]> CyberLeo.Net >> Repos - FreeBSD/FreeBSD.git/log
FreeBSD/FreeBSD.git
25 years agoDon't ignore weak symbols.
Dmitrij Tejblum [Sat, 3 Jul 1999 12:30:04 +0000 (12:30 +0000)]
Don't ignore weak symbols.

EGCS assign weak symbols to inline functions it couldn't inline (e.g. virtual
inline functions), template functions, etc. Omitting them result in quite bogus
profile.

Weak symbols created by __weak_reference are not really problem.

Caught by: Ilya Segalovich <iseg@comptek.ru>

25 years agoReverse the FTP_PASSIVE_MODE check, so it checks for "is not NO" rather
Dag-Erling Smørgrav [Sat, 3 Jul 1999 10:10:50 +0000 (10:10 +0000)]
Reverse the FTP_PASSIVE_MODE check, so it checks for "is not NO" rather
than "is YES".

25 years agoAdded copyright.
KATO Takenori [Sat, 3 Jul 1999 08:50:45 +0000 (08:50 +0000)]
Added copyright.

Pointed out by: yokota

25 years agoRemoved unused files.
KATO Takenori [Sat, 3 Jul 1999 08:45:16 +0000 (08:45 +0000)]
Removed unused files.

25 years agoMoved LCD controle routine for certain models of EPSON laptops into
KATO Takenori [Sat, 3 Jul 1999 08:44:32 +0000 (08:44 +0000)]
Moved LCD controle routine for certain models of EPSON laptops into
suitable place.

Submitted by: yokota

25 years agoSync with sys/i386/i386/machdep.c revision 1.349.
KATO Takenori [Sat, 3 Jul 1999 08:31:32 +0000 (08:31 +0000)]
Sync with sys/i386/i386/machdep.c revision 1.349.

25 years agoBe more informative and try to ask the user in some instances if we can't
Poul-Henning Kamp [Sat, 3 Jul 1999 08:24:00 +0000 (08:24 +0000)]
Be more informative and try to ask the user in some instances if we can't
figure out the root device.

25 years agoBail after 5 attempts to read very noisy signals.
Poul-Henning Kamp [Sat, 3 Jul 1999 08:23:00 +0000 (08:23 +0000)]
Bail after 5 attempts to read very noisy signals.

25 years agoWarn about drivers which take over other drivers cdevsw entries, but still
Poul-Henning Kamp [Sat, 3 Jul 1999 08:22:30 +0000 (08:22 +0000)]
Warn about drivers which take over other drivers cdevsw entries, but still
grant them squatters right.

25 years agoSync with sys/i386/conf/majors.i386 revision 1.78.
KATO Takenori [Sat, 3 Jul 1999 08:18:27 +0000 (08:18 +0000)]
Sync with sys/i386/conf/majors.i386 revision 1.78.

25 years agoSync with sys/i386/conf/Makefile.i386 revision 1.157.
KATO Takenori [Sat, 3 Jul 1999 08:17:40 +0000 (08:17 +0000)]
Sync with sys/i386/conf/Makefile.i386 revision 1.157.

25 years agoFixed regression of $Id$ to "from: $Header$" in previous commit.
Bruce Evans [Sat, 3 Jul 1999 06:48:02 +0000 (06:48 +0000)]
Fixed regression of $Id$ to "from: $Header$" in previous commit.
The `from:' was already corrupted by expansion of $Header$ in in
rev.1.1; rev.1.2 replaced it by $Id$.  Put $Id$ in the usual place
this time.

25 years agoAn SMP-specific change: Add the lock prefix to RMW operations
Alan Cox [Sat, 3 Jul 1999 06:33:48 +0000 (06:33 +0000)]
An SMP-specific change: Add the lock prefix to RMW operations
on ipending.

25 years agowcd -> acd
Michael Haro [Sat, 3 Jul 1999 05:42:38 +0000 (05:42 +0000)]
wcd -> acd

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

25 years agosys/buf.h needs to have included sys/systm.h for spl prototypes.
Brian Feldman [Sat, 3 Jul 1999 04:56:57 +0000 (04:56 +0000)]
sys/buf.h needs to have included sys/systm.h for spl prototypes.

