]> CyberLeo.Net >> Repos - FreeBSD/FreeBSD.git/log
FreeBSD/FreeBSD.git
24 years agoFix breakage when NO_RSA specified.
Kris Kennaway [Fri, 14 Jan 2000 05:24:08 +0000 (05:24 +0000)]
Fix breakage when NO_RSA specified.

Reviewed by: Ben Laurie <ben@openssl.org>

24 years agoDon't do device_set_desc() until after checking for PnP probes. Otherwise
Peter Wemm [Fri, 14 Jan 2000 05:03:23 +0000 (05:03 +0000)]
Don't do device_set_desc() until after checking for PnP probes.  Otherwise
things like sound cards can get called "Parallel port".  A note to the
unwary; the isa-pnp devices in the system are probed like PCI - each
device ID is passed to *all* isa probe routines to find the best match.
If the driver is not prepared to deal with this, it must abort in this
scenario or it will try and claim all PnP devices.

24 years ago Non-operational change, fix compiler warning.
Matthew Dillon [Fri, 14 Jan 2000 04:39:28 +0000 (04:39 +0000)]
Non-operational change, fix compiler warning.

Reviewed by:  mckusick

24 years agoThe error status for a scsi status error is "CAM_SCSI_STATUS_ERROR", not 0.
Justin T. Gibbs [Fri, 14 Jan 2000 03:39:30 +0000 (03:39 +0000)]
The error status for a scsi status error is "CAM_SCSI_STATUS_ERROR", not 0.

24 years agoadv_pci.c:
Justin T. Gibbs [Fri, 14 Jan 2000 03:33:38 +0000 (03:33 +0000)]
adv_pci.c:
Update list of supported products.
Adjust probe message to include the ASC3030.

advansys.c:
Fix a long standing bug in the error recovery strategy.  In order
to keep recovery simple, we freeze the SIMQ, stopping the XPT from
submitting new requests.  Unfortunately, we also will freeze the
SIMQ if bus_dmamap_load blocks or we run out of controller resources.
On cards with limited resources it was possible to freeze the
SIM a second time and never unfreeze it.  Now we more carefully
track our exception state so we never freeze the SIMQ more than
once.

Don't rely on pointers fitting in a 32bit field stored in the
per-transaction data structures on the card.  Use an index to
an array of transaction mapping structures instead.  This should
allow this driver to work on the Alpha.

Deal with the ASC3030 which is almost idistinguishable from the
ASC3050.  Unfortunately the ASC3030 does not work at Ultra speeds,
so if we can't find an eeprom, we must assume that ultra is disabled.
The SIIG cards using the 3030 do not have eeproms.  As a side effect,
we now honor the ultra disable bit in the eeprom if it is present.

Don't bother attempting to write corrected eeprom data back to the
eeprom.  We can function just fine if the data is corrupted and
I'd rather not risk messing up the user's eeprom.

Modify the interrupt handler to catch latched external bus rests.

Dynamically determine the maximum number of S/G elements we can
map at a single time.  The nature of the firmware interface for
these cards makes this value dependent on the number of "queues"
the card can support.

advlib.c:
advlib.h:
advmcode.c:
advmcode.h:
Synchronize with the latest firmware image released in the
Linux Advansys driver.

24 years agoAdd device driver support for USB ethernet adapters based on the CATC
Bill Paul [Fri, 14 Jan 2000 03:14:49 +0000 (03:14 +0000)]
Add device driver support for USB ethernet adapters based on the CATC
USB-EL1202A chipset. Between this and the other two drivers, we should
have support for pretty much every USB ethernet adapter on the market.
The only other USB chip that I know of is the SMC USB97C196, and right
now I don't know of any adapters that use it (including the ones made
by SMC :/ ).

Note that the CATC chip supports a nifty feature: read and write combining.
This allows multiple ethernet packets to be transfered in a single USB
bulk in/out transaction. However I'm again having trouble with large
bulk in transfers like I did with the ADMtek chip, which leads me to
believe that our USB stack needs some work before we can really make
use of this feature. When/if things improve, I intend to revisit the
aue and cue drivers. For now, I've lost enough sanity points.

24 years agoAdd aio_waitcomplete(). Make aio work correctly for socket descriptors.
Jason Evans [Fri, 14 Jan 2000 02:53:29 +0000 (02:53 +0000)]
Add aio_waitcomplete().  Make aio work correctly for socket descriptors.
Make gratuitous style(9) fixes (me, not the submitter) to make the aio
code more readable.

