]> CyberLeo.Net >> Repos - FreeBSD/FreeBSD.git/log
FreeBSD/FreeBSD.git
26 years agoDocument behaviour of "-" and "--" on the command line.
Joseph Koshy [Thu, 30 Jul 1998 04:47:56 +0000 (04:47 +0000)]
Document behaviour of "-" and "--" on the command line.

PR: docs/5399

26 years agoDocument meanings of flags to the -r and -s options.
Joseph Koshy [Thu, 30 Jul 1998 04:17:12 +0000 (04:17 +0000)]
Document meanings of flags to the -r and -s options.

26 years agoadd anti-panic workaround from chris radek (cradek@in221.inetnebr.com)
Julian Elischer [Thu, 30 Jul 1998 03:22:52 +0000 (03:22 +0000)]
add anti-panic workaround from chris radek (cradek@in221.inetnebr.com)
Not sure why this is needed but but does stop crashes.

26 years agoTypo fix: teh --> (the|they)
Alexander Langer [Thu, 30 Jul 1998 02:27:41 +0000 (02:27 +0000)]
Typo fix: teh --> (the|they)

26 years agoRemove ppp 3000/tcp. Ppp has nothing in common with port 3000.
Brian Somers [Wed, 29 Jul 1998 19:30:44 +0000 (19:30 +0000)]
Remove ppp 3000/tcp.  Ppp has nothing in common with port 3000.

26 years agoAdd singlehand dvorak keyboard layouts, one for right hand use and
Søren Schmidt [Wed, 29 Jul 1998 18:49:45 +0000 (18:49 +0000)]
Add singlehand dvorak keyboard layouts, one for right hand use and
one for lefthand use.

Submitted by: "Chris Csanady" <cc@tarsier.ca.sandia.gov>

26 years agoFixed sign extension bugs awoken by changing speed_t to an unsigned
Bruce Evans [Wed, 29 Jul 1998 18:48:20 +0000 (18:48 +0000)]
Fixed sign extension bugs awoken by changing speed_t to an unsigned
type.  19200, 1200 and other relatively uninteresting speeds were
broken.

Submitted by: Rob Mallory <rmallory@qualcomm.com>

26 years agoOnly access an int for READU/WRITEU since that is what ptrace is declared to
Doug Rabson [Wed, 29 Jul 1998 18:41:30 +0000 (18:41 +0000)]
Only access an int for READU/WRITEU since that is what ptrace is declared to
return.

26 years agoDefault to FreeBSD if no brand detected. This makes life easier when
Doug Rabson [Wed, 29 Jul 1998 18:39:35 +0000 (18:39 +0000)]
Default to FreeBSD if no brand detected.  This makes life easier when
bootstrapping from NetBSD/alpha.

26 years agoFix msgbuf so that it actually works properly.
Doug Rabson [Wed, 29 Jul 1998 18:36:29 +0000 (18:36 +0000)]
Fix msgbuf so that it actually works properly.

26 years agoAllow an optional ``!'' in the open, dial & call commands.
Brian Somers [Wed, 29 Jul 1998 18:21:17 +0000 (18:21 +0000)]
Allow an optional ``!'' in the open, dial & call commands.
When used, the redial timer is ignored and the modem is
opened immediately.

26 years agoUse tcsetattr TCSANOW instead of TCSADRAIN so that dedicated mode doesn't
Brian Somers [Wed, 29 Jul 1998 18:20:53 +0000 (18:20 +0000)]
Use tcsetattr TCSANOW instead of TCSADRAIN so that dedicated mode doesn't
end up blocking indefinitely when there's no carrier.

26 years agoDon't try to dereference a pointer to freed memory.
Dag-Erling Smørgrav [Wed, 29 Jul 1998 17:50:06 +0000 (17:50 +0000)]
Don't try to dereference a pointer to freed memory.

PR: bin/7393

26 years agoFixed printf format errors.
Bruce Evans [Wed, 29 Jul 1998 17:38:14 +0000 (17:38 +0000)]
Fixed printf format errors.

26 years agoFixed print format errors.
Bruce Evans [Wed, 29 Jul 1998 16:43:00 +0000 (16:43 +0000)]
Fixed print format errors.

26 years agoFixed printf format errors.
Bruce Evans [Wed, 29 Jul 1998 15:50:41 +0000 (15:50 +0000)]
Fixed printf format errors.

