]> CyberLeo.Net >> Repos - FreeBSD/FreeBSD.git/log
FreeBSD/FreeBSD.git
23 years ago"$FreeBSD $" -> "$FreeBSD$" (the space causes CVS to think this isn't
dd [Sat, 2 Jun 2001 23:12:23 +0000 (23:12 +0000)]
"$FreeBSD $" -> "$FreeBSD$" (the space causes CVS to think this isn't
a keyword, so it wasn't expanding it)

23 years agoOnly touch the PCR register in order to set bits for the fxp driver.
jlemon [Sat, 2 Jun 2001 19:51:02 +0000 (19:51 +0000)]
Only touch the PCR register in order to set bits for the fxp driver.
The 3C509-TX card apparently had a slightly different version of the
chip, and has problems when this register is set.  The problem does
not appear on the 3C509{BC} cards, but since only the fxp driver needs
specific bits set, conditionalize on that.

23 years agoRemove double //'s in path.
obrien [Sat, 2 Jun 2001 19:09:37 +0000 (19:09 +0000)]
Remove double //'s in path.

23 years agoSupport GCC 3.0
obrien [Sat, 2 Jun 2001 17:35:50 +0000 (17:35 +0000)]
Support GCC 3.0

23 years agoadded Sergio de Souza Prallon (prallon@tmp.com.br), the author of
hm [Sat, 2 Jun 2001 12:27:37 +0000 (12:27 +0000)]
added Sergio de Souza Prallon (prallon@tmp.com.br), the author of
the itjc hardware driver.

23 years agoupdate the isdn4bsd general documentation
hm [Sat, 2 Jun 2001 12:21:16 +0000 (12:21 +0000)]
update the isdn4bsd general documentation

23 years agobring the roadmap to isdn4bsd up to date.
hm [Sat, 2 Jun 2001 11:48:50 +0000 (11:48 +0000)]
bring the roadmap to isdn4bsd up to date.

23 years agoupdate the postcard list
hm [Sat, 2 Jun 2001 11:31:33 +0000 (11:31 +0000)]
update the postcard list

23 years agoProperly wrap mtx_intr_enable() macro in "do $bla while (0)"
phk [Sat, 2 Jun 2001 08:17:42 +0000 (08:17 +0000)]
Properly wrap mtx_intr_enable() macro in "do $bla while (0)"

23 years agoMerged from sys/i386/i386/machdep.c revision 1.454.
nyan [Sat, 2 Jun 2001 06:30:56 +0000 (06:30 +0000)]
Merged from sys/i386/i386/machdep.c revision 1.454.

23 years agoDisabled unneeded code for PC98.
nyan [Sat, 2 Jun 2001 05:51:21 +0000 (05:51 +0000)]
Disabled unneeded code for PC98.

Submitted by: NOKUBI Hirotaka <nokubi@ff.iij4u.or.jp>

23 years agoMerged from sys/i386/isa/npx.c revision 1.101.
nyan [Sat, 2 Jun 2001 05:00:08 +0000 (05:00 +0000)]
Merged from sys/i386/isa/npx.c revision 1.101.

23 years agoARP works on networks other than Ethernet.
dd [Sat, 2 Jun 2001 04:09:53 +0000 (04:09 +0000)]
ARP works on networks other than Ethernet.

PR: 22062
Submitted by: Christian Weisgerber <naddy@mips.inka.de>,
Brooks Davis <brooks@one-eyed-alien.net>

23 years agoMention the 'ether' parameter.
dd [Sat, 2 Jun 2001 04:05:51 +0000 (04:05 +0000)]
Mention the 'ether' parameter.

PR: 23767
Submitted by: Maxime Henrion <mux@qualys.com>,
Brooks Davis <brooks@one-eyed-alien.net>

23 years agoMention the kern.ps_showallprocs sysctl.
dd [Sat, 2 Jun 2001 04:02:36 +0000 (04:02 +0000)]
Mention the kern.ps_showallprocs sysctl.

