]> CyberLeo.Net >> Repos - FreeBSD/FreeBSD.git/log
FreeBSD/FreeBSD.git
26 years agovsprintf -> vnsprintf paranoia.
eivind [Tue, 7 Oct 1997 16:33:22 +0000 (16:33 +0000)]
vsprintf -> vnsprintf paranoia.

Obtained from: Todd Miller <millert@openbsd.org> (OpenBSD rev 1.3)

26 years agoDon't honor $HOME if >= MAXPATHLEN
eivind [Tue, 7 Oct 1997 16:27:48 +0000 (16:27 +0000)]
Don't honor $HOME if >= MAXPATHLEN

Obtained from: Todd Miller <millert@openbsd.org> (OpenBSD csh.c rev 1.3 and 1.4)

26 years agoRemoved uneeded ${.MAKEFLAGS} again.
bde [Tue, 7 Oct 1997 11:32:05 +0000 (11:32 +0000)]
Removed uneeded ${.MAKEFLAGS} again.
Originally committed in: rev.1.141
Spammed by: rev.1.149 (mouldy parallel make diffs)

26 years ago`numdirtybuffers' was not maintained properly. This caused excessive
bde [Tue, 7 Oct 1997 11:10:18 +0000 (11:10 +0000)]
`numdirtybuffers' was not maintained properly.  This caused excessive
flushing of buffers in an attempt to reduce numdirtybuffers, and
perhaps other problems.

26 years agoPreserve old SC_STATIC value after units exchange
ache [Tue, 7 Oct 1997 09:13:06 +0000 (09:13 +0000)]
Preserve old SC_STATIC value after units exchange

26 years agoOoops, this should have made it into the same commit, but didn't.
joerg [Tue, 7 Oct 1997 07:40:35 +0000 (07:40 +0000)]
Ooops, this should have made it into the same commit, but didn't.

Introduce the SIOC[SG]IFGENERIC hooks that can be used to pass an
arbritrary ioctl subcommand into an interface driver.  Surprisingly
enough, there was no provision for this already present (except of the
option of abusing SIOC[SG]IFMEDIA for this).

The idea is that an interface driver can establish ioctl subcommands
of its own that can't be meaningfully interpreted by the upper layer
interface ioctl function.  Something like this is required to
implement a clean solution of passing down things like CHAP secrets or
PPP options to the /sys/net/if_sppp* files.  (Yes, my CHAP is now
finally working with it, but i gotta update my kernel to the new
callout interface before being able to commit _that_.)

Reviewed by:    peter   [long ago, actually]

26 years agoIntroduce the SIOC[SG]IFGENERIC hooks that can be used to pass an
joerg [Tue, 7 Oct 1997 07:39:40 +0000 (07:39 +0000)]
Introduce the SIOC[SG]IFGENERIC hooks that can be used to pass an
arbritrary ioctl subcommand into an interface driver.  Surprisingly
enough, there was no provision for this already present (except of the
option of abusing SIOC[SG]IFMEDIA for this).

The idea is that an interface driver can establish ioctl subcommands
of its own that can't be meaningfully interpreted by the upper layer
interface ioctl function.  Something like this is required to
implement a clean solution of passing down things like CHAP secrets or
PPP options to the /sys/net/if_sppp* files.  (Yes, my CHAP is now
finally working with it, but i gotta update my kernel to the new
callout interface before being able to commit _that_.)

Reviewed by: peter [long ago, actually]

26 years agoRemove the claim that UUCP locking were not atomic. It is since
joerg [Tue, 7 Oct 1997 07:24:50 +0000 (07:24 +0000)]
Remove the claim that UUCP locking were not atomic.  It is since
revision 1.8 of uucplock.c.

26 years agoInclude optional support for HDB-style config files, so the true masochists
joerg [Tue, 7 Oct 1997 07:23:28 +0000 (07:23 +0000)]
Include optional support for HDB-style config files, so the true masochists
can get their rights as well. ;-)  The default remains, of course, Taylor
config.

Demanded by: some people on -hackers

I think this is safe enough to go into RELENG_2_2 as well, if there's
demand.

