]> CyberLeo.Net >> Repos - FreeBSD/FreeBSD.git/log
FreeBSD/FreeBSD.git
26 years agoDocument the use of sendmail options on the commandline.
Steve Price [Tue, 14 Apr 1998 00:59:29 +0000 (00:59 +0000)]
Document the use of sendmail options on the commandline.

PR: 4778
Submitted by: Ruslan Ermilov <ru@ucb.crimea.ua>

26 years agoDon't let secs variable wrap if <= 2.
Steve Price [Tue, 14 Apr 1998 00:23:54 +0000 (00:23 +0000)]
Don't let secs variable wrap if <= 2.

PR: 6290
Submitted by: Ruslan Ermilov <ru@ucb.crimea.ua>

26 years agoAdd tutorials directories.
Wolfram Schneider [Mon, 13 Apr 1998 21:44:46 +0000 (21:44 +0000)]
Add tutorials directories.

26 years agoAdded EIDRM & ENOMSG errno in translation table.
Søren Schmidt [Mon, 13 Apr 1998 17:52:41 +0000 (17:52 +0000)]
Added EIDRM & ENOMSG errno in translation table.

26 years agoAdded EIDRM & ENOMSG to errno translation table.
Søren Schmidt [Mon, 13 Apr 1998 17:49:51 +0000 (17:49 +0000)]
Added EIDRM & ENOMSG to errno translation table.

26 years agoDon't use INTR when only one device supports it.
Mike Smith [Mon, 13 Apr 1998 17:45:56 +0000 (17:45 +0000)]
Don't use INTR when only one device supports it.
Submitted by: Satoh Junichi <junichi@astec.co.jp>

26 years agoAdd EIDRM errno (PR 176), ENOMSG errno (myself) for prober sysv_ipc.
Søren Schmidt [Mon, 13 Apr 1998 17:45:00 +0000 (17:45 +0000)]
Add EIDRM errno (PR 176), ENOMSG errno (myself) for prober sysv_ipc.

26 years agoWrong header length used for certain reassembled IP packets.
Poul-Henning Kamp [Mon, 13 Apr 1998 17:27:08 +0000 (17:27 +0000)]
Wrong header length used for certain reassembled IP packets.
PR: 6177
Reviewed by: phk, wollman
Submitted by: Eric Sprinkle <eric@ennovatenetworks.com>

26 years agoFinished previous commit. Just revert to rev.1.9.
Bruce Evans [Mon, 13 Apr 1998 15:05:43 +0000 (15:05 +0000)]
Finished previous commit.  Just revert to rev.1.9.

Bruce

26 years agoDocument the requirement for TCPMUX to also be enabled as an internal
Garrett Wollman [Mon, 13 Apr 1998 15:05:14 +0000 (15:05 +0000)]
Document the requirement for TCPMUX to also be enabled as an internal
service if any external TCPMUX servers are desired.

PR: 826

26 years agoFixed autonegotiation. Card registers are now accessed via memory
Semen Ustimenko [Mon, 13 Apr 1998 14:15:40 +0000 (14:15 +0000)]
Fixed  autonegotiation. Card registers are now accessed via memory
not i/o space.

26 years ago-pg was causing a link with -lc_r... :-]
Peter Wemm [Mon, 13 Apr 1998 14:12:34 +0000 (14:12 +0000)]
-pg was causing a link with -lc_r... :-]

Submitted by:  Dmitry Khrustalev <dima@xyzzy.machaon.ru>
PR: 6287

26 years agoFixes for using amd with non-NFSv3 servers and for choosing the right address
Doug Rabson [Mon, 13 Apr 1998 12:11:24 +0000 (12:11 +0000)]
Fixes for using amd with non-NFSv3 servers and for choosing the right address
for a multihomed server.

PR: bin/6231, bin/4678
Submitted by: Mika Nystrom <mika@cs.caltech.edu

26 years agoSync with sys/i386/isa/wd.c revision 1.156.
KATO Takenori [Mon, 13 Apr 1998 08:35:37 +0000 (08:35 +0000)]
Sync with sys/i386/isa/wd.c revision 1.156.

26 years agoFormat changes in an attempt to address some of Bruce's comments
John Birrell [Sun, 12 Apr 1998 20:29:24 +0000 (20:29 +0000)]
Format changes in an attempt to address some of Bruce's comments
about spaces and tabs.

