]> CyberLeo.Net >> Repos - FreeBSD/FreeBSD.git/log
FreeBSD/FreeBSD.git
23 years agoThis commit was generated by cvs2svn to compensate for changes in r70635,
David E. O'Brien [Wed, 3 Jan 2001 17:16:04 +0000 (17:16 +0000)]
This commit was generated by cvs2svn to compensate for changes in r70635,
which included commits to RCS files with non-trunk default branches.

23 years agoEnlist the FreeBSD-CURRENT users as testers of GCC 2.95.3 Release Candidate #1
David E. O'Brien [Wed, 3 Jan 2001 17:16:04 +0000 (17:16 +0000)]
Enlist the FreeBSD-CURRENT users as testers of GCC 2.95.3 Release Candidate #1

23 years agoDocument fsck_ffs's new SIGINFO handler.
Ian Dowse [Wed, 3 Jan 2001 13:53:48 +0000 (13:53 +0000)]
Document fsck_ffs's new SIGINFO handler.

Reviewed by: sheldonh

23 years agoRemove struct.h, which has been punted into the Attic.
Sheldon Hearn [Wed, 3 Jan 2001 12:35:39 +0000 (12:35 +0000)]
Remove struct.h, which has been punted into the Attic.

23 years agoAdd the Id of the Epson 1640 scanner.
Nick Hibma [Wed, 3 Jan 2001 11:46:09 +0000 (11:46 +0000)]
Add the Id of the Epson 1640 scanner.

Submitted by: Chris Shenton <chris@shenton.org>

23 years agoAdd a few Ids. From NetBSD.
Nick Hibma [Wed, 3 Jan 2001 11:45:03 +0000 (11:45 +0000)]
Add a few Ids. From NetBSD.

23 years agoCatch up with rev 1.19 of vnconfig.c .
Sheldon Hearn [Wed, 3 Jan 2001 10:55:23 +0000 (10:55 +0000)]
Catch up with rev 1.19 of vnconfig.c .

23 years agoFinally put this mistake to rest.
Poul-Henning Kamp [Wed, 3 Jan 2001 10:04:45 +0000 (10:04 +0000)]
Finally put this mistake to rest.

23 years agoadd new PC-Card entry, COREGA FEther PCC-TXD
MIHIRA Sanpei Yoshiro [Wed, 3 Jan 2001 08:42:08 +0000 (08:42 +0000)]
add new PC-Card entry, COREGA FEther PCC-TXD

ed1 at port 0x300-0x31f iomem 0xd0000-0xd0fff irq 3 slot 0 on pccard0
ed1: address 00:90:99:xx:xx:xx, type NE2000 (16 bit)

23 years agoMake sppp a kld.
John Hay [Wed, 3 Jan 2001 08:12:30 +0000 (08:12 +0000)]
Make sppp a kld.

23 years agoRemove alpm numbering.
Nicolas Souchu [Wed, 3 Jan 2001 07:39:07 +0000 (07:39 +0000)]
Remove alpm numbering.

23 years agoSpell while the more traditional way.
Warner Losh [Wed, 3 Jan 2001 05:30:30 +0000 (05:30 +0000)]
Spell while the more traditional way.

Submitted by: knu

Suggest that maybe rebuilding just libc might also work around this.

Suggested by: rgrimes and phk.

23 years agoRename the loader tunable from hw.sndunit to hw.snd.unit.
John Baldwin [Wed, 3 Jan 2001 02:09:31 +0000 (02:09 +0000)]
Rename the loader tunable from hw.sndunit to hw.snd.unit.

Submitted by: cg

23 years ago- When acknowledging interrupts, write the value 0xff to the interrupt
John Baldwin [Wed, 3 Jan 2001 01:32:57 +0000 (01:32 +0000)]
- When acknowledging interrupts, write the value 0xff to the interrupt
  status register rather than 0.  Without this, a single hardware volume
  event triggers an interrupt storm.
- Implement hardware volume control for the Maestro chips.  This version
  only handles the case where both channels are adjusted at the same time.

