]> CyberLeo.Net >> Repos - FreeBSD/FreeBSD.git/log
FreeBSD/FreeBSD.git
25 years agoFix a width.
ru [Sat, 5 Jun 1999 15:03:47 +0000 (15:03 +0000)]
Fix a width.

25 years agoFloppy driver options moved to conf/options.
dfr [Sat, 5 Jun 1999 13:57:10 +0000 (13:57 +0000)]
Floppy driver options moved to conf/options.

25 years ago* Implement ISA dma (only for CIA now but more chipsets to follow).
dfr [Sat, 5 Jun 1999 13:30:13 +0000 (13:30 +0000)]
* Implement ISA dma (only for CIA now but more chipsets to follow).
* Port the fd driver to alpha.

25 years agoDon't discard error output from sysctl(8).
bde [Sat, 5 Jun 1999 12:06:19 +0000 (12:06 +0000)]
Don't discard error output from sysctl(8).

Do discard standard output from the sysctl for approxy_all, and echo
what this sysctl is doing in the usual way.  This fix is probably
backwards.  We should probably just use the standard sysctl output
in all cases (it needs to have a newline filtered out).

Echo what the sysctls for nfs_reserved_port_only and nfs_access_cache
are doing.

25 years agoWhen incrementing through a SIOCGIFCONF list, enforce a lower limit of
archie [Sat, 5 Jun 1999 05:55:07 +0000 (05:55 +0000)]
When incrementing through a SIOCGIFCONF list, enforce a lower limit of
sizeof(ifr->ifr_addr) for the variable length field ifr->ifr_addr.sa_len.
Otherwise the increment will be wrong in certain cases.

Obtained from:  Whistle source tree
For the record: Garrett Wollman <wollman@khavrinen.lcs.mit.edu> suggests
                SIOCGIFCONF should be dropped in favor of a sysctl mechanism.

25 years agoWhen incrementing through a SIOCGIFCONF list, enforce a lower limit of
archie [Sat, 5 Jun 1999 05:52:52 +0000 (05:52 +0000)]
When incrementing through a SIOCGIFCONF list, enforce a lower limit of
sizeof(ifr->ifr_addr) for the variable length field ifr->ifr_addr.sa_len.
Otherwise the increment will be wrong in certain cases.

Obtained from: Whistle source tree
For the record: Garrett Wollman <wollman@khavrinen.lcs.mit.edu> suggests
SIOCGIFCONF should be dropped in favor of a sysctl mechanism.

25 years agoAdd handle to control global TCP keepalives and turn them on as
phk [Sat, 5 Jun 1999 05:45:57 +0000 (05:45 +0000)]
Add handle to control global TCP keepalives and turn them on as
default.

Despite their name it doesn't keep TCP sessions alive, it kills
them if the other end has gone AWOL.  This happens a lot with
clients which use NAT, dynamic IP assignment or which has a 2^32
* 10^-3 seconds upper bound on their uptime.

There is no detectable increase in network trafic because of this:
two minimal TCP packets every two hours for a live TCP connection.

Many servers already enable keepalives themselves.

The host requirements RFC is 10 years old, and doesn't know about
the loosing clients of todays InterNet.

25 years agoVarious changes lifted from the OpenBSD cvs tree:
peter [Sat, 5 Jun 1999 05:35:03 +0000 (05:35 +0000)]
Various changes lifted from the OpenBSD cvs tree:

txdr_hyper and fxdr_hyper tweaks to avoid excessive CPU order knowledge.

nfs_serv.c: don't call nfsm_adj() with negative values, windows clients
could crash servers when doing a readdir of a large directory.

nfs_socket.c: Use IP_PORTRANGE to get a priviliged port without a spin
loop trying to bind().  Don't clobber a mbuf pointer or we get panics
on a NFS3ERR_JUKEBOX error from a server when reusing a freed mbuf.

nfs_subs.c: Don't loose st_blocks on NFSv2 mounts when > 2GB.