Use offsetof() instead of null pointer hacks.  Use a home made offsetof()
because including <stddef.h> is not permitted in LKMs.

26 years agoHonor NOSHARED for tools. All tools should be built static for
Bruce Evans [Wed, 29 Jul 1998 14:19:48 +0000 (14:19 +0000)]
Honor NOSHARED for tools.  All tools should be built static for
`make world' to avoid problems with picking up (new) target or (stale)
host shared libraries.

Don't honor -static in LDFLAGS for linking LKMs.  LDFLAGS is not
actually for ld, but we use it anyway, and must prevent -static being
misinterpreted as -s.

Don't hide any of the link steps.

26 years agoPrint uid/gid as u_long per bde suggestion
Andrey A. Chernov [Wed, 29 Jul 1998 14:05:01 +0000 (14:05 +0000)]
Print uid/gid as u_long per bde suggestion

26 years agoFixed disordering in previous commit.
Bruce Evans [Wed, 29 Jul 1998 13:43:06 +0000 (13:43 +0000)]
Fixed disordering in previous commit.

26 years agoFixed error handling:
Bruce Evans [Wed, 29 Jul 1998 13:00:42 +0000 (13:00 +0000)]
Fixed error handling:
- Call isa_dmadone() whenever necessary to stop DMA and/or free bounce
  buffers.  Undead DMA corrupted the malloc freelist fairly consistently
  in the following configuration: SLICE kernel, 2 floppy drives, no disk
  in fd0, disk in fd1.
- Don't call fdc_reset() from fd_timeout().  Doing so gave an "extra"
  interrupt which was usually misinterpreted as being for completion
  of the next FDC command; the interrupt for completion of the next
  FDC command was then usually misinterpreted...  There were further
  complications for interrupts latched by the soft-spl mechanism so
  that they were delivered after all the h/w interrupts went away.
  This caused at least wrong head settle delays and may be why the
  FreeBSD floppy driver seems to munch floppies more than most floppy
  drivers.  The reset was unnecessary anyway in cases that didn't have
  the bug described next, since is was repeated a little later for
  the IOTIMEDOUT state.  The state machine has complications to handle
  resets correctly, so just use it.
- Don't call retrier() from fd_timeout().  The IOTIMEDOUT state needs
  to be processed next, and it isn't valid to set to that state if
  retrier() has aborted the current transfer.  Doing so caused null
  pointer panics after the previous bug was fixed.

Improved error handling:
- If an i/o is aborted, arrange to reset in the state machine before
  doing the next i/o.  New fdc flag for this.  This fixes spurious
  warnings and lengthy busy-waiting for the next i/o.
- Split STARTRECAL into RESETCOMPLETE and STARTRECAL and only check
  for the results from reset if we actually reset.  This fixes spurious
  warnings for other paths to STARTRECAL.  [Oops, it may break reset
  handling for motor-off resets.]

Cleanups in fd_timeout():
- Renamed to fd_iotimeout() to make it clearer that it is only used
  for i/o.
- Don't handle the bp == 0 case.  This case can't happen for i/o.
- Don't check for controller-busy.  We know it must be.
- Don't print anything.  retrier() already prints too much for normal
  errors.
- Fudge the state differently so that the state machine advances
  fdc->retry and the status is invalid (perhaps this should fudge a
  valid state like the one for WP).
- Style fixes.

26 years agoCentralized and optimized handling of large sectors. Centralized
Bruce Evans [Wed, 29 Jul 1998 11:15:54 +0000 (11:15 +0000)]
Centralized and optimized handling of large sectors.  Centralized
checking of transfer sizes and alignments.

Old version tested with 2K-sectors on od disks by: Shunsuke Akiyama
<akiyama@kme.mei.co.jp>.

26 years agoDocument the kern.logsigexit sysctl variable (sort of).
Joerg Wunsch [Wed, 29 Jul 1998 08:30:37 +0000 (08:30 +0000)]
Document the kern.logsigexit sysctl variable (sort of).

Reminded by: bde

26 years agoUse the slice-relative blkno in all parts of the label write
Bruce Evans [Wed, 29 Jul 1998 08:24:23 +0000 (08:24 +0000)]
Use the slice-relative blkno in all parts of the label write
protection checks.  Using the partition-relative blkno in some
parts broke the write protection for partitions at unusual
offsets (only for partitions at offset 1 on i386's).

