)]
Sort sections. Avoid using "The -? option" at the start of option descriptions.

22 years agoUse an usage() rather than doing the fprintf inside option parsing.
Juli Mallett [Fri, 14 Jun 2002 09:54:48 +0000 (09:54 +0000)]
Use an usage() rather than doing the fprintf inside option parsing.

22 years agoDon't treat the trailing ']' of an equivalence class expression as a
Tim J. Robbins [Fri, 14 Jun 2002 09:53:11 +0000 (09:53 +0000)]
Don't treat the trailing ']' of an equivalence class expression as a
character in the set. tr -d '[=a=]' was deleting ]'s as well as a's.
Noticed by the textutils test suite.

22 years agoAdd the P1003.1-2001 -C option which complements the set of characters
Tim J. Robbins [Fri, 14 Jun 2002 08:58:30 +0000 (08:58 +0000)]
Add the P1003.1-2001 -C option which complements the set of characters
(not byte values) specified by the first string argument.

22 years agoReserve two constants for managing socket MAC labels via socket options.
Robert Watson [Fri, 14 Jun 2002 08:49:04 +0000 (08:49 +0000)]
Reserve two constants for managing socket MAC labels via socket options.

22 years agoWhitespaec consistency.
Robert Watson [Fri, 14 Jun 2002 08:46:07 +0000 (08:46 +0000)]
Whitespaec consistency.

22 years agoNotify functions can destroy the pcb, so they have to return an
Jeffrey Hsu [Fri, 14 Jun 2002 08:35:21 +0000 (08:35 +0000)]
Notify functions can destroy the pcb, so they have to return an
indication of whether this happenned so the calling function
knows whether or not to unlock the pcb.

Submitted by: Jennifer Yang (yangjihui@yahoo.com)
Bug reported by:  Sid Carter (sidcarter@symonds.net)

22 years agoImplement support for equivalence classes ([=e=]) when the mapping is
Tim J. Robbins [Fri, 14 Jun 2002 07:37:08 +0000 (07:37 +0000)]
Implement support for equivalence classes ([=e=]) when the mapping is
one-to-one (SUSv3)

22 years agoAdd a comment about how we should use vn_open() here instead of directly
Robert Watson [Fri, 14 Jun 2002 07:24:01 +0000 (07:24 +0000)]
Add a comment about how we should use vn_open() here instead of directly
invoking VOP_OPEN().  This would reduce code redundancy with the rest
of the kernel, and also is required for MAC to work properly.

22 years agoI'm keeping an eye on xten
Warner Losh [Fri, 14 Jun 2002 06:59:49 +0000 (06:59 +0000)]
I'm keeping an eye on xten

22 years agodocument hw.pcic.ignore_pci
Warner Losh [Fri, 14 Jun 2002 06:56:35 +0000 (06:56 +0000)]
document hw.pcic.ignore_pci

22 years agoAdd US Robotics Wireless Card 2410
Warner Losh [Fri, 14 Jun 2002 06:56:03 +0000 (06:56 +0000)]
Add US Robotics Wireless Card 2410

Also add a snarky note about telling me about changes to this file so I can
keep OpenBSD and FreeBSD's copy of this man page more or less in sync.

22 years agoMissed in earlier commit -- I did cvs commit src/lib/libc. Oops.
Robert Watson [Fri, 14 Jun 2002 04:02:25 +0000 (04:02 +0000)]
Missed in earlier commit -- I did cvs commit src/lib/libc.  Oops.

22 years agoRe-commit w/fix:
Mike Silbersack [Fri, 14 Jun 2002 03:08:05 +0000 (03:08 +0000)]
Re-commit w/fix:

  Ensure that the syn cache's syn-ack packets contain the same
  ip_tos, ip_ttl, and DF bits as all other tcp packets.

  PR:             39141
  MFC after:      2 weeks

This time, make sure that ipv4 specific code (aka all of the above)
is only run in the ipv4 case.

22 years agoBack out ip_tos/ip_ttl/DF "fix", it just panic'd my box. :)
Mike Silbersack [Fri, 14 Jun 2002 02:43:20 +0000 (02:43 +0000)]
Back out ip_tos/ip_ttl/DF "fix", it just panic'd my box. :)

Pointy-hat to: silby