Reviewed by: cg

23 years agoAdd a new API for soundcards that have hardware volume control:
John Baldwin [Wed, 3 Jan 2001 01:29:47 +0000 (01:29 +0000)]
Add a new API for soundcards that have hardware volume control:
- The mixer_hwmute() function can be called when a soundcard receives a
  mute request.
- The mixer_hwstep() function can be used to adjust the volume of one or
  both channels.
- The 'hw.snd.hwvol_step' sysctl determines the amount that mixer_hwstep()
  adjusts the volume by on each call.
- The 'hw.snd.hwvol_mixer' sysctl specifies the mixer device to adjust the
  volume on for both functions.  The values used correspond to the
  SOUNDCARD_MIXER_* constants.

23 years agoCreate a new sysctl node 'hw.snd' and move 'hw.sndunit' to
John Baldwin [Wed, 3 Jan 2001 01:25:26 +0000 (01:25 +0000)]
Create a new sysctl node 'hw.snd' and move 'hw.sndunit' to
'hw.snd.unit'.

Reviewed by: cg

23 years agoAdd device id for the OZ Micro 6832. I didn't try to init it like the
Warner Losh [Tue, 2 Jan 2001 21:42:48 +0000 (21:42 +0000)]
Add device id for the OZ Micro 6832.  I didn't try to init it like the
Cirrus Logic parts that end in 6832, but maybe some machines will need
that in the future.

23 years agoadd pcic_p to the list of things built
Warner Losh [Tue, 2 Jan 2001 21:40:41 +0000 (21:40 +0000)]
add pcic_p to the list of things built

23 years agoCleanup and improve mode detection. Now, you should get what you actually
Nicolas Souchu [Tue, 2 Jan 2001 21:29:06 +0000 (21:29 +0000)]
Cleanup and improve mode detection. Now, you should get what you actually
want according to the modes set with the ppc(4) flags. Especially, it
should fix some problems with mode detection of parallel chipsets
configured to EPP but which have timing troubles with the drives. In such
a case, the driver should now fall back to slower modes (PS2, NIBBLE).

23 years agoNote problems with vi core dumping for a comple of days, plus
Warner Losh [Tue, 2 Jan 2001 21:28:48 +0000 (21:28 +0000)]
Note problems with vi core dumping for a comple of days, plus
workaround.

23 years agoRemove old PCI compatibility.
Nicolas Souchu [Tue, 2 Jan 2001 21:19:32 +0000 (21:19 +0000)]
Remove old PCI compatibility.

Submitted by: Andrew Gallatin <gallatin@freebsd.org>

23 years agoTidy up if_init routine so that it does not return an error.
Duncan Barclay [Tue, 2 Jan 2001 20:29:25 +0000 (20:29 +0000)]
Tidy up if_init routine so that it does not return an error.

23 years agoThe previous commit wasn't entirely correct. At least one goto to the
John Baldwin [Tue, 2 Jan 2001 18:54:09 +0000 (18:54 +0000)]
The previous commit wasn't entirely correct.  At least one goto to the
out: label in psignal() did not grab sched_lock before trying to release
it.  Also, the previous version had several cases where it grabbed
sched_lock before jumping to out: unneccessarily, so rework this a bit.
The runfast: and out: labels must be called with sched_lock released, and
the run: label must be called with it held.  Appropriate mtx_assert()'s
have been added that should catch any bugs that may still be in this
code.

Noticed by: bde

23 years agoCope with ports that have multiple versions *and* have embedded ``-''s
Brian Somers [Tue, 2 Jan 2001 15:17:59 +0000 (15:17 +0000)]
Cope with ports that have multiple versions *and* have embedded ``-''s
in their name.

23 years agodo not touch ra_addr if it is NULL. from IIJ SEIL team
Hajimu UMEMOTO [Tue, 2 Jan 2001 15:17:19 +0000 (15:17 +0000)]
do not touch ra_addr if it is NULL.  from IIJ SEIL team

Obtained from: KAME

23 years agoUnderstand ``multiple versions'' lines emitted from pkg_version.
Brian Somers [Tue, 2 Jan 2001 13:28:47 +0000 (13:28 +0000)]
Understand ``multiple versions'' lines emitted from pkg_version.

23 years agoFix the <sys/queue.h> abuse.
Poul-Henning Kamp [Tue, 2 Jan 2001 11:51:55 +0000 (11:51 +0000)]
Fix the <sys/queue.h> abuse.