PR: kern/12053
Submitted by: Chris Sedore <cmsedore@maxwell.syr.edu>

24 years agoEliminate PC-card installation floppy and add PC-card support for
Tatsumi Hosokawa [Fri, 14 Jan 2000 02:17:57 +0000 (02:17 +0000)]
Eliminate PC-card installation floppy and add PC-card support for
generic installation floppy.

24 years agoPull my head out of my ass and actually make the tx netisr stuff work right.
Bill Paul [Fri, 14 Jan 2000 01:36:16 +0000 (01:36 +0000)]
Pull my head out of my ass and actually make the tx netisr stuff work right.
Do not not not call m_freem() in the txeof routines. Let the netisr routine
do it. This also makes the tx netisr queuing much simpler (I can just use
another ifqueue instead of the mess I had before.)

Thanks to Bosko Milekic for making me actually think about what I was
doing for a minute.

24 years agoClean up rxeof routines a little.
Bill Paul [Fri, 14 Jan 2000 00:49:28 +0000 (00:49 +0000)]
Clean up rxeof routines a little.

24 years agoPort of ppbus standalone framework to the newbus system.
Nicolas Souchu [Fri, 14 Jan 2000 00:18:06 +0000 (00:18 +0000)]
Port of ppbus standalone framework to the newbus system.

Note1: the correct interrupt level is invoked correctly for each driver.
       For this purpose, drivers request the bus before being able to
       call BUS_SETUP_INTR and BUS_TEARDOWN_INTR call is forced by the ppbus
       core when drivers release it. Thus, when BUS_SETUP_INTR is called
       at ppbus driver level, ppbus checks that the caller owns the
       bus and stores the interrupt handler cookie (in order to unregister
       it later).

       Printing is impossible while plip link is up is still TRUE.
       vpo (ZIP driver) and lpt are make in such a way that
       using the ZIP and printing concurrently is permitted is also TRUE.

Note2: specific chipset detection is not done by default. PPC_PROBE_CHIPSET
       is now needed to force chipset detection. If set, the flags 0x40
       still avoid detection at boot.

Port of the pcf(4) driver to the newbus system (was previously directly
connected to the rootbus and attached by a bogus pcf_isa_probe function).

24 years agoFix instance of AUE_BUFSZ that should have been AUE_CUTOFF.
Bill Paul [Thu, 13 Jan 2000 23:40:00 +0000 (23:40 +0000)]
Fix instance of AUE_BUFSZ that should have been AUE_CUTOFF.

24 years ago Thresh-out the nfs manual page references a bit
Matthew Dillon [Thu, 13 Jan 2000 21:47:21 +0000 (21:47 +0000)]
Thresh-out the nfs manual page references a bit

Reviewed by: Julian Elischer <julian@elischer.org>

24 years agoRemove debug printf I left in by mistake.
Bill Paul [Thu, 13 Jan 2000 20:53:49 +0000 (20:53 +0000)]
Remove debug printf I left in by mistake.

24 years agoReadd ipfilter devices
Guido van Rooij [Thu, 13 Jan 2000 20:27:22 +0000 (20:27 +0000)]
Readd ipfilter devices

PR: 16093

24 years ago The alpha build cuases the 'nfsuid bloated' warning to occur. Well,
Matthew Dillon [Thu, 13 Jan 2000 20:18:25 +0000 (20:18 +0000)]
The alpha build cuases the 'nfsuid bloated' warning to occur.  Well,
    there is nothing we can do about it.  In fact, after further review
    there simply are not very many instances of the two structures NFS
    checks for 'bloat' so I've decided to simply rip the checks out entirely.

Submitted by:  Andrew Gallatin <gallatin@cs.duke.edu>

24 years agoRegenerate.
Bill Paul [Thu, 13 Jan 2000 20:17:11 +0000 (20:17 +0000)]
Regenerate.

24 years agoBunch of updates:
Bill Paul [Thu, 13 Jan 2000 20:13:58 +0000 (20:13 +0000)]
Bunch of updates:

- Add vendor/device ID for Corega USB-T ethernet adapter to necessary
  places so that it will work with the kue driver.

- Add vendor/device ID for CATC Netmate devices for driver to be added
  soon.

- Get really crazy about netisr stuff: avoid doing any mbuf allocations
  or deallocations at splbio/splusb.

- Fix if_aue driver so that it works with LinkSys USB100TX: you need
  to flip the GPIO bits just the right way to put the PHY in the right
  mode.

