Commit Graph

102192 Commits

Author SHA1 Message Date
bmah
7d1143b1dd New release notes: SA-04:08, SA-04:09. 2004-05-09 22:23:30 +00:00
simon
84916e2d8a Trim the fxp(4) hardware notes. 2004-05-09 21:27:15 +00:00
simon
0457db4498 - Sync with Hardware Notes.
- Spell adapter names more like Intel do.
- Make the page title reflect reality a bit better.
- Spell Fast Ethernet correctly.
2004-05-09 21:25:49 +00:00
alc
600fc14c5f Correct the implementation of pmap_page_is_mapped(): It should return TRUE
only if the page has one or more managed mappings.
2004-05-09 19:09:14 +00:00
pjd
52fd45ee97 Don't confuse "GEOM" with "geom".
Approved by:	phk
2004-05-09 16:16:10 +00:00
tjr
a8117b04ca Use a binary search to find the range containing a character in
RuneRange arrays. This is much faster when there are hundreds of
ranges (as is the case in UTF-8 locales) and was inspired by a
similar change made by Apple in Darwin.
2004-05-09 13:04:49 +00:00
brueffer
0b4230fd70 o fix a sentence to match with the synopsis [1]
o fix grammar nit

PR:		66289 [1]
Submitted by:	Michel Lavondès <fox@vader.aacc.cc.md.us>
Obtained from:	OpenBSD [1]
MFC after:	3 days
2004-05-09 11:11:21 +00:00
imp
b5c3c9d311 This file has never been used, nor will it ever be used in FreeBSD, so
remove it.
2004-05-09 07:09:30 +00:00
sam
c89a947571 set m_len to reflect mbuf contents on return from m_dup1; fixes an obscure
m_pullup case that contributed to breaking ipcomp in tunnel mode for kame

Submitted by:	itojun
Obtained from:	kame
2004-05-09 05:57:58 +00:00
marcel
7cd85074fc This commit was generated by cvs2svn to compensate for changes in r129059,
which included commits to RCS files with non-trunk default branches.
2004-05-09 03:06:25 +00:00
marcel
19196476e4 Update to BETA 7. Besides C++ support, which is irrelevant to us,
this version mostly has bugs fixes.
2004-05-09 03:06:25 +00:00
csjp
02de288baa Remove redundant sanity check before add_mac() when adding
mac ipfw rules. The exact same sanity check is performed as
the first operation of add_mac(), so there is no sense
in doing it twice.

Approved by:	bmilekic (mentor)
PR:		bin/55981
2004-05-09 01:53:31 +00:00
alc
0049080776 Cache queue pages are not mapped. Thus, the pmap_remove_all() by
vm_page_alloc() is unnecessary.
2004-05-09 01:00:15 +00:00
bmah
3209d7f80d Add 4.10 and 5.2.1 entries so that pkg_add -r fetches from the right
package sets.

PR:		66251
Submitted by:	eik
2004-05-08 23:45:31 +00:00
alc
baf71676e7 Since revision 1.280 of vm/vm_page.c, vm_page_grab() always returns a
zeroed page when passed VM_ALLOC_ZERO.  Thus, we can eliminate the check
against PG_ZERO from pmap_pinit().
2004-05-08 23:26:11 +00:00
peadar
ea85333e1c The FTS_NOSTAT option is an optimisation that reduces the number
of stat(2) calls by keeping an eye of the number of links a directory
has. It assumes that each subdirectory will have a hard link to its
parent, to represent the ".." node, and stops calling stat(2) when
all links are accounted for in a given directory.

This assumption is really only valid for UNIX-like filesystems: A
concrete example is NTFS. The NTFS "i-node" does contain a link
count, but most/all directories have a link count between 0 and 2
inclusive. The end result is that find on an NTFS volume won't
actually traverse the entire hierarchy of the directories passed
to it. (Those with a link count of two are not traversed at all)

The fix checks the "UFSness" of the filesystem before enabling the
optimisation.

Reviewed By: Tim Kientzle (kientzle@)
2004-05-08 15:09:02 +00:00
marius
bc0d53456e - Remove the old sparc64 OFW PCI code (as opposed to the former
"options OFW_NEWPCI").
  This is a bit overdue, the new sparc64 OFW PCI code which is
  meant to replace the old one is in place for 10 months and
  enabled by default in GENERIC for 8 months. FreeBSD 5.2 and
  5.2.1 also shipped with the new code enabled by default.
- Some minor clean-up, e.g. remove functions that encapsulated
  the #ifdefs for OFW_NEWPCI, remove unused resp. no longer
  required includes, etc.

