10527 Commits

Author SHA1 Message Date
bde
004d5369f3 Partially fixed write clustering for cases where cluster_wbuild() is
called from vfs_bio_awrite() without going through cluster_write()
or ufs_bmaparray(), in particular for all writes to block disk devices.
Only ufs_bmaparray() sets vp->v_maxio in a correct way, and it doesn't
seem to be called early enough even for regular files.
1998-05-01 16:29:27 +00:00
peter
dfa49715ca vm_page_is_valid() wasn't expecting a large offset argument, it's
expecting a sub-page offset.  We were passing the file position,
and vm_page_bits() could do some interesting things when base was
larger PAGE_SIZE.
if (size > PAGE_SIZE - base)
	size = PAGE_SIZE - base;
is interesting when (PAGE_SIZE - base) is negative.  I could imagine that
this could have interesting consequences for memory page -> device block
bit validation.
1998-05-01 15:10:59 +00:00
peter
e5ab5108e0 Fix one problem with NFSv3 > 2GB file support.
Submitted by: bde
1998-05-01 15:04:35 +00:00
dyson
7835943279 Fix minor bug with new over used swap fix. 1998-05-01 02:25:29 +00:00
phk
ee66bf4cbc Loopback network interface driver (net/if_loop.c) has no SIOCSIFFLAGS
ioctl handler.

PR:		6466
Reviewed by:	phk
Submitted by:	Ruslan Ermilov <ru@ucb.crimea.ua>
1998-04-30 19:37:00 +00:00
jb
0e07a10674 Make errno finally and permanently thread-aware so that it is no longer
necessary to compile with _THREAD_SAFE defined. This means that people
will get thread-aware code whether they like it or not. This change
is required to allow a process to link against libpthread and libc
to use kernel threads (and prevent each thread from clobbering another
thread's errno just be doing a syscall).

This is bound to break some ports, but it is strictly allowed by ANSI C,
so anything that breaks as a result was already broken anyway 8-).
"Sorry".
1998-04-30 09:38:03 +00:00
dyson
224581813c Fix an error that I made with an optimization. In the case
of softupdates, we need to do vtruncbuf the old way.  Luoqi
caught, found the bug and submitted this fix.
Submitted by:	Luoqi Chen <luoqi@chen.ml.org>
1998-04-30 05:28:53 +00:00
imp
307c5832d7 Use ${.TARGET} rather than $@. i386 tested, pc98 untested because config
on my box doesn't grok machine type pc98.
PR:		3272
Submitted by:	jhs
1998-04-29 18:21:35 +00:00
andreas
e4bced7382 Fixes incorrect company and product names in uha(4) manpage and
LINT config file. Should be merged to -STABLE as well.
PR:		closes 6447
Submitted by:	Steven Plite splite@purdue.edu
1998-04-29 17:09:41 +00:00
dt
0e69a77105 Use DFLTBSIZE instead of MAXBSIZE for pm_fatblksize.
In msdosfs_sync: spelling fix, formatting changes; fix MNT_LAZY (sync
modified denodes, don't sync device)

Mostly submitted by (and with hints from):	bde

Increase limit for maximum disk size: as far as I can see previous limit was
gratuitously too low.
1998-04-29 12:55:51 +00:00
kato
a44088edc8 Sync with sys/i386/i386/trap.c revision 1.127. 1998-04-29 09:45:38 +00:00
dyson
a7cf05f1b7 Add a needed prototype, and fix a panic problem with the new
memory code.
1998-04-29 06:59:08 +00:00
max
42ba449fff Add "flags 0x10" to the sio0 entry so that SMP-GENERIC acts in thesame
manner as the GENERIC kernel to the -h option.
1998-04-29 06:58:43 +00:00
dyson
b5a79794cd Tighten up management of memory and swap space during map allocation,
deallocation cycles.  This should provide a measurable improvement
on swap and memory allocation on loaded systems.  It is unlikely a
complete solution.  Also, provide more map info with procfs.
Chuck Cranor spurred on this improvement.
1998-04-29 04:28:22 +00:00
eivind
67c7bb9c04 Translate T_PROTFLT to SIGSEGV instead of SIGBUS when running under
Linux emulation.  This make Allegro Common Lisp 4.3 work under
FreeBSD!

Submitted by: Fred Gilham <gilham@csl.sri.com>
Commented on by: bde, dg, msmith, tg
Hoping he got everything right:  eivind
1998-04-28 18:15:08 +00:00
dyson
36e092b938 Fix a pseudo-swap leak problem. This mitigates "leaks" due to
freeing partial objects, not freeing entire objects didn't
free any of it.  Simple fix to the map code.
Reviewed by:	dg
1998-04-28 05:54:47 +00:00
julian
7353226d84 Submitted by: Doug Ambrisko <ambrisko@whistle.com>
make standard links for the first detected sound card.
1998-04-28 00:10:53 +00:00
des
1278274147 Cast return values to the appropriate fp_*_t. Note that the man page
incorrectly refers to them as e.g. fp_except rather than fp_except_t.