25 years agoIt's really mP6.
Brian Feldman [Sat, 3 Jul 1999 01:56:58 +0000 (01:56 +0000)]
It's really mP6.
:)

25 years agoWell, don't try and probe 65535 luns- things just don't really work well
Matt Jacob [Sat, 3 Jul 1999 01:44:05 +0000 (01:44 +0000)]
Well, don't try and probe 65535 luns- things just don't really work well
when this happens. Limit to 16 luns for the 2100/2200 for now.

25 years agoBump maxusers to 32 to match the i386.
Mike Smith [Sat, 3 Jul 1999 01:35:51 +0000 (01:35 +0000)]
Bump maxusers to 32 to match the i386.
Remove the gzip pseudo-device, since the Alpha never did a.out.
Add the 'wb' driver since it is supported.

25 years agoAdd definitions for simplified config commands
Greg Lehey [Fri, 2 Jul 1999 23:57:45 +0000 (23:57 +0000)]
Add definitions for simplified config commands
Change definition of sflag

Reported-by: Johan Karlsson <k@numeri.campus.luth.se>
Doug <Doug@gorean.org>
green

World-broken-by: grog
Pointy-hat-passed-to: grog

25 years agoRemove pre-CAM code. Add in getenv_int calls for variables isp_mem_map,
Matt Jacob [Fri, 2 Jul 1999 23:18:03 +0000 (23:18 +0000)]
Remove pre-CAM code. Add in getenv_int calls for variables isp_mem_map,
isp_io_map, isp_no_fwload, isp_fwload, isp_no_nvram, isp_fcduplex
which are all bitmaps of isp instances that should or shouldn't
map memory space, I/O space, not load f/w, load f/w, ignore nvram,
not ignore nvarm, set full duplex mode. Also have an isp_seed value
that we can use to generate a pseudo seed for a synthetic WWN.
Other minor cosmetic cleanup. Add in support for the Qlogic ISP
2200. Very important change where we actually check now to see
whether we were successful in mapping request and response queues
(and fibre channel scratch space).

25 years agoRemove all pre-CAM code. Fix breakage for SCCLUN when it is in fact
Matt Jacob [Fri, 2 Jul 1999 23:12:58 +0000 (23:12 +0000)]
Remove all pre-CAM code. Fix breakage for SCCLUN when it is in fact
ISP2100_SCCLUN. Make changes for how ISPASYNC_PDB_CHANGE_COMPLETE
is now ISPASYNC_PDB_CHANGED. Add in ISPASYNC_FABRIC_DEV case.

25 years agoMerge defunct isp_freebsd_cam.h into this file. Do some appropriate
Matt Jacob [Fri, 2 Jul 1999 23:10:34 +0000 (23:10 +0000)]
Merge defunct isp_freebsd_cam.h into this file. Do some appropriate
changes like rolling minor revision levels and defines where we now
do default WWN seeding in the platform files.

25 years agoRoll revision levels. Add support for the Qlogic 2200 (warn about
Matt Jacob [Fri, 2 Jul 1999 23:06:38 +0000 (23:06 +0000)]
Roll revision levels. Add support for the Qlogic 2200 (warn about
not having SCSI_ISP_SCCLUN config defined if we don't have f/w for
the 2200- it's resident firmware uses SCCLUN (65535 luns)). Change
the way the default LoopID is gathered (it's now a platform specific
define so that some attempt at a synthetic WWN can be made in case
NVRAM isn't readable).

Change initialization of options a bit- don't use ADISC. Set
FullDuplex mode if config options tells us to do so. Do not use
FULL_LOGIN after LIP- it's the right thing to do but it causes too
much loop disruption (Loop Resets). Sanity check some default
values. Redo construction of port and node WWNs based upon what we
have- if we have 2 in the top nibble, we can have distinct port
and node WWNs. Clean up some SCCLUN related code that we obviously
had never compiled (:-(). Audit commands coming int ispscsicmd and
don't throw commands at Fibre devices that do not have Class 3
service parameters TARGET ROLE defined.

