]> CyberLeo.Net >> Repos - FreeBSD/FreeBSD.git/log
FreeBSD/FreeBSD.git
27 years agoMerge from 2.2: auto-gen init.c and add ser-tcp.c
Paul Traina [Thu, 1 May 1997 00:18:51 +0000 (00:18 +0000)]
Merge from 2.2: auto-gen init.c and add ser-tcp.c

27 years ago 1.15 4/18/97 John-Mark Gurney <gurney_j@resnet.uoregon.edu>
Steve Passe [Thu, 1 May 1997 00:16:31 +0000 (00:16 +0000)]
 1.15          4/18/97    John-Mark Gurney <gurney_j@resnet.uoregon.edu>
                           Added [SR]RGBMASKs ioctl for byte swapping.

 1.16          4/20/97    Randall Hopper <rhh@ct.picker.com>
                          Generalized RGBMASK ioctls for general pixel
                          format setting [SG]ACTPIXFMT, and added query API
                          to return driver-supported pix fmts GSUPPIXFMT.

 1.17          4/21/97    hasty@rah.star-gate.com
                          Clipping support added.

 1.18          4/23/97    Clean up after failed CAP_SINGLEs where bt
                          interrupt isn't delivered, and fixed fixing
                          CAP_SINGLEs that for ODD_ONLY fields.

Submitted by:   individuals in above log messages.

27 years ago 1.15 4/18/97 John-Mark Gurney <gurney_j@resnet.uoregon.edu>
Steve Passe [Thu, 1 May 1997 00:15:30 +0000 (00:15 +0000)]
 1.15          4/18/97    John-Mark Gurney <gurney_j@resnet.uoregon.edu>
                           Added [SR]RGBMASKs ioctl for byte swapping.

 1.16          4/20/97    Randall Hopper <rhh@ct.picker.com>
                          Generalized RGBMASK ioctls for general pixel
                          format setting [SG]ACTPIXFMT, and added query API
                          to return driver-supported pix fmts GSUPPIXFMT.

 1.17          4/21/97    hasty@rah.star-gate.com
                          Clipping support added.

 1.18          4/23/97    Clean up after failed CAP_SINGLEs where bt
                          interrupt isn't delivered, and fixed fixing
                          CAP_SINGLEs that for ODD_ONLY fields.

Submitted by: individuals in above log messages.

27 years agofixed minor typo, bumped to version 2.0.9
Steve Passe [Wed, 30 Apr 1997 23:54:55 +0000 (23:54 +0000)]
fixed minor typo, bumped to version 2.0.9

27 years agoremoved:
Steve Passe [Wed, 30 Apr 1997 21:54:01 +0000 (21:54 +0000)]
removed:
   opt_smp_invltlb.h
from:
SMPHDRS=        opt_smp.h opt_smp_invltlb.h

SMP_INVLTLB is no longer a valid config option, the invalidation of the TLB
via inter-CPU IPIs is now standard when APIC_IO is used.

27 years agoFixed the SMP fixes:
Bruce Evans [Wed, 30 Apr 1997 20:56:37 +0000 (20:56 +0000)]
Fixed the SMP fixes:
- fixed dependencies so that `make depend' isn't necessary.
- added new files to CLEANFILES.
- fixed style.

27 years agoFixed wrong mnemonic `setnel' that gas happened to generate correct object
Bruce Evans [Wed, 30 Apr 1997 20:37:52 +0000 (20:37 +0000)]
Fixed wrong mnemonic `setnel' that gas happened to generate correct object
code for.

Obtained from: a slightly different fix in NetBSD

27 years agoRemoved unused groff `generate' directories.
Bruce Evans [Wed, 30 Apr 1997 20:27:34 +0000 (20:27 +0000)]
Removed unused groff `generate' directories.

27 years agochanged expect_lock() to try_lock(), the real name used in mplock.s
Steve Passe [Wed, 30 Apr 1997 19:04:26 +0000 (19:04 +0000)]
changed expect_lock() to try_lock(), the real name used in mplock.s

27 years agoBack out msmith's recent commit which breaks using symbolic names for netmasks.
Paul Traina [Wed, 30 Apr 1997 18:40:12 +0000 (18:40 +0000)]
Back out msmith's recent commit which breaks using symbolic names for netmasks.
Instead, reverse the order of the testing, so if a symbolic name starts with
a digit, we'll see if we can make a network address out of it first.  If
that fails, then we'll call getnet...