24 years agoConfirming Peter's fix (locking 101: release the lock before you go
Kirk McKusick [Thu, 13 Jan 2000 20:03:22 +0000 (20:03 +0000)]
Confirming Peter's fix (locking 101: release the lock before you go
to sleep). Locking 101, part 2: do not look at buffer contents after
you have been asleep. There is no telling what wonderous changes may
have occurred.

24 years agoAdd ipfilter 3.3.6
Guido van Rooij [Thu, 13 Jan 2000 20:00:48 +0000 (20:00 +0000)]
Add ipfilter 3.3.6

24 years agoBring over ipfilter kernel sources, including merging the local modifications.
Guido van Rooij [Thu, 13 Jan 2000 19:01:33 +0000 (19:01 +0000)]
Bring over ipfilter kernel sources, including merging the local modifications.

24 years agoFree the global softupdates lock prior to tsleep() in getdirtybuf().
Peter Wemm [Thu, 13 Jan 2000 18:48:12 +0000 (18:48 +0000)]
Free the global softupdates lock prior to tsleep() in getdirtybuf().
This seems to be responsible for a bunch of panics where the process
sleeps and something else finds softupdates "locked" when it shouldn't
be.  This commit is unreviewed, but has been a big help here.
Previously my boxes would panic pretty much on the first fsync() that
wrote something to disk.

24 years agoadd MAINTAINER file for clarity
Matt Jacob [Thu, 13 Jan 2000 18:45:14 +0000 (18:45 +0000)]
add MAINTAINER file for clarity

24 years agoThis commit was generated by cvs2svn to compensate for changes in r55924,
Guido van Rooij [Thu, 13 Jan 2000 18:30:37 +0000 (18:30 +0000)]
This commit was generated by cvs2svn to compensate for changes in r55924,
which included commits to RCS files with non-trunk default branches.

24 years agoImport of ipfilter 3.3.6 (freebsd relevant part)
Guido van Rooij [Thu, 13 Jan 2000 18:30:37 +0000 (18:30 +0000)]
Import of ipfilter 3.3.6 (freebsd relevant part)

Obtained from: ftp://coombs.anu.edu.au/pub/net/firewall/ip-filter/ip_fil3.3.6.tar.gz

24 years agoFix dhclient-exit-hooks location. (PR 14253)
Jeroen Ruigrok van der Werven [Thu, 13 Jan 2000 16:27:20 +0000 (16:27 +0000)]
Fix dhclient-exit-hooks location. (PR 14253)
Add $FreeBSD$ CVS identifier.

PR: 14253
Submitted by: Arcady Genkin

24 years agofix wrong name which is hidden by wrong ifdef.
Yoshinobu Inoue [Thu, 13 Jan 2000 15:27:50 +0000 (15:27 +0000)]
fix wrong name which is hidden by wrong ifdef.
Sorry for build failure. There was a mistake when I moved the patch
from my build check machine to commit machine.

Specified by: peter

24 years agolibc rcmd update for IPv6.
Yoshinobu Inoue [Thu, 13 Jan 2000 15:09:48 +0000 (15:09 +0000)]
libc rcmd update for IPv6.
A new function bindresvport2(), AF independent version of bindresvport()
is also added.

Reviewed by: sumikawa
Obtained from: KAME project

24 years agoChange struct sockaddr_storage member name, because following change
Yoshinobu Inoue [Thu, 13 Jan 2000 14:52:53 +0000 (14:52 +0000)]
Change struct sockaddr_storage member name, because following change
is very likely to become consensus as recent ietf/ipng mailing list
discussion. Also recent KAME repository and other KAME patched BSDs
also applied it.

  s/__ss_family/ss_family/
  s/__ss_len/ss_len/

Makeworld is confirmed, and no application should be affected by this change
yet.

24 years agoCorrect placement of $FreeBSD$ CVS identifier.
Jeroen Ruigrok van der Werven [Thu, 13 Jan 2000 14:27:44 +0000 (14:27 +0000)]
Correct placement of $FreeBSD$ CVS identifier.

24 years agoChange ``from'' to ``to''.
Jeroen Ruigrok van der Werven [Thu, 13 Jan 2000 14:26:23 +0000 (14:26 +0000)]
Change ``from'' to ``to''.

PR: 15729
Submitted by: Kim Toms

24 years agoClear rt after RTFREE. This might have sometime caused kernel panic at rtfree()
Yoshinobu Inoue [Thu, 13 Jan 2000 14:21:30 +0000 (14:21 +0000)]
Clear rt after RTFREE. This might have sometime caused kernel panic at rtfree()
on INET6 enabled environment.

