]> CyberLeo.Net >> Repos - FreeBSD/FreeBSD.git/log
FreeBSD/FreeBSD.git
23 years agoDuh, it is mtree(8) that creates boot/kernel for us as refered to in
David E. O'Brien [Sun, 8 Oct 2000 18:45:01 +0000 (18:45 +0000)]
Duh, it is mtree(8) that creates boot/kernel for us as refered to in
rev 1.572.  Of course if you are manually running the various Makefile
targets, it may or may not get run.

23 years agoDon't install manpages.
David E. O'Brien [Sun, 8 Oct 2000 16:56:04 +0000 (16:56 +0000)]
Don't install manpages.
They are being moved elsewhere, and they are causing problems being here.

23 years agoUnbreak detection of breakage in cy driver.
Bruce Evans [Sun, 8 Oct 2000 14:46:04 +0000 (14:46 +0000)]
Unbreak detection of breakage in cy driver.

23 years agoMade this file compile again after move of stuff from <machine.ipl.h>
Bruce Evans [Sun, 8 Oct 2000 14:41:13 +0000 (14:41 +0000)]
Made this file compile again after move of stuff from <machine.ipl.h>
to <sys/ipl.h>.

23 years agoUse schedsofttty() again (SMPng casualty with intentionally wrong fix
Bruce Evans [Sun, 8 Oct 2000 14:37:39 +0000 (14:37 +0000)]
Use schedsofttty() again (SMPng casualty with intentionally wrong fix
in rev.1.101).

Made this file compile again after move of stuff from <machine.ipl.h>
to <sys/ipl.h>.

23 years agoUse schedsofttty() again so that siopoll() gets scheduled as designed
Bruce Evans [Sun, 8 Oct 2000 14:13:31 +0000 (14:13 +0000)]
Use schedsofttty() again so that siopoll() gets scheduled as designed
(SMPng casualty in rev.1.308 with wrong fix in rev.1.310).

23 years agoKASSERT that mbuf external reference counts are > 0 before we remove
David Malone [Sun, 8 Oct 2000 09:52:50 +0000 (09:52 +0000)]
KASSERT that mbuf external reference counts are > 0 before we remove
a reference. (Agreed to by Bosko some time ago).

23 years agoFixed printf format errors which corrupted /var/db/mounttab on i386's with
Bruce Evans [Sun, 8 Oct 2000 09:24:45 +0000 (09:24 +0000)]
Fixed printf format errors which corrupted /var/db/mounttab on i386's with
64-bit longs (don't assume that time_t is long; assume that time_t's are
representable as longs).

23 years agoUse the correct timestamp for the -C -p case when the comparison fails.
Bruce Evans [Sun, 8 Oct 2000 09:17:56 +0000 (09:17 +0000)]
Use the correct timestamp for the -C -p case when the comparison fails.
A garbage timestamp was used for at least installing /dev/null with -C -p
when the target doesn't already exist.

23 years agoEnable use of poll()
Kris Kennaway [Sun, 8 Oct 2000 09:09:05 +0000 (09:09 +0000)]
Enable use of poll()

23 years agoDon't overflow our fd_set. This is not a full sync with KAME because there
Kris Kennaway [Sun, 8 Oct 2000 08:02:35 +0000 (08:02 +0000)]
Don't overflow our fd_set. This is not a full sync with KAME because there
are a whole lot of other changes which may not be suitable for us.

Obtained from: KAME

23 years agoDon't write past the end of our fd_set
Kris Kennaway [Sun, 8 Oct 2000 07:58:32 +0000 (07:58 +0000)]
Don't write past the end of our fd_set

Obtained from: OpenBSD, via KAME

23 years agoFormat string paranoia.
Kris Kennaway [Sun, 8 Oct 2000 07:54:43 +0000 (07:54 +0000)]
Format string paranoia.

23 years agoSync with KAME. Don't write past the end of the fd_set.
Kris Kennaway [Sun, 8 Oct 2000 07:52:08 +0000 (07:52 +0000)]
Sync with KAME. Don't write past the end of the fd_set.