27 years agoFixed overflow for attempts to seek to offsets > ULONG_MAX. See the
Bruce Evans [Wed, 30 Apr 1997 17:51:29 +0000 (17:51 +0000)]
Fixed overflow for attempts to seek to offsets > ULONG_MAX.  See the
OpenBSD version for more complete fixes.  E.g., to seek to offsets >
INT_MAX using a block size of 1, block numbers need to be off_t's
instead of u_long's.

27 years agoAdded trailing slash to ${CANONICALOBJDIR} in tests for directories,
Bruce Evans [Wed, 30 Apr 1997 17:04:11 +0000 (17:04 +0000)]
Added trailing slash to ${CANONICALOBJDIR} in tests for directories,
so that /usr/obj can be a symlink to an existing directory.

27 years agoOops. The previous change duplicated the grossness instead of removing it.
Bruce Evans [Wed, 30 Apr 1997 17:03:09 +0000 (17:03 +0000)]
Oops.  The previous change duplicated the grossness instead of removing it.

27 years agoRemoved gross dir file bootstrap rule. Keep depending on the dir
Bruce Evans [Wed, 30 Apr 1997 16:33:43 +0000 (16:33 +0000)]
Removed gross dir file bootstrap rule.  Keep depending on the dir
file so that we fail if it isn't already installed.  The bootstrap
is now done by `make install' in the top-level src directory.  As
well as being gross, the rule didn't actually work for src trees
other than /usr/src, and this is difficult to fix since the whole
src tree may not exist.

27 years agoBootstrap the info `dir' file from here.
Bruce Evans [Wed, 30 Apr 1997 16:02:03 +0000 (16:02 +0000)]
Bootstrap the info `dir' file from here.

27 years agoFetch the registers from struct members in the pcb instead of
Bruce Evans [Wed, 30 Apr 1997 15:33:56 +0000 (15:33 +0000)]
Fetch the registers from struct members in the pcb instead of
punning the pcb to an array of ints and using magic indices to
access values in it.  This should prevent silent breakage from
changes in the pcb.

Supply 0 for unavailable registers instead of punning the tss to
an array of ints and using magic indices to access garbage values
in it.  (The registers are in the pcb; there is nothing interesting
in the tss.  This should change someday.  At least for dumps, all
the registers should be saved, and common_tss is a good place to
put them.)

Removed ancient wrong (disabled) method for reading eip.

27 years agoRemove SO_PRIVSTATE socket option; it is no longer necessary, nor implemented
Garrett Wollman [Wed, 30 Apr 1997 15:32:52 +0000 (15:32 +0000)]
Remove SO_PRIVSTATE socket option; it is no longer necessary, nor implemented
in the kernel.  inetd should automatically notic that it has gone away
once it is recompiled.

27 years agoRestore CRTC start address check back after looking into more sources
Andrey A. Chernov [Wed, 30 Apr 1997 15:31:22 +0000 (15:31 +0000)]
Restore CRTC start address check back after looking into more sources

27 years agoBacked out previous change. It just gave a more verbose Makefile by
Bruce Evans [Wed, 30 Apr 1997 15:23:02 +0000 (15:23 +0000)]
Backed out previous change.  It just gave a more verbose Makefile by
repeating the default for MAN1.

27 years agoAdd D_DISK flag to vn_bdevsw.
Doug Rabson [Wed, 30 Apr 1997 11:16:25 +0000 (11:16 +0000)]
Add D_DISK flag to vn_bdevsw.

27 years agoAllow NULL rpcs on non-privileged ports at all times to work around broken
Doug Rabson [Wed, 30 Apr 1997 09:51:37 +0000 (09:51 +0000)]
Allow NULL rpcs on non-privileged ports at all times to work around broken
clients.