Obtained from:  OpenBSD

25 years agoFix a malloc race
peter [Sat, 5 Jun 1999 05:26:36 +0000 (05:26 +0000)]
Fix a malloc race

Obtained from:  OpenBSD (csapuntz)

25 years agoDon't mistake a non-async block that needs to be committed for an
peter [Sat, 5 Jun 1999 05:25:37 +0000 (05:25 +0000)]
Don't mistake a non-async block that needs to be committed for an
interrupted write.

Obtained from: fvdl@NetBSD.org via OpenBSD.

25 years agoRestructure pipe_read in order to eliminate several race conditions.
alc [Sat, 5 Jun 1999 03:53:57 +0000 (03:53 +0000)]
Restructure pipe_read in order to eliminate several race conditions.

Submitted by: Matthew Dillon <dillon@apollo.backplane.com> and myself

25 years agoSupport the IANA definition of Shift Jis nameley ja_JP.Shift_JIS
julian [Sat, 5 Jun 1999 03:04:03 +0000 (03:04 +0000)]
Support the IANA definition of Shift Jis nameley ja_JP.Shift_JIS
as well as the X11 version ja_JP.SJIS

25 years agoChange an internal variable from "class" to "ns_class" for C++'s benefit.
peter [Fri, 4 Jun 1999 22:38:27 +0000 (22:38 +0000)]
Change an internal variable from "class" to "ns_class" for C++'s benefit.
Apparently BIND-8.2.1 uses ns_class.

Submitted by: John Plevyak <jplevyak@inktomi.com>

25 years agoNo, the space required for a string is strlen() _plus_ one... Oops.
hoek [Fri, 4 Jun 1999 19:35:22 +0000 (19:35 +0000)]
No, the space required for a string is strlen() _plus_ one...  Oops.

25 years agoFixed null setting of `rc_started'. This was fixed for the
bde [Fri, 4 Jun 1999 18:53:47 +0000 (18:53 +0000)]
Fixed null setting of `rc_started'.  This was fixed for the
corresponding variable `rc_wakeup_started' in rev.1.36 but broken
again in rev.1.37.  This bug only caused excessive polling (it gave
NRC activations for each of the SWI handler and the timeout handler
instead of 1 of each).

Moved cdevsw attachment from the driver probe routine to the driver
attach routine.

25 years agoReally fix cy-driver-related panics when SMP is configured. Rev.1.88 only
bde [Fri, 4 Jun 1999 18:13:25 +0000 (18:13 +0000)]
Really fix cy-driver-related panics when SMP is configured.  Rev.1.88 only
fixed half the problem.

Tested by: Michael Scott Boers <mboers@datacompusa.com>

25 years ago${.CURDIR} doesn't have to be /usr/src, especially in a cvs checkout. Make
billf [Fri, 4 Jun 1999 17:34:28 +0000 (17:34 +0000)]
${.CURDIR} doesn't have to be /usr/src, especially in a cvs checkout. Make
sure what is reported to the user is accurate.

Stolen From: mharo

25 years agoFix the cdevsw_add change made by PHK.
roger [Fri, 4 Jun 1999 13:24:54 +0000 (13:24 +0000)]
Fix the cdevsw_add change made by PHK.
(It was accidentally added to the BSDI bktr_probe and not the
 FreeBSD bktr_probe)

25 years agoAdd remark about where bInterfaceProtocol number for USB Zip drive comes
n_hibma [Fri, 4 Jun 1999 12:41:42 +0000 (12:41 +0000)]
Add remark about where bInterfaceProtocol number for USB Zip drive comes
from. 0x50 == 'P' for protoype.

Obtained from: Pat LaVarre <LAVARRE@iomega.com> on linux-usb@suse.com

25 years agoFix the parsing of ip addresses on a command line.
ru [Fri, 4 Jun 1999 11:20:59 +0000 (11:20 +0000)]
Fix the parsing of ip addresses on a command line.

