]> CyberLeo.Net >> Repos - FreeBSD/FreeBSD.git/log
FreeBSD/FreeBSD.git
25 years agoThe way the interrupt id was calculated was wrong and the lpt
Nicolas Souchu [Sun, 14 Feb 1999 17:09:59 +0000 (17:09 +0000)]
The way the interrupt id was calculated was wrong and the lpt
driver was thinking irq was enabled although it wasn't.
This case was particular to a no-interrupt static configuration.

Reported by: "Norman C. Rice" <nrice@emu.sourcee.com>

25 years agoMove lp.4 to man4. It is not i386 dependent.
Nicolas Souchu [Sun, 14 Feb 1999 16:22:27 +0000 (16:22 +0000)]
Move lp.4 to man4. It is not i386 dependent.

Fix ppbus and lpt declaration ('0' forgotten). Add info about retrieving
printer status with 'cat /dev/lpt0'.

25 years agoMove away from lpt.h data structure definition that only lpt.c relies on.
Nicolas Souchu [Sun, 14 Feb 1999 16:19:16 +0000 (16:19 +0000)]
Move away from lpt.h data structure definition that only lpt.c relies on.

Remove if_plip.c dependency to lpt.h.

25 years agoReturn approriate errors to userland.
Nicolas Souchu [Sun, 14 Feb 1999 14:36:45 +0000 (14:36 +0000)]
Return approriate errors to userland.

25 years agoMention that chflags can fail with EOPNOTSUPP.
Dag-Erling Smørgrav [Sun, 14 Feb 1999 13:58:18 +0000 (13:58 +0000)]
Mention that chflags can fail with EOPNOTSUPP.

25 years agoIgnore errors from chflags. This makes it possible to make installworld
Dag-Erling Smørgrav [Sun, 14 Feb 1999 13:56:15 +0000 (13:56 +0000)]
Ignore errors from chflags. This makes it possible to make installworld
with DESTDIR set to an NFS-mounted file system.

25 years agoAdd -e option and change unit to device access in lptcontrol.8
Nicolas Souchu [Sun, 14 Feb 1999 12:23:49 +0000 (12:23 +0000)]
Add -e option and change unit to device access in lptcontrol.8

Change unit to device access in lptcontrol.c. Now usage is

lptcontrol -i | -p | -e [-d /dev/lpt?]

25 years agoDescribe manual dialing in greater detail.
Brian Somers [Sun, 14 Feb 1999 12:16:41 +0000 (12:16 +0000)]
Describe manual dialing in greater detail.
Mention more rfc numbers.
Don't ``.Nm Ppp'' (just use ``.Nm'').

25 years agoCreate new lpt manpage with ppbus info and the original lpt info.
Nicolas Souchu [Sun, 14 Feb 1999 12:06:16 +0000 (12:06 +0000)]
Create new lpt manpage with ppbus info and the original lpt info.

Fix lp.4 ppbus terminology.

25 years agoFix interrupt handling with DMA. Bit nFault was tested in the control reg.
Nicolas Souchu [Sun, 14 Feb 1999 12:03:35 +0000 (12:03 +0000)]
Fix interrupt handling with DMA. Bit nFault was tested in the control reg.
instead of the status reg. and check ECP mode before considering nFault.

25 years agoRename nlpt to lpt.
Nicolas Souchu [Sun, 14 Feb 1999 12:00:00 +0000 (12:00 +0000)]
Rename nlpt to lpt.

Remove from ppi.c the old depreciated module stuff.
Print info when if_plip can't use interrupts.

25 years agouse minimum length for strcoll too
Andrey A. Chernov [Sun, 14 Feb 1999 09:44:56 +0000 (09:44 +0000)]
use minimum length for strcoll too

25 years agoUse more sensible defaults.
Jordan K. Hubbard [Sun, 14 Feb 1999 07:43:07 +0000 (07:43 +0000)]
Use more sensible defaults.

25 years agorc.conf vars no longer "dirty" by default.
Jordan K. Hubbard [Sun, 14 Feb 1999 07:35:27 +0000 (07:35 +0000)]
rc.conf vars no longer "dirty" by default.

25 years agoPreserve existing rc.conf contents.
Jordan K. Hubbard [Sun, 14 Feb 1999 05:52:57 +0000 (05:52 +0000)]
Preserve existing rc.conf contents.