PR: kern/3298
Submitted by: Tor Egge <Tor.Egge@idi.ntnu.no>

27 years agoAdded two missing committers to the list of developer and
Masafumi Max NAKANE [Wed, 30 Apr 1997 09:08:49 +0000 (09:08 +0000)]
Added two missing committers to the list of developer and
several missing port maintainers to the list of additional contributor.

27 years agoAdded new entities, a.jgreco and a.tedm.
Masafumi Max NAKANE [Wed, 30 Apr 1997 09:05:32 +0000 (09:05 +0000)]
Added new entities, a.jgreco and a.tedm.
(They have been mssing from the committers list although they are in
the group ncvs on freefall.)

27 years agoCatch up with the original.
Hiroyuki Hanai [Wed, 30 Apr 1997 06:43:05 +0000 (06:43 +0000)]
Catch up with the original.
Submitted by: iwasaki@jp.FreeBSD.org

27 years agoCatch up with the original.
Hiroyuki Hanai [Wed, 30 Apr 1997 06:41:51 +0000 (06:41 +0000)]
Catch up with the original.

27 years agobugfix: 'make depend' cause all LKMs rebuild
Andrey A. Chernov [Wed, 30 Apr 1997 06:04:17 +0000 (06:04 +0000)]
bugfix: 'make depend' cause all LKMs rebuild

27 years agoFix parsing of mount options with '=' in their name.
Mike Smith [Wed, 30 Apr 1997 05:48:22 +0000 (05:48 +0000)]
Fix parsing of mount options with '=' in their name.
PR: bin/3027
Submitted by: Louis Mamakos <louie@TransSys.COM>

27 years agoFix a bug that caused the relocs for linker set members in shared
John Polstra [Wed, 30 Apr 1997 05:18:08 +0000 (05:18 +0000)]
Fix a bug that caused the relocs for linker set members in shared
libraries to come out as 1-byte relocations instead of 4-byte
relocations.

Submitted by: Doug Rabson <dfr@nlsystems.com>

27 years agoWe don't support the AdvanSys product line of SCSI controllers, so
Justin T. Gibbs [Wed, 30 Apr 1997 04:46:01 +0000 (04:46 +0000)]
We don't support the AdvanSys product line of SCSI controllers, so
don't say that we do.

27 years agoUse our copy of dlfcn.h, not the version in /usr/include, which may not
Mike Smith [Wed, 30 Apr 1997 03:12:09 +0000 (03:12 +0000)]
Use our copy of dlfcn.h, not the version in /usr/include, which may not
be up-to-date when we are building.
Submitted by: Terry Lambert <terry@lambert.org>

27 years ago(1) Change logic for dependency: instead of defining "is_depended"
Satoshi Asami [Wed, 30 Apr 1997 03:12:05 +0000 (03:12 +0000)]
(1) Change logic for dependency: instead of defining "is_depended"
    (default: IS_DEPENDED?=install) target on the depended port, call
    DEPENDS_TARGET (default: install) from the depending port.

    Other than being more flexible (some ports don't require the
    dependency to be fully installed: see ghostscript4), this seems
    like a more natural thing to do.  (I never understood the
    convoluted logic that was used before.)

    By building packages with "DEPENDS_TARGET=package", I can avoid
    file-sharing problems (like gs3 binary going into the gs4
    package).

(2) Add new variable PATCH_SITE_SUBDIR and separate it from
    MASTER_SITE_SUBDIR.  Fixes linuxls port and is a correct thing to
    do anyway.
Initially suggested and reviewed by: fenner

(3) Add new variable MASTER_SITE_LOCAL for local ports.  Defaults to
    LOCAL_PORTS subdir on ftp.freebsd.org's distfiles dir.
Submitted by: jkh

27 years agoStaticize an unnecessarily global function: vputrele.
John Dyson [Wed, 30 Apr 1997 03:09:15 +0000 (03:09 +0000)]
Staticize an unnecessarily global function: vputrele.
Submitted by:  Michael Hancock <michaelh@cet.co.jp>

27 years agoIn comp_vgaregs skip cursor shape and 4 unused bytes (6 total)
Andrey A. Chernov [Wed, 30 Apr 1997 01:19:21 +0000 (01:19 +0000)]
In comp_vgaregs skip cursor shape and 4 unused bytes (6 total)
instead of 2 unused.
Previus code broke syscons on Diamond Stealth 3D 3000
(video mode ptr was NULL with no real reason)

27 years agoEnabled 'FIX_MP_TABLE_WORKS' code.
Steve Passe [Tue, 29 Apr 1997 22:12:32 +0000 (22:12 +0000)]
Enabled 'FIX_MP_TABLE_WORKS' code.
This code re-numbers PCI busses in the MP table to match PCI semantics
when the MP BIOS fails to do it properly.