PR: 5047
Reviewed by: des
Test case: ipfw add allow ip from 127.1 to any

25 years agoSwitch to proper mbr.
ru [Fri, 4 Jun 1999 10:01:53 +0000 (10:01 +0000)]
Switch to proper mbr.
OK'ed by: jkh,rnordier

25 years agoOops, remove redundant comments.
kato [Fri, 4 Jun 1999 08:19:29 +0000 (08:19 +0000)]
Oops, remove redundant comments.

25 years agoDeclare Crtat and Atrat because screen savers use these variables.
kato [Fri, 4 Jun 1999 08:17:11 +0000 (08:17 +0000)]
Declare Crtat and Atrat because screen savers use these variables.

Pointed out by: Nobuyuki Koganemaru <kogane@koganemaru.co.jp>

25 years agoModify the boot loader to recognize the -C flag and pass the RB_CDROM
ghelmer [Fri, 4 Jun 1999 03:18:28 +0000 (03:18 +0000)]
Modify the boot loader to recognize the -C flag and pass the RB_CDROM
flag to the kernel to mount a CDROM as the root filesystem.  Alternatively,
the boot_cdrom env var can be set.

As Mike Smith noted, "-C is the "wrong" way to do this", but this is
an acceptable stopgap in lieu of a better way.

PR: bin/11884
Reviewed by: msmith@freebsd.org

25 years agosunsite is now metalab.
billf [Fri, 4 Jun 1999 03:17:04 +0000 (03:17 +0000)]
sunsite is now metalab.

25 years agoBlindly commit an entry for a 'Castlewood ORB2-E', which is a 2.2gig
billf [Fri, 4 Jun 1999 02:49:27 +0000 (02:49 +0000)]
Blindly commit an entry for a 'Castlewood ORB2-E', which is a 2.2gig
removable disk.

Submitted by: Troy Settle <st@i-Plus.net>

25 years agoAdd the 3Dfx Voodoo Banshee
mharo [Fri, 4 Jun 1999 02:38:18 +0000 (02:38 +0000)]
Add the 3Dfx Voodoo Banshee

25 years agoApply patch for PR #12008: remember to reset _yp_enabled flag in
wpaul [Fri, 4 Jun 1999 02:27:35 +0000 (02:27 +0000)]
Apply patch for PR #12008: remember to reset _yp_enabled flag in
endpwent() so we don't trip over a NULL db pointer later.

PR: 12008
Submitted by: Valentin Netchayev <netch@lucky.net>

25 years agoPlug a mbuf leak in tcp_usr_send(). pru_send() routines are expected
peter [Fri, 4 Jun 1999 02:27:06 +0000 (02:27 +0000)]
Plug a mbuf leak in tcp_usr_send().  pru_send() routines are expected
to either enqueue or free their mbuf chains, but tcp_usr_send() was
dropping them on the floor if the tcpcb/inpcb has been torn down in the
middle of a send/write attempt.  This has been responsible for a wide
variety of mbuf leak patterns, ranging from slow gradual leakage to rather
rapid exhaustion.  This has been a problem since before 2.2 was branched
and appears to have been fixed in rev 1.16 and lost in 1.23/1.28.

Thanks to Jayanth Vijayaraghavan <jayanth@yahoo-inc.com> for checking
(extensively) into this on a live production 2.2.x system and that it
was the actual cause of the leak and looks like it fixes it.  The machine
in question was loosing (from memory) about 150 mbufs per hour under
load and a change similar to this stopped it.  (Don't blame Jayanth
for this patch though)

An alternative approach to this would be to recheck SS_CANTSENDMORE etc
inside the splnet() right before calling pru_send() after all the potential
sleeps, interrupts and delays have happened.  However, this would mean
exposing knowledge of the tcp stack's reset handling and removal of the
pcb to the generic code.  There are other things that call pru_send()
directly though.

Problem originally noted by:  John Plevyak <jplevyak@inktomi.com>

25 years agoRemove a rather bogus search path reference..
peter [Thu, 3 Jun 1999 22:07:41 +0000 (22:07 +0000)]
Remove a rather bogus search path reference..