25 years agolpt -> nlpt and ppbus due to Dag's changes.
Warner Losh [Sun, 14 Feb 1999 05:18:35 +0000 (05:18 +0000)]
lpt -> nlpt and ppbus due to Dag's changes.

Submitted by: Joseph Koshy <jkoshy@FreeBSD.ORG>

P.S.  Forgive the delay on committing this.  I've been out of town at
a funeral until just a few minutes ago.

25 years agoFix .Fx usage.
Bill Fenner [Sat, 13 Feb 1999 21:45:58 +0000 (21:45 +0000)]
Fix .Fx usage.

25 years agoMake sure correct path is set in upgrade.
Jordan K. Hubbard [Sat, 13 Feb 1999 18:49:02 +0000 (18:49 +0000)]
Make sure correct path is set in upgrade.

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

25 years agoFix the date and add an smbus declaration
Nicolas Souchu [Sat, 13 Feb 1999 18:08:35 +0000 (18:08 +0000)]
Fix the date and add an smbus declaration

25 years agoHandle correctly iicbus request/release mechanism. Add iicbus allocation
Nicolas Souchu [Sat, 13 Feb 1999 18:01:55 +0000 (18:01 +0000)]
Handle correctly iicbus request/release mechanism. Add iicbus allocation
to the general purpose i/o iic(4) driver.

25 years agoFix smbus allocation and add the alsmb (see alpm(4)) driver.
Nicolas Souchu [Sat, 13 Feb 1999 17:57:19 +0000 (17:57 +0000)]
Fix smbus allocation and add the alsmb (see alpm(4)) driver.

25 years agoAdd alpm, Acer Aladdin-IV/V/Pro2 Power Management Unit. See alpm(4).
Nicolas Souchu [Sat, 13 Feb 1999 17:54:32 +0000 (17:54 +0000)]
Add alpm, Acer Aladdin-IV/V/Pro2 Power Management Unit. See alpm(4).

25 years agoAdd alpm.c, Aladdin Power Management SMBus support for the SMBus framework.
Nicolas Souchu [Sat, 13 Feb 1999 17:51:46 +0000 (17:51 +0000)]
Add alpm.c, Aladdin Power Management SMBus support for the SMBus framework.

Update pcisupport.c with NALPM and Aladdin USB detection printf.

25 years agoAdd pci/alpm.c, Aladdin SMBus driver (alpm(4)).
Nicolas Souchu [Sat, 13 Feb 1999 17:49:57 +0000 (17:49 +0000)]
Add pci/alpm.c, Aladdin SMBus driver (alpm(4)).

25 years agoAdd alpm.4, Aladdin Power Management SMBus support for the SMBus framework.
Nicolas Souchu [Sat, 13 Feb 1999 17:48:41 +0000 (17:48 +0000)]
Add alpm.4, Aladdin Power Management SMBus support for the SMBus framework.

25 years agoDon't pass PSL_NT to vm86 signal handlers. Some vm86/real mode
Bruce Evans [Sat, 13 Feb 1999 17:45:15 +0000 (17:45 +0000)]
Don't pass PSL_NT to vm86 signal handlers.  Some vm86/real mode
programs, including msdos, set PSL_NT in probes for old cpu types,
although PSL_NT doesn't do anything useful in vm86 or real mode.
PSL_NT is even less useful in the signal handlers.  It just causes
T_TSSFLT faults on return from syscalls made by the handlers.
These faults are fixed up lazily so that Xsyscall() doesn't have
to be slowed down to prevent them.  The fault handler recently
started complaining about these faults occurring "with interrupts
disabled".  It should not have, but the complaints pointed to this
bug.

PR: 9211

25 years agoremove collating optimization - not work with substituted characters
Andrey A. Chernov [Sat, 13 Feb 1999 15:45:30 +0000 (15:45 +0000)]
remove collating optimization - not work with substituted characters

25 years agosyntax change: allow symbolic names as substitute first arg
Andrey A. Chernov [Sat, 13 Feb 1999 14:15:25 +0000 (14:15 +0000)]
syntax change: allow symbolic names as substitute first arg
use this to substitute <ss>

25 years agosyntax change: allow symbolic names as substitute first arg
Andrey A. Chernov [Sat, 13 Feb 1999 14:14:47 +0000 (14:14 +0000)]
syntax change: allow symbolic names as substitute first arg

25 years agouse strcoll() where it is easy to sort german ss and similar properly
Andrey A. Chernov [Sat, 13 Feb 1999 13:05:04 +0000 (13:05 +0000)]
use strcoll() where it is easy to sort german ss and similar properly

25 years agoAdd "beep" subcommand.
Jun Kuriyama [Sat, 13 Feb 1999 11:32:01 +0000 (11:32 +0000)]
Add "beep" subcommand.