Reviewed by: Peter Wemm <peter@spinner.DIALix.COM>

27 years agoUse a common numbering of the tty and net software interrupt levels
Peter Wemm [Tue, 29 Apr 1997 20:05:48 +0000 (20:05 +0000)]
Use a common numbering of the tty and net software interrupt levels
between the SMP and non-SMP case.  It simplifies the #ifdef's, since
NHWI changes (at least for the moment) when APIC's are involved.

27 years agoRemove stale comment, as suggested by Bruce.
Peter Wemm [Tue, 29 Apr 1997 20:00:41 +0000 (20:00 +0000)]
Remove stale comment, as suggested by Bruce.

27 years agoDocument -k flag here, too.
Garrett Wollman [Tue, 29 Apr 1997 17:48:58 +0000 (17:48 +0000)]
Document -k flag here, too.

27 years agoImplement Kerberized rcmd for rdump/rrestore. This is lacking the
Garrett Wollman [Tue, 29 Apr 1997 17:46:27 +0000 (17:46 +0000)]
Implement Kerberized rcmd for rdump/rrestore.   This is lacking  the
options one would normally expect to set the realm, enable encryption,
and whatnot, but this actually is able to contact the remote server,
so at least it's a start.  (As a bonus, the stripped static binary is
unquestionably exportable.)

27 years agoFor multi-session CD-ROMs, we have to account for previous sessions as
Joerg Wunsch [Tue, 29 Apr 1997 17:11:51 +0000 (17:11 +0000)]
For multi-session CD-ROMs, we have to account for previous sessions as
well in volume_space_size.  Otherwise, NFS exports won't work.

27 years agoUserland part of ISO9660 multi-session support. mount_cd9660(8) will
Joerg Wunsch [Tue, 29 Apr 1997 15:56:40 +0000 (15:56 +0000)]
Userland part of ISO9660 multi-session support.  mount_cd9660(8) will
now by default mount the last data track (thus last session), as
opposed to the very first session it has been mounting previously.
This is consistent with the ISO9660 multi-session idea, and the way
other operating systems are working.

There's support to mount arbitrary sessions using the -s option.  This
way, you can simulate multi-session CDs on something like vn devices
that don't support CDIO* ioctl commands.  You can also force the
historic behaviour with

mount -t cd9660 -o -s=0 /dev/cd0a /cdrom

27 years agoAdd support for ISO9660 multi-session CD-ROMs. This is just nothing
Joerg Wunsch [Tue, 29 Apr 1997 15:52:53 +0000 (15:52 +0000)]
Add support for ISO9660 multi-session CD-ROMs.  This is just nothing
but searching the directory on something else than the default
location.

NB: this comprises an interface change to the mount_cd9660(8)
utility (commit will follow).  You need to rebuild both.

I've got similar patches for RELENG_2_2, should i commit them too?

27 years agoAdds anon ftp virtual host capability to ftpd, using /etc/ftphosts for
David Nugent [Tue, 29 Apr 1997 12:42:08 +0000 (12:42 +0000)]
Adds anon ftp virtual host capability to ftpd, using /etc/ftphosts for
definition of a system's virtual hosts.

27 years agoEnable whiteout code since we now have the lite2 support for them.
Doug Rabson [Tue, 29 Apr 1997 10:03:10 +0000 (10:03 +0000)]
Enable whiteout code since we now have the lite2 support for them.

27 years agoYAMF22
Jordan K. Hubbard [Tue, 29 Apr 1997 09:14:24 +0000 (09:14 +0000)]
YAMF22

27 years agoDon't try to look up unionfs' <above>, <below> keywords as hostnames.
Doug Rabson [Tue, 29 Apr 1997 09:10:11 +0000 (09:10 +0000)]
Don't try to look up unionfs' <above>, <below> keywords as hostnames.