25 years agoGC unused prototype
peter [Thu, 3 Jun 1999 22:03:35 +0000 (22:03 +0000)]
GC unused prototype

25 years agoQuieten the bt_isa_probe() messages since they get a bit much when
peter [Thu, 3 Jun 1999 20:56:09 +0000 (20:56 +0000)]
Quieten the bt_isa_probe() messages since they get a bit much when
the isa probe has gone hunting for a card on it's own.

25 years agoremove references to isa_device, it's no longer associated with interrupts.
peter [Thu, 3 Jun 1999 20:41:00 +0000 (20:41 +0000)]
remove references to isa_device, it's no longer associated with interrupts.

25 years agoCheck and warn about unknown devices in the config file along the same
peter [Thu, 3 Jun 1999 20:37:19 +0000 (20:37 +0000)]
Check and warn about unknown devices in the config file along the same
lines as the pseudo-device checking.  Previously 'device xyz0' would have
been silently accepted without comment.

25 years agoRemove VM86 option.
kato [Thu, 3 Jun 1999 13:50:54 +0000 (13:50 +0000)]
Remove VM86 option.

25 years agoSync with sys/i386/i386/machdep.c revision 1.339.
kato [Thu, 3 Jun 1999 13:49:52 +0000 (13:49 +0000)]
Sync with sys/i386/i386/machdep.c revision 1.339.

New function getmemsize_pc98 is added in this commit, since PC98 is
quite different in obtaining memory size from IBM-PC.  Many lines of
this function is shareable with IBM-PC's getmemsize function, but
sharing needs many #ifdef PC98 statements.  Therefore, I gave up
sharing code with IBM-PC's and just added new function.

25 years agoFix some MP sequence number comparison bogons that are tickled by
brian [Thu, 3 Jun 1999 13:29:32 +0000 (13:29 +0000)]
Fix some MP sequence number comparison bogons that are tickled by
having different speed links in a bundle.  This would manifest itself
by having the link occasionally hang, but revive when a new connection
is made....
Make ``show mp'' a bit prettier.

25 years agoSync with sys/i386/conf/files.i386 and options.i386 revisions 1.246
kato [Thu, 3 Jun 1999 13:04:00 +0000 (13:04 +0000)]
Sync with sys/i386/conf/files.i386 and options.i386 revisions 1.246
and 1.115, respectively.

25 years agoMention that 8x8 font is required for 43, 50 and 60 line text modes,
yokota [Thu, 3 Jun 1999 12:44:16 +0000 (12:44 +0000)]
Mention that 8x8 font is required for 43, 50 and 60 line text modes,
and that font files can be specified in /etc/rc.conf[.local].

25 years agoMake it clear that PS/2 type mice will work only with "auto" and
yokota [Thu, 3 Jun 1999 12:43:16 +0000 (12:43 +0000)]
Make it clear that PS/2 type mice will work only with "auto" and
"ps/2" protocol types, and that serial mouse protocols shouldn't be
specified for these mice.

25 years agoFix PS/2 MouseMan+ protocol. We have been looking at a wrong place
yokota [Thu, 3 Jun 1999 12:42:10 +0000 (12:42 +0000)]
Fix PS/2 MouseMan+ protocol.  We have been looking at a wrong place
for the sign bit for roller movement!

25 years agoUse MACHINE_ARCH instead of MACHINE to check x86 architecture.
kato [Thu, 3 Jun 1999 10:38:07 +0000 (10:38 +0000)]
Use MACHINE_ARCH instead of MACHINE to check x86 architecture.

Pointed out by: Akio Morita <amorita@meadow.scphys.kyoto-u.ac.jp>

25 years agogetvfsbyname() returns zero on success.
ru [Thu, 3 Jun 1999 09:03:50 +0000 (09:03 +0000)]
getvfsbyname() returns zero on success.

PR: 12000
Submitted by: Anatoly A. Orehovsky <tolik@mpeks.tomsk.su>

25 years agoOops, quieten a compiler warning.
brian [Wed, 2 Jun 1999 23:06:21 +0000 (23:06 +0000)]
Oops, quieten a compiler warning.