The externs in the static inline functions remain 'cause (a) they're
required; and (b) I can't find any gcc -W* cases where they generate
warnings.

26 years agoFix dependent-package loading when pkg_adding with an FTP URL. I broke
Jordan K. Hubbard [Sun, 12 Apr 1998 16:01:10 +0000 (16:01 +0000)]
Fix dependent-package loading when pkg_adding with an FTP URL.  I broke
this when I changed the playpen code to disallow recursion.  I know
this fixes a PR someplace but I can't seem to find it with the GNATs
search tool.

26 years agoFix comment.
Jordan K. Hubbard [Sun, 12 Apr 1998 14:02:33 +0000 (14:02 +0000)]
Fix comment.
Scolded by: bde

26 years agoomehow Jordan got 07.lpd in -current enabled again...
Søren Schmidt [Sun, 12 Apr 1998 12:33:59 +0000 (12:33 +0000)]
omehow Jordan got 07.lpd in -current enabled again...

26 years agoAic APM driver support for e.g. Toshiba T2400 notebook with internal
Poul-Henning Kamp [Sun, 12 Apr 1998 12:11:31 +0000 (12:11 +0000)]
Aic APM driver support for e.g. Toshiba T2400 notebook with internal
Adaptec SCSI (aic6360/aha152x) controller.

PR: 5382
Reviewed by: phk
Submitted by: Oliver Breuninger <ob@seicom.NET>

26 years agoopenlog() needs to have LOG_NDELAY added, or else the syslog() calls after
Poul-Henning Kamp [Sun, 12 Apr 1998 11:15:54 +0000 (11:15 +0000)]
openlog() needs to have LOG_NDELAY added, or else the syslog() calls after
the chroot will not get sent to syslogd.

PR: 4910
Reviewed by: phk
Submitted by: Jim Mercer <jim@komodo.reptiles.org>

26 years agoRemove these copies, they now live the right place.
Poul-Henning Kamp [Sun, 12 Apr 1998 10:45:11 +0000 (10:45 +0000)]
Remove these copies, they now live the right place.

26 years agoChange noc-tun to nos-tun after Peter did the repository copy to cover
Poul-Henning Kamp [Sun, 12 Apr 1998 10:44:24 +0000 (10:44 +0000)]
Change noc-tun to nos-tun after Peter did the repository copy to cover
up the tracks of my blunder.

26 years agoEnable the SecureRPC bits in rc.conf, if the Administrator wants them.
Mark Murray [Sun, 12 Apr 1998 09:47:43 +0000 (09:47 +0000)]
Enable the SecureRPC bits in rc.conf, if the Administrator wants them.

26 years agoSync with sys/i386/isa/wd.c revision 1.155.
KATO Takenori [Sun, 12 Apr 1998 05:05:19 +0000 (05:05 +0000)]
Sync with sys/i386/isa/wd.c revision 1.155.

26 years agoFix the problem when SCSI ID is not contiguous.
KATO Takenori [Sun, 12 Apr 1998 04:48:11 +0000 (04:48 +0000)]
Fix the problem when SCSI ID is not contiguous.

Submitted by: URATA Shuichiro <s-urata@nmit.tmg.nec.co.jp>

26 years agoAdd aio_error decl.
John Dyson [Sun, 12 Apr 1998 03:09:43 +0000 (03:09 +0000)]
Add aio_error decl.

26 years agoImplement Helbig idea of offset calculation reducing mktime() calls
Andrey A. Chernov [Sat, 11 Apr 1998 21:18:55 +0000 (21:18 +0000)]
Implement Helbig idea of offset calculation reducing mktime() calls
Immediately exit if /etc/wall_cmos_clock not present

26 years agoIn these enlightened times there are more and better places to find
Poul-Henning Kamp [Sat, 11 Apr 1998 20:45:16 +0000 (20:45 +0000)]
In these enlightened times there are more and better places to find
this information.  PR 2789 mentions:

     http://www.pacbell.com/about-pb/areacodes/

     http://www.bellcore.com/NANP/

PR: 2789 (in a backewards kind of way)

26 years agosetsockopt() transports user option data in an mbuf. if the user
Poul-Henning Kamp [Sat, 11 Apr 1998 20:31:46 +0000 (20:31 +0000)]
setsockopt() transports user option data in an mbuf. if the user
data is greater than MLEN, setsockopt is unable to pass it onto
the protocol handler.  Allocate a cluster in such case.

