]> CyberLeo.Net >> Repos - FreeBSD/FreeBSD.git/log
FreeBSD/FreeBSD.git
25 years agoUpdates for alignment rounding. XXX this is highly machine dependent and
Peter Wemm [Fri, 9 Oct 1998 23:59:01 +0000 (23:59 +0000)]
Updates for alignment rounding.  XXX this is highly machine dependent and
should probably be moved to i386/i386/link_machdep.c (and the same for the
alpha).
Implement "deleting" a preloaded module by destroying it's tags. This is a
hack.  We cannot reuse the data, it's been destroyed by relocation,
statically initialized variables have been modified, etc.  Note that to
reclaim the load space is going to be more machine-dependent work.
Implement a relocate hook for machdep.c to call so that the physical
addresses get converted to the equivalent KVM addresses.

25 years agoFully implement KLD and preloading.
Peter Wemm [Fri, 9 Oct 1998 23:55:31 +0000 (23:55 +0000)]
Fully implement KLD and preloading.
- seperate unload for preloaded linker objects.
- Don't build a kernel object if running as an a.out kernel.
- extract the real kernel name rather than hardwiring "kernel" for kldstat.
  (sysctl kern.bootfile getst the full name via bootinfo)
- use real addresses on the kernel "module" rather than fictitious ones.
- preloaded module support
- search module path for file modules.
- symbols are checked to see if they are in the right containing file
  before using their indexes into string tables.  This is to help ddb
  since it only supplies a pointer to an opaque symbol and there is no
  telling which file/object/module/whatever it came from.

25 years agoUpdates for KLD backends.
Peter Wemm [Fri, 9 Oct 1998 23:49:28 +0000 (23:49 +0000)]
Updates for KLD backends.
 - symbol_values checks that the symbol is indeed belonging to the
   correct symbol and string table pairs before looking up. (since there
   could be many pairs, and KLD/DDB need to find out).
 - different ops for files versus preload modules - the unload mechanism
   is different.  (a preloaded module has to be deleted on unload since
   the in-core image is tainted by relocation and variables used)
 - Do not build an a.out kernel module if we're running on an elf
   kernel. :-)  Note that it should theoretically be possible to
   mix a.out and elf KLD modules providing -mno-underscores was used
   to compile it, or some other symbol conversion takes place.
 - Support preload modules (even though /boot/loader doesn't yet)
 - Search the module path when loading files.

25 years agoImplement merging SYSINIT's from preloaded KLD modules. This means we
Peter Wemm [Fri, 9 Oct 1998 23:42:47 +0000 (23:42 +0000)]
Implement merging SYSINIT's from preloaded KLD modules.  This means we
check off SYSINIT entries as they are run, and when more arrive, we re-sort
and restart (skipping the already-run entries).
This can *only* be done after KMEM (and malloc) is up and running - this is
fine because KLD is the only consumer of this and it's done after that.
The nice thing about this is that the SYSINIT's within preloaded KLD modules
are executed in their natural order.  It should be possible to register
devices for the probes which follow, etc.  (soon.. several key things
prevent this, such as use of linker sets for things like pci devices).

25 years agoMODINFO_ADDR has real addresses now, remove the manual relocation based
Peter Wemm [Fri, 9 Oct 1998 23:37:37 +0000 (23:37 +0000)]
MODINFO_ADDR has real addresses now, remove the manual relocation based
on cpu type.

25 years agoRelocate the preload module info from machdep specifically rather than
Peter Wemm [Fri, 9 Oct 1998 23:36:26 +0000 (23:36 +0000)]
Relocate the preload module info from machdep specifically rather than
trying to do it in locore.  We also walk through the module table
and relocate any MODINFO_ADDR pointers so that they become KVM relative
rather than physical addresses.  This means that hacks for adding
0xf0000000 in places like MFS go away.

