]> CyberLeo.Net >> Repos - FreeBSD/FreeBSD.git/log
FreeBSD/FreeBSD.git
26 years agoSupply the symbol table as an argument to X_db_symbol_values (required for elf).
dfr [Sun, 28 Jun 1998 00:55:02 +0000 (00:55 +0000)]
Supply the symbol table as an argument to X_db_symbol_values (required for elf).

26 years agoUse db_addr_t for the breakpoint address in db_set_single_step.
dfr [Sun, 28 Jun 1998 00:52:50 +0000 (00:52 +0000)]
Use db_addr_t for the breakpoint address in db_set_single_step.

26 years agoAdd a macro for testing the e_machine field of Elf64_Ehdr.
dfr [Sun, 28 Jun 1998 00:50:35 +0000 (00:50 +0000)]
Add a macro for testing the e_machine field of Elf64_Ehdr.

26 years agoUse elf symbols in DDB.
dfr [Sun, 28 Jun 1998 00:48:48 +0000 (00:48 +0000)]
Use elf symbols in DDB.

26 years agoAllow gentrap.
dfr [Sun, 28 Jun 1998 00:47:50 +0000 (00:47 +0000)]
Allow gentrap.

26 years agoCall ddb_trap() for gentrap on SimOS since I am using gentrap for DDB
dfr [Sun, 28 Jun 1998 00:47:20 +0000 (00:47 +0000)]
Call ddb_trap() for gentrap on SimOS since I am using gentrap for DDB
breakpoints (SimOS uses bpt itself for external debuggers).

26 years agoAdd a hack to get the ELF symbols from where SIMOS leaves them.
dfr [Sun, 28 Jun 1998 00:45:50 +0000 (00:45 +0000)]
Add a hack to get the ELF symbols from where SIMOS leaves them.

26 years agomachine/conf.h no longer exists.
dfr [Sun, 28 Jun 1998 00:44:51 +0000 (00:44 +0000)]
machine/conf.h no longer exists.

26 years agoo Fix remaining sizeof problems for 64 bit machines.
brian [Sat, 27 Jun 1998 23:48:54 +0000 (23:48 +0000)]
o Fix remaining sizeof problems for 64 bit machines.
o Allow ``set ....'' when we have multiple links but aren't in
  multilink mode.
o Do a TLS when we receive a ``Open'' event in ``Closed'' state,
  despite the rfc state transition table.  This is clearly an
  error in the RFC as TLS cannot have yet been called (without
  TLF) in the ``Closed'' state.
  I've posted a message to comp.protocols.ppp for confirmation.

26 years agoNote in log file if messages are being forwarded from a
steve [Sat, 27 Jun 1998 21:45:59 +0000 (21:45 +0000)]
Note in log file if messages are being forwarded from a
remote machine.

PR: 7055
Submitted by: Edwin Groothus <edwin@nwm.wan.philips.com>

26 years agoTypo.
steve [Sat, 27 Jun 1998 21:29:35 +0000 (21:29 +0000)]
Typo.

PR: 7093
Submitted by: Stefan Eggers <seggers@semyam.dinoco.de>

26 years agoA few tweaks to the spanish keymap.
steve [Sat, 27 Jun 1998 20:41:59 +0000 (20:41 +0000)]
A few tweaks to the spanish keymap.

PR: 7080
Submitted by: Jose M. Acaide <jose@we.lc.ehu.es>

26 years agoFix a few typos.
steve [Sat, 27 Jun 1998 20:32:41 +0000 (20:32 +0000)]
Fix a few typos.

PR: 7092
Submitted by: Kauzo Horikawa <h-horik@yk.rim.or.jp>

26 years agoMore u_long -> u_int32_t
brian [Sat, 27 Jun 1998 16:24:52 +0000 (16:24 +0000)]
More u_long -> u_int32_t

26 years agoFixes assorted problems with the history edit functions in libedit. This
brian [Sat, 27 Jun 1998 16:10:40 +0000 (16:10 +0000)]
Fixes assorted problems with the history edit functions in libedit.  This
affects sh, ftp (and others?).