PR: 24804
Submitted by: Mike Meyer <mwm@mired.org>

23 years agoUse S_IXUSR instead of hardcoded values.
mikeh [Sat, 2 Jun 2001 03:10:07 +0000 (03:10 +0000)]
Use S_IXUSR instead of hardcoded values.

Reviewed by: md5

23 years agoRemove this file that slipped thru FREEBSD-Xlist's actions.
obrien [Sat, 2 Jun 2001 00:03:06 +0000 (00:03 +0000)]
Remove this file that slipped thru FREEBSD-Xlist's actions.

23 years agoChange this to support the new way swap device information is exported
tmm [Fri, 1 Jun 2001 22:57:07 +0000 (22:57 +0000)]
Change this to support the new way swap device information is exported
via sysctl, and clean up some style and (size_t != int) issues.

23 years agoChange the way information about swap devices is exported to be more
tmm [Fri, 1 Jun 2001 22:53:10 +0000 (22:53 +0000)]
Change the way information about swap devices is exported to be more
canonical: define a versioned struct xswdev, and add a sysctl node
handler that allows the user to get this structure for a certain device
index by specifying this index as last element of the MIB.
This new node handler, vm.swap_info, replaces the old vm.nswapdev
and vm.swapdevX.* (where X was the index) sysctls.

23 years agoFix typo (opt_acc.h -> opt_aac.h) for AAC_COMPAT_LINUX.
dirk [Fri, 1 Jun 2001 22:39:50 +0000 (22:39 +0000)]
Fix typo (opt_acc.h -> opt_aac.h) for AAC_COMPAT_LINUX.

23 years agoFix another unprotected instance of chdir() by extending the
kris [Fri, 1 Jun 2001 21:53:50 +0000 (21:53 +0000)]
Fix another unprotected instance of chdir() by extending the
fts_safe_changedir() function and using that instead for both of the
chdir()s.

Partially submitted by: Todd Miller <millert@OpenBSD.org>, bde

23 years agoRevert the last bits of my bogus move of NMBCLUSTERS
jesper [Fri, 1 Jun 2001 21:47:34 +0000 (21:47 +0000)]
Revert the last bits of my bogus move of NMBCLUSTERS
to <sys/param.h>

23 years agoMan page for orm driver.
imp [Fri, 1 Jun 2001 21:02:18 +0000 (21:02 +0000)]
Man page for orm driver.

PR: 22078
Submitted by: Nikolai Saoukh <nms@otdel-1.org>

23 years agoThe orm device. This device gobbles up the Option ROMs in the ISA
imp [Fri, 1 Jun 2001 20:58:32 +0000 (20:58 +0000)]
The orm device.  This device gobbles up the Option ROMs in the ISA
memory I/O space.  Otherwise, our resource allocation system might
mistakenly assign pccard, plug and play devices or other things
addresses that conflict with ROMs.

I cleaned up his code a little from the submited driver: style(9)
issues, commentary on why something that looks incorrect really is
correct.  Also noted that while a checksum field is defined for the
ROMs, enough common hardware neglects it to make it not worthwhile
checking.

Submitted by: Nikolai Saoukh <nms@otdel-1.org>
PR: 22078

23 years agoResurrect platform.pci_intr_map() and essentially undo the effects of
gallatin [Fri, 1 Jun 2001 17:39:11 +0000 (17:39 +0000)]
Resurrect platform.pci_intr_map() and essentially undo the effects of
the interface conversion to platform.pci_intr_route().  I've left the
platform.pci_intr_route() function pointer in place, as well as
alpha_pci_route_interrupt(), but no platform currently implements it.

To work around the removal of alpha_platform_assign_pciintr(cfg);
from the pci probe code, I've hooked in calls to platform.pci_intr_map()
in pcib_read_config (similar to the x86 APIC_IO ifdef in pci_cfgregread)
for every chipset that has a platform which needs it.

While here, I've removed the interupt mapping/routing code from the
AS2x00 platform because its not required (it has never been present in
-stable).