Submitted by: Dima Dorfman <dima@unixfreak.org>
Reviewed by: /sbin/md5

23 years agoUse a slightly better documented way of calculating easter and
David Malone [Tue, 2 Jan 2001 11:35:57 +0000 (11:35 +0000)]
Use a slightly better documented way of calculating easter and
include a reference to the Calendar FAQ. Also make argument parsing
a little more robust.

PR: 23881

23 years agoRemove conflicts between unsinged char *s and int s.
Noriaki Mitsunaga [Tue, 2 Jan 2001 10:41:50 +0000 (10:41 +0000)]
Remove conflicts between unsinged char *s and int s.
It was possible cause of kernel panic.

Pointed Out by: phk@FreeBSD.ORG

23 years agoAdd an XXX about a <sys/queue.h> transgression which needs cleaned up.
Poul-Henning Kamp [Tue, 2 Jan 2001 10:34:09 +0000 (10:34 +0000)]
Add an XXX about a <sys/queue.h> transgression which needs cleaned up.

23 years agoEither cvs(1) or I forgot this file in my last commit.
Poul-Henning Kamp [Tue, 2 Jan 2001 09:42:47 +0000 (09:42 +0000)]
Either cvs(1) or I forgot this file in my last commit.

Please see commit log for rev 1.4 of src/sbin/mdconfig/mdconfig.c

23 years agoNewbusify the PCI PDQ (fddi) attachment. This isn't as clean as
Peter Wemm [Tue, 2 Jan 2001 09:30:48 +0000 (09:30 +0000)]
Newbusify the PCI PDQ (fddi) attachment.  This isn't as clean as
the EISA attachment and has not been tested (no hardware!), but at least
it stands a chance at working.  At least it compiles now.

23 years agoFix SCCS id string abuse I introduced.
David E. O'Brien [Tue, 2 Jan 2001 09:10:14 +0000 (09:10 +0000)]
Fix SCCS id string abuse I introduced.

23 years agoChange Embedded Controller lock to ACPI Global Lock.This is needed for
Takanori Watanabe [Tue, 2 Jan 2001 05:22:35 +0000 (05:22 +0000)]
Change Embedded Controller lock to ACPI Global Lock.This is needed for
mutual execution between  BIOS and OS.

23 years agoFix typo: s/memberhip/membership/
Peter Wemm [Tue, 2 Jan 2001 04:01:59 +0000 (04:01 +0000)]
Fix typo: s/memberhip/membership/

Submitted by:  Peter Avalos <pavalos@theshell.com>

23 years agoPowerPC platform-specific definitions (modeled on sys/i386/include/setjmp.h)
David E. O'Brien [Tue, 2 Jan 2001 00:34:24 +0000 (00:34 +0000)]
PowerPC platform-specific definitions (modeled on sys/i386/include/setjmp.h)

23 years agoPowerPC platform-specific definitions (modeled on sys/i386/include/types.h)
David E. O'Brien [Tue, 2 Jan 2001 00:30:49 +0000 (00:30 +0000)]
PowerPC platform-specific definitions (modeled on sys/i386/include/types.h)

23 years agoMinor style tweaks.
David E. O'Brien [Tue, 2 Jan 2001 00:11:41 +0000 (00:11 +0000)]
Minor style tweaks.

23 years agoPowerPC platform-specific definitions (modeled on sys/i386/include/param.h)
David E. O'Brien [Tue, 2 Jan 2001 00:06:45 +0000 (00:06 +0000)]
PowerPC platform-specific definitions (modeled on sys/i386/include/param.h)

23 years agoMP shells for the PowerPC platform.
David E. O'Brien [Mon, 1 Jan 2001 23:45:11 +0000 (23:45 +0000)]
MP shells for the PowerPC platform.

23 years agoMinor layout fixes.
Ben Smithurst [Mon, 1 Jan 2001 23:30:51 +0000 (23:30 +0000)]
Minor layout fixes.

PR: 24004
Submitted by: Jimmy Olgeni <olgeni@uli.it>

23 years agoAdd getenv(3) and setenv(3) into ``See also'' section.
Maxim Sobolev [Mon, 1 Jan 2001 23:26:54 +0000 (23:26 +0000)]
Add getenv(3) and setenv(3) into ``See also'' section.