24 years agoAdd the LSI-Logic 53C875e chipset and the DawiControl DC2976UW entries
Jeroen Ruigrok van der Werven [Thu, 13 Jan 2000 13:43:21 +0000 (13:43 +0000)]
Add the LSI-Logic 53C875e chipset and the DawiControl DC2976UW entries
to our hardware list.

PR: 15744
Submitted by: Sascha Blank <blank@uni-tier.de>

24 years agoAdd missing 'DEVICE_SYSCTLS' to opt_bus.h. This is the (experimental?)
Peter Wemm [Thu, 13 Jan 2000 13:29:51 +0000 (13:29 +0000)]
Add missing 'DEVICE_SYSCTLS' to opt_bus.h.  This is the (experimental?)
code that exports the bus heirarchy to hw.devices.*

24 years agoAdd myself as ntp docs maintainer, with roberto's permission.
Sheldon Hearn [Thu, 13 Jan 2000 13:02:00 +0000 (13:02 +0000)]
Add myself as ntp docs maintainer, with roberto's permission.

24 years agoFix page fault in -vv mode.
Ruslan Ermilov [Thu, 13 Jan 2000 12:59:58 +0000 (12:59 +0000)]
Fix page fault in -vv mode.

PR: 16098
Submitted by: Alan.Judge@indigo.ie
Reviewed by: ru

24 years agoAdd dhclient(8) crossreference to the dhcp_flags section.
Jeroen Ruigrok van der Werven [Thu, 13 Jan 2000 12:53:54 +0000 (12:53 +0000)]
Add dhclient(8) crossreference to the dhcp_flags section.

Submitted by: sheldonh

24 years agoSynced with sys/dev/syscons/syscons.c rev 1.331.
KATO Takenori [Thu, 13 Jan 2000 12:24:43 +0000 (12:24 +0000)]
Synced with sys/dev/syscons/syscons.c rev 1.331.

24 years agoFix old style /fB /fR emphasis to the correct usage of .Em or other
Jeroen Ruigrok van der Werven [Thu, 13 Jan 2000 12:06:41 +0000 (12:06 +0000)]
Fix old style /fB /fR emphasis to the correct usage of .Em or other
more appropriate uses of mdoc macros given the context they were
used in.

24 years agoAdd crossreference to dhclient(8) since our interfaces nowadays
Jeroen Ruigrok van der Werven [Thu, 13 Jan 2000 11:54:35 +0000 (11:54 +0000)]
Add crossreference to dhclient(8) since our interfaces nowadays
support DHCP and more information would be handy.

24 years agoSynced with following changes:
KATO Takenori [Thu, 13 Jan 2000 11:47:42 +0000 (11:47 +0000)]
Synced with following changes:

>yokota      2000/01/11 05:39:06 PST
>
>  Modified files:
>    sys/dev/usb          ukbd.c
>    sys/dev/kbd          atkbd.c kbd.c kbdreg.h
>  Log:
>  Rework shifta/ctla/alta key handling.  It appears that there was
>  misunderstanding between the PR originator and me.  I hope I got it
>  right this time.
>
>  Revision  Changes    Path
>  1.22      +4 -1      src/sys/dev/usb/ukbd.c
>  1.21      +1 -8      src/sys/dev/kbd/atkbd.c
>  1.16      +19 -10    src/sys/dev/kbd/kbd.c
>  1.9       +2 -2      src/sys/dev/kbd/kbdreg.h

Submitted by: yokota

24 years agoSynced with sys/isa/sio.c rev 1.284.
KATO Takenori [Thu, 13 Jan 2000 11:45:33 +0000 (11:45 +0000)]
Synced with sys/isa/sio.c rev 1.284.

24 years agoAdd description for how to enable DHCP for network interfaces.
Jeroen Ruigrok van der Werven [Thu, 13 Jan 2000 11:44:28 +0000 (11:44 +0000)]
Add description for how to enable DHCP for network interfaces.

PR: 15980
Submitted by: John Reynolds <jjreynold@home.com>
Reviewed by: sheldonh

24 years agoFixed missing include and missing arg in synopsis.
Bruce Evans [Thu, 13 Jan 2000 10:22:50 +0000 (10:22 +0000)]
Fixed missing include and missing arg in synopsis.

24 years agoFixed missing include in synopsis.
Bruce Evans [Thu, 13 Jan 2000 10:21:25 +0000 (10:21 +0000)]
Fixed missing include in synopsis.

