openbsd
27 years agoBe more careful about enabling nubus interrupts.
briggs [Thu, 27 Mar 1997 14:58:53 +0000 (14:58 +0000)]
Be more careful about enabling nubus interrupts.

27 years agoUse P_MD_REGS instead of P_MD + MD_REGS (consistent with hp300)
niklas [Thu, 27 Mar 1997 12:06:45 +0000 (12:06 +0000)]
Use P_MD_REGS instead of P_MD + MD_REGS (consistent with hp300)

27 years agoUse genassym.sh
niklas [Thu, 27 Mar 1997 11:52:06 +0000 (11:52 +0000)]
Use genassym.sh

27 years agoImport genassym.sh method from NetBSD
niklas [Thu, 27 Mar 1997 11:50:45 +0000 (11:50 +0000)]
Import genassym.sh method from NetBSD

27 years agoAdd NCR variant of SVR4, as it has some specifics
niklas [Thu, 27 Mar 1997 08:40:36 +0000 (08:40 +0000)]
Add NCR variant of SVR4, as it has some specifics

27 years agoImplement NOFILE_MAX--hard limit on max descriptors per proccess.
millert [Thu, 27 Mar 1997 05:35:25 +0000 (05:35 +0000)]
Implement NOFILE_MAX--hard limit on max descriptors per proccess.
Future direction, might make sense to make this a kernel symbol tha
is sysctl-able.

27 years agoMake kern.osrevision (sysctl) and /kern/osrev report value of
millert [Thu, 27 Mar 1997 04:39:35 +0000 (04:39 +0000)]
Make kern.osrevision (sysctl) and /kern/osrev report value of
OpenBSD define, not BSD which is a bit more useful.

27 years agoyet another random() converted to arc4random()
provos [Thu, 27 Mar 1997 01:09:38 +0000 (01:09 +0000)]
yet another random() converted to arc4random()

27 years agouse arc4random instead of random
provos [Thu, 27 Mar 1997 00:36:32 +0000 (00:36 +0000)]
use arc4random instead of random

27 years agoSlight cleanup, more needed
weingart [Thu, 27 Mar 1997 00:30:52 +0000 (00:30 +0000)]
Slight cleanup, more needed

27 years agomd5 stuff
provos [Thu, 27 Mar 1997 00:10:52 +0000 (00:10 +0000)]
md5 stuff

27 years agoadd comment on md5
provos [Thu, 27 Mar 1997 00:07:14 +0000 (00:07 +0000)]
add comment on md5

27 years agosupport for md5 passwords
provos [Thu, 27 Mar 1997 00:05:41 +0000 (00:05 +0000)]
support for md5 passwords

27 years agoadded ipsec subdir
provos [Wed, 26 Mar 1997 23:14:57 +0000 (23:14 +0000)]
added ipsec subdir

27 years agoAdd OpenBSD tag. Remove redundant test.
niklas [Wed, 26 Mar 1997 22:14:41 +0000 (22:14 +0000)]
Add OpenBSD tag.  Remove redundant test.

27 years agoUse MKDEP instead of mkdep. Split STRIP into STRIP & STRIPFLAGS.
niklas [Wed, 26 Mar 1997 22:11:20 +0000 (22:11 +0000)]
Use MKDEP instead of mkdep.  Split STRIP into STRIP & STRIPFLAGS.
Use HOSTCC for genassym.

27 years agoAdd ast, tel and IPSEC
niklas [Wed, 26 Mar 1997 21:53:29 +0000 (21:53 +0000)]
Add ast, tel and IPSEC

27 years agoGot rid of an unused function
niklas [Wed, 26 Mar 1997 21:51:31 +0000 (21:51 +0000)]
Got rid of an unused function

27 years agofix a bunch of errors
deraadt [Wed, 26 Mar 1997 20:18:53 +0000 (20:18 +0000)]
fix a bunch of errors

27 years agocalculate width for select() more carefully
deraadt [Wed, 26 Mar 1997 19:41:59 +0000 (19:41 +0000)]
calculate width for select() more carefully

27 years agooxtabs is converse of tabs; gwing@primenet.com.au
deraadt [Wed, 26 Mar 1997 19:12:14 +0000 (19:12 +0000)]
oxtabs is converse of tabs; gwing@primenet.com.au