25 years agoMention physical and sync logging in the ``set log''
brian [Wed, 2 Jun 1999 21:28:02 +0000 (21:28 +0000)]
Mention physical and sync logging in the ``set log''
usage message.

25 years agoBacked out the previous backout. We may as well have a version of ucmpdi2.c
bde [Wed, 2 Jun 1999 16:39:14 +0000 (16:39 +0000)]
Backed out the previous backout.  We may as well have a version of ucmpdi2.c
that actually compiles, and I use it for i386's with 64 bit longs.

25 years agoFixed bitrot in synopsis (a missing include).
bde [Wed, 2 Jun 1999 16:33:22 +0000 (16:33 +0000)]
Fixed bitrot in synopsis (a missing include).

25 years agoo Alter the mbuf type as it's processed by different layers.
brian [Wed, 2 Jun 1999 15:59:09 +0000 (15:59 +0000)]
o Alter the mbuf type as it's processed by different layers.
o Show more information about missing MP fragments in ``show mp''.
o Do away with mbuf_Log().  It was showing mbuf stats twice on
  receipt of LCP/CCP/IPCP packets.... ???!!?
o Pre-allocate a bit extra when creating LQR packets to avoid having
  to allocate another mbuf in mbuf_Prepend().

25 years agoBack out last commit (did I get all of them?)
newton [Wed, 2 Jun 1999 15:35:46 +0000 (15:35 +0000)]
Back out last commit (did I get all of them?)

Submitted by: Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>

25 years agoRemove vinum from the build until Greg reviews phk's cdevsw changes.
jb [Wed, 2 Jun 1999 07:15:17 +0000 (07:15 +0000)]
Remove vinum from the build until Greg reviews phk's cdevsw changes.

Preferred by: phk (rather than committing the patch without review).

25 years agoSpelling corrections for dummynet.
ru [Wed, 2 Jun 1999 05:59:48 +0000 (05:59 +0000)]
Spelling corrections for dummynet.
Reviewed by: des,luigi

25 years agoDon't print backspaces in first column since they backup to the prev line.
hoek [Wed, 2 Jun 1999 03:42:39 +0000 (03:42 +0000)]
Don't print backspaces in first column since they backup to the prev line.

25 years agoIntroduce the ``keep-session'' option. Refer to the man
brian [Wed, 2 Jun 1999 00:46:55 +0000 (00:46 +0000)]
Introduce the ``keep-session'' option.  Refer to the man
page for details.  This allows MP over non-tty devices where
the original ppp process must not exit (such as sshd-spawned
ppp sessions).

25 years agoUnbreak memory sizing for SMP.
jlemon [Tue, 1 Jun 1999 23:45:48 +0000 (23:45 +0000)]
Unbreak memory sizing for SMP.

25 years ago|| vs && confusion in cdevsw_add().
dt [Tue, 1 Jun 1999 20:41:26 +0000 (20:41 +0000)]
|| vs && confusion in cdevsw_add().

25 years agoRemove an unused variable.
dt [Tue, 1 Jun 1999 20:29:58 +0000 (20:29 +0000)]
Remove an unused variable.

25 years agocdevsw_add(&cn_devsw), not cdevsw_add(&dev).
dt [Tue, 1 Jun 1999 20:26:04 +0000 (20:26 +0000)]
cdevsw_add(&cn_devsw), not cdevsw_add(&dev).

25 years agoGrok gtags too.
hoek [Tue, 1 Jun 1999 20:02:34 +0000 (20:02 +0000)]
Grok gtags too.

25 years agoIncrease the length of an individual device name to LINE_LEN.
brian [Tue, 1 Jun 1999 19:08:59 +0000 (19:08 +0000)]
Increase the length of an individual device name to LINE_LEN.
Adjust the base physical device name correctly after a link
transfer (allowing correct multilink callbacks).