24 years agoMove the Variables section of the old ntp_misc(8) manual page to
Sheldon Hearn [Thu, 13 Jan 2000 10:03:53 +0000 (10:03 +0000)]
Move the Variables section of the old ntp_misc(8) manual page to
the correct location in the ntpd(8) page.

24 years agoAbandon hope of keeping in line with the author's structure,
Sheldon Hearn [Thu, 13 Jan 2000 09:59:55 +0000 (09:59 +0000)]
Abandon hope of keeping in line with the author's structure,
in favour of placing information in the correct sections.

The ntp_acc(8), ntp_auth(8), ntp_clock(8), ntp_conf(8),
ntp_misc(8) and ntp_mon(8) pages have been merged into
ntp.conf(5) and ntp.keys(5).

Requested by: rgrimes, wollman

24 years agoRemove old copy of if_ed.c; it is no longer used.
Matthew N. Dodd [Thu, 13 Jan 2000 09:30:18 +0000 (09:30 +0000)]
Remove old copy of if_ed.c; it is no longer used.

24 years agoFix unresolved _libc_*() references in libc by creating weak aliases
Jason Evans [Thu, 13 Jan 2000 09:26:50 +0000 (09:26 +0000)]
Fix unresolved _libc_*() references in libc by creating weak aliases
to the respective system call entry points.

24 years agoSet ifq_maxlen. (to default IFQ_MAXLEN)
Matthew N. Dodd [Thu, 13 Jan 2000 09:13:22 +0000 (09:13 +0000)]
Set ifq_maxlen. (to default IFQ_MAXLEN)

24 years agoAllow SMP systems with an MCA bus to work properly.
Matthew N. Dodd [Thu, 13 Jan 2000 09:09:02 +0000 (09:09 +0000)]
Allow SMP systems with an MCA bus to work properly.

Reviewed by: peter

24 years agoImplement BUS_{GET,SET,DELETE}_RESOURCE methods.
Matthew N. Dodd [Thu, 13 Jan 2000 09:01:46 +0000 (09:01 +0000)]
Implement BUS_{GET,SET,DELETE}_RESOURCE methods.

24 years agoCorrectly test CF.
Matthew N. Dodd [Thu, 13 Jan 2000 08:49:47 +0000 (08:49 +0000)]
Correctly test CF.

24 years agoRemove duplicate DEC 21050 PCI-PCI bridge (0x00011011)
Peter Wemm [Thu, 13 Jan 2000 08:40:10 +0000 (08:40 +0000)]
Remove duplicate DEC 21050 PCI-PCI bridge (0x00011011)
Add an entry for the Toshiba Fast Infra Red controller (0x07011179)

24 years agoAdd a few devices IDs and clarify the SiS 85c496 entry.
Matthew N. Dodd [Thu, 13 Jan 2000 08:01:53 +0000 (08:01 +0000)]
Add a few devices IDs and clarify the SiS 85c496 entry.

24 years agoBecause cylinder group blocks are now written in background,
Kirk McKusick [Thu, 13 Jan 2000 07:20:01 +0000 (07:20 +0000)]
Because cylinder group blocks are now written in background,
it is no longer sufficient to get a lock on a buffer to know
that its write has been completed. We have to first get the
lock on the buffer, then check to see if it is doing a
background write. If it is doing background write, we have
to wait for the background write to finish, then check to see
if that fullfilled our dependency, and if not to start another
write. Luckily the explanation is longer than the fix.

24 years agoA panic occurs during an fsync when a dirty block associated with
Kirk McKusick [Thu, 13 Jan 2000 07:17:39 +0000 (07:17 +0000)]
A panic occurs during an fsync when a dirty block associated with
a vnode has not been written (which would clear certain of its
dependencies). The problems arises because fsync with MNT_NOWAIT
no longer pushes all the dirty blocks associated with a vnode. It
skips those that require rollbacks, since they will just get instantly
dirty again. Such skipped blocks are marked so that they will not be
skipped a second time (otherwise circular dependencies would never
clear). So, we fsync twice to ensure that everything will be written
at least once.

24 years agoRemove the 'at isa? ...' bits for ex0.
Matthew N. Dodd [Thu, 13 Jan 2000 07:01:13 +0000 (07:01 +0000)]
Remove the 'at isa? ...' bits for ex0.

Remove the confusing text about pccard and unit numbers for ep0.

24 years agoCosmetic cleanups.
Matthew N. Dodd [Thu, 13 Jan 2000 06:55:35 +0000 (06:55 +0000)]
Cosmetic cleanups.

