]> CyberLeo.Net >> Repos - FreeBSD/FreeBSD.git/log
FreeBSD/FreeBSD.git
27 years agoThis commit was generated by cvs2svn to compensate for changes in r26790,
Peter Wemm [Sun, 22 Jun 1997 09:32:32 +0000 (09:32 +0000)]
This commit was generated by cvs2svn to compensate for changes in r26790,
which included commits to RCS files with non-trunk default branches.

27 years agoClean import of Matt Thomas's if_de.c driver as of 970508, rev 1.85. The
Peter Wemm [Sun, 22 Jun 1997 09:32:32 +0000 (09:32 +0000)]
Clean import of Matt Thomas's if_de.c driver as of 970508, rev 1.85.  The
slightly later one with optional if_media will be imported shortly as well.

Obtained from: Matt Thomas via http://www.3am-software.com/

27 years agoClean import of Matt Thomas's if_de.c driver as of 970508, rev 1.85. The
Peter Wemm [Sun, 22 Jun 1997 09:32:32 +0000 (09:32 +0000)]
Clean import of Matt Thomas's if_de.c driver as of 970508, rev 1.85.  The
slightly later one with optional if_media will be imported shortly as well.

Obtained from: Matt Thomas via http://www.3am-software.com/

27 years agoReshuffle some of the distribution names to make them
Jordan K. Hubbard [Sun, 22 Jun 1997 08:37:49 +0000 (08:37 +0000)]
Reshuffle some of the distribution names to make them
easier to remember & allow overrides.

27 years agoHook natd
Brian Somers [Sun, 22 Jun 1997 04:23:10 +0000 (04:23 +0000)]
Hook natd

27 years agoWhat twit commited this ? Duh !
Brian Somers [Sun, 22 Jun 1997 04:20:43 +0000 (04:20 +0000)]
What twit commited this ?  Duh !

27 years agoBring natd into main source tree now that the
Brian Somers [Sun, 22 Jun 1997 04:19:08 +0000 (04:19 +0000)]
Bring natd into main source tree now that the
pppd/natd combination works ok.

Submitted by: Ari Suutari <ari.suutari@ps.carel.fi>

27 years agoRemove a window during running down a file vnode. Also, the OBJ_DEAD
John Dyson [Sun, 22 Jun 1997 03:00:24 +0000 (03:00 +0000)]
Remove a window during running down a file vnode.  Also, the OBJ_DEAD
flag wasn't being respected during vref(), et. al.  Note that this
isn't the eventual fix for the locking problem.  Fine grained SMP
in the VM and VFS code will require (lots) more work.

27 years agoFix this damn mbuf with a negative m_len. It turns
Brian Somers [Sun, 22 Jun 1997 02:19:53 +0000 (02:19 +0000)]
Fix this damn mbuf with a negative m_len.  It turns
out to be a problem with VJ header compression.
davidg spotted this in usr.sbin/ppp/slcompress.c
a while ago, but I believe gave the wrong reasons -
it's too easy to reproduce !  The only scenario that
I've been able to reproduce the problem under is when
m_len is *exactly* 40 !  So go figure !

PR: 3749
Submitted elsewhere by: davidg
Obtained from: usr.sbin/ppp/slcompress.c

27 years agoAdd some cross references and alphabetize them.
John Polstra [Sat, 21 Jun 1997 23:05:44 +0000 (23:05 +0000)]
Add some cross references and alphabetize them.

27 years agoMake docs go to the right place finally.
Jordan K. Hubbard [Sat, 21 Jun 1997 19:39:48 +0000 (19:39 +0000)]
Make docs go to the right place finally.

27 years agoFixed va_arg() to work for small args (as in stdarg.h).
Bruce Evans [Sat, 21 Jun 1997 16:20:55 +0000 (16:20 +0000)]
Fixed va_arg() to work for small args (as in stdarg.h).

27 years agoBlock all write operations to /proc/1/* when securelevel > 0.
Alexander Langer [Sat, 21 Jun 1997 16:09:49 +0000 (16:09 +0000)]
Block all write operations to /proc/1/* when securelevel > 0.
The additional check in procfs_ctl.c could be backed out, but
I'm leaving it in for good measure.