23 years agoPowerPC platform-specific page size setting.
David E. O'Brien [Mon, 1 Jan 2001 23:26:39 +0000 (23:26 +0000)]
PowerPC platform-specific page size setting.

23 years agoFix typo: ispec -> ipsec
Ben Smithurst [Mon, 1 Jan 2001 23:26:16 +0000 (23:26 +0000)]
Fix typo: ispec -> ipsec

PR: 24005
Submitted by: Jimmy Olgeni <olgeni@uli.it>

23 years agoPowerPC platform-specific definitions.
David E. O'Brien [Mon, 1 Jan 2001 23:19:22 +0000 (23:19 +0000)]
PowerPC platform-specific definitions.

Obtained from: NetBSD (parts)

23 years agoInstall sysinstall.8 as part of buildworld.
Ben Smithurst [Mon, 1 Jan 2001 23:17:11 +0000 (23:17 +0000)]
Install sysinstall.8 as part of buildworld.

PR: 19818
Submitted by: Brian Candler <B.Candler@pobox.com>

23 years agoRemove a bogus #ifdef KTR stanza.
Poul-Henning Kamp [Mon, 1 Jan 2001 23:09:53 +0000 (23:09 +0000)]
Remove a bogus #ifdef KTR stanza.

Noticed by: Alexander Langer <alex@big.endian.de>

23 years agoAdd padding space and a version number to the md_ioctl structre for future
Poul-Henning Kamp [Mon, 1 Jan 2001 23:08:26 +0000 (23:08 +0000)]
Add padding space and a version number to the md_ioctl structre for future
extension.

Add ability to create a preload disk giving an address and a length
(suggested by imp)

Fix bug relating to very small md(4) devices.

Update md.c copyright to reflect the status of code copied from vn.c.
(noticed by dillon)

23 years agoShells for the atomic operations FreeBSD needs.
David E. O'Brien [Mon, 1 Jan 2001 23:06:59 +0000 (23:06 +0000)]
Shells for the atomic operations FreeBSD needs.
This is just waiting for a budding PowerPC ASM guy to fill in the blanks.

23 years agoPowerPC platform-specific type definitions.
David E. O'Brien [Mon, 1 Jan 2001 22:43:52 +0000 (22:43 +0000)]
PowerPC platform-specific type definitions.

23 years agoSort some of the _BSD_* types.
David E. O'Brien [Mon, 1 Jan 2001 22:26:56 +0000 (22:26 +0000)]
Sort some of the _BSD_* types.

23 years agoUpdate the e_machine values to include x86-64, PowerPC and [Strong]ARM;
David E. O'Brien [Mon, 1 Jan 2001 21:56:57 +0000 (21:56 +0000)]
Update the e_machine values to include x86-64, PowerPC and [Strong]ARM;
and fix the Sparc values.  Also update the ELFOSABI list while I'm here.

23 years agoPowerPC specific ELF ABI definitions.
David E. O'Brien [Mon, 1 Jan 2001 21:54:48 +0000 (21:54 +0000)]
PowerPC specific ELF ABI definitions.

23 years agoFurther tidy up the sbin/init and release builds.
Peter Wemm [Mon, 1 Jan 2001 21:39:00 +0000 (21:39 +0000)]
Further tidy up the sbin/init and release builds.

23 years agos/rsa0/sa0/g
David E. O'Brien [Mon, 1 Jan 2001 19:50:48 +0000 (19:50 +0000)]
s/rsa0/sa0/g

23 years agoThis is not necessarily the correct fix, but at least sbin/init compiles
Poul-Henning Kamp [Mon, 1 Jan 2001 19:46:43 +0000 (19:46 +0000)]
This is not necessarily the correct fix, but at least sbin/init compiles
in a sterile environment like "make release"

23 years agoCheck that the hostname field in received packets is NUL-terminated.
Ben Smithurst [Mon, 1 Jan 2001 18:43:21 +0000 (18:43 +0000)]
Check that the hostname field in received packets is NUL-terminated.

23 years agoSanity check ptr for legal values so it is less likely
Matt Jacob [Mon, 1 Jan 2001 16:49:31 +0000 (16:49 +0000)]
Sanity check ptr for legal values so it is less likely
(but not impossible) to get stuck in an infinite loop.