Tested on: UP1000, Miata(GL), XP1000, AS2100, AS500

23 years agoSmall adjustment to whitespace in output
dougb [Fri, 1 Jun 2001 16:40:57 +0000 (16:40 +0000)]
Small adjustment to whitespace in output

23 years agoNew release notes: lge(4) driver (and update hardware compatability
bmah [Fri, 1 Jun 2001 16:29:21 +0000 (16:29 +0000)]
New release notes:  lge(4) driver (and update hardware compatability
list from manpage), RANDOM_IP_ID, msdos->msdosfs, chown(8) doesn't
use "." for a delimiter anymore.

23 years agoSupport /dev/tun cloning. Ansify if_tun.c while I'm there.
brian [Fri, 1 Jun 2001 15:51:10 +0000 (15:51 +0000)]
Support /dev/tun cloning.  Ansify if_tun.c while I'm there.

Only tun0 -> tun32767 may now be opened as struct ifnet's if_unit
is a short.

It's now possible to open /dev/tun and get a handle back for an available
tun device (use devname to find out what you got).

The implementation uses rman by popular demand (and against my judgement)
to track opened devices and uses the new dev_depends() to ensure that
all make_dev()d devices go away before the module is unloaded.

Reviewed by: phk

23 years agoCopy the sockaddr from the netbuf data area, not from the netbuf
iedowse [Fri, 1 Jun 2001 15:20:45 +0000 (15:20 +0000)]
Copy the sockaddr from the netbuf data area, not from the netbuf
data pointer. This bug has been here since the ti-rpc import; it
apparently broke the clnt_control CLGET_SVC_ADDR options.

PR: misc/27813
Submitted by: Jean-Luc Richier <Jean-Luc.Richier@imag.fr>

23 years agoDon't treat the pipe symbol in capabilities as an alternate
ru [Fri, 1 Jun 2001 14:45:00 +0000 (14:45 +0000)]
Don't treat the pipe symbol in capabilities as an alternate
record name indicator; this causes a spurious warning:

$ cat x
record:\
:capability=|value:
$ cap_mkdb x
cap_mkdb: ignored duplicate: record

Spotted by: dcs

23 years agoDo retry next entry in addrinfo during connect.
ume [Fri, 1 Jun 2001 13:43:20 +0000 (13:43 +0000)]
Do retry next entry in addrinfo during connect.

MFC after: 1 week

23 years agoThis file is not needed any more, the definitions and declarations it
tmm [Fri, 1 Jun 2001 13:33:30 +0000 (13:33 +0000)]
This file is not needed any more, the definitions and declarations it
contained have been moved to i386/isa/icu.h and sys/interrupt.h.

23 years agoReplace a use of the hw.nintr sysctl as it has just gone away, cast
tmm [Fri, 1 Jun 2001 13:29:19 +0000 (13:29 +0000)]
Replace a use of the hw.nintr sysctl as it has just gone away, cast
size_t variables when passing them to a printf-like function, and some
minor cleanups.

23 years agoClean up the code exporting interrupt statistics via sysctl a bit:
tmm [Fri, 1 Jun 2001 13:23:28 +0000 (13:23 +0000)]
Clean up the code exporting interrupt statistics via sysctl a bit:
- move the sysctl code to kern_intr.c
- do not use INTRCNT_COUNT, but rather eintrcnt - intrcnt to determine
  the length of the intrcnt array
- move the declarations of intrnames, eintrnames, intrcnt and eintrcnt
  from machine-dependent include files to sys/interrupt.h
- remove the hw.nintr sysctl, it is not needed.
- fix various style bugs

Requested by: bde
Reviewed by: bde (some time ago)

23 years agoAdd a quirk entry for ARCHIVE Python 06408.
non [Fri, 1 Jun 2001 12:36:24 +0000 (12:36 +0000)]
Add a quirk entry for ARCHIVE Python 06408.

Approved by: mjacob

23 years agomsdos -> msdosfs.
ru [Fri, 1 Jun 2001 12:16:09 +0000 (12:16 +0000)]
msdos -> msdosfs.

