]> CyberLeo.Net >> Repos - FreeBSD/FreeBSD.git/log
FreeBSD/FreeBSD.git
23 years agoFix a warning (the prototypes probably shouldn't be so over-zealously
peter [Fri, 19 Jan 2001 09:07:16 +0000 (09:07 +0000)]
Fix a warning (the prototypes probably shouldn't be so over-zealously
#ifdef'ed though)

23 years agoWarning police:
jhb [Fri, 19 Jan 2001 09:04:56 +0000 (09:04 +0000)]
Warning police:
- Share the ex_stop() prototype in if_exvar.h
- Remove an unused local variable.

23 years agoFix a maybe-not-so-harmless warning.
peter [Fri, 19 Jan 2001 09:02:40 +0000 (09:02 +0000)]
Fix a maybe-not-so-harmless warning.

23 years agoAdd in an interrupt type for this driver. I have no idea how this could
jhb [Fri, 19 Jan 2001 08:55:14 +0000 (08:55 +0000)]
Add in an interrupt type for this driver.  I have no idea how this could
have ever worked without this.

Reviewed by: peter

23 years agoRevert revision 1.102. I don't think p_nice needs to be protected with
jhb [Fri, 19 Jan 2001 08:23:22 +0000 (08:23 +0000)]
Revert revision 1.102.  I don't think p_nice needs to be protected with
sched_lock, and I'm fairly certain P_TRACED will be protected with the
proc lock instead.

Pointed out indirectly by: bde

23 years agoBSD/OS was called BSD/386 prior to 2.0.
itojun [Fri, 19 Jan 2001 07:36:36 +0000 (07:36 +0000)]
BSD/OS was called BSD/386 prior to 2.0.

23 years agoDo not cluster with B_LOCKED buffers.
dillon [Fri, 19 Jan 2001 05:31:07 +0000 (05:31 +0000)]
Do not cluster with B_LOCKED buffers.

This is an odd one.  This patch appears to fix a panic related to background
bitmap writes (for FFS), though neither Kirk, Ian, or I can figure out how
B_CLUSTEROK could possibly be set on a bitmap block to cause the clustering
code to improperly cluster with a buffer undergoing a background write.

In anycase, the clustering code is very fragile and this patch helps with
that, as well as possibly fixing a bug Andre was having.

Suggested by: Ian Dowse <iedowse@maths.tcd.ie>
Testing by: Andre Albsmeier <andre.albsmeier@mchp.siemens.de>

23 years agoImplement MTX_RECURSE flag for mtx_init().
bmilekic [Fri, 19 Jan 2001 01:59:14 +0000 (01:59 +0000)]
Implement MTX_RECURSE flag for mtx_init().
All calls to mtx_init() for mutexes that recurse must now include
the MTX_RECURSE bit in the flag argument variable. This change is in
preparation for an upcoming (further) mutex API cleanup.
The witness code will call panic() if a lock is found to recurse but
the MTX_RECURSE bit was not set during the lock's initialization.

The old MTX_RECURSE "state" bit (in mtx_lock) has been renamed to
MTX_RECURSED, which is more appropriate given its meaning.

The following locks have been made "recursive," thus far:
eventhandler, Giant, callout, sched_lock, possibly some others declared
in the architecture-specific code, all of the network card driver locks
in pci/, as well as some other locks in dev/ stuff that I've found to
be recursive.

Reviewed by: jhb

23 years agoLook for modules in /boot/kernel these days.
phk [Thu, 18 Jan 2001 22:36:34 +0000 (22:36 +0000)]
Look for modules in /boot/kernel these days.

23 years agoNew release notes: Condition variables in kernel, ConnectionsPerPeriod
bmah [Thu, 18 Jan 2001 19:25:40 +0000 (19:25 +0000)]
New release notes:  Condition variables in kernel, ConnectionsPerPeriod
deprecated in sshd_config, cdcontrol(1) and CDROM, sysctl -N,
ports "origin", sysinstall->/usr/sbin.

Note MFC for LinkSys cards note, clarify and MFC syslogd(8) and
LOG_CONSOLE.

23 years agoDocument I386_CPU kernel options, and note removal of I386_CPU
bmah [Thu, 18 Jan 2001 17:25:34 +0000 (17:25 +0000)]
Document I386_CPU kernel options, and note removal of I386_CPU
from the GENERIC kernel.