Obtained from: PAO3
Reviewed by: -current list

25 years agoAdd "beep" ioctl. (#10 is reserved for PIOCSVIR)
Jun Kuriyama [Sat, 13 Feb 1999 11:32:00 +0000 (11:32 +0000)]
Add "beep" ioctl.  (#10 is reserved for PIOCSVIR)

Obtained from: PAO3
Reviewed by: -current list

25 years agoFix ufs mounting support
David E. O'Brien [Sat, 13 Feb 1999 11:06:20 +0000 (11:06 +0000)]
Fix ufs mounting support

PR: 10044
Submitted by: Andre Albsmeier <andre.albsmeier@mchp.siemens.de>

25 years ago Add -z option to vmstat to dump data from the zone allocator
Matthew Dillon [Sat, 13 Feb 1999 09:59:24 +0000 (09:59 +0000)]
Add -z option to vmstat to dump data from the zone allocator

25 years agoremove BUGS section, now behaviour is predictable enough
Andrey A. Chernov [Sat, 13 Feb 1999 09:57:58 +0000 (09:57 +0000)]
remove BUGS section, now behaviour is predictable enough

25 years ago General additional cleanup of VOP API for NFS ops - mainly NFS ignoring
Matthew Dillon [Sat, 13 Feb 1999 09:47:30 +0000 (09:47 +0000)]
General additional cleanup of VOP API for NFS ops - mainly NFS ignoring
    the API for freeing up cnp's.  This cleanup should not effect nominal
    operation one way or the other since NFS VOPs just happen to be called
    with flags that match what it actually does to the NAMEI components it
    gets.  Still, if an NFS error occured, there was probably some memory
    leakage of NAMEI components with certain NFS VOP ops.

25 years ago Remove XXX comment in regarsd to why NFS doesn't use VOP_ABORT(). NFS
Matthew Dillon [Sat, 13 Feb 1999 08:38:28 +0000 (08:38 +0000)]
Remove XXX comment in regarsd to why NFS doesn't use VOP_ABORT().  NFS
    is being fixed now.

25 years agoPR: kern/9970
Matthew Dillon [Sat, 13 Feb 1999 08:01:59 +0000 (08:01 +0000)]
PR: kern/9970

    Remove incorrect vput() in nfs_link()

25 years ago Clarify more of what VOP_LINK expects.
Matthew Dillon [Sat, 13 Feb 1999 07:56:49 +0000 (07:56 +0000)]
Clarify more of what VOP_LINK expects.

25 years ago Update man page - directory vnode should not be unlocked or released
Matthew Dillon [Sat, 13 Feb 1999 07:44:14 +0000 (07:44 +0000)]
Update man page - directory vnode should not be unlocked or released
    by the VOP routine.  The caller does it.

25 years agoCorrect various mentions of rc.conf
Jordan K. Hubbard [Sat, 13 Feb 1999 05:30:49 +0000 (05:30 +0000)]
Correct various mentions of rc.conf

25 years agoUpdate to current state of /etc
Jordan K. Hubbard [Sat, 13 Feb 1999 05:18:06 +0000 (05:18 +0000)]
Update to current state of /etc

25 years agoAdd some examples of how to throw a window up on your display when
Brian Somers [Sat, 13 Feb 1999 00:59:19 +0000 (00:59 +0000)]
Add some examples of how to throw a window up on your display when
the peer demands authentication, and add some more detail to the
example configurations.

This is the first time I've written any TCL, so I'd appreciate it
if someone eyeballed the *-auth stuff and fixed any glaring problems.