26 years agoadd en ATM driver manual.
Kenjiro Cho [Wed, 29 Jul 1998 05:42:20 +0000 (05:42 +0000)]
add en ATM driver manual.

26 years agoadd Native ATM Protocol manual.
Kenjiro Cho [Wed, 29 Jul 1998 05:41:20 +0000 (05:41 +0000)]
add Native ATM Protocol manual.

26 years agoupdate ATM driver. (base version: midway.c 1.67 --> 1.68)
Kenjiro Cho [Wed, 29 Jul 1998 05:35:16 +0000 (05:35 +0000)]
update ATM driver. (base version: midway.c 1.67 --> 1.68)

several new features are added:
- support vc/vp shaping
- support pvc shadow interface

code cleanup:
- remove WMAYBE related code.  ENI WMAYBE DMA doen't work.
- remove updating if_lastchange for every packet.
- BPF related code is moved to midway.c as it should be.
  (bpfwrite should work if atm_pseudohdr and LLC/SNAP are
  prepended.)
- BPF link type is changed to DLT_ATM_RFC1483.
  BPF now understands only LLC/SNAP!! (because bpf can't
  handle variable link header length.)
  It is recommended to use LLC/SNAP instead of NULL
  encapsulation for various reasons.  (BPF, IPv6,
  interoperability, etc.)

the code has been used for months in ALTQ and KAME IPv6.

OKed by phk long time ago.

26 years agoAdd missing period. We don't allow maternity leaves.
Tim Vanderhoek [Wed, 29 Jul 1998 05:13:39 +0000 (05:13 +0000)]
Add missing period.  We don't allow maternity leaves.

26 years agoMAXPATHLEN is long enough
Warner Losh [Wed, 29 Jul 1998 03:31:03 +0000 (03:31 +0000)]
MAXPATHLEN is long enough

26 years agowhen changing mktemp -> mkstemp, make sure to include the new s
Warner Losh [Wed, 29 Jul 1998 03:25:08 +0000 (03:25 +0000)]
when changing mktemp -> mkstemp, make sure to include the new s

26 years agoFix temp file race in unused file. If someone wanted to clean the
Warner Losh [Wed, 29 Jul 1998 03:23:18 +0000 (03:23 +0000)]
Fix temp file race in unused file.  If someone wanted to clean the
tree a little, many files here are likely ripe for removal...

26 years agoAllow env. variable LS_COLWIDTHS to specify minimum column widths,
Tim Vanderhoek [Wed, 29 Jul 1998 00:46:13 +0000 (00:46 +0000)]
Allow env. variable LS_COLWIDTHS to specify minimum column widths,
effectively overriding the dynamically-sized-column feature.  This
is mostly useful for non-interactive use, where it may be necessary
to ensure that listings taken at different times have columns that
line-up correctly.  I have been assured that at least one large,
well-known program will soon be taking advantage of this.  :-)

PR: bin/7011
Submitted by: Joel Ray Holveck <joelh@gnu.org>

26 years agoMake the logging of abnormally exiting processes optional by a sysctl.
Joerg Wunsch [Tue, 28 Jul 1998 22:34:12 +0000 (22:34 +0000)]
Make the logging of abnormally exiting processes optional by a sysctl.
PR: kern/1711
Submitted by: Nick Sayer <nsayer@kfu.com>

26 years agoAdd var "CD_MOUNTPT" set to /cdrom by default.
David E. O'Brien [Tue, 28 Jul 1998 22:27:16 +0000 (22:27 +0000)]
Add var "CD_MOUNTPT" set to /cdrom by default.
Ok'ed by:  discussed with Satoshi at USENIX

26 years agopseudo-device ssc requires su.c (pseudo-device su)
Brian Somers [Tue, 28 Jul 1998 22:06:16 +0000 (22:06 +0000)]
pseudo-device ssc requires su.c (pseudo-device su)
Submitted by: woods@zeus.leitch.com (Greg A. Woods)
PR: 7413

26 years agoSilence ``Network unreachable'' warnings when using
Brian Somers [Tue, 28 Jul 1998 21:54:54 +0000 (21:54 +0000)]
Silence ``Network unreachable'' warnings when using
``add .... HISADDR''.  The network will never be
reachable at this point unless we're in -auto or reading
the command from ppp.linkup.

