]> CyberLeo.Net >> Repos - FreeBSD/FreeBSD.git/log
FreeBSD/FreeBSD.git
11 years agoAdd some further BAR TX debugging; it was useful when figuring out
adrian [Sun, 28 Oct 2012 04:18:49 +0000 (04:18 +0000)]
Add some further BAR TX debugging; it was useful when figuring out
when BAR TX was actually failing.

11 years agoBetter comments.
imp [Sun, 28 Oct 2012 02:55:51 +0000 (02:55 +0000)]
Better comments.

11 years agoExtend dim's hack from r228978: not only clang but gcc on non-x86 platforms
nwhitehorn [Sun, 28 Oct 2012 02:15:35 +0000 (02:15 +0000)]
Extend dim's hack from r228978: not only clang but gcc on non-x86 platforms
warns about unused variables in this code, so always add -Wno-unused to
the warning flags. Why gcc on x86 *doesn't* warn about this, I will never
know. The code itself should probably be fixed at some point.

11 years agoThe fields of struct timespec32 should be int32_t and not uint32_t.
davide [Sat, 27 Oct 2012 23:42:41 +0000 (23:42 +0000)]
The fields of struct timespec32 should be int32_t and not uint32_t.
Make this change.

Reviewed by: bde, davidxu
Tested by: pho
MFC after: 1 week

11 years agoAdd missing return that broke 8-bit CF support in refactoring in r222671.
jmallett [Sat, 27 Oct 2012 23:36:41 +0000 (23:36 +0000)]
Add missing return that broke 8-bit CF support in refactoring in r222671.
Tested on a Cavium CN5860-EVB-NIC4.  This was broken for over a year.

11 years agoDon't try to build Linux compatibility stuff on platforms without
nwhitehorn [Sat, 27 Oct 2012 23:14:37 +0000 (23:14 +0000)]
Don't try to build Linux compatibility stuff on platforms without
COMPAT_LINUX.

11 years agoEliminate a redundant TLB invalidation from pmap_pv_reclaim().
alc [Sat, 27 Oct 2012 22:43:30 +0000 (22:43 +0000)]
Eliminate a redundant TLB invalidation from pmap_pv_reclaim().

11 years agoMissing paren.
kientzle [Sat, 27 Oct 2012 22:13:42 +0000 (22:13 +0000)]
Missing paren.

Pointy hat:me

11 years agoFix bug introduced by r241902 (MANIFEST uses TAB delimiter).
dteske [Sat, 27 Oct 2012 19:56:57 +0000 (19:56 +0000)]
Fix bug introduced by r241902 (MANIFEST uses TAB delimiter).

PR: bin/173140
Approved by: adrian (co-mentor)

11 years agoAdd setfib(1) support for services as <name>_fib in rc.conf.
hrs [Sat, 27 Oct 2012 19:09:09 +0000 (19:09 +0000)]
Add setfib(1) support for services as <name>_fib in rc.conf.

11 years agoAllow spaces in _chroot
crees [Sat, 27 Oct 2012 17:43:30 +0000 (17:43 +0000)]
Allow spaces in _chroot

Noticed by: adj (IRC/#bsdports)
Approved by: hrs
MFC after: 1 month

11 years agoFollow clang lead and include mm_malloc.h only in hosted configurations.
kan [Sat, 27 Oct 2012 17:39:36 +0000 (17:39 +0000)]
Follow clang lead and include  mm_malloc.h only in hosted configurations.

This makes the use of intrinsics easier in kernel environment, according
to the submitter.

Requested by: jmg

11 years agoFix an issue when ipv6_enable=YES && ipv6_gateway_enable=YES which could
hrs [Sat, 27 Oct 2012 17:06:26 +0000 (17:06 +0000)]
Fix an issue when ipv6_enable=YES && ipv6_gateway_enable=YES which could
prevent rtadvd(8) from working as intended.

Spotted by: brian
Discussed with: brian

11 years agodrm(4) works just fine on PowerPC, so connect it to the build.
nwhitehorn [Sat, 27 Oct 2012 16:07:38 +0000 (16:07 +0000)]
drm(4) works just fine on PowerPC, so connect it to the build.