23 years agomount_msdos(8) -> mount_msdosfs(8).
ru [Fri, 1 Jun 2001 12:06:06 +0000 (12:06 +0000)]
mount_msdos(8) -> mount_msdosfs(8).

23 years agomount_msdos(8) -> mount_msdosfs(8).
ru [Fri, 1 Jun 2001 11:25:40 +0000 (11:25 +0000)]
mount_msdos(8) -> mount_msdosfs(8).

23 years ago- VFS_SET(msdos) -> VFS_SET(msdosfs)
ru [Fri, 1 Jun 2001 10:57:26 +0000 (10:57 +0000)]
- VFS_SET(msdos) -> VFS_SET(msdosfs)
- msdos.ko -> msdosfs.ko
- mount_msdos(8) -> mount_msdosfs(8)
- "msdos" -> "msdosfs" compatibility glue in mount(8)

23 years agoDon't try to print a completion percentage for zero-size files.
des [Fri, 1 Jun 2001 10:24:58 +0000 (10:24 +0000)]
Don't try to print a completion percentage for zero-size files.

PR: 27780
Submitted by: Craig Leres <leres@ee.lbl.gov>

23 years agoRemove vestiges of MFS.
ru [Fri, 1 Jun 2001 10:07:28 +0000 (10:07 +0000)]
Remove vestiges of MFS.

23 years agoAdd ``options RANDOM_IP_ID'' which randomizes the ID field of IP packets.
kris [Fri, 1 Jun 2001 10:02:28 +0000 (10:02 +0000)]
Add ``options RANDOM_IP_ID'' which randomizes the ID field of IP packets.
This closes a minor information leak which allows a remote observer to
determine the rate at which the machine is generating packets, since the
default behaviour is to increment a counter for each packet sent.

Reviewed by:    -net
Obtained from:  OpenBSD

23 years agoBoth X11 (3.x and 4.x) currently lack a libdata/ subdirectory,
obrien [Fri, 1 Jun 2001 09:56:15 +0000 (09:56 +0000)]
Both X11 (3.x and 4.x) currently lack a libdata/ subdirectory,
(which is present both in /usr and /usr/local) which is causing a few
ports to erroneously report mtree failures on bento.

Submitted by: ade

23 years agoBack out jesper's 2001/05/31 14:58:11 PDT commit. It does not compile.
obrien [Fri, 1 Jun 2001 09:51:14 +0000 (09:51 +0000)]
Back out jesper's 2001/05/31 14:58:11 PDT commit.  It does not compile.

23 years agoMinor style reformatting.
obrien [Fri, 1 Jun 2001 05:31:02 +0000 (05:31 +0000)]
Minor style reformatting.

23 years agoUpgrade to Binutils 2.11.0.
obrien [Fri, 1 Jun 2001 05:30:31 +0000 (05:30 +0000)]
Upgrade to Binutils 2.11.0.

23 years agoChange plxic to plxcard, per phk. He thnks plxic is too generic a
imp [Fri, 1 Jun 2001 05:20:38 +0000 (05:20 +0000)]
Change plxic to plxcard, per phk.  He thnks plxic is too generic a
name.  I didn't do repo magic because this is so new.

23 years agouucp.daemon -> uucp:daemon
imp [Fri, 1 Jun 2001 04:53:13 +0000 (04:53 +0000)]
uucp.daemon -> uucp:daemon

23 years agoRemove unused variable (descr) in show_var().
dd [Fri, 1 Jun 2001 02:58:09 +0000 (02:58 +0000)]
Remove unused variable (descr) in show_var().

PR: 22582
Submitted by: Giorgos Keramidas <charon@gray.westgate.gr>

23 years agoCorrect a typo.
dd [Fri, 1 Jun 2001 02:53:32 +0000 (02:53 +0000)]
Correct a typo.

PR: 22885
Submitted by: Rene de Vries <rene@tunix.nl>

23 years agoCorrect some typos.
dd [Fri, 1 Jun 2001 02:28:15 +0000 (02:28 +0000)]
Correct some typos.

