]> CyberLeo.Net >> Repos - FreeBSD/FreeBSD.git/log
FreeBSD/FreeBSD.git
14 years agoBump .Dd for r203620
gavin [Sun, 7 Feb 2010 18:06:55 +0000 (18:06 +0000)]
Bump .Dd for r203620

MFC after: 1 week

14 years agoDocument support for the D-Link DFE520-TX card (supported with the vr(4)
gavin [Sun, 7 Feb 2010 18:05:12 +0000 (18:05 +0000)]
Document support for the D-Link DFE520-TX card (supported with the vr(4)
driver)

PR: kern/135989
Submitted by: "Rashid N. Achilov"  citycat4 ngs.ru

14 years agoSpelling nit
gavin [Sun, 7 Feb 2010 18:00:13 +0000 (18:00 +0000)]
Spelling nit

14 years agoTell the compiler these structures are aligned to a byte boundary.
imp [Sun, 7 Feb 2010 17:05:22 +0000 (17:05 +0000)]
Tell the compiler these structures are aligned to a byte boundary.
All the elements of these structs are char anyway, so it won't hurt
performance.

Bump warns back up to the default.

# we likely should have CTASSERTS to make sure they are the right size.
# but with libarchive based tar maybe we shouldn't bother.

14 years agoRemove statistics from the TTY queues.
ed [Sun, 7 Feb 2010 15:42:15 +0000 (15:42 +0000)]
Remove statistics from the TTY queues.

I added counters to see how often fast copying to userspace was actually
performed, which was only useful during development. Remove these
statistics now we know it to be effective.

14 years agoFix whitespace in pho's entry, results in nicer output files.
brueffer [Sun, 7 Feb 2010 13:59:03 +0000 (13:59 +0000)]
Fix whitespace in pho's entry, results in nicer output files.

PR: 143613
Submitted by: pluknet <pluknet@gmail.com>
Committed from: Debian/kFreeBSD talk at FOSDEM

14 years agoInitialize fromlen before calling recvfrom to avoid passing in random
imp [Sun, 7 Feb 2010 07:50:41 +0000 (07:50 +0000)]
Initialize fromlen before calling recvfrom to avoid passing in random
stack garbage.

Obtained from: NetBSD 1.13

14 years agoSimplify, remove unnecessary code.
kientzle [Sun, 7 Feb 2010 02:04:58 +0000 (02:04 +0000)]
Simplify, remove unnecessary code.

14 years agoMerge a bunch of refactoring from Joerg Sonnenberger to
kientzle [Sun, 7 Feb 2010 02:00:26 +0000 (02:00 +0000)]
Merge a bunch of refactoring from Joerg Sonnenberger to
isolate common code used by tar and cpio (and useful to other
libarchive clients).  The functions here are prefixed with
"lafe" (libarchive front-end) to indicate their use.

14 years agoStyle & Portability: Use archive_entry methods to examine
kientzle [Sun, 7 Feb 2010 01:35:28 +0000 (01:35 +0000)]
Style & Portability:  Use archive_entry methods to examine
file information, change some functions to static, remove
some unused headers.

14 years agoVarious portability workarounds for non-FreeBSD platforms.
kientzle [Sun, 7 Feb 2010 01:26:45 +0000 (01:26 +0000)]
Various portability workarounds for non-FreeBSD platforms.

14 years agoRestructure the logic that determines when we're crossing a mount
kientzle [Sun, 7 Feb 2010 01:22:55 +0000 (01:22 +0000)]
Restructure the logic that determines when we're crossing a mount
point.  In particular, this carves out a place for detecting and
excluding synthetic or network filesystems.

14 years agoMinor code rework.
kientzle [Sun, 7 Feb 2010 01:16:05 +0000 (01:16 +0000)]
Minor code rework.

14 years agoTrim out some unused configuration variables, remove
kientzle [Sun, 7 Feb 2010 01:07:58 +0000 (01:07 +0000)]
Trim out some unused configuration variables, remove
some unused headers, etc.

14 years agoAdd files for NO_MAIL, NO_SENDMAIL and NO_MAILWRAPPER.
delphij [Sun, 7 Feb 2010 00:38:31 +0000 (00:38 +0000)]
Add files for NO_MAIL, NO_SENDMAIL and NO_MAILWRAPPER.