27 years agoadd an Id line, minor mdoc fix ups
John-Mark Gurney [Tue, 29 Apr 1997 09:09:40 +0000 (09:09 +0000)]
add an Id line, minor mdoc fix ups

add ftp to list of facilities in syslog.conf
Closes PR#3389
Submitted-by: Pius Fischer
27 years agoUse realpath() on the source as well as the target otherwise it can be
Doug Rabson [Tue, 29 Apr 1997 09:09:09 +0000 (09:09 +0000)]
Use realpath() on the source as well as the target otherwise it can be
impossible to unmount a union which was mounted with a relative pathname.

27 years agoFix so that amd can build with hesiod support.
Doug Rabson [Tue, 29 Apr 1997 08:43:52 +0000 (08:43 +0000)]
Fix so that amd can build with hesiod support.

Submitted by: Kevin Van Maren <vanmaren@marker.cs.utah.edu>

27 years agoFix makefiles so they work without having to 'make depend'
Doug Rabson [Tue, 29 Apr 1997 08:42:34 +0000 (08:42 +0000)]
Fix makefiles so they work without having to 'make depend'

27 years agoadd md5(1) to SEE ALSO section.
John-Mark Gurney [Tue, 29 Apr 1997 08:41:26 +0000 (08:41 +0000)]
add md5(1) to SEE ALSO section.

Closes PR#3409
Submitted-by: Josh Gilliam
27 years agoDynamically adjust size of displayed username to the longest username which
Jordan K. Hubbard [Tue, 29 Apr 1997 05:26:05 +0000 (05:26 +0000)]
Dynamically adjust size of displayed username to the longest username which
appears, not the longest _maximum_ username (this should probably also go
into 2.2, for the day when we bump up the username length there too).

Submitted-By: Terry Lambert <terry@lambert.org>
27 years agoFix a bug that caused some false mismatches when both FNM_PATHNAME
John Polstra [Tue, 29 Apr 1997 03:24:57 +0000 (03:24 +0000)]
Fix a bug that caused some false mismatches when both FNM_PATHNAME
and FNM_LEADING_DIR were specified and the pattern ended with "*".
Example: pattern="src/usr.sbin/w*", string="src/usr.sbin/watch/watch.8,v".
This should match, but did not.

27 years agoSynchronize with sys/i386/conf/options.i386 and sys/i386/isa/wd.c
KATO Takenori [Tue, 29 Apr 1997 03:15:47 +0000 (03:15 +0000)]
Synchronize with sys/i386/conf/options.i386 and sys/i386/isa/wd.c
revisions 1.43 and 1.129, respectively.

27 years agoadded mptable to directory list of usr.sbin programs.
Steve Passe [Tue, 29 Apr 1997 03:09:01 +0000 (03:09 +0000)]
added mptable to directory list of usr.sbin programs.

27 years ago#include <stdlib.h> in the C++ case and declare getenv and realloc
Steve Price [Tue, 29 Apr 1997 03:07:45 +0000 (03:07 +0000)]
#include <stdlib.h> in the C++ case and declare getenv and realloc
in the C case so that we don't have to depend on stdlib.h being
present.

Submitted by: Bruce Evans <bde@freefall.freebsd.org>

27 years agoadd a userland program for SMP support
Steve Passe [Tue, 29 Apr 1997 03:06:35 +0000 (03:06 +0000)]
add a userland program for SMP support
mptable examines and displays the MP table of an SMP motherboard.

27 years agoKeep style consistent.
Steve Price [Tue, 29 Apr 1997 03:05:38 +0000 (03:05 +0000)]
Keep style consistent.

Submitted by: Bruce Evans <bde@freefall.freebsd.org>

27 years agoSet the N_EXT (external) flag for all weak symbols. It makes no
John Polstra [Tue, 29 Apr 1997 02:11:48 +0000 (02:11 +0000)]
Set the N_EXT (external) flag for all weak symbols.  It makes no
sense to have a weak symbol that is not externally visible.  This
fixes many of the "relocation burb" warnings produced when compiling
C++ code with "-fpic".  Beyond eliminating warnings, it also makes
some things work that didn't work before.