PR: 2575
Reviewed by:  phk
Submitted by: Julian Assange proff@iq.org

26 years agoMinor typo in the wd driver. The manifestation of this bug
John Dyson [Sat, 11 Apr 1998 20:09:39 +0000 (20:09 +0000)]
Minor typo in the wd driver.  The manifestation of this bug
is a tremendous perf decrease due to the disabling of advanced
features such as DMA, Ultra DMA, and 32bit mode.  This patch
might have been reported by someone else (I seem to remember
it.)

26 years agoAdd noc-tun to list
Poul-Henning Kamp [Sat, 11 Apr 1998 20:03:05 +0000 (20:03 +0000)]
Add noc-tun to list

26 years agoProgram which implements "nos" alias "ka9q" alias "IP-IP" tunnels.
Poul-Henning Kamp [Sat, 11 Apr 1998 19:33:06 +0000 (19:33 +0000)]
Program which implements "nos" alias "ka9q" alias "IP-IP" tunnels.

PR: 1154
Reviewed by: phk
Manpage by: phk
Submitted by: Nickolay N. Dudorov nnd@itfs.nsk.su

26 years agoWhen pmap_pinit0() allocates a page for proc0's page directory,
Poul-Henning Kamp [Sat, 11 Apr 1998 17:24:06 +0000 (17:24 +0000)]
When pmap_pinit0() allocates a page for proc0's page directory,
kernal page table may need to be extended.  But while growing the
kernel page table (pmap_growkernel()), newly allocated kernel page
table pages are entered into every process' page directory. For
proc0, the page directory is not allocated yet, and results in a
page fault.  Eventually, the machine panics with "lockmgr: not
holding exclusive lock".

PR: 5458
Reviewed by: phk
Submitted by: Luoqi Chen <luoqi@luoqi.watermarkgroup.com>

26 years agoArgh!.. Spot the deliberate mistake that was probably causing the
Peter Wemm [Sat, 11 Apr 1998 16:12:07 +0000 (16:12 +0000)]
Argh!..  Spot the deliberate mistake that was probably causing the
leftover files in /tmp..  (this commit brought to you by 'cvs update -j')

26 years agoMF22: build lpd docs
Jordan K. Hubbard [Sat, 11 Apr 1998 15:10:35 +0000 (15:10 +0000)]
MF22: build lpd docs

26 years agoChange the FILE locking to be by FILE, not by the underlying fd as
John Birrell [Sat, 11 Apr 1998 07:47:22 +0000 (07:47 +0000)]
Change the FILE locking to be by FILE, not by the underlying fd as
it was. Add a FILE_WAIT state and queue threads waiting for a FILE
lock. Start using the sys/queue.h macros instead of the way that MIT
pthreads did it.

Add a thread name to the private thread structure and a non-POSIX
function to set this. This helps (me at least) when sending a SIGINFO
to a threaded process to get a /tmp/uthread.dump to see what the
<expletive deleted> threads are doing this time. It is nice to be
able to recognise (yes, I spell that with an 's' too) which threads
are which.

26 years agoAdd FILE locking stubs for libc.
John Birrell [Sat, 11 Apr 1998 07:40:47 +0000 (07:40 +0000)]
Add FILE locking stubs for libc.

Change the FILE locking to support kernel threads when linked with
libpthread (which you haven't see yet). This requires that libc become
thread-safe and thread-aware, testing __isthreaded before attempting
to do lock/unlock calls. The impact on non-threaded programs is minor.
This change works with libc_r, so it's the best compromise.

26 years agoRemove a nolonger implented "BUGS" description.
Poul-Henning Kamp [Sat, 11 Apr 1998 07:37:44 +0000 (07:37 +0000)]
Remove a nolonger implented "BUGS" description.
PR: 6240
Reviewed by: phk
Submitted by: Niall Smart rotel@indigo.ie

26 years agoAdd *_unlocked() macros according to POSIX. In the _THREAD_SAFE case,
John Birrell [Sat, 11 Apr 1998 07:33:46 +0000 (07:33 +0000)]
Add *_unlocked() macros according to POSIX. In the _THREAD_SAFE case,
test for __isthreaded before calling the lock/unlock functions to
try to save some performance. The _THREAD_SAFE case should become the
default, but since it tests for a global variable in libc, people won't
be able to build -current on pre-3.0 systems unless the default leaves
it out. Such is life.