MFC after: 2 weeks

14 years agosh: Do not stat() $MAIL/$MAILPATH in non-interactive shells.
jilles [Sat, 6 Feb 2010 22:57:24 +0000 (22:57 +0000)]
sh: Do not stat() $MAIL/$MAILPATH in non-interactive shells.

These may be NFS mounted, and we should not touch them unless we are going
to do something useful with the information.

14 years agoAdd sane-port (Scanner Access Now Easy) as port 6566.
bms [Sat, 6 Feb 2010 21:22:01 +0000 (21:22 +0000)]
Add sane-port (Scanner Access Now Easy) as port 6566.

Obtained from:  http://www.iana.org/assignments/port-numbers
MFC after:      3 days

14 years agoFix single-stepping when the kernel was entered through the EPC syscall
marcel [Sat, 6 Feb 2010 20:46:14 +0000 (20:46 +0000)]
Fix single-stepping when the kernel was entered through the EPC syscall
path. When the taken branch leaves the kernel and enters the process,
we still need to execute the instruction at that address. Don't raise
SIGTRAP when we branch into the process, but enable single-stepping
instead.

14 years agoFill in some missing error handling, be a little more careful about
kientzle [Sat, 6 Feb 2010 20:41:25 +0000 (20:41 +0000)]
Fill in some missing error handling, be a little more careful about
error reporting, prefer int64_t to off_t.

14 years agobsdtar doesn't actually know what compression is supported by libarchive
kientzle [Sat, 6 Feb 2010 20:36:14 +0000 (20:36 +0000)]
bsdtar doesn't actually know what compression is supported by libarchive
and it should not pretend that it does.  It should just pass along the
user's request and handle an error if it's not supported.

14 years agoRefactor the siginfo/sigusr1 handling. The read/write reporting
kientzle [Sat, 6 Feb 2010 20:27:36 +0000 (20:27 +0000)]
Refactor the siginfo/sigusr1 handling.  The read/write reporting
is sufficiently different that it was simpler to just put separate
reporting functions into read.c and write.c rather than try to have
a single all-purpose reporting function.
Switch to a custom function for converting int64_t to a string; in
the portable version, this saves a lot of configuration headaches
trying to decipher the platform printf().

14 years agoUpdate files to remove when MK_KERBEROS=no.
antoine [Sat, 6 Feb 2010 20:21:03 +0000 (20:21 +0000)]
Update files to remove when MK_KERBEROS=no.

14 years agoAdd files to remove when MK_JAIL=no.
antoine [Sat, 6 Feb 2010 20:11:23 +0000 (20:11 +0000)]
Add files to remove when MK_JAIL=no.

14 years agoUpdate files to remove when MK_IPX=no.
antoine [Sat, 6 Feb 2010 20:09:55 +0000 (20:09 +0000)]
Update files to remove when MK_IPX=no.

14 years agoAdd files to remove when MK_IPFW=no.
antoine [Sat, 6 Feb 2010 20:08:46 +0000 (20:08 +0000)]
Add files to remove when MK_IPFW=no.

14 years agoMove rescue/ping6 from files to remove when MK_INET6=no to file to remove
antoine [Sat, 6 Feb 2010 20:06:06 +0000 (20:06 +0000)]
Move rescue/ping6 from files to remove when MK_INET6=no to file to remove
when MK_INET6_SUPPORT=no.  (MK_INET6=no implies MK_INET6_SUPPORT=no)

14 years agoAdd file to remove when MK_IDEA=no.
antoine [Sat, 6 Feb 2010 20:02:56 +0000 (20:02 +0000)]
Add file to remove when MK_IDEA=no.

14 years agoAdd files to remove when MK_GSSAPI=no.
antoine [Sat, 6 Feb 2010 20:00:35 +0000 (20:00 +0000)]
Add files to remove when MK_GSSAPI=no.

14 years agoRemove all traces of an experiment for handling "root" on Windows systems.
kientzle [Sat, 6 Feb 2010 19:56:32 +0000 (19:56 +0000)]
Remove all traces of an experiment for handling "root" on Windows systems.