27 years agomore careful strncpy
deraadt [Wed, 26 Mar 1997 19:08:06 +0000 (19:08 +0000)]
more careful strncpy

27 years agouse support.s from m68k/
niklas [Wed, 26 Mar 1997 19:04:25 +0000 (19:04 +0000)]
use support.s from m68k/

27 years agoJust some initial KNF in preparation for some other changes coming soon
niklas [Wed, 26 Mar 1997 18:45:31 +0000 (18:45 +0000)]
Just some initial KNF in preparation for some other changes coming soon

27 years agoRemove stuff now in m68k/
niklas [Wed, 26 Mar 1997 18:30:50 +0000 (18:30 +0000)]
Remove stuff now in m68k/

27 years agocopy.s gets done from files.m68k now
niklas [Wed, 26 Mar 1997 18:25:37 +0000 (18:25 +0000)]
copy.s gets done from files.m68k now

27 years agoproper english
deraadt [Wed, 26 Mar 1997 18:03:57 +0000 (18:03 +0000)]
proper english

27 years agobail like ruptime; netbsd
deraadt [Wed, 26 Mar 1997 18:02:37 +0000 (18:02 +0000)]
bail like ruptime; netbsd

27 years agos/main/listmain/; seebs@solon.com
deraadt [Wed, 26 Mar 1997 17:46:44 +0000 (17:46 +0000)]
s/main/listmain/; seebs@solon.com

27 years agologinname copying in fill_eproc; gwing@primenet.com.au
deraadt [Wed, 26 Mar 1997 17:23:50 +0000 (17:23 +0000)]
loginname copying in fill_eproc; gwing@primenet.com.au

27 years agoReplace "isa_io" with "bus". Not isa specific.
pefo [Wed, 26 Mar 1997 14:44:38 +0000 (14:44 +0000)]
Replace "isa_io" with "bus". Not isa specific.

27 years agosync
downsj [Wed, 26 Mar 1997 08:41:42 +0000 (08:41 +0000)]
sync

27 years agoMerge changes from NetBSD, up to 3/19/97; undoes some local changes.
downsj [Wed, 26 Mar 1997 08:32:36 +0000 (08:32 +0000)]
Merge changes from NetBSD, up to 3/19/97; undoes some local changes.
Changed to match new arch/m68k code.
genassym.cf is currently just a place holder.

27 years agoMerge changes from NetBSD, 3/19/97; enforce use of new files in file.m68k.
downsj [Wed, 26 Mar 1997 08:23:52 +0000 (08:23 +0000)]
Merge changes from NetBSD, 3/19/97; enforce use of new files in file.m68k.

All of the m68k ports need fixing, in various ways.  From a casual look,
mvme68k and sun3 need the most work.

Tested on hp300, from which sig_machdep.c is based.

27 years agoRegen.
downsj [Wed, 26 Mar 1997 08:12:39 +0000 (08:12 +0000)]
Regen.

27 years agoMerge in NetBSD changes as if 3/19/97, undo some of ours.
downsj [Wed, 26 Mar 1997 08:11:04 +0000 (08:11 +0000)]
Merge in NetBSD changes as if 3/19/97, undo some of ours.

27 years agocorrect a typo
deraadt [Wed, 26 Mar 1997 05:23:57 +0000 (05:23 +0000)]
correct a typo

27 years agoinstall ktrace(2)
deraadt [Wed, 26 Mar 1997 05:11:54 +0000 (05:11 +0000)]
install ktrace(2)

27 years agoshow a backslash properly; david@mono.org
deraadt [Wed, 26 Mar 1997 02:43:17 +0000 (02:43 +0000)]
show a backslash properly; david@mono.org

27 years agofix *link.2 Xr thingies; bgrayson@r2d2.ece.utexas.edu + more
deraadt [Wed, 26 Mar 1997 02:29:03 +0000 (02:29 +0000)]
fix *link.2 Xr thingies; bgrayson@r2d2.ece.utexas.edu + more

27 years agomore pedantic description of secure levels on file flags; rickb@iaw.on.ca
deraadt [Wed, 26 Mar 1997 01:59:37 +0000 (01:59 +0000)]
more pedantic description of secure levels on file flags; rickb@iaw.on.ca

27 years agodeal with CD drives that indicate 2352 bytes per block...
deraadt [Wed, 26 Mar 1997 01:53:48 +0000 (01:53 +0000)]
deal with CD drives that indicate 2352 bytes per block...