24 years agoImplement a DEVICE_IDENTIFY method. I want to revisit some of this later
Matthew N. Dodd [Thu, 13 Jan 2000 06:52:51 +0000 (06:52 +0000)]
Implement a DEVICE_IDENTIFY method.  I want to revisit some of this later
but this is enough to get us going.

24 years agoWhen I converted this driver to newbus I also cleaned up the code
Matthew N. Dodd [Thu, 13 Jan 2000 06:46:02 +0000 (06:46 +0000)]
When I converted this driver to newbus I also cleaned up the code
layout.  It seems that I cleaned it up a bit too much and confused a few

if () {
if () {
} else {
}
}

statements in the obvious manner.

This allows the driver to transmit packets again.  *sigh*

24 years agoStop the recurring feeling of deja vu
Warner Losh [Thu, 13 Jan 2000 06:32:33 +0000 (06:32 +0000)]
Stop the recurring feeling of deja vu
Stop the recurring feeling of deja vu
Stop the recurring feeling of deja vu
Stop the recurring feeling of deja vu

and debounce the eject messages.  We now mark the socket empty in the
interrupt handler, rather than after we've disabled the socket which
happens "much later".

24 years agoattach to non-ac97 neomagic chips and hope that ac97 ones do not have the
Cameron Grant [Thu, 13 Jan 2000 06:11:32 +0000 (06:11 +0000)]
attach to non-ac97 neomagic chips and hope that ac97 ones do not have the
same pnpbios id

24 years agobe less verbose
Cameron Grant [Thu, 13 Jan 2000 06:00:57 +0000 (06:00 +0000)]
be less verbose

24 years agoadded IPv6 unspecified addr check for getipnodebyaddr.
Yoshinobu Inoue [Thu, 13 Jan 2000 05:47:11 +0000 (05:47 +0000)]
added IPv6 unspecified addr check for getipnodebyaddr.

24 years agoadded IPv4 mapped IPv6 addr consideration for getaddrinfo() reverse lookup case
Yoshinobu Inoue [Thu, 13 Jan 2000 05:37:51 +0000 (05:37 +0000)]
added IPv4 mapped IPv6 addr consideration for getaddrinfo() reverse lookup case

24 years agoadd a comment for some possible? IPv4 option processing.
Yoshinobu Inoue [Thu, 13 Jan 2000 05:21:05 +0000 (05:21 +0000)]
add a comment for some possible? IPv4 option processing.

24 years agoremoved incorrect ip6 length setting for IPv6 tcp reset packet.
Yoshinobu Inoue [Thu, 13 Jan 2000 05:18:30 +0000 (05:18 +0000)]
removed incorrect ip6 length setting for IPv6 tcp reset packet.

24 years agoremoved an ours case which think a packet destined to loopback interface
Yoshinobu Inoue [Thu, 13 Jan 2000 05:12:48 +0000 (05:12 +0000)]
removed an ours case which think a packet destined to loopback interface
with IPv6 loopback addr for its dest or src addr as ours.

24 years agoadded missing IPV6_PORTRANGE case.
Yoshinobu Inoue [Thu, 13 Jan 2000 05:07:42 +0000 (05:07 +0000)]
added missing IPV6_PORTRANGE case.

24 years agoremove unnecessary "$ifdef INET6"
Yoshinobu Inoue [Thu, 13 Jan 2000 05:01:27 +0000 (05:01 +0000)]
remove unnecessary "$ifdef INET6"

24 years agoMake mktemp.c a "normal" C implimentation when building this library.
David E. O'Brien [Thu, 13 Jan 2000 04:55:41 +0000 (04:55 +0000)]
Make mktemp.c a "normal" C implimentation when building this library.

24 years agoadd entries for several DEC PCI-PCI bridges & the Cypress PCI-ISA bridge found
Andrew Gallatin [Thu, 13 Jan 2000 02:47:36 +0000 (02:47 +0000)]
add entries for several DEC PCI-PCI bridges & the Cypress PCI-ISA bridge found
on newer alpha workstations and servers

24 years agoMake this compile on alpha
Andrew Gallatin [Thu, 13 Jan 2000 02:09:44 +0000 (02:09 +0000)]
Make this compile on alpha

24 years agoo Remove entry that has 'any' for the interrupt and 'auto' for config.
Warner Losh [Thu, 13 Jan 2000 00:29:06 +0000 (00:29 +0000)]
o Remove entry that has 'any' for the interrupt and 'auto' for config.
  We don't have that yet.
o Uncomment FLASH cards so when the ata flash patches go into the tree
  they will work.