Obtained from: msmith@freebsd.org

23 years agoChange prototype to getlogin_r to reflect recent change to its
Daniel Eischen [Mon, 1 Jan 2001 13:35:04 +0000 (13:35 +0000)]
Change prototype to getlogin_r to reflect recent change to its
implementation (conformance to '96 POSIX standard).

23 years agoChange the interface of getlogin_r to return an int. The former
Daniel Eischen [Mon, 1 Jan 2001 13:29:19 +0000 (13:29 +0000)]
Change the interface of getlogin_r to return an int.  The former
interface was based on a draft version of POSIX whereas the final
(1996) version of POSIX specified that the error is returned.

While I'm here, fix getlogin_r so that it works for more than just
the first time it's called.

Reviewed by: wes, wollman (man page)

23 years agoAdd a missing vnode_if.h.
Poul-Henning Kamp [Mon, 1 Jan 2001 11:28:32 +0000 (11:28 +0000)]
Add a missing vnode_if.h.

Submitted by: Jim Bloom <bloom@acm.org>

23 years agoWe are now in the year 2001, so ajust copyrights.
Maxim Sobolev [Mon, 1 Jan 2001 08:05:48 +0000 (08:05 +0000)]
We are now in the year 2001, so ajust copyrights.

Happy New Year!

23 years agoMake it a bit clearer that asprintf doesn't actually "return" a pointer in
Ben Smithurst [Mon, 1 Jan 2001 05:19:52 +0000 (05:19 +0000)]
Make it a bit clearer that asprintf doesn't actually "return" a pointer in
the normal sense of the word, but does it through one of its arguments which
is a pointer to a pointer.

PR: 23717
Submitted by: phk

23 years agoPush down sched_lock in psignal(). sched_lock was being held across
John Baldwin [Mon, 1 Jan 2001 02:31:08 +0000 (02:31 +0000)]
Push down sched_lock in psignal().  sched_lock was being held across
recursive calls into psignal() as well as calls to signotify(),
forward_signal(), etc.

23 years agoru_SU -> ru_RU for nls too
Andrey A. Chernov [Mon, 1 Jan 2001 02:19:59 +0000 (02:19 +0000)]
ru_SU -> ru_RU for nls too

23 years agoAdd in a missing release of the proctree lock.
John Baldwin [Mon, 1 Jan 2001 02:19:51 +0000 (02:19 +0000)]
Add in a missing release of the proctree lock.

Submitted by: Sja <sakari.jalovaara@eqonline.fi>

23 years agoAdd a dopey makefile to do the tags dance, which is probably not perfect
Matt Jacob [Sun, 31 Dec 2000 23:26:34 +0000 (23:26 +0000)]
Add a dopey makefile to do the tags dance, which is probably not perfect
but is better than nothing by a good deal.

23 years agothere is no more miscfs/devfs
Matt Jacob [Sun, 31 Dec 2000 23:12:20 +0000 (23:12 +0000)]
there is no more miscfs/devfs

23 years agoFix problems with incomplete conversions from printf to isp_prt.
Matt Jacob [Sun, 31 Dec 2000 20:50:56 +0000 (20:50 +0000)]
Fix problems with incomplete conversions from printf to isp_prt.

23 years agoOops, I'm such an idiot. For reasons I don't really agree with,
Matt Jacob [Sun, 31 Dec 2000 20:29:56 +0000 (20:29 +0000)]
Oops, I'm such an idiot. For reasons I don't really agree with,
all devices are by default known by their 'cooked' name, so
my change was wrong. I thought it was a hangover from old 'block
tape device' support which hasn't worked (if it ever did) since
v6/PWB.

So, the default tape name is now the same as Linux. Far out, man....

23 years agoIn call to realloc, pass the number of bytes needed, not simply the
Guy Helmer [Sun, 31 Dec 2000 16:15:06 +0000 (16:15 +0000)]
In call to realloc, pass the number of bytes needed, not simply the
number of login time structures.

Forward the name of the deny capability rather than hard-coding it
in login_hostok.

23 years ago... and ru_RU.KOI8-R.
Poul-Henning Kamp [Sun, 31 Dec 2000 15:08:14 +0000 (15:08 +0000)]
... and ru_RU.KOI8-R.