27 years agoNote that default blocking factor is 20.
millert [Wed, 26 Mar 1997 01:20:45 +0000 (01:20 +0000)]
Note that default blocking factor is 20.

27 years agounterminated buffer; charnier@xp11.frmug.org
deraadt [Wed, 26 Mar 1997 01:19:49 +0000 (01:19 +0000)]
unterminated buffer; charnier@xp11.frmug.org

27 years agoBlocking factor in tar is in 512byte blocks. Makes -b act like
millert [Wed, 26 Mar 1997 01:14:13 +0000 (01:14 +0000)]
Blocking factor in tar is in 512byte blocks.  Makes -b act like
``real'' tars.

27 years agoclean; netbsd
deraadt [Wed, 26 Mar 1997 00:45:56 +0000 (00:45 +0000)]
clean; netbsd

27 years agoxref rwhod; netbsd
deraadt [Wed, 26 Mar 1997 00:45:46 +0000 (00:45 +0000)]
xref rwhod; netbsd

27 years agosync to rwhod.h
deraadt [Wed, 26 Mar 1997 00:39:34 +0000 (00:39 +0000)]
sync to rwhod.h

27 years agono libcrypt, fix DPADD
deraadt [Wed, 26 Mar 1997 00:33:39 +0000 (00:33 +0000)]
no libcrypt, fix DPADD

27 years agocloselog() more often
deraadt [Tue, 25 Mar 1997 23:12:23 +0000 (23:12 +0000)]
closelog() more often

27 years agocareful with name lengths and the xid
deraadt [Tue, 25 Mar 1997 23:08:57 +0000 (23:08 +0000)]
careful with name lengths and the xid

27 years agospelling error
deraadt [Tue, 25 Mar 1997 22:53:33 +0000 (22:53 +0000)]
spelling error

27 years agoSplat .message file if it exists when user cd's to a directory
millert [Tue, 25 Mar 1997 22:47:08 +0000 (22:47 +0000)]
Splat .message file if it exists when user cd's to a directory
like wu-ftpd.  Based on changes from Oliver.

27 years agoexec with argv[0] as shell, not "sh"; era@iki.fi
deraadt [Tue, 25 Mar 1997 21:58:09 +0000 (21:58 +0000)]
exec with argv[0] as shell, not "sh"; era@iki.fi

27 years agoconfig file paranoia
bitblt [Tue, 25 Mar 1997 21:51:43 +0000 (21:51 +0000)]
config file paranoia

27 years agocorrect string empty test; yokota@freebsd
deraadt [Tue, 25 Mar 1997 21:28:12 +0000 (21:28 +0000)]
correct string empty test; yokota@freebsd

27 years agoensure strftime buf has NUL; yokota@freebsd
deraadt [Tue, 25 Mar 1997 21:24:12 +0000 (21:24 +0000)]
ensure strftime buf has NUL; yokota@freebsd

27 years agodo not chmod file before unlink; firebug@atlantic.net
deraadt [Tue, 25 Mar 1997 21:09:47 +0000 (21:09 +0000)]
do not chmod file before unlink; firebug@atlantic.net

27 years agoGet rid of userland includes
niklas [Tue, 25 Mar 1997 20:30:41 +0000 (20:30 +0000)]
Get rid of userland includes

27 years agomove the .; enami@ba2.so-net.or.jp
deraadt [Tue, 25 Mar 1997 20:20:51 +0000 (20:20 +0000)]
move the .; enami@ba2.so-net.or.jp

27 years agoOk here is the mvme88k userland, minus a few important pieces.
rahnds [Tue, 25 Mar 1997 17:06:34 +0000 (17:06 +0000)]
Ok here is the mvme88k userland, minus a few important pieces.
as, ld change need to be merged. It is known that shared libraries and C++
do not work.
If anyone wants to step forward and take over the mvme88k port please do.
It was dumped on me when the original developer, Nivas, left his previous
job and started a new one.

I am trying to get the port in a shape that someone could take over.