14 years agoReformat the Makefile slightly.
kientzle [Sat, 6 Feb 2010 19:53:48 +0000 (19:53 +0000)]
Reformat the Makefile slightly.

14 years agoAllow -b up to 8192. I've had reports from people who routinely
kientzle [Sat, 6 Feb 2010 19:48:59 +0000 (19:48 +0000)]
Allow -b up to 8192.  I've had reports from people who routinely
use -b 2048 (1MiB block size).  Setting the limit to 8192 should
allow some room for growth while still helping people who mistakenly
put in byte counts here instead of block counts.

14 years agoDiff reduction compared to portable bsdtar 2.8: Move the
kientzle [Sat, 6 Feb 2010 19:44:37 +0000 (19:44 +0000)]
Diff reduction compared to portable bsdtar 2.8:  Move the
program name into a global, which eliminates an extra
argument from a lot of places.

14 years agoRevert unwanted changes in revision 203422.
rpaulo [Sat, 6 Feb 2010 19:24:16 +0000 (19:24 +0000)]
Revert unwanted changes in revision 203422.

Spotted by: sam

14 years agoAdd myself.
bschmidt [Sat, 6 Feb 2010 18:10:58 +0000 (18:10 +0000)]
Add myself.

Approved by: rpaulo (mentor)

14 years ago- Cast intptr_t, pid_t and time_t values to intmax_t and use %jd with
jh [Sat, 6 Feb 2010 16:01:38 +0000 (16:01 +0000)]
- Cast intptr_t, pid_t and time_t values to intmax_t and use %jd with
  printf.
- Cast the system call return value to long and use %ld in a printf in
  ktrsysret().

PR: bin/123774
MFC after: 2 weeks

14 years agoAdd missing coma.
trasz [Sat, 6 Feb 2010 14:10:45 +0000 (14:10 +0000)]
Add missing coma.

14 years agoPropagate the vlan eventis to the underlying interfaces/members so they can do initia...
eri [Sat, 6 Feb 2010 13:49:35 +0000 (13:49 +0000)]
Propagate the vlan eventis to the underlying interfaces/members so they can do initialization of hw related features.

PR: kern/141646
Reviewed by: thompsa
Approved by: thompsa(co-mentor)
MFC after: 2 weeks

14 years agoAdd the -i option to the synopsis.
gavin [Sat, 6 Feb 2010 13:39:08 +0000 (13:39 +0000)]
Add the -i option to the synopsis.

Submitted by: dhw
MFC after: 1 week (with r203310)

14 years agoacpi_cpu: prefer _OSC over _PDC, just in case
avg [Sat, 6 Feb 2010 12:48:06 +0000 (12:48 +0000)]
acpi_cpu: prefer _OSC over _PDC, just in case

_PDC was deprecated in favor of _OSC long time ago, but it
seems that they still peacefully coexist and in some case
only _PDC is present.
Still _OSC provides a reacher interface and is capable to
report back its status.
If the status is non-zero, then report it, we may find
it useful to understand what firmware expects from OS.
Also clean up some comments that became less useful over time.

Reviewed by: njl, jhb, rpaulo
MFC after: 3 weeks

14 years agoDocument one more file descriptor type and two more vnode types.
antoine [Sat, 6 Feb 2010 11:29:06 +0000 (11:29 +0000)]
Document one more file descriptor type and two more vnode types.

MFC after: 2 weeks

14 years agoMap and report actual video memory we need.
jkim [Sat, 6 Feb 2010 00:52:42 +0000 (00:52 +0000)]
Map and report actual video memory we need.

14 years agoCorrect two typos.
delphij [Sat, 6 Feb 2010 00:25:46 +0000 (00:25 +0000)]
Correct two typos.

Reported by: Brandon Falk <falkman gamozo org>
MFC after: 1 week

14 years agoRemove two files that are not needed by FreeBSD.
delphij [Fri, 5 Feb 2010 23:17:59 +0000 (23:17 +0000)]
Remove two files that are not needed by FreeBSD.

Approved by: pjd
MFC after: 2 weeks

14 years agoFix gcore so that it can have the '-s' flag without hanging.
mjacob [Fri, 5 Feb 2010 18:28:43 +0000 (18:28 +0000)]
Fix gcore so that it can have the '-s' flag without hanging.