We can now run the following lines and get the expected
results:

  set ifaddr 1.2.3.4/0 5.6.7.8/0
  add default HISADDR

where a route is added immediately in auto mode and the
whole thing is delayed 'till the IP numbers have been
agreed in other modes.

Essentially, ppp.linkup is no longer required.

26 years agoo Tidy up PAP and CHAP diagnostics. They're now all logged as PHASE
Brian Somers [Tue, 28 Jul 1998 21:54:30 +0000 (21:54 +0000)]
o Tidy up PAP and CHAP diagnostics.  They're now all logged as PHASE
  diagnostics (which are on by default).
o Deal correctly with both sides wanting CHAP.
o Output a warning if we're using an empty ``authname''.  This is
  *not* what we want to do.

26 years agoSet bp->b_resid for failed transfers in dscheck(). This is the
Bruce Evans [Tue, 28 Jul 1998 19:39:09 +0000 (19:39 +0000)]
Set bp->b_resid for failed transfers in dscheck().  This is the
best place to set it, and the wd and wfd strategy routines don't
set it (for failed transfers) because they expect dscheck() to
initialize everything necessary.  dscheck() has always set B_ERROR,
but this is not quite sufficient, because b_resid is used by physio()
to decide how much of a B_ERROR'ed i/o was done.

26 years agoSet the disk type to SCSI in the in-core label for the whole disk.
Bruce Evans [Tue, 28 Jul 1998 18:59:49 +0000 (18:59 +0000)]
Set the disk type to SCSI in the in-core label for the whole disk.

26 years agoCommit patch from Tor Egge to fix the "large filesystem restore" problem.
Warner Losh [Tue, 28 Jul 1998 18:50:01 +0000 (18:50 +0000)]
Commit patch from Tor Egge to fix the "large filesystem restore" problem.
This appears to work for me in the old case, but I don't have large
enough filesystems to test the fix case.

Reported working by: karl@mcs.net

26 years agoUsed daddr_t's, not ints, to store disk block numbers. Updated printf
Bruce Evans [Tue, 28 Jul 1998 18:25:51 +0000 (18:25 +0000)]
Used daddr_t's, not ints, to store disk block numbers.  Updated printf
formats and args to match.  Fixed old printf format errors (all related;
most were hidden by calling printf indirectly).

This change somehow avoids compiler bugs for 64-bit longs on i386's,
although it increases the number of 64-bit calculations.

26 years agoSet p->p_switchtime to switchtime instead of to the current time in
Bruce Evans [Tue, 28 Jul 1998 17:55:37 +0000 (17:55 +0000)]
Set p->p_switchtime to switchtime instead of to the current time in
fork_trampoline() if switchtime is valid.  This fixes not accounting
for the time between the previous context switch and and the current
time (when the forked child starts up here) in most cases - the time
is now counted in the child's runtime.  I think it actually fixes
all cases, and switchtime is always valid here, since there must have
been a context switch just before the forked child starts up.  Some
code should be removed if this is correct.  The check that switchtime
is valid sometimes gives a false negative because the check isn't
correct until the after the first context switch after the system
has been up for >= 1 second.

26 years agoMicro-optimized and cleaned up the clearing of switchtime in idle().
Bruce Evans [Tue, 28 Jul 1998 17:35:09 +0000 (17:35 +0000)]
Micro-optimized and cleaned up the clearing of switchtime in idle().

Cleaned up the conditionals in the disgusting SMP ifdef in idle().

26 years agoFixed two spl nesting bugs. They caused (at least) the entire pageout
Bruce Evans [Tue, 28 Jul 1998 15:30:01 +0000 (15:30 +0000)]
Fixed two spl nesting bugs.  They caused (at least) the entire pageout
daemon to run at splvm() forever after swap_pager_putpages() is called
from vm_pageout_scan().

