21323 Commits

Author SHA1 Message Date
mckusick
a3d0c189ea Add snapshots to the fast filesystem. Most of the changes support
the gating of system calls that cause modifications to the underlying
filesystem. The gating can be enabled by any filesystem that needs
to consistently suspend operations by adding the vop_stdgetwritemount
to their set of vnops. Once gating is enabled, the function
vfs_write_suspend stops all new write operations to a filesystem,
allows any filesystem modifying system calls already in progress
to complete, then sync's the filesystem to disk and returns. The
function vfs_write_resume allows the suspended write operations to
begin again. Gating is not added by default for all filesystems as
for SMP systems it adds two extra locks to such critical kernel
paths as the write system call. Thus, gating should only be added
as needed.

Details on the use and current status of snapshots in FFS can be
found in /sys/ufs/ffs/README.snapshot so for brevity and timelyness
is not included here. Unless and until you create a snapshot file,
these changes should have no effect on your system (famous last words).
2000-07-11 22:07:57 +00:00
dmlb
a1d52bdfdc Add attribute memory setup warnings 2000-07-11 21:31:59 +00:00
markm
5812cc4cc4 Storing to a pointer is (effectively) atomic; no need to protect this
with splhigh(). However, the entropy-harvesting routine needs pretty
serious irq-protection, as it is called out of irq handlers etc.

Clues given by:	bde
2000-07-11 19:37:25 +00:00
mckusick
b2ed023a03 Clean up warning about undeclared function by declaring softdep_fsync
in mount.h instead of ffs_extern.h. The correct solution is to use
an indirect function pointer so that the kernel does not have to be
built with options FFS, but that will be left for another day.
2000-07-11 19:28:26 +00:00
markm
e07f67dda2 I think I need to move the newly static variables to the random_state
structure; remind myself in the cooments. Also regroup all the Yarrow
variables at the top of the variable list; they are "special".
(no functional change).
2000-07-11 18:35:33 +00:00
nsayer
dc89f21663 Reserve 149 for the TAP device. Look for it in a source tree near you
soon.
2000-07-11 15:13:49 +00:00
nyan
b5738d479e Merge from the following changes.
sys/conf/files.i386		1.321
sys/dev/syscons/syscons.c	1.343
sys/i386/isa/spkr.c		1.46
sys/isa/fd.c			1.183 and 1.185
sys/isa/syscons_isa.c		1.14
sys/isa/vga_isa.c		1.18
2000-07-11 12:50:34 +00:00
tanimura
7716c5370a Finally merge newmidi.
(I had been busy for my own research activity until the last weekend)

Supported devices:

SB Midi Port			(sbc + midi)
SB OPL3				(sbc + midi)
16550 UART			(midi, needs a trick in your hint)
CS461x Midi Port		(csa + midi)

OSS-compatible sequencer	(seq)

Supported playing software:

playmidi			(We definitely need more)

Notes:

/dev/midistat now reports installed midi drivers. /dev/sndstat reports
only pcm drivers. We need the new name(pcmstat?).

EMU8000(SB AWE) does not sound yet but does get probed so that the OPL3
synth on an AWE card works.

TODO:

MSS/PCI bridge drivers
Midi-tty interface to support general serial devices
Modules
2000-07-11 11:49:33 +00:00
alfred
4036af0a9c #elsif -> #elif
Noticed by: green
2000-07-11 09:41:29 +00:00
green
7402d1359e One should never allocate 4-kilobyte structs and such on the interrupt
stack.  It's bad for your machine's health.

Make the two huge structs in reseed() static to prevent crashes.  This
is the bug that people have been running into and panic()ing on for the
past few days.

Reviewed by:	phk
2000-07-11 06:47:38 +00:00
phk
7e67bebec5 Newbusify this driver at an expense of 40 extra lines of code. 2000-07-10 19:33:41 +00:00
jhb
41134498c4 Always install loader.4th, defaults/loader.conf, and support.4th instead of
only doing so if loader.rc does not exist.  This fixes the problem where
installworld doesn't update /boot/loader.4th, resulting in device.hints not
being loaded after updating past the config(8) changes, which resulted in
mcclock0 not being probed, and a nice kernel panic during boot.
2000-07-10 16:52:42 +00:00
imp
3dbed00e16 Remove the XE_DEBUG define. It shouldn't be defined now that things
are working.  Add a small blurb about XE_DEBUG as it might be useful
to some people troubelshooting problems in the future.

