]> CyberLeo.Net >> Repos - FreeBSD/FreeBSD.git/log
FreeBSD/FreeBSD.git
21 years agoPrepare for switching to unlimited chains format
Andrey A. Chernov [Fri, 30 Aug 2002 20:34:36 +0000 (20:34 +0000)]
Prepare for switching to unlimited chains format

21 years ago - Convert the function definition to declare its arguments
Robert Drehmel [Fri, 30 Aug 2002 20:33:05 +0000 (20:33 +0000)]
 - Convert the function definition to declare its arguments
   in the ANSI-C format.
 - Change the code a bit to hopefully save some cycles.
   I.e. (simplified) change

     a = b + 1;
     while (--b & 0x7)
/* ... */
   to
     a = b;
     for (; b & 0x7; b--)
/* ... */
   and
     while (--a >= 0)
/* ... */
   to
     for (; a > 0; a--)
/* ... */
 - Equip two function arguments of swab() with the 'restrict'
   type qualifier in form of the '__restrict' macro.  This is
   specified by POSIX.1-2001.

21 years agoPrepare for switching to unlimited chains format.
Andrey A. Chernov [Fri, 30 Aug 2002 20:26:02 +0000 (20:26 +0000)]
Prepare for switching to unlimited chains format.
Optimize chains lookup a bit.

21 years ago - Update the manual pages of index() and rindex() to show
Robert Drehmel [Fri, 30 Aug 2002 19:42:07 +0000 (19:42 +0000)]
 - Update the manual pages of index() and rindex() to show
   <strings.h> as the associated header file.
   The prototypes have been moved there from <string.h> because
   POSIX.1-2001 said so.
 - Conditionally include either <strings.h> or <string.h> based
   on whether the [r]index() or str[r]chr() functions are
   compiled, respectively.
 - Style(9) tells us to
    - put a space after the return keyword
    - to check for a NUL character without using the ! operator.
    - use NULL instead of (type *)NULL where the compiler knows
      the type.
   Apply these rules.
 - Rather use ANSI-C function definitions than K&R ones.
 - For index(3), correct second function argument's type; it was
   declared to be a `const char' before and is now an `int'.

21 years ago - Update the manual page to show that the associated header file
Robert Drehmel [Fri, 30 Aug 2002 19:08:53 +0000 (19:08 +0000)]
 - Update the manual page to show that the associated header file
   is <strings.h> and not <string.h> anymore.
 - Tell the reader about this change in the HISTORY section.
 - Switch to use an ANSI-C function definition.
 - Include <strings.h> instead of <string.h> in the source file.

21 years agoAdd IDs for upcoming LSI U320 products that use the MegaRAID interface.
Justin T. Gibbs [Fri, 30 Aug 2002 18:28:00 +0000 (18:28 +0000)]
Add IDs for upcoming LSI U320 products that use the MegaRAID interface.

Adjust format of one comment so it doesn't wrap at 80 columns.
Submitted by: "Moore, Eric Dean" <emoore@lsil.com>

21 years agoImplement data, text, and vmem limit checking in the elf loader and svr4
Matthew Dillon [Fri, 30 Aug 2002 18:09:46 +0000 (18:09 +0000)]
Implement data, text, and vmem limit checking in the elf loader and svr4
compat code.  Clean up accounting for multiple segments.  Part 1/2.

Submitted by: Andrey Alekseyev <uitm@zenon.net> (with some modifications)
MFC after: 3 days

21 years agoReserve majors 171 and 172 for the "mide" (LSI MegaRAID IDE control device)
Justin T. Gibbs [Fri, 30 Aug 2002 16:54:02 +0000 (16:54 +0000)]
Reserve majors 171 and 172 for the "mide" (LSI MegaRAID IDE control device)
and "mided" (LSI MegaRAID IDE disk device).

Submitted by: "Moore, Eric Dean" <emoore@lsil.com>

21 years ago - Update strcasecmp(3)/strncasecmp(3) to reflect the fact that
Robert Drehmel [Fri, 30 Aug 2002 15:40:01 +0000 (15:40 +0000)]
 - Update strcasecmp(3)/strncasecmp(3) to reflect the fact that
   the prototypes for both functions are now in the <strings.h>
   header, as required by IEEE Std 1003.1-2001.
 - Add one sentence about that in the HISTORY section.
 - Include <strings.h> in the source file to have the prototypes
   in scope when the _ANSI_SOURCE macro is defined.