26 years agoFixed spammed Id.
bde [Tue, 7 Oct 1997 06:40:16 +0000 (06:40 +0000)]
Fixed spammed Id.
Spammed in: rev.1.5 (long ago)

26 years agoLatest round of patches from the author.
markm [Tue, 7 Oct 1997 06:30:01 +0000 (06:30 +0000)]
Latest round of patches from the author.

This driver includes the following patches submitted by:

1.0 Hideyuki Suzuki <hideyuki@sat.t.u-tokyo.ac.jp>
Japanese Cable support

2.0 Keith Sklower <sklower@CS.Berkeley.EDU>
Minor update to the BSDI section so it compiles cleanly on BSDI

3.0 Joao Carlos Mendes Luis <jonny@coppe.ufrj.br>
ioctl interface to select video format , NTSC, PAL, etc...

26 years agoAdd passwd(5) to "SEE ALSO".
obrien [Tue, 7 Oct 1997 05:40:36 +0000 (05:40 +0000)]
Add passwd(5) to "SEE ALSO".

ISSUES:
        An example and better explansion on how to specify a user's login
class in /etc/master passwd is needed.
(As I don't seem to be specifiying it right, I can't do it).

26 years agoAdd login.conf(5) to "SEE ALSO".
obrien [Tue, 7 Oct 1997 05:35:52 +0000 (05:35 +0000)]
Add login.conf(5) to "SEE ALSO".

ISSUES:
        An example and better explansion on how to use login classes is needed
        in both passwd.5 and login.conf.5.

26 years agoSupport VJ maximum slot identifiers != 15.
brian [Tue, 7 Oct 1997 00:56:58 +0000 (00:56 +0000)]
Support VJ maximum slot identifiers != 15.
Support VJ slot id compression.

Previously, ppp would negotiate a max slot between 2 & 15
(if asked), and would agree to slot id compression (if asked).
It would then proceed to use 16 slots and no compression
anyway.  The result was a rather unusable connection.

26 years agoMention the LOG_NTP change.
joerg [Mon, 6 Oct 1997 20:43:05 +0000 (20:43 +0000)]
Mention the LOG_NTP change.

26 years agoDocumentation for the LOG_NTP facility. While i was at it, added a
joerg [Mon, 6 Oct 1997 20:37:50 +0000 (20:37 +0000)]
Documentation for the LOG_NTP facility.  While i was at it, added a
bit of a warning about not all systems providing for the same set of
syslog facilities.

26 years agoSince i've just touched <sys/syslog.h>, thereby causing a large part
joerg [Mon, 6 Oct 1997 20:37:01 +0000 (20:37 +0000)]
Since i've just touched <sys/syslog.h>, thereby causing a large part
of the system to be rebuilt anyway, this is a good time to introduce
LOG_NTP.

The reasoning for a separate facility is that xntpd can sometimes
cause exaggerative log message at high prioritites which are,
depending on your environment and available clock sources, not
necessarily as important as other LOG_DAEMON messages.  However, they
used to clutter log files and system console in the existing setup.
Note that this situation could not be resolved using the !xntpd option
(think about it).

xntpd(8) is supposed to automatically pick up the change, it had
already all necessary #ifdef's.

The chosen value does, as far as my inquiries yielded, not clash with
any other operating system.

26 years agoDisable LPR/LPD by default.
imp [Mon, 6 Oct 1997 19:41:21 +0000 (19:41 +0000)]
Disable LPR/LPD by default.
Reviewed by: jkh, guido, imp

26 years agoBe more specific about -f/-t options (they're not yet implemented),
dima [Mon, 6 Oct 1997 18:52:16 +0000 (18:52 +0000)]
Be more specific about -f/-t options (they're not yet implemented),
not just exit with 0 status.

This definitely should go to 2.2.5.
If i won't have any objections - I'll commit it tonight.

26 years agoShift the value for INTERNAL_MARK left by 3, as done for all other
joerg [Mon, 6 Oct 1997 18:20:23 +0000 (18:20 +0000)]
Shift the value for INTERNAL_MARK left by 3, as done for all other
facilities.  This fixes the problems reported about syslogd not
recording marks correctly.