PR: 27251
Submitted by: Pekka Savola <pekkas@netcore.fi>

23 years agoFix a spacing issue. (No content changes.)
dd [Fri, 1 Jun 2001 01:36:03 +0000 (01:36 +0000)]
Fix a spacing issue.  (No content changes.)

23 years agoUnlock the process returned from pfind() if it does not return NULL.
jake [Fri, 1 Jun 2001 01:30:51 +0000 (01:30 +0000)]
Unlock the process returned from pfind() if it does not return NULL.
This fixes a witness lock violation for nfssvc returning with locks
held.

Submitted by: Jean-Luc Richier <Jean-Luc.Richier@imag.fr>
PR: kern/27776

23 years agoDon't install setgid kmem.
dd [Fri, 1 Jun 2001 01:14:28 +0000 (01:14 +0000)]
Don't install setgid kmem.

23 years agoGrrr. Fix PR 27742 correctly this time. (At least I got -stable right.)
wpaul [Fri, 1 Jun 2001 00:34:50 +0000 (00:34 +0000)]
Grrr. Fix PR 27742 correctly this time. (At least I got -stable right.)

23 years agoFix how /bin/sh handles 'for' and 'case' statements when it is called to do
gad [Fri, 1 Jun 2001 00:07:09 +0000 (00:07 +0000)]
Fix how /bin/sh handles 'for' and 'case' statements when it is called to do
errexit (-e) processing.  This solves a problem where 'make clean' would
fail with an unspecified error in certain automake-generated makefiles.

Reviewed by: no objections from -hackers...
MFC after: 2 weeks

23 years agoremove unneeded #include <vm/vm_map.h>
gallatin [Thu, 31 May 2001 23:31:34 +0000 (23:31 +0000)]
remove unneeded #include <vm/vm_map.h>

23 years agoCast size_t variables before printing them to cope with the case of
tmm [Thu, 31 May 2001 22:36:51 +0000 (22:36 +0000)]
Cast size_t variables before printing them to cope with the case of
sizeof(size_t) != sizeof(int).

23 years agoDocument the half-supported WITHOUT_X.
dd [Thu, 31 May 2001 22:36:35 +0000 (22:36 +0000)]
Document the half-supported WITHOUT_X.

PR: 27709
Submitted by: Mike Meyer <mwm@mired.org>

23 years agoConvert to use the sysctl interface to obtain the data when operating
dd [Thu, 31 May 2001 22:30:39 +0000 (22:30 +0000)]
Convert to use the sysctl interface to obtain the data when operating
on a running system.  Using kvm is still supported, primarily for
postmortem analysis.

Reviewed by: tmm

23 years agoClose PR #27742: allow the xl driver to receive VLAN tagged frames by
wpaul [Thu, 31 May 2001 22:08:45 +0000 (22:08 +0000)]
Close PR #27742: allow the xl driver to receive VLAN tagged frames by
setting the 'max packet size' register in window 3. This only
works for cards based on the cyclone or newer chipsets (i.e. it
won't work with the original 3c905/boomerang cards).

There is a trick which will work with the boomerang, which is to turn
on the 'large packets ok' bit in the MAC control register, however this
lets the chip accept any frame up to 4K in length, which is larger than
the mbuf cluster buffers we use to receive frames. If somebody sends us
such a frame and the chip DMAs it to us, it could write past the end
of the cluster buffer and clobber something.

PR: kern/27742

23 years agoMove the definition of NMBCLUSTERS from src/sys/kern/uipc_mbuf.c
jesper [Thu, 31 May 2001 21:58:10 +0000 (21:58 +0000)]
Move the definition of NMBCLUSTERS from src/sys/kern/uipc_mbuf.c
to <sys/param.h>, so it's available to src/sys/netinet/ip_input.c

MFC after: 1 week

23 years agoChange the default value of net.inet6.ip6.maxfragpackets from
jesper [Thu, 31 May 2001 21:57:53 +0000 (21:57 +0000)]
Change the default value of net.inet6.ip6.maxfragpackets from
200 to NMBCLUSTERS/4 to match the IPv4 case.