Submitted by: Max Euston <meuston@jmrodgers.com>
PR:           6516

26 years agoMake db_tok_number a long to allow typing 64bit numbers on alpha.
dfr [Sat, 27 Jun 1998 15:40:56 +0000 (15:40 +0000)]
Make db_tok_number a long to allow typing 64bit numbers on alpha.

26 years agoUse %+11ln for printing in r format to make it work for 64bit registers.
dfr [Sat, 27 Jun 1998 15:39:51 +0000 (15:39 +0000)]
Use %+11ln for printing in r format to make it work for 64bit registers.

26 years agoAdd some stack backtrace support. Pretty fragile but its a start.
dfr [Sat, 27 Jun 1998 15:37:43 +0000 (15:37 +0000)]
Add some stack backtrace support.  Pretty fragile but its a start.

26 years agoUse %lx for printing register contents.
dfr [Sat, 27 Jun 1998 15:35:57 +0000 (15:35 +0000)]
Use %lx for printing register contents.

26 years agoMake uoff_t compile.
dfr [Sat, 27 Jun 1998 15:35:00 +0000 (15:35 +0000)]
Make uoff_t compile.

26 years agoI find it *really* irritating that you can't ``cvs rm''
brian [Sat, 27 Jun 1998 15:33:17 +0000 (15:33 +0000)]
I find it *really* irritating that you can't ``cvs rm''
things without removing them first.

26 years agoRemove redundant includes
brian [Sat, 27 Jun 1998 14:18:15 +0000 (14:18 +0000)]
Remove redundant includes

26 years agoDon't dlopen()/dlsym() libalias, use it in the same way
brian [Sat, 27 Jun 1998 14:17:28 +0000 (14:17 +0000)]
Don't dlopen()/dlsym() libalias, use it in the same way
as the rest of the world uses libraries.

26 years agoDon't assume ``sizeof(u_long) == 4''
brian [Sat, 27 Jun 1998 12:03:50 +0000 (12:03 +0000)]
Don't assume ``sizeof(u_long) == 4''
Submitted by: Theo

26 years agoCast void * to char * for log_Printf().
brian [Sat, 27 Jun 1998 12:03:46 +0000 (12:03 +0000)]
Cast void * to char * for log_Printf().

26 years ago``struct tunnel_header'' -> u_int32_t in OpenBSD.
brian [Sat, 27 Jun 1998 12:03:43 +0000 (12:03 +0000)]
``struct tunnel_header'' -> u_int32_t in OpenBSD.
struct tunnel_header is going soon.

Suggested by: Theo

26 years agoadditionally warnings
andreas [Sat, 27 Jun 1998 11:13:59 +0000 (11:13 +0000)]
additionally warnings
- login failures
- tcp_wrapper messages about refused connections

26 years agoOnly make struct xtcpcb visable if _NETINET_IN_PCB_H_ and _SYS_SOCKETVAR_H_
jhay [Sat, 27 Jun 1998 07:30:45 +0000 (07:30 +0000)]
Only make struct xtcpcb visable if _NETINET_IN_PCB_H_ and _SYS_SOCKETVAR_H_
are defined.
Reviewed by: bde

26 years agoMake vprint() print dev_t in hex also.
phk [Sat, 27 Jun 1998 07:28:49 +0000 (07:28 +0000)]
Make vprint() print dev_t in hex also.

26 years agoReport the type from the inode, not the vnode.
phk [Sat, 27 Jun 1998 06:45:04 +0000 (06:45 +0000)]
Report the type from the inode, not the vnode.

26 years agoReport the mode as the result of the VOP_GETATTR rather than the
phk [Sat, 27 Jun 1998 06:43:09 +0000 (06:43 +0000)]
Report the mode as the result of the VOP_GETATTR rather than the
vnodes type, they may not correspond.