Submitted by: Alexander B. Povolotsky <tarkhil@mgt.msk.ru>

26 years agoClarify the actions of -s and the list of allowable names.
imp [Mon, 6 Oct 1997 16:28:47 +0000 (16:28 +0000)]
Clarify the actions of -s and the list of allowable names.

26 years agoRemove dereference of uninitialized (and unused) variable. This enables
jlemon [Mon, 6 Oct 1997 15:56:16 +0000 (15:56 +0000)]
Remove dereference of uninitialized (and unused) variable.  This enables
netatalk to work again.

26 years agoSync usage string with man page.
charnier [Mon, 6 Oct 1997 11:38:30 +0000 (11:38 +0000)]
Sync usage string with man page.

26 years agoUse err(3). Add usage()s.
charnier [Mon, 6 Oct 1997 11:36:08 +0000 (11:36 +0000)]
Use err(3). Add usage()s.

26 years agoHaving to review all changes to this file _very_ carefully
jkh [Mon, 6 Oct 1997 10:21:36 +0000 (10:21 +0000)]
Having to review all changes to this file _very_ carefully
led me to find 2 more errors which were there before. :)

26 years agoHooboy!
jkh [Mon, 6 Oct 1997 09:58:11 +0000 (09:58 +0000)]
Hooboy!

Did I ever spam this file good with that last commit.  Despite 3
reviewers, we still managed to revoke the eBones fixes, TCL 8.0 support,
libvgl and a host of other new things from this file in the process of
parallelizing the Makefile.  DOH!   I think we need more pointy hats - this
particular incident is worthy of a small children's birthday party's worth of
pointy hats. ;-)

I certainly intend to take more care with the processing of aged diffs
in the future, even if it does mean reading through 20K's worth of them.
I might also be a bit more anal about asking for more up-to-date changes
before looking at them. ;)

26 years agoAdd an extra sanity check to previous commit.
jkh [Mon, 6 Oct 1997 08:36:12 +0000 (08:36 +0000)]
Add an extra sanity check to previous commit.

26 years agoThis should put boot.help in its rightful place (finally!)
jkh [Mon, 6 Oct 1997 08:34:56 +0000 (08:34 +0000)]
This should put boot.help in its rightful place (finally!)

26 years agoAck! Fix excessive cut/paste blunder during poll mods. Who had the
peter [Mon, 6 Oct 1997 08:30:08 +0000 (08:30 +0000)]
Ack!  Fix excessive cut/paste blunder during poll mods.  Who had the
pointy hat last? :-]

When one is selecting (or polling) for write, it helps if we use the
write side of the pipe when requesting wakeups instead of the read side.
This broke ghostview (at least) - I'm suprised it wasn't noticed for
so long.

Reviewed by:  Greg Lehey <grog@lemis.com>

26 years agoAdded two Cyrix 6x86/6x86MX options.
kato [Mon, 6 Oct 1997 08:08:41 +0000 (08:08 +0000)]
Added two Cyrix 6x86/6x86MX options.

- CPU_CYRIX_NO_LOCK enables weak locking.  If this option is not set and
  FAILESAFE is defined, NO_LOCK bit of CCR1 is cleared.
- CPU_WT_ALLOC enables write-through allocation.

26 years agoWhoops! Whomever added the de_DE.ISO_8859-1 entry missed a "..", causing
jkh [Mon, 6 Oct 1997 08:08:14 +0000 (08:08 +0000)]
Whoops!  Whomever added the de_DE.ISO_8859-1 entry missed a "..", causing
an incomplete /usr/local to be populated (this is why the ports collection
is mostly broken in the out-of-box configuration with 2.2.5-BETA).

26 years agoCosmetic in err() strings.
charnier [Mon, 6 Oct 1997 07:46:08 +0000 (07:46 +0000)]
Cosmetic in err() strings.