MFC after: 1 week

23 years agoPrevent denial of service using bogus fragmented IPv4 packets.
jesper [Thu, 31 May 2001 21:57:29 +0000 (21:57 +0000)]
Prevent denial of service using bogus fragmented IPv4 packets.

A attacker sending a lot of bogus fragmented packets to the target
(with different IPv4 identification field - ip_id), may be able
to put the target machine into mbuf starvation state.

By setting a upper limit on the number of reassembly queues we
prevent this situation.

This upper limit is controlled by the new sysctl
net.inet.ip.maxfragpackets which defaults to NMBCLUSTERS/4

If you want old behaviour (no upper limit) set this sysctl
to a negative value.

If you don't want to accept any fragments (not recommended)
set the sysctl to 0 (zero)

Obtained from: NetBSD (partially)
MFC after: 1 week

23 years agoMove the definition of NMBCLUSTERS from src/sys/kern/uipc_mbuf.c
jesper [Thu, 31 May 2001 21:56:44 +0000 (21:56 +0000)]
Move the definition of NMBCLUSTERS from src/sys/kern/uipc_mbuf.c
to <sys/param.h>, so it's available to src/sys/netinet/ip_input.c,
and remove the now unneeded includes of "opt_param.h".

MFC after: 1 week

23 years agoBackout previous revision. While it fixed many platforms, it broke
gallatin [Thu, 31 May 2001 21:47:25 +0000 (21:47 +0000)]
Backout previous revision.  While it fixed many platforms, it broke
all alphas with devices behind ppb's.  I'm working on a better solution now.

Note that all alphas that use per-platform interrupt mapping are broken
again (as they have been for several months)

23 years agoAdd device driver support for the Level 1 LXT1001 NetCellerator
wpaul [Thu, 31 May 2001 21:44:26 +0000 (21:44 +0000)]
Add device driver support for the Level 1 LXT1001 NetCellerator
gigabit ethernet controller chip. This device is used on some
fiber optic gigE cards from SMC, D-Link and Addtron. Jumbograms and
TCP/IP checksum offload on receive are supported. Hardware VLAN
filtering is not, because it doesn't play well with our existing
VLAN code. Also add manual page.

There is a 4.x version of this driver available at
http://www.freebsd.org/~wpaul/Level1/4.x if anyone feels adventurous
and wants to test it. I still need to do performance testing and
tuning with this device.

(For my next trick, I will make the 3Com 3cR990 sit up and beg.)

23 years agoStarting with 5.0-R, do not support `.' as a owner / group delimiter.
obrien [Thu, 31 May 2001 19:30:19 +0000 (19:30 +0000)]
Starting with 5.0-R, do not support `.' as a owner / group delimiter.

23 years agoDisable rfc1323 and rfc1644 TCP extensions if we havn't got
jesper [Thu, 31 May 2001 19:24:49 +0000 (19:24 +0000)]
Disable rfc1323 and rfc1644 TCP extensions if we havn't got
any response to our third SYN to work-around some broken
terminal servers (most of which have hopefully been retired)
that have bad VJ header compression code which trashes TCP
segments containing unknown-to-them TCP options.

PR: kern/1689
Submitted by: jesper
Reviewed by: wollman
MFC after: 2 weeks

23 years agoCross reference pkg_update(1).
eric [Thu, 31 May 2001 19:15:25 +0000 (19:15 +0000)]
Cross reference pkg_update(1).

Approved by: murray

23 years agoplxic device
imp [Thu, 31 May 2001 19:06:46 +0000 (19:06 +0000)]
plxic device

23 years agoAdd a simple plx pci9052 based pccard bridges. This doesn't work yet,
imp [Thu, 31 May 2001 19:04:29 +0000 (19:04 +0000)]
Add a simple plx pci9052 based pccard bridges.  This doesn't work yet,
but I'll be fleshing this out as I have time.  This should mean we no
longer need to have an and wi pci attachments, but that's a ways off.