Reviewed by: Theo de Raadt <deraadt@OpenBSD.org>

27 years agoShut up catman output.
Jordan K. Hubbard [Sat, 21 Jun 1997 15:56:50 +0000 (15:56 +0000)]
Shut up catman output.

27 years agoWhoops, get the proper rev of this.
Jordan K. Hubbard [Sat, 21 Jun 1997 15:45:51 +0000 (15:45 +0000)]
Whoops, get the proper rev of this.

27 years agoChange the way that X configuration method is selected.
Jordan K. Hubbard [Sat, 21 Jun 1997 15:45:15 +0000 (15:45 +0000)]
Change the way that X configuration method is selected.

27 years agoDon't attempt to generate errors for unpromoted types in va_arg(),
Bruce Evans [Sat, 21 Jun 1997 15:45:13 +0000 (15:45 +0000)]
Don't attempt to generate errors for unpromoted types in va_arg(),
since it is impossible to distinguish unpromoted types from small
(struct) types.  Renamed __va_promote() to __va_size() since it is
related to sizes of args on the stack and not to promotion.

PR: 3884
Submitted by: mostly by arnej@math.ntnu.no (Arne Henrik Juul)
Obtained from: name of__va_size and some parentheses fixes from NetBSD

27 years agoNew build scripts for these distributions.
Jordan K. Hubbard [Sat, 21 Jun 1997 15:42:17 +0000 (15:42 +0000)]
New build scripts for these distributions.

27 years agoChange the way distributions are built such that they can be built
Jordan K. Hubbard [Sat, 21 Jun 1997 15:41:55 +0000 (15:41 +0000)]
Change the way distributions are built such that they can be built
either in the Makefile or by an external script.  Move some of the
existing stuff into scripts.

27 years agoChange the distribute targets so that a given item in our source tree
Jordan K. Hubbard [Sat, 21 Jun 1997 15:40:34 +0000 (15:40 +0000)]
Change the distribute targets so that a given item in our source tree
can place itself into n distributions, where n >= 1.

27 years agoFix a typo, so the -a option will by documented in the synopsis.
Joerg Wunsch [Sat, 21 Jun 1997 09:17:45 +0000 (09:17 +0000)]
Fix a typo, so the -a option will by documented in the synopsis.

27 years agoFix "delete all".
Brian Somers [Fri, 20 Jun 1997 23:43:35 +0000 (23:43 +0000)]
Fix "delete all".

PR: 3913

27 years agoHm... wonder how long this has been here.
Bill Paul [Fri, 20 Jun 1997 17:54:11 +0000 (17:54 +0000)]
Hm... wonder how long this has been here.

The logic in get_myaddress() is broken: it always returns the loopback
address due to the following rule:

                if ((ifreq.ifr_flags & IFF_UP) &&
                    ifr->ifr_addr.sa_family == AF_INET &&
                    (loopback == 1 && (ifreq.ifr_flags & IFF_LOOPBACK))) {

The idea is that we want to select the interface address only if it's
up and it's in the AF_INET family. If it turns uout we don't have
such an interface available, we make a second pass through the loop,
this time settling for the loopback interface. But the logic inadvertently
locks out all cases when loopback == 0, so nothing is ever selected until
the second pass (when loopback == 1).

This is changed to:

                if (((ifreq.ifr_flags & IFF_UP) &&
                    ifr->ifr_addr.sa_family == AF_INET) ||
                    (loopback == 1 && (ifreq.ifr_flags & IFF_LOOPBACK))) {

which I think does the right thing.

This is yet another bogon I discovered during NIS+ testing; I need
get_myaddress() to work correctly so that the callback code in the
client library will work.

27 years agoAdd a couple of strategic screen clears.
Jordan K. Hubbard [Fri, 20 Jun 1997 07:04:16 +0000 (07:04 +0000)]
Add a couple of strategic screen clears.

27 years agoBack out previous fix - this bug's got diplomatic immunity as a registered
Jordan K. Hubbard [Thu, 19 Jun 1997 17:57:06 +0000 (17:57 +0000)]
Back out previous fix - this bug's got diplomatic immunity as a registered
political issue.