25 years agoCall some helper routines to be supplied by kern_linker.c in order to
Peter Wemm [Fri, 9 Oct 1998 23:34:09 +0000 (23:34 +0000)]
Call some helper routines to be supplied by kern_linker.c in order to
get to all the symbol tables for all modules, not just the core kernel
symbol table.  Yes, DDB can see KLD module symbols with this, both by
lookup and in tracebacks.  No more references to _end from tracebacks
within an LKM. :-)

25 years agoAt the moment, the alpha tentatively uses the ddb elf code rather than KLD,
Peter Wemm [Fri, 9 Oct 1998 23:32:03 +0000 (23:32 +0000)]
At the moment, the alpha tentatively uses the ddb elf code rather than KLD,
because the alpha boot loader hasn't been converted yet, and because
it needs the full symbol tables with local symbols in order to make sense
of stack tracebacks.  KLD will implement this (using full sybmol table
rather than the globals only) shortly.

25 years agoTurn off the new /sys/boot stuff (except boot0) unless OBJFORMAT
Robert Nordier [Fri, 9 Oct 1998 23:30:16 +0000 (23:30 +0000)]
Turn off the new /sys/boot stuff (except boot0) unless OBJFORMAT
is elf.  (The BTX client must be ELF, though it is packaged as a.out
for compatibility.)

25 years agoWe don't compile this on an elf kernel (and explicitly not on alpha in
Peter Wemm [Fri, 9 Oct 1998 23:29:44 +0000 (23:29 +0000)]
We don't compile this on an elf kernel (and explicitly not on alpha in
case it's possible to compile in something like ECOFF)
The three db_xxx.c symbol interfaces are "standard" because config isn't
flexible enough without forcing the user to know about it.

25 years agoRemove DDB_ELF_SYMBOLS and DDB_KLD_SYMBOLS options. KLD will be the sole
Peter Wemm [Fri, 9 Oct 1998 23:26:26 +0000 (23:26 +0000)]
Remove DDB_ELF_SYMBOLS and DDB_KLD_SYMBOLS options.  KLD will be the sole
interface to symbols in an elf kernel.

25 years agoAttempt to at least align MODINFO_* blocks. IMHO, decoding this stuff
Peter Wemm [Fri, 9 Oct 1998 23:24:55 +0000 (23:24 +0000)]
Attempt to at least align MODINFO_* blocks.  IMHO, decoding this stuff
should be MD code since one day we'll have to recover pages from deleted
preload data.  MI code can't be expected to know how to deal with pmap
internals, assuming it gets done via pmap that is. :-)

25 years agoTurn symbol table info passing back on, although it's probably not
Peter Wemm [Fri, 9 Oct 1998 23:22:30 +0000 (23:22 +0000)]
Turn symbol table info passing back on, although it's probably not
particularly useful in this form.  KLD can deal with it much better.
setenv kernelname moved earlier.

25 years agoRemove kernelname setenv, the common code does it.
Peter Wemm [Fri, 9 Oct 1998 23:20:47 +0000 (23:20 +0000)]
Remove kernelname setenv, the common code does it.
Use the metadata for symtab pointers, mainly to keep it common with
elf_freebsd.c.

25 years agoImplement preloading for elf modules
Peter Wemm [Fri, 9 Oct 1998 23:18:43 +0000 (23:18 +0000)]
Implement preloading for elf modules
 - get dependency info from PT_DYNAMIC's DT_NEEDED tags.
 - store MODINFOMD_DYNAMIC for the kernel's later use
setenv kernelname when we have it
Fix firstaddr/lastaddr calculation (duh! :-)
Explicitly skip string table with section names in it.

25 years agoPreload support for a.out KLD not implemented yet. (almost)
Peter Wemm [Fri, 9 Oct 1998 23:15:39 +0000 (23:15 +0000)]
Preload support for a.out KLD not implemented yet. (almost)
KLD modules are *not* PIC.  (Shared libs are pic to avoid relocations
causing copy-on-write, that's irrelevant here).
setenv kernelname when we load it.
Use MODINFO_SSYM/ESYM for each symbol section when (if) there are
more than one being loaded.

25 years agoUse a consistant module search path (same as kernel will be).
Peter Wemm [Fri, 9 Oct 1998 23:12:34 +0000 (23:12 +0000)]
Use a consistant module search path (same as kernel will be).
Use new dependency mechanism.

25 years agoloader <-> kernel preload module interface constants moved to common file.
Peter Wemm [Fri, 9 Oct 1998 23:11:05 +0000 (23:11 +0000)]
loader <-> kernel preload module interface constants moved to common file.
Remove Mike's explicit data structures for dependency info.  This is
done via DT_NEEDED etc in the dynamic section for now.  This may need
to be revisited later on.