Reported by: hosokawa-san

24 years agoReintroduce the dc_coal() workaround routine for coalescing outbound
Bill Paul [Wed, 12 Jan 2000 22:24:05 +0000 (22:24 +0000)]
Reintroduce the dc_coal() workaround routine for coalescing outbound
packets into a single buffer, and set the DC_TX_COALESCE flag for the
Davicom DM9102 chip. I thought I had escaped this problem, but... This
chip appears to silently corrupt or discard transmitted frames when
using scatter/gather DMA (i.e. DMAing each packet fragment in place
with a separate descriptor). The only way to insure reliable transmission
is to coalesce transmitted packets into a single cluster buffer. (There
may also be an alignment constraint here, but mbuf cluster buffers are
naturally aligned on 2K boundaries, which seems to be good enough.)

The DM9102 driver for Linux written by Davicom also uses this workaround.
Unfortunately, the Davicom datasheet has no errata section describing
this or any other apparently known defect.

Problem noted by: allan_chou@davicom.com.tw

24 years agoAs we parse the interface list, ignore sockaddr_dl's that
Brian Somers [Wed, 12 Jan 2000 21:53:46 +0000 (21:53 +0000)]
As we parse the interface list, ignore sockaddr_dl's that
don't have an interface index that's the same as the if_msghdr
interface index.

This prevents the occasional perror("SIOCGIFFLAGS") from appearing
at boot time.

While I'm there:
  Make a couple of error messages more useful.
  Add a missing include.
  Add some braces to silence gccs dumb complaints.
  Add some consts
  Ansify decls
  Add copyright to pmap_check.h (well, you could say it's been rewritten)

24 years agoRemove a stray splx(s) call. This fixes the compiler warning.
Matthew N. Dodd [Wed, 12 Jan 2000 17:51:01 +0000 (17:51 +0000)]
Remove a stray splx(s) call.  This fixes the compiler warning.

Submitted by: gibbs

24 years agoApply the same netisr mechanism to transmissions as well. In order to
Bill Paul [Wed, 12 Jan 2000 17:46:40 +0000 (17:46 +0000)]
Apply the same netisr mechanism to transmissions as well. In order to
drive the transmitter, we have to check the interface's send queue in the
TX end of frame handler (i.e. the usb bulk out callback) and push out new
transmissions if the queue has packets in it and the transmitter is
ready. But the txeof handler is also called from a USB callback running
at splusb() too.

Grrr.

24 years agoSeconds to ticks conversion was done at the wrong place.
Luoqi Chen [Wed, 12 Jan 2000 17:26:42 +0000 (17:26 +0000)]
Seconds to ticks conversion was done at the wrong place.

24 years agoRemove the cautionary note concerning the move of the chown binary
Sheldon Hearn [Wed, 12 Jan 2000 14:49:51 +0000 (14:49 +0000)]
Remove the cautionary note concerning the move of the chown binary
and its link chgrp, since the change has been reverted.

Reported by: Kelly Yancey <kbyanc@posi.net>

24 years agoWe do not support the -B option.
Jeroen Ruigrok van der Werven [Wed, 12 Jan 2000 14:49:38 +0000 (14:49 +0000)]
We do not support the -B option.

PR: 15925
Submitted by: Thierry Herbelot

24 years agoAdd selected manual pages transcribed from the HTML documentation.
Sheldon Hearn [Wed, 12 Jan 2000 14:41:00 +0000 (14:41 +0000)]
Add selected manual pages transcribed from the HTML documentation.
Those pages which have not been transcribed are referenced as
gracefully as possible.

There is no perfect section for the ntp_* files, which document
configuration options for the NTP suite, so I'm putting them in
the same section as the pages for the utilities themselves.

24 years agoAdd gross hack to work around bogus dependency information created
Marcel Moolenaar [Wed, 12 Jan 2000 14:33:00 +0000 (14:33 +0000)]
Add gross hack to work around bogus dependency information created
by gnu/usr.bin/cc/cc_tools/Makefile. This bug is painfully visible
when making buildworld with -DNOCLEAN. This work around is beyond
dirty...

24 years agoFix a bungle with the CAM static wiring tables. Write CAMCONF_UNSPEC
Peter Wemm [Wed, 12 Jan 2000 14:20:12 +0000 (14:20 +0000)]
Fix a bungle with the CAM static wiring tables.  Write CAMCONF_UNSPEC
instead of -2.  This (I believe) caused static wirings to not match.
This should fix Bill Pechter's problem but we'll see.