Clean up f/w initialization a bit. Add Fabric support (or at least
the first blush of it). Whew - way too much to describe here.
Basically, after a LIP, hang out until we see a Loop Up or a Port
DataBase Change async event, then see if we're on a Fabric
(GET_PORT_NAME of FL_PORT_ID). If we are, try and scan the fabric
controller for fabric devices using the GetAllNext SNS subcommand.
As we find devices, announce them to the outer layer. Try and do
some guard code for broken (Brocade) SNS servers (that get stuck
in loops- gotta maybe do this a different way using the GP_ID3 cmd
instead).  Then do a scan of the lower (local loop) ids using a
GET_PORT_NAME to see if the f/w has logged into anything at that
loop id. If so, then do a GET_PORT_DATABASE command.  Do this scan
into a local database. At this point we can say the loop is 'Ready'.
After this, we merge our local loop port database with our stored
port database- in a as yet to be really fully exercised fashion we
try and follow the logic of something having moved around. The
first time we see something at a Loop ID, we fix it, for the purpose
of this system instance, at that Loop ID. If things shift around
so it ends up somewhere else, we still keep it at this Loop ID (our
'Target') but use the new (moved) Loop ID when we actually throw
commands at it. Check for insane cases of different Loop IDs both
claiming to have the same WWN- if that happens, invalidate both.
Notify the outer layer of devices that have arrived and devices
that have gone away. *Finally*, when this is done, search the
softc's database of Fabric devices and perform logout/login actions.
The Qlogic f/w maintains logout/login for all local loop devices.
We have to maintain logout/login for fabric devices- total PITA.
Expect to see this area undergo more change over time.

25 years agoMention the Rise MP6, list merged items. There are probably a pile more
Mike Smith [Fri, 2 Jul 1999 23:04:28 +0000 (23:04 +0000)]
Mention the Rise MP6, list merged items.  There are probably a pile more
changes that need to be listed here.

25 years agoRoll revision levels. Move DEFAULT_LOOPID definition to platform files.
Matt Jacob [Fri, 2 Jul 1999 22:46:31 +0000 (22:46 +0000)]
Roll revision levels. Move DEFAULT_LOOPID definition to platform files.
Change some fcp parameter structures such that we can get the portid
(24 bit value), get both node and port WWN, know whether we're on a fabric
or not, note whether we've ever seen the loop up, and note the current
state of the loop.

Replace the isp_pdb_t structure in fcparams with a reduced cost structure
that maintains a static relationship to 'Target', but can have the actual
loop ID used change (in case, post LIP, we discover things have moved
around). This also retains portid and node/port WWNs.  This array gets
larger if we have fabric support compiled in.

Note special loop IDs that are invariate for this device- FL_PORT_ID
(0x7e) which tells us if there's a fabric controller present, FC_PORT_ID
and FC_SNS_ID (fabric controller port and fabric SNS server port). We don't
use the latter two for anything. IDs above FC_SNS_ID up through 255 are
available for mapping fabric devices to 'target' ids.

Add in a config define to set FC full duplex mode. Add in a define to
recognize the Qlogic 2200 boards. Add comments about ISPCTL commands.
Add and change some ISPASYNC enumes.

25 years agoEliminate some varargs abuse.
Jordan K. Hubbard [Fri, 2 Jul 1999 22:36:14 +0000 (22:36 +0000)]
Eliminate some varargs abuse.

25 years agoAdd in SNS and Fabric login/logout commands. Clean up ICBOPT defines.
Matt Jacob [Fri, 2 Jul 1999 22:31:11 +0000 (22:31 +0000)]
Add in SNS and Fabric login/logout commands. Clean up ICBOPT defines.
Remove INVALID_PDB_OPTIONS defines. Define generic SNS request and
response structures and the bare minimum GAN and GP3 subcommands.

25 years agoThis file is now obsolete.
Matt Jacob [Fri, 2 Jul 1999 22:26:10 +0000 (22:26 +0000)]
This file is now obsolete.

25 years agoRemove unused variable.
Bill Fumerola [Fri, 2 Jul 1999 20:46:03 +0000 (20:46 +0000)]
Remove unused variable.

Approved By: phk

25 years agoDocument the hw.physmem tunable. (There should be arch-specific tunable
Mike Smith [Fri, 2 Jul 1999 20:38:05 +0000 (20:38 +0000)]
Document the hw.physmem tunable.  (There should be arch-specific tunable
help sections.)

25 years agoLightly overhaul the memory sizing code again.
Mike Smith [Fri, 2 Jul 1999 20:33:32 +0000 (20:33 +0000)]
Lightly overhaul the memory sizing code again.