Reviewed by: peter, jhb

23 years agomdoc(7) police: add missing .Ed call.
ru [Thu, 18 Jan 2001 17:23:15 +0000 (17:23 +0000)]
mdoc(7) police: add missing .Ed call.

23 years agoman(7) -> mdoc(7).
ru [Thu, 18 Jan 2001 16:04:48 +0000 (16:04 +0000)]
man(7) -> mdoc(7).

23 years agoWhen blocksize is set to 2352, alwas read 2352 bytes.
sos [Thu, 18 Jan 2001 15:20:18 +0000 (15:20 +0000)]
When blocksize is set to 2352, alwas read 2352 bytes.
This means it might read more than just data, but that is exactly
what we want here (for reading VCD's fx)

23 years agoman(7) -> mdoc(7).
ru [Thu, 18 Jan 2001 12:56:17 +0000 (12:56 +0000)]
man(7) -> mdoc(7).

23 years agoremove reference to nonexistent screenmap(5) manual page.
ben [Thu, 18 Jan 2001 11:59:01 +0000 (11:59 +0000)]
remove reference to nonexistent screenmap(5) manual page.

PR: 24305
Submitted by: Marco van de Voort <marcov@stack.nl>

23 years agomdoc(7) police:
ru [Thu, 18 Jan 2001 09:42:50 +0000 (09:42 +0000)]
mdoc(7) police:
Use proper macros for variables and their types, removed
hard sentence breaks, added a bunch of cross references.

23 years agoProtect p_stat and p_oncpu with sched_lock in forward_signal().
jhb [Thu, 18 Jan 2001 08:19:25 +0000 (08:19 +0000)]
Protect p_stat and p_oncpu with sched_lock in forward_signal().

23 years agoNote that /stand/sysinstall is deprecated.
jkh [Thu, 18 Jan 2001 08:13:38 +0000 (08:13 +0000)]
Note that /stand/sysinstall is deprecated.

Forgotten by: jhb

23 years agoDeal with table parsing bugs caused by out-of-order record entries.
peter [Thu, 18 Jan 2001 07:42:02 +0000 (07:42 +0000)]
Deal with table parsing bugs caused by out-of-order record entries.
Complain about out-of-order entries.  This fixes the 'mp extended
table HOSED!" report on the DL360 we have here with the "fixed" compaq
bios rom to fix the table length off-by-one error.

23 years agoAdd a smal line of text which indicates I am working on updating
asmodai [Thu, 18 Jan 2001 07:19:25 +0000 (07:19 +0000)]
Add a smal line of text which indicates I am working on updating
this.  <almost exactly 6 years after Garrett>

23 years agoworkaround; be sure to initialize nd6 interface information when IPv6
itojun [Thu, 18 Jan 2001 06:07:53 +0000 (06:07 +0000)]
workaround; be sure to initialize nd6 interface information when IPv6
interface address gets added.  this will avoid presenting EMSGSIZE when
outgoing interface is down (and never brought up).

sync with kame.

23 years agoNikon E990 and Nomad MP3 player Ids.
n_hibma [Thu, 18 Jan 2001 00:15:49 +0000 (00:15 +0000)]
Nikon E990 and Nomad MP3 player Ids.

Submitted by: pete@altadena.net

23 years agoPrevious commit was:
peter [Wed, 17 Jan 2001 22:37:52 +0000 (22:37 +0000)]
Previous commit was:

Submitted by: Bart Thate <freebsd@1st.dudi.org>

23 years agoArgh! add card_if.h to the generated list. This fixes kernel builds for
peter [Wed, 17 Jan 2001 22:36:49 +0000 (22:36 +0000)]
Argh!  add card_if.h to the generated list.  This fixes kernel builds for
those who haven't added 'NO_MODULES=off' to /etc/make.conf.

23 years agoDocument the restrictions on changing times in high securelevels.
ben [Wed, 17 Jan 2001 21:27:54 +0000 (21:27 +0000)]
Document the restrictions on changing times in high securelevels.

PR: 23502
Submitted by: Robert Boyd <rboyd@pokerspot.com>

23 years agoDocument EHOSTDOWN error.
ben [Wed, 17 Jan 2001 21:02:50 +0000 (21:02 +0000)]
Document EHOSTDOWN error.