26 years ago- Bring in APM_PCCARD_RESUME support from PAO, FreeBSD style. This new
nate [Mon, 6 Oct 1997 05:46:03 +0000 (05:46 +0000)]
- Bring in APM_PCCARD_RESUME support from PAO, FreeBSD style.  This new
  sysctl option 'fakes' like a card was removed and inserted when the
  machine is brought up again from a suspend.  It is disabled by default,
  and the old code is used.
Obtained from: PAO

26 years agoFixed spot in twintr() where tw_control and TWC_SYNC were used instead
stark [Mon, 6 Oct 1997 04:43:46 +0000 (04:43 +0000)]
Fixed spot in twintr() where tw_control and TWC_SYNC were used instead
of tw_zcport and tw_zcmask.  Could cause problems for people who use
pin 12 for sync.

26 years ago- Include <sys/interrupt.h> to avoid compiler errors.
nate [Mon, 6 Oct 1997 04:32:42 +0000 (04:32 +0000)]
- Include <sys/interrupt.h> to avoid compiler errors.

26 years ago- Hide the 'device doesn't supported shared interrupts' code behind
nate [Mon, 6 Oct 1997 04:27:32 +0000 (04:27 +0000)]
- Hide the 'device doesn't supported shared interrupts' code behind
  bootverbose, since the older register_intr() code didn't print out
  anything, and the laptop support will cause lots of these un-necessary
  messages.

26 years agoHmm.. where did _that_ come from? :-)
peter [Mon, 6 Oct 1997 04:10:24 +0000 (04:10 +0000)]
Hmm.. where did _that_ come from? :-)

26 years ago- Fix braino in last commit.
nate [Mon, 6 Oct 1997 04:02:42 +0000 (04:02 +0000)]
- Fix braino in last commit.

26 years agoBe more paranoid about unlinking files. From mhpower@MIT.EDU by way of
imp [Mon, 6 Oct 1997 03:58:48 +0000 (03:58 +0000)]
Be more paranoid about unlinking files.  From mhpower@MIT.EDU by way of
Theo de Raadt.

Likely 2.2.5R candidate.
Obtained from: OpenBSD

26 years agoModem cards send the speaker audio (dialing noises) to the host's
nate [Mon, 6 Oct 1997 03:17:36 +0000 (03:17 +0000)]
Modem cards send the speaker audio (dialing noises) to the host's
speaker.  Cirrus Logic PCIC chips must enable this.  There is also a Low
Power Dynamic Mode bit that claims to reduce power consumption by 30%,
so enable it and hope for the best.

PR: 4650
Submitted by: Nick Sayer <nsayer@quack.kfu.com>

26 years ago- Allocate the 'PCIC' interrupt from the last available (higher #) IRQ
nate [Mon, 6 Oct 1997 02:56:15 +0000 (02:56 +0000)]
- Allocate the 'PCIC' interrupt from the last available (higher #) IRQ
  instead of the first available, like Win95 does.  This appears to help
  on some machines, and avoids potential problems with built-in serial
  ports which tend to live at IRQ 3, which is usually picked with the
  old method.

26 years ago- Don't start at IRQ 0 in build-freelist, but instead start at IRQ 1.
nate [Mon, 6 Oct 1997 02:54:19 +0000 (02:54 +0000)]
- Don't start at IRQ 0 in build-freelist, but instead start at IRQ 1.
  (Not sure if this makes muich difference, but you never know.)

26 years agoImprove management of pages moving from the inactive to active queue. Additionally,
dyson [Mon, 6 Oct 1997 02:48:16 +0000 (02:48 +0000)]
Improve management of pages moving from the inactive to active queue.  Additionally,
add some much needed comments.

26 years ago- Style police
nate [Mon, 6 Oct 1997 02:46:38 +0000 (02:46 +0000)]
- Style police
- Updated some comments using data from the most recent PAO release.

Obtained (partially) from: PAO-970616

26 years agoRelax the vnode locking for read only operations.
dyson [Mon, 6 Oct 1997 02:38:30 +0000 (02:38 +0000)]
Relax the vnode locking for read only operations.

