Nick Hibma [Sun, 23 Jan 2000 11:40:10 +0000 (11:40 +0000)]
Add the unlpt\d+ device to MAKEDEV. It is the same as ulpt\d+, except
that it does not reset the printer when opened. This fixes the problem
of printing a document almost till the end and then resetting the
printer when the next print job is started.
Submitted by: Christopher Masto <chris@netmonger.net>
Also, remove all but the ums0 device from the fixit target 'to save
precious i-nodes on the fixit floppy'.
Brian Somers [Sun, 23 Jan 2000 10:41:48 +0000 (10:41 +0000)]
Hmm, don't compile in INET6, IPX or NETATALK support for the
moment :-( I think we need some sort of stub variable and a
``is this queue available'' flag.
Matthew N. Dodd [Sun, 23 Jan 2000 08:46:21 +0000 (08:46 +0000)]
Fix a bug in the last commit. >> 4 instead of >> 8. *bap*
Fix a long-standing bug where I used a 'break;' instead of a 'continue;';
you had to have multiple ISA boards in non-PnP mode with an 'unknown' board
with a lower MAC address to find this one. Since I have 4 3c5x9 boards
in my test box I was somewhat confused when this happened. :)
Make the messages printed by ep_isa_identify() a little more consistent;
we'll only see them in verbose boot mode but it makes me feel better if
they look nice.
Matt Jacob [Sun, 23 Jan 2000 03:19:49 +0000 (03:19 +0000)]
Correctly put the place to mark EOP where we actually do the packet
transmission- this handles the odd and rare case of a list terminating
with a zero length mbuf.
Matt Jacob [Sun, 23 Jan 2000 01:49:11 +0000 (01:49 +0000)]
Get rid of the WX_XMIT_SMALL code- we've fixed that problem. Restructure
the receive code so that it correctly chains receive descriptors together
and handles the case that only a part of a packet is done at the time
we get here.
Mike Pritchard [Sun, 23 Jan 2000 01:48:16 +0000 (01:48 +0000)]
Fxi various man pages to stop abusing the .Bx macro to generate
the string "FreeBSD". Use the .Fx macro instead. Also did some
minor re-wording/formatting to work around a deficiency with
the .Fx macro when it comes to puncuation characters other than
periods and commas.
Brian Somers [Sun, 23 Jan 2000 01:47:12 +0000 (01:47 +0000)]
Implement TUN[GS]IFHEAD ioctls. Passing a non-zero int to TUNSIFHEAD
tells that tun unit to prepend a four byte address family to packets
queued for tunread() and to expect a four byte address family at the
front of data received by tunwrite().
We queue any protocol received from the interface for tunread(), but
only accept INET, INET6, IPX and NETATALK from tunwrite(). There is
support for Xerox's NS stuff, but AFAICT config(8) doesn't ever
define NS.
Matt Jacob [Sun, 23 Jan 2000 01:41:17 +0000 (01:41 +0000)]
Add a pointer in the softc that will point to partially received packet.
Set up index increments for receive descriptors based on whether
the PADDED_CELL define is set.
Wilko Bulte [Sat, 22 Jan 2000 18:33:24 +0000 (18:33 +0000)]
Check if the various .TXT files exist before trying to copy them.
Oh why did I select a first project that needed to touch release/Makefile..
The fact that my release-building Alpha panics on me does not help either :(
Brian Feldman [Sat, 22 Jan 2000 18:11:58 +0000 (18:11 +0000)]
Cleanups, new options, and syntax changes w.r.t. the -e fiag I had just
added. This completes the addition of apmconf(8) features into apm(8),
and will be followed shortly by asmodai with a manpage commit.
David E. O'Brien [Sat, 22 Jan 2000 16:31:39 +0000 (16:31 +0000)]
I wanted to preserve this analysis of the bug fixed by the previous commit.
From "Alexander N. Kabaev" <ak03@gte.com>:
The problem is that default builtin_setjmp implementation does not restore
any registers except for stack pointers when doing nonlocal jump. This
means, that every platform, that needs to store/restore some other state
between jumps, has to provide appropriate definitions for
builtin_setjmp_setup and builtin_setjmp_receiver RTL expansions.
Every plaftorm which needs to do some additional processing at setjmp
receive point, such as restoring additional information previously saved by
builtin_setjmp_setup, can provide definition for builtin_setjmp_receiver
RTL expansion. [That is what the previous commit does.] The i386 code
compiled with -fpic option requires the value of the PIC register (EBX) to
be restored in order to function properly. One should not provide
expansion for builtin_setjmp_setup because GCC documentation explicitly
states that values should be recalculated if possibe, rather than stored in
setjmp buffer and proper value for EBX can easlity be determined from the
current EIP contents.
Matthew Dillon [Fri, 21 Jan 2000 20:17:01 +0000 (20:17 +0000)]
Fix a deadlock between msync(..., MS_INVALIDATE) and vm_fault. The
invalidation code cannot wait for paging to complete while holding a
vnode lock, so we don't wait. Instead we simply allow the lower level
code to simply block on any busy pages it encounters. I think Yahoo
may be the only entity in the entire world that actually uses this
msync feature :-).
Wilko Bulte [Fri, 21 Jan 2000 19:24:33 +0000 (19:24 +0000)]
This file has gone away. It's contents have become the starting point for the
GENERIC HARDWARE.TXT now located one directory level up. This saves
people from duplicating all updates in i386 that are in fact generic.
Wilko Bulte [Fri, 21 Jan 2000 19:13:46 +0000 (19:13 +0000)]
This is an ugly way to make HARDWARE.TXT live here again (instead of
in [i386,alpha]HARDWARE.TXT. This particular file is destined to be
the generic HARDWARE.TXT. In [i386,alpha]HARWDARE.TXT the machdep
information will live from now on. This should fix the make release
failures people were experiencing.
Warner Losh [Fri, 21 Jan 2000 16:55:21 +0000 (16:55 +0000)]
Cleanup of the card database:
o Make sure every entry has a logger event on insert/delete
o Make sure that the order of loggers is consistant
o Add D-Link DEF-650
o use /sbin/ifconfig consistantly
o Add Elecom Laneed LD-CDE, NTT DoCoMo Paldio 321S and 341S
(from shige@FreeBSD.org)
Warner Losh [Fri, 21 Jan 2000 06:57:52 +0000 (06:57 +0000)]
Fix the style bugs in the style bugs fix. The style bug fix made the
new function inconsistant with the rest of this file. The spelling
and grammer fixes were good and remain.
Jason Evans [Thu, 20 Jan 2000 21:58:27 +0000 (21:58 +0000)]
Remove unnecessary alternate entry points for *setjmp(). Make the main
entry point the standard name when not compiling libc_r (for example,
longjmp is the main entry point instead of __longjmp).
Bill Fumerola [Thu, 20 Jan 2000 20:48:51 +0000 (20:48 +0000)]
Brucify:
add an upper limit to -t
match the types of return values and the variables they are stuffed in
make the man page and usage() a little more consistantly ugly
less obfuscation.
Bill Paul [Thu, 20 Jan 2000 19:57:43 +0000 (19:57 +0000)]
Fix test for deciding when a bulk IN transfer got truncated. I thought
I fixed this last night, but apparently I only applied the patch to the
copy of the code in /dev/brain0.
Andrew Gallatin [Thu, 20 Jan 2000 16:49:18 +0000 (16:49 +0000)]
Make devices behind PPBs work on DS10s (and any other single pchip tsunamis)
I was wagering on DEC being elegant & numbering PCI buses normally on
machines with one pchip. It looks like they went with consistent -- buses
behind ppbs begin with bus 2.