14 years agoFix a bug in previous revision.
delphij [Fri, 5 Feb 2010 18:17:17 +0000 (18:17 +0000)]
Fix a bug in previous revision.

The bc(1) program may need to deal with files when it's being run in
interactive mode, so we can not blindly use interactive mode (in turn
use libedit) but need to check if the input source is really the standard
input.

This commit should fix a regression where 'bc -l' would not parse the
mathlib.

Reported by: trasz

14 years agoReplace some homegrown functions with better/correct ones.
jkim [Fri, 5 Feb 2010 18:00:24 +0000 (18:00 +0000)]
Replace some homegrown functions with better/correct ones.

14 years agoUse new helper functions to set PCIe max read request size.
mav [Fri, 5 Feb 2010 17:20:48 +0000 (17:20 +0000)]
Use new helper functions to set PCIe max read request size.

14 years agoAdd pci_get|set_max_read_req() helper functions to control maximum PCIe
mav [Fri, 5 Feb 2010 17:18:48 +0000 (17:18 +0000)]
Add pci_get|set_max_read_req() helper functions to control maximum PCIe
read request size.

Reviewed by: jhb@

14 years agoAdd a description of supported PowerPC hardware to the hardware notes.
nwhitehorn [Fri, 5 Feb 2010 16:41:19 +0000 (16:41 +0000)]
Add a description of supported PowerPC hardware to the hardware notes.

14 years agoReport SATA300 chips also as SATA.
mav [Fri, 5 Feb 2010 14:41:18 +0000 (14:41 +0000)]
Report SATA300 chips also as SATA.

14 years agoWhen hacking INQUIRY result, make sure that it is right INQUIRY and there
mav [Fri, 5 Feb 2010 12:40:18 +0000 (12:40 +0000)]
When hacking INQUIRY result, make sure that it is right INQUIRY and there
is enough of result to hack.

14 years agoCorrect a comment - we are not setting the exception level but rather are
neel [Fri, 5 Feb 2010 06:36:03 +0000 (06:36 +0000)]
Correct a comment - we are not setting the exception level but rather are
disabling interrupts.

Simplify register usage - we can directly load 'curpcb' into 'k1' after
interrupts are disabled. There is no need to do so indirectly through 'a1'.

14 years agoInitialize interrupt controller early on.
neel [Fri, 5 Feb 2010 03:22:04 +0000 (03:22 +0000)]
Initialize interrupt controller early on.

14 years agoReimplement all functions to access the system control unit in C.
neel [Fri, 5 Feb 2010 03:20:47 +0000 (03:20 +0000)]
Reimplement all functions to access the system control unit in C.

The only reason we need to have the sb_load64() and sb_store64()
functions in assembly is to cheat the compiler and generate the
'ld' and 'sd' instructions which it otherwise will not do when
compiling for a 32-bit architecture. There are some 64-bit
registers in the SCD unit that must be accessed using 64-bit
load and store instructions.

14 years agostyle: don't need to use braces for single line control statements.
neel [Fri, 5 Feb 2010 02:40:42 +0000 (02:40 +0000)]
style: don't need to use braces for single line control statements.

14 years agoThe ZTE MF633R modem has a different type of cdrom driver disk, add the product
thompsa [Fri, 5 Feb 2010 01:47:43 +0000 (01:47 +0000)]
The ZTE MF633R modem has a different type of cdrom driver disk, add the product
ID and use a standard scsi eject.

Reported by: Patrick Lamaiziere
MFC after: 3 days

14 years agoProperly name the 0x0016 ZTE product as MF633R now that its known.
thompsa [Thu, 4 Feb 2010 23:34:35 +0000 (23:34 +0000)]
Properly name the 0x0016 ZTE product as MF633R now that its known.

14 years agoAdd the long missing "destroy" option.
mjacob [Thu, 4 Feb 2010 21:51:33 +0000 (21:51 +0000)]
Add the long missing "destroy" option.

MFC after: 2 weeks

14 years agoOpen provider for writting when we find the right one. Opening too much
pjd [Thu, 4 Feb 2010 21:11:44 +0000 (21:11 +0000)]
Open provider for writting when we find the right one. Opening too much
providers for writing provokes huge traffic related to taste events send
by GEOM on close. This can lead to various problems with opening GEOM
providers that are created on top of other GEOM providers.