26 years agoIt is possible that MB's with really broken bios's not set up more of
dyson [Mon, 6 Oct 1997 02:11:32 +0000 (02:11 +0000)]
It is possible that MB's with really broken bios's not set up more of
the mtrr registers.  This just fills in more of the registers.

26 years agoUse the RFC1738 interpretation of ftp: URL's, i.e. CWD to each
fenner [Mon, 6 Oct 1997 01:09:56 +0000 (01:09 +0000)]
Use the RFC1738 interpretation of ftp: URL's, i.e. CWD to each
slash-seperated element of the URL and then RETR the last element.

PR: bin/4670
Reviewed by: wollman

26 years agorecord the envelope-from address in the received headers
jmb [Mon, 6 Oct 1997 00:09:17 +0000 (00:09 +0000)]
record the envelope-from address in the received headers
for all mail received at hub.freebsd.org

Obtained from: Peter Wemm <peter@freebsd.org>

26 years agoTurn NODOC back on - this is going to take a bit more work to get right.
jkh [Sun, 5 Oct 1997 22:31:40 +0000 (22:31 +0000)]
Turn NODOC back on - this is going to take a bit more work to get right.

26 years agoRemoved redundant 'all' arg to " Making Make" target.
fsmp [Sun, 5 Oct 1997 22:28:50 +0000 (22:28 +0000)]
Removed redundant 'all' arg to " Making Make" target.
Removed "-r RELENG_2_2" from 'update' target.

26 years agoFor SMP, add a space between the state name and the CPU#.
fsmp [Sun, 5 Oct 1997 21:20:56 +0000 (21:20 +0000)]
For SMP, add a space between the state name and the CPU#.
Everything following bumps right 1 character.

26 years agoAdded -D_ANSI_SOURCE as discussed with bde,
helbig [Sun, 5 Oct 1997 18:44:37 +0000 (18:44 +0000)]
Added -D_ANSI_SOURCE as discussed with bde,
removed bogus -trigraph (pointed out by bde),
included check of operands (suggested by joerg)

26 years agoHow the heck did this get back??
markm [Sun, 5 Oct 1997 15:39:47 +0000 (15:39 +0000)]
How the heck did this get back??

26 years agoChange things around a bit when bringing down the layers.
brian [Sun, 5 Oct 1997 15:00:10 +0000 (15:00 +0000)]
Change things around a bit when bringing down the layers.

o LcpLayerDown() no longer does a NewPhase(PHASE_TERMINATE).
  Instead, it's done in LcpLayerFinish().  LayerFinish() gets
  called by the FSM after the LCP FSM goes through the Stopping
  and Stopped states.

o -direct and -background mode exit at PHASE_TERMINATE, not
  PHASE_DEAD.

The result is that LCP, CCP & IPCP are brought down cleanly on both
sides of the link (not just our side).  Killing ppp rather than just
closing it still makes it get out after the LCP SendTerminateReq().
I'll have a look at that soon.  We're probably not actually sending
the REQ :-(

26 years agoRefer a bit to pppctl.
brian [Sun, 5 Oct 1997 14:27:08 +0000 (14:27 +0000)]
Refer a bit to pppctl.
Suggested (far to subtly for his own good) by: joerg

26 years agoAdd a few examples.
brian [Sun, 5 Oct 1997 14:21:30 +0000 (14:21 +0000)]
Add a few examples.

26 years agoAdd the REGAL CDC-4X as a known CD-ROM changer device.
joerg [Sun, 5 Oct 1997 13:38:17 +0000 (13:38 +0000)]
Add the REGAL CDC-4X as a known CD-ROM changer device.

Submitted by: shanee@augusta.de (Andreas Kohout)

26 years agoAdd a new special called `keep' to the list of special options. It
joerg [Sun, 5 Oct 1997 13:35:05 +0000 (13:35 +0000)]
Add a new special called `keep' to the list of special options.  It
provides for a means to specify an argument for crunchide's -k option.
(This is required by ntpdate.)