23 years agoCleanup TurboLaser a bit. Add non-Adaptec statement for Miata SRM.
wilko [Thu, 31 May 2001 18:54:06 +0000 (18:54 +0000)]
Cleanup TurboLaser a bit. Add non-Adaptec statement for Miata SRM.

23 years agoSupport a range of registers to read.
imp [Thu, 31 May 2001 18:31:43 +0000 (18:31 +0000)]
Support a range of registers to read.
eg pciconf -r pci0:10:0 0:0xff
and keep the output of the old singleton the same.

Reviewed by: audit@, dd

MFC after: 10 days

23 years agoChange the ownership of the symbolic link for hidden game
ru [Thu, 31 May 2001 15:05:05 +0000 (15:05 +0000)]
Change the ownership of the symbolic link for hidden game
rather than /usr/games/dm wrapper.

Spotted by: bde
Reviewed by: bde

23 years agoRemove commented out line that doesn't belong here.
sobomax [Thu, 31 May 2001 14:42:04 +0000 (14:42 +0000)]
Remove commented out line that doesn't belong here.

23 years agoFix typos and modify obscure wording
tom [Thu, 31 May 2001 11:51:20 +0000 (11:51 +0000)]
Fix typos and modify obscure wording

MFC after: 5

23 years agoFixed the bug from the previous revision.
ru [Thu, 31 May 2001 11:47:20 +0000 (11:47 +0000)]
Fixed the bug from the previous revision.

``chown -h owner symlink'' did not set the symlink's owner
if the file the symlink points to already had that owner:

# ls -l alink afile
-rw-r--r--  1 nobody  ru  0 May 31 14:14 afile
lrwxr-xr-x  1 root    ru  5 May 31 14:14 alink -> afile
# ./chown -h -v nobody alink
# ls -l alink afile
-rw-r--r--  1 nobody  ru  0 May 31 14:14 afile
lrwxr-xr-x  1 root    ru  5 May 31 14:14 alink -> afile

Similarly for chgrp(1) and chmod(1).

23 years agoPERL_THREADED is too experimental at this stage. Remove.
markm [Thu, 31 May 2001 11:19:00 +0000 (11:19 +0000)]
PERL_THREADED is too experimental at this stage. Remove.

23 years agoFor the time being, Perl threading is considered dangerous. How, when
markm [Thu, 31 May 2001 11:04:53 +0000 (11:04 +0000)]
For the time being, Perl threading is considered dangerous. How, when
and why this is re-enabled will be addressed again in the future.

23 years agoRecently, other BSDs had faith support in inetd. Though our inetd has
ume [Thu, 31 May 2001 10:09:36 +0000 (10:09 +0000)]
Recently, other BSDs had faith support in inetd.  Though our inetd has
it already, their syntax is not compatible with ours.  It will confuse
users.  So, we have compatibility with their syntex.

Approved by: dwmalone
Obtained from: NetBSD

23 years agoTruly limit the path to local filesystems.
dougb [Thu, 31 May 2001 09:53:53 +0000 (09:53 +0000)]
Truly limit the path to local filesystems.

23 years agoDisplay -tcp w/o load average so that all statistics fits on a 80x25 screen.
ru [Thu, 31 May 2001 07:14:52 +0000 (07:14 +0000)]
Display -tcp w/o load average so that all statistics fits on a 80x25 screen.

23 years agoShow TCP checksum failures with -tcp display.
ru [Thu, 31 May 2001 07:08:49 +0000 (07:08 +0000)]
Show TCP checksum failures with -tcp display.

PR: bin/27786
Submitted by: Brooks Davis <brooks@one-eyed-alien.net>

23 years agoMinor style(9) tweak: if ( instead of if(.
imp [Thu, 31 May 2001 05:57:49 +0000 (05:57 +0000)]
Minor style(9) tweak: if ( instead of if(.

23 years agoAdd suggested parens around truth value.
imp [Thu, 31 May 2001 04:37:17 +0000 (04:37 +0000)]
Add suggested parens around truth value.