MFC after: 2 weeks

11 years agoRemove priority enforcement from xpt_ation(). It is not good and even not
mav [Sat, 27 Oct 2012 10:14:12 +0000 (10:14 +0000)]
Remove priority enforcement from xpt_ation().  It is not good and even not
safe in some cases to reduce CCB priority after it was scheduled with high
priority.  This fixes reproducible deadlock when command sent through the
pass interface while ATA XPT recovers from command timeout.

Instead of that enforce priority at passioctl().  libcam provides no obvious
interface to specify CCB priority and so much (all?) code specifies zero
(highest) priority.  This change limits pass CCBs priority to NORMAL run
level, allowing XPT to complete bus and device recovery after reset before
running any payload.

11 years agoRemove several uses of numeric priorities from immediate CCB setups.
mav [Sat, 27 Oct 2012 09:40:29 +0000 (09:40 +0000)]
Remove several uses of numeric priorities from immediate CCB setups.

11 years agoRemove one more numeric priority constant.
mav [Sat, 27 Oct 2012 08:52:33 +0000 (08:52 +0000)]
Remove one more numeric priority constant.

11 years agoComment out the other BOOTP option
kientzle [Sat, 27 Oct 2012 04:02:12 +0000 (04:02 +0000)]
Comment out the other BOOTP option
This should make PANDABOARD suitable for building
bootable SD images.

Submitted by: Giovanni Trematerra

11 years agostack_machdep.c is dependent on ddb or stack options, not standard.
imp [Fri, 26 Oct 2012 21:25:10 +0000 (21:25 +0000)]
stack_machdep.c is dependent on ddb or stack options, not standard.

11 years agoo Remove last argument to ip_fragment(), and obtain all needed information
glebius [Fri, 26 Oct 2012 21:06:33 +0000 (21:06 +0000)]
o Remove last argument to ip_fragment(), and obtain all needed information
  on checksums directly from mbuf flags. This simplifies code.
o Clear CSUM_IP from the mbuf in ip_fragment() if we did checksums in
  hardware. Some driver may not announce CSUM_IP in theur if_hwassist,
  although try to do checksums if CSUM_IP set on mbuf. Example is em(4).
o While here, consistently use CSUM_IP instead of its alias CSUM_DELAY_IP.
  After this change CSUM_DELAY_IP vanishes from the stack.

Submitted by: Sebastian Kuzminsky <seb lineratesystems.com>

11 years agoSiba, in theory, is a architecturally neutral bus, so place it in
imp [Fri, 26 Oct 2012 20:43:30 +0000 (20:43 +0000)]
Siba, in theory, is a architecturally neutral bus, so place it in
files.  It used to be in files.mips before the clean-room rewrite and
really doesn't belong there.  If we need to grow arch specific code,
we can move it into $ARCH/$ACH/siba_machdep.c.

11 years agoWhile 'make universe' passed this didn't work with clang.
eadler [Fri, 26 Oct 2012 20:25:05 +0000 (20:25 +0000)]
While 'make universe' passed this didn't work with clang.

This reverts r242120

Submitted by: Jan Beich
Approved by: cperciva (implicit)

11 years agoA little bit easier to read.
obrien [Fri, 26 Oct 2012 20:24:13 +0000 (20:24 +0000)]
A little bit easier to read.

11 years agoTest both active and non-active cases.
obrien [Fri, 26 Oct 2012 20:14:40 +0000 (20:14 +0000)]
Test both active and non-active cases.

11 years agoImplement CAM_ATAIO_NEEDRESULT (fetching full set of result registers) for
mav [Fri, 26 Oct 2012 20:03:08 +0000 (20:03 +0000)]
Implement CAM_ATAIO_NEEDRESULT (fetching full set of result registers) for
ata(4) driver in ATA_CAM mode.  That slighty improves error reporting and
also should fix `smartctl -l scterc /dev/adaX` operation.

MFC after: 3 weeks

11 years agoOops, missed in my last commit.
adrian [Fri, 26 Oct 2012 19:46:55 +0000 (19:46 +0000)]
Oops, missed in my last commit.

11 years agoAllow net80211 to be built on -9 and -8.
adrian [Fri, 26 Oct 2012 19:06:24 +0000 (19:06 +0000)]
Allow net80211 to be built on -9 and -8.

There are some people who use the -HEAD net80211 and wireless drivers
on earlier FreeBSD versions in order to get the updated 802.11n support.
The previous if_clone API changes broke this.