26 years agoCorrectly figure out that the remove cannot do passive mode.
Poul-Henning Kamp [Sat, 11 Apr 1998 07:28:53 +0000 (07:28 +0000)]
Correctly figure out that the remove cannot do passive mode.

PR: 6259
Reviewed & slightly modified by: phk
Submitted by: Archie Cobbs <archie@whistle.com>

26 years agoAdd a global variable called __isthreaded that can be tested throughout
John Birrell [Sat, 11 Apr 1998 07:02:58 +0000 (07:02 +0000)]
Add a global variable called __isthreaded that can be tested throughout
libc to determine if locking is required. This is needed in libc
for use with kernel threads, but until a thread is created, we don't
really want to bother locking things. The variable was added here
because the crt code calls exit(main()) so all programs will get the
variable.

26 years agoAdd a private header file for libc/libc_r/libpthread to contain
John Birrell [Sat, 11 Apr 1998 06:57:33 +0000 (06:57 +0000)]
Add a private header file for libc/libc_r/libpthread to contain
definitions for things like locking etc.

26 years agoAdd a function prototype to set the name of a thread for debugging
John Birrell [Sat, 11 Apr 1998 02:50:59 +0000 (02:50 +0000)]
Add a function prototype to set the name of a thread for debugging
purposes.

26 years agoAdded comments about LC_ALL and setlocale(3).
Dag-Erling Smørgrav [Fri, 10 Apr 1998 16:13:29 +0000 (16:13 +0000)]
Added comments about LC_ALL and setlocale(3).
PR: conf/6145

26 years agoAdd a reference to the gcc linker option -pthread that is part of
John Birrell [Fri, 10 Apr 1998 09:33:09 +0000 (09:33 +0000)]
Add a reference to the gcc linker option -pthread that is part of
the FreeBSD configuration LIB_SPEC.

26 years agoDocument the FreeBSD specific options from LIB_SPEC.
John Birrell [Fri, 10 Apr 1998 09:19:11 +0000 (09:19 +0000)]
Document the FreeBSD specific options from LIB_SPEC.

26 years agoDo not clobber "heads" by &= in wdcommand use only &.
Søren Schmidt [Fri, 10 Apr 1998 08:00:24 +0000 (08:00 +0000)]
Do not clobber "heads" by &= in wdcommand use only &.

Spotted by: bde

26 years agoFix the assembler so that it produces object files with the correct
John Polstra [Fri, 10 Apr 1998 01:20:40 +0000 (01:20 +0000)]
Fix the assembler so that it produces object files with the correct
magic number byte ordering for FreeBSD.  This makes "file" describe
our object files as "FreeBSD/i386 object" instead of as NetBSD
object files.  In case this seems drastic and risky, Bruce points
out that the "ld -r -x" step that is done on every object file when
building libraries fixes the byte ordering in the same way.  I have
been running with this patch for over a month and have seen no
problems.

26 years agoGrammar police.
Alexander Langer [Fri, 10 Apr 1998 00:09:04 +0000 (00:09 +0000)]
Grammar police.

26 years agoAdd the LBA addressing flag.
Søren Schmidt [Thu, 9 Apr 1998 22:31:47 +0000 (22:31 +0000)]
Add the LBA addressing flag.

26 years agoAdd the new LBA mode support in the wd? config examples.
Søren Schmidt [Thu, 9 Apr 1998 22:28:57 +0000 (22:28 +0000)]
Add the new LBA mode support in the wd? config examples.

26 years agoTry to use the "right" CHS translations of a LBA device.
Søren Schmidt [Thu, 9 Apr 1998 17:46:45 +0000 (17:46 +0000)]
Try to use the "right" CHS translations of a LBA device.
Drives bigger than 8.4G is still in question until I get a drive
to test on...

26 years ago- Whoops, better have the structure created before trying to use it in
Nate Williams [Thu, 9 Apr 1998 14:01:13 +0000 (14:01 +0000)]
- Whoops, better have the structure created before trying to use it in
  the patch I just submitted.

Noticed by:     phk

26 years agoAdd LBA mode support for large drives.
Søren Schmidt [Wed, 8 Apr 1998 20:04:39 +0000 (20:04 +0000)]
Add LBA mode support for large drives.
Use config flags 0x1000 to enable LBA mode. It should be enabled in
the BIOS too to avoid geometry confusion.

One catch though, I'm not sure all BIOS's uses the 64head/63secs
translation, all mine does but....