27 years agoFix multiple documentation bogons.
Jordan K. Hubbard [Thu, 19 Jun 1997 15:18:34 +0000 (15:18 +0000)]
Fix multiple documentation bogons.
PR: 3798
Submitted by: Jan Kuriyama <kuriyama@opt.phys.waseda.ac.jp>

27 years ago>Number: 3780
Jordan K. Hubbard [Thu, 19 Jun 1997 15:14:01 +0000 (15:14 +0000)]
>Number:         3780
>Category:       bin
>Synopsis:       WEXITSTATUS() may return nagative value, which causes sh to generate bad $?

PR: 3780
Submitted by: sanewo@ba2.so-net.or.jp

27 years agoSync usage string according to man page.
Philippe Charnier [Thu, 19 Jun 1997 14:44:24 +0000 (14:44 +0000)]
Sync usage string according to man page.

27 years agoCosmetic in usage string.
Philippe Charnier [Thu, 19 Jun 1997 14:42:10 +0000 (14:42 +0000)]
Cosmetic in usage string.

27 years agoUse warn(3).
Philippe Charnier [Thu, 19 Jun 1997 14:40:41 +0000 (14:40 +0000)]
Use warn(3).

27 years agoUse err(3). Add missing prototypes. Cosmetic in usage string.
Philippe Charnier [Thu, 19 Jun 1997 14:38:53 +0000 (14:38 +0000)]
Use err(3). Add missing prototypes. Cosmetic in usage string.

27 years agoUse err(3). Add usage().
Philippe Charnier [Thu, 19 Jun 1997 14:33:42 +0000 (14:33 +0000)]
Use err(3). Add usage().

27 years agoUse err(3).
Philippe Charnier [Thu, 19 Jun 1997 14:28:32 +0000 (14:28 +0000)]
Use err(3).

27 years agoTry again to explain the existence of lchown() and the absence of
John Polstra [Thu, 19 Jun 1997 01:07:05 +0000 (01:07 +0000)]
Try again to explain the existence of lchown() and the absence of
lchmod() and lutimes().

27 years agoFix crash when halting where APM was configured but not enabled
Garrett Wollman [Thu, 19 Jun 1997 00:25:03 +0000 (00:25 +0000)]
Fix crash when halting where APM was configured but not enabled
by checking whether APM is active in apm_power_off() and returning
if not.  (The code was already written with the expectation that
this function would return if it fails.)

27 years agoAdd cron_enable and inetd_enable flags to rc.conf.
Paul Traina [Wed, 18 Jun 1997 16:01:19 +0000 (16:01 +0000)]
Add cron_enable and inetd_enable flags to rc.conf.
WARNING: don't update rc and forget to update rc.conf, or you won't be
able to telnet back into your box after a reboot.

27 years agoVarious typos in man page. Cosmetic in usage string. Use err(3) instead
Philippe Charnier [Wed, 18 Jun 1997 06:37:23 +0000 (06:37 +0000)]
Various typos in man page. Cosmetic in usage string. Use err(3) instead
of fprintf. Remove unused variables.

27 years agoSync man page and usage string. Remove unused -p option from getopt call.
Philippe Charnier [Wed, 18 Jun 1997 06:34:28 +0000 (06:34 +0000)]
Sync man page and usage string. Remove unused -p option from getopt call.

27 years agoSync man page and usage string.
Philippe Charnier [Wed, 18 Jun 1997 06:33:11 +0000 (06:33 +0000)]
Sync man page and usage string.

27 years agoSync beetween man page and usage string.
Philippe Charnier [Wed, 18 Jun 1997 06:30:34 +0000 (06:30 +0000)]
Sync beetween man page and usage string.

27 years agoFprintf -> warn.
Philippe Charnier [Wed, 18 Jun 1997 06:28:46 +0000 (06:28 +0000)]
Fprintf -> warn.

27 years agoTypo.
Philippe Charnier [Wed, 18 Jun 1997 06:26:13 +0000 (06:26 +0000)]
Typo.

27 years ago1. Add a noWarn flag so that scripts can turn warnings off.
Jordan K. Hubbard [Wed, 18 Jun 1997 05:11:37 +0000 (05:11 +0000)]
1. Add a noWarn flag so that scripts can turn warnings off.
2. If hostname not set from script, bring up interactive dialog anyway.