Obtained from: KAME

23 years agoSync with KAME. Format string auditing and manpage cleanups.
Kris Kennaway [Sun, 8 Oct 2000 07:43:38 +0000 (07:43 +0000)]
Sync with KAME. Format string auditing and manpage cleanups.

Obtained from: KAME
Submitted by: Bill Sommerfeld <sommerfeld@netbsd.org>

23 years agoadd COREGA FEther PCC-TXF
MIHIRA Sanpei Yoshiro [Sun, 8 Oct 2000 07:42:52 +0000 (07:42 +0000)]
add COREGA FEther PCC-TXF
(MITA Yoshio <mita@ee.t.u-tokyo.ac.jp> in [bsd-nomads:14572])

23 years agoSync with KAME. Format string auditing and manpage cleanups.
Kris Kennaway [Sun, 8 Oct 2000 07:36:35 +0000 (07:36 +0000)]
Sync with KAME. Format string auditing and manpage cleanups.

Obtained from: KAME
Submitted by: Bill Sommerfeld <sommerfeld@netbsd.org>

23 years agoFormat string paranoia.
Kris Kennaway [Sun, 8 Oct 2000 04:33:21 +0000 (04:33 +0000)]
Format string paranoia.

23 years agoCleanup comment in machine/param.h regarding mbuf-related sizes, and get rid
Bosko Milekic [Sun, 8 Oct 2000 03:52:27 +0000 (03:52 +0000)]
Cleanup comment in machine/param.h regarding mbuf-related sizes, and get rid
of MCLOFSET, which does not appear to be used anywhere anymore, and if it is,
it probably shouldn't be.

23 years agoCorrect corruption of the qinfifo in ahc_search_qinififo() for all
Justin T. Gibbs [Sun, 8 Oct 2000 03:37:52 +0000 (03:37 +0000)]
Correct corruption of the qinfifo in ahc_search_qinififo() for all
non-LVD controllers.  We only need to take special action on the qinfifo
if we have dectected the case of an SCB that has been removed from the
qinfifo but has not been fully DMAed to the controller.  A missing
conditional caused this code to be executed every time an SCB was
aborted from the queue

Don't attempt to print the path of an SCB that has been freed.

Clean up the traversal of the pending scb list in
ahc_update_pending_syncrates().  This has no functional change.

Correct ahc_timeout()'s requeing of a timedout SCB to effect a
recovery action.  We now use ahc_qinfifo_requeue() and a
new function ahc_qinfifo_count() instead of performing the
requeue inline.  The old code did not conform to the new qinfifo
method.

Clear the timedout SCB from the disconnected list.  This ensures
that the SCB_NEXT field is free to be used for queuing us to
the qinfifo.

23 years agoConform to BDECFLAGS
Bill Fumerola [Sun, 8 Oct 2000 03:24:30 +0000 (03:24 +0000)]
Conform to BDECFLAGS

23 years agoRemove headers not needed.
Alfred Perlstein [Sat, 7 Oct 2000 23:15:17 +0000 (23:15 +0000)]
Remove headers not needed.

Pointed out by: phk

23 years agoForgot to remove the directory from mtree for the obsolete sendmail
Gregory Neil Shapiro [Sat, 7 Oct 2000 22:46:42 +0000 (22:46 +0000)]
Forgot to remove the directory from mtree for the obsolete sendmail
documentation.

23 years agoremove unused #include.
Poul-Henning Kamp [Sat, 7 Oct 2000 19:38:31 +0000 (19:38 +0000)]
remove unused #include.

23 years agoGive the post-fix example a better description.
Alexander Langer [Sat, 7 Oct 2000 14:34:26 +0000 (14:34 +0000)]
Give the post-fix example a better description.

PR: 21646
Submitted by: "Peter Pentchev" <roam@orbitel.bg>

23 years agoDocument the fact that sockstat now lists Unix domain sockets, and the new
Dag-Erling Smørgrav [Sat, 7 Oct 2000 14:11:43 +0000 (14:11 +0000)]
Document the fact that sockstat now lists Unix domain sockets, and the new
command-line options for selecting what to display.