26 years agoExtend the existing hackery to set the disk type to SCSI for the
jdp [Sat, 27 Jun 1998 02:01:25 +0000 (02:01 +0000)]
Extend the existing hackery to set the disk type to SCSI for the
"da" and "od" devices.  Formerly ESDI was used for them, causing
problems for fresh installations of CAM systems.

26 years agoDon't depend on sizeof(u_long) == 4.
brian [Fri, 26 Jun 1998 19:02:40 +0000 (19:02 +0000)]
Don't depend on sizeof(u_long) == 4.

26 years agoDon't make clean after building the crunch file. Build
brian [Fri, 26 Jun 1998 18:54:06 +0000 (18:54 +0000)]
Don't make clean after building the crunch file.  Build
a des & non-des version of ppp in the same way as init
and ed.

26 years agoDISTRIBUTION=des
brian [Fri, 26 Jun 1998 18:50:29 +0000 (18:50 +0000)]
DISTRIBUTION=des

26 years agoRemove isdisk() hacks.
phk [Fri, 26 Jun 1998 18:14:25 +0000 (18:14 +0000)]
Remove isdisk() hacks.

26 years agoAdd D_DISK flag.
phk [Fri, 26 Jun 1998 18:13:57 +0000 (18:13 +0000)]
Add D_DISK flag.

26 years agoCorrect typo: loosing --> losing
rnordier [Fri, 26 Jun 1998 14:25:18 +0000 (14:25 +0000)]
Correct typo: loosing --> losing

26 years agoFlesh this document out just a little in response to some user
jkh [Fri, 26 Jun 1998 10:35:55 +0000 (10:35 +0000)]
Flesh this document out just a little in response to some user
questions and also recommend linking over copying since, at this stage,
a stale copy is a real concern.

26 years agoDocument environment variables used by pkg_{add,create,info}. These
jkoshy [Fri, 26 Jun 1998 07:15:38 +0000 (07:15 +0000)]
Document environment variables used by pkg_{add,create,info}.  These
include PKG_TMPDIR, TMPDIR, PKG_DBDIR.  Add a FILES section to these
manual pages.

PR: docs/7064

26 years agoMF22: doc pointer.
jkh [Fri, 26 Jun 1998 01:48:04 +0000 (01:48 +0000)]
MF22: doc pointer.

26 years agoMF22: die sup die.
jkh [Fri, 26 Jun 1998 01:44:57 +0000 (01:44 +0000)]
MF22: die sup die.

26 years agoAdd ``ipcp'' as an optional argument to ``open'', and make
brian [Thu, 25 Jun 1998 22:33:31 +0000 (22:33 +0000)]
Add ``ipcp'' as an optional argument to ``open'', and make
open capable of re-negotiatiating the various layers.

It is now possible to change various link options and then
re-open the relevant layer, making the changes effective -
for example, switching off VJ compression or starting ECHO
LQRs on-the-fly.

26 years agoUpdate vmstat portion of man page to reflect the statistics currently
ghelmer [Thu, 25 Jun 1998 22:04:48 +0000 (22:04 +0000)]
Update vmstat portion of man page to reflect the statistics currently
shown by systat's vmstat display.
PR: docs/3764

26 years agoA long time ago I hacked mountd so that it would deal intelligently
wpaul [Thu, 25 Jun 1998 21:26:06 +0000 (21:26 +0000)]
A long time ago I hacked mountd so that it would deal intelligently
with export lines where the same hostname was specified more than once
(this happens a lot with netgroups sometimes). Recently I discovered
that it needs to be hacked to deal with multiple instances of the
same IP address too.

I've been using this modification locally for several months with no
hassles.