21 years agoRevert previous untested revision. The i386 loader consists of three parts:
John Baldwin [Fri, 30 Aug 2002 14:59:47 +0000 (14:59 +0000)]
Revert previous untested revision.  The i386 loader consists of three parts:
At the front is btxldr, in the middle is BTX itself (our mini-kernel), and
then the 'client' (loader.bin) which is the actual loader itself.  boot2
just executes a raw ELF or a.out binary with the only setup provided being
that a bootinfo structure is passed on the stack.  Now, since loader.bin
is a BTX client, the loader needs to be able to locate a BTX kernel for
the client to execute in the context.  Thus, just like pxelder, btxldr
uses the a.out header on the loader binary to find the BTX kernel stored
in the loader and set it up.  It does _not_ just reuse the BTX kernel
that boot2 invoked it with.  This is because it can't assume that it will
_have_ a "spare" BTX kernel lying around.  For example, when cdboot
loads the loader there isn't an existing BTX kernel.  In fact, cdboot
will only work with an a.out loader as well since it also "borrows" the
BTX kernel in the loader binary (which it finds by parsing the a.out
header) just as pxeldr does.  The only difference between cdboot and
pxeldr is where they get /boot/loader from.

If we wanted to make /boot/loader be an actual ELF binary we would need
to change the following utilites to handle that (and they all have to be
able to handle locating the BTX kernel inside of an ELF binary somehow):
- btxldr
- pxeldr
- cdboot

If we didn't want to require a flag day but make the transition smooth
then we need to be able to support both a.out and ELF versions of
/boot/loader which isn't exactly trivial since all three of these utilities
are written in assembly.

Pointy-hat to: peter

21 years agoWhitespace-only: don't mix tabs and spaces for doing identation.
Maxim Sobolev [Fri, 30 Aug 2002 14:21:01 +0000 (14:21 +0000)]
Whitespace-only: don't mix tabs and spaces for doing identation.

21 years agoFix a silly typo in user-setable promisc mode code.
Maxim Sobolev [Fri, 30 Aug 2002 13:37:13 +0000 (13:37 +0000)]
Fix a silly typo in user-setable promisc mode code.

Pointed out by: Yann Berthier <yb@sainte-barbe.org>
MFC after: 1 day

21 years agoestablish default values for /etc/rc.d/pcvt script
Hellmuth Michaelis [Fri, 30 Aug 2002 13:01:42 +0000 (13:01 +0000)]
establish default values for /etc/rc.d/pcvt script

21 years agodocument options for integrating pcvt configuration into the new /etc/rc.d
Hellmuth Michaelis [Fri, 30 Aug 2002 12:44:44 +0000 (12:44 +0000)]
document options for integrating pcvt configuration into the new /etc/rc.d
startup system.

21 years agointegrate pcvt configuration into the new /etc/rc.d startup system
Hellmuth Michaelis [Fri, 30 Aug 2002 12:43:23 +0000 (12:43 +0000)]
integrate pcvt configuration into the new /etc/rc.d startup system

PR:           i386/7100
Reviewed by:  Gordon Tetlow <gordon@FreeBSD.org>

21 years agos/hint.acpi.0.disable/hint.acpi.0.disabled/
Mitsuru IWASAKI [Fri, 30 Aug 2002 11:11:07 +0000 (11:11 +0000)]
s/hint.acpi.0.disable/hint.acpi.0.disabled/

Fix device hints entry for disabling acpi(4).
This also should fix the arbitration with apm(4) when both drivers
are enabled.

Note that your /boot/device.hints needs to be updated if you want to
stop auto-loading acpi.ko or disable acpi(4).

21 years agoUse correct arguments to sysctl in code example.
Johan Karlsson [Fri, 30 Aug 2002 10:52:17 +0000 (10:52 +0000)]
Use correct arguments to sysctl in code example.

PR: 40038
Submitted by: perky
Approved by: sheldonh (mentor)
MFC after: 1 week

21 years agoCorrectly handle setting, getting and deleting EA's with zero length content.
Poul-Henning Kamp [Fri, 30 Aug 2002 08:57:09 +0000 (08:57 +0000)]
Correctly handle setting, getting and deleting EA's with zero length content.

Sponsored by: DARPA & NAI Labs.

21 years agoAdd the new userland tool for manipulating extended attributes.
Poul-Henning Kamp [Fri, 30 Aug 2002 08:53:03 +0000 (08:53 +0000)]
Add the new userland tool for manipulating extended attributes.

This will replace the existing getextattr(8) and setextattr(8) with
a single binary responding to the names getextattr, setextattr,
rmextattr and lsextattr.

This program is not yet connected to the build.

Sponsored by: DARPA and NAI Labs.