27 years agoAdd `B' to list of flags to pass through from C*FLAGS when using mkdep
Satoshi Asami [Wed, 18 Jun 1997 03:39:34 +0000 (03:39 +0000)]
Add `B' to list of flags to pass through from C*FLAGS when using mkdep
(bsd.dep.mk) and compiling assembly language sources (bsd.lib.mk).
This doesn't change anything for our current source tree, but if you
want to use the -B switch in C*FLAGS to specify the location of
compiler subprograms, now you can do it.

Reviewed by: bde (implicitly)

27 years agoMove the "umask ?= 0022" check (a warning only) from the default
Satoshi Asami [Wed, 18 Jun 1997 03:20:22 +0000 (03:20 +0000)]
Move the "umask ?= 0022" check (a warning only) from the default
"do-install" target to the beginning of the "install" meta-target, so
that ports that define their own do-install will also run it without
having to duplicate it themselves.

Tested by: rebuilding all packages

27 years agoAdd "-I${DESTDIR}/usr/include" to CFLAGS if DESTDIR is defined, just
Satoshi Asami [Wed, 18 Jun 1997 03:10:31 +0000 (03:10 +0000)]
Add "-I${DESTDIR}/usr/include" to CFLAGS if DESTDIR is defined, just
like bsd.lib.mk and bsd.prog.mk.  It doesn't add it to CXXINCLUDES, I
don't think anybody has written a kernel module with C++.  (Not that I
think DavidG will allow it anyway. :)

Reviewed by: bde

27 years agoProtect against garbage mbufs in pppstart.
Brian Somers [Wed, 18 Jun 1997 02:50:40 +0000 (02:50 +0000)]
Protect against garbage mbufs in pppstart.
Remove previous hack in pppfcs().

This is still not the correct solution.  We shouldn't
have any incorrect mbufs.  This patch does however make
pppd/natd work (rather than jamming the interface).

27 years agoCheck named_enable rather than just named_flags.
Jordan K. Hubbard [Wed, 18 Jun 1997 01:55:19 +0000 (01:55 +0000)]
Check named_enable rather than just named_flags.
PR: 3893 (sort of)

27 years agoAdd for public examination the beginnings of the per-host cache support
Garrett Wollman [Wed, 18 Jun 1997 01:24:28 +0000 (01:24 +0000)]
Add for public examination the beginnings of the per-host cache support
which will for the basis of RTF_PRCLONING's more efficient, better-
designed replacement.

27 years agoTypo police.
Alexander Langer [Wed, 18 Jun 1997 00:35:53 +0000 (00:35 +0000)]
Typo police.

27 years agoCorrect the section number in the cross-reference for the publickey
John Polstra [Tue, 17 Jun 1997 20:24:33 +0000 (20:24 +0000)]
Correct the section number in the cross-reference for the publickey
file.

27 years agoWork around a bug (deficiency?) in the libdes Secure RPC compat interface.
Bill Paul [Tue, 17 Jun 1997 18:03:52 +0000 (18:03 +0000)]
Work around a bug (deficiency?) in the libdes Secure RPC compat interface.
The way Secure RPC is set up, the ecb_crypt() routine is expected to
be able to encrypt a buffer of any size up to 8192 bytes. However, the
des_ecb_encrypt() routine in libdes only encrypts 8 bytes (64 bits) at a
time. The rpc_enc.c module should compensate for this by calling
des_ecb_encrypt() repeatedly until it has encrypted the entire supplied
buffer, but it does not do this.

As a workaround, keyserv now handles this itself: if we're using DES
encryption, and the caller requested ECB mode, keyserv will do the right
thing.

Also changed all references to 'rc4' into 'arcfour' just in case some
litigious bastard from RSA is watching.

Note that I discovered and fixed this problem while trying to get
a part of NIS+ working: rpc.nisd signs directory objects with a 16-byte
MD5 digest that is encrypted with ecb_crypt(). Previously, only the
first 8 bytes of the digest were being properly encrypted, which caused
the Sun nis_cachemgr to reject the signatures as invalid. I failed to
notice this before since Secure RPC usually never has to encrypt more
than 8 bytes of data during normal operations.

27 years agoAdded CONTEC C-NET(9N) and C-NET(98)P2 support.
KATO Takenori [Tue, 17 Jun 1997 11:26:50 +0000 (11:26 +0000)]
Added CONTEC C-NET(9N) and C-NET(98)P2 support.