26 years agoImplement new argument -l (OpenBSD has -a but we already use that).
guido [Thu, 25 Jun 1998 19:39:19 +0000 (19:39 +0000)]
Implement new argument -l (OpenBSD has -a but we already use that).
This allows one to specify additional sockets in the unix domain
that syslogd listens to. Its primary use is to create log sockets in
chroot environments.
Obtained from:OpenBSD (with a bug fixed d

26 years agoFix btree problems. This passes regressions tests.
guido [Thu, 25 Jun 1998 18:50:06 +0000 (18:50 +0000)]
Fix btree problems.  This passes regressions tests.
PR: 7009
Obtained from: http://www.sleepycat.com/update/patch.185.html, patches 1.2,
1.3 and 1.4

26 years agoRemove "not hungly" panics. Cookies now used by the linux and ibcs2
dt [Thu, 25 Jun 1998 16:54:41 +0000 (16:54 +0000)]
Remove "not hungly" panics. Cookies now used by the linux and ibcs2
emulators. The emulators assume that filesystem may just ignore cookies, and
handle this case correctly. So we just ignore cookies.

Also sync *_readdir "prototypes" with reality.

26 years agoRemove bdevsw_add(), change the only two users to use bdevsw_add_generic().
phk [Thu, 25 Jun 1998 11:28:07 +0000 (11:28 +0000)]
Remove bdevsw_add(), change the only two users to use bdevsw_add_generic().
Extend cdevsw to be superset of bdevsw.
Remove non-functional bdev lkm support.
Teach wcd what the open() args mean.

26 years agoSync with sys/i386/conf/Makefile.i386 revision 1.115.
kato [Thu, 25 Jun 1998 10:49:49 +0000 (10:49 +0000)]
Sync with sys/i386/conf/Makefile.i386 revision 1.115.

26 years agoFix syntax error from last commit
brian [Thu, 25 Jun 1998 06:35:20 +0000 (06:35 +0000)]
Fix syntax error from last commit

26 years agoMention `kern_securelevel' variables in manual page.
jkoshy [Thu, 25 Jun 1998 04:27:56 +0000 (04:27 +0000)]
Mention `kern_securelevel' variables in manual page.

Reminded-by: max
26 years agoAdd the missing {} that caused the function to return ESRCH if it
jb [Thu, 25 Jun 1998 00:04:21 +0000 (00:04 +0000)]
Add the missing {} that caused the function to return ESRCH if it
had to wait for the thread to exit and if the caller didn't want the
thread exit status.

26 years agoReviewed by: guido@mouse.gvr.org
julian [Wed, 24 Jun 1998 23:50:20 +0000 (23:50 +0000)]
Reviewed by: guido@mouse.gvr.org
Submitted by: Archie@whistle.com
redo the last patch to allow differentiated kernel logging in
a much better way.

26 years agoOops, add ppc to the 'tty' imask.
msmith [Wed, 24 Jun 1998 22:01:32 +0000 (22:01 +0000)]
Oops, add ppc to the 'tty' imask.

26 years agoWhen some functions were added to sys/queue.h (1.15, 1.16),
phk [Wed, 24 Jun 1998 20:51:09 +0000 (20:51 +0000)]
When some functions were added to sys/queue.h (1.15, 1.16),
the queue diagram (in comment, showing member functions)
update were slipped.

PR: 7048
Submitted by: SAWADA Mizuki <miz@qb3.so-net.ne.jp>

26 years agoOops - forgot these with the last commit.
brian [Wed, 24 Jun 1998 19:36:37 +0000 (19:36 +0000)]
Oops - forgot these with the last commit.

26 years agoo If we come out of select() with only write descriptors that
brian [Wed, 24 Jun 1998 19:33:36 +0000 (19:33 +0000)]
o If we come out of select() with only write descriptors that
  end up writing zero bytes, sleep for 1/10 of a second so that
  we don't end up using up too much cpu.
  This should only ever happen on systems that wrongly report a
  descriptor as writable despite the tty buffer being full.
  Discussed with: Jeff Evarts

o Do an initial run-time check to see if select() alters the passed
  timeval.  This knowledge isn't yet used, but will be soon.

26 years agoFix race condition for creation of vnode_if.c and vnode_if.h.
eivind [Wed, 24 Jun 1998 13:55:32 +0000 (13:55 +0000)]
Fix race condition for creation of vnode_if.c and vnode_if.h.