 - The kernel environment variable 'hw.physmem' can be used to set the
   amount of physical memory space, based at 0, that FreeBSD will use.
   Any memory detected over this limit is ignored.  Documentation for
   this is available under 'help set tunables' in the loader.

 - In the case where system memory size can't be accurately determined,
   hw.physmem is used as a best-guess memory size, but speculative
   probing will be used to determine actual memory size if any of the
   guesses or hints are 16M or more.

 - If RB_VERBOSE, we list the memory regions as we test them.

 - The compile-time option MAXMEM supplies a default value for
   'hw.physmem'.

25 years ago* The OBSTACK_CHUNK_SIZE #define was in the "Code generation parameters"
David E. O'Brien [Fri, 2 Jul 1999 19:26:45 +0000 (19:26 +0000)]
* The OBSTACK_CHUNK_SIZE #define was in the "Code generation parameters"
  section rather than the "Miscellaneous parameters" where it belongs.

* Clarify what we are locating using the -m rule.

Submitted by: bde

25 years agoThe GNU standard about putting 2 spaces after sentences in comments
David E. O'Brien [Fri, 2 Jul 1999 19:20:44 +0000 (19:20 +0000)]
The GNU standard about putting 2 spaces after sentences in comments
was not consistantly followed.

Submitted by: bde

25 years agoUpdate these files to match current reality.
Jordan K. Hubbard [Fri, 2 Jul 1999 17:11:58 +0000 (17:11 +0000)]
Update these files to match current reality.

25 years agoroll firmware for FC
Matt Jacob [Fri, 2 Jul 1999 17:07:39 +0000 (17:07 +0000)]
roll firmware for FC

25 years agoAdd SSL versions of imap, pop3 and telnet (imaps, pop3s and telnets)
Nick Sayer [Fri, 2 Jul 1999 16:55:04 +0000 (16:55 +0000)]
Add SSL versions of imap, pop3 and telnet (imaps, pop3s and telnets)

25 years agoMake sure that stat(2) and friends always return a valid st_dev field.
Poul-Henning Kamp [Fri, 2 Jul 1999 16:29:47 +0000 (16:29 +0000)]
Make sure that stat(2) and friends always return a valid st_dev field.

Pseudo-FS need not fill in the va_fsid anymore, the syscall code
will use the first half of the fsid, which now looks like a udev_t
with major 255.

25 years agoAllow service alias names from /etc/services to be used when specifying
Sheldon Hearn [Fri, 2 Jul 1999 16:21:13 +0000 (16:21 +0000)]
Allow service alias names from /etc/services to be used when specifying
internal services in inetd.conf .

The inetd(8) manpage used to say that the official name of a service
_must_ be used, yet inetd itself was hardcoded to used a service alias for
the auth service, namely ident!

Rather than change inetd.conf and break existing configurations on next
upgrade, we now allow service aliases as well as official names. This
allows the software to work as expected and still support existing
configurations.

This should not breaking existing wrapped configurations either and the
inetd(8) manpage already states that it is the service name specified in
inetd.conf that is used for calls to hosts_access(3).

PR: 11796
Reported by: Alex Charalabidis <alex@wnm.net>
Approved by: des

25 years agoClarify that the services name, as specified in inetd.conf, for an
Sheldon Hearn [Fri, 2 Jul 1999 15:58:32 +0000 (15:58 +0000)]
Clarify that the services name, as specified in inetd.conf, for an
internal service should be used as the daemon name when constructing
hosts_access(5) rules.

25 years agoAdd braces to avoid ambigious else.
Bill Fumerola [Fri, 2 Jul 1999 15:26:53 +0000 (15:26 +0000)]
Add braces to avoid ambigious else.

25 years agoo use getprotobynumber(3) instead of hardcoded protocol names.
Ruslan Ermilov [Fri, 2 Jul 1999 13:25:49 +0000 (13:25 +0000)]
o use getprotobynumber(3) instead of hardcoded protocol names.
  Divert(4) sockets will now appear as "internet raw divert"
  instead of "internet raw 254".

o -Wall tweak.

Reviewed by: bde

25 years agoBack out previous commit. Allowing `-' as a pseudonym for /dev/stdin is
Sheldon Hearn [Fri, 2 Jul 1999 12:31:31 +0000 (12:31 +0000)]
Back out previous commit. Allowing `-' as a pseudonym for /dev/stdin is
legacy behaviour inherited from systems that don't have /dev/stdin.