25 years agoChange the warning message issued if pci_map_port() fails. I think this
wpaul [Tue, 1 Jun 1999 19:04:23 +0000 (19:04 +0000)]
Change the warning message issued if pci_map_port() fails. I think this
happens if you have a BIOS with a 'Plug & Play OS' setting and you leave
it set to 'Yes.' This is wrong for FreeBSD (and LoseNT): it should be set
to 'No.' Apparently it's still possible to map the iobase of the NIC and
have the card work by reading the config space manually (which is what
the driver does if pci_map_port() fails) but we need to warn the user to
do fix their machine anyway. Anyway, warn the user to check the 'Plug &
Play OS' setting in their BIOS if mapping the io space fails.

25 years agoIntroduce the makebdev() function, it does the same as the makedev()
phk [Tue, 1 Jun 1999 18:56:26 +0000 (18:56 +0000)]
Introduce the makebdev() function, it does the same as the makedev()
function for now, but that will change.

25 years agoNull commit; note that there is a new memory sizing routine that uses
jlemon [Tue, 1 Jun 1999 18:25:26 +0000 (18:25 +0000)]
Null commit; note that there is a new memory sizing routine that uses
the BIOS calls to determine the memory configuration.  This should fix
problems with >64M for good.

Reviewed by: Mike Smith

25 years agoUnifdef VM86.
jlemon [Tue, 1 Jun 1999 18:20:36 +0000 (18:20 +0000)]
Unifdef VM86.

Reviewed by: silence on on -current

25 years agoMake vm86 a standard component
jlemon [Tue, 1 Jun 1999 18:18:39 +0000 (18:18 +0000)]
Make vm86 a standard component

Reviewed by: silence on on -current

25 years agoUnifdef VM86
jlemon [Tue, 1 Jun 1999 18:17:50 +0000 (18:17 +0000)]
Unifdef VM86

Reviewed by: silence on on -current

25 years agoUpdated to support new Meteor/PPB cards (PCI-PCI Bridge).
roger [Tue, 1 Jun 1999 17:30:08 +0000 (17:30 +0000)]
Updated to support new Meteor/PPB cards (PCI-PCI Bridge).
The driver now identifies the IBM PCI-PCI Bridge fitted to newer
Matrox cards and initialises it.

Sumitted by: Anton Berezin <tobez@plab.ku.dk>
The Protein Laboratory, University of Copenhagen

25 years agoShorten a detour around dev_t to get a udev_t created.
phk [Tue, 1 Jun 1999 17:11:27 +0000 (17:11 +0000)]
Shorten a detour around dev_t to get a udev_t created.

25 years agoMake async/sync/physical/hdlc dumps prettier by showing printable
brian [Tue, 1 Jun 1999 16:01:48 +0000 (16:01 +0000)]
Make async/sync/physical/hdlc dumps prettier by showing printable
characters at the end of the line in hexdump style.

25 years agoFixed locations of include files.
kato [Tue, 1 Jun 1999 13:14:35 +0000 (13:14 +0000)]
Fixed locations of include files.

25 years agoGet rid of segfaults in a set-uid program.
ru [Tue, 1 Jun 1999 13:14:07 +0000 (13:14 +0000)]
Get rid of segfaults in a set-uid program.
PR: 11823
Reviewed by: des

25 years agoSync with sys/i386/isa/clock.c revision 1.136.
kato [Tue, 1 Jun 1999 12:32:54 +0000 (12:32 +0000)]
Sync with sys/i386/isa/clock.c revision 1.136.

25 years agoDamn, forgot this: time -> when.
ru [Tue, 1 Jun 1999 11:20:15 +0000 (11:20 +0000)]
Damn, forgot this: time -> when.

25 years agoCorrect dumb mistake in previous commit.
dfr [Tue, 1 Jun 1999 11:16:23 +0000 (11:16 +0000)]
Correct dumb mistake in previous commit.