27 years agoFixed obj-dir building, compacted the bootblock enough to fit (at least)
niklas [Tue, 25 Mar 1997 17:05:49 +0000 (17:05 +0000)]
Fixed obj-dir building, compacted the bootblock enough to fit (at least)
the HD one, floppy is still too large.  Added our BOOT_CONFIG -c option.
Some small style nits.

27 years agoCorrect typo
niklas [Tue, 25 Mar 1997 16:53:56 +0000 (16:53 +0000)]
Correct typo

27 years agoCorrect typo + moved Teles S0/16.3 port to what the manual says
niklas [Tue, 25 Mar 1997 16:52:13 +0000 (16:52 +0000)]
Correct typo + moved Teles S0/16.3 port to what the manual says

27 years agoUse GENERIC instead of the obsolete INSTALL kernel
niklas [Tue, 25 Mar 1997 16:48:59 +0000 (16:48 +0000)]
Use GENERIC instead of the obsolete INSTALL kernel

27 years agoMake AS, AR & LD overrideable
niklas [Tue, 25 Mar 1997 14:31:07 +0000 (14:31 +0000)]
Make AS, AR & LD overrideable

27 years agoBack out my previous kludge to make tar files with short trailers work in
millert [Tue, 25 Mar 1997 09:30:19 +0000 (09:30 +0000)]
Back out my previous kludge to make tar files with short trailers work in
favor of a simpler change--never try to do a media change when invoked
as tar.  This fixes problems that crop up if gzip exits unhappily and
others.

27 years agoAddition of the ed ethernet driver
niklas [Tue, 25 Mar 1997 08:00:42 +0000 (08:00 +0000)]
Addition of the ed ethernet driver

27 years agoFrom scottr@netbsd.org:
briggs [Tue, 25 Mar 1997 05:02:39 +0000 (05:02 +0000)]
From scottr@netbsd.org:
Several minor changes:

 - Indent some comments appropriately per KNF.
 - Output all diagnostics using the device name rather than a static string.
 - Fix a (mostly marmless) bug in snintr(), where loss of heartbeat wasn't
   properly ignored and could mask the `receive descriptors exhausted'
   diagnostic.

27 years agoFrom scottr@netbsd.org:
briggs [Tue, 25 Mar 1997 04:58:43 +0000 (04:58 +0000)]
From scottr@netbsd.org:
Several more changes to move us toward MI-ness:

 - Use more consistent and portable types in the softc.
 - Map registers using an array of bus_size_t offsets, and set up the
   mapping in the attach code (thanks to Jason Thorpe for suggesting
   this!).
 - Disable the ae-specific watchdog, which is no longer necessary in
   the general case.

Still remaining:  split out functions used to copy data to/from the
card, and retain a way to have a local driver name with the MI code.

27 years agoMake sure head, tail, and nitems get zero'd even if we
millert [Mon, 24 Mar 1997 02:54:15 +0000 (02:54 +0000)]
Make sure head, tail, and nitems get zero'd even if we
can't cd to the target dir.  Fixes rogue pointer problem
introduced with safe chdir changes.

27 years agoAdding more arches require more structuring and cleanup. More to come...
pefo [Sun, 23 Mar 1997 11:34:25 +0000 (11:34 +0000)]
Adding more arches require more structuring and cleanup. More to come...

27 years agoSome changes for arc local bus com & lpt.
pefo [Sun, 23 Mar 1997 11:02:58 +0000 (11:02 +0000)]
Some changes for arc local bus com & lpt.

27 years agoMoved from pica since they are not Pica bus specific
pefo [Sun, 23 Mar 1997 10:14:58 +0000 (10:14 +0000)]
Moved from pica since they are not Pica bus specific

27 years agoUse POSIX regexp, not v8 to avoid linking with -lcompat.
millert [Sun, 23 Mar 1997 04:43:22 +0000 (04:43 +0000)]
Use POSIX regexp, not v8 to avoid linking with -lcompat.
Parsing config files w/o regular expressions may be slightly
slower but since this is normally done only once it's not
a big deal.

27 years agoMinor updates from lite2 + some KNF and strdup(3) checking.
millert [Sun, 23 Mar 1997 03:52:13 +0000 (03:52 +0000)]
Minor updates from lite2 + some KNF and strdup(3) checking.
Also, no need to change umask now that we use mkstemp(3).