25 years agomore words about return value
Andrey A. Chernov [Sat, 13 Feb 1999 00:04:54 +0000 (00:04 +0000)]
more words about return value

25 years agore-phrase things a bit
Andrey A. Chernov [Fri, 12 Feb 1999 23:52:29 +0000 (23:52 +0000)]
re-phrase things a bit

25 years agofix wrong return result
Andrey A. Chernov [Fri, 12 Feb 1999 23:40:41 +0000 (23:40 +0000)]
fix wrong return result
fix n=0 case
improve manpage

25 years agosubstitute ss
Andrey A. Chernov [Fri, 12 Feb 1999 21:14:41 +0000 (21:14 +0000)]
substitute ss
Submitted by: "D. Rock" <rock@cs.uni-sb.de>

25 years agoadd more checks for substitution
Andrey A. Chernov [Fri, 12 Feb 1999 20:56:49 +0000 (20:56 +0000)]
add more checks for substitution
dissalow substituted character be ordered

25 years ago jkh accidently removed rc.diskless1 & 2, then put rc.diskless back when
Matthew Dillon [Fri, 12 Feb 1999 20:51:38 +0000 (20:51 +0000)]
jkh accidently removed rc.diskless1 & 2, then put rc.diskless back when
    he moved rc.conf.  Then he deleted rc.diskless when it ( of course ) didn't
    work.  Now I'm putting the originally accidently removed rc.diskless{1,2}
    back in.

25 years ago Minor optimization to madvise() MADV_FREE to make page as freeable as
Matthew Dillon [Fri, 12 Feb 1999 20:42:19 +0000 (20:42 +0000)]
Minor optimization to madvise() MADV_FREE to make page as freeable as
    possible without actually unmapping it from the process.

    As of now, I declare madvise() on OBJT_DEFAULT/OBJT_SWAP objects to be
    'working and complete'.

25 years agofix English
Andrey A. Chernov [Fri, 12 Feb 1999 20:39:06 +0000 (20:39 +0000)]
fix English
detect recursive substitutions
allow substituted character not present in the order

25 years agoMerge from RELENG_3.
Andrzej Bialecki [Fri, 12 Feb 1999 20:36:13 +0000 (20:36 +0000)]
Merge from RELENG_3.

25 years agofix tabs lost apparently in copy&paste
Andrey A. Chernov [Fri, 12 Feb 1999 19:45:53 +0000 (19:45 +0000)]
fix tabs lost apparently in copy&paste

25 years agoImproved reporting of autodetected speed and duplex.
Julian Elischer [Fri, 12 Feb 1999 17:56:23 +0000 (17:56 +0000)]
Improved reporting of autodetected speed and duplex.
Now should be able to report speed for cards using NatSemi PHY.
(if you have one please let me know if it works as I
only have the Intel version)

Reviewed by: David Greenman <dg@root.com>

25 years agoRemove float-point dependency from i386's version of setjmp()/longjmp().
Daniel C. Sobral [Fri, 12 Feb 1999 17:22:30 +0000 (17:22 +0000)]
Remove float-point dependency from i386's version of setjmp()/longjmp().

25 years agoCopy _setjmp.S from libc in preparation for loader-specific version.
Daniel C. Sobral [Fri, 12 Feb 1999 17:20:19 +0000 (17:20 +0000)]
Copy _setjmp.S from libc in preparation for loader-specific version.

25 years agoMerge from OpenBSD up to rev 1.7 (matches NetBSD rev 1.4):
Eivind Eklund [Fri, 12 Feb 1999 15:06:55 +0000 (15:06 +0000)]
Merge from OpenBSD up to rev 1.7 (matches NetBSD rev 1.4):

Misc small cleanups.

25 years agoMerge from OpenBSD up to rev 1.5 (matches NetBSD up to rev 1.3):
Eivind Eklund [Fri, 12 Feb 1999 14:42:31 +0000 (14:42 +0000)]
Merge from OpenBSD up to rev 1.5 (matches NetBSD up to rev 1.3):
* Clean up waitpid parameter handling.

25 years agoBack out the NetBSD .Nm changes - we don't have the required
Eivind Eklund [Fri, 12 Feb 1999 13:24:45 +0000 (13:24 +0000)]
Back out the NetBSD .Nm changes - we don't have the required
infrastructure, and I don't seem to find time to merge it.

25 years agoFixed -Wcast-qual warnings.
Bruce Evans [Fri, 12 Feb 1999 12:44:19 +0000 (12:44 +0000)]
Fixed -Wcast-qual warnings.

25 years agoFixed fatal type mismatches in the aout case. Const poisoning in
Bruce Evans [Fri, 12 Feb 1999 12:15:07 +0000 (12:15 +0000)]
Fixed fatal type mismatches in the aout case.  Const poisoning in
db_sym.h had not reached here.