PR: 24410
Submitted by: Martin Horcicka <horcicka@vol.cz>

23 years agoClarify comments referring to strlcat() usage
jedgar [Wed, 17 Jan 2001 20:51:20 +0000 (20:51 +0000)]
Clarify comments referring to strlcat() usage

Obtained from: OpenBSD

23 years ago- Reverse the order of two loop invariant to ensure strlcat() does not
jedgar [Wed, 17 Jan 2001 20:51:16 +0000 (20:51 +0000)]
- Reverse the order of two loop invariant to ensure strlcat() does not
  attempt to read memory when siz is 0
- Clarify comments referring to strlcat() usage

PR: 24278, 24295
Submitted by: Tony Finch <dot@dotat.at>
Richard Kettlewell <rjk@greenend.org.uk>
Reviewed by: -audit

23 years agoNote fix for isp that made certain CDROM attach fail.
wilko [Wed, 17 Jan 2001 20:42:29 +0000 (20:42 +0000)]
Note fix for isp that made certain CDROM attach fail.

PR: 22650

23 years agoThe example install.cfg now lives in /usr/src/usr.sbin/sysinstall.
jhb [Wed, 17 Jan 2001 19:52:16 +0000 (19:52 +0000)]
The example install.cfg now lives in /usr/src/usr.sbin/sysinstall.

23 years agoLook for help files in /usr/src/usr.sbin/sysinstall instead of
jhb [Wed, 17 Jan 2001 19:51:08 +0000 (19:51 +0000)]
Look for help files in /usr/src/usr.sbin/sysinstall instead of
/usr/src/release/sysinstall.

23 years agoTell users to run /usr/sbin/sysinstall now instead of /stand/sysinstall.
jhb [Wed, 17 Jan 2001 19:29:53 +0000 (19:29 +0000)]
Tell users to run /usr/sbin/sysinstall now instead of /stand/sysinstall.

Reminded by: jkh

23 years agoMerge the documentation for sigsetmask() and sigblock() into a
ben [Wed, 17 Jan 2001 19:20:58 +0000 (19:20 +0000)]
Merge the documentation for sigsetmask() and sigblock() into a
single manual page, appropriately linked, since this removes the
decision of which page the (previously non-existent) sigmask.2
MLINK should point at.

Submitted by: will

23 years agoman(7) -> mdoc(7).
ru [Wed, 17 Jan 2001 18:26:21 +0000 (18:26 +0000)]
man(7) -> mdoc(7).

23 years agoAdd librpcsvc.
ru [Wed, 17 Jan 2001 18:19:06 +0000 (18:19 +0000)]
Add librpcsvc.

23 years agoCatchup from my holiday lull:
imp [Wed, 17 Jan 2001 17:59:09 +0000 (17:59 +0000)]
Catchup from my holiday lull:
o libcrypt not a symlink.
o hw.sndunit -> hw.snd.unit
o burncd / kernel changes
o I386_CPU now mutually exclusive with all other CPU types.

23 years agoTake advantage of the fixes to the pcic code that allows multiple
dmlb [Wed, 17 Jan 2001 17:55:00 +0000 (17:55 +0000)]
Take advantage of the fixes to the pcic code that allows multiple
active memory maps. This removes the need to change the memory
map from common to attribute every time a packet is sent/received.

This increases performance and decreases cpu load (ping times on
slow machines improve by about 1.5ms).

Move out the old common memory/attrbiute memory hack functions to a
new header file to tidy up the main code. I want to keep them available
for a while.

23 years agoActivate Greek message catalog
ache [Wed, 17 Jan 2001 16:54:43 +0000 (16:54 +0000)]
Activate Greek message catalog

23 years agoAdd Greek collate
ache [Wed, 17 Jan 2001 16:47:28 +0000 (16:47 +0000)]
Add Greek collate

PR: 24324
Submitted by: past@netmode.ntua.gr

23 years agoAdd Greek keyboard
ache [Wed, 17 Jan 2001 16:27:47 +0000 (16:27 +0000)]
Add Greek keyboard

PR: 24324
Submitted by: past@netmode.ntua.gr

23 years agoAdd iso07 fonts
ache [Wed, 17 Jan 2001 15:57:22 +0000 (15:57 +0000)]
Add iso07 fonts