Broken in: rev.1.189 (1998/02/23)

26 years agou_int --> unsigned int, remove (now unneeded) <sys/types.h>
Jonathan Lemon [Tue, 28 Jul 1998 15:22:51 +0000 (15:22 +0000)]
u_int --> unsigned int, remove (now unneeded) <sys/types.h>

26 years agoDon't exit() from cleanup() if we got there via a non-signal since
Jordan K. Hubbard [Tue, 28 Jul 1998 11:55:39 +0000 (11:55 +0000)]
Don't exit() from cleanup() if we got there via a non-signal since
we're only going to exit (after printing a more detailed error message)
afterwards anyway.
PR: 7020

26 years agoBrucifixion.
Dag-Erling Smørgrav [Tue, 28 Jul 1998 10:08:16 +0000 (10:08 +0000)]
Brucifixion.

26 years agoRecord the new physical address of the pcb in pmap_swapin_proc.
Doug Rabson [Tue, 28 Jul 1998 09:34:50 +0000 (09:34 +0000)]
Record the new physical address of the pcb in pmap_swapin_proc.
Third attempt at pmap_remove().  This one actually works properly :-).

26 years agoI have a PD drive (Matsushita PD/CD Drive LF-1600).
Poul-Henning Kamp [Tue, 28 Jul 1998 09:03:37 +0000 (09:03 +0000)]
I have a PD drive (Matsushita PD/CD Drive LF-1600).
The machine can not recognize this PD drive
as neither PD drive nor CD-ROM Drive.
So I can not use CD-ROMs and PDs from FreeBSD.

PR: 7423
Reviewed by: phk
Submitted by: Takura Koyama <takura@popweb.ne.jp>

26 years agoSpelling. Errx() on malloc() failure.
Philippe Charnier [Tue, 28 Jul 1998 06:42:38 +0000 (06:42 +0000)]
Spelling. Errx() on malloc() failure.

26 years agoCorrect use of .Nm, add rcsid, remove unused #include.
Philippe Charnier [Tue, 28 Jul 1998 06:38:57 +0000 (06:38 +0000)]
Correct use of .Nm, add rcsid, remove unused #include.

26 years agoSpelling, add rcsid, remove unused #includes.
Philippe Charnier [Tue, 28 Jul 1998 06:36:31 +0000 (06:36 +0000)]
Spelling, add rcsid, remove unused #includes.
Convert 1000000 usec to 1 sec 0 usec.
Use provided safe malloc (rtmalloc()) instead of malloc(): exit on allocation
failure.
Correct use of .Nm
Add usage() and use errx().

26 years agoComment is .\" not .|'. Add -d and -t flags to SYNOPSIS and usage(). Getopt
Philippe Charnier [Tue, 28 Jul 1998 06:25:35 +0000 (06:25 +0000)]
Comment is .\" not .|'. Add -d and -t flags to SYNOPSIS and usage(). Getopt
returns -1. Correct use of .Nm. Spelling. Add rcsid and remove unused
#include.

26 years agoAdd rcsid. Remove unused #includes. Add missing prototypes and others -Wall
Philippe Charnier [Tue, 28 Jul 1998 06:20:16 +0000 (06:20 +0000)]
Add rcsid. Remove unused #includes. Add missing prototypes and others -Wall
cleanings. Spelling.

26 years agoChange doscmd to use i386_set_ioperm() call instead of sysarch().
Jonathan Lemon [Tue, 28 Jul 1998 03:39:59 +0000 (03:39 +0000)]
Change doscmd to use i386_set_ioperm() call instead of sysarch().

26 years agooops, forgot to do ``cvs add'' first.
Jonathan Lemon [Tue, 28 Jul 1998 03:39:04 +0000 (03:39 +0000)]
oops, forgot to do ``cvs add'' first.

26 years agoAdd wrappers for i386_*_ioperm, i386_vm86 so userland code does
Jonathan Lemon [Tue, 28 Jul 1998 03:33:27 +0000 (03:33 +0000)]
Add wrappers for i386_*_ioperm, i386_vm86 so userland code does
not have to call sysarch() directly.
Added man pages for above, as well as sysarch()

26 years agoFix an off-by-one error when setting the iomap bits.
Jonathan Lemon [Tue, 28 Jul 1998 03:29:32 +0000 (03:29 +0000)]
Fix an off-by-one error when setting the iomap bits.
Change struct i386_*_iomap to use ints instead of shorts/chars.
  (pointed out by bde long ago, prodded into action by msmith)

26 years agocast arg to (long) to match format
Andrey A. Chernov [Tue, 28 Jul 1998 01:30:16 +0000 (01:30 +0000)]
cast arg to (long) to match format