11 years agoConst-ify the zone name argument to uma_zcreate(9).
mdf [Fri, 26 Oct 2012 17:51:05 +0000 (17:51 +0000)]
Const-ify the zone name argument to uma_zcreate(9).

MFC after: 3 days

11 years agoMove the corresponding MTX_SYSINIT() next to their struct mtx declaration
andre [Fri, 26 Oct 2012 17:31:35 +0000 (17:31 +0000)]
Move the corresponding MTX_SYSINIT() next to their struct mtx declaration
to make their relationship more obvious as done with the other such mutexs.

11 years agoFix up some initial issues with creation and deletion of hotplugged
adrian [Fri, 26 Oct 2012 16:56:55 +0000 (16:56 +0000)]
Fix up some initial issues with creation and deletion of hotplugged
net80211 devices and vaps.

* vnet sets vnet0 during kldload and device probe/attach, but not for
  the hotplug event.  Thus, plugging in a NIC causes things to panic.
  So, add a CURVNET_SET(vnet0) for now during the attach phase, until
  the hotplug code is taught to set CURVNET_SET(vnet0).

* there's also no implied detach vnet context - so teach the detach
  path about ifp->if_vnet.

* When creating/deleting vaps, also set the vnet context appropriately.
  These can be done at any time.

Now, the problems!

* ieee80211.c is supposed to be OS-portable code, with no OS-specific stuff
  like vnet. That should be fixed.

* When the device hotplug code gets taught about CURVNET_SET(vnet0), the
  device vnet set can go away; but the VAP vnet set still needs to be there.

* .. and there still is the question about potentially adding an implied
  CURVNET_SET(ifp->if_vnet) on if_free(), since any/all devices may end up
  being detached by a hotplug event in today's world.  That's going to be
  a topic of a subsequent commit.

11 years agomdoc: sort sections.
joel [Fri, 26 Oct 2012 16:40:03 +0000 (16:40 +0000)]
mdoc: sort sections.

11 years agoFix loader crash when some unhalted exception happens during `include`
mav [Fri, 26 Oct 2012 16:32:20 +0000 (16:32 +0000)]
Fix loader crash when some unhalted exception happens during `include`
command execution.  In case of such unhandled exception, vmReset() inside
ficlExecC() flushes the VM state.  Attempt to return back to Forth after
that cause garbage dereference with unexpected results.  To avoid that
situation call vmThrow() directly instead of expecting Forth to do it.

11 years agoSince it's not immediately obvious whether the current TX path handles
adrian [Fri, 26 Oct 2012 16:31:12 +0000 (16:31 +0000)]
Since it's not immediately obvious whether the current TX path handles
fragment rate lookups correctly, add a comment describing exactly that.

The assumption in the fragment duration code is the duration of the next
fragment will match the rate used by the current fragment.  But I think
a rate lookup is being done for _each_ fragment.  For older pre-sample
rate control this would almost always be the case, but for sample
it may be incorrect more often then correct.

11 years agoAdd pcpu to the the rctl(8) manual page.
trasz [Fri, 26 Oct 2012 16:03:13 +0000 (16:03 +0000)]
Add pcpu to the the rctl(8) manual page.

11 years agoAdd CPU percentage limit enforcement to RCTL. The resouce name is "pcpu".
trasz [Fri, 26 Oct 2012 16:01:08 +0000 (16:01 +0000)]
Add CPU percentage limit enforcement to RCTL.  The resouce name is "pcpu".
It was implemented by Rudolf Tomori during Google Summer of Code 2012.

11 years agoAdd "-f" to also output filemon(4) information.
obrien [Fri, 26 Oct 2012 15:56:28 +0000 (15:56 +0000)]
Add "-f" to also output filemon(4) information.

11 years agoIterate rather than use recursion. We can blow out the kernel stack if there
obrien [Fri, 26 Oct 2012 15:44:29 +0000 (15:44 +0000)]
Iterate rather than use recursion.  We can blow out the kernel stack if there
is a long chain of fork(2)s.

11 years agoPrefer an example users born after myself might use.
eadler [Fri, 26 Oct 2012 15:21:23 +0000 (15:21 +0000)]
Prefer an example users born after myself might use.