27 years agoRevised fix for locking violation when unionfs calls vput with
KATO Takenori [Tue, 29 Apr 1997 02:06:07 +0000 (02:06 +0000)]
Revised fix for locking violation when unionfs calls vput with
UN_KLOCK flag.

When UN_KLOCK is set, VOP_UNLOCK should keep uppervp locked and clear
UN_ULOCK flag.  To do this, when UN_KLOCK is set, (1) union_unlock
clears UN_ULOCK and does not clear UN_KLOCK, (2) union_lock() does not
access uppervp and does not clear UN_KLOCK, and (3) callers of
vput/VOP_UNLOCK should clear UN_KLOCK.  For example, vput becomes:

SETKLOCK(union_node);
vput(vnode);
CLEARKLOCK(union_node);

where SETKLOCK macro sets UN_KLOCK and CLEARKLOCK macro clears
UN_KLOCK.

27 years agoMove CMD640 option from kernel Makefile into opt_wd.h
Stefan Eßer [Mon, 28 Apr 1997 19:26:18 +0000 (19:26 +0000)]
Move CMD640 option from kernel Makefile into opt_wd.h

Submitted by: Wolfgang Helbig <helbig@mx.ba-stuttgart.de>

27 years agoSynchronize with sys/i386/isa/isa.c revision 1.83.
KATO Takenori [Mon, 28 Apr 1997 15:51:15 +0000 (15:51 +0000)]
Synchronize with sys/i386/isa/isa.c revision 1.83.

27 years agoSynchornize with sys/i386/conf/options.i386 revision 1.42.
KATO Takenori [Mon, 28 Apr 1997 15:49:55 +0000 (15:49 +0000)]
Synchornize with sys/i386/conf/options.i386 revision 1.42.

27 years agoSynchronize with sys/i386/conf/Makefile.i386 revision 1.95.
KATO Takenori [Mon, 28 Apr 1997 15:48:49 +0000 (15:48 +0000)]
Synchronize with sys/i386/conf/Makefile.i386 revision 1.95.

27 years agoBack out unnecessary overly-paranoid paranoia test from here too; yp_access()
Bill Paul [Mon, 28 Apr 1997 14:22:31 +0000 (14:22 +0000)]
Back out unnecessary overly-paranoid paranoia test from here too; yp_access()
does the right thing already.

27 years agoBack out one of my paranoia tests from the last commit here; yp_access()
Bill Paul [Mon, 28 Apr 1997 14:18:38 +0000 (14:18 +0000)]
Back out one of my paranoia tests from the last commit here; yp_access()
already deals with it appropriately. (You know you've been working too
hard when you forget how your own code works.)

27 years agoCondition SO_PRIVSTATE twiddling on the definedness of SO_PRIVSTATE.
Garrett Wollman [Mon, 28 Apr 1997 13:55:07 +0000 (13:55 +0000)]
Condition SO_PRIVSTATE twiddling on the definedness of SO_PRIVSTATE.
(This was done as a #ifdef to keep source compatibility between 2.2 and 3.0.)

27 years agoAdapt to rc.conf file format.
Jordan K. Hubbard [Mon, 28 Apr 1997 10:31:14 +0000 (10:31 +0000)]
Adapt to rc.conf file format.

27 years agoYAMF22
Jordan K. Hubbard [Mon, 28 Apr 1997 10:14:45 +0000 (10:14 +0000)]
YAMF22

27 years agoYAMF22
Jordan K. Hubbard [Mon, 28 Apr 1997 10:07:21 +0000 (10:07 +0000)]
YAMF22

27 years agoBack out mount change here (which is the only branch it should have
Jordan K. Hubbard [Mon, 28 Apr 1997 09:35:59 +0000 (09:35 +0000)]
Back out mount change here (which is the only branch it should have
occupied in the first place :)

27 years agoMount MFS read/write as in days of yore.
Jordan K. Hubbard [Mon, 28 Apr 1997 09:15:18 +0000 (09:15 +0000)]
Mount MFS read/write as in days of yore.