26 years agoParanoia: use mkdtemp instead of mktemp
Jacques Vidrine [Tue, 28 Jul 1998 01:18:02 +0000 (01:18 +0000)]
Paranoia: use mkdtemp instead of mktemp
PR: bin/3212
Reviewed by: jkh@FreeBSD.ORG

26 years agomake a bounds file for msgs(1) if there isn't one already
Jacques Vidrine [Tue, 28 Jul 1998 01:15:12 +0000 (01:15 +0000)]
make a bounds file for msgs(1) if there isn't one already
PR: bin/3023
Reviewed by: jkh@FreeBSD.ORG

26 years agoStyle fixes and a bug fix: don't remove the exit handler if unmount
Alexander Langer [Mon, 27 Jul 1998 22:47:17 +0000 (22:47 +0000)]
Style fixes and a bug fix: don't remove the exit handler if unmount
fails.

Submitted by: bde

26 years agoParanoia: use mkstemp instead of mktemp.
Jacques Vidrine [Mon, 27 Jul 1998 22:26:25 +0000 (22:26 +0000)]
Paranoia: use mkstemp instead of mktemp.
PR: bin/3211
Reported by: Mark Pritchard <mpp@FreeBSD.ORG>

26 years agoDocument the printf built-in.
Jacques Vidrine [Mon, 27 Jul 1998 21:31:19 +0000 (21:31 +0000)]
Document the printf built-in.

26 years agocsh built-in printf now expands arguments. e.g.
Jacques Vidrine [Mon, 27 Jul 1998 21:15:52 +0000 (21:15 +0000)]
csh built-in printf now expands arguments. e.g.
  printf '%d\n' `echo 23`
now works as expected.
PR: bin/2969
Submitted by: Gareth McCaughan <gjm11@dpmms.cam.ac.uk>

26 years agoFixed double counting of runtime after a process exits. The last
Bruce Evans [Mon, 27 Jul 1998 19:16:21 +0000 (19:16 +0000)]
Fixed double counting of runtime after a process exits.  The last
timeslice of the exiting process was counted for both the exiting
process and the next process to run if the next process runs
immediately.

Broken in: mostly in kern_clock.c rev.1.70 (1998/05/28)

26 years agoUse an error message that more clearly indites(sp?) an out-of-sync
Tim Vanderhoek [Mon, 27 Jul 1998 18:53:45 +0000 (18:53 +0000)]
Use an error message that more clearly indites(sp?) an out-of-sync
kernel+userland when that is the case.

PR: bin/7414

26 years agofixed: tar will extract file even if -C command fails
Jacques Vidrine [Mon, 27 Jul 1998 18:40:09 +0000 (18:40 +0000)]
fixed: tar will extract file even if -C command fails
tar now exits with new exit code EX_BADDIR after a failed chdir()
in name_next(), name_match(), and name_from_list().
PR: bin/2394
Submitted by: Satoshi Asami <asami@freebsd.org>

26 years agoBehave like GNU time. More cleanup. Reword man page.
Dag-Erling Smørgrav [Mon, 27 Jul 1998 16:54:05 +0000 (16:54 +0000)]
Behave like GNU time. More cleanup. Reword man page.

Pointed out by: Bruce Evans

26 years agoExtend cpl workaround so that it applies when we are returning to
Jonathan Lemon [Mon, 27 Jul 1998 16:51:33 +0000 (16:51 +0000)]
Extend cpl workaround so that it applies when we are returning to
user-mode as well as vm86 mode.

26 years agoRe-arrange the page layout used by vm86_bioscall so that we can
Jonathan Lemon [Mon, 27 Jul 1998 16:45:05 +0000 (16:45 +0000)]
Re-arrange the page layout used by vm86_bioscall so that we can
potentially re-use the stack page.

Cosmetic cleanup of the code to de-obfuscate it and make it easier
to follow.  There should be no functional changes in this commit.

26 years agoClean up the previous commit.
Dag-Erling Smørgrav [Mon, 27 Jul 1998 16:08:58 +0000 (16:08 +0000)]
Clean up the previous commit.

26 years agoMade lazy syncing of timestamps for special files non-optional.
Bruce Evans [Mon, 27 Jul 1998 15:37:00 +0000 (15:37 +0000)]
Made lazy syncing of timestamps for special files non-optional.