Remove ru_SU

Forgotten by: phantom on 2000/12/29 14:58:01

23 years agoAdd ru_RU.ISO_8859-5 also.
Poul-Henning Kamp [Sun, 31 Dec 2000 15:06:08 +0000 (15:06 +0000)]
Add ru_RU.ISO_8859-5 also.

23 years agoThis is the first snapshot of the new all-singing-and-dancing md(4).
Poul-Henning Kamp [Sun, 31 Dec 2000 13:03:42 +0000 (13:03 +0000)]
This is the first snapshot of the new all-singing-and-dancing md(4).

Using the mdconfig(8) program you can now configure memory disks
on malloc(9), swap or a file/vnode.  preloaded md disks also work
as usual.

23 years agoAdd forgotten entry for ru_RU.CP866 to try to make make release work.
Poul-Henning Kamp [Sun, 31 Dec 2000 12:02:55 +0000 (12:02 +0000)]
Add forgotten entry for ru_RU.CP866 to try to make make release work.

23 years agoUse macro API to <sys/queue.h>
Poul-Henning Kamp [Sun, 31 Dec 2000 11:22:42 +0000 (11:22 +0000)]
Use macro API to <sys/queue.h>

Submitted by: Dima Dorfman <dima@unixfreak.org>
Reviewed by: phk

23 years agoImprove cmdline parsing and messages.
Poul-Henning Kamp [Sun, 31 Dec 2000 11:20:49 +0000 (11:20 +0000)]
Improve cmdline parsing and messages.

Add a rudimentary man-page.  (I'm no mdoc wizard, feel free to attack my
mistakes and blunders)

23 years agoUse macro API to <sys/queue.h>
Poul-Henning Kamp [Sun, 31 Dec 2000 10:28:01 +0000 (10:28 +0000)]
Use macro API to <sys/queue.h>

Submitted by: "Peter Avalos" <pavalos@theshell.com>
Reviewed by: /sbin/md5 [*]

[*] This line appears courtesy of Mr. Warner Losch, all rights reversed.

23 years agoUse macro API to <sys/queue.h>
Poul-Henning Kamp [Sun, 31 Dec 2000 10:24:19 +0000 (10:24 +0000)]
Use macro API to <sys/queue.h>

23 years agoAdd compress option at the instigation of rgrimes.
Poul-Henning Kamp [Sun, 31 Dec 2000 10:22:52 +0000 (10:22 +0000)]
Add compress option at the instigation of rgrimes.

23 years agoThe default is /dev/nrsa0 (now), not /dev/nsa0.
Matt Jacob [Sun, 31 Dec 2000 08:17:58 +0000 (08:17 +0000)]
The default is /dev/nrsa0 (now), not /dev/nsa0.

23 years agoUnbelievable how this slipped through. There is
Matt Jacob [Sun, 31 Dec 2000 08:17:19 +0000 (08:17 +0000)]
Unbelievable how this slipped through. There is
no such device as /dev/nsa0. The DEFTAPE should
"/dev/nrsa0". *Groan*.

23 years agoIf the first argument doesn't start with '-' assume that it is an
Warner Losh [Sun, 31 Dec 2000 01:51:23 +0000 (01:51 +0000)]
If the first argument doesn't start with '-' assume that it is an
interface.  This augments the default to an appropriate interface
code.

# These programs should be merged into ifconfig, ala NetBSD, but that's
# a fight for another day.

Idea from: OpenBSD

23 years agoIgnore a net interrupt if the corresponding handler is not
Seigo Tanimura [Sun, 31 Dec 2000 01:31:55 +0000 (01:31 +0000)]
Ignore a net interrupt if the corresponding handler is not
registered.

This fixes panic on my laptop where a spurious arp packet
is received when arp is not ready to run.

23 years agoBackout rev 1.57 & 1.58. While the previous revisions fixed
Paul Saab [Sun, 31 Dec 2000 01:30:27 +0000 (01:30 +0000)]
Backout rev 1.57 & 1.58.  While the previous revisions fixed
attaching to running processes, it completely breaks normal debugging.
A better fix is in the works, but cannot be properly tested until
the problem with gdb hanging the system in -current is solved.