26 years agoFix a minor bug (|| instead of |)
Søren Schmidt [Wed, 8 Apr 1998 20:00:03 +0000 (20:00 +0000)]
Fix a minor bug (|| instead of |)

26 years agoNew mount option nosymfollow. If enabled, the kernel lookup()
Wolfram Schneider [Wed, 8 Apr 1998 18:31:59 +0000 (18:31 +0000)]
New mount option nosymfollow. If enabled, the kernel lookup()
function will not follow symbolic links on the mounted
file system and return EACCES (Permission denied).

26 years ago- Fix bug I introduced a few months ago. If a driver fails the probe,
Nate Williams [Wed, 8 Apr 1998 15:00:02 +0000 (15:00 +0000)]
- Fix bug I introduced a few months ago.  If a driver fails the probe,
  correctly unregister the interrupt from the system.

[ My fix is much simpler than the one provided in the PR ]

PR: 6249

26 years ago(evil) hackers -> crackers
Martin Cracauer [Wed, 8 Apr 1998 12:00:48 +0000 (12:00 +0000)]
(evil) hackers -> crackers

26 years agoMinor adjustments to the timecounting and proc0.
Poul-Henning Kamp [Wed, 8 Apr 1998 09:01:53 +0000 (09:01 +0000)]
Minor adjustments to the timecounting and proc0.

Mostly Submitted by: bde

26 years agoFix a bug which clobbered linker set symbols that had forward references.
John Polstra [Tue, 7 Apr 1998 17:10:01 +0000 (17:10 +0000)]
Fix a bug which clobbered linker set symbols that had forward references.
With -O3, egcs generates such forward references.

PR: gnu/6055
Reviewed by: jdp
Submitted by: Dmitrij Tejblum <tejblum@arc.hq.cti.ru> in slightly different form

26 years agoSync with sys/i386/boot/biosboot/README.serial revision 1.9.
KATO Takenori [Tue, 7 Apr 1998 15:33:47 +0000 (15:33 +0000)]
Sync with sys/i386/boot/biosboot/README.serial revision 1.9.

26 years agoGuess the position of the drive number in the device name better so
Bruce Evans [Tue, 7 Apr 1998 12:43:57 +0000 (12:43 +0000)]
Guess the position of the drive number in the device name better so
that `fsck -p' doesn't check multiple slices on the same drive
concurrently.  Don't invoke undefined behaviour when searching for
the drive number in strange device names.

PR: 6129
Reviewed by: phk
Submitted by: Yuichi MATSUTAKA <matutaka@osa.att.ne.jp>, but rewritten
by me.

26 years agoSync with sys/i386/isa/npx.c revision 1.58.
KATO Takenori [Tue, 7 Apr 1998 09:09:41 +0000 (09:09 +0000)]
Sync with sys/i386/isa/npx.c revision 1.58.

26 years agoSync with sys/i386/conf/files.i386 revision 1.196.
KATO Takenori [Tue, 7 Apr 1998 09:08:27 +0000 (09:08 +0000)]
Sync with sys/i386/conf/files.i386 revision 1.196.

26 years agoSync with sys/i386/i386/machdep.c revision 1.294.
KATO Takenori [Tue, 7 Apr 1998 09:07:53 +0000 (09:07 +0000)]
Sync with sys/i386/i386/machdep.c revision 1.294.

26 years agoToday is not my lucky day. Fix missing brace and I got a request
Peter Wemm [Mon, 6 Apr 1998 19:32:37 +0000 (19:32 +0000)]
Today is not my lucky day.  Fix missing brace and I got a request
to use EMLINK instead.

26 years agoFix VM86 compiles. a #include "opt_vm86.h" was missing, and the my_tr
Peter Wemm [Mon, 6 Apr 1998 18:59:15 +0000 (18:59 +0000)]
Fix VM86 compiles.  a #include "opt_vm86.h" was missing, and the my_tr
variable was needed in the non-SMP case.

Submitted by:  Jonathan Lemon <jlemon@americantv.com>

26 years agoUse a different errno (ELOOP (as sef mentioned) since the text that goes
Peter Wemm [Mon, 6 Apr 1998 18:43:28 +0000 (18:43 +0000)]
Use a different errno (ELOOP (as sef mentioned) since the text that goes
with the error sounds ok for the condition) if O_NOFOLLOW gets a link.

