38986 Commits

Author SHA1 Message Date
yokota
934ebfaf1b Fill in tp->t_windowsize AFTER the call to (*linesw[tp->t_line].l_open)(),
rather than BEFORE the call.  Otherwise the structure will be `zero'ed out
by l_open, which actually is ttyopen(), if !TS_OPEN.

PR: kern/12420
1999-06-29 17:30:33 +00:00
roger
ec30ce339b Stop signals being generated after meteor_close.
Update METEORSSIGNAL to disable signals by setting the signal to 0.

PR:		i386/10533
Submitted by:	Frode Vatvedt Fjeld <frodef@dslab7.cs.uit.no>
1999-06-29 16:45:51 +00:00
billf
a9d802a64d Clarify what happens if fd is set to -1
Submitted by:	Alfred Perlstein <bright@rush.net>
1999-06-29 16:32:22 +00:00
peter
893e0d6d36 (mostly) fix ordering. 1999-06-29 16:14:20 +00:00
mckusick
582bbe6a3b No longer need to set B_ASYNC flag since BUF_KERNPROC now
unconditionally sets the identity of the buffer.
1999-06-29 15:57:40 +00:00
peter
5f8ebc1b91 Hopefully fix the remaining glitches with the BUF_*() changes. This should
(really this time) fix pageout to swap and a couple of clustering cases.

This simplifies BUF_KERNPROC() so that it unconditionally reassigns the
lock owner rather than testing B_ASYNC and having the caller decide when
to do the reassign.  At present this is required because some places use
B_CALL/b_iodone to free the buffers without B_ASYNC being set.  Also,
vfs_cluster.c explicitly calls BUF_KERNPROC() when attaching the buffers
rather than the parent walking the cluster_head tailq.

Reviewed by:	Kirk McKusick <mckusick@mckusick.com>
1999-06-29 05:59:47 +00:00
grog
0a692c6a09 Lock buffer before calling strategy.
Idea-stolen-from:	peter (sys/dev/ccd/ccd.c revision 1.49)
1999-06-29 04:10:05 +00:00
grog
3645b5b2f4 close_drive:
Correct race condition between caller and daemon.

  Tripped-over-by:	Zach Heilig <zach@uffdaonline.net>
			Bernd Walter <ticso@cicely.de>
			Niels Chr. Bank-Pedersen <ncbp@bank-pedersen.dk>
1999-06-29 04:08:51 +00:00
grog
37d4cc3c46 Correct type of intializer for (undocumented) cdevsw.d_parms.
Submitted-by:	peter
1999-06-29 04:07:55 +00:00
pb
0ff8a29fa6 Move call to umask(0) back into pw_util(), because the latter
function is also used by chpass(1) and passwd(1).
1999-06-29 01:04:10 +00:00
luoqi
4ee0d623ca Save common_tssd before it's loaded and the busy bit set.
Submitted by:	bde
1999-06-28 15:34:54 +00:00
peter
bbe9a3f06f Fix a bug that was almost certainly making breadn() fail. BUF_KERNPROC()
was being called on the wrong bp - it should be called on the one that's
just about to be fed to VOP_STRATEGY().
1999-06-28 15:32:10 +00:00
kato
836f488cc5 Typo: BUF_INITLOCK -> BUF_LOCKINIT and BUF_FREELOCK -> BUF_LOCKFREE. 1999-06-28 14:01:03 +00:00
des
81a1333c81 Use the correct value for banksize so splash_pcx works in LFB modes. 1999-06-28 13:52:29 +00:00
kato
d13ed34863 Sync with sys/i386/isa/clock.c revision 1.138. 1999-06-28 13:11:16 +00:00
kato
b601ed220a Sync with sys/i386/i386/userconfig.c revision 1.146. 1999-06-28 13:10:03 +00:00
kato
a6bac2371e Sync with sys/i386/i386/machdep.c revision 1.344. 1999-06-28 13:08:59 +00:00
kato
4e506f57f0 Sync with sys/i386/conf/Makefile.i386 revision 1.156. 1999-06-28 13:07:58 +00:00
peter
5ecb2e0dad Fix a KASSERT() that was negated and lead to:
nfs_strategy: buffer 0xxxxx not locked
when you attempted to write and had INVARIANTS turned on.
1999-06-28 12:34:40 +00:00
peter
debb5dbb90 Fix page fault in visual userconfig's save code. (I only use normal
userconfig, my original tweaks to visual mode were not well tested)

Submitted by:	Peter Holm <peter@holm.cc>
1999-06-28 12:07:36 +00:00
sheldonh
fa84406525 Sync usage() with the manpage.
Approved by:	mpp
1999-06-28 11:27:14 +00:00
mpp
f90c97ac2f Update the SYNOPSIS to reflect that the -l option can be specified
more than once.

Pointed-out-by: sheldonh
1999-06-28 10:50:47 +00:00
phk
7faf98030f make va_fsid be of type udev_t 1999-06-28 10:35:07 +00:00
peter
9a460a1161 Add some extra alpha_pal_imb()'s in overkill mode. This makes this PC164SX
run quite reliably now.  I've explicitly tagged them as /* XXX overkill? */
although one does actually check for VM_PROT_EXECUTE.