PR: 24324
Submitted by: past@netmode.ntua.gr

23 years agoAdd Greek support
ache [Wed, 17 Jan 2001 13:43:20 +0000 (13:43 +0000)]
Add Greek support

PR: 24324
Submitted by: past@netmode.ntua.gr

23 years agoAdd Greek support
ache [Wed, 17 Jan 2001 13:30:16 +0000 (13:30 +0000)]
Add Greek support

PR: 24324
Submitted by: past@netmode.ntua.gr

23 years agoAdd el_GR.ISO_8859-7
ache [Wed, 17 Jan 2001 13:15:46 +0000 (13:15 +0000)]
Add el_GR.ISO_8859-7

23 years agoAdd NEWCARD hooks. This seems to work just fine.
peter [Wed, 17 Jan 2001 12:31:59 +0000 (12:31 +0000)]
Add NEWCARD hooks.  This seems to work just fine.

23 years agodocument ${KERNEL}
ben [Wed, 17 Jan 2001 11:51:43 +0000 (11:51 +0000)]
document ${KERNEL}

PR: 24404
Submitted by: Mike Meyer <mwm@mired.org>

23 years agorstat(1) and rstat_svc(8) are the early versions of
ru [Wed, 17 Jan 2001 11:50:42 +0000 (11:50 +0000)]
rstat(1) and rstat_svc(8) are the early versions of
the rup(1) and rpc.rstatd(8) manpages respectively.

23 years agoActually add build.7 to the Makefile, 2 months late.
ben [Wed, 17 Jan 2001 11:11:02 +0000 (11:11 +0000)]
Actually add build.7 to the Makefile, 2 months late.

PR: 24405
Submitted by: Mike Meyer <mwm@mired.org>

23 years agoAdd first cut support for the ATA100 capable VIA82c686b.
sos [Wed, 17 Jan 2001 09:22:13 +0000 (09:22 +0000)]
Add first cut support for the ATA100 capable VIA82c686b.

23 years agoman(7) -> mdoc(7).
ru [Wed, 17 Jan 2001 09:15:13 +0000 (09:15 +0000)]
man(7) -> mdoc(7).

23 years agomdoc(7) police: compact a few lists for better output.
ru [Wed, 17 Jan 2001 09:07:31 +0000 (09:07 +0000)]
mdoc(7) police: compact a few lists for better output.

23 years agomdoc(7) police: removed empty line (mdocNG does not like them),
ru [Wed, 17 Jan 2001 08:32:08 +0000 (08:32 +0000)]
mdoc(7) police: removed empty line (mdocNG does not like them),
columnate ENVIRONMENT variables.

23 years agoSysinstall has been repocopied to src/usr.sbin/sysinstall and everything
jhb [Wed, 17 Jan 2001 07:14:45 +0000 (07:14 +0000)]
Sysinstall has been repocopied to src/usr.sbin/sysinstall and everything
now uses the sources found there.

23 years ago- sysinstall now gets built as part of world, so don't throw in extra
jhb [Wed, 17 Jan 2001 07:13:40 +0000 (07:13 +0000)]
- sysinstall now gets built as part of world, so don't throw in extra
  commands to build sysinstall manually during release.
- Get the sysinstall help files from /usr/src/usr.sbin/sysinstall/help
  rather than ${.CURDIR}/sysinstall/help.

23 years agoNow that sysinstall lives in /usr/src/usr.sbin, we don't need to pull it
jhb [Wed, 17 Jan 2001 07:11:05 +0000 (07:11 +0000)]
Now that sysinstall lives in /usr/src/usr.sbin, we don't need to pull it
from /usr/src/release.

23 years agoRemoved unused copy of ../mount/getmtopts.3.
bde [Wed, 17 Jan 2001 04:10:06 +0000 (04:10 +0000)]
Removed unused copy of ../mount/getmtopts.3.

Removed used copies of ../mount/getmntopts.c ../mount/mntopts.h.  Use the
versions in ../mount directly like all other mount utilities.

Removed used copy of ../mount/pathnames.h.  Use the version in ../mount
for free as a side effect of using the mntopts files there.  We should
not use it at all, since the 2/3 of the definitions in it are in <paths.h>
and the other 1/3 should be in ../mountd/pathnames.h and is not used by
mount_ifs anyway.