27 years ago-Wall happiness, use mkstemp(3) instead of mktemp(3), and some
millert [Sun, 23 Mar 1997 03:04:28 +0000 (03:04 +0000)]
-Wall happiness, use mkstemp(3) instead of mktemp(3), and some
minor KNF.

27 years agoremoved code that was inside #if 0 ... #endif
kstailey [Sat, 22 Mar 1997 02:51:06 +0000 (02:51 +0000)]
removed code that was inside #if 0 ... #endif

27 years agoget __LDPGSZ def
niklas [Sat, 22 Mar 1997 00:20:48 +0000 (00:20 +0000)]
get __LDPGSZ def

27 years agoMake this build work with objdirs, at least as links
niklas [Sat, 22 Mar 1997 00:19:56 +0000 (00:19 +0000)]
Make this build work with objdirs, at least as links
Still need work for plain obj dirs.

27 years agoFix two typos in fts(3). NetBSD PR#3364 and NetBSD PR#3368
millert [Fri, 21 Mar 1997 21:36:36 +0000 (21:36 +0000)]
Fix two typos in fts(3).  NetBSD PR#3364 and NetBSD PR#3368
from Enami Tsugutomo.

27 years agoMake struct rusage comments consistent between header file
millert [Fri, 21 Mar 1997 21:27:21 +0000 (21:27 +0000)]
Make struct rusage comments consistent between header file
and man page.  NetBSD PR #3367

27 years agoAdd in recent NetBSD changes we didn't already have:
millert [Fri, 21 Mar 1997 20:59:26 +0000 (20:59 +0000)]
Add in recent NetBSD changes we didn't already have:
    Always compile complete.c but ifdef out the bits if -DSMALL (christos)
    reset interactive mode correctly in auto_fetch() mget mode (lukem)

27 years agouse ${COMPILE.c} even for the custom build
gvf [Fri, 21 Mar 1997 14:42:42 +0000 (14:42 +0000)]
use ${COMPILE.c} even for the custom build

27 years agoFix alignment in srt0.c, from NetBSD.
downsj [Fri, 21 Mar 1997 07:16:20 +0000 (07:16 +0000)]
Fix alignment in srt0.c, from NetBSD.

27 years agocruft
downsj [Fri, 21 Mar 1997 04:00:35 +0000 (04:00 +0000)]
cruft

27 years agosince cpio is now pax you can have it too.
kstailey [Fri, 21 Mar 1997 03:33:23 +0000 (03:33 +0000)]
since cpio is now pax you can have it too.

27 years agoYet more DDB type correctness
niklas [Fri, 21 Mar 1997 02:10:42 +0000 (02:10 +0000)]
Yet more DDB type correctness

27 years agoMake ddb variables long
niklas [Fri, 21 Mar 1997 01:02:58 +0000 (01:02 +0000)]
Make ddb variables long

27 years agoNew FIXUP_PC_AFTER_BREAK convention from NetBSD
niklas [Fri, 21 Mar 1997 00:48:41 +0000 (00:48 +0000)]
New FIXUP_PC_AFTER_BREAK convention from NetBSD

27 years agoInstead of explicit extern, include the include file which should do it
niklas [Fri, 21 Mar 1997 00:44:18 +0000 (00:44 +0000)]
Instead of explicit extern, include the include file which should do it

27 years agoMake ddb variables long (for 64bit archs). from netbsd
niklas [Fri, 21 Mar 1997 00:41:17 +0000 (00:41 +0000)]
Make ddb variables long (for 64bit archs).  from netbsd

27 years agoNetBSD sync to sth like 970218
niklas [Fri, 21 Mar 1997 00:36:34 +0000 (00:36 +0000)]
NetBSD sync to sth like 970218

27 years agoAdd the bus_space_{read,write}_raw_multi_[248] extensions
niklas [Fri, 21 Mar 1997 00:16:32 +0000 (00:16 +0000)]
Add the bus_space_{read,write}_raw_multi_[248] extensions

27 years ago64-bit clean. if_ed.c now works on alpha.
niklas [Thu, 20 Mar 1997 23:59:58 +0000 (23:59 +0000)]
64-bit clean.  if_ed.c now works on alpha.