25 years agoelf_machdep.c and rindex.c are now standard
Peter Wemm [Fri, 9 Oct 1998 23:08:14 +0000 (23:08 +0000)]
elf_machdep.c and rindex.c are now standard

25 years agoDebug typo fixes.
Peter Wemm [Fri, 9 Oct 1998 23:07:27 +0000 (23:07 +0000)]
Debug typo fixes.
Change symbol_values return type to int, on the chance that we've
been given the wrong symbol table.  Symbols have a string index that
must be relative to the correct string table.
Add prototypes for better kld support for ddb.

25 years agoA hook for storing the origin of a module, this is important when
Peter Wemm [Fri, 9 Oct 1998 23:05:45 +0000 (23:05 +0000)]
A hook for storing the origin of a module, this is important when
it's being registered in sorted sysinit sequence and not in load order.

25 years agoAdd SI_SUB_KLD
Peter Wemm [Fri, 9 Oct 1998 23:03:27 +0000 (23:03 +0000)]
Add SI_SUB_KLD
First part of support for merging SYSINIT sets.

This, and the following KLD commits have been OK'ed by jkh and msmith
based on my assertion that it works here (barring merge errors :-).

25 years agoClean up the boot targets a bit.
Jordan K. Hubbard [Fri, 9 Oct 1998 22:22:34 +0000 (22:22 +0000)]
Clean up the boot targets a bit.
Suggested by: jdp

25 years agoCall dpt_intr from our timeout routine to clear any pending commands before
Justin T. Gibbs [Fri, 9 Oct 1998 21:42:19 +0000 (21:42 +0000)]
Call dpt_intr from our timeout routine to clear any pending commands before
performing actual timeout processing.

Modify a few printf statements.

Submitted by: Simon Shapiro <shimon@simon-shapiro.org>

25 years agoOnly pull 16 bits of residual information from completing queues. This
Justin T. Gibbs [Fri, 9 Oct 1998 21:40:50 +0000 (21:40 +0000)]
Only pull 16 bits of residual information from completing queues.  This
is a work-around from an LRAM access bug on the 940UA.  In a future
microcode revision, the high 16bits of residual information will be moved
to a safe location and we'll return to 32bit residuals.  Since we only
allow 64KB I/O, 16bits is enough.

25 years agoHook up ahapoll so that dumps, synchronize cache commands or any other
Justin T. Gibbs [Fri, 9 Oct 1998 21:38:36 +0000 (21:38 +0000)]
Hook up ahapoll so that dumps, synchronize cache commands or any other
command that comes in through xpt_polled_action works correctly.

25 years agoDon't try to initialise the environment out of a sysinit, it's handled
Mike Smith [Fri, 9 Oct 1998 21:21:34 +0000 (21:21 +0000)]
Don't try to initialise the environment out of a sysinit, it's handled
in MD code instead.

25 years agoNull commit.. CVS aborted on freefall last time (reaonly file).
Peter Wemm [Fri, 9 Oct 1998 20:38:03 +0000 (20:38 +0000)]
Null commit.. CVS aborted on freefall last time (reaonly file).

An elf_reloc() function for the i386.  Based on alpha/alpha/elf_machdep.c
and rtld-elf/i386/reloc.c.

25 years agoAn elf_reloc() function for the i386. Based on alpha/alpha/elf_machdep.c
Peter Wemm [Fri, 9 Oct 1998 20:35:45 +0000 (20:35 +0000)]
An elf_reloc() function for the i386.  Based on alpha/alpha/elf_machdep.c
and rtld-elf/i386/reloc.c.

25 years agoAdd JKH's auth.conf parser to turn on/off Kerberos in userland
Mark Murray [Fri, 9 Oct 1998 20:14:48 +0000 (20:14 +0000)]
Add JKH's auth.conf parser to turn on/off Kerberos in userland

25 years agoFix some bugs in pthread scheduler:
Dmitrij Tejblum [Fri, 9 Oct 1998 19:01:30 +0000 (19:01 +0000)]
Fix some bugs in pthread scheduler:
  make pthread_yield() more reliable,
  threads always (I hope) preempted at least every 0.1 sec, as intended.