21 years agoRewrite wargames(6) in C. A program in C in the public domain is better than
Juli Mallett [Fri, 30 Aug 2002 07:14:42 +0000 (07:14 +0000)]
Rewrite wargames(6) in C.  A program in C in the public domain is better than
a shell script with a big copyright.  Or maybe just a good way to spend an hour
after watching a Matthew Broderick flick.

21 years agoWas mising ftpchroot.5
David E. O'Brien [Fri, 30 Aug 2002 06:50:08 +0000 (06:50 +0000)]
Was mising ftpchroot.5

PR: 40717
Submitted by: Andrew <andrew@ugh.net.au>

21 years agoUnbreak kernel build by printing Maxmem using %ld instead of old (now changed)
Andrey A. Chernov [Fri, 30 Aug 2002 06:13:39 +0000 (06:13 +0000)]
Unbreak kernel build by printing Maxmem using %ld instead of old (now changed)
%u

21 years agoThe aac driver needs 256K of contiguous physical memory in order to
Scott Long [Fri, 30 Aug 2002 05:02:07 +0000 (05:02 +0000)]
The aac driver needs 256K of contiguous physical memory in order to
attach.  If it can't get that (highly likely if loaded as a module on
a system that's been up for a while), give a more descriptive error
message.
Also clean up some nearby style nits.

MFC after: 2 days

21 years agoChange hw.physmem and hw.usermem to unsigned long like they used to be
Peter Wemm [Fri, 30 Aug 2002 04:04:37 +0000 (04:04 +0000)]
Change hw.physmem and hw.usermem to unsigned long like they used to be
in the original hardwired sysctl implementation.

The buf size calculator still overflows an integer on machines with large
KVA (eg: ia64) where the number of pages does not fit into an int.  Use
'long' there.

Change Maxmem and physmem and related variables to 'long', mostly for
completeness.  Machines are not likely to overflow 'int' pages in the
near term, but then again, 640K ought to be enough for anybody.  This
comes for free on 32 bit machines, so why not?

21 years agoSigh. Ken Merry convinced me that my attempts to DTRT were wrong.
Matt Jacob [Fri, 30 Aug 2002 03:36:50 +0000 (03:36 +0000)]
Sigh. Ken Merry convinced me that my attempts to DTRT were wrong.
Replace dual copyright with a plain BSD style copyright assigned
to LSI Logic. This is still within the intents of express consent
from LSI.

MFC after: 2 days

21 years agoNew release note: XFree86-derived termcap(5) entry for xterm.
Bruce A. Mah [Fri, 30 Aug 2002 02:56:24 +0000 (02:56 +0000)]
New release note:  XFree86-derived termcap(5) entry for xterm.

21 years agoFinish bringing support for the vendor attribute to the stylesheets
Bruce A. Mah [Fri, 30 Aug 2002 02:52:14 +0000 (02:52 +0000)]
Finish bringing support for the vendor attribute to the stylesheets
used for the release documentation.

21 years agoAdd 909A PCI id.
Matt Jacob [Fri, 30 Aug 2002 02:35:31 +0000 (02:35 +0000)]
Add 909A PCI id.

MFC after: 2 days

21 years agoAdd an additional copyright (with the express consent of LSI Logic) that
Matt Jacob [Fri, 30 Aug 2002 02:34:19 +0000 (02:34 +0000)]
Add an additional copyright (with the express consent of LSI Logic) that
specifically allows for (via 'BSD Style' licensing) source && binary
redistribution.

Pointy hat to: Matt, for not getting this done ahead of time.

MFC after: 2 days

21 years agoHopefully unbreak world. ke_slptime is gone. It should really have been
Peter Wemm [Fri, 30 Aug 2002 02:18:38 +0000 (02:18 +0000)]
Hopefully unbreak world.  ke_slptime is gone.  It should really have been
looking at p_ksegrp.kg_slptime anyway.

21 years agoAARGH! btoc() is used in the MI buffer sizing routines to calculate
Peter Wemm [Fri, 30 Aug 2002 00:29:52 +0000 (00:29 +0000)]
AARGH!  btoc() is used in the MI buffer sizing routines to calculate
the minimum of either physmem or KVA.  But.. btoc() casts the address
to (unsigned int).  This is NOT GOOD on 64 bit machines and on alpha and
ia64, this results in a buffer limit of around 500K (not megs).  This
causes extreme disk access problems on alpha and ia64.  Since this cast
is simply to ensure that it is unsigned, use 'vm_offset_t' instead.  This
is available because it is already defined in types.h.