Exact problem pointed out by: Cyrus Rahman <cr@jcmax.com>

26 years agoSync with sys/i386/isa/syscons.c revision 1.264.
kato [Wed, 24 Jun 1998 13:37:50 +0000 (13:37 +0000)]
Sync with sys/i386/isa/syscons.c revision 1.264.

26 years agoSync with sys/i386/isa/syscons.c revision 1.208.
kato [Wed, 24 Jun 1998 13:37:23 +0000 (13:37 +0000)]
Sync with sys/i386/isa/syscons.c revision 1.208.

26 years agoInsert a missing newline.
dt [Wed, 24 Jun 1998 13:30:25 +0000 (13:30 +0000)]
Insert a missing newline.

26 years agoDon't refer to mouse_info.u.datea.[xyz] while processing MOUSE_BUTTON_EVENT
yokota [Wed, 24 Jun 1998 10:21:30 +0000 (10:21 +0000)]
Don't refer to mouse_info.u.datea.[xyz] while processing MOUSE_BUTTON_EVENT
ioctl.  These fields don't have valid values for this ioctl.
PR: kern/6928
Submitted by: Bill Fenner <fenner@parc.xerox.com>

26 years agoAvoid some panics in sio_unload which gets called with various dirt
phk [Wed, 24 Jun 1998 09:23:32 +0000 (09:23 +0000)]
Avoid some panics in sio_unload which gets called with various dirt
if the attach of a modem card failed.

26 years agoAdd CUSEEME support. This has *not* been tested, nor
brian [Wed, 24 Jun 1998 06:58:00 +0000 (06:58 +0000)]
Add CUSEEME support.  This has *not* been tested, nor
could I find anyone to test it, so please report any
problems to me.

26 years agoClean the crunchgen targets after creating the crunch binary.
brian [Wed, 24 Jun 1998 06:55:18 +0000 (06:55 +0000)]
Clean the crunchgen targets after creating the crunch binary.
This *should* avoid the -DRELEASE_CRUNCH polution that's
mis-building user-ppp.

26 years agoMF22: more explanatory message from config on version mismatch detection.
jkh [Wed, 24 Jun 1998 06:16:32 +0000 (06:16 +0000)]
MF22: more explanatory message from config on version mismatch detection.

26 years agoMerge ppp 2.3.3 -> 2.3.5 changes (for what it's worth)
peter [Tue, 23 Jun 1998 21:58:41 +0000 (21:58 +0000)]
Merge ppp 2.3.3 -> 2.3.5 changes (for what it's worth)

26 years agoForgot to resync the RELNOTES to 2.3.5
peter [Tue, 23 Jun 1998 21:41:58 +0000 (21:41 +0000)]
Forgot to resync the RELNOTES to 2.3.5

26 years agoMinor grammar and punctuation cleanup.
ghelmer [Tue, 23 Jun 1998 16:52:59 +0000 (16:52 +0000)]
Minor grammar and punctuation cleanup.
Reference /etc/periodic/daily instead of /etc/crontab.

26 years agolable -> label
hoek [Tue, 23 Jun 1998 05:30:26 +0000 (05:30 +0000)]
lable -> label

26 years agoCross-ref login.conf(5).
hoek [Tue, 23 Jun 1998 04:39:35 +0000 (04:39 +0000)]
Cross-ref login.conf(5).

26 years agoGrammar touch-up.
hoek [Tue, 23 Jun 1998 04:38:50 +0000 (04:38 +0000)]
Grammar touch-up.

26 years agoUse a default blanktime (otherwise it shows up as "NO" in sysinstall's menus).
jkh [Tue, 23 Jun 1998 03:09:26 +0000 (03:09 +0000)]
Use a default blanktime (otherwise it shows up as "NO" in sysinstall's menus).

26 years agoTypo fix.
phk [Mon, 22 Jun 1998 21:09:10 +0000 (21:09 +0000)]
Typo fix.