Submitted by: peter@rhiannon.clari.net.au (Peter Hawkins)

26 years agoDOH! I cannot believe that none of us caught this. Do *not*
jkh [Sun, 5 Oct 1997 12:31:08 +0000 (12:31 +0000)]
DOH!  I cannot believe that none of us caught this.  Do *not*
turn TCPWRAPPER support on by default, eh? :-)

26 years agoReverse rev 1.56 and rev 1.59. These made NFS too flakey.
phk [Sun, 5 Oct 1997 12:28:59 +0000 (12:28 +0000)]
Reverse rev 1.56 and rev 1.59.  These made NFS too flakey.

26 years agoCleanup.
wosch [Sun, 5 Oct 1997 12:28:02 +0000 (12:28 +0000)]
Cleanup.

26 years agoCosmetic: Be specific about using TABs in syslog.conf.
brian [Sun, 5 Oct 1997 10:29:32 +0000 (10:29 +0000)]
Cosmetic: Be specific about using TABs in syslog.conf.

26 years agoRemove NODOC setting - time to figure out how to get this working again.
jkh [Sun, 5 Oct 1997 09:47:28 +0000 (09:47 +0000)]
Remove NODOC setting - time to figure out how to get this working again.

26 years agoChanges to support full make parallelism (-j<n>) in the world
jkh [Sun, 5 Oct 1997 09:40:24 +0000 (09:40 +0000)]
Changes to support full make parallelism (-j<n>) in the world
target.
Reviewed by: <many different folks>
Submitted by: Nickolay N. Dudorov" <nnd@nnd.itfs.nsk.su>

26 years agoMake sure that the memory type registers are the same for each CPU
dyson [Sun, 5 Oct 1997 03:19:29 +0000 (03:19 +0000)]
Make sure that the memory type registers are the same for each CPU
in a P6 SMP system.  Some MB bios'es don't set the registers up correctly
for the AP's.  Additionally, set the memory between 0xa0000 and 0xbffff
as write combining.

26 years agoFix syntax error I introduced earlier. :)
jkh [Sat, 4 Oct 1997 20:51:42 +0000 (20:51 +0000)]
Fix syntax error I introduced earlier. :)

26 years agoWhile booting diskless we have no proc pointer.
phk [Sat, 4 Oct 1997 18:21:15 +0000 (18:21 +0000)]
While booting diskless we have no proc pointer.

26 years agoWhups! Missed a spot in the kerberos src changes.
jkh [Sat, 4 Oct 1997 15:50:09 +0000 (15:50 +0000)]
Whups!  Missed a spot in the kerberos src changes.

26 years agomv(1) is too silent if used with the option -i. It should
wosch [Sat, 4 Oct 1997 13:02:06 +0000 (13:02 +0000)]
mv(1) is too silent if used with the option -i. It should
print which input mv(1) expect (y/n) and print a warning if the
file was not overwritten.

26 years agoDelete redundant entries.
wosch [Sat, 4 Oct 1997 13:00:56 +0000 (13:00 +0000)]
Delete redundant entries.
Run tex twice due cross references.
Cleanup many tex generated files in `make clean'
Format latin1 output in 80 characters column (was ~110 characters).

26 years agoDo not suppress the first 5 lines of updated ports.
wosch [Sat, 4 Oct 1997 12:20:16 +0000 (12:20 +0000)]
Do not suppress the first 5 lines of updated ports.

Detected by: Ville Eerola <ville@vlsi.fi>

26 years agoRecord the envelope from address in the received: line. This has turned
peter [Sat, 4 Oct 1997 11:10:19 +0000 (11:10 +0000)]
Record the envelope from address in the received: line.  This has turned
out to be very useful for spam tracking and other problem debugging.

I first noticed this in one of Paul Vixie's machine's headers a while ago.

26 years agoSupport crypto and kerberos src dists.
jkh [Sat, 4 Oct 1997 09:08:14 +0000 (09:08 +0000)]
Support crypto and kerberos src dists.

26 years agoSupport crypto and kerberos source dists.
jkh [Sat, 4 Oct 1997 09:05:06 +0000 (09:05 +0000)]
Support crypto and kerberos source dists.