27 years agoupdate for adaptec support
chuck [Thu, 20 Mar 1997 22:03:45 +0000 (22:03 +0000)]
update for adaptec support

27 years agoMAJOR CHANGES: [contributed by Chuck Cranor <chuck@ccrc.wustl.edu> and
chuck [Thu, 20 Mar 1997 22:03:03 +0000 (22:03 +0000)]
MAJOR CHANGES: [contributed by Chuck Cranor <chuck@ccrc.wustl.edu> and
                Anne Hutton <hutton@isi.edu>]:
   - add support for Adaptec 155 PCI ATM cards (e.g. ANA-5940)
          - add sc->is_adaptec to handle differences between cards.
          - break out MID_MK_TXQ/MID_MK_RXQ seperate macros to handle
                  the new Adaptec format TXQ/RXQ.
          - adjust en_dqneed to return 1 on ADP (since the Adaptec can
                  DMA anything in one DRQ/DTQ!)
          - add hook for a bus specific reset function (adaptec has
                  a seperate reset register that needs to be hit when
                  resettting the midway).
          - adjust DMA test to not worry about burst sizes on the
                  adaptec (since it handles it all for us!) and to handle
                  the new DTQ/DRQ format.
          - add Adaptec DMA support to en_txlaunch() and en_service()

BUG FIXES:
   - fixed receiver panic under heavy load ("lost mbuf in slot 0!").
          when the reassembly buffer overflows, the T-bit is set in
          the RDB and the data field is empty.  en_service() sets up
          a 4-byte (RDB size) dummy DMA without IF_ENQUEUE.  but the
          recv intr handling in en_intr() always does IF_DEQUEUE.
          as a result, a successive recv intr loses its mbuf and
          leads to a panic.  the solution is to only IF_DEQUEUE if
          the interrupt has non-zero length (indicating that there
          is an mbuf to get).   in order for this to work, EN_DQ_MK
          must always be non-zero.   we do this by or'ing in an unused
          bit (0x80000).
                  reported by: Kenjiro Cho <kjc@csl.sony.co.jp>

   - fix setting of transmit channel when txspeed[] is non-zero
          (e.g. traffic shaping).    the old scheme didn't work
          properly (it allowed the same VCI to use multiple tx channels
          thus defeating the txspeed[] parameter).   the new scheme
          statically assigns a VC to a channel when txspeed[] is set.
          [note that the code to set txspeed[] isn't in the driver right
          now since a MI interface to do this hasn't been made yet]
          we add sc->txvc2slot[] and sc->txslot[n].nref for this.
                  reported by: Kenjiro Cho <kjc@csl.sony.co.jp>,
                                  Milind M Buddihikot <milind@ccrc.wustl.edu>,
                                  Dong Lin <dong@eecs.harvard.edu>

    - if aal5 frame has a CRC error then the length field in the aal5 trailer
          may not be valid, so we can not use it [and we must dump the frame]
                  contributed by: Yuhang Sun <sunyh@dworkin.wustl.edu> & chuck

   - when doing SRAM copies, be sure to round up the length to the next
          largest word (otherwise the driver will try to do a byte clean
          up DMA and then get an ID error interrupt).

MINOR CLEANUPS:
   - add some extra support for a few more versions of FreeBSD
                  contributed by: Kenjiro Cho <kjc@csl.sony.co.jp>

   - clean up loops in DMA test
                  contributed by: Kenjiro Cho <kjc@csl.sony.co.jp>

   - restructure and cleanup of en_read/en_write macros/inlines

   - clean up some byte ordering stuff so that we are consistant throughout
          the driver

27 years agoregen
chuck [Thu, 20 Mar 1997 20:31:58 +0000 (20:31 +0000)]
regen

27 years ago - add Adaptec ATM ID's (0x5900, 0x5905). still trying to find out how they
chuck [Thu, 20 Mar 1997 20:29:55 +0000 (20:29 +0000)]
 - add Adaptec ATM ID's (0x5900, 0x5905).   still trying to find out how they
        correspond to the ANA numbers...  certainly includes ANA-5940.
 - add Efficient Nets product 0x0000.  according to linux pci.h
        the 0x0000 is an FPGA version of the midway card and the 0x0002
        is the ASIC version.

27 years agouse ${COMPILE.c} even for the custom build
niklas [Thu, 20 Mar 1997 19:10:20 +0000 (19:10 +0000)]
use ${COMPILE.c} even for the custom build