23 years agoRemoved unused rotting copy of ../mount/mount.8. It should never have
bde [Wed, 17 Jan 2001 03:39:53 +0000 (03:39 +0000)]
Removed unused rotting copy of ../mount/mount.8.  It should never have
been repo-copied.

23 years agosysinstall.8 lives in src/usr.sbin/sysinstall now and gets installed via
jhb [Wed, 17 Jan 2001 03:33:04 +0000 (03:33 +0000)]
sysinstall.8 lives in src/usr.sbin/sysinstall now and gets installed via
normal means, so remove the hack to install it.

23 years agoActivate sysinstall.
jhb [Wed, 17 Jan 2001 03:32:19 +0000 (03:32 +0000)]
Activate sysinstall.

23 years ago- We live in /usr/sbin now, so don't set special veriables to build a
jhb [Wed, 17 Jan 2001 03:32:00 +0000 (03:32 +0000)]
- We live in /usr/sbin now, so don't set special veriables to build a
  static version that installs in /stand.  Also, don't use an extra
  before-install target to create /stand.
- Add missing $FreeBSD$.
- Fix dependencies to handle keymap.h. (*)

Submitted by: obrien (*)

23 years agoo When returning NULL, return (NULL) instead of return (0).
rwatson [Wed, 17 Jan 2001 02:40:39 +0000 (02:40 +0000)]
o When returning NULL, return (NULL) instead of return (0).

Submitted by: jedgar
Obtained from: TrustedBSD Project

23 years ago- Sort of lie and say that %eax is an output only and not an input for the
jhb [Wed, 17 Jan 2001 02:15:11 +0000 (02:15 +0000)]
- Sort of lie and say that %eax is an output only and not an input for the
  non-386 atomic_load_acq().  %eax is an input since its value is used in
  the cmpxchg instruction, but we don't care what value it is, so setting
  it to a specific value is just wasteful.  Thus, it is being used without
  being initialized as the warning stated, but it is ok for it to be used
  because its value isn't important.  Thus, we are only sort of lying when
  we say it is an output only operand.
- Add "cc" to the clobber list for atomic_load_acq() since the cmpxchgl
  changes ZF.

23 years agoThe stallion.c pci driver is broken. Emit a warning if compiling LINT.
peter [Wed, 17 Jan 2001 01:21:08 +0000 (01:21 +0000)]
The stallion.c pci driver is broken.  Emit a warning if compiling LINT.

23 years agoUse pci_get_revid() rather than groping around the PCI register space
peter [Wed, 17 Jan 2001 01:17:00 +0000 (01:17 +0000)]
Use pci_get_revid() rather than groping around the PCI register space
using the removed compatability register names.  (this broke LINT)

23 years agoFix breakage unconvered by LINT - dont refer to undefined variables in
peter [Wed, 17 Jan 2001 01:10:23 +0000 (01:10 +0000)]
Fix breakage unconvered by LINT - dont refer to undefined variables in
KASSERT()

23 years agoDocument data structures and operation on dummynet so next time
luigi [Wed, 17 Jan 2001 01:09:40 +0000 (01:09 +0000)]
Document data structures and operation on dummynet so next time
I or someone else browse through this code I do not have a hard
time understanding what is going on.

23 years agoMinor tweaks to get these to stop breaking LINT. They still dont work
peter [Wed, 17 Jan 2001 01:08:50 +0000 (01:08 +0000)]
Minor tweaks to get these to stop breaking LINT.  They still dont work
and emit warnings, but we need to get the test coverage elsewhere.

23 years agoSpecial case for compiling LINT - just give a warning and continue on.
peter [Wed, 17 Jan 2001 00:47:46 +0000 (00:47 +0000)]
Special case for compiling LINT - just give a warning and continue on.
At least we have a chance at getting test compile coverage for the rest
of the kernel now.

23 years agoRemove unneeded (and compile-breaking) #include "alpm.h"
peter [Wed, 17 Jan 2001 00:38:06 +0000 (00:38 +0000)]
Remove unneeded (and compile-breaking) #include "alpm.h"

23 years agoSome dummynet patches that I forgot to commit last summer.
luigi [Tue, 16 Jan 2001 23:49:49 +0000 (23:49 +0000)]
Some dummynet patches that I forgot to commit last summer.
One of them fixes a potential panic when bridging is used and
you run out of mbufs (though i have no idea if the bug has
ever hit anyone).