PR: bin/7744
Submitted by: "Richard Seaman, Jr." <dick@tar.com>

25 years agoPass termination information via softc flags to the core driver. The
Justin T. Gibbs [Fri, 9 Oct 1998 17:42:28 +0000 (17:42 +0000)]
Pass termination information via softc flags to the core driver.  The
EISA probe missed out on this change in attach->core API.

25 years agoWhen 'using defaults' ensure that termination is enabled.
Justin T. Gibbs [Fri, 9 Oct 1998 17:41:39 +0000 (17:41 +0000)]
When 'using defaults' ensure that termination is enabled.

Removed some unused code.

25 years agoGet this building as a.out or ELF.
Robert Nordier [Fri, 9 Oct 1998 17:19:51 +0000 (17:19 +0000)]
Get this building as a.out or ELF.

With thanks to: jdp

25 years agoHand me the pointy hat, and make it big.
Dag-Erling Smørgrav [Fri, 9 Oct 1998 17:11:14 +0000 (17:11 +0000)]
Hand me the pointy hat, and make it big.

25 years agoPut some #ifdef's around debugging code.
Andrzej Bialecki [Fri, 9 Oct 1998 12:42:56 +0000 (12:42 +0000)]
Put some #ifdef's around debugging code.

25 years agoFix for bad calculation of netmask bits. I'm amazed it worked thus far...
Andrzej Bialecki [Fri, 9 Oct 1998 12:42:20 +0000 (12:42 +0000)]
Fix for bad calculation of netmask bits. I'm amazed it worked thus far...

25 years agoSync with sys/i386/i386/machdep.c revision 1.312.
KATO Takenori [Fri, 9 Oct 1998 12:36:25 +0000 (12:36 +0000)]
Sync with sys/i386/i386/machdep.c revision 1.312.

25 years agofix a bug which could lock up a transmitter.
Kenjiro Cho [Fri, 9 Oct 1998 11:48:22 +0000 (11:48 +0000)]
fix a bug which could lock up a transmitter.
don't use the entire buffer space.  if WRTX becomes equal
to RDTX, the transmitter stops assuming the buffer is empty.

25 years agoNow take stdio.h out of files that don't require it.
Jordan K. Hubbard [Fri, 9 Oct 1998 11:24:20 +0000 (11:24 +0000)]
Now take stdio.h out of files that don't require it.

25 years agoFix damaged comment.
Jordan K. Hubbard [Fri, 9 Oct 1998 11:03:46 +0000 (11:03 +0000)]
Fix damaged comment.

25 years agofread() returns 0 on eof or error, not EOF. This fixes the following
Dag-Erling Smørgrav [Fri, 9 Oct 1998 10:33:46 +0000 (10:33 +0000)]
fread() returns 0 on eof or error, not EOF. This fixes the following
bug:

  "head -c <n>" never exit and loops forever (until it is killed),
  if the input stream has fewer bytes than specified (n).

PR: bin/8225
Submitted-by: FUJIMOTO Kensaku <fujimoto@oscar.elec.waseda.ac.jp>

25 years agoAdjust documentation to note the 3 boot images available now.
Jordan K. Hubbard [Fri, 9 Oct 1998 09:18:49 +0000 (09:18 +0000)]
Adjust documentation to note the 3 boot images available now.

25 years agoMove auth.conf path in here. Doesn't affect published interface.
Jordan K. Hubbard [Fri, 9 Oct 1998 07:34:47 +0000 (07:34 +0000)]
Move auth.conf path in here.  Doesn't affect published interface.

25 years agoUpdate docs to match interface change.
Jordan K. Hubbard [Fri, 9 Oct 1998 07:33:58 +0000 (07:33 +0000)]
Update docs to match interface change.

25 years agoo move path in libutil.h to paths.h
Jordan K. Hubbard [Fri, 9 Oct 1998 07:32:38 +0000 (07:32 +0000)]
o move path in libutil.h to paths.h
o make property_read() take a fd instead to avoid stdio.h mess
o update auth to new interface.

25 years agoTake the path spec back out.
Jordan K. Hubbard [Fri, 9 Oct 1998 07:28:14 +0000 (07:28 +0000)]
Take the path spec back out.