Submitted by: Chiharu Shibata <chi@rd.njk.co.jp>

27 years agocorrect the wrong ATM option name for native atm access
Kenjiro Cho [Tue, 17 Jun 1997 05:58:15 +0000 (05:58 +0000)]
correct the wrong ATM option name for native atm access
NETNATM --> NATM

reported by Bruce Evans.

Bruce also pointed out that NATM is confusing since config(8) defines
NATM as the number of atm pseudo device in "BUILD_DIR/atm.h".
We might change the name in the future but leave it as it is for now.

27 years agoCosmetic: Fix help syntax.
Brian Somers [Tue, 17 Jun 1997 02:04:51 +0000 (02:04 +0000)]
Cosmetic: Fix help syntax.

27 years agoHandle redial tries correctly under -ddial
Brian Somers [Tue, 17 Jun 1997 01:46:05 +0000 (01:46 +0000)]
Handle redial tries correctly under -ddial

27 years agoFix infinite loop.
David Nugent [Mon, 16 Jun 1997 23:38:01 +0000 (23:38 +0000)]
Fix infinite loop.
PR: 3878
Submitted by: roman@rpd.univ.kiev.ua

27 years agoPut TCP/IP logs back on one line.
Brian Somers [Mon, 16 Jun 1997 21:20:00 +0000 (21:20 +0000)]
Put TCP/IP logs back on one line.

27 years agoSet netfd to STDOUT_FILENO rather than STDIN_FILENO.
Brian Somers [Mon, 16 Jun 1997 19:59:41 +0000 (19:59 +0000)]
Set netfd to STDOUT_FILENO rather than STDIN_FILENO.

27 years agoCorrect filter docs.
Brian Somers [Mon, 16 Jun 1997 13:52:10 +0000 (13:52 +0000)]
Correct filter docs.

PR: 3464
Submitted by: Stephen J. Roznowski <sjr1@flash.net>

27 years agoRemove call to daemon(). Just call setsid() and close
Brian Somers [Mon, 16 Jun 1997 13:24:59 +0000 (13:24 +0000)]
Remove call to daemon().  Just call setsid() and close
any relevent descriptors (0,2,[1]).  We've already fork()d.
PR: 3874

27 years agoOuch!!! This should fix a serious build problem after the addition of the
John Dyson [Mon, 16 Jun 1997 12:10:21 +0000 (12:10 +0000)]
Ouch!!!  This should fix a serious build problem after the addition of the
new preliminary AIO support.  Unfortunately, I had a stray copy of aio.h
that made me think that things worked.

27 years agoMerge from Lite2.
Bruce Evans [Mon, 16 Jun 1997 11:20:05 +0000 (11:20 +0000)]
Merge from Lite2.
- use new getvfsbyname() interface.
- new -A option, like -a except only mounted file systems are unmounted.

All non-cosmetic FreeBSD changes in umount.c, except ignoring of
realpath() failures, went away because they are done better in Lite2.
realpath() failures must be ignored so that non-pathnames like
"<above>:/foo" and "host:/bar" get as far as mount(2).

Reviewed by: dfr

27 years agoFixed comment.
KATO Takenori [Mon, 16 Jun 1997 09:56:57 +0000 (09:56 +0000)]
Fixed comment.

Submitted by: Chiharu Shibata <chi@rd.njk.co.jp>

27 years agoGive up automatic detection of card model. Since DP8390 based NICs
KATO Takenori [Mon, 16 Jun 1997 09:54:30 +0000 (09:54 +0000)]
Give up automatic detection of card model.  Since DP8390 based NICs
are similar to each other, it is difficult to detec card type
automatically.

Reviewed by: Chiharu Shibata <chi@rd.njk.co.jp>

27 years agoTypo fix.
Masafumi Max NAKANE [Mon, 16 Jun 1997 06:54:30 +0000 (06:54 +0000)]
Typo fix.
PR: 3868
Submitted by: Kazuo Horikawa <k-horik@yk.rim.or.jp>

27 years agoUse err(3) instead of local copy.
Philippe Charnier [Mon, 16 Jun 1997 06:43:13 +0000 (06:43 +0000)]
Use err(3) instead of local copy.

27 years agoUse err(3) instead of local err() (a errx() like function).
Philippe Charnier [Mon, 16 Jun 1997 06:38:14 +0000 (06:38 +0000)]
Use err(3) instead of local err() (a errx() like function).