26 years agoRather than let users get fd's to symlink files, make O_NOFOLLOW cause
Peter Wemm [Mon, 6 Apr 1998 18:25:21 +0000 (18:25 +0000)]
Rather than let users get fd's to symlink files, make O_NOFOLLOW cause
an error if it gets a link (like it does if it gets a socket).  The
implications of letting users try and do file operations on symlinks
themselves were too worrying.

26 years agoBack this out, allowing users to get a fd connected to a symlink is
Peter Wemm [Mon, 6 Apr 1998 18:18:50 +0000 (18:18 +0000)]
Back this out, allowing users to get a fd connected to a symlink is
just too dangerous.

26 years agoDon't panic if a VOP_READ() gets through on a short link, Just Do It
Peter Wemm [Mon, 6 Apr 1998 17:44:40 +0000 (17:44 +0000)]
Don't panic if a VOP_READ() gets through on a short link, Just Do It
(because we can :-).  This means you can open a link file (or pseudo-file
in the case of short links where the data is stored in the inode rather
than disk blocks) and read the contents.
However, trap any writes from the user as it's difficult to do the right
thing in all cases.  A link may be short and the user may be trying to
extend it beyond the limit and so on.  Although.. being able to re-target
a symlink without deleting it first might have been nice.
This stuff is a bit perverse since symlink() and readlink() calls can
end up actually being implemented as read/write vnode ops.

Reviewed by: phk

26 years agoImplement a new open(2) flag: O_NOFOLLOW. This will instruct open
Peter Wemm [Mon, 6 Apr 1998 17:38:43 +0000 (17:38 +0000)]
Implement a new open(2) flag: O_NOFOLLOW.  This will instruct open
to not follow symlinks, but to open a handle on the link itself(!).
As strange as this might sound, it has several useful applications
safe race-free ways of opening files in hostile areas (eg: /tmp, a mode
1777 /var/mail, etc).  It also would allow things like fchown() to work
on the link rather than having to implement a new syscall specifically for
that task.

Reviewed by: phk

26 years agoAdd a new -F flag which is a superset of -f. It will cause tail to
Peter Wemm [Mon, 6 Apr 1998 16:13:49 +0000 (16:13 +0000)]
Add a new -F flag which is a superset of -f.  It will cause tail to
stat() the file being followed and do a close/reopen if the file has been
renamed and/or rotated.  This is damn useful for leaving running on files
in /var/log when newsyslog(8) rotates them.

26 years agocurproc is initialized in locore at the same time for both SMP and UP now.
Peter Wemm [Mon, 6 Apr 1998 15:51:22 +0000 (15:51 +0000)]
curproc is initialized in locore at the same time for both SMP and UP now.

26 years agoremove #ifdef declaration of npxproc, use globals.s and the extern always.
Peter Wemm [Mon, 6 Apr 1998 15:50:29 +0000 (15:50 +0000)]
remove #ifdef declaration of npxproc, use globals.s and the extern always.

26 years agoadd globals.s for data that is treated differently on SMP.
Peter Wemm [Mon, 6 Apr 1998 15:49:35 +0000 (15:49 +0000)]
add globals.s for data that is treated differently on SMP.

26 years agoUse real types for the SMP pages being allocated rather than arrays of
Peter Wemm [Mon, 6 Apr 1998 15:48:30 +0000 (15:48 +0000)]
Use real types for the SMP pages being allocated rather than arrays of
ints.  Remove some no longer needed casts.  Initialize the per-cpu
global data area using the structs rather than knowing too much about
layout, alignment, etc.

26 years agoclean up #ifdefs, define the variables that have to be per-cpu on SMP
Peter Wemm [Mon, 6 Apr 1998 15:46:17 +0000 (15:46 +0000)]
clean up #ifdefs, define the variables that have to be per-cpu on SMP
in globals.s only and use externs always.

26 years ago_curpcb is always defined in globals.s instead of here in #ifdefs
Peter Wemm [Mon, 6 Apr 1998 15:44:31 +0000 (15:44 +0000)]
_curpcb is always defined in globals.s instead of here in #ifdefs

26 years agoBogus casts
Peter Wemm [Mon, 6 Apr 1998 15:43:29 +0000 (15:43 +0000)]
Bogus casts

26 years agoDefunct, now part of globals.s
Peter Wemm [Mon, 6 Apr 1998 15:42:54 +0000 (15:42 +0000)]
Defunct, now part of globals.s

26 years agoRather than filling this file up with SMP .sets, use those from
Peter Wemm [Mon, 6 Apr 1998 15:42:26 +0000 (15:42 +0000)]
Rather than filling this file up with SMP .sets, use those from
globals.s instead.
Initialize curproc in the same place for both UP and SMP.