Reviewed by: phk
Submitted by: John Hay <jhay@mikom.csir.co.za>

26 years agoSync with sys/i386/isa/if_ed.c and npx.c revisions 1.143 and 1.61,
kato [Mon, 22 Jun 1998 08:06:58 +0000 (08:06 +0000)]
Sync with sys/i386/isa/if_ed.c and npx.c revisions 1.143 and 1.61,
respectively.

26 years agoSync with sys/i386/i386/machdep.c revision 1.301.
kato [Mon, 22 Jun 1998 08:05:12 +0000 (08:05 +0000)]
Sync with sys/i386/i386/machdep.c revision 1.301.

26 years agoSync with sys/i386/isa/if_fe.c revision 1.42.
kato [Mon, 22 Jun 1998 08:03:52 +0000 (08:03 +0000)]
Sync with sys/i386/isa/if_fe.c revision 1.42.

26 years agoAdd rc.conf option to set kernel security level and modify "/etc/rc" to
jkoshy [Mon, 22 Jun 1998 06:34:12 +0000 (06:34 +0000)]
Add rc.conf option to set kernel security level and modify "/etc/rc" to
use this.

Requested by: max, andreas

Note: This change just provides a convenient way to exercise existing
functionality.  Whether `kern.securelevel' is effective in increasing
system security is another issue, and one that has been well thrashed
out in the lists.

26 years agoFixed (?) races in mark_buffer_dirty(). We abuse the buffer cache
bde [Sun, 21 Jun 1998 21:06:04 +0000 (21:06 +0000)]
Fixed (?) races in mark_buffer_dirty().  We abuse the buffer cache
by hacking on locked buffers without getblk()ing them, and we didn't
even use splbio() to prevent biodone() changing the buffer underneath
use when a write completes.  I think there was no problem in practice
on i386's because the operations on b_flags and numdirtybufs happen to
be atomic.  We still depend on biodone()'s operations on b_flags not
interfering with ours.  I think there is only interference for B_ERROR,
and this is harmless because errors for async writes are ignored anyway.

Don't use mark_buffer_dirty() except for superblock-related metadata.
It was used in just one case where ordinary BSD buffering is more
natural.

26 years agoRemoved unused function ll_w_block(). It has always had races due
bde [Sun, 21 Jun 1998 19:56:31 +0000 (19:56 +0000)]
Removed unused function ll_w_block().  It has always had races due
to not using splbio(), and has rotted a little.  The races were
probably harmless in practice because this function was only used
for superblock updates, and separate superblock updates are probably
prevented from running into each other by doing part of the update
synchronously.

26 years agoRemoved unused includes.
bde [Sun, 21 Jun 1998 18:02:50 +0000 (18:02 +0000)]
Removed unused includes.

26 years agoMake value of SIZE accurate.
dt [Sun, 21 Jun 1998 18:00:34 +0000 (18:00 +0000)]
Make value of SIZE accurate.

Obtained from: ps(1)

26 years agoRemoved unused includes.
bde [Sun, 21 Jun 1998 17:46:53 +0000 (17:46 +0000)]
Removed unused includes.
Ifdefed conditionally used includes.

26 years agoRemoved unused includes.
bde [Sun, 21 Jun 1998 17:08:08 +0000 (17:08 +0000)]
Removed unused includes.
Ifdefed conditionally used includes.

26 years agoRemoved unused includes.
bde [Sun, 21 Jun 1998 16:51:06 +0000 (16:51 +0000)]
Removed unused includes.
Ifdefed conditionally used includes.
Finished rev.1.39.  (I only submitted the request.)

26 years agoRemoved unused includes.
bde [Sun, 21 Jun 1998 15:54:11 +0000 (15:54 +0000)]
Removed unused includes.
Don't depend on "implicit int".

26 years agoRemoved unused includes.
bde [Sun, 21 Jun 1998 15:49:39 +0000 (15:49 +0000)]
Removed unused includes.
Don't depend on "implicit int".
Don't bloat the data section with explicit initialization to 0.