23 years agoTotal rewrite. This was actually the first non-trivial Perl script I ever
Dag-Erling Smørgrav [Sat, 7 Oct 2000 14:10:37 +0000 (14:10 +0000)]
Total rewrite. This was actually the first non-trivial Perl script I ever
wrote, and as such was not very pretty.

Changes that may cause problems for people who use sockstat in scripts:

 - sockstat(8) now displays Unix domain sockets in addition to IPv4 and IPv6
 - the last period in local and foreign addresses is changed to a colon to
   make the port number easier to spot
 - IPv4 and IPv6 sockets are listed separately (IPv4 first, then IPv6, then
   Unix)

23 years agoRegen.
Nick Hibma [Sat, 7 Oct 2000 11:11:12 +0000 (11:11 +0000)]
Regen.

23 years agoSync with NetBSD. More ids, more!
Nick Hibma [Sat, 7 Oct 2000 11:10:53 +0000 (11:10 +0000)]
Sync with NetBSD. More ids, more!

23 years agoRegen.
Nick Hibma [Sat, 7 Oct 2000 11:02:25 +0000 (11:02 +0000)]
Regen.

23 years agoZyxel keyboard Id
Nick Hibma [Sat, 7 Oct 2000 11:01:54 +0000 (11:01 +0000)]
Zyxel keyboard Id

23 years agoCy driver doesn't compile and nobody seems to care.
Poul-Henning Kamp [Sat, 7 Oct 2000 10:07:36 +0000 (10:07 +0000)]
Cy driver doesn't compile and nobody seems to care.

23 years agoCaught up with the changes in interruption stuff.
KATO Takenori [Sat, 7 Oct 2000 06:01:38 +0000 (06:01 +0000)]
Caught up with the changes in interruption stuff.

23 years agoSeems ${RD}/trees/bin/boot/kernel can exist before the `release.6:' target
David E. O'Brien [Sat, 7 Oct 2000 05:52:41 +0000 (05:52 +0000)]
Seems ${RD}/trees/bin/boot/kernel can exist before the `release.6:' target
runs.  So perform a copy that take that into account.

23 years agoo Implement get_ivars so matching routines will work (this causes
Warner Losh [Sat, 7 Oct 2000 05:48:51 +0000 (05:48 +0000)]
o Implement get_ivars so matching routines will work (this causes
  compat probe routines to work).
o Have a null driver_added routines.  We need to be careful about probing
  until after we know we have a card.  For the moment, we do nothing
  (which is safe).  This fixes a panic when a driver is loaded w/o a
  card in the slot.

XXX still need to fix the resource list code.  It is totally busted and
XXX causes a panic in the child printing routine after the probe has
XXX succeeded.

23 years agoFix arguments to kthread_create (submitted by takawata@freebsd.org).
Warner Losh [Sat, 7 Oct 2000 05:42:43 +0000 (05:42 +0000)]
Fix arguments to kthread_create (submitted by takawata@freebsd.org).
Remove unnecessary includes

23 years agoSync with KAME. Format string auditing, and add -a flag to autoprobe
Kris Kennaway [Fri, 6 Oct 2000 23:46:52 +0000 (23:46 +0000)]
Sync with KAME. Format string auditing, and add -a flag to autoprobe
interfaces.

Obtained from: KAME

23 years agoMore complete fix for multi-link sequence number handling bugs.
Archie Cobbs [Fri, 6 Oct 2000 23:42:02 +0000 (23:42 +0000)]
More complete fix for multi-link sequence number handling bugs.
Add a new control message for querying the sequence number state.