25 years agoCorrect i/o addresses for dgb & dgm
Brian Somers [Fri, 12 Feb 1999 12:15:02 +0000 (12:15 +0000)]
Correct i/o addresses for dgb & dgm
Submitted for dgm by: Andre Oppermann <opi@opi.flirtbox.ch>

25 years ago Fix non-fatal bug in vm_map_insert() which improperly cleared
Matthew Dillon [Fri, 12 Feb 1999 09:51:43 +0000 (09:51 +0000)]
Fix non-fatal bug in vm_map_insert() which improperly cleared
    OBJ_ONEMAPPING in the case where an object is extended by an
    additional vm_map_entry must be allocated.

    In vm_object_madvise(), remove calll to vm_page_cache() in MADV_FREE
    case in order to avoid a page fault on page reuse.  However, we still
    mark the page as clean and destroy any swap backing store.

Submitted by: Alan Cox <alc@cs.rice.edu>

25 years agoSync with sys/i386/i386/machdep.c revision 1.325.
KATO Takenori [Fri, 12 Feb 1999 09:15:33 +0000 (09:15 +0000)]
Sync with sys/i386/i386/machdep.c revision 1.325.

25 years agoRemoved occurrences of consecutive repeated words (such as "the the").
Alexander Langer [Fri, 12 Feb 1999 02:12:08 +0000 (02:12 +0000)]
Removed occurrences of consecutive repeated words (such as "the the").

25 years agoWhen executing a command as part of a dial/login/hangup
Brian Somers [Fri, 12 Feb 1999 00:52:30 +0000 (00:52 +0000)]
When executing a command as part of a dial/login/hangup
script, expand words in the same way as !bg does.

25 years agoAllow both old and new rc.conf syntax.
Jordan K. Hubbard [Fri, 12 Feb 1999 00:17:12 +0000 (00:17 +0000)]
Allow both old and new rc.conf syntax.

25 years agoDo not issue RX and TX reset commands in xl_stop() and xl_init(). On the
Bill Paul [Thu, 11 Feb 1999 23:59:29 +0000 (23:59 +0000)]
Do not issue RX and TX reset commands in xl_stop() and xl_init(). On the
3c905B, the RX and TX reset commands also reset the cyclone chip's internal
PHY, which causes it to restart its autonegotiation session. This takes a
second or two to complete, which makes the interface seem to stop responding
for a few seconds every time you do something that reinitializes it.

Issuing the RX and TX resets on the older 3c905 boomerang adapters doesn't
cause any delay because the boomerang chip requires an external PHY.

This should fix the problem where people doing network installs via 3c905B
cards experience a delay after the interface is first initialized, among
other things.

25 years agoDefine more registers and fix incorrect (but unused) register bit definitions.
Julian Elischer [Thu, 11 Feb 1999 23:41:21 +0000 (23:41 +0000)]
Define more registers and fix incorrect (but unused) register bit definitions.

25 years agoDefine more registers in the PHY unit and use them to report back
Julian Elischer [Thu, 11 Feb 1999 21:47:10 +0000 (21:47 +0000)]
Define more registers in the PHY unit and use them to report back
the result of the media auto negotiation.

Reviewed by: David Greenman <dg@freebsd.org>

25 years agoAdd a note about the insecurity of MD4 and potential vulnerability of
Garrett Wollman [Thu, 11 Feb 1999 20:31:49 +0000 (20:31 +0000)]
Add a note about the insecurity of MD4 and potential vulnerability of
MD5 to similar attacks.

25 years agoUse the fictional server "CHANGE_THIS.FreeBSD.org". Add comments
John Polstra [Thu, 11 Feb 1999 18:14:03 +0000 (18:14 +0000)]
Use the fictional server "CHANGE_THIS.FreeBSD.org".  Add comments
telling the user to (duh) change it.

25 years agoRELENG_3 is now "the" stable branch.
John Polstra [Thu, 11 Feb 1999 18:04:17 +0000 (18:04 +0000)]
RELENG_3 is now "the" stable branch.

25 years agoReplace the out-of-date list of mirror sites with a reference to the
John Polstra [Thu, 11 Feb 1999 17:56:11 +0000 (17:56 +0000)]
Replace the out-of-date list of mirror sites with a reference to the
list in the FreeBSD Handbook.