26 years agoHarlan.Stenn@pfcs.com added two (very interesting) options to syslogd for
Poul-Henning Kamp [Mon, 27 Jul 1998 13:04:14 +0000 (13:04 +0000)]
Harlan.Stenn@pfcs.com added two (very interesting) options to syslogd for
-current (Thanks Harald). However, on my attempt to try this on -STABLE,
I found that when forwarding to another host the actual messages gets lost.
This is due to a wrong index because when the -v option was added, the
indexes shifted one place.

PR: 7407
Submitted by: Andre Albsmeier <andre.albsmeier@mchp.siemens.de>

26 years agoRound - not cut - the real cpu time.
Wolfram Schneider [Mon, 27 Jul 1998 12:21:58 +0000 (12:21 +0000)]
Round - not cut - the real cpu time.

26 years agoSync with sys/i386/isa/diskslice_machdep.c revision 1.30.
KATO Takenori [Mon, 27 Jul 1998 09:49:23 +0000 (09:49 +0000)]
Sync with sys/i386/isa/diskslice_machdep.c revision 1.30.

26 years agoMacros for accessing alpha sparse device ports and memory.
Doug Rabson [Mon, 27 Jul 1998 09:40:35 +0000 (09:40 +0000)]
Macros for accessing alpha sparse device ports and memory.

26 years agoSupport the disabled keyword for isa devices.
Doug Rabson [Mon, 27 Jul 1998 09:38:26 +0000 (09:38 +0000)]
Support the disabled keyword for isa devices.

26 years agoOnly call m_reclaim() if M_WAIT since calling it from an interrupt can
David Greenman [Mon, 27 Jul 1998 03:59:48 +0000 (03:59 +0000)]
Only call m_reclaim() if M_WAIT since calling it from an interrupt can
cause problems.
PR: 7403

26 years agoA better solution to the rm_at_exit problem: Register the exit function
Alexander Langer [Mon, 27 Jul 1998 01:07:01 +0000 (01:07 +0000)]
A better solution to the rm_at_exit problem: Register the exit function
during first mount.  Unregister the exit function at last unmount.

Concept by: sef
Reviewed by: sef
Implemented by: alex

26 years agofd_set is documented to be a typedef, not a struct. Conform more
Warner Losh [Sun, 26 Jul 1998 18:49:36 +0000 (18:49 +0000)]
fd_set is documented to be a typedef, not a struct.  Conform more
closely to the documented interface.

Inspired-by: similar changes in OpenBSD
26 years agoFix a stupid bug preventing gdb from seeing the right value of 'v0'.
Doug Rabson [Sun, 26 Jul 1998 18:47:47 +0000 (18:47 +0000)]
Fix a stupid bug preventing gdb from seeing the right value of 'v0'.

26 years agoNotify pmap when a page is freed on the alpha to allow it to clean up
Doug Rabson [Sun, 26 Jul 1998 18:15:20 +0000 (18:15 +0000)]
Notify pmap when a page is freed on the alpha to allow it to clean up
its emulated modified/referenced bits.

26 years agoFix a stupid bug in the new pmap_remove().
Doug Rabson [Sun, 26 Jul 1998 18:13:28 +0000 (18:13 +0000)]
Fix a stupid bug in the new pmap_remove().
Arrange for soft modified/referenced flags to be cleared when a page is freed.

26 years agoUse malloc + sprintf rather than asprintf to ensure portability to
Warner Losh [Sun, 26 Jul 1998 17:06:05 +0000 (17:06 +0000)]
Use malloc + sprintf rather than asprintf to ensure portability to
other, less advanced architecutres.  This should minorly help porting
efforts of FreeBSD.  I've done several make worlds since this came up
with this change, as well as debugging several interesting nits with
-V (which is the only thing this change will affect really).

26 years agoRemove comment relevant to 16bit integers only.
Warner Losh [Sun, 26 Jul 1998 16:49:20 +0000 (16:49 +0000)]
Remove comment relevant to 16bit integers only.

26 years agoFix compilation for the case where neither ioctl(TIOCGWINSZ)
Tim Vanderhoek [Sun, 26 Jul 1998 16:15:26 +0000 (16:15 +0000)]
Fix compilation for the case where neither ioctl(TIOCGWINSZ)
nor ioctl(WIOCGETD) exist.  This case doesn't occur under FreeBSD.