25 years agoRemove some debugging code.
Mike Smith [Fri, 9 Oct 1998 07:11:19 +0000 (07:11 +0000)]
Remove some debugging code.
Do a much better job of DWIM with partial device specifications.
Fix the module metadata build process, which was completely broken.
Use a larger read buffer when copying large objects in; this
improves performance marginally and will avoid flushning any small caches
we might choose to implement.

25 years agoSync the MODINFO constants with <sys/linker.h>
Mike Smith [Fri, 9 Oct 1998 07:09:22 +0000 (07:09 +0000)]
Sync the MODINFO constants with <sys/linker.h>
Remove debugging in command_read().
Correctly strip leading controls on script commands.
Make 'ls' more DWIM in regard to pathnames.  We can still do better.

25 years agoSmall refinements to boot script.
Jordan K. Hubbard [Fri, 9 Oct 1998 07:07:55 +0000 (07:07 +0000)]
Small refinements to boot script.

25 years agoMODINFO_NAME can't be 0, that's the end of the metadata area.
Mike Smith [Fri, 9 Oct 1998 07:06:43 +0000 (07:06 +0000)]
MODINFO_NAME can't be 0, that's the end of the metadata area.

25 years agoUse KJH's auth.conf parser to turn on/off Kerberos in userland.
Mark Murray [Fri, 9 Oct 1998 06:47:57 +0000 (06:47 +0000)]
Use KJH's auth.conf parser to turn on/off Kerberos in userland.

25 years agoUse KJH's auth.conf parser to turn on/off Kerberos in userland.
Mark Murray [Fri, 9 Oct 1998 06:38:33 +0000 (06:38 +0000)]
Use KJH's auth.conf parser to turn on/off Kerberos in userland.

25 years agoAdd JKH's auth.conf-file parser to turn Kerberos on/off in userland.
Mark Murray [Fri, 9 Oct 1998 06:31:45 +0000 (06:31 +0000)]
Add JKH's auth.conf-file parser to turn Kerberos on/off in userland.

25 years agoInitialize boot configuration files on floppy correctly. We now
Jordan K. Hubbard [Fri, 9 Oct 1998 06:28:28 +0000 (06:28 +0000)]
Initialize boot configuration files on floppy correctly.  We now
support 3 kinds of floppies:  boot+mfs (standard), boot, mfs.

25 years agoAdd some evil temporary phys-to-kern translation for mfs.
Jordan K. Hubbard [Fri, 9 Oct 1998 06:21:12 +0000 (06:21 +0000)]
Add some evil temporary phys-to-kern translation for mfs.

25 years agoAdd support for yet another "cyclone" board, with PCI device id 0x905A.
Bill Paul [Fri, 9 Oct 1998 03:59:24 +0000 (03:59 +0000)]
Add support for yet another "cyclone" board, with PCI device id 0x905A.
This is a 100BaseFX board with SC fiber media connectors. I don't actually
have one of these but I've been told it works with the xl driver.

Submitted by: Jason Wright from the openbsd group

25 years agoMissing defines for the kernel environment and module metadata lookup
Mike Smith [Fri, 9 Oct 1998 01:44:09 +0000 (01:44 +0000)]
Missing defines for the kernel environment and module metadata lookup
functions

25 years agoDon't try to print out the signal number that caused core to be
John Polstra [Fri, 9 Oct 1998 01:43:08 +0000 (01:43 +0000)]
Don't try to print out the signal number that caused core to be
dumped.  It is not in a fixed location in the file, so at best,
garbage comes out.  I don't think this code was ever valid.

25 years agoinclude proper header for Mike's new stuff.
Jordan K. Hubbard [Fri, 9 Oct 1998 01:40:56 +0000 (01:40 +0000)]
include proper header for Mike's new stuff.

25 years agoAdd new variable NO_FILTER_SHLIBS -- it will disable ELF <-> a.out shlib
Satoshi Asami [Fri, 9 Oct 1998 01:27:21 +0000 (01:27 +0000)]
Add new variable NO_FILTER_SHLIBS -- it will disable ELF <-> a.out shlib
name conversion.  Use it for binary ports that come with its own private
shlib dirs, ports that install linux compatibility libraries (thus following
their naming conventions and not ours), etc.

Reviewed by: Martin Cracauer <cracauer@FreeBSD.ORG>

25 years agoAll these have to include stdio.h now.
Jordan K. Hubbard [Fri, 9 Oct 1998 00:39:09 +0000 (00:39 +0000)]
All these have to include stdio.h now.