PR:		misc/6310
Submitted by:	Niall Smart
1998-04-27 14:17:30 +00:00
des
1e8b395cf1 Fixed spelling of "fictitious" in diagnostics and comments
PR:		kern/6429
1998-04-27 11:36:06 +00:00
brian
5270239bec Support more than 256 tun devices:
$ ls -l /dev/tun25[4-7]
crw-------  1 fax  dialer   52, 254 Apr 27 02:27 /dev/tun254
crw-------  1 fax  dialer   52, 255 Apr 27 02:27 /dev/tun255
crw-------  1 fax  dialer   52, 0x00010000 Apr 27 02:31 /dev/tun256
crw-------  1 fax  dialer   52, 0x00010001 Apr 27 02:31 /dev/tun257
1998-04-27 01:44:22 +00:00
dyson
776fdaf7cf Add the PAT cpuid feature. 1998-04-26 03:18:38 +00:00
kato
803f1dcc1d Sync with sys/i386/isa/wd.c revision up to 1.163. 1998-04-25 05:19:33 +00:00
dyson
888e1a851b Correct copyright. 1998-04-25 04:50:03 +00:00
kato
b73332da4c Sync with sys/i386/isa/wd.c revision 1.161. 1998-04-25 04:27:54 +00:00
kato
68daaf494d Sync with sys/i386/conf/GENERIC revision 1.108. 1998-04-25 04:27:15 +00:00
mjacob
7d48e85bb7 Oops- osreldate.h is not around like I thought it would be... hmmm.. quick fix to get around this 1998-04-24 18:23:20 +00:00
obrien
fea3a33427 Discussed with: bde 1998-04-24 11:50:30 +00:00
obrien
82338a65ac Create virgin disklabels with 8 (MAXPARTITIONS) partitions rather than
three (RAW_PART + 1);
This makes ``disklabel -Brw sdN auto'' do the Right Thing.
1998-04-24 11:49:57 +00:00
dg
c0b0bc1742 At the request of Garrett, changed sysctl:
net.inet.tcp.delack_enabled -> net.inet.tcp.delayed_ack
1998-04-24 10:08:57 +00:00
dg
a30ffaa134 Ensure that TCP_REXMTVAL doesn't return a value less than t_rttmin. This
is believed to have been broken with the Brakmo/Peterson srtt
calculation changes. The result of this bug is that TCP connections
could time out extremely quickly (in 12 seconds).
Also backed out jdp's partial fix for this problem in rev 1.17 of
tcp_timer.c as it is obsoleted by this commit.
Bug was pointed out by Kevin Lehey <kml@roller.nas.nasa.gov>.

PR:	6068
1998-04-24 09:25:39 +00:00
julian
de78d409f5 slice_device.c: permissions changes for SLICE devices
vn.c:		change time of SYSINIT scheduling.
wd.c		don't revert to fully closed state. ( may require more)
all in SLICE mode only.
1998-04-24 07:54:00 +00:00
dg
3253b48e39 Added kern.ipc.nmbclusters 1998-04-24 04:15:52 +00:00
julian
740cb0899a When calling the open function, specify either FREAD and/or FWRITE
as leaving them both 0 has the same effect as not openning the device at all.
1998-04-23 22:09:55 +00:00
julian
7b5f18bb24 Oops forgot to include opt_bootp.h
Bootp works a lot better with devfs if this is present.
1998-04-23 20:59:42 +00:00
eivind
2297a9666d Convert timespecclear() back to an expression (this time without
sideeffects).
1998-04-23 14:12:06 +00:00
eivind
a325280132 Wrap timerclear in do { ... } while(0) 1998-04-23 00:11:32 +00:00
eivind
1c00798cb4 Add devfs support.
Submitted by:	somebody whose name escapes me :-(
1998-04-23 00:06:43 +00:00
tegge
3135dafcdb Mask the interrupt before setting the corresponding bit in ipending if
the interrupt is already active.
Don't use lock prefix for operations on ipending.
Always use lock prefix for operations on iactive.
1998-04-22 22:49:29 +00:00
julian
ed0da47a02 Bad144 support for the slice system (!)
Submitted by: luoqi@watermarkgroup.com (Luoqi Chen)

I'm amazed by this. Slice has only been checked in for 2 days..
1998-04-22 19:27:54 +00:00
mjacob
efb500c7ac Add support for the Qlogic ISP SCSI && FC/AL Adapters 1998-04-22 18:18:16 +00:00
mjacob
8cd3efb7df Add support for the Qlogic ISP SCSI && FC/AL Adapters 1998-04-22 18:12:29 +00:00
mjacob
a9113b6f9b Add support for the Qlogic ISP SCSI && FC/AL Adapters 1998-04-22 17:54:58 +00:00
kato
d272f774b9 Sync with sys/i386/conf/majors.i386 revision 1.39. 1998-04-22 13:35:21 +00:00
julian
db37f632a0 close() is no longer a SLICE method.
Close is simply an open with no-read and no-write once internal to SLICE
(it still exports a close to the rest of the kernel)
1998-04-22 10:25:27 +00:00
peter
2fe90b3e89 Add a couple of missing symbols referred to in asm code. 1998-04-22 09:46:16 +00:00
msmith
cf01114fc4 Add a block device entry for the 'myx' driver too. 1998-04-22 05:10:51 +00:00
msmith
055b4ba647 Add 'myx' for Ulf Zimmerman's Mylex Raid driver (ulf@alameda.net> 1998-04-22 03:57:07 +00:00
julian
b8c5fe2677 Don't allow an open to succeed if the device is invalidated during the
open process.
1998-04-22 01:51:34 +00:00
brian
7914b665ae - breakage of the warnings about pessimized i/o port types.
- pessimized i/o port types.
- other pessimized types.
- Don't use DEBUG (causes LINT warnings).  Use DGB_DEBUG instead.
- commented out code.
- cloned code that doesn't apply ("Smarts" is for the cy driver only).

Submitted by: bde
1998-04-21 21:06:57 +00:00
brian
c38ab8d4d9 Correct the digiboard device names in agreement
with the man page (POLA).
1998-04-21 21:06:22 +00:00