Reorted by: Kurt Touet <ktouet@gmail.com>, mr
Tested by: mr, Baginski Darren <kickbsd@ya.ru>
MFC after: 2 weeks

14 years agoRestore the checksum received before processing the packet.
tuexen [Thu, 4 Feb 2010 21:02:29 +0000 (21:02 +0000)]
Restore the checksum received before processing the packet.

MFC after: 1 week

14 years agoOn probe error, if restart requested, skip any retries and recovery.
mav [Thu, 4 Feb 2010 18:56:38 +0000 (18:56 +0000)]
On probe error, if restart requested, skip any retries and recovery.
Just restart probe from the beginning immediately.

14 years agoUse libedit when interacting with tty, which provided history
delphij [Thu, 4 Feb 2010 18:43:05 +0000 (18:43 +0000)]
Use libedit when interacting with tty, which provided history
functionality, etc. as did by GNU bc.

This also fixes an issue where BSDL bc can not handle very long
line.

Reported by: imp
Reviewed by: imp

14 years agoIntroduce '[ipaddr]:path' notation.
ume [Thu, 4 Feb 2010 15:17:49 +0000 (15:17 +0000)]
Introduce '[ipaddr]:path' notation.
Since the existing implementation searches ':' backward, a path which
includes ':' could not be mounted.  You can now mount such path by
enclosing an IP address by '[]'.
Though we should change to search ':' forward, it will break
'ipv6addr:path' which is currently working.  So, it still searches ':'
backward, at least for now.

MFC after: 2 weeks

14 years agoReturn CAM_RELEASE_SIMQ flag only when it is needed, when SIM really
mav [Thu, 4 Feb 2010 14:40:08 +0000 (14:40 +0000)]
Return CAM_RELEASE_SIMQ flag only when it is needed, when SIM really
was frozen before and should be released.

14 years agoShortening a passphrase caused wrong authentication key to be used.
ru [Thu, 4 Feb 2010 11:43:22 +0000 (11:43 +0000)]
Shortening a passphrase caused wrong authentication key to be used.
Fix this in a FreeBSD and OpenBSD compatible way.

MFC after: 3 days

14 years agoMark up "unsigned char".
ru [Thu, 4 Feb 2010 11:23:28 +0000 (11:23 +0000)]
Mark up "unsigned char".

14 years agoDo not release device, when changing number of openings.
mav [Thu, 4 Feb 2010 08:54:14 +0000 (08:54 +0000)]
Do not release device, when changing number of openings.

14 years agoInstead of spamming the console on each curvnet recursion event, print
zec [Thu, 4 Feb 2010 07:55:42 +0000 (07:55 +0000)]
Instead of spamming the console on each curvnet recursion event, print
out each such call graph only once, along with a stack backtrace.  This
should make kernels built with VNET_DEBUG reasonably usable again in
busy / production environments.

Introduce a new DDB command "show vnetrcrs" which dumps the whole log
of distinctive curvnet recursion events.  This might be useful when
recursion reports get burried / lost too deep in the message buffer.
In the later case stack backtraces are not available.

Reviewed by: bz
MFC after: 3 days

14 years agoSince APM is a legacy technology, that won't be adapted to other
imp [Thu, 4 Feb 2010 07:26:26 +0000 (07:26 +0000)]
Since APM is a legacy technology, that won't be adapted to other
platforms, move the test for i386 from the Makefile to powerd.c.  The
powerpc specific checks are already done this way...

14 years agoArm doesn't seem to need such kit gloves. Bump the warnings to 3 here
imp [Thu, 4 Feb 2010 07:18:19 +0000 (07:18 +0000)]
Arm doesn't seem to need such kit gloves.  Bump the warnings to 3 here
as well.

14 years agoAlways compile in the speaker code. There's little savings by
imp [Thu, 4 Feb 2010 07:08:06 +0000 (07:08 +0000)]
Always compile in the speaker code.  There's little savings by
omitting it, and other platforms may implement /dev/speaker in the
future.