23 years agoImply -c when -S is specified, so 'vnconfig -s reserve -S 4 vn1' configures the
Matthew Dillon [Sun, 31 Dec 2000 01:27:55 +0000 (01:27 +0000)]
Imply -c when -S is specified, so 'vnconfig -s reserve -S 4 vn1' configures the
device as expected.

23 years agoFix a lockup problem that occurs with 'cvs update'. specfs's fsync can
Matthew Dillon [Sat, 30 Dec 2000 23:32:24 +0000 (23:32 +0000)]
Fix a lockup problem that occurs with 'cvs update'.  specfs's fsync can
get into the same sort of infinite loop that ffs's fsync used to get
into, probably due to background bitmap writes.  The solution is
the same.

23 years agoUse macro API for <sys/queue.h>
Poul-Henning Kamp [Sat, 30 Dec 2000 22:06:19 +0000 (22:06 +0000)]
Use macro API for <sys/queue.h>

23 years agoMy bad, committed the submitted patch rather than the fixed patch.
Poul-Henning Kamp [Sat, 30 Dec 2000 21:56:18 +0000 (21:56 +0000)]
My bad, committed the submitted patch rather than the fixed patch.

23 years agoUse macro API to <sys/queue.h>
Ben Smithurst [Sat, 30 Dec 2000 21:52:34 +0000 (21:52 +0000)]
Use macro API to <sys/queue.h>

23 years agoUse macro API to <sys/queue.h>
Poul-Henning Kamp [Sat, 30 Dec 2000 21:22:54 +0000 (21:22 +0000)]
Use macro API to <sys/queue.h>

Submitted by: "Jason" <jsmethers@pdq.net>
Reviewed by: phk

23 years agoUse macro API to <sys/queue.h>
Poul-Henning Kamp [Sat, 30 Dec 2000 21:17:03 +0000 (21:17 +0000)]
Use macro API to <sys/queue.h>

Submitted by: "Jason" <jsmethers@pdq.net>
Reviewed by: phk

23 years agoUse macro API to <sys/queue.h>
Poul-Henning Kamp [Sat, 30 Dec 2000 21:05:45 +0000 (21:05 +0000)]
Use macro API to <sys/queue.h>

Submitted by: "Peter Avalos" <pavalos@theshell.com>
Reviewed by: /sbin/md5

23 years agoUse macro API to <sys/queue.h>
Poul-Henning Kamp [Sat, 30 Dec 2000 21:02:25 +0000 (21:02 +0000)]
Use macro API to <sys/queue.h>

Submitted by: "Jason" <jsmethers@pdq.net>
Reviewed by: /sbin/md5

23 years agoUse macro API to <sys/queue.h>
Poul-Henning Kamp [Sat, 30 Dec 2000 20:56:04 +0000 (20:56 +0000)]
Use macro API to <sys/queue.h>

Submitted by: "Jason" <jsmethers@pdq.net>
Reviewed by: /sbin/md5

23 years agoUse <sys/queue.h> macro API.
Poul-Henning Kamp [Sat, 30 Dec 2000 20:40:16 +0000 (20:40 +0000)]
Use <sys/queue.h> macro API.

Submitted by: "Jason" <jsmethers@pdq.net>
Reviewed by: /sbin/md5

23 years agoSync up the list of supported AMI MegaRAID controllers with reality.
Bruce A. Mah [Sat, 30 Dec 2000 20:11:55 +0000 (20:11 +0000)]
Sync up the list of supported AMI MegaRAID controllers with reality.

Submitted by: msmith

23 years agoChange the modification of what could be a const string. Apparently the
Matt Jacob [Sat, 30 Dec 2000 20:09:26 +0000 (20:09 +0000)]
Change the modification of what could be a const string. Apparently the
construct:

char *foo;
...
foo = "XXX";
...
foo[1] = 'Y';

is wrong. IT blew up on NetBSD-sparc64 because that platform write-protects
constant strings.

23 years agoRemove comma after final .Nm entry before .Nd
Ben Smithurst [Sat, 30 Dec 2000 18:58:15 +0000 (18:58 +0000)]
Remove comma after final .Nm entry before .Nd

23 years agoRemove CIRCLEQ manual page links now that CIRCLEQ has been axed.
Ben Smithurst [Sat, 30 Dec 2000 18:54:35 +0000 (18:54 +0000)]
Remove CIRCLEQ manual page links now that CIRCLEQ has been axed.