26 years agoUn-nuke krb and do what I *meant* to do, darn it.
jkh [Sat, 4 Oct 1997 09:01:20 +0000 (09:01 +0000)]
Un-nuke krb and do what I *meant* to do, darn it.

26 years agoAdapt for new crypto source layout.
jkh [Sat, 4 Oct 1997 08:58:16 +0000 (08:58 +0000)]
Adapt for new crypto source layout.

26 years agoSynchronize with sys/i386/isa/if_ed.c revision 1.122.
kato [Sat, 4 Oct 1997 05:19:49 +0000 (05:19 +0000)]
Synchronize with sys/i386/isa/if_ed.c revision 1.122.

26 years agoMake the blank screen saver work with MDA and CGA. The fade and green
yokota [Sat, 4 Oct 1997 04:24:18 +0000 (04:24 +0000)]
Make the blank screen saver work with MDA and CGA.  The fade and green
savers are also modified so that they behave in the same way as the
blank saver on MDA and CGA, although it's not the way these screen
savers are supposed to work, but fading monitor and tickling green
monitor cannot be done on MDA and CGA, AFAIK.  So, this is the second
best solution.

As of this change, the current state of support of screen savers is
summarized in the following table.

MDA CGA EGA VGA
blank OK OK NA OK
daemon OK OK OK OK
fade * * NA OK
green * * NA OK
snake OK OK OK OK
star OK OK OK OK

OK: works
NA: doesn't work, the module cannot be loaded for this adapter.
*: behave the same way as the blank saver.

As you can see, EGA is left out for now. But, we can do no better, as
EGA registers cannot be read...

Reviewed by: sos

26 years agoCheck for invalid lower controls to not claim binary data with chars < 128 as
ache [Sat, 4 Oct 1997 01:03:28 +0000 (01:03 +0000)]
Check for invalid lower controls to not claim binary data with chars < 128 as
ASCII _text_
Cleanup names detection code

26 years agoCosmetic: Log a strerror(errno) when failing to write to
brian [Sat, 4 Oct 1997 00:14:39 +0000 (00:14 +0000)]
Cosmetic: Log a strerror(errno) when failing to write to
          the tun device.

26 years agoAllow interfaces to be attached to bpf at times other than boot.
julian [Fri, 3 Oct 1997 21:32:05 +0000 (21:32 +0000)]
Allow interfaces to be attached to bpf at times other than boot.
doing so without this patch leads to an infinite loop in the kernel.

26 years agoMerged in better support of ISO 8601 from elsie.nci.nih.gov.
helbig [Fri, 3 Oct 1997 19:06:57 +0000 (19:06 +0000)]
Merged in better support of ISO 8601 from elsie.nci.nih.gov.
Added the conversion specifiers %g and %G, that are replaced
by the year which contains the greater part of the week in question.

26 years agocp(1) is too silent if used with the option -i. It should
wosch [Fri, 3 Oct 1997 18:11:14 +0000 (18:11 +0000)]
cp(1) is too silent if used with the option -i. It should
print which input cp(1) expect (y/n) and print a warning if the
file was not overwritten.

26 years agoReinsert telnet.h into the list of files installed into include/arpa/
markm [Fri, 3 Oct 1997 18:10:35 +0000 (18:10 +0000)]
Reinsert telnet.h into the list of files installed into include/arpa/
I suspect a commit of mine had this (bogusly).

26 years agoChange example from US to international standard date notation.
wosch [Fri, 3 Oct 1997 18:09:17 +0000 (18:09 +0000)]
Change example from US to international standard date notation.

26 years agoBacked out part of a previous change: don't set ZWS on '790 chips; it has
dg [Fri, 3 Oct 1997 16:26:15 +0000 (16:26 +0000)]
Backed out part of a previous change: don't set ZWS on '790 chips; it has
shown to be harmful in that it results in the card not being detected
properly on warmboot due to the station address failing to be read
correctly from the NVRAM.