14 years agoYet another target mode compilation error.
mjacob [Thu, 4 Feb 2010 06:45:14 +0000 (06:45 +0000)]
Yet another target mode compilation error.

14 years agoCompile SWARM with KTRACE support.
neel [Thu, 4 Feb 2010 06:44:42 +0000 (06:44 +0000)]
Compile SWARM with KTRACE support.

14 years agoGet system call tracing using ktrace working for mips.
neel [Thu, 4 Feb 2010 06:42:30 +0000 (06:42 +0000)]
Get system call tracing using ktrace working for mips.

14 years agoReinstate the ptrace patch to restore the 'gp' register after calling
neel [Thu, 4 Feb 2010 05:49:59 +0000 (05:49 +0000)]
Reinstate the ptrace patch to restore the 'gp' register after calling
a function. I made a mistake in assuming that the .cprestore directive
will cause the assembler to automatically restore 'gp' after the 'jalr'.

The .cprestore directive does its magic only after 'jal' and 'bal'
instructions - not the 'jalr'.

Pointed out by: c.jayachandran@gmail.com

14 years agoClean up all places in exception.S that fiddle with 'pcpup' directly. We now
neel [Thu, 4 Feb 2010 05:25:59 +0000 (05:25 +0000)]
Clean up all places in exception.S that fiddle with 'pcpup' directly. We now
use the GET_CPU_PCPU() macro exclusively.

This isolates the users of pcpu data from its implementation details.

Reviewed by: imp

14 years agoFix target mode compilation problem with previous delta
mjacob [Thu, 4 Feb 2010 00:40:12 +0000 (00:40 +0000)]
Fix target mode compilation problem with previous delta

14 years agostatic'ify function prototypes and convert K&R to ANSI.
delphij [Wed, 3 Feb 2010 23:59:52 +0000 (23:59 +0000)]
static'ify function prototypes and convert K&R to ANSI.

MFC after: 1 month

14 years agopukeText is an internal function so define it as static rather than
delphij [Wed, 3 Feb 2010 23:58:00 +0000 (23:58 +0000)]
pukeText is an internal function so define it as static rather than
exporting it.

MFC after: 1 month

14 years agoPlug two memory leaks in error case.
delphij [Wed, 3 Feb 2010 23:56:12 +0000 (23:56 +0000)]
Plug two memory leaks in error case.

MFC after: 1 month

14 years agoRemove dead code and fix style(9) bugs.
jkim [Wed, 3 Feb 2010 22:20:57 +0000 (22:20 +0000)]
Remove dead code and fix style(9) bugs.

14 years agoUse bytes per scan line from mode table. The previous implementation did
jkim [Wed, 3 Feb 2010 22:17:30 +0000 (22:17 +0000)]
Use bytes per scan line from mode table.  The previous implementation did
not reflect actual number of bytes when it was not exactly width * bpp * 8.

14 years agoCorrect virtual address of frame buffer for non-linear mode.
jkim [Wed, 3 Feb 2010 22:07:50 +0000 (22:07 +0000)]
Correct virtual address of frame buffer for non-linear mode.

Reported by: Marc UBM Bocklet (ubm dot freebsd at googlemail dot com)

14 years agoImplement poll method for atapicam.
mav [Wed, 3 Feb 2010 21:45:09 +0000 (21:45 +0000)]
Implement poll method for atapicam.
It is not perfect, but better then nothing.

14 years agoReport to CAM unit number of underlying ataX bus device, instead of this
mav [Wed, 3 Feb 2010 21:26:54 +0000 (21:26 +0000)]
Report to CAM unit number of underlying ataX bus device, instead of this
atapicamX. It is more apropriate and useful together with "ata" name.

14 years agoRedo how commands handles are created and managed and implement sequence
mjacob [Wed, 3 Feb 2010 21:09:32 +0000 (21:09 +0000)]
Redo how commands handles are created and managed and implement sequence
numbers and handle types in rational way. This will better protect from
(unwittingly) dealing with stale handles/commands.

Fix the watchdog timeout code to better protect itself from mistakes.

If we run an abort on a putatively timed out command, the command
may in fact get completed, so check to make sure the command we're
timing it out is still around. If the abort succeeds, btw, the command
should get returned via a different path.