PR: bin/5996
Submitted by: Max Euston <meuston@jmrodgers.com>

26 years agoo Fix Y2K buffer overflow.
Warner Losh [Sun, 26 Jul 1998 16:00:16 +0000 (16:00 +0000)]
o Fix Y2K buffer overflow.
o Fix zero filling bug in all years % 100 < 10.

26 years agoY2K fix. Fix by jsm28@cam.ac.uk via OpenBSD
Warner Losh [Sun, 26 Jul 1998 15:30:48 +0000 (15:30 +0000)]
Y2K fix.  Fix by jsm28@cam.ac.uk via OpenBSD

26 years agoAdd Xref to ldd(1).
Joseph Koshy [Sun, 26 Jul 1998 04:16:04 +0000 (04:16 +0000)]
Add Xref to ldd(1).

PR: 7397

26 years agoDon't pass the label to diskerr(), since the label is being constructed
Bruce Evans [Sat, 25 Jul 1998 16:35:06 +0000 (16:35 +0000)]
Don't pass the label to diskerr(), since the label is being constructed
and may be invalid.  In particular, d_secpercyl may be 0, and diskerr()
divides by it.

26 years agoFixed bugs in `disklabel -R...':
Bruce Evans [Sat, 25 Jul 1998 16:19:10 +0000 (16:19 +0000)]
Fixed bugs in `disklabel -R...':
- nonstandard sector sizes didn't work because the sector size in the
  ASCII label was not read before a (default, wrong) sector size was used.
- the exit status was 0 after exiting early due to an invalid ASCII label.

Abort `disklabel -B...' if the secondary bootstrap doesn't fit.

26 years agoOverride the default VFS LKM dispatch functions so that a module
Alexander Langer [Sat, 25 Jul 1998 15:52:44 +0000 (15:52 +0000)]
Override the default VFS LKM dispatch functions so that a module
unload function can be provided (this is necessary to unregister
the at_exit handler).

26 years agoAllow VFS LKMs to override the default module dispatch functions if
Alexander Langer [Sat, 25 Jul 1998 15:52:10 +0000 (15:52 +0000)]
Allow VFS LKMs to override the default module dispatch functions if
VFS_LKM_NO_DEFAULT_DISPATCH is defined.

26 years agoMake __weak_reference work on the alpha.
Doug Rabson [Sat, 25 Jul 1998 14:37:37 +0000 (14:37 +0000)]
Make __weak_reference work on the alpha.

26 years agoReally delete -# option.
Tim Vanderhoek [Sat, 25 Jul 1998 05:58:04 +0000 (05:58 +0000)]
Really delete -# option.

26 years agoNuke obsolote and broken -# option.
Tim Vanderhoek [Sat, 25 Jul 1998 05:56:49 +0000 (05:56 +0000)]
Nuke obsolote and broken -# option.
Order options.

PR: bin/5996
Submitted by: Max Euston <meuston@jmrodgers.com>

26 years agoKeep "-c" (top_scroll) through viewing help.
Tim Vanderhoek [Sat, 25 Jul 1998 05:28:32 +0000 (05:28 +0000)]
Keep "-c" (top_scroll) through viewing help.

PR: bin/5996
Submitted by: Max Euston <meuston@jmrodgers.com>

26 years agominphys means something different in FreeBSD than in NetBSD- not needed here and...
Matt Jacob [Sat, 25 Jul 1998 04:48:25 +0000 (04:48 +0000)]
minphys means something different in FreeBSD than in NetBSD- not needed here and in the way

26 years agoFix pmap_enter_quick() to stop it hanging when the level 2 pte exists
Doug Rabson [Fri, 24 Jul 1998 09:43:27 +0000 (09:43 +0000)]
Fix pmap_enter_quick() to stop it hanging when the level 2 pte exists
but isn't valid.

Reimplement pmap_remove() to be much more efficient at removing large
stretches of addresses.

As part of reimplementing pmap_remove() fix pmap_protect() so that it stands
a hope of working.

26 years agoAdd (struct timezone *) cast to NULL for K&R
Andrey A. Chernov [Fri, 24 Jul 1998 08:28:33 +0000 (08:28 +0000)]
Add (struct timezone *) cast to NULL for K&R