26 years agoCall identifycyrix() when 6x86MX CPU is found. The identifycyrix()
kato [Fri, 3 Oct 1997 14:23:47 +0000 (14:23 +0000)]
Call identifycyrix() when 6x86MX CPU is found.  The identifycyrix()
function sets cyrix_did.  Old code could not display correct variable.

Reviewed by: Hideyuki Suzuki <hideyuki@sat.t.u-tokyo.ac.jp>

26 years agoNuke these two out of date readme files. This stuff should be in de(4)
peter [Fri, 3 Oct 1997 14:22:10 +0000 (14:22 +0000)]
Nuke these two out of date readme files.  This stuff should be in de(4)
man pages rather than hidden here anyway.

26 years agoInstall boot.help differently.
jkh [Fri, 3 Oct 1997 14:17:46 +0000 (14:17 +0000)]
Install boot.help differently.

26 years agoGet boot.help from a different place.
jkh [Fri, 3 Oct 1997 14:14:40 +0000 (14:14 +0000)]
Get boot.help from a different place.

26 years agoThe eBones src distribution is gone, and crypto is here. So remove
markm [Fri, 3 Oct 1997 14:05:29 +0000 (14:05 +0000)]
The eBones src distribution is gone, and crypto is here. So remove
references to sebones.* and bring in scrypto.*.

26 years agoDon't attempt to deref OpenConn after it's been set to NULL. :)
jkh [Fri, 3 Oct 1997 14:00:09 +0000 (14:00 +0000)]
Don't attempt to deref OpenConn after it's been set to NULL. :)

26 years agoBack out suppress '\n' mod.
danny [Fri, 3 Oct 1997 12:54:30 +0000 (12:54 +0000)]
Back out suppress '\n' mod.

26 years ago*GULP* Punt this into the attic. It is no longer used.
markm [Fri, 3 Oct 1997 09:49:50 +0000 (09:49 +0000)]
*GULP* Punt this into the attic. It is no longer used.

26 years agoCorrect the MAKE_EBONES macro --> MAKE_KERBEROS4
markm [Fri, 3 Oct 1997 06:37:35 +0000 (06:37 +0000)]
Correct the MAKE_EBONES macro --> MAKE_KERBEROS4

26 years agoCorrect typo.
jkh [Fri, 3 Oct 1997 03:34:00 +0000 (03:34 +0000)]
Correct typo.
PR: 4681
Submitted by: Masahiro Sekiguchi <seki@sysrap.cs.fujitsu.co.jp>

26 years agoFor people running paranoid ftpds, use the real username to log in
jkh [Fri, 3 Oct 1997 03:32:03 +0000 (03:32 +0000)]
For people running paranoid ftpds, use the real username to log in
with when running multi-user and we actually have one.
Suggested by: Joao Carlos Mendes Luis <jonny@coppe.ufrj.br>

26 years agoadded icmp6 and ip6. they are becoming common.
itojun [Fri, 3 Oct 1997 01:32:41 +0000 (01:32 +0000)]
added icmp6 and ip6.  they are becoming common.
Source: ftp://ftp.isi.edu/in-notes/iana/assignments/protocol-numbers

26 years agoTeach ftpErrString to format UNIX errnos, since at least ftpLogin()
fenner [Thu, 2 Oct 1997 23:26:03 +0000 (23:26 +0000)]
Teach ftpErrString to format UNIX errnos, since at least ftpLogin()
can return UNIX errnos.  When UNIX errnos catch up with FTP status
codes (e.g. at 100) a new way will have to be found to tell which
is which.

This allows fetch to print errors like
fetch: ftp.fu-berlin.de: No route to host
instead of
fetch: ftp.fu-berlin.de: Unknown error

26 years agoTake pity on Mark and fix the bogus paths to dump and restore in the
jkh [Thu, 2 Oct 1997 22:59:48 +0000 (22:59 +0000)]
Take pity on Mark and fix the bogus paths to dump and restore in the
help-distribute target myself.  They live in /usr/src/sbin, not
/usr/src/usr.sbin. :-)

I'm also assuming the omission of telnetd and telnet from this target
are intentional so I'll leave that alone.