14 years ago- style(9)
gabor [Wed, 3 Feb 2010 21:06:13 +0000 (21:06 +0000)]
- style(9)

Approved by: delphij (mentor)

14 years ago-White space cleanup (missing spaces in new line)
rrs [Wed, 3 Feb 2010 20:54:04 +0000 (20:54 +0000)]
-White space cleanup (missing spaces in new line)
-Remove extra tab.
-Took out the duplicate code that cprestore does.
All suggested by Neel.

14 years agoPlacate new binutils, by using 16-bit %ax instead of 32-bit %eax as an
kib [Wed, 3 Feb 2010 20:23:47 +0000 (20:23 +0000)]
Placate new binutils, by using 16-bit %ax instead of 32-bit %eax as an
argument for fnstsw. Explicitely specify sizes for the XMM control and
status word and X87 control and status words.

Reviewed by: das
Tested by: avg
MFC after: 2 weeks

14 years agoDisable kern.cam.power_down ATM. It doesn't work fine on some controllers.
mav [Wed, 3 Feb 2010 20:00:56 +0000 (20:00 +0000)]
Disable kern.cam.power_down ATM. It doesn't work fine on some controllers.

14 years agoAlso add a cross-reference to NetBSD updlastlogx().
ed [Wed, 3 Feb 2010 19:31:02 +0000 (19:31 +0000)]
Also add a cross-reference to NetBSD updlastlogx().

14 years ago- Remove dead variable assignments
gabor [Wed, 3 Feb 2010 19:13:41 +0000 (19:13 +0000)]
- Remove dead variable assignments
- Fix -x

Submitted by: uqs
Approved by: delphij (mentor)

14 years ago- style.Makefile(5)
gabor [Wed, 3 Feb 2010 19:12:27 +0000 (19:12 +0000)]
- style.Makefile(5)

Submitted by: uqs
Approved by: delphij (mentor)

14 years agoRemove alpha support from here too. FreeBSD 6 was the last branch to
imp [Wed, 3 Feb 2010 18:32:29 +0000 (18:32 +0000)]
Remove alpha support from here too.  FreeBSD 6 was the last branch to
support it.

14 years agoRemove more traces of the Alpha port.
imp [Wed, 3 Feb 2010 18:29:35 +0000 (18:29 +0000)]
Remove more traces of the Alpha port.

14 years agoAdd rc.d script for the rtsold(8) daemon.
ume [Wed, 3 Feb 2010 16:18:42 +0000 (16:18 +0000)]
Add rc.d script for the rtsold(8) daemon.

The rtsol(8) handles just one RA then exit.  So, the OtherConfig flag
may not be handled well by rtsol(8) in the environment where there are
multiple RA servers on the segment.  In such case, rtsold(8) will be
your friend.

Reviewed by: hrs
MFC after: 2 weeks

14 years agoacpi_cpu: correct capabilities arguments for Processor _OSC evaluation
avg [Wed, 3 Feb 2010 14:35:33 +0000 (14:35 +0000)]
acpi_cpu: correct capabilities arguments for Processor _OSC evaluation

Populate capabilities buffer according to
Intel Processor Vendor-Specific ACPI Interface Specification.

MFC after: 2 weeks

14 years agoThis fix makes it so the args are malloc'd before
rrs [Wed, 3 Feb 2010 13:03:47 +0000 (13:03 +0000)]
This fix makes it so the args are malloc'd before
we use them (always a good idea ;-0)

This was found and fixed by JC

Obtained from: JC (c.jayachandran@gmail.com)

14 years agoThis fixes a bug found and fixed by JC. Basically
rrs [Wed, 3 Feb 2010 12:59:44 +0000 (12:59 +0000)]
This fixes a bug found and fixed by JC. Basically
no save was being done of the ra and gp pointers
before we call the __error function.

Obtained from: JC (c.jayachandran@gmail.com)

14 years agoDisable PHY of unconnected ports when interface power management enabled.
mav [Wed, 3 Feb 2010 12:00:57 +0000 (12:00 +0000)]
Disable PHY of unconnected ports when interface power management enabled.
It allows to save a bit more power (about 0.5W on 2 unused ports of ICH8M).