Approved by:	tmm, no objections on freebsd-sparc64
2004-05-08 13:53:47 +00:00
julian
98d03d5d06 Fix rtprio() to do sensible things when called from threaded processes.
It's not quite correct from a posix Point Of view, but it is a lot better
than what was there before. This will be revisited later
when we decide what form our priority extensions will take. Posix doesn't
specify  how a system scope thread can change its priority so you need to
add non-standard extensions to be able to do it..
For now make this slightly non standard to allow it to be done.

Submitted by:	Dan Eischen originally, changed by myself.
2004-05-08 08:56:05 +00:00
alc
b0b9413238 Avoid pointless zeroing of the bogus page in vfs_bio_clrbuf().
Suggested by:	tegge@	(from October of last year)
2004-05-08 06:46:40 +00:00
imp
eb0b0158bc We don't need the dependency on the pccard module here. 2004-05-08 06:06:13 +00:00
imp
eaa54ff558 It turns out that the module dependency on pccard is in error. Since
there's not dependencies on pccard symboles, such a dependency is not
necessary.  This means that drivers that have multiple attachments can
not drag bogus devices into the kernel at load time.

We can't (yet) do this with pci and isa.  Drivers written for them
actually do seem to have symbols that depend on these busses'
implementation code.

ndis not touched until other things can be tested.
2004-05-08 06:04:06 +00:00
sanpei
40227f903a Add support CS4294
PR:		kern/66280
Submitted by:	Christian Brueffer <chris@unixpages.org>
2004-05-08 03:41:40 +00:00
rwatson
7128a3b5cc Unconditionally lock Giant in do_sendfile(), rather than locking it
conditional on debug.mpsafenet.  We can try pushing down Giant here
later, but we don't want to enter VFS without holding Giant.

Bumped into by:	kris
2004-05-08 02:24:21 +00:00
cjc
54af59dd3a It was pointed out[0] that ctags(1) uses some potentially dangerous
system(3) calls where user-supplied data is used with no sanity
checking. Since ctags(1) is not setuid and is not likely to be used
in a privileged situation, this is not a big deal. However, the
fix is relatively easy and less ugly than the current code, let's be
safe. (I'm sure there are about 2^134 other system(3) calls like this
out there.)

[0] On freebsd-security by Roman Bogorodskiy <bogorodskiy@inbox.ru>
with subject "ctags(1) command execution vulnerability."

MFC after:	3 days
2004-05-07 19:44:40 +00:00
jhb
7664209e3a Minor fixes for ia64 installs:
- Don't look for partitions inside a FreeBSD chunk on ia64 when mounting
  the filesystems just before the chroot and install.
- Write entries out to /etc/fstab for filesystems that aren't inside a
  FreeBSD chunk, but are a top-level chunk under the disk.
2004-05-07 19:15:56 +00:00
stefanf
2a0970f8ce Add an implementation of copysignl(), a long double version of copysign().
Approved by:	das (mentor)
2004-05-07 18:56:31 +00:00
stefanf
65ef3a8ab7 Add an MLINK for fabsl().
Approved by:	das (mentor)
2004-05-07 17:55:07 +00:00
rik
79864ee9b7 Add manual page for Cronyx Tau-PCI adapters.
It is PCI family, but we never test it on other than i386 arch.
2004-05-07 16:28:30 +00:00
cognet
50108a5412 Define KINFO_PROC_SIZE for arm. 2004-05-07 15:37:56 +00:00
cognet
6897229d63 Compare t_brkc against (char)_POSIX_VDISABLE, not against -1.
Discussed with:	bde
2004-05-07 15:35:38 +00:00
csjp
60a433f4c8 Teach route(8) how to deal with root being in a prison. If prison
root is allowed to create raw sockets, then they will be able to create
routing sockets, too.  However prison-root is not able to manipulate
routing tables.  So when route(8) attempts to write to a routing
socket and recieves EPERM from the kernel, exit rather than moving
on with execution.