Based on a suggestion by:  Dmitrij Tejblum <tejblum@arc.hq.cti.ru>
1999-06-28 09:38:09 +00:00
sheldonh
b86772ac10 Fix broken logic: (!wrap || log) -> (!wrap && log) .
Reported by:	David Malone <dwmalone@maths.tcd.ie>
1999-06-28 09:28:17 +00:00
peter
88bef0081c Use the same -UKERNEL strategy as the alpha to avoid the inlines etc. 1999-06-28 09:21:41 +00:00
peter
7f6f706641 Don't #include i386/isa/isa_dma.h - it's in isa/isavar.h now. This
driver is probably not far from being MI now anyway.
1999-06-28 09:19:58 +00:00
peter
cafc37474e Revert back to not using -DKERNEL - it exposes inlines etc.
Zap symbols.raw and glue to make symbols.* - it's not used on the ELF-only
alpha kernel.  Symbol sorting is dead-end anyway once libkvm uses the
in-kernel linker symbol lookup.
1999-06-28 09:18:44 +00:00
peter
b27aea1e8e Revert back to not using -DKERNEL 1999-06-28 09:15:35 +00:00
peter
467b1db545 Move struct prochd out of #ifdef KERNEL so the Alpha genassym can get
at it.
1999-06-28 09:14:18 +00:00
peter
64f5069512 Tweak include ordering so sys/systm.h is before sys/buf.h to keep buf.h's
inlines happy.
1999-06-28 09:12:46 +00:00
obrien
11285d42bb Define "OBSTACK_CHUNK_SIZE" as getpagesize().
Our malloc can allocte pagesized blocks efficiently and the EGCS default size
of 4072 bytes is not optimal.

Submitted by:	Hidetoshi Shimokawa <simokawa@sat.t.u-tokyo.ac.jp>
1999-06-28 09:05:56 +00:00
mckusick
0d5e2c5c4f When requesting an exclusive lock with LK_NOWAIT, do not panic
if LK_RECURSIVE is not set, as we will simply return that the
lock is busy and not actually deadlock. This allows processes
to use polling locks against buffers that they may already
hold exclusively locked.
1999-06-28 07:54:58 +00:00
phk
50faed161f Hmm, might as well make all of BEFORE_DEPEND first in case make depend
wasn't run.
1999-06-28 07:19:51 +00:00
phk
771a9642c3 Also make pci_if.h early if make depend wasn't run. 1999-06-28 07:10:55 +00:00
phk
2e0584f725 Make bus_if.h and device_if.h as early as possible if make depend hasn't
been run.
1999-06-28 07:06:21 +00:00
obrien
8447bc90a5 Amd now mounts to "/.amd_mnt" rather than "/net". "/net/<host>" now works
in addition to "/host/<host>".  This make us consistant with Sun (as we
already were with SGI).
1999-06-28 04:09:14 +00:00
obrien
d3ca6b9c8b Syntax for user/group is changed from "user.group" to "user:group" to be
consistant with chown(8).
1999-06-28 03:15:02 +00:00
obrien
69070f26c3 Fix minor alignment problem. 1999-06-28 03:03:17 +00:00
billf
e3def919df New ports/{java,irc,x11-servers} categories, Step #5 - update misc files. 1999-06-28 02:37:34 +00:00
n_hibma
188adeba87 Add MultiTech Atlas modem 1999-06-27 22:28:02 +00:00
n_hibma
7725b5c306 Remove superfluous semi-colon 1999-06-27 22:24:20 +00:00
jseger
adb097eb8e Standardize appearance of subsection headers (two spaces for section name) 1999-06-27 22:12:35 +00:00
peter
5d3d376190 Kirk missed a required BUF_KERNPROC(). Even though this is a non-async
transfer, the b_iodone hook causes biodone() to release it from interrupt
context.
1999-06-27 22:08:38 +00:00
alc
f03495d396 An SMP-specific change: Remove an unnecessary lock acquire and release
from every system call.  (Storing a 32-bit constant is inherently
atomic.)

Reviewed by:	Matthew Dillon <dillon@apollo.backplane.com>
1999-06-27 21:31:43 +00:00
sheldonh
9b5c2ba429 Fix the SYNOPSIS to reflect that the -w option can be specified twice.
Requested by:	obrien
Approved by:	mpp
1999-06-27 21:07:55 +00:00
peter
9ecc5dac27 Make <sys/buf.h>'s inlines happier. 1999-06-27 20:52:22 +00:00
mjacob
63f64fbefe add description of Qlogic ISP FC Full Duplex option 1999-06-27 19:35:23 +00:00
rnordier
db2d7d3fb6 Drop the embedded boot code in favour of a -B option which reads
the boot code from /boot/mbr, or elsewhere as defined by the revised
"-b bootcode" option; use getopt(3); clarify usage(); partially
revise man page; etc.
1999-06-27 19:29:15 +00:00
mjacob
bdf6f33fa0 add config option for fibre channel full duplex 1999-06-27 19:28:26 +00:00