Commit Graph

82582 Commits

Author SHA1 Message Date
Jeff Roberson
eff3a9fd92 - Enable the use of UMA_MD_PAGE_ALLOC on alpha.
- Define uma_small_alloc and uma_small_free using K0SEG for virtual
   addresses.

Approved by:	re
2002-11-01 03:34:51 +00:00
Tony Finch
3229ed9140 Variables that come in in and out pairs should be named accordingly.
Different code that processes the input in similar ways should be
called in similar ways. File-local stuff should be static. Output
errors should be checked for. Diffs sometimes have to be big.
2002-11-01 02:08:51 +00:00
Jun Kuriyama
b46e66a5a5 debugMsg() should end with "\n". 2002-11-01 02:05:05 +00:00
Ian Dowse
9ce4f7a0f0 Specify the correct protocol for the Olympus C-1 camera. It appears
that this entry was accidentally given the wrong protocol in revision
1.61.

PR:		kern/42417
Submitted by:	"Moriki, Toshiomi" <toshiomi@whi.m-net.ne.jp>
2002-11-01 01:13:20 +00:00
Jeff Roberson
48eea37508 - Add support for machine dependant page allocation routines. MD code
may define UMA_MD_SMALL_ALLOC to make use of this feature.

Reviewed by:	peter, jake
2002-11-01 01:01:27 +00:00
Jeff Roberson
026aa839a4 - Add a new flag to vm_page_alloc, VM_ALLOC_NOOBJ. This tells
vm_page_alloc not to insert this page into an object.  The pindex is
   still used for colorization.
 - Rework vm_page_select_* to accept a color instead of an object and
   pindex to work with VM_PAGE_NOOBJ.
 - Document other VM_ALLOC_ flags.

Reviewed by:	peter, jake
2002-11-01 00:59:03 +00:00
Tony Finch
735838d3fe Be much more paranoid about where uudecode writes its output, especially
when the filename comes from the untrusted input. This is a work-around
for careless people who don't routinely check the begin line of the file
or run uudecode -i and instead report "vulnerabilities" to CERT.
	http://www.kb.cert.org/vuls/id/336083
2002-11-01 00:58:00 +00:00
Tony Finch
eae1ff453a Instead of abusing stdin and stdout, use our own file pointers.
Check for errors when closing the output.
2002-11-01 00:29:00 +00:00
Julian Elischer
94d31abc4e More formatting fixes 2002-10-31 23:36:16 +00:00
Julian Elischer
5bec8a73ac Slight formatting change. 2002-10-31 23:26:02 +00:00
Jeff Roberson
1645d0903e - Consistently update snd_wl1, snd_wl2, and rcv_up in the header
prediction code.  Previously, 2GB worth of header predicted data
   could leave these variables too far out of sequence which would cause
   problems after receiving a packet that did not match the header
   prediction.

Submitted by:	Bill Baumann <bbaumann@isilon.com>
Sponsored by:	Isilon Systems, Inc.
Reviewed by:	hsu, pete@isilon.com, neal@isilon.com, aaronp@isilon.com
2002-10-31 23:24:13 +00:00
Archie Cobbs
83999f5a32 Re-apply the previously backed-out commit that fixes the problem where
HUGE_VAL is not properly aligned on some architectures. The previous
fix now works because the two versions of 'math.h' (include/math.h
and lib/msun/src/math.h) have since been merged into one.

PR:	bin/43544
2002-10-31 23:05:20 +00:00
Bruce A. Mah
06527f6dc3 Modified release notes: Note that check_utility_compat(3) is designed
for FreeBSD base system programs, rewrite expr(1) entry and note its
use of check_utility_compat(3), mention autogeneration of
nsswitch.conf and hosts.conf.

Mostly submitted by:	wollman

(But any mistakes are solely mine.)
2002-10-31 23:04:28 +00:00
Julian Elischer
585ff168dc Add the netgraph 'source' module.
This is NOT YET CONVERTED TO -current.
This node is a source for preprogrammed packets at a known rate for testing.

I will convert it to -current "in place" but will MFC teh original
pre-conversion variant as that is  what is originally submitted.
Man page my me, info from Dave's README.

Submitted by:	Dave Chapeskie <dchapeskie@SANDVINE.com>
Obtained from:	Sandvine inc.
MFC after:	1 week
2002-10-31 23:03:09 +00:00
Jeff Roberson
24b50116ed - Introduce a new macro, since that's what nfs loves, called
nfsm_srvpathsiz.  This macro plucks a length out of an rpc request and
   verifies that its size does not exceed NFS_MAXPATHLEN.  If it does
   it generates an ENAMETOOLONG response.
 - Use this macro, and the existing nfsm_srvnamsiz macro in two places
   where we deal with paths passed in by the client.

This fixes a linux interoperability bug.  Linux was sending oversized path
components which would cause us to ignore the request all together.  This
causes linux to hang indefinitly while it waits for a response.  This
could still happen in other cases where we error out with EBADRPC.

Sponsored by:	Isilon Systems, Inc.
Reviewed by:	alfred, fabbri@isilon.com, neal@isilon.com
2002-10-31 22:35:03 +00:00
Poul-Henning Kamp
bfccaa1df2 Spruce up bootverbose output a bit.
Allow extended partitions to have flag=0x80
2002-10-31 22:18:49 +00:00
Mitsuru IWASAKI
21ac81ad19 Don't examine ACPI_STA_ENABLE bit of device's_STA object on initial
state.  Instead, use ACPI_STA_PRESENT and ACPI_STA_FUNCTIONAL for it.
In some ACPI BIOS implementations, boot disabled devices don't have
ACPI_STA_ENABLE bit in _STA object.
Also it is not fatal if getting current IRQ of boot disabled devices
is failed in initial state.