23 years agoIndex: Makefile
Brian Somers [Fri, 6 Oct 2000 23:29:11 +0000 (23:29 +0000)]
Index: Makefile
===================================================================
RCS file: /home/ncvs/src/share/man/man9/Makefile,v
retrieving revision 1.75
diff -u -r1.75 Makefile
--- Makefile 2000/09/17 12:15:12 1.75
+++ Makefile 2000/10/06 23:24:26
@@ -69,6 +69,8 @@
 MLINKS+=make_dev.9 destroy_dev.9
 MLINKS+=malloc.9 FREE.9 malloc.9 MALLOC.9 malloc.9 free.9
 MLINKS+=mi_switch.9 cpu_switch.9
+MLINKS+=namei.9 NDINIT.9
+MLINKS+=namei.9 NDFREE.9
 MLINKS+=posix4.9 p1003_1b.9
 MLINKS+=psignal.9 gsignal.9 psignal.9 pgsignal.9
 MLINKS+=rtalloc.9 rtalloc1.9 rtalloc.9 rtalloc_ign.9

23 years agoFormat string fix.
Kris Kennaway [Fri, 6 Oct 2000 23:24:45 +0000 (23:24 +0000)]
Format string fix.

23 years agoWhen leaving suspend mode after enabling/disabling the promisc mode bit,
Bill Paul [Fri, 6 Oct 2000 22:54:41 +0000 (22:54 +0000)]
When leaving suspend mode after enabling/disabling the promisc mode bit,
make sure the chip is restarted by issuing a start command to the command
register. Sometimes the receiver doesn't restart after leaving suspend
mode.

23 years agoArgh. Fix typo and sync to alpha/RELNOTES.TXT.
Bruce A. Mah [Fri, 6 Oct 2000 22:15:06 +0000 (22:15 +0000)]
Argh.  Fix typo and sync to alpha/RELNOTES.TXT.

Must remember to save changes to a file *before* a cvs commit,
not after.

23 years agoNote string-handling library calls fixed, stronger randomness in TCP
Bruce A. Mah [Fri, 6 Oct 2000 22:07:09 +0000 (22:07 +0000)]
Note string-handling library calls fixed, stronger randomness in TCP
initial sequence numbers.

23 years agoFix bug in handling of multi-link sequence numbers.
Archie Cobbs [Fri, 6 Oct 2000 20:36:17 +0000 (20:36 +0000)]
Fix bug in handling of multi-link sequence numbers.

Reported by: Becca Anderson <becca@worldint.com>

23 years ago.Os FreeBSD instead of just plain .Os
Archie Cobbs [Fri, 6 Oct 2000 19:47:00 +0000 (19:47 +0000)]
.Os FreeBSD instead of just plain .Os

Submitted by: ru

23 years ago* I don't think that ed-devices are still "high performance ethernet ...
Alexander Langer [Fri, 6 Oct 2000 19:44:09 +0000 (19:44 +0000)]
* I don't think that ed-devices are still "high performance ethernet              device"s.  Change that to "ethernet device".

* Update synopsis for the new config-style: just "device ed".

* PCI and PC Card devices are also supported.

23 years agoFix single character typo in legacy transfer settings function
Justin T. Gibbs [Fri, 6 Oct 2000 19:34:40 +0000 (19:34 +0000)]
Fix single character typo in legacy transfer settings function
thereby re-enable tagged queuing.

23 years agoNote security profiles added to sysinstall(8).
Bruce A. Mah [Fri, 6 Oct 2000 18:17:03 +0000 (18:17 +0000)]
Note security profiles added to sysinstall(8).
Add $FreeBSD$.

23 years agoFix typo: kereros -> kerberos
Alexander Langer [Fri, 6 Oct 2000 17:36:05 +0000 (17:36 +0000)]
Fix typo:  kereros -> kerberos

PR: 21468
Submitted by: "Giorgos Keramidas" <keramida@westgate.gr>

23 years agoMake it compile without -DNO_PAM again.
Ruslan Ermilov [Fri, 6 Oct 2000 17:08:31 +0000 (17:08 +0000)]
Make it compile without -DNO_PAM again.

23 years agoArgh, make P_ALRMPEND and P_PROFPEND be different flags.
John Baldwin [Fri, 6 Oct 2000 16:17:12 +0000 (16:17 +0000)]
Argh, make P_ALRMPEND and P_PROFPEND be different flags.

Submitted by: bde
Pointy hat to: jhb

23 years agoo Move initialization of ump from mp to the top of the function so that
Robert Watson [Fri, 6 Oct 2000 15:31:28 +0000 (15:31 +0000)]
o Move initialization of ump from mp to the top of the function so that
  it is defined whenm used in ufs_extattr_uepm_destroy(), fixing a panic
  due to a NULL pointer dereference.