23 years agoFix date. This man page was not written on July 11, 2001, because that
wpaul [Thu, 31 May 2001 00:18:19 +0000 (00:18 +0000)]
Fix date. This man page was not written on July 11, 2001, because that
day hasn't happened yet. It should be May. Ignore failed time-travel
experiment, K PLZ THX.

23 years agoCorrect the documentation for snprintf() and vsprintf() which actually
gshapiro [Wed, 30 May 2001 23:47:14 +0000 (23:47 +0000)]
Correct the documentation for snprintf() and vsprintf() which actually
return the number of characters that would have been in the new string.

Obtained from: OpenBSD
MFC after: 3 days

23 years agoRevert the previous commit in favor of the fix in rev 1.42 of
jhb [Wed, 30 May 2001 23:09:19 +0000 (23:09 +0000)]
Revert the previous commit in favor of the fix in rev 1.42 of
ufs/ffs/ffs_extern.h instead.

Requested by: bde

23 years agoForward declare struct cg to quiet a warning.
jhb [Wed, 30 May 2001 23:08:40 +0000 (23:08 +0000)]
Forward declare struct cg to quiet a warning.

Submitted by: bde

23 years agoCatch up to the axeing of MFS and fix the ia64 build.
jhb [Wed, 30 May 2001 23:06:14 +0000 (23:06 +0000)]
Catch up to the axeing of MFS and fix the ia64 build.

Forgotten by: a Danish axe-wielder

23 years agofix typo
cg [Wed, 30 May 2001 22:46:40 +0000 (22:46 +0000)]
fix typo

23 years agono longer used
cg [Wed, 30 May 2001 22:41:04 +0000 (22:41 +0000)]
no longer used

23 years agoupdate this driver to use new firmware and incorporate many fixes.
cg [Wed, 30 May 2001 22:38:31 +0000 (22:38 +0000)]
update this driver to use new firmware and incorporate many fixes.

this works on cs4630 chips, and should implement the clkrun hack for
thinkpads- this will display diagnostic messages when triggered until its
correctness is established.

23 years agonew (gpl licensed) firmware image for cs4280/cs46xx sound chips
cg [Wed, 30 May 2001 22:28:30 +0000 (22:28 +0000)]
new (gpl licensed) firmware image for cs4280/cs46xx sound chips

23 years agoQuiet warnings by adding a prototype for set_user_ldt_rv() and making it
jhb [Wed, 30 May 2001 21:57:52 +0000 (21:57 +0000)]
Quiet warnings by adding a prototype for set_user_ldt_rv() and making it
conditional on #ifdef SMP.

23 years agoFrom the PR:
imp [Wed, 30 May 2001 21:45:50 +0000 (21:45 +0000)]
From the PR:
The PCCard daemon can hang indefinately while reading its
configuration file. If the last line of the file is a comment line
that does not end in a newline, the program goes into an infinite
loop searching for the non-existent newline.

This fix, provided by the PR, will allow files ending without a newline
to be read without hanging.

Submitted by: Crist J. Clark <cjclark@alum.mit.edu>
PR: bin/25791

23 years agoAdd Billionton LNT-10TN
imp [Wed, 30 May 2001 21:30:40 +0000 (21:30 +0000)]
Add Billionton LNT-10TN

Submitted by: Miklos Niedermayer <mico@bsd.hu>
PR: conf/27726

23 years agoWhen doing the chdir("..") in the !FTS_NOCHDIR case, stat() after we get
kris [Wed, 30 May 2001 20:35:44 +0000 (20:35 +0000)]
When doing the chdir("..") in the !FTS_NOCHDIR case, stat() after we get
there and compare the inode and device numbers to the values we remember,
to guard against the directory having been moved around in the meantime.

Reported by: Nick Cleaton <nick@cleaton.net>

23 years agoDefault daily_accounting_flags to -q. I thought this was a typo in the
brian [Wed, 30 May 2001 20:23:43 +0000 (20:23 +0000)]
Default daily_accounting_flags to -q.  I thought this was a typo in the
originally submitted patch (oops!).

Also check for an empty $daily_accounting_save.

Submitted by: Udo Schweigert <Udo.Schweigert@cert.siemens.de>