Alpha has been suffering from this for ages. It always felt like the
caching wasn't working, and unfortunately it turned out that way. :-(

21 years agoRejig the code to figure out estcpu and work out how long a KSEGRP has been
Julian Elischer [Fri, 30 Aug 2002 00:25:49 +0000 (00:25 +0000)]
Rejig the code to figure out estcpu and work out how long a KSEGRP has been
idle. What was there before was surprisingly ALMOST correct.

Peter and I fried our brains on this for a couple of hours figuring out
what this actually means in the context of multiple threads.

Reviewed by: peter@freebsd.org

21 years agoTry #2 at having /boot/loader default to ELF. Have pxeldr build its
Peter Wemm [Thu, 29 Aug 2002 23:37:44 +0000 (23:37 +0000)]
Try #2 at having /boot/loader default to ELF.  Have pxeldr build its
own a.out version of loader.bin rather than depend on ../loader/loader
being a.out.

21 years agoRemove much of the dereferencing of the fd table entries to look
Daniel Eischen [Thu, 29 Aug 2002 23:06:07 +0000 (23:06 +0000)]
Remove much of the dereferencing of the fd table entries to look
at file flags and replace it with functions that will avoid null
pointer checks.

MFC to be done by archie ;-)

PR: 42100
Reviewed by: archie, robert
MFC after: 3 days

21 years agoActually remove the a.out kld loader. While I am not 100% sure, I believe
Peter Wemm [Thu, 29 Aug 2002 23:04:05 +0000 (23:04 +0000)]
Actually remove the a.out kld loader.  While I am not 100% sure, I believe
it is broken.  It certainly has been suffering neglect.  It is not needed
because we never shipped a.out kld's and they never really worked right.

21 years agoActually remove the stale a.out kld support. This is the stuff that was
Peter Wemm [Thu, 29 Aug 2002 23:02:37 +0000 (23:02 +0000)]
Actually remove the stale a.out kld support.  This is the stuff that was
never updated for the metadata infrastructure.

21 years agoOur gnats sets Confidential to "no" by default.
Giorgos Keramidas [Thu, 29 Aug 2002 22:59:22 +0000 (22:59 +0000)]
Our gnats sets Confidential to "no" by default.

PR: docs/41727
Submitted by: osgene@web.de
Reviewed by: des

21 years agoMake the libc_r version of select() set the readable or writable
Archie Cobbs [Thu, 29 Aug 2002 21:39:19 +0000 (21:39 +0000)]
Make the libc_r version of select() set the readable or writable
file descriptor bit if poll() returns POLLERR, POLLHUP, or POLLNVAL.
Othewise, it's possible for select() to return successfully but
with no bits set.

Reviewed by: deischen
MFC after: 3 days
PR: bin/42175

21 years agoCleanup grammer related to PR 40907 and add a section on TCP delayed
Matthew Dillon [Thu, 29 Aug 2002 20:34:06 +0000 (20:34 +0000)]
Cleanup grammer related to PR 40907 and add a section on TCP delayed
acks as suggested by Mike Harding <mvh@ix.netcom.com>

21 years agoFix crack-smoking code that was panicing on the quad xeon:
Julian Elischer [Thu, 29 Aug 2002 19:49:53 +0000 (19:49 +0000)]
Fix crack-smoking code that was panicing on the quad xeon:
        - If either of proc or kse are NULL during thread_exit(), then
          the kernel is going to fault because parts of the function
          assume they aren't NULL.  Instead, just assert they aren't NULL
          (as well as the kse group) and assume they are in all of the
          code.  It doesn't make sense for them to be NULL here anyways.
        - Move the PROC_UNLOCK(p) up above clearing td_proc, etc. since
          otherwise we will panic if the proc's lock is contested.

Submitted by: jhb@freebsd.org

21 years agoFix a silly off my one error that caused crashes on resume in certain
Scott Long [Thu, 29 Aug 2002 15:06:26 +0000 (15:06 +0000)]
Fix a silly off my one error that caused crashes on resume in certain
circumstances.  The problem was only reported with -stable, but it's
obviously wrong in -current also.  MFC is forthcoming.

Submitted by: doconnor@dsoft.com.au

21 years agoFix a warning of "possibly used before initialisation".
Giorgos Keramidas [Thu, 29 Aug 2002 14:29:09 +0000 (14:29 +0000)]
Fix a warning of "possibly used before initialisation".

Reviewed by: tjr

21 years agoAdd <stdlib.h> to get a prototype for exit().
Giorgos Keramidas [Thu, 29 Aug 2002 14:26:49 +0000 (14:26 +0000)]
Add <stdlib.h> to get a prototype for exit().