Requested by: bde

25 years agoActually fixed ambiguous else. The previous revision had no effect.
Bruce Evans [Fri, 2 Jul 1999 11:20:59 +0000 (11:20 +0000)]
Actually fixed ambiguous else.  The previous revision had no effect.

25 years ago"transfered" -> "transferred"
Brian Feldman [Fri, 2 Jul 1999 08:57:06 +0000 (08:57 +0000)]
"transfered" -> "transferred"

Reviewed by: dict(1)

25 years agoRemove the volume, plex, sd and drive commands which allowed the user
Greg Lehey [Fri, 2 Jul 1999 08:01:12 +0000 (08:01 +0000)]
Remove the volume, plex, sd and drive commands which allowed the user
to bypass the create command.

vinum_stop:
  Allow -f option.

Add commands for simplified configuration:
  concat - create a volume with one concatenated plex
  stripe - create a volume with one striped plex
  mirror - create a volume with two concatenated or striped plexes

Egged-on-by:   jkh

25 years agoChange the name of the s option to indicate that it wears more than
Greg Lehey [Fri, 2 Jul 1999 08:00:31 +0000 (08:00 +0000)]
Change the name of the s option to indicate that it wears more than
one hat.

Change formats as the result of changes in drive structure.

25 years agoChange the name of the s option to indicate that it wears more than
Greg Lehey [Fri, 2 Jul 1999 07:59:45 +0000 (07:59 +0000)]
Change the name of the s option to indicate that it wears more than
one hat.

Add -n option for specifying names of volumes created with the
simplified commands.

Add function find_drive_by_devname to help the simplified
configuration commands.

25 years agoDocument simplified configuration commands.
Greg Lehey [Fri, 2 Jul 1999 07:58:54 +0000 (07:58 +0000)]
Document simplified configuration commands.

25 years agoAdd keywords 'mirror' and 'stripe' when not in kernel mode.
Greg Lehey [Fri, 2 Jul 1999 07:58:09 +0000 (07:58 +0000)]
Add keywords 'mirror' and 'stripe' when not in kernel mode.

25 years agoAllow to unconditionally up subdisks with the -f option.
Greg Lehey [Fri, 2 Jul 1999 07:57:28 +0000 (07:57 +0000)]
Allow to unconditionally up subdisks with the -f option.

Change returns for failed state changes from EINVAL to EBUSY.

Suggested-by: Bernd Walter <ticso@cicely.de>
25 years agoChange the type of the freelist sectors variable to u_int64_t in
Greg Lehey [Fri, 2 Jul 1999 07:56:47 +0000 (07:56 +0000)]
Change the type of the freelist sectors variable to u_int64_t in
order to ensure that initializers will be of correct type.

25 years agoCorrect some ugly formatting. Remember to initialize the alignment tag.
Matt Jacob [Fri, 2 Jul 1999 05:12:11 +0000 (05:12 +0000)]
Correct some ugly formatting. Remember to initialize the alignment tag.
Honor and pass a callers request to contigalloc if they had a non-zero
alignment constraint.

25 years agoFix last delta which reinitialized nextpaddr to zero each time through the
Matt Jacob [Fri, 2 Jul 1999 05:05:51 +0000 (05:05 +0000)]
Fix last delta which reinitialized nextpaddr to zero each time through the
loop it was supposed to be in. Correct some ugly formatting. Remember to
initialize the alignment tag. Honor and pass a callers request to contigalloc
if they had a non-zero alignment constraint.

25 years agoFix time of last dump handling.
Jordan K. Hubbard [Fri, 2 Jul 1999 04:44:43 +0000 (04:44 +0000)]
Fix time of last dump handling.

Submittted by: Richard Wiwatowski <rjwiwat@ozemail.com.au>

25 years agoSync with GENERIC.. (Gee, this is a useful file..)
Peter Wemm [Fri, 2 Jul 1999 04:36:48 +0000 (04:36 +0000)]
Sync with GENERIC..  (Gee, this is a useful file..)

25 years agoZap totally the npx0 memory size override. It only worked if statically
Peter Wemm [Fri, 2 Jul 1999 04:33:05 +0000 (04:33 +0000)]
Zap totally the npx0 memory size override.  It only worked if statically
specified in the kernel config file - but setting options MAXMEM works
exactly the same.  Userconfig overrides of this have not worked for
ages.