27 years agocleaned out an old FIXME.
Steve Passe [Mon, 28 Apr 1997 08:39:48 +0000 (08:39 +0000)]
cleaned out an old FIXME.

27 years agoYAMF22
Jordan K. Hubbard [Mon, 28 Apr 1997 07:35:37 +0000 (07:35 +0000)]
YAMF22

27 years agoYAMF22
Jordan K. Hubbard [Mon, 28 Apr 1997 06:15:49 +0000 (06:15 +0000)]
YAMF22

27 years agoYAMF22
Jordan K. Hubbard [Mon, 28 Apr 1997 05:51:57 +0000 (05:51 +0000)]
YAMF22

27 years agoChange spelling of ECBDIC to EBCDIC. Closes PR #3402
Steve Price [Mon, 28 Apr 1997 04:07:29 +0000 (04:07 +0000)]
Change spelling of ECBDIC to EBCDIC.  Closes PR #3402

Submitted by: h-nokubi@nmit.mt.nec.co.jp

27 years agoTurn on the new type builtin.
Steve Price [Mon, 28 Apr 1997 03:50:07 +0000 (03:50 +0000)]
Turn on the new type builtin.

27 years agoFix problems using -Wwrite-strings and -Wcast-qual with yacc generated
Steve Price [Mon, 28 Apr 1997 03:36:13 +0000 (03:36 +0000)]
Fix problems using -Wwrite-strings and -Wcast-qual with yacc generated
parsers.  Closes PR #2792.

Submitted by: Tim Vanderhoek

27 years agoSync with NetBSD's revision 1.29 of this file.
Steve Price [Mon, 28 Apr 1997 03:28:43 +0000 (03:28 +0000)]
Sync with NetBSD's revision 1.29 of this file.

Obtained from: NetBSD

27 years agoRemove an extra semicolon and nuke register keyword usage.
Steve Price [Mon, 28 Apr 1997 03:27:12 +0000 (03:27 +0000)]
Remove an extra semicolon and nuke register keyword usage.

27 years agoFix redirection of unopened file descriptors and nuke register
Steve Price [Mon, 28 Apr 1997 03:25:13 +0000 (03:25 +0000)]
Fix redirection of unopened file descriptors and nuke register
keyword usage.

Obtained from: NetBSD

27 years agoRevert changes from rev 1.16 to 1.17 for now. Closes PR 2879.
Steve Price [Mon, 28 Apr 1997 03:22:09 +0000 (03:22 +0000)]
Revert changes from rev 1.16 to 1.17 for now.  Closes PR 2879.

27 years agomdoc-ify and cleanup UNIMPLEMENTED messages. Closes PR 2880.
Steve Price [Mon, 28 Apr 1997 03:20:09 +0000 (03:20 +0000)]
mdoc-ify and cleanup UNIMPLEMENTED messages.  Closes PR 2880.

27 years agoReorder declarations slightly.
Steve Price [Mon, 28 Apr 1997 03:17:51 +0000 (03:17 +0000)]
Reorder declarations slightly.

27 years agoAvoid a possible NULL pointer dereference, nuke register usage,
Steve Price [Mon, 28 Apr 1997 03:16:12 +0000 (03:16 +0000)]
Avoid a possible NULL pointer dereference, nuke register usage,
and remove an unnecessary reset.

Obtained from: NetBSD

27 years agoNuke register keyword usage, #if -> #ifdef, and avoid void *
Steve Price [Mon, 28 Apr 1997 03:14:04 +0000 (03:14 +0000)]
Nuke register keyword usage, #if -> #ifdef, and avoid void *
arithmetic.

Obtained from: NetBSD

27 years agoNuke register keyword usage, rename pread to preadfd, and open
Steve Price [Mon, 28 Apr 1997 03:11:52 +0000 (03:11 +0000)]
Nuke register keyword usage, rename pread to preadfd, and open
input files FD_CLOEXEC.

Obtained from: NetBSD

27 years agoWhen NO_HISTORY is defined and the history command is used
Steve Price [Mon, 28 Apr 1997 03:10:31 +0000 (03:10 +0000)]
When NO_HISTORY is defined and the history command is used
show an error message.