Submitted by: Wesley Morgan <morganw@chemicals.tacorp.com>

23 years agoEnable the if_ed module.
Alexander Langer [Fri, 6 Oct 2000 15:23:32 +0000 (15:23 +0000)]
Enable the if_ed module.

23 years agoAdd card_if.h and sort the header files, while I'm here.
Alexander Langer [Fri, 6 Oct 2000 15:21:10 +0000 (15:21 +0000)]
Add card_if.h and sort the header files, while I'm here.

23 years agoFix the `missing footer' problem. See the comment for details.
Ruslan Ermilov [Fri, 6 Oct 2000 13:22:24 +0000 (13:22 +0000)]
Fix the `missing footer' problem.  See the comment for details.

Obtained from: NetBSD

23 years agoIf a process is over its resource limit for datasize, still allow
David Malone [Fri, 6 Oct 2000 13:03:50 +0000 (13:03 +0000)]
If a process is over its resource limit for datasize, still allow
it to lower its memory usage. This was mentioned on the mailing
lists ages ago, and I've lost the name of the person who brought
it up.

Reviewed by: alc

23 years agoDon't try to compare other attributes if the type doesn't match, it's
Poul-Henning Kamp [Fri, 6 Oct 2000 12:48:55 +0000 (12:48 +0000)]
Don't try to compare other attributes if the type doesn't match, it's
nothing but trouble, and no use anyway.

23 years agoDon't whine about modes on symlinks.
Poul-Henning Kamp [Fri, 6 Oct 2000 12:27:36 +0000 (12:27 +0000)]
Don't whine about modes on symlinks.

23 years agoThis brings support for IP Filter into rc.network and rc.conf with
Darren Reed [Fri, 6 Oct 2000 12:24:45 +0000 (12:24 +0000)]
This brings support for IP Filter into rc.network and rc.conf with
the appropriate documentation added to rc.conf(5).  If all goes well
with this over the next few weeks, the PR will be closed with the
pullup of patches back to 4-STABLE.

PR: 20202
Submitted by: Gerhard Sittig <Gerhard.Sittig@gmx.net>
Reviewed by: Darren Reed <darrenr@freebsd.org>
Approved by: Darren Reed <darrenr@freebsd.org>
Obtained from: Gerhard Sittig <Gerhard.Sittig@gmx.net>

23 years agoAs we now may check the TCP header window field, make sure we pullup
Ruslan Ermilov [Fri, 6 Oct 2000 12:12:09 +0000 (12:12 +0000)]
As we now may check the TCP header window field, make sure we pullup
enough into the mbuf data area.  Solve this problem once and for all
by pulling up the entire (standard) header for TCP and UDP, and four
bytes of header for ICMP (enough for type, code and cksum fields).

23 years agoList the -t option in usage().
Alexander Langer [Fri, 6 Oct 2000 12:10:51 +0000 (12:10 +0000)]
List the -t option in usage().

PR: 21734
Submitted by: Giorgos Keramidas <keramida@ceid.upatras.gr>

23 years agoList the options in alphabetic order.
Alexander Langer [Fri, 6 Oct 2000 12:07:46 +0000 (12:07 +0000)]
List the options in alphabetic order.

PR: 21734
Submitted by: Johan Karlsson <k@numeri.campus.luth.se>

23 years agoWork around a bug by adding struct tags. gcc-2.95 apparently gets the
Bruce Evans [Fri, 6 Oct 2000 11:53:32 +0000 (11:53 +0000)]
Work around a bug by adding struct tags.  gcc-2.95 apparently gets the
check in the [basic.link] section of the C++ standard wrong.  gcc-2.7.2.3
apparently doesn't do the check, so the bug doesn't affect RELENG_3.

PR: 16170, 21427
Submitted by: Max Khon <fjoe@lark.websci.ru> (i386 version)
Discussed with: jdp

23 years agoMerged from sys/isa/sio.c revision 1.316.
KATO Takenori [Fri, 6 Oct 2000 11:52:32 +0000 (11:52 +0000)]
Merged from sys/isa/sio.c revision 1.316.