25 years ago$Id$ misspelled as $Id
Wolfram Schneider [Thu, 11 Feb 1999 17:09:27 +0000 (17:09 +0000)]
$Id$ misspelled as $Id

Submitted by: bde

25 years ago/etc/ppp/ppp.*.sample -> /usr/share/examples/ppp/ppp.*.sample
Brian Somers [Thu, 11 Feb 1999 16:34:15 +0000 (16:34 +0000)]
/etc/ppp/ppp.*.sample -> /usr/share/examples/ppp/ppp.*.sample

25 years agoAdd ppp example files.
Brian Somers [Thu, 11 Feb 1999 16:33:14 +0000 (16:33 +0000)]
Add ppp example files.

25 years agoAdd /usr/share/examples/ppp directory
Brian Somers [Thu, 11 Feb 1999 16:30:56 +0000 (16:30 +0000)]
Add /usr/share/examples/ppp directory
Don't install /etc/ppp/ppp.*.example
Remove ppp/ppp.{conf,linkup,linkdown,secret}.sample

25 years agoThis tries to install rc.diskless, but there is no such file.
Jordan K. Hubbard [Thu, 11 Feb 1999 15:46:22 +0000 (15:46 +0000)]
This tries to install rc.diskless, but there is no such file.
I don't know which of the two to install in its place so just
remove it for now and get the release build working again.

25 years agoAdd spanish mirrors to FTP menu.
Jordan K. Hubbard [Thu, 11 Feb 1999 14:42:39 +0000 (14:42 +0000)]
Add spanish mirrors to FTP menu.

25 years agoList diamond fireport cards under Symbios section.
Jordan K. Hubbard [Thu, 11 Feb 1999 14:00:14 +0000 (14:00 +0000)]
List diamond fireport cards under Symbios section.

Submitted by: Stephen McKay <syssgm@detir.qld.gov.au>

25 years agofix conflicts
David E. O'Brien [Thu, 11 Feb 1999 12:49:40 +0000 (12:49 +0000)]
fix conflicts

25 years agoThis commit was generated by cvs2svn to compensate for changes in r43892,
David E. O'Brien [Thu, 11 Feb 1999 12:40:22 +0000 (12:40 +0000)]
This commit was generated by cvs2svn to compensate for changes in r43892,
which included commits to RCS files with non-trunk default branches.

25 years agoVirgin import of ISC-DHCP v2.0b1pl11
David E. O'Brien [Thu, 11 Feb 1999 12:40:22 +0000 (12:40 +0000)]
Virgin import of ISC-DHCP v2.0b1pl11

25 years agoupdate for latest import
David E. O'Brien [Thu, 11 Feb 1999 12:39:45 +0000 (12:39 +0000)]
update for latest import

25 years agoAdded ioctl REMOTE_GETKEY for Hauppauge Infra-Red Remote Control.
Roger Hardiman [Thu, 11 Feb 1999 10:22:30 +0000 (10:22 +0000)]
Added ioctl REMOTE_GETKEY for Hauppauge Infra-Red Remote Control.
Submitted by Roger Hardiman.

Added ioctl TVTUNER_GETCHANSET to discover which regions the bktr driver
supports. Submitted by Vsevolod Lobko <seva@alex-ua.com>

Added BT848_GPIO_SET_EN,BT848_GPIO_SET_DATA (and GETs) to allow user land
control of the GPIO pins. This allows a Radio module on the GPIO port
to be controlled. Submitted by Vsevolod Lobko <seva@alex-ua.com>
The kernel option BKTR_GPIO_ACCESS must be used to enable the GPIO ioctls.

Submitted by: Roger Hardiman and Vsevolod Lobko <seva@alex-ua.com>

25 years agoAdded new ioctls for Hauppage Infra-Red Remote Control support,
Roger Hardiman [Thu, 11 Feb 1999 10:18:04 +0000 (10:18 +0000)]
Added new ioctls for Hauppage Infra-Red Remote Control support,
Obtaining the supported channel sets and Controlling the GPIO port.

Submitted by: Roger Hardiman and Vsevolod Lobko <seva@alex-ua.com>

25 years agoWhen resending chap challenges, resend the same challenge
Brian Somers [Thu, 11 Feb 1999 10:14:08 +0000 (10:14 +0000)]
When resending chap challenges, resend the same challenge
each time rather than making up a new one.

Increase the authname/authkey max sizes to 100 characters.