26 years agoRemoved unused includes.
bde [Sun, 21 Jun 1998 14:53:44 +0000 (14:53 +0000)]
Removed unused includes.

26 years agoIfdefed a conditionally used include.
bde [Sun, 21 Jun 1998 14:45:00 +0000 (14:45 +0000)]
Ifdefed a conditionally used include.

26 years agoSince I got no objections to this patch, and no one has offered any
mckay [Sun, 21 Jun 1998 14:22:29 +0000 (14:22 +0000)]
Since I got no objections to this patch, and no one has offered any
alternative, I present .. ta! da! .. the __error() hack.

This patch to the a.out dynamic loader provides old a.out binaries
with __error() if they are linked with an older libc that lacks it,
but are also linked against a library that needs it.

There is a smaller, tricker hack that takes advantage of the fact
that ld.so has __error() too, courtesy of the new libc, but this
hack is the straightforward version.

26 years agoRemove bogus comment that teleported in from sys/i386/i386/mp_machdep.c.
mckay [Sun, 21 Jun 1998 14:08:27 +0000 (14:08 +0000)]
Remove bogus comment that teleported in from sys/i386/i386/mp_machdep.c.

26 years agoMoved P1003 options from the middle of the devices section to the
bde [Sun, 21 Jun 1998 13:36:53 +0000 (13:36 +0000)]
Moved P1003 options from the middle of the devices section to the
end of the main options section.

Turned on documented option OVERRIDE_TUNER.  LINT is primarily
for turning on options, not for documenting them.

Don't list IPFILTER twice (once as broken).

26 years agoFixed typo in ifdefed code. (NFS_ACDEBUG is not in LINT. Therefore,
bde [Sun, 21 Jun 1998 12:50:12 +0000 (12:50 +0000)]
Fixed typo in ifdefed code.  (NFS_ACDEBUG is not in LINT.  Therefore,
code controlled by it did not even compile.)

26 years agoAdded a missing options include.
bde [Sun, 21 Jun 1998 12:36:12 +0000 (12:36 +0000)]
Added a missing options include.

26 years agoRound tickadj up. This prevents tickadj from being 0 when HZ > 500,
bde [Sun, 21 Jun 1998 12:22:35 +0000 (12:22 +0000)]
Round tickadj up.  This prevents tickadj from being 0 when HZ > 500,
which makes adjtime(2) useless and confuses xntpd(8) into refusing
to start even when it would use the kernel PLL instead of adjtime().
The result is the same as recommended by tickadj(8), at least when
HZ divides 10^6.  Of course, you wouldn't want to actually use
adjtime() when HZ is large.  In the silly boundary case of HZ == 10^6,
tickadj == tick == 1 so the clock stops while adjtime() is active.

26 years agoConverted add_interrupt_randomness() to take a `void *' arg. Rewrote
bde [Sun, 21 Jun 1998 11:33:32 +0000 (11:33 +0000)]
Converted add_interrupt_randomness() to take a `void *' arg.  Rewrote
mmioctl() to fix hundreds of style bugs and a few error handling bugs
(don't check for superuser privilege for inappropriate ioctls, don't
check the input arg for the output-only MEM_RETURNIRQ ioctl, and don't
return EPERM for null changes).

26 years agoSuggest the use of ``accept dns'', not ``enable dns'' in
brian [Sun, 21 Jun 1998 11:14:50 +0000 (11:14 +0000)]
Suggest the use of ``accept dns'', not ``enable dns'' in
server mode.

26 years agoremove old hack that's no longer needed and now breaks the compile
peter [Sun, 21 Jun 1998 04:47:21 +0000 (04:47 +0000)]
remove old hack that's no longer needed and now breaks the compile

26 years agoSay goodbye to Airport 71
jkh [Sun, 21 Jun 1998 04:46:53 +0000 (04:46 +0000)]
Say goodbye to Airport 71