25 years agoSync header line with newsyslog(8).
ru [Tue, 1 Jun 1999 10:57:38 +0000 (10:57 +0000)]
Sync header line with newsyslog(8).
PR: 11511
Reviewed by: des

25 years agoIgnore the quota.user and quota.group files rather than the
brian [Tue, 1 Jun 1999 10:47:53 +0000 (10:47 +0000)]
Ignore the quota.user and quota.group files rather than the
``quotas'' file when cleaning /tmp.

25 years agoMove fd driver back to files.${arch} since pc98 has a special fd driver.
dfr [Tue, 1 Jun 1999 09:02:27 +0000 (09:02 +0000)]
Move fd driver back to files.${arch} since pc98 has a special fd driver.

25 years agodevsw and cdevsw_add() changes for i4b.
phk [Tue, 1 Jun 1999 08:46:56 +0000 (08:46 +0000)]
devsw and cdevsw_add() changes for i4b.

Reviewed by: hm

25 years agoMention that using MYADDR and HISADDR with ``set filter''
brian [Tue, 1 Jun 1999 08:46:53 +0000 (08:46 +0000)]
Mention that using MYADDR and HISADDR with ``set filter''
will result in the rules being updated any time MYADDR
or HISADDR change.

25 years agoUSB stopped working as of the recent cdevsw cleanup. This fixes that.
n_hibma [Tue, 1 Jun 1999 07:22:01 +0000 (07:22 +0000)]
USB stopped working as of the recent cdevsw cleanup. This fixes that.

25 years agoRemove the stub driver. It's useless.
n_hibma [Tue, 1 Jun 1999 07:18:53 +0000 (07:18 +0000)]
Remove the stub driver. It's useless.

25 years agousbd_get_request_status now only returns the value if the associated
n_hibma [Tue, 1 Jun 1999 06:39:03 +0000 (06:39 +0000)]
usbd_get_request_status now only returns the value if the associated
pointer is passed.

25 years ago'make update' in the ports directory.
billf [Tue, 1 Jun 1999 02:55:44 +0000 (02:55 +0000)]
'make update' in the ports directory.

25 years agoRemember if MYADDR or HISADDR is used in a filter add tweak all
brian [Mon, 31 May 1999 23:57:40 +0000 (23:57 +0000)]
Remember if MYADDR or HISADDR is used in a filter add tweak all
filters any time either value changes.

25 years ago1) Remove the definition of usb_cdc_generic_descriptor_t. It is
n_hibma [Mon, 31 May 1999 22:35:55 +0000 (22:35 +0000)]
1) Remove the definition of usb_cdc_generic_descriptor_t. It is
equivalent to usb_descriptor_t

2) Rename USB_CDC_CM_CM_OVER_DATA to USB_CDC_CM_OVER_DATA

25 years agoOn the new Meteor cards, the Philips SAA 7116 is connected to the PCI bus
roger [Mon, 31 May 1999 22:13:37 +0000 (22:13 +0000)]
On the new Meteor cards, the Philips SAA 7116 is connected to the PCI bus
via an IBM PCI-PCI bridge (82351 or 82352 or 82353)

The driver must identify if it is on a secondary PCI bus, which is
created via the IBM PCI-PCI bridge. If it is, then it must initialise
the IBM PCI-PCI bridge correctly.

To do this, the following new functions are added.
Because they use the pcici_t tag, they are considered 2.2 compatibility APIs
  pcici_t * pci_get_parent_from_tag(pcici_t tag);
  int       pci_get_bus_from_tag(pcici_t tag);

(The _from_tag suffix is used to prevent clashes with similarly named
 newbus PCI API functions)

Submitted by: Anton Berezin <tobez@plab.ku.dk>
Reviewed by:  Doug Rabson <dfr@nlsystems.com>
Reworked by:  Me (roger)

25 years agoDon't use nblkdev.
dt [Mon, 31 May 1999 20:44:11 +0000 (20:44 +0000)]
Don't use nblkdev.

25 years agoRemove declaration of the nblkdev and nchrdev variables.
dt [Mon, 31 May 1999 20:42:41 +0000 (20:42 +0000)]
Remove declaration of the nblkdev and nchrdev variables.