Reviewed by: tjr

21 years agosome ipfilter files that accidently got imported here
Darren Reed [Thu, 29 Aug 2002 13:27:26 +0000 (13:27 +0000)]
some ipfilter files that accidently got imported here

21 years agoCorrectly handle ifr.ifr_flags/ifr.ifr_flagshigh like ifconfig(8) does.
Maxim Sobolev [Thu, 29 Aug 2002 12:52:28 +0000 (12:52 +0000)]
Correctly handle ifr.ifr_flags/ifr.ifr_flagshigh like ifconfig(8) does.

MFC after: 1 day

21 years agoAdd sanity check seeing if adjusted start address exceeds end address
Mitsuru IWASAKI [Thu, 29 Aug 2002 12:39:21 +0000 (12:39 +0000)]
Add sanity check seeing if adjusted start address exceeds end address
after boundary and alignment adjustment.

21 years agoThe mode can be "r+" as well on PUT, but only "a" on APPE.
Yaroslav Tykhiy [Thu, 29 Aug 2002 09:53:51 +0000 (09:53 +0000)]
The mode can be "r+" as well on PUT, but only "a" on APPE.

21 years agoFix lexer jam on unimplemented commands.
Yaroslav Tykhiy [Thu, 29 Aug 2002 09:23:08 +0000 (09:23 +0000)]
Fix lexer jam on unimplemented commands.

Submitted by: maxim
MFC after: 5 days

21 years agoFix kernel build breakage when ACPI_DEBUG option is specified.
Mitsuru IWASAKI [Thu, 29 Aug 2002 08:31:06 +0000 (08:31 +0000)]
Fix kernel build breakage when ACPI_DEBUG option is specified.

21 years agoMention that fdisk(8) should be used to grow the containing slice
Giorgos Keramidas [Thu, 29 Aug 2002 08:12:09 +0000 (08:12 +0000)]
Mention that fdisk(8) should be used to grow the containing slice
before growing partitions with growfs(8), if necessary.

PR: docs/42148
Submitted by: Chris S.J. Peron <maneo@bsdpro.com>

21 years agoRenamed poorly named setregs to exec_setregs. Moved its prototype to
Jake Burkholder [Thu, 29 Aug 2002 06:17:48 +0000 (06:17 +0000)]
Renamed poorly named setregs to exec_setregs.  Moved its prototype to
imgact.h with the other exec support functions.

21 years agoFixed printf format errors.
Jake Burkholder [Thu, 29 Aug 2002 05:49:04 +0000 (05:49 +0000)]
Fixed printf format errors.

21 years agoWargames was PG-13, not PGP 13. I just watched it with my girlfriend, and
Juli Mallett [Thu, 29 Aug 2002 04:07:20 +0000 (04:07 +0000)]
Wargames was PG-13, not PGP 13.  I just watched it with my girlfriend, and
there was no mention of PGP, despite what the manpage said!  Total let down!

MFC After: 65456665556666545666655654 DTMF tones.

21 years ago- made ppp compliant to RFC 2472 (based on a patch from another
Brian Somers [Thu, 29 Aug 2002 02:44:58 +0000 (02:44 +0000)]
- made ppp compliant to RFC 2472 (based on a patch from another
  contributor)
- support ipv6cpretry and ipv6cpretries, which are IPv6 versions
  of ipcpretry and ipcpretries.
- improve handling of IPv6 link-local addresses

Submitted by: JINMEI Tatuya <jinmei@isl.rdc.toshiba.co.jp>

21 years agoMinor cleanup.
Jake Burkholder [Thu, 29 Aug 2002 02:39:39 +0000 (02:39 +0000)]
Minor cleanup.

21 years agoInitiate deorbit burn of i386 a.out kld "support" in loader. Note that
Peter Wemm [Thu, 29 Aug 2002 02:02:28 +0000 (02:02 +0000)]
Initiate deorbit burn of i386 a.out kld "support" in loader.  Note that
this was quite broken, it never was updated for metadata support.
The a.out kld file support was never really used, as it wasn't necessary.
You could always load elf kld's, even in an a.out kernel.

21 years agoRemoved legacy signal trampoline.
Jake Burkholder [Thu, 29 Aug 2002 01:59:54 +0000 (01:59 +0000)]
Removed legacy signal trampoline.

21 years agoRemoved support for in-kernel signal code.
Jake Burkholder [Thu, 29 Aug 2002 01:55:24 +0000 (01:55 +0000)]
Removed support for in-kernel signal code.