Approved by:    bmilekic (mentor)
2004-05-07 15:33:17 +00:00
brueffer
79e214c972 o fix a typo
o remove stray word
o add missing words

PR:		66293
Submitted by:	Michel Lavondès <fox@vader.aacc.cc.md.us>
MFC after:	3 days
2004-05-07 13:00:01 +00:00
rik
e290adea78 Use better way of closing fr support before current sppp get it. 2004-05-07 11:59:54 +00:00
rik
5e58127328 Sync with RELENG_4 2004-05-07 11:56:07 +00:00
rik
f13a054c86 Use better way of closing fr support before current sppp doesn't have it. 2004-05-07 11:45:25 +00:00
green
129c6b32bf In r1.190, vslock() and vsunlock() were bogusly made to do a "user wire"
and a "system unwire."  Make this a "system wire" and "system unwire."

Reviewed by:	alc
2004-05-07 11:43:24 +00:00
rik
1e881321c1 Sync with RELENG_4. 2004-05-07 11:11:13 +00:00
rik
70641422ab Delete unused cx_slow_ih.
Pointed by:	jhb
2004-05-07 11:06:25 +00:00
des
c0f6282aac Document SHLIB, kind of. 2004-05-07 09:59:03 +00:00
des
e24f4ce2d9 Add SHLIB as a shortcut for shared-only libraries.
Not objected to by: bde, ru
2004-05-07 09:58:36 +00:00
marcel
27622b9560 Revert previous commit. We should not get any FP traps from within
the kernel. We can guarantee this by resetting the FP status register.
This masks all FP traps. The reason we did get FP traps was that we
didn't reset the FP status register in all cases.

Make sure to reset the FP status register in syscall(). This is one of
the places where it was forgotten.

While on the subject, reset the FP status register only when we trapped
from user space.
2004-05-07 05:35:31 +00:00
marcel
254a0ce34e Make sure to sanitize the FP status register. Specifically this
masks all FP traps, which should not happen in the kernel.
2004-05-07 05:29:12 +00:00
njl
055815ed7e Change hw.acpi.cpu.cx_lowest to accept values in the form of C1,
C2, ...  Update power_profile to use the new format.  Update the
man page to reflect this and give more info on Cx states.
2004-05-07 05:22:38 +00:00
marcel
f822beb736 Remove unwinder files that are commented-out. 2004-05-07 03:56:00 +00:00
maxim
5839c11830 o IFNAMSIZ does include the trailing \0.
Approved by:	andre

o Document net.inet.icmp.reply_src.
2004-05-07 01:24:53 +00:00
green
41ba6d4a43 Properly remove MAP_FUTUREWIRE when a vm_map_entry gets torn down.
Previously, mlockall(2) usage would leak MAP_FUTUREWIRE of the process's
vmspace::vm_map and subsequent processes would wire all of their memory.
Coupled with a wired-page leak in vm_fault_unwire(), this would run the
system out of free pages and cause programs to randomly SIGBUS when
faulting in new pages.

(Note that this is not the fix for the latter part; pages are still
 leaked when a wired area is unmapped in some cases.)

Reviewed by:	alc
PR		kern/62930
2004-05-07 00:17:07 +00:00
andre
832d1bd181 Provide the sysctl net.inet.ip.process_options to control the processing
of IP options.

 net.inet.ip.process_options=0  Ignore IP options and pass packets unmodified.
 net.inet.ip.process_options=1  Process all IP options (default).
 net.inet.ip.process_options=2  Reject all packets with IP options with ICMP
  filter prohibited message.

This sysctl affects packets destined for the local host as well as those
only transiting through the host (routing).

IP options do not have any legitimate purpose anymore and are only used
to circumvent firewalls or to exploit certain behaviours or bugs in TCP/IP
stacks.

Reviewed by:	sam (mentor)
2004-05-06 18:46:03 +00:00
njl
d33cf3a8dd Rename acpi_cpu to cpu. Change the probe routine to early on reject
devices it cannot attach to.  This gets rid of extraneous but harmless
device_probe_and_attach() errors.  While I'm here, make the device
description more useful.  The !acpi case for cpu is handled by legacy0.
2004-05-06 17:25:23 +00:00
njl
e9ec8dbd49 Move the CPU newbus attachment to i386 legacy. The acpi_cpu device will
become just "cpu" and provide attachments in the !legacy case.

Tested by:	des
2004-05-06 15:54:02 +00:00
joerg
d5ea7dd1b1 Mention that hme(4) now supports long frames for vlan(4) natively. 2004-05-06 13:51:00 +00:00