25 years ago* Change include file locations.
dfr [Mon, 31 May 1999 18:39:17 +0000 (18:39 +0000)]
* Change include file locations.
* Fix some misunderstandings about the return value of resource_int_value().
* Make it build on alpha (doesn't work yet...)

25 years agoRemove fd driver from its old home and change files which include rtc.h
dfr [Mon, 31 May 1999 18:36:14 +0000 (18:36 +0000)]
Remove fd driver from its old home and change files which include rtc.h
to account for its new location.

25 years agoThe fd driver has moved from i386/isa to isa.
dfr [Mon, 31 May 1999 18:33:24 +0000 (18:33 +0000)]
The fd driver has moved from i386/isa to isa.

25 years agoSimplify cdevsw registration.
phk [Mon, 31 May 1999 11:29:30 +0000 (11:29 +0000)]
Simplify cdevsw registration.

The cdevsw_add() function now finds the major number(s) in the
struct cdevsw passed to it.  cdevsw_add_generic() is no longer
needed, cdevsw_add() does the same thing.

cdevsw_add() will print an message if the d_maj field looks bogus.

Remove nblkdev and nchrdev variables.  Most places they were used
bogusly.  Instead check a dev_t for validity by seeing if devsw()
or bdevsw() returns NULL.

Move bdevsw() and devsw() functions to kern/kern_conf.c

Bump __FreeBSD_version to 400006

This commit removes:
        72 bogus makedev() calls
        26 bogus SYSINIT functions

if_xe.c bogusly accessed cdevsw[], author/maintainer please fix.

I4b and vinum not changed.  Patches emailed to authors.  LINT
probably broken until they catch up.

25 years agoYou'll see them eventually :-)
phk [Mon, 31 May 1999 07:43:44 +0000 (07:43 +0000)]
You'll see them eventually :-)

25 years agoThe PPSAPI group defines the polarity in draft 3, we guessed it wrong.
phk [Mon, 31 May 1999 06:57:31 +0000 (06:57 +0000)]
The PPSAPI group defines the polarity in draft 3, we guessed it wrong.

25 years agoAdd the 'advocacy' category, sort this document.
billf [Sun, 30 May 1999 23:59:36 +0000 (23:59 +0000)]
Add the 'advocacy' category, sort this document.

25 years agoprocedure, not proceedure.
billf [Sun, 30 May 1999 21:31:18 +0000 (21:31 +0000)]
procedure, not proceedure.

25 years agoAdd support for SiS 900 chipset (PCI ID 0x1039/0x0900), which appears to
wpaul [Sun, 30 May 1999 18:55:20 +0000 (18:55 +0000)]
Add support for SiS 900 chipset (PCI ID 0x1039/0x0900), which appears to
be yet another rebadged RealTek 8139.

25 years agoEnable attachment of multiple drivers to a single device.
n_hibma [Sun, 30 May 1999 18:49:17 +0000 (18:49 +0000)]
Enable attachment of multiple drivers to a single device.

25 years agoNotify CAM of the removed device on detach.
n_hibma [Sun, 30 May 1999 18:46:04 +0000 (18:46 +0000)]
Notify CAM of the removed device on detach.

25 years agoWhoops, forgot to update xl_mediacheck() to account for the 3c905C.
wpaul [Sun, 30 May 1999 18:11:47 +0000 (18:11 +0000)]
Whoops, forgot to update xl_mediacheck() to account for the 3c905C.

25 years agoHead for the hills friends and neighbors, 3Com has yet another 3c90X
wpaul [Sun, 30 May 1999 18:09:17 +0000 (18:09 +0000)]
Head for the hills friends and neighbors, 3Com has yet another 3c90X
chipset. First you thrilled to the 3c905, then you trembled at the
3c905B, now gaze in wonder at: the 3c905C! This appears to be another
3c90X series chip called the Tornado (PCI ID 0x10B7/0x9200) and should
be equivalent (from the driver API perspective) to the 3c905B, so all
we have to do is add the PCI ID to the list.