25 years agoKernel environment access, preloaded module lookup.
Mike Smith [Fri, 9 Oct 1998 00:31:29 +0000 (00:31 +0000)]
Kernel environment access, preloaded module lookup.

25 years agoInitialise kernel environment and module metadata pointers.
Mike Smith [Fri, 9 Oct 1998 00:31:06 +0000 (00:31 +0000)]
Initialise kernel environment and module metadata pointers.

25 years agoNew files for kernel environment and module metadata interfaces.
Mike Smith [Fri, 9 Oct 1998 00:30:50 +0000 (00:30 +0000)]
New files for kernel environment and module metadata interfaces.

25 years agoFix a panic on SMP systems, caused by sleeping while holding a
John Polstra [Fri, 9 Oct 1998 00:24:49 +0000 (00:24 +0000)]
Fix a panic on SMP systems, caused by sleeping while holding a
simple-lock.

The reviewer raises the following caveat: "I believe these changes
open a non-critical race condition when adding memory to the pool
for the zone. I think what will happen is that you could have two
threads that are simultaneously adding additional memory when the
pool runs out. This appears to not be a problem, however, since
the re-aquisition of the lock will protect the list pointers."
The submitter agrees that the race is non-critical, and points out
that it already existed for the non-SMP case.  He suggests that
perhaps a sleep lock (using the lock manager) should be used to
close that race.  This might be worth revisiting after 3.0 is
released.

Reviewed by: dg (David Greenman)
Submitted by: tegge (Tor Egge)

25 years agoUse lstat() rather than stat in determining whether a file exists,
Jordan K. Hubbard [Fri, 9 Oct 1998 00:01:16 +0000 (00:01 +0000)]
Use lstat() rather than stat in determining whether a file exists,
otherwise we'll miss the "symlink exists but points nowhere" case.
Submitted by: asami and/or possibly <ru@ucb.crimea.ua>

25 years agoAllow the module area to be used in order to find the MFS image
Jordan K. Hubbard [Thu, 8 Oct 1998 23:34:44 +0000 (23:34 +0000)]
Allow the module area to be used in order to find the MFS image
(in addition to allowing it to be compiled in) and stop overloading
the MFS_ROOT variable to store size information.

25 years agoThis thing has its own puts function, so use it.
Jordan K. Hubbard [Thu, 8 Oct 1998 23:14:02 +0000 (23:14 +0000)]
This thing has its own puts function, so use it.
Submitted by: Matthew Jacob <mjacob@nas.nasa.gov>

25 years agoremove stdio.h include; I forgot Bruce's cardinal rule that header files
Jordan K. Hubbard [Thu, 8 Oct 1998 23:10:41 +0000 (23:10 +0000)]
remove stdio.h include; I forgot Bruce's cardinal rule that header files
shouldn't include other ones (which, unfortunately, is also a hellish
rule since he broke interfaces like sysctl this way by requiring undocumented
header files to be included just in order to be able to use them now - SIGH!).

25 years agoFix up the kernel environment and module data pointers in the bootinfo if
Mike Smith [Thu, 8 Oct 1998 21:03:41 +0000 (21:03 +0000)]
Fix up the kernel environment and module data pointers in the bootinfo if
they are present.
If we are told where the end of the loaded kernel image is, believe it.

25 years agoMake that annoying "device busy" message dependent on DEBUG (using the
Dag-Erling Smørgrav [Thu, 8 Oct 1998 20:55:54 +0000 (20:55 +0000)]
Make that annoying "device busy" message dependent on DEBUG (using the
DEB macro). There are probably quite a few other messages that warrant
a similar treatment, and many more that should be converted to plain
log messages (e.g. "WARNING: wrintr but write DMA inactive!"). Now
that I think of it, same goes for the CAM code (e.g. the famed "tagged
openings" message)

25 years agoFix a memory leak in rpc.rstatd that shows up when it's run in standalone
Kenneth D. Merry [Thu, 8 Oct 1998 19:59:40 +0000 (19:59 +0000)]
Fix a memory leak in rpc.rstatd that shows up when it's run in standalone
mode.  (i.e., not from inetd)

PR: bin/8212

25 years agoWhile I'm at it, remove the tickadj manpage from src/share/man0/man8,
Dag-Erling Smørgrav [Thu, 8 Oct 1998 18:52:17 +0000 (18:52 +0000)]
While I'm at it, remove the tickadj manpage from src/share/man0/man8,
though I'm afraid there's a lot more that needs fixing in this file,
judging by 'find /usr/src -name "*.8" -print'.