Submitted by:	"Kevin Oberman" <oberman@es.net>
2000-07-10 16:46:21 +00:00
mjacob
f5a6b75d84 Beef up a bit descriptions of SCSI devices and what the drive. 2000-07-10 15:37:03 +00:00
mjacob
8e224e7a68 Removing commented out devices I added. 2000-07-10 15:18:20 +00:00
phk
9241ff9fc6 Finish repo-copy:
Move ufs/ufs/ufs_disksubr.c to kern/subr_disklabel.c.

These functions are not UFS specific and are in fact used all over the place.
2000-07-10 13:48:06 +00:00
kris
baccd04090 Don't call printf without a format string. 2000-07-10 07:13:36 +00:00
markm
950c011f72 Provide more splsofttq() protection for the reseed task (running out of
taskqueue_swi).
2000-07-10 06:40:23 +00:00
kris
78f990a287 Don't call printf with no format string. 2000-07-10 06:40:06 +00:00
kris
0d6709d662 Don't call printf with no format string. This is technically a security
vulnerability and could in principle be used to upload a new kernel from the
bootloader :-)
2000-07-10 06:33:55 +00:00
kris
5fae984c0b Don't call printf with no format string.
Reviewed by:	msmith
2000-07-10 05:51:56 +00:00
mjacob
4175feb6ac Do the same thing for TurboLaser that was done for Rawhide- make room
for secondary (bridged) PCI busses by making primary PCI instances
16 units apart.
2000-07-10 02:40:49 +00:00
mjacob
448fe174f0 Coordinate with change to mcpcia_pci.c- major primary busses on each
hose are 16 PCI instances apart. This allows us to recognize secondary
PCI busses (at least to a first level) until the pci infrastructure is
fixed.

Turn on support for secondary cycles, too.  Redo debug printouts.
2000-07-10 00:34:18 +00:00
mjacob
e2f854c9d6 Don't let the infrastructure assign the 'next' PCI bus for us.
Instead, for now (until we get a pci infrastructure cleanup),
assign the PCI bus number to be mcpcia bus instance << 4. This
is to allow secondary bridges some room to be recongnized on
4100 systems.
2000-07-10 00:32:02 +00:00
bp
b25a59d797 Correct SYSINIT execution order in the case when KLD contains more
than one SYSINIT with the same 'subsystem' id and different 'order' id.

Reviewed by:	peter
2000-07-09 23:58:56 +00:00
markm
40b3a303c2 Make sure that tasks (running out of taskqueue_swi at splsofttq)
are not interfered with by the harvester.
2000-07-09 16:00:31 +00:00
alex
9474a996bc Add missing "a" in "Soft updates is technique".
PR:		19770
Submitted by:	Udo Erdelhoff <ue@nathan.ruhr.de>
2000-07-09 15:44:02 +00:00
itojun
56bc1eab2d be more cautious about tcp option length field. drop bogus ones earlier.
not sure if there is a real threat or not, but it seems that there's
possibility for overrun/underrun (like non-NOP option with optlen > cnt).
2000-07-09 13:01:59 +00:00
alex
2cdcc841c4 Since this file is doc now, reorganize its structure.
Currently, many drivers support more than one bus of ISA, EISA, MCA,
PCI.

Before this commit, we had, for example, some SCSI devices listed more
than once, iirc, some up to three times (ISA/EISA, MCA, PCI).

Since now the "device" line is common for all of them and they only
differ for the hints stuff, I did the following:

First, list Busses: (E)ISA, MCA, PCI and explain, that only ISA
needs the hints stuff.

Move NIC/SCSI stuff, which were the only split sections, behind these
stuff.  Describe all drivers only one time and list all supported
chips.

List all device (+ hints for ISA, if possible).

I've also added few additional supported chips to some drivers, xl for
example and some SCSI drivers.