23 years agoMinor grammar nit.
ben [Tue, 16 Jan 2001 22:41:25 +0000 (22:41 +0000)]
Minor grammar nit.

PR: 23742
Submitted by: Rich Morin <rdm@cfcl.com>

23 years agoSIGABRT is *not* the same as calling abort(), so don't claim that it is.
ben [Tue, 16 Jan 2001 22:25:26 +0000 (22:25 +0000)]
SIGABRT is *not* the same as calling abort(), so don't claim that it is.
(abort() flushes all open stdio streams for one thing.)

PR: 24249
Submitted by: Edwin Groothuis <mavetju@chello.nl>

23 years agoOoops, the -M and -N flags were already documented, though not very clearly.
ben [Tue, 16 Jan 2001 22:09:33 +0000 (22:09 +0000)]
Ooops, the -M and -N flags were already documented, though not very clearly.
Remove the old description in favour of the new description which lists the
-M and -N flags along with all the other flags.  This is consistent with the
manual pages for ps, netstat, iostat, etc.

23 years agodocument -M and -N flags.
ben [Tue, 16 Jan 2001 22:05:28 +0000 (22:05 +0000)]
document -M and -N flags.

PR: 24323
Submitted by: Jesse Monroy <opentrax@email.com>

23 years agoThe bit about sigpending not detecting any errors is a lie, it can return
ben [Tue, 16 Jan 2001 21:57:42 +0000 (21:57 +0000)]
The bit about sigpending not detecting any errors is a lie, it can return
EFAULT.

PR: 24360
Submitted by: Kenneth Ingham <ingham@i-pi.com>

23 years agoFreeBSD doesn't run named in a sandbox by default, so change a comment so it
ben [Tue, 16 Jan 2001 20:57:18 +0000 (20:57 +0000)]
FreeBSD doesn't run named in a sandbox by default, so change a comment so it
doesn't imply we do.

23 years agoDeprecate the use of the CD_DRIVE, CDPLAY, DISC and MUSIC_CD environment
joe [Tue, 16 Jan 2001 20:31:53 +0000 (20:31 +0000)]
Deprecate the use of the CD_DRIVE, CDPLAY, DISC and MUSIC_CD environment
variables in favour of CDROM.

Discussed on: stable@FreeBSD.org

23 years agoAdd a sysvipc_enable variable that causes the SysV IPC modules to be loaded.
des [Tue, 16 Jan 2001 20:05:05 +0000 (20:05 +0000)]
Add a sysvipc_enable variable that causes the SysV IPC modules to be loaded.

Reviewed by: dougb

23 years agoman(7) -> mdoc(7).
ru [Tue, 16 Jan 2001 17:29:42 +0000 (17:29 +0000)]
man(7) -> mdoc(7).

23 years agoman(7) -> mdoc(7).
ru [Tue, 16 Jan 2001 17:04:17 +0000 (17:04 +0000)]
man(7) -> mdoc(7).

23 years agoAdd tags quirk for IBM DNES drives.
mjacob [Tue, 16 Jan 2001 17:02:16 +0000 (17:02 +0000)]
Add tags quirk for IBM DNES drives.

PR: 21139
Obtained from: campt@miralink.com

23 years agoman(7) -> mdoc(7).
ru [Tue, 16 Jan 2001 15:28:12 +0000 (15:28 +0000)]
man(7) -> mdoc(7).

23 years agoman(7) -> mdoc(7).
ru [Tue, 16 Jan 2001 13:12:21 +0000 (13:12 +0000)]
man(7) -> mdoc(7).

23 years agoEnhance find_matching_entry_incoming() to check if the controller, on
hm [Tue, 16 Jan 2001 13:10:51 +0000 (13:10 +0000)]
Enhance find_matching_entry_incoming() to check if the controller, on
which the call gets in, matches the one given in the entry section, or
-1 for any controller.

Submitted by: Oliver von Bueren <maillist@ovb.ch>

23 years agomdoc(7) police: fixed broken references.
ru [Tue, 16 Jan 2001 11:52:00 +0000 (11:52 +0000)]
mdoc(7) police: fixed broken references.

23 years agoChange NSWAPDEV to something else than the default value.
des [Tue, 16 Jan 2001 10:48:58 +0000 (10:48 +0000)]
Change NSWAPDEV to something else than the default value.