Allow ``authkey'' specifications beginning with ``!''.
When a challenge is received, the text following the
``!'' is executed as a program (expanding stuff in the same
way that ``sh'' and ``!bg'' do).  The program is passed the
peer name, peer challenge and local ``authname'' on standard
input and is expected to output the name/key combination that
should be used to build the CHAP response.

This provides support for Secure ID cards (guess what I was
given at work recently!) using CHAP.

Examples will follow.

25 years agoZero p->retval[1] when starting a process. This value ends up in %edx
Mike Smith [Thu, 11 Feb 1999 07:53:28 +0000 (07:53 +0000)]
Zero p->retval[1] when starting a process.  This value ends up in %edx
when the process starts, and having it nonzero causes statically-linked
Linux binaries to fail.

PR: i386/10015
Submitted by: Marcel Moolenaar <marcel@scc.nl>

25 years agoSet the ultra enable bits in the adaptor's scratch ram area only after
Justin T. Gibbs [Thu, 11 Feb 1999 07:12:16 +0000 (07:12 +0000)]
Set the ultra enable bits in the adaptor's scratch ram area only after
we finish modifying our in kernel version during seeprom configuration.

25 years agoMake the ahc_eisa file also optional on 'eisa'.
Justin T. Gibbs [Thu, 11 Feb 1999 07:11:00 +0000 (07:11 +0000)]
Make the ahc_eisa file also optional on 'eisa'.

25 years agoEnsure that the ACK from the last information transfer phase has been
Justin T. Gibbs [Thu, 11 Feb 1999 07:10:20 +0000 (07:10 +0000)]
Ensure that the ACK from the last information transfer phase has been
de-asserted by the initiator before changing phase or releasing the bus.
Corrects a SCSI specification violation in target mode.

25 years agoCompletely set information in the transaction description before performing
Justin T. Gibbs [Thu, 11 Feb 1999 07:08:58 +0000 (07:08 +0000)]
Completely set information in the transaction description before performing
a selwakeup for it.  Cosmetic.

25 years agoCorrect a race condition where a target mode SCB dma could occur
Justin T. Gibbs [Thu, 11 Feb 1999 07:07:27 +0000 (07:07 +0000)]
Correct a race condition where a target mode SCB dma could occur
before we finished setting the pending phase information.

Use the ALL_TARGETS_MASK #define instead of 0xffff for the default
ultraenb mask.

25 years agoCorrect some indentation
Greg Lehey [Thu, 11 Feb 1999 06:44:46 +0000 (06:44 +0000)]
Correct some indentation

Implicitly-desired-by: bde
update_plex_state:  Handle empty RAID-5 plexes correctly

25 years agobuild_write_request: correct logic for handling writes to volumes with
Greg Lehey [Thu, 11 Feb 1999 06:44:12 +0000 (06:44 +0000)]
build_write_request: correct logic for handling writes to volumes with
failed plexes.

allocrqg: Remove dead code

25 years agocheck_drive: Make sure that we don't open two different drives with
Greg Lehey [Thu, 11 Feb 1999 06:43:15 +0000 (06:43 +0000)]
check_drive: Make sure that we don't open two different drives with
the same name.  Silently return EEXIST if this happens.

vinum_scandisk: Collect drive numbers, not pointers, to avoid problems
of relocated drives.

Tripped-over-by: Bernd Walter <ticso@cicely.de>
25 years agoIn some cases, it's possible that I/O errors will cause the daemon to
Greg Lehey [Thu, 11 Feb 1999 06:42:39 +0000 (06:42 +0000)]
In some cases, it's possible that I/O errors will cause the daemon to
time out on an operation.  Under these circumstances, vinum(8) will
automatically start another daemon.  Add a pid for the daemon, so that
an overtaken daemon will discover that it's no longer in favour, and
will crawl into a corner and die.

25 years agoModify example configuration file to show more optimum stripe size of
Greg Lehey [Thu, 11 Feb 1999 06:14:49 +0000 (06:14 +0000)]
Modify example configuration file to show more optimum stripe size of
256 kB instead of 16 kB.

Pointed-out-by: Steve Taylor <staylor@cybernet.com>
Modify description of start command to include 'start' with no
parameters, which reads the config from all drives found on the
system.

25 years agoExtend 'start' command: if used without any parameters, vinum scans
Greg Lehey [Thu, 11 Feb 1999 06:14:15 +0000 (06:14 +0000)]
Extend 'start' command: if used without any parameters, vinum scans
all disks known to devstat for vinum drives and reads their
configuration.