23 years agoMerged from sys/i386/isa/npx.c revision 1.85.
KATO Takenori [Fri, 6 Oct 2000 11:51:20 +0000 (11:51 +0000)]
Merged from sys/i386/isa/npx.c revision 1.85.

23 years agoMerged from sys/i386/isa/clock.c revisions 1.158 and 1.159.
KATO Takenori [Fri, 6 Oct 2000 11:50:19 +0000 (11:50 +0000)]
Merged from sys/i386/isa/clock.c revisions 1.158 and 1.159.

23 years agoMerged from sys/i386/i386/machdep.c revision 1.415.
KATO Takenori [Fri, 6 Oct 2000 11:43:51 +0000 (11:43 +0000)]
Merged from sys/i386/i386/machdep.c revision 1.415.

23 years agoRemove references to recently removed voxware snd driver.
Maxim Sobolev [Fri, 6 Oct 2000 11:40:44 +0000 (11:40 +0000)]
Remove references to recently removed voxware snd driver.

23 years agoConvert this Makefile to the usual style.
Ruslan Ermilov [Fri, 6 Oct 2000 11:18:11 +0000 (11:18 +0000)]
Convert this Makefile to the usual style.

23 years agoDocument the latest firewall knobs.
Ruslan Ermilov [Fri, 6 Oct 2000 11:17:06 +0000 (11:17 +0000)]
Document the latest firewall knobs.

23 years agoCorrect who pioneered the accept filters and Yahoo!'s company name.
Paul Saab [Fri, 6 Oct 2000 10:51:24 +0000 (10:51 +0000)]
Correct who pioneered the accept filters and Yahoo!'s company name.

23 years agoAdd missing escape sequences, -mdoc'ify.
Ruslan Ermilov [Fri, 6 Oct 2000 08:40:52 +0000 (08:40 +0000)]
Add missing escape sequences, -mdoc'ify.

23 years agoSimple -mdoc fixes.
Ruslan Ermilov [Fri, 6 Oct 2000 08:33:43 +0000 (08:33 +0000)]
Simple -mdoc fixes.

23 years agoReduce userland namespace polution.
Jason Evans [Fri, 6 Oct 2000 08:11:11 +0000 (08:11 +0000)]
Reduce userland namespace polution.

#include <proc.h>, since curproc is needed.

23 years agoDo not call lockdestroy() for v_vnlock, which may point to a lock in a
Jason Evans [Fri, 6 Oct 2000 08:04:48 +0000 (08:04 +0000)]
Do not call lockdestroy() for v_vnlock, which may point to a lock in a
deeper vfs stacking layer.

Submitted by: bp

23 years agoCorrect a warning where the r_debug_state() dummy function used to trigger
John Baldwin [Fri, 6 Oct 2000 05:20:02 +0000 (05:20 +0000)]
Correct a warning where the r_debug_state() dummy function used to trigger
a breakpoint in the kernel didn't use the proper argument list.  To avoid
having to include the userland link.h header everyhwere that sys/linker.h
is used, make r_debug_state() a static function in link_elf.c as well.

23 years agoBring in a slew of fixes that were supposed to be in the last commit.
Justin T. Gibbs [Fri, 6 Oct 2000 04:01:06 +0000 (04:01 +0000)]
Bring in a slew of fixes that were supposed to be in the last commit.

In ahc_search_qinfifo, the SEARCH_REMOVE case must also handle
an SCB that has been removed from the QINFIFO but not yet been
fully dmaed to the card.

Correct locking for ahc_get_scb() calls.

Set SCB syncrate settings in ahc_execute_scb() to avoid a race
condition that could allow a newly queued SCB to be missed
by ahc_update_pending_syncrates().

When notifying the system of transfer negotiation updates, only
set the valid bits for tagged queuing and disconnection if the
path is fully qualified.  Sync/Wide settins apply to all luns
of a target, but tagged queuing and disconnection may change
on a per-lun basis.