Also, softupdates is no longer disabled by default, so the comment should
not say, it's not enabled by default due to license issues.

Approved by:	asmodai

To come:	include PAO devices (imp volunteered for help)
2000-07-09 12:34:53 +00:00
markm
edc351fc59 Small style change; make function names less likely to clash with
existing names. "null" is too common a string; use "null_".
2000-07-09 12:29:24 +00:00
markm
f0aab2e149 Add entropy-harvesting calls.
/dev/random now has new-and-improved entropy!
2000-07-09 12:26:38 +00:00
markm
89ca691934 Yarrow tweaks; separate the fast and slow reseed tasks so that they don't
stomp on each other; provide constant names (as enums) for the harvester
to use (makes it more self-documenting).
2000-07-09 11:52:12 +00:00
markm
be58a54d86 Fix bug with a vraiable that needs to be per-process, not static;
fix formatting of long macros.

Pointed out by:	bde
2000-07-09 11:49:07 +00:00
itojun
fec7b00bf4 repair IPV6_JOIN_GROUP to IPv6 all multi.
From: ume
2000-07-09 11:17:17 +00:00
mjacob
712e1316bf Thanks for Andrew Gallatin pointing out that freeing contigmalloc'd
items via free is bad.
2000-07-09 00:18:21 +00:00
mjacob
974a998625 Use %p to print a pointer. 2000-07-08 23:24:08 +00:00
mjacob
6301565880 Update SCSI device section, per Peter Wemm. I still
think the commented entries in GENERIC are the
right thing to do.
2000-07-08 07:44:01 +00:00
mjacob
ed6b1cf022 Oops- remove the '0' appended to targbh. 2000-07-08 07:05:40 +00:00
mjacob
c711d7f105 Add in the commented out SCSI device entries of
#device         ses             # SCSI Environmental Services (and SAF-TE)
#device         targ            # SCSI Target Mode Code
#device         targbh          # SCSI Target Mode Blackhole Device
#define         pt              # SCSI Processor Target Device

so that people know that they are there.
2000-07-08 07:05:06 +00:00
mjacob
14d741fcc2 Add in the commented out SCSI device entries of
#device         ses             # SCSI Environmental Services (and SAF-TE)
#device         targ            # SCSI Target Mode Code
#device         targbh0         # SCSI Target Mode Blackhole Device
#define         pt              # SCSI Processor Target Device

so that people know that they are there.
2000-07-08 07:02:58 +00:00
mckusick
92dfcced5b Delete README as it is now obsolete. Relevant information is in
README.softupdates.
2000-07-08 02:32:49 +00:00
mckusick
0ab089c771 Update to reflect current status. 2000-07-08 02:31:21 +00:00
gallatin
27334af100 use contigfree() rather than free() to free memory allocated with
contigmalloc().

reviewed by: wpaul
2000-07-08 00:14:12 +00:00
green
efb639324d Remove two micro-pessimizations I made. Bruce is teaching me well :)
KTRPOINT(p, KTR_GENIO) is more uncommon than error == 0, so it should
be first in the && statement.
2000-07-07 22:11:37 +00:00
green
f3eed69a97 Change that &@!$# UIO_READ to be UIO_WRITE. I tested the ktrace stuff,
but somehow... pass the pointy hat, again!
2000-07-07 21:52:15 +00:00
jhb
1ee4aa44d5 Remove commented out NOMAN variable. 2000-07-07 20:37:11 +00:00
dmlb
adc2b92260 Move newbus detection code to alloc routines.
Work around pccard nasties.
2000-07-07 19:13:11 +00:00
obrien
950539b64e MFS: adjust University of California's copyright. 2000-07-07 18:35:11 +00:00
bp
8a86977869 Fix support for more than 256 simultaneous mounts. Theoretical limit
is 2^16 mounts per fs type.

Reported by:	Troy Arie Cobb <tcobb@staff.circle.net> via phk
Reviewed by:	bde
2000-07-07 14:01:08 +00:00
markm
94bc8a09a9 Honour appropriate no-crypto flags. 2000-07-07 09:38:01 +00:00