21 years agoResolve conflicts arising from the ACPI CA 20020815 import.
Mitsuru IWASAKI [Thu, 29 Aug 2002 01:52:27 +0000 (01:52 +0000)]
Resolve conflicts arising from the ACPI CA 20020815 import.

21 years agoThis commit was generated by cvs2svn to compensate for changes in r102550,
Mitsuru IWASAKI [Thu, 29 Aug 2002 01:51:24 +0000 (01:51 +0000)]
This commit was generated by cvs2svn to compensate for changes in r102550,
which included commits to RCS files with non-trunk default branches.

21 years agoVendor import of the Intel ACPI CA 20020815 drop.
Mitsuru IWASAKI [Thu, 29 Aug 2002 01:51:24 +0000 (01:51 +0000)]
Vendor import of the Intel ACPI CA 20020815 drop.

21 years agoInitiate deorbit burn for sys/kern/link_aout.c. We never shipped a.out
Peter Wemm [Thu, 29 Aug 2002 01:48:09 +0000 (01:48 +0000)]
Initiate deorbit burn for sys/kern/link_aout.c.  We never shipped a.out
kld's anywhere, and it was always possible to load ELF kld's even in an
a.out kernel.  There is no reason for this to exist anymore, and a.out
kld support has been suffering serious bitrot over the years.  They have
not been fully functional for quite some time.

21 years agoDon't require that sysentvec.sv_szsigcode be non-NULL.
Jake Burkholder [Thu, 29 Aug 2002 01:28:27 +0000 (01:28 +0000)]
Don't require that sysentvec.sv_szsigcode be non-NULL.

21 years agoUnrot SPARSE_MAPPING code (vm_map_pageable -> vm_map_wire).
Jake Burkholder [Thu, 29 Aug 2002 01:16:14 +0000 (01:16 +0000)]
Unrot SPARSE_MAPPING code (vm_map_pageable -> vm_map_wire).

21 years agoWhen poll(2)'ing for readability or writability of a file descriptor
Archie Cobbs [Thu, 29 Aug 2002 00:44:11 +0000 (00:44 +0000)]
When poll(2)'ing for readability or writability of a file descriptor
on behalf of a thread, we should check the POLLERR, POLLHUP, and
POLLNVAL flags as well to wake up the thread in these cases.

Suggested by: deischen
MFC after: 3 days

21 years agoAdd date of foundation of Federal Republic of Germany.
Greg Lehey [Thu, 29 Aug 2002 00:26:26 +0000 (00:26 +0000)]
Add date of foundation of Federal Republic of Germany.
Remove bogus event of 22 September 1953.

Submitted by:  Jan-Oliver Neumann <jneumann@ix.urz.uni-heidelberg.de>

21 years agoupdatepri() works on a ksegrp (where the scheduling parameters are), so
Peter Wemm [Wed, 28 Aug 2002 23:45:15 +0000 (23:45 +0000)]
updatepri() works on a ksegrp (where the scheduling parameters are), so
directly give it the ksegrp instead of the thread.  The only thing it used
to use in the thread was the ksegrp.

Reviewed by: julian

21 years agoOK, I have had it with losing my console because the AP's print their "I am
Peter Wemm [Wed, 28 Aug 2002 23:24:05 +0000 (23:24 +0000)]
OK, I have had it with losing my console because the AP's print their "I am
alive!" message right as the scsi probe messages happen.  This is a bit
nasty, but it seems to work.  At the point that we unlock the AP's, briefly
wait till they are all done while we hold the console on their behalf.

21 years agoOnly try to flush the FIFO of 500 characters, return EIO if that fails.
Poul-Henning Kamp [Wed, 28 Aug 2002 22:25:41 +0000 (22:25 +0000)]
Only try to flush the FIFO of 500 characters, return EIO if that fails.

This at least prevents the total hang of the machine when I open a
PCCARD sio device on -current, but it does not solve the problem.

21 years agoFix the generation of the HTML manual page link according to new
Marc Fonvieille [Wed, 28 Aug 2002 21:26:33 +0000 (21:26 +0000)]
Fix the generation of the HTML manual page link according to new
doc/share/sgml/freebsd.dsl

Signaled by: bmah

21 years agoaccept(2) on a socket that has been shutdown(2) normally returns
Archie Cobbs [Wed, 28 Aug 2002 20:56:01 +0000 (20:56 +0000)]
accept(2) on a socket that has been shutdown(2) normally returns
ECONNABORTED. Make this happen in the non-blocking case as well.
The previous behavior was to return EAGAIN, which (a) is not
consistent with the blocking case and (b) causes the application
to think the socket is still valid.