27 years agoUse err(3).
Philippe Charnier [Mon, 16 Jun 1997 06:35:23 +0000 (06:35 +0000)]
Use err(3).

27 years agoFix 2 typos.
Philippe Charnier [Mon, 16 Jun 1997 06:32:51 +0000 (06:32 +0000)]
Fix 2 typos.

27 years agoMinor optimization in fxp_intr.
David Greenman [Mon, 16 Jun 1997 04:45:57 +0000 (04:45 +0000)]
Minor optimization in fxp_intr.

27 years agoModifications to existing files to support the initial AIO/LIO and
John Dyson [Mon, 16 Jun 1997 00:29:36 +0000 (00:29 +0000)]
Modifications to existing files to support the initial AIO/LIO and
kernel based threading support.

27 years agoAdd initial AIO/LIO kernel thread support files. This is preliminary, and
John Dyson [Mon, 16 Jun 1997 00:27:26 +0000 (00:27 +0000)]
Add initial AIO/LIO kernel thread support files.  This is preliminary, and
further features will be added.

27 years agoUpgrade NFS to support the new vfs_bio resource/buffer management.
John Dyson [Mon, 16 Jun 1997 00:23:40 +0000 (00:23 +0000)]
Upgrade NFS to support the new vfs_bio resource/buffer management.

27 years agoCorrect the return code for the mlock system call. Also add the stubs
John Dyson [Sun, 15 Jun 1997 23:35:32 +0000 (23:35 +0000)]
Correct the return code for the mlock system call.  Also add the stubs
for mlockall and munlockall.

27 years agoFix a reference problem with maps. Only appears to manifest itself when
John Dyson [Sun, 15 Jun 1997 23:33:52 +0000 (23:33 +0000)]
Fix a reference problem with maps.  Only appears to manifest itself when
sharing address spaces.

27 years agoRemember to zero sockaddr_in struct before calling uaddr_to_sockaddr() to
Bill Paul [Sun, 15 Jun 1997 21:03:32 +0000 (21:03 +0000)]
Remember to zero sockaddr_in struct before calling uaddr_to_sockaddr() to
populate it. Not doing this can result in a garbage sockaddr_in, which
will cause connect() to block inside clnttcp_create().

27 years agosrandomdev: use stack junk value in the fallback code too
Andrey A. Chernov [Sun, 15 Jun 1997 18:23:19 +0000 (18:23 +0000)]
srandomdev: use stack junk value in the fallback code too

27 years agoFix a problem with the VN device. Specifically, the VN device can
John Dyson [Sun, 15 Jun 1997 17:56:53 +0000 (17:56 +0000)]
Fix a problem with the VN device.  Specifically, the VN device can
cause a problem of spiraling death due to buffer resource limitations.
The vfs_bio code in general had little ability to handle buffer resource
management, and now it does.  Also, there are a lot more knobs for tuning the
vfs_bio code now.  The knobs came free because of the need that there
always be some immediately available buffers (non-delayed or locked) for
use.  Note that the buffer cache code is much less likely to get bogged
down with lots of delayed writes, even more so than before.

27 years agoFixed iomem address of SMIT mode in an example of bs driver.
KATO Takenori [Sun, 15 Jun 1997 16:38:12 +0000 (16:38 +0000)]
Fixed iomem address of SMIT mode in an example of bs driver.

27 years agoSynchronize with sys/i386/i386/machdep.c revision up to 1.249.
KATO Takenori [Sun, 15 Jun 1997 16:35:13 +0000 (16:35 +0000)]
Synchronize with sys/i386/i386/machdep.c revision up to 1.249.

27 years agoSynchronize with sys/i386/boot/biosboot/boot.c revision 1.67.
KATO Takenori [Sun, 15 Jun 1997 16:33:52 +0000 (16:33 +0000)]
Synchronize with sys/i386/boot/biosboot/boot.c revision 1.67.

27 years agoFix another power down braino.
Garrett Wollman [Sun, 15 Jun 1997 02:24:06 +0000 (02:24 +0000)]
Fix another power down braino.

27 years agoFix definition of apm_power off (was suffering from cut&paste syndrome).
Garrett Wollman [Sun, 15 Jun 1997 02:19:40 +0000 (02:19 +0000)]
Fix definition of apm_power off (was suffering from cut&paste syndrome).