Spotted-by: glimpse -H /usr/src tickadj

25 years agoRemove all references to tickadj(8) from rc, rc.conf and rc.conf.5.
Dag-Erling Smørgrav [Thu, 8 Oct 1998 18:47:30 +0000 (18:47 +0000)]
Remove all references to tickadj(8) from rc, rc.conf and rc.conf.5.

Disable building tickadj(8) by removing util from SUBDIR in the xntpd
Makefile. Note that the sources are still there and tickadj can still
be built and installed by doing:

# cd /usr/src/usr.sbin/xntpd/util
# make all install

There are enough references to tickadj in e.g. the xntpd documentation
(not to mention the sysctl variables it uses etc.) that I don't feel
up to implementing the final solution right now.

Kinda-approved-by: phk

25 years agoDiscard previous commit.
KATO Takenori [Thu, 8 Oct 1998 17:20:58 +0000 (17:20 +0000)]
Discard previous commit.

25 years agoOops, discard my previous commits becase sumitted code is for RELENG_2_2.
KATO Takenori [Thu, 8 Oct 1998 17:16:41 +0000 (17:16 +0000)]
Oops, discard my previous commits becase sumitted code is for RELENG_2_2.

25 years agoAdd entry for EtherEZ98 and Access/PC N98C+ cards.
KATO Takenori [Thu, 8 Oct 1998 17:07:44 +0000 (17:07 +0000)]
Add entry for EtherEZ98 and Access/PC N98C+ cards.

Reviewed by: kato
Submitted by: chi@bd.mbn.or.jp (Chiharu Shibata)

25 years agoAdd ungermann-Bass Access/PC N98C+ support to fe driver (PC-98).
KATO Takenori [Thu, 8 Oct 1998 17:06:32 +0000 (17:06 +0000)]
Add ungermann-Bass Access/PC N98C+ support to fe driver (PC-98).

Reviewed by: kato
Submitted by: chi@bd.mbn.or.jp (Chiharu Shibata)

25 years agoAdd SMC EtherEZ98 support to ed driver (PC-98).
KATO Takenori [Thu, 8 Oct 1998 17:04:59 +0000 (17:04 +0000)]
Add SMC EtherEZ98 support to ed driver (PC-98).

Reviewed by: kato
Submitted by: chi@bd.mbn.or.jp (Chiharu Shibata)

25 years agoBIOS ROM base address is 0xe8000 on PC-98.
KATO Takenori [Thu, 8 Oct 1998 16:15:22 +0000 (16:15 +0000)]
BIOS ROM base address is 0xe8000 on PC-98.

25 years agoCorrectly update the tail pointer of the transmit queue in tl_start()
Bill Paul [Thu, 8 Oct 1998 15:45:36 +0000 (15:45 +0000)]
Correctly update the tail pointer of the transmit queue in tl_start()
(one-liner). I have yet to actually encounter any problems due to this
bug, but why take chances.

25 years agoSync with sys/i386/conf/GENERIC revision up to 1.123.
KATO Takenori [Thu, 8 Oct 1998 12:10:07 +0000 (12:10 +0000)]
Sync with sys/i386/conf/GENERIC revision up to 1.123.

25 years agoSync with sys/i386/i386/userconfig.c revision 1.111.
KATO Takenori [Thu, 8 Oct 1998 12:09:38 +0000 (12:09 +0000)]
Sync with sys/i386/i386/userconfig.c revision 1.111.

25 years agoFinally settle on a name for the mfsroot image.
Jordan K. Hubbard [Thu, 8 Oct 1998 10:18:55 +0000 (10:18 +0000)]
Finally settle on a name for the mfsroot image.

25 years agoOnly install boot blocks if !MFSKERN. Too big otherwise!
Jordan K. Hubbard [Thu, 8 Oct 1998 10:02:28 +0000 (10:02 +0000)]
Only install boot blocks if !MFSKERN.  Too big otherwise!