Add missing ahc_unlock() calls in ahc_timeout() for the target
mode case.

23 years ago- Change fast interrupts on x86 to push a full interrupt frame and to
John Baldwin [Fri, 6 Oct 2000 02:20:21 +0000 (02:20 +0000)]
- Change fast interrupts on x86 to push a full interrupt frame and to
  return through doreti to handle ast's.  This is necessary for the
  clock interrupts to work properly.
- Change the clock interrupts on the x86 to be fast instead of threaded.
  This is needed because both hardclock() and statclock() need to run in
  the context of the current process, not in a separate thread context.
- Kill the prevproc hack as it is no longer needed.
- We really need Giant when we call psignal(), but we don't want to block
  during the clock interrupt.  Instead, use two p_flag's in the proc struct
  to mark the current process as having a pending SIGVTALRM or a SIGPROF
  and let them be delivered during ast() when hardclock() has finished
  running.
- Remove CLKF_BASEPRI, which was #ifdef'd out on the x86 anyways.  It was
  broken on the x86 if it was turned on since cpl is gone.  It's only use
  was to bogusly run softclock() directly during hardclock() rather than
  scheduling an SWI.
- Remove the COM_LOCK simplelock and replace it with a clock_lock spin
  mutex.  Since the spin mutex already handles disabling/restoring
  interrupts appropriately, this also lets us axe all the *_intr() fu.
- Back out the hacks in the APIC_IO x86 cpu_initclocks() code to use
  temporary fast interrupts for the APIC trial.
- Add two new process flags P_ALRMPEND and P_PROFPEND to mark the pending
  signals in hardclock() that are to be delivered in ast().

Submitted by: jakeb (making statclock safe in a fast interrupt)
Submitted by: cp (concept of delaying signals until ast())

23 years agocurrentldt is now a "special" global-data variable, and as such, there
John Baldwin [Fri, 6 Oct 2000 01:59:07 +0000 (01:59 +0000)]
currentldt is now a "special" global-data variable, and as such, there
is no actual currentldt integer variable directly.  Thus, don't claim that
there is.

23 years agoInterrupt frames don't include the saved cpl anymore since cpl is dead.
John Baldwin [Fri, 6 Oct 2000 01:57:28 +0000 (01:57 +0000)]
Interrupt frames don't include the saved cpl anymore since cpl is dead.

23 years agoVarious whitespace cleanups after the SMPng commit, which jumbled things
John Baldwin [Fri, 6 Oct 2000 01:55:07 +0000 (01:55 +0000)]
Various whitespace cleanups after the SMPng commit, which jumbled things
around a bit in the trap handling code.

23 years agoDon't treat a kernel stack fault the same as a general protect fault or
John Baldwin [Fri, 6 Oct 2000 01:50:43 +0000 (01:50 +0000)]
Don't treat a kernel stack fault the same as a general protect fault or
a segment not present fault in the non-vm86 case.

23 years agoRemove an unnecessary sti and spl0() in fork_trampoline. Interrupts
John Baldwin [Fri, 6 Oct 2000 01:47:24 +0000 (01:47 +0000)]
Remove an unnecessary sti and spl0() in fork_trampoline.  Interrupts
should be enabled by MTX_EXIT() now when it releases the sched_lock.

23 years ago- Include opt_vesa.h in vesa.c so that the VESA_DEBUG option is actually
John Baldwin [Fri, 6 Oct 2000 01:43:59 +0000 (01:43 +0000)]
- Include opt_vesa.h in vesa.c so that the VESA_DEBUG option is actually
  propagated from the kernel config file to the source.
- Add some more debug messages to list each mode that is rejected or found.

23 years agoDuring a verbose boot, call the null device 'null' rather than 'null0' to
John Baldwin [Fri, 6 Oct 2000 00:46:29 +0000 (00:46 +0000)]
During a verbose boot, call the null device 'null' rather than 'null0' to
be more consistent with the rest of the kernel.