Also, change the getenv for the loader override to hw.physmem based on a
prior suggestion from Mike Smith.  I think he still wants to change this
some, but this shouldn't get in his way.  This is a forced setting of
the memory size, not a "cap".  We probably should have a plain 'maxmem'
variable as well which does do a cap, without loosing the bios memory
configuration data.

25 years agoChange the cast in pci_map_port() from u_short * to pci_port_t * so it
Peter Wemm [Fri, 2 Jul 1999 04:17:16 +0000 (04:17 +0000)]
Change the cast in pci_map_port() from u_short * to pci_port_t * so it
compiles cleanly on the Alpha.  (On the alpha, the port type is an int,
not a short).
Cast a couple of pointers to ints via 'uintptr_t' rather than 'unsigned
int' since uintptr_t is long (64 bit) on Alpha, as are pointers.

25 years agoReplicate changes to i386 to do tags. I doesn't work yet 'coz ctags dumps
Matt Jacob [Fri, 2 Jul 1999 04:16:57 +0000 (04:16 +0000)]
Replicate changes to i386 to do tags. I doesn't work yet 'coz ctags dumps
core, but when that is fixed it ought to.

25 years agoWorking kernel tags!
Jordan K. Hubbard [Fri, 2 Jul 1999 04:00:01 +0000 (04:00 +0000)]
Working kernel tags!

Submitted by: Craig Leres <leres@ee.lbl.gov>
PR: 2806

25 years ago127 fire Firewire driver <ikob@koganei.wide.ad.jp>
Jordan K. Hubbard [Fri, 2 Jul 1999 02:49:24 +0000 (02:49 +0000)]
127     fire            Firewire driver <ikob@koganei.wide.ad.jp>

25 years agoFix stupid error where more bits where set than actual distributions
Jordan K. Hubbard [Fri, 2 Jul 1999 02:00:23 +0000 (02:00 +0000)]
Fix stupid error where more bits where set than actual distributions
used.  Doh!

Embarassingly-pointed-out-by: Brian Dean <brdean@unx.sas.com>

25 years ago#include <machine/md_var.h> to bring the prototype for
Peter Wemm [Thu, 1 Jul 1999 22:58:03 +0000 (22:58 +0000)]
#include <machine/md_var.h> to bring the prototype for
alpha_platform_assign_pciintr() into scope (!).

25 years agoFix a warning - the code is correct but gcc can't tell.
Peter Wemm [Thu, 1 Jul 1999 22:54:55 +0000 (22:54 +0000)]
Fix a warning - the code is correct but gcc can't tell.

25 years agoMoving the initialization for write sooner quiets a warning.
Peter Wemm [Thu, 1 Jul 1999 22:52:40 +0000 (22:52 +0000)]
Moving the initialization for write sooner quiets a warning.

25 years agoUnused variable.
Peter Wemm [Thu, 1 Jul 1999 22:49:03 +0000 (22:49 +0000)]
Unused variable.

25 years agoPull some prototypes into scope
Peter Wemm [Thu, 1 Jul 1999 22:48:31 +0000 (22:48 +0000)]
Pull some prototypes into scope

25 years agoAdd alpha_platform_setup_ide_intr() and alpha_platform_assign_pciintr()
Peter Wemm [Thu, 1 Jul 1999 22:47:53 +0000 (22:47 +0000)]
Add alpha_platform_setup_ide_intr() and alpha_platform_assign_pciintr()
prototypes.

25 years agoFix a printf int/long problem on the Alpha
Peter Wemm [Thu, 1 Jul 1999 22:14:51 +0000 (22:14 +0000)]
Fix a printf int/long problem on the Alpha

25 years agomove <sys/systm.h> before <sys/buf.h>
Peter Wemm [Thu, 1 Jul 1999 22:12:39 +0000 (22:12 +0000)]
move <sys/systm.h> before <sys/buf.h>

25 years ago#include <machine/md_var.h> to get prototype for alpha_register_pci_scsi().
Peter Wemm [Thu, 1 Jul 1999 22:00:12 +0000 (22:00 +0000)]
#include <machine/md_var.h> to get prototype for alpha_register_pci_scsi().