And minor fixes.
2002-10-31 20:43:02 +00:00
Mitsuru IWASAKI
498d464fe9 Interpret new loader tunable "hw.acpi.verbose" to turn
verbose mode on at boot time.
2002-10-31 20:23:41 +00:00
John Baldwin
97af632bcb Add an isp(4) module. sbus support is only compiled in on sparc64.
pci support is always compiled in.
2002-10-31 19:50:18 +00:00
Hajimu UMEMOTO
35f6695bb2 plugged memory leakage in some erroneous cases
Obtained from:	KAME
MFC after:	1 week
2002-10-31 19:45:48 +00:00
John Baldwin
bc96e1c7cf Add a module for mpt(4). 2002-10-31 19:39:23 +00:00
John Baldwin
de5df1d716 Comment out 'mpt' for the time being. This driver doesn't have a module
either.
2002-10-31 19:02:07 +00:00
Daniel Eischen
0ece26b0ba Use the strong symbol'd version of pthread_mutex_init so we don't
accidentally call a stub or application provided version of the
same routine.

Submitted by:	dfr
2002-10-31 18:17:58 +00:00
Mitsuru IWASAKI
91da7c40b2 Invoke 3 ACPI task threads as default if option ACPI_MAX_THREADS is
not defined.
To make previous default behavior (ACPI_MAX_THREADS undefined), define
option ACPI_MAX_THREADS as 0.
2002-10-31 17:58:39 +00:00
Poul-Henning Kamp
5c63c8dd25 Give a meaningfull diagnostic when we cannot determine the filesystem type. 2002-10-31 15:32:39 +00:00
John Baldwin
96d13b89c2 There is no 'isp' module to move off to the 3rd floppy, so comment out the
entry for now.
2002-10-31 14:38:21 +00:00
Eric Moore
f462788495 Incorrect typecast. 2002-10-31 14:10:00 +00:00
Tim J. Robbins
01e5f16641 Do not print a header line if it would be empty; required by 1003.1-2001. 2002-10-31 13:41:37 +00:00
Maxim Sobolev
1acb5329c8 Back out previous commit - it was not well thought out.
Clue beaten in by:	ume
Pointy hat to:		sobomax
2002-10-31 13:05:56 +00:00
Yoshihiro Takahashi
abb69175f8 Enable GEOM by default. 2002-10-31 12:14:05 +00:00
Thomas Quinot
a1e2d9ff4d Add ATAPI/CAM man page.
Reviewed by:	roberto
2002-10-31 10:52:47 +00:00
Ruslan Ermilov
b9889f93c2 Document the new behavior of -C with respect to inode-change-time.
Document that -p also preserves the access time of the source.
2002-10-31 10:43:32 +00:00
Ruslan Ermilov
1067b3a3fb Do not change the target file's inode-change-time if -C is
specified, the target file already exists and the files are
the same, and the target's file flags and mode need not to
be changed.
2002-10-31 10:41:51 +00:00
Maxim Sobolev
ef1e4234eb Fix POLA breakage in 1.29: IPv4 should be default. This makes `-4' option
a nop, but we'll probably want to keep it for compatibility with other
KAME-based systems.

Complained about by:	Andrey Lakhno <land@dnepr.net>
MFC after:		3 days
2002-10-31 10:36:12 +00:00
Ruslan Ermilov
7472551efd ade@ no longer has the FreeBSD account. 2002-10-31 08:56:34 +00:00
Ruslan Ermilov
d1c009cec8 Sort by month/day/year/login. 2002-10-31 08:55:33 +00:00
David Xu
adac9400a7 KSE-enabled processes only. 2002-10-31 08:00:51 +00:00
Poul-Henning Kamp
fd4a61e809 Style(9) improvements. 2002-10-31 07:56:40 +00:00
Poul-Henning Kamp
63abca411d Set the sector size for the disk. 2002-10-31 07:55:38 +00:00
Yoshihiro Takahashi
fe1f2a9a84 Restore to pc98 support. 2002-10-31 05:51:25 +00:00
Yoshihiro Takahashi
fecafa765a Use ${MACHINE} variable instead of using '.if .endif' each machines. 2002-10-31 05:43:49 +00:00
Yoshihiro Takahashi
c540b79f8b Add '#include <err.h>' for warn(). 2002-10-31 05:38:48 +00:00
Yoshihiro Takahashi
d2f5398613 Rename the speaker device for pc98 to 'pcspeaker'. 2002-10-31 05:19:33 +00:00
Yoshihiro Takahashi
4d5b1746b2 MFi386: revision 1.55.
Rename from atspeaker to pcspeaker. (PC98 is not PC/AT)
2002-10-31 05:12:22 +00:00
Yoshihiro Takahashi
dacf0b85b7 Disable COMPAT_FREEBSD4. 2002-10-31 04:51:05 +00:00
Jake Burkholder
e03494dd17 Actually save the bootblock in the disk structure. Write the bootblock
to the right place on the disk instead of srewn all over it.
2002-10-31 04:25:17 +00:00
Gordon Tetlow
cbde04a9b7 Wrap test case in an if statement to avoid crunching a return value. 2002-10-31 02:06:05 +00:00
Alfred Perlstein
c313cb2abf Make __sF static. This can not be allowed to exist in 5.x. 2002-10-31 01:54:27 +00:00
Peter Wemm
700f053192 Add tunable "hw.eisa_slots" to allow overriding the default number
of slots (10).
2002-10-31 00:45:49 +00:00
Thomas Quinot
17b1e1df1a Remove bogus advertising clauses.
Reviewed by:	n_hibma, roberto
2002-10-30 22:27:54 +00:00