25 years agoQuiet many compiler warnings.
David E. O'Brien [Thu, 8 Oct 1998 09:56:10 +0000 (09:56 +0000)]
Quiet many compiler warnings.

25 years agoAdd boot bits to boot floppy.
Jordan K. Hubbard [Thu, 8 Oct 1998 09:53:32 +0000 (09:53 +0000)]
Add boot bits to boot floppy.

25 years agoMake both types of boot targets.
Jordan K. Hubbard [Thu, 8 Oct 1998 09:40:17 +0000 (09:40 +0000)]
Make both types of boot targets.

25 years agoFinal cleanup pass over boot floppy changes.
Jordan K. Hubbard [Thu, 8 Oct 1998 09:06:16 +0000 (09:06 +0000)]
Final cleanup pass over boot floppy changes.

25 years agoAvoid using dmesg to find devices, the buffer may not be big enough.
Poul-Henning Kamp [Thu, 8 Oct 1998 08:56:01 +0000 (08:56 +0000)]
Avoid using dmesg to find devices, the buffer may not be big enough.

Reviewed by: phk
Submitted by: Mike Spengler <mks@networkcs.com>

25 years agoRemove a gratuitous copy.
Jordan K. Hubbard [Thu, 8 Oct 1998 07:12:37 +0000 (07:12 +0000)]
Remove a gratuitous copy.

25 years agoAdd some rudimentary documentation for my new functions.
Jordan K. Hubbard [Thu, 8 Oct 1998 06:53:32 +0000 (06:53 +0000)]
Add some rudimentary documentation for my new functions.

25 years agoRemove debug output on disk change.
Søren Schmidt [Thu, 8 Oct 1998 06:41:44 +0000 (06:41 +0000)]
Remove debug output on disk change.

25 years agoAdd auth.conf. JKH Added the code to understand this to libutil, and
Mark Murray [Thu, 8 Oct 1998 06:12:04 +0000 (06:12 +0000)]
Add auth.conf. JKH Added the code to understand this to libutil, and
I will be following up with commits to use it in KerberosIV userland.

25 years agoAdd the quirk entry framework to handle disabling the synchronize cache
Kenneth D. Merry [Thu, 8 Oct 1998 05:46:38 +0000 (05:46 +0000)]
Add the quirk entry framework to handle disabling the synchronize cache
command on drives that don't like it.  Right now, there's just a bogus
quirk entry in the table that doesn't do anything, but that should be
changed once we get actual inquiry data for drives that don't like the
synchronize cache command.

Also, add a shutdown hook that runs through all direct access peripherals
and runs a synchronize cache on them if they're still open, and if
synchronize cache isn't disabled via a quirk entry.

Add a synchronize cache call at the end of dadump() (again, conditionalized
on the quirk entry), so we can insure that the disk cache contents get
flushed to physical media after a dump.

Check the new quirk entry in daclose() to decide whether or not to
synchronize the cache for a disk at final close.

Reviewed by: gibbs

25 years agoCorrect a build error that got past my build test somehow.
Jordan K. Hubbard [Thu, 8 Oct 1998 01:56:49 +0000 (01:56 +0000)]
Correct a build error that got past my build test somehow.

25 years agoCheck the timeval passed to BIOCSRTIMEOUT with itimerfix. Use tvtohz()
Alexander Langer [Thu, 8 Oct 1998 00:32:08 +0000 (00:32 +0000)]
Check the timeval passed to BIOCSRTIMEOUT with itimerfix.  Use tvtohz()
to convert the timeval into a tick count.

Suggested by: bde
Reviewed by: bde

Handle hz > 1000 in BIOCGRTIMEOUT.

Pointed out by: bde
Reviewed by: bde
Obtained from: OpenBSD

25 years agoUnreverse a conditional and gzip the mfs image by default.
Jordan K. Hubbard [Wed, 7 Oct 1998 22:58:22 +0000 (22:58 +0000)]
Unreverse a conditional and gzip the mfs image by default.

25 years agoSupport 3 floppy images as standard now: boot kernel, mfs image and
Jordan K. Hubbard [Wed, 7 Oct 1998 22:54:04 +0000 (22:54 +0000)]
Support 3 floppy images as standard now:  boot kernel, mfs image and
kernel+mfs image.