PR: bin/42100
Reviewed by: freebsd-net
MFC after: 3 days

21 years agoAllow one to grab the definition of struct ucred by defining _WANT_UCRED
Alfred Perlstein [Wed, 28 Aug 2002 20:39:48 +0000 (20:39 +0000)]
Allow one to grab the definition of struct ucred by defining _WANT_UCRED
instead of forcing _KERNEL.

Move the include of sys/_label.h in ucred.h under the
_KERNEL || _WANT_UCRED case.

21 years agoAdd a device description for Intel 82801CA/CAM (ICH3) USB controller
Josef Karthauser [Wed, 28 Aug 2002 20:24:49 +0000 (20:24 +0000)]
Add a device description for Intel 82801CA/CAM (ICH3) USB controller
USB-C.

PR: kern/41963

21 years agoMention new termcap and TERM=xterm-color -> xterm issue.
Jens Schweikhardt [Wed, 28 Aug 2002 19:34:49 +0000 (19:34 +0000)]
Mention new termcap and TERM=xterm-color -> xterm issue.

21 years agoChange default value of hw.acpi.sleep_delay to 0.
Mitsuru IWASAKI [Wed, 28 Aug 2002 18:46:09 +0000 (18:46 +0000)]
Change default value of hw.acpi.sleep_delay to 0.
This caused problems (reset or lock up) at wakeup.

21 years agosendmail 8.12.6 has been imported
Gregory Neil Shapiro [Wed, 28 Aug 2002 18:14:13 +0000 (18:14 +0000)]
sendmail 8.12.6 has been imported

21 years agoMove the $FreeBSD$ CVS labels far enough away from the $Id$ CVS labels so
Gregory Neil Shapiro [Wed, 28 Aug 2002 18:12:33 +0000 (18:12 +0000)]
Move the $FreeBSD$ CVS labels far enough away from the $Id$ CVS labels so
I don't have to manually resolve conflicts on every import.  Ironically,
most of these files have no changes from the vendor version except the
$FreeBSD$ line.

21 years agoChanges for the import of sendmail 8.12.6
Gregory Neil Shapiro [Wed, 28 Aug 2002 17:59:45 +0000 (17:59 +0000)]
Changes for the import of sendmail 8.12.6

21 years agoResolve conflicts from sendmail 8.12.6 import
Gregory Neil Shapiro [Wed, 28 Aug 2002 17:58:45 +0000 (17:58 +0000)]
Resolve conflicts from sendmail 8.12.6 import

21 years agoThis commit was generated by cvs2svn to compensate for changes in r102528,
Gregory Neil Shapiro [Wed, 28 Aug 2002 17:57:52 +0000 (17:57 +0000)]
This commit was generated by cvs2svn to compensate for changes in r102528,
which included commits to RCS files with non-trunk default branches.

21 years agoImport sendmail 8.12.6
Gregory Neil Shapiro [Wed, 28 Aug 2002 17:57:52 +0000 (17:57 +0000)]
Import sendmail 8.12.6

21 years agoAdd IFF_POLLING into the list of flags which are protected from changing via
Maxim Sobolev [Wed, 28 Aug 2002 15:55:49 +0000 (15:55 +0000)]
Add IFF_POLLING into the list of flags which are protected from changing via
ioctl(SIOCSIFFLAGS).

MFC after: 1 day

21 years agoMarginally simplify dsp_open error handling by adding an early test to
Orion Hodson [Wed, 28 Aug 2002 15:19:30 +0000 (15:19 +0000)]
Marginally simplify dsp_open error handling by adding an early test to
determine liklihood of opening device in requested directions.  Makes for
simpler error handling and change should close kern/35004.

PR: kern/35004.

21 years agomerge the changes from 3.4.29 that didn't go cleanly due to conflicts
Darren Reed [Wed, 28 Aug 2002 13:45:36 +0000 (13:45 +0000)]
merge the changes from 3.4.29 that didn't go cleanly due to conflicts

21 years agoThis commit was generated by cvs2svn to compensate for changes in r102521,
Darren Reed [Wed, 28 Aug 2002 13:42:48 +0000 (13:42 +0000)]
This commit was generated by cvs2svn to compensate for changes in r102521,
which included commits to RCS files with non-trunk default branches.

21 years agoWith a bit of luck, this will be a first-time right import of ipfilter 3.4.29
Darren Reed [Wed, 28 Aug 2002 13:42:48 +0000 (13:42 +0000)]
With a bit of luck, this will be a first-time right import of ipfilter 3.4.29
on to the vendor branch.