Problem discovered by: Bill Pechter <pechter@shell.monmouth.com>

24 years agoMerge from sys/conf/files.i386 revision 1.296.
Yoshihiro Takahashi [Wed, 12 Jan 2000 13:43:23 +0000 (13:43 +0000)]
Merge from sys/conf/files.i386 revision 1.296.

24 years agoAdd necessary tmac files to the tmac Makefile and install them
Jeroen Ruigrok van der Werven [Wed, 12 Jan 2000 13:37:57 +0000 (13:37 +0000)]
Add necessary tmac files to the tmac Makefile and install them
in /usr/share/tmac like the rest.

24 years agoAllow devhtml and grohtml to be built.
Jeroen Ruigrok van der Werven [Wed, 12 Jan 2000 13:17:51 +0000 (13:17 +0000)]
Allow devhtml and grohtml to be built.
This is a new feature of groff and is a html driver for groff.

From the manual page:
"grohtml translates the output of GNU troff to html."

This is very interesting for people working on documentation.

24 years agoAdd bmake glue for devhtml and grohtml of the groff package.
Jeroen Ruigrok van der Werven [Wed, 12 Jan 2000 13:14:27 +0000 (13:14 +0000)]
Add bmake glue for devhtml and grohtml of the groff package.

24 years agoAdd devhtml entry for the groff_font directory.
Jeroen Ruigrok van der Werven [Wed, 12 Jan 2000 13:06:44 +0000 (13:06 +0000)]
Add devhtml entry for the groff_font directory.
This is needed for the new devhtml in the groff 1.15 package.

24 years agoMake the mouse cursor char code configurable via the CONS_MOUSECTL
Kazutaka YOKOTA [Wed, 12 Jan 2000 12:30:33 +0000 (12:30 +0000)]
Make the mouse cursor char code configurable via the CONS_MOUSECTL
ioctl.

By popular demand.

24 years agoAdd support for detection of the ESS 1878 chip.
Daniel Eischen [Wed, 12 Jan 2000 11:16:23 +0000 (11:16 +0000)]
Add support for detection of the ESS 1878 chip.

24 years agoThe groff package is not maintained by jjc@jclark.com anymore.
Jeroen Ruigrok van der Werven [Wed, 12 Jan 2000 10:35:57 +0000 (10:35 +0000)]
The groff package is not maintained by jjc@jclark.com anymore.
Reflect this change in the upgrade file.

24 years agoAllow grog to be made during compilation.
Jeroen Ruigrok van der Werven [Wed, 12 Jan 2000 10:31:55 +0000 (10:31 +0000)]
Allow grog to be made during compilation.

24 years agoAdd bmake glue for grog(1), an utility which guesses groff's
Jeroen Ruigrok van der Werven [Wed, 12 Jan 2000 10:30:27 +0000 (10:30 +0000)]
Add bmake glue for grog(1), an utility which guesses groff's
flags and options.

Requested by: andreas

24 years agoFix conflicts which arose during import, also add $FreeBSD$.
Jeroen Ruigrok van der Werven [Wed, 12 Jan 2000 10:26:36 +0000 (10:26 +0000)]
Fix conflicts which arose during import, also add $FreeBSD$.

24 years agoFix broken installkernel target. Don't use the WMAKE context to
Marcel Moolenaar [Wed, 12 Jan 2000 10:00:36 +0000 (10:00 +0000)]
Fix broken installkernel target. Don't use the WMAKE context to
install; use the IMAKE context.

Reported by: sheldonh

24 years agoThis commit was generated by cvs2svn to compensate for changes in r55839,
Jeroen Ruigrok van der Werven [Wed, 12 Jan 2000 09:51:43 +0000 (09:51 +0000)]
This commit was generated by cvs2svn to compensate for changes in r55839,
which included commits to RCS files with non-trunk default branches.

24 years agoVirgin import of FSF groff v1.15
Jeroen Ruigrok van der Werven [Wed, 12 Jan 2000 09:51:43 +0000 (09:51 +0000)]
Virgin import of FSF groff v1.15

24 years agoTrack libc's three-tier symbol naming. libc_r must currently implement
Jason Evans [Wed, 12 Jan 2000 09:28:58 +0000 (09:28 +0000)]
Track libc's three-tier symbol naming.  libc_r must currently implement
the _libc_*() entry points and add *() weak aliases.  This will all
change for the better when libc_r becomes libpthread.