25 years agoDeclare busdma_swi() like on i386 version.
Peter Wemm [Thu, 1 Jul 1999 21:58:38 +0000 (21:58 +0000)]
Declare busdma_swi() like on i386 version.

25 years agoNuke the BUGS sections of these man pages because they are not appropriate.
Mike Pritchard [Thu, 1 Jul 1999 21:13:06 +0000 (21:13 +0000)]
Nuke the BUGS sections of these man pages because they are not appropriate.

25 years agoFix typos/spelling errors.
Mike Pritchard [Thu, 1 Jul 1999 21:09:57 +0000 (21:09 +0000)]
Fix typos/spelling errors.

25 years agoAvoid warning on nextpaddr, just like on the x86.
Peter Wemm [Thu, 1 Jul 1999 20:59:56 +0000 (20:59 +0000)]
Avoid warning on nextpaddr, just like on the x86.

25 years agoModify code to be -Wall'able.
Mike Spengler [Thu, 1 Jul 1999 20:48:59 +0000 (20:48 +0000)]
Modify code to be -Wall'able.

PR: bin/11315

25 years agoTidy up a few warnings when used without splash. (things defined but not
Peter Wemm [Thu, 1 Jul 1999 20:43:03 +0000 (20:43 +0000)]
Tidy up a few warnings when used without splash. (things defined but not
used)

25 years agoPrototype for alpha_fix_srm_checksum().
Peter Wemm [Thu, 1 Jul 1999 20:34:28 +0000 (20:34 +0000)]
Prototype for alpha_fix_srm_checksum().

25 years agoFix warnings on the Alpha.
Peter Wemm [Thu, 1 Jul 1999 20:29:25 +0000 (20:29 +0000)]
Fix warnings on the Alpha.

25 years agoGet prototype for isa_init_intr().
Peter Wemm [Thu, 1 Jul 1999 20:25:39 +0000 (20:25 +0000)]
Get prototype for isa_init_intr().

25 years agoFix a warning, and pull prototypes into scope for ide/isa interrupt setup.
Peter Wemm [Thu, 1 Jul 1999 20:23:01 +0000 (20:23 +0000)]
Fix a warning, and pull prototypes into scope for ide/isa interrupt setup.

25 years agoFix printf int/long format problems on the Alpha.
Peter Wemm [Thu, 1 Jul 1999 20:01:03 +0000 (20:01 +0000)]
Fix printf int/long format problems on the Alpha.

25 years agoAdd mising aio_* man pages. Fixed a minor typo in aio_read.2,
Wes Peters [Thu, 1 Jul 1999 19:58:12 +0000 (19:58 +0000)]
Add mising aio_* man pages.  Fixed a minor typo in aio_read.2,
and "corrected" statement of Posix conformance.

25 years agoint/long printf format problems on the Alpha.
Peter Wemm [Thu, 1 Jul 1999 19:56:50 +0000 (19:56 +0000)]
int/long printf format problems on the Alpha.

25 years agoFix some int/long printf problems for the Alpha
Peter Wemm [Thu, 1 Jul 1999 19:53:43 +0000 (19:53 +0000)]
Fix some int/long printf problems for the Alpha

25 years agoQuiet warnings on an Alpha. CBSIZE has long type and causes the other
Peter Wemm [Thu, 1 Jul 1999 19:46:36 +0000 (19:46 +0000)]
Quiet warnings on an Alpha. CBSIZE has long type and causes the other
ints to promote to long.

25 years agoQuiet warnings on Alpha. (db_expr_t is a long on alpha, int on x86)
Peter Wemm [Thu, 1 Jul 1999 19:42:56 +0000 (19:42 +0000)]
Quiet warnings on Alpha.  (db_expr_t is a long on alpha, int on x86)

25 years agoRemove the now-bogus comment about using iosiz with npx0 for memory
Jordan K. Hubbard [Thu, 1 Jul 1999 18:39:23 +0000 (18:39 +0000)]
Remove the now-bogus comment about using iosiz with npx0 for memory
sizing - environment does this properly  now.  Thanks, Peter!

25 years agoLook up the kernel environment for MAXMEM as a final override for the
Peter Wemm [Thu, 1 Jul 1999 18:33:22 +0000 (18:33 +0000)]
Look up the kernel environment for MAXMEM as a final override for the
memory size.  If somebody wants to change the name, fine - I used this
since it's consistant with the config variable it replaces.
This is intended to replace the npx0 msize hack (which no longer works).