26 years agoGenerate #defines that the asm code can access for the per-cpu data
Peter Wemm [Mon, 6 Apr 1998 15:40:10 +0000 (15:40 +0000)]
Generate #defines that the asm code can access for the per-cpu data
structures.

26 years agogenerate .sets for variables used in asm and C that are stored in per-cpu
Peter Wemm [Mon, 6 Apr 1998 15:39:09 +0000 (15:39 +0000)]
generate .sets for variables used in asm and C that are stored in per-cpu
space under SMP.

26 years agoA pair of C structures used for laying out the SMP per-cpu data space.
Peter Wemm [Mon, 6 Apr 1998 15:37:21 +0000 (15:37 +0000)]
A pair of C structures used for laying out the SMP per-cpu data space.

26 years agosome missing symbols
Peter Wemm [Mon, 6 Apr 1998 15:33:02 +0000 (15:33 +0000)]
some missing symbols

26 years agoUse getmicrotime() for if_lastchange, 10msec is plenty precision.
Poul-Henning Kamp [Mon, 6 Apr 1998 11:43:12 +0000 (11:43 +0000)]
Use getmicrotime() for if_lastchange, 10msec is plenty precision.

26 years agoUse random() to find our initial xid.
Poul-Henning Kamp [Mon, 6 Apr 1998 11:41:07 +0000 (11:41 +0000)]
Use random() to find our initial xid.

26 years agoUse read_random()
Poul-Henning Kamp [Mon, 6 Apr 1998 11:40:42 +0000 (11:40 +0000)]
Use read_random()

26 years agoUse random() for seq numbers and read_random for CHAP challenge.
Poul-Henning Kamp [Mon, 6 Apr 1998 11:40:17 +0000 (11:40 +0000)]
Use random() for seq numbers and read_random for CHAP challenge.

26 years agoUse random() rather then than homegrown stuff.
Poul-Henning Kamp [Mon, 6 Apr 1998 11:39:04 +0000 (11:39 +0000)]
Use random() rather then than homegrown stuff.

26 years agoRemove stuff related to microtime.s, which is gone.
Poul-Henning Kamp [Mon, 6 Apr 1998 11:38:18 +0000 (11:38 +0000)]
Remove stuff related to microtime.s, which is gone.

26 years agoUse getmicrotime insted of microtime, we only use the second part.
Poul-Henning Kamp [Mon, 6 Apr 1998 11:37:17 +0000 (11:37 +0000)]
Use getmicrotime insted of microtime, we only use the second part.

26 years agoMake read_random() take a (void *) argument instead of (char *)
Poul-Henning Kamp [Mon, 6 Apr 1998 09:30:42 +0000 (09:30 +0000)]
Make read_random() take a (void *) argument instead of (char *)

26 years agoMake a kernel version of the timer* functions called timerval* to be
Poul-Henning Kamp [Mon, 6 Apr 1998 08:26:08 +0000 (08:26 +0000)]
Make a kernel version of the timer* functions called timerval* to be
more consistent.

OK'ed by: bde

26 years agotake out opt_bktr.h from brooktree848.c
Amancio Hasty [Mon, 6 Apr 1998 07:57:55 +0000 (07:57 +0000)]
take out opt_bktr.h from brooktree848.c

26 years agoRemove the last traces of TUBA.
Poul-Henning Kamp [Mon, 6 Apr 1998 06:52:47 +0000 (06:52 +0000)]
Remove the last traces of TUBA.

Inspired by: PR kern/3317

26 years agoPrevent rdist from dumping core: do not free() pointer before using it.
Philippe Charnier [Mon, 6 Apr 1998 06:18:32 +0000 (06:18 +0000)]
Prevent rdist from dumping core: do not free() pointer before using it.
Add rcsid. Document -D flag (debug mode). Remove unused includes.
PR:bin/3158 (part 1)

26 years agoReally make this compile..
Peter Wemm [Mon, 6 Apr 1998 05:11:42 +0000 (05:11 +0000)]
Really make this compile..

Pointed out by: bde

26 years agoSync with sys/i386/isa/clock.c revision 1.119.
KATO Takenori [Mon, 6 Apr 1998 03:38:18 +0000 (03:38 +0000)]
Sync with sys/i386/isa/clock.c revision 1.119.