23 years agoCorrect spelling of millennium.
dwmalone [Tue, 16 Jan 2001 10:07:04 +0000 (10:07 +0000)]
Correct spelling of millennium.

PR: 24369
Submitted by: Tony Finch <dot@dotat.at>

23 years agoThese files have been on deathrow for a couple of months, no appeal.
phk [Tue, 16 Jan 2001 10:01:56 +0000 (10:01 +0000)]
These files have been on deathrow for a couple of months, no appeal.

23 years agoPrepare for mdoc(7)NG.
ru [Tue, 16 Jan 2001 09:39:23 +0000 (09:39 +0000)]
Prepare for mdoc(7)NG.

23 years agoPrepare for mdoc(7)NG.
ru [Tue, 16 Jan 2001 09:32:40 +0000 (09:32 +0000)]
Prepare for mdoc(7)NG.

23 years agoPrepare for mdoc(7)NG.
ru [Tue, 16 Jan 2001 09:24:36 +0000 (09:24 +0000)]
Prepare for mdoc(7)NG.

23 years agoPrepare for mdoc(7)NG.
ru [Tue, 16 Jan 2001 09:15:57 +0000 (09:15 +0000)]
Prepare for mdoc(7)NG.

23 years agoStop doing runtime checking on i386 cpus for cpu class. The cpu is
peter [Tue, 16 Jan 2001 09:10:34 +0000 (09:10 +0000)]
Stop doing runtime checking on i386 cpus for cpu class.  The cpu is
slow enough as it is, without having to constantly check that it really
is an i386 still.  It was possible to compile out the conditionals for
faster cpus by leaving out 'I386_CPU', but it was not possible to
unconditionally compile for the i386.  You got the runtime checking whether
you wanted it or not.  This makes I386_CPU mutually exclusive with the
other cpu types, and tidies things up a little in the process.

Reviewed by:  alfred, markm, phk, benno, jlemon, jhb, jake, grog, msmith,
              jasone, dcs, des (and a bunch more people who encouraged it)

23 years agoPrepare for mdoc(7)NG.
ru [Tue, 16 Jan 2001 09:08:22 +0000 (09:08 +0000)]
Prepare for mdoc(7)NG.

23 years agomdoc(7) police: use .Fa for function arguments, use .Vt for variable type.
ru [Tue, 16 Jan 2001 08:53:04 +0000 (08:53 +0000)]
mdoc(7) police: use .Fa for function arguments, use .Vt for variable type.

23 years ago- Fix a nasty bug where pkg_create doesn't gzip package if lengh of package
sobomax [Tue, 16 Jan 2001 08:27:28 +0000 (08:27 +0000)]
- Fix a nasty bug where pkg_create doesn't gzip package if lengh of package
  name is less than 5 and doesn't contain recognizeable suffix (one of .tar or
  .tgz), while gzip's it if lengh of the name greater than 4. For example
  `pkg_create [options] pkg1' will create pkg1.tar, while
  `pkg_create [options] pkg11' will create pkg11.tgz;
- use TRUE/FALSE as a values for boolean variables instead of explicit 1/0 and
  erroneous YES in one case.

MFC candidate.

23 years agoGuard against overflow of the calculated timeout value.
mjacob [Tue, 16 Jan 2001 07:15:36 +0000 (07:15 +0000)]
Guard against overflow of the calculated timeout value.

23 years agoArgh, disable the micro-ops again. I didn't test these adequately and
jhb [Tue, 16 Jan 2001 04:48:38 +0000 (04:48 +0000)]
Argh, disable the micro-ops again.  I didn't test these adequately and
managed to lock up one of my machines in world again.

Pointy-hat to: me

23 years ago- Use "+a" instead of "=&a" for several constraints. This should fix
jhb [Tue, 16 Jan 2001 03:45:54 +0000 (03:45 +0000)]
- Use "+a" instead of "=&a" for several constraints.  This should fix
  compiling errors where gcc would run out of registers.
- Add "cc" to the list of clobbers for micro-ops where we perform
  instructions that alter %eflags.
- Use xchgl instead of cmpxchgl to release a spin lock.  This could allow
  for more efficient register allocation as we no longer mandate that %eax
  be used.
- Reenable the optimized mutex micro-ops in the non-i386 case.