Approved by: bcr (mentor)
MFC after: 3 days

11 years agozfs: fix label validation code in vdev_geom_read_config
avg [Fri, 26 Oct 2012 14:50:16 +0000 (14:50 +0000)]
zfs: fix label validation code in vdev_geom_read_config

POOL_STATE_SPARE and POOL_STATE_L2CACHE were not handled correctly
and thus the cache and spare disks would not be correctly probed.

Reported by: Michael Schmiedgen <schmiedgen@gmx.net>,
Matthew D. Fuller <fullermd@over-yonder.net>
Tested by: Michael Schmiedgen <schmiedgen@gmx.net>,
flo
MFC after: 5 days

11 years agoThese no longer appera to be necessary for either 32-bit or 64-bit
imp [Fri, 26 Oct 2012 14:36:02 +0000 (14:36 +0000)]
These no longer appera to be necessary for either 32-bit or 64-bit
builds, eliminate them.

11 years agoMake sure to define __BIG_ENDIAN and __LITTLE_ENDIAN with the appropriate
nwhitehorn [Fri, 26 Oct 2012 14:29:57 +0000 (14:29 +0000)]
Make sure to define __BIG_ENDIAN and __LITTLE_ENDIAN with the appropriate
number of underscores for Linux compatibility in drm(4).

Submitted by: sendtomatt at gmail dot com
MFC after: 2 weeks

11 years agoImprove libgeom XML parsing error handling.
jh [Fri, 26 Oct 2012 12:46:33 +0000 (12:46 +0000)]
Improve libgeom XML parsing error handling.

- Abort parsing and return an error if we run out of memory.
- Return EILSEQ from geom_xml2tree() for XML syntax errors.

Silence on: freebsd-geom

11 years agoMake sure the vendor class quirk only matches the vendor specific
hselasky [Fri, 26 Oct 2012 12:41:01 +0000 (12:41 +0000)]
Make sure the vendor class quirk only matches the vendor specific
class, else multiple pcm devices can appear on certain USB audio
devices.

MFC after: 1 week

11 years agoAdd support for Fast Track Ultra 8R from M-audio.
hselasky [Fri, 26 Oct 2012 11:41:57 +0000 (11:41 +0000)]
Add support for Fast Track Ultra 8R from M-audio.

MFC after: 1 week

11 years agoAdd missing CTLFLAG_TUN flag to tunable sysctls in USB stack.
hselasky [Fri, 26 Oct 2012 06:04:47 +0000 (06:04 +0000)]
Add missing CTLFLAG_TUN flag to tunable sysctls in USB stack.
Rearrange the tunables and belonging sysctl declarations, so
that they are next to eachother.

Submitted by: n_hibma @
MFC after: 1 week

11 years agoset the kernelname from the boot loader environment.
kientzle [Fri, 26 Oct 2012 05:48:53 +0000 (05:48 +0000)]
set the kernelname from the boot loader environment.
This fixes kern.bootfile sysctl.

Submitted by: Giovanni Trematerra

11 years agoComment out the BOOTP/NFSROOT support. Transition this
kientzle [Fri, 26 Oct 2012 05:41:58 +0000 (05:41 +0000)]
Comment out the BOOTP/NFSROOT support.  Transition this
config file to support a production kernel mounted on an
SD card.

11 years agoResurrect and integrate stable/9/usr.sbin/sysinstall/help/usermgmt.hlp
dteske [Fri, 26 Oct 2012 03:20:04 +0000 (03:20 +0000)]
Resurrect and integrate stable/9/usr.sbin/sysinstall/help/usermgmt.hlp

Approved by: adrian (co-mentor) (implicit)

11 years agoPort the new PV entry allocator from amd64/i386. This allocator has two
alc [Fri, 26 Oct 2012 03:02:39 +0000 (03:02 +0000)]
Port the new PV entry allocator from amd64/i386.  This allocator has two
advantages.  First, PV entries are roughly half the size.  Second, this
allocator doesn't access the paging queues, and thus it allows for the
removal of the page queues lock from this pmap.

Replace all uses of the page queues lock by a R/W lock that is private
to this pmap.

Tested by: marcel

11 years agoThis utility builds without NO_WCAST_ALIGN
eadler [Fri, 26 Oct 2012 02:50:16 +0000 (02:50 +0000)]
This utility builds without NO_WCAST_ALIGN

Tested with make universe

No objections from: stas
Approved by: cperciva
MFC after: 3 days

11 years agoRemove unnecessary quotation to clean things up.
dteske [Fri, 26