25 years agoMove kern_envp and preload initialization a little earlier so that we
Peter Wemm [Thu, 1 Jul 1999 18:27:15 +0000 (18:27 +0000)]
Move kern_envp and preload initialization a little earlier so that we
can do a getenv_int() inside the memory sizing routines to override the
memory limit.

25 years agoFix ambigious else.
Bill Fumerola [Thu, 1 Jul 1999 17:59:17 +0000 (17:59 +0000)]
Fix ambigious else.

Reviewed By: eivind

25 years agoDon't include trailing whitespace in ID tokens before comments. ie:
Peter Wemm [Thu, 1 Jul 1999 16:20:16 +0000 (16:20 +0000)]
Don't include trailing whitespace in ID tokens before comments. ie:
options FOO=10 # comment
would give FOO the value of "10 " and that caused unwanted
touches on the opt_*.h files.

I hope I've got this right..

25 years agoDetect and remove defunct or unknown options from opt_*.h files. This
Peter Wemm [Thu, 1 Jul 1999 16:17:13 +0000 (16:17 +0000)]
Detect and remove defunct or unknown options from opt_*.h files.  This
can happen when options are removed from the options files.

25 years agoFix two easy warnings when using BRIDGE without IPFIREWALL.
Peter Wemm [Thu, 1 Jul 1999 15:07:08 +0000 (15:07 +0000)]
Fix two easy warnings when using BRIDGE without IPFIREWALL.

25 years agoFix some warnings, unused functions etc.
Peter Wemm [Thu, 1 Jul 1999 15:05:11 +0000 (15:05 +0000)]
Fix some warnings, unused functions etc.

25 years agoDocument that FTP_PASSIVE_MODE should be explicitly
Ruslan Ermilov [Thu, 1 Jul 1999 14:32:07 +0000 (14:32 +0000)]
Document that FTP_PASSIVE_MODE should be explicitly
set to `YES' for passive mode to be used by default.
Just setting FTP_PASSIVE_MODE is not sufficient, as
it was before.

Noted by: eivind
Reviewed by: des

25 years agoTurn System V command line syntax ``on'' by default.
Ruslan Ermilov [Thu, 1 Jul 1999 13:33:56 +0000 (13:33 +0000)]
Turn System V command line syntax ``on'' by default.

Requested by: peter
Reviewed by: des, billf

25 years agoFix warning. va_fsid is udev_t, which is int32_t. No need to use %lx.
Peter Wemm [Thu, 1 Jul 1999 13:32:54 +0000 (13:32 +0000)]
Fix warning. va_fsid is udev_t, which is int32_t.  No need to use %lx.

25 years agoSpelll 'asynchronous' coriectly
Kris Kennaway [Thu, 1 Jul 1999 13:31:30 +0000 (13:31 +0000)]
Spelll 'asynchronous' coriectly

25 years agoSlight reorganization of kernel thread/process creation. Instead of using
Peter Wemm [Thu, 1 Jul 1999 13:21:46 +0000 (13:21 +0000)]
Slight reorganization of kernel thread/process creation.  Instead of using
SYSINIT_KT() etc (which is a static, compile-time procedure), use a
NetBSD-style kthread_create() interface.  kproc_start is still available
as a SYSINIT() hook.  This allowed simplification of chunks of the
sysinit code in the process.  This kthread_create() is our old kproc_start
internals, with the SYSINIT_KT fork hooks grafted in and tweaked to work
the same as the NetBSD one.

One thing I'd like to do shortly is get rid of nfsiod as a user initiated
process.  It makes sense for the nfs client code to create them on the
fly as needed up to a user settable limit.  This means that nfsiod
doesn't need to be in /sbin and is always "available".  This is a fair bit
easier to do outside of the SYSINIT_KT() framework.

25 years agoFix the NAME section of the kldload.2 manpage, which made reference to
Sheldon Hearn [Thu, 1 Jul 1999 12:16:06 +0000 (12:16 +0000)]
Fix the NAME section of the kldload.2 manpage, which made reference to
kldunload. Add proper cross-references to the whole family.

PR: 12472
Submitted by: Chris Costello <chris@calldei.com>