27 years agoWhen APM is configured, turn off the power when halting for good.
Garrett Wollman [Sun, 15 Jun 1997 02:03:03 +0000 (02:03 +0000)]
When APM is configured, turn off the power when halting for good.

27 years agoBelgian keyboard fixes
Andrey A. Chernov [Sat, 14 Jun 1997 22:26:52 +0000 (22:26 +0000)]
Belgian keyboard fixes
Submitted by: bovynf@awe.be

27 years agoTypo fix.
Masafumi Max NAKANE [Sat, 14 Jun 1997 19:45:00 +0000 (19:45 +0000)]
Typo fix.
PR: 3869
Submitted by: Kazuo Horikawa <k-horik@yk.rim.or.jp>

27 years agoTypo fix.
Masafumi Max NAKANE [Sat, 14 Jun 1997 19:30:27 +0000 (19:30 +0000)]
Typo fix.
PR: 3662
Submitted by: Matthew Hunt <mph@pobox.com>

27 years agoAdded 2 new defines:
Steve Passe [Sat, 14 Jun 1997 19:10:53 +0000 (19:10 +0000)]
Added 2 new defines:
 - OVERRIDE_TUNER: allows you to manually choose the tuner type for those
                   cards that fail to probe properly.  See source for legal
                   values.
 - OVERRIDE_DBX:   allows you to manually choose DBX or NO DBX for those
                   cards that fail to probe properly.
                   0 == no DBX circuit present, 1 == DBX circuit present.

27 years agoUpdate some text to reflect the fact that symbolic links now have
John Polstra [Sat, 14 Jun 1997 18:44:03 +0000 (18:44 +0000)]
Update some text to reflect the fact that symbolic links now have
owner and group attributes.

27 years agoRemoved unused #includes.
Bruce Evans [Sat, 14 Jun 1997 15:19:48 +0000 (15:19 +0000)]
Removed unused #includes.

27 years agoRemoved unused #includes.
Bruce Evans [Sat, 14 Jun 1997 15:11:07 +0000 (15:11 +0000)]
Removed unused #includes.

27 years agoFixed the type of a timeout function and removed casts that hid the
Bruce Evans [Sat, 14 Jun 1997 14:54:55 +0000 (14:54 +0000)]
Fixed the type of a timeout function and removed casts that hid the
problem.  There was no problem in practice (at least on 386's).

27 years agoRemoved unused #includes.
Bruce Evans [Sat, 14 Jun 1997 14:17:07 +0000 (14:17 +0000)]
Removed unused #includes.

27 years agoRemoved unused #includes.
Bruce Evans [Sat, 14 Jun 1997 13:56:12 +0000 (13:56 +0000)]
Removed unused #includes.

27 years agoRemoved unused #includes.
Bruce Evans [Sat, 14 Jun 1997 11:52:37 +0000 (11:52 +0000)]
Removed unused #includes.

27 years agoRemoved unused #includes.
Bruce Evans [Sat, 14 Jun 1997 11:38:46 +0000 (11:38 +0000)]
Removed unused #includes.

27 years agoDon't require superuser privileges for creating fifos. The v2 case was
Bruce Evans [Sat, 14 Jun 1997 11:19:35 +0000 (11:19 +0000)]
Don't require superuser privileges for creating fifos.  The v2 case was
broken when support for v3 was introduced in rev.1.16.  The v3 case has
always been broken in FreeBSD.

Should be in 2.2.

PR: 3838

27 years agoMove machine/types.h to non-standard section
Andrey A. Chernov [Sat, 14 Jun 1997 10:45:25 +0000 (10:45 +0000)]
Move machine/types.h to non-standard section
Change order of arc4* functions
Pointed-by: bde
27 years agoLoad the kernel config file where userconfig (with option USERCONFIG_BOOT)
Bruce Evans [Sat, 14 Jun 1997 10:00:42 +0000 (10:00 +0000)]
Load the kernel config file where userconfig (with option USERCONFIG_BOOT)
will see it.  The name of the file is obtained by appending ".config"
to the name of the kernel file.  If the file doesn't exist, then an
annoying message is printed.

Reduced the command line buffer size yet again from 4096 to 1024.  It is
oversized to defend against overrun in gets(), but data space is starting
to run out so the buffer size can't be as grossly oversized as before.