21 years agoFinally merge in the changes from ipfilter 3.4.29 to freebsd-current.
Darren Reed [Wed, 28 Aug 2002 13:41:36 +0000 (13:41 +0000)]
Finally merge in the changes from ipfilter 3.4.29 to freebsd-current.
Main changes here are related to the ftp proxy and making that work better.

21 years agoThis commit was generated by cvs2svn to compensate for changes in r102514,
Darren Reed [Wed, 28 Aug 2002 13:26:01 +0000 (13:26 +0000)]
This commit was generated by cvs2svn to compensate for changes in r102514,
which included commits to RCS files with non-trunk default branches.

21 years agoImport IPfilter 3.4.29. Main purpose is to address ftp proxy problems.
Darren Reed [Wed, 28 Aug 2002 13:26:01 +0000 (13:26 +0000)]
Import IPfilter 3.4.29.  Main purpose is to address ftp proxy problems.

21 years agoAdd a new command: show pciregs, equivalent to pciconf -l
Guido van Rooij [Wed, 28 Aug 2002 10:02:59 +0000 (10:02 +0000)]
Add a new command: show pciregs, equivalent to pciconf -l

Reviewed by: Doug Rabson (quite some time ago)
MFC after: 1 week

21 years agoLast missed bit of user-setable promisc mode.
Maxim Sobolev [Wed, 28 Aug 2002 07:24:20 +0000 (07:24 +0000)]
Last missed bit of user-setable promisc mode.

Pointy hat to:  sobomax
MFC after:      1 day

21 years agoprintf -> CAM_DEBUG so debugging prints are tunable.
Nate Lawson [Wed, 28 Aug 2002 07:03:44 +0000 (07:03 +0000)]
printf -> CAM_DEBUG so debugging prints are tunable.

MFC after: 1 week

21 years agoUse csio pointer instead of dereferencing hcb.
Nate Lawson [Wed, 28 Aug 2002 07:00:58 +0000 (07:00 +0000)]
Use csio pointer instead of dereferencing hcb.

MFC after: 1 week

21 years agoForward declare struct thread so that this header doesn't depend on its
Bruce Evans [Wed, 28 Aug 2002 05:14:55 +0000 (05:14 +0000)]
Forward declare struct thread so that this header doesn't depend on its
own namespace pollution/compatibility cruft.

Removed the main part of the pollution.  All clients have been converted
to either not depend on getting old locking interfaces from this new
locking header, or usual case to get it from another header (typically
vnode.h, where declaring old loccking interfaces is less bogus because
vnode.h uses them internally).

21 years ago- Clarify events which happened in the USA.
Greg Lehey [Wed, 28 Aug 2002 00:33:11 +0000 (00:33 +0000)]
- Clarify events which happened in the USA.
- Note that the USA has no King.
- Unify spelling "US", not "U.S.".
- Remove USSR holiday (anniversary of the revolution).
- Don't remove claim that the Federal Republic of Germany was founded
  in 1953 (22 September).  Somebody should clarify if something
  reallly did happen on that day.

21 years agoInclude the correct file (stdarg.h) and use va_list rather than _BSD_VA_LIST_
Brian Somers [Tue, 27 Aug 2002 20:11:58 +0000 (20:11 +0000)]
Include the correct file (stdarg.h) and use va_list rather than _BSD_VA_LIST_

Suggested by: mike

21 years agoPrint a '-' sign for negative zero. Tested with
Jens Schweikhardt [Tue, 27 Aug 2002 20:11:08 +0000 (20:11 +0000)]
Print a '-' sign for negative zero. Tested with

#include <stdio.h>
int main(void)
{
printf("%+f\n", -0.0);
printf("%+f\n", +0.0);
printf("%+f\n",  0.0);
return 0;
}

to output
-0.000000
+0.000000
+0.000000

PR: bin/41823
Submitted by: GOTO Kentaro <gotoken@notwork.org>
Liked by: bde
MFC after: 3 weeks

21 years agoDo not repeat the name of the flag in its description
Philippe Charnier [Tue, 27 Aug 2002 19:14:08 +0000 (19:14 +0000)]
Do not repeat the name of the flag in its description

21 years agoUse strchr() instead of index(). Use getopt(). Use err().
Philippe Charnier [Tue, 27 Aug 2002 19:06:44 +0000 (19:06 +0000)]
Use strchr() instead of index(). Use getopt(). Use err().

21 years agoAdd \n to various printfs. grep shows these should be the last ones.
Nate Lawson [Tue, 27 Aug 2002 18:43:14 +0000 (18:43 +0000)]
Add \n to various printfs.  grep shows these should be the last ones.