23 years agoWorking down my list of release note items: SMP configuration
Bruce A. Mah [Fri, 6 Oct 2000 00:41:15 +0000 (00:41 +0000)]
Working down my list of release note items:  SMP configuration
option changes (+MFC), ahc(4) updates (+MFC), ACPI support, ,
tcpdump(1) updates (+MFC), ppp(8) updates (+MFC).

23 years agoNote new pcm(4) hardware support. Synchronize the lists of supported
Bruce A. Mah [Fri, 6 Oct 2000 00:32:13 +0000 (00:32 +0000)]
Note new pcm(4) hardware support.  Synchronize the lists of supported
sound hardware among the various .TXT files.

23 years agoMan page for the ichsmb SMBus controller driver.
Archie Cobbs [Fri, 6 Oct 2000 00:10:31 +0000 (00:10 +0000)]
Man page for the ichsmb SMBus controller driver.

23 years agoDriver for the Intel 82801AA (ICH) SMBus controller and compatibles.
Archie Cobbs [Fri, 6 Oct 2000 00:09:46 +0000 (00:09 +0000)]
Driver for the Intel 82801AA (ICH) SMBus controller and compatibles.

Obtained from: Whistle source tree

23 years agoPR: 21729
Murray Stokely [Thu, 5 Oct 2000 23:27:32 +0000 (23:27 +0000)]
PR: 21729
Approved by: jkh

Write kern_securelevel_enable variable to rc.conf if user selects
medium or low security in sysinstall.  This overrides the case where a
user selects fascist security and then tries to go back to a lower
setting.

23 years agoreturn correct type for process directory entries, DT_DIR not DT_REG
Alfred Perlstein [Thu, 5 Oct 2000 23:19:51 +0000 (23:19 +0000)]
return correct type for process directory entries, DT_DIR not DT_REG

23 years agoAdd upcoming ports layout changes
Warner Losh [Thu, 5 Oct 2000 23:13:47 +0000 (23:13 +0000)]
Add upcoming ports layout changes

23 years agoRegen.
Nick Hibma [Thu, 5 Oct 2000 23:11:13 +0000 (23:11 +0000)]
Regen.

23 years ago- Heavyweight interrupt threads on the alpha for device I/O interrupts.
John Baldwin [Thu, 5 Oct 2000 23:09:57 +0000 (23:09 +0000)]
- Heavyweight interrupt threads on the alpha for device I/O interrupts.
- Make softinterrupts (SWI's) almost completely MI, and divorce them
  completely from the x86 hardware interrupt code.
  - The ihandlers array is now gone.  Instead, there is a MI shandlers array
    that just contains SWI handlers.
  - Most of the former machine/ipl.h files have moved to a new sys/ipl.h.
- Stub out all the spl*() functions on all architectures.

Submitted by: dfr

23 years agoAdditional Ids
Nick Hibma [Thu, 5 Oct 2000 23:09:48 +0000 (23:09 +0000)]
Additional Ids

23 years agoReplace loadandclear() with atomic_readandclear_int().
John Baldwin [Thu, 5 Oct 2000 22:22:31 +0000 (22:22 +0000)]
Replace loadandclear() with atomic_readandclear_int().

23 years agoAdd atomic_readandclear_int and atomic_readandclear_long.
John Baldwin [Thu, 5 Oct 2000 22:19:50 +0000 (22:19 +0000)]
Add atomic_readandclear_int and atomic_readandclear_long.

23 years agoThe aac driver manual page
Scott Long [Thu, 5 Oct 2000 20:44:17 +0000 (20:44 +0000)]
The aac driver manual page

Reviewed by: sheldonh@freebsd.org

23 years agoAdd the manual page for the aac device.
Scott Long [Thu, 5 Oct 2000 20:42:42 +0000 (20:42 +0000)]
Add the manual page for the aac device.

Reviewed by: sheldonh@freebsd.org

23 years agoMake the gd_currentldt member in struct globaldata unconditional so
John Baldwin [Thu, 5 Oct 2000 20:30:36 +0000 (20:30 +0000)]
Make the gd_currentldt member in struct globaldata unconditional so
that this header doesn't depend on USER_LDT.  This fixes the USER_LDT
breakage with SMP kernels.