Inspired by: NetBSD

27 years agoAdd a type builtin and nuke register keyword usage.
Steve Price [Mon, 28 Apr 1997 03:08:38 +0000 (03:08 +0000)]
Add a type builtin and nuke register keyword usage.

Obtained from: NetBSD

27 years agoNuke register keyword usage and #if -> #ifdef.
Steve Price [Mon, 28 Apr 1997 03:06:52 +0000 (03:06 +0000)]
Nuke register keyword usage and #if -> #ifdef.

Obtained from: NetBSD

27 years agoShamelessly pilfer most of NetBSD's Makefile so that the
Steve Price [Mon, 28 Apr 1997 03:03:23 +0000 (03:03 +0000)]
Shamelessly pilfer most of NetBSD's Makefile so that the
problem with 'make -j n' and no .depend file goes away.
I think Bruce mentioned this somewhere on one of the
mailing lists.

Obtained from: NetBSD

27 years agoAlways include PROT_READ for Linux mmap operations.
Mike Smith [Mon, 28 Apr 1997 02:53:17 +0000 (02:53 +0000)]
Always include PROT_READ for Linux mmap operations.
Submitted by: Hannu Savolainen <hannu@voxware.pp.fi> via jkh

27 years agoremoved TEST_CPUHITS code.
Steve Passe [Mon, 28 Apr 1997 01:47:55 +0000 (01:47 +0000)]
removed TEST_CPUHITS code.

replaced push/pop of %ds with use of 'ss' prefix in Xinvltlb.

Submitted by:  Bruce Evans <bde@zeta.org.au>

27 years agoremoved TEST_CPUHITS code.
Steve Passe [Mon, 28 Apr 1997 01:46:31 +0000 (01:46 +0000)]
removed TEST_CPUHITS code.

27 years agoremoved all the TEST_UPPERPRIO crud.
Steve Passe [Mon, 28 Apr 1997 01:08:47 +0000 (01:08 +0000)]
removed all the TEST_UPPERPRIO crud.

27 years agoremove all the SMP_INVLTLB defines, making the code default for APIC_IO.
Steve Passe [Mon, 28 Apr 1997 00:25:00 +0000 (00:25 +0000)]
remove all the SMP_INVLTLB defines, making the code default for APIC_IO.

Reviewed by: informal discussion with Peter Wemm <peter@spinner.DIALix.COM>

27 years agoremove the SMP_INVLTLB option, making the code default for APIC_IO.
Steve Passe [Mon, 28 Apr 1997 00:24:28 +0000 (00:24 +0000)]
remove the SMP_INVLTLB option, making the code default for APIC_IO.

Reviewed by: informal discussion with Peter Wemm <peter@spinner.DIALix.COM>

27 years agoremove all the SMP_INVLTLB defines, making the code default for APIC_IO.
Steve Passe [Mon, 28 Apr 1997 00:24:00 +0000 (00:24 +0000)]
remove all the SMP_INVLTLB defines, making the code default for APIC_IO.
replace invldebug with invltlb_ok for throttling smp_invltlb() during boot.

Reviewed by: informal discussion with Peter Wemm <peter@spinner.DIALix.COM>

27 years agoSubmitted by: Wolfgang Helbig <helbig@MX.BA-Stuttgart.De>
Steve Passe [Mon, 28 Apr 1997 00:02:22 +0000 (00:02 +0000)]
Submitted by: Wolfgang Helbig <helbig@MX.BA-Stuttgart.De>
make world fails when it tries to build LKMs because the files
opt_smp.h
opt_smp_invltbl.h
are missing.

This patch to /usr/src/share/mk/bsd.kmod.mk is a temporary workaround.

Note that LKMs built in this way may or may NOT work properly with an
SMP kernel.

27 years agochange kvm-fbsd.c so that kernel debugging works again.
Gary Jennejohn [Sun, 27 Apr 1997 21:36:49 +0000 (21:36 +0000)]
change kvm-fbsd.c so that kernel debugging works again.

document the -k and -wcore options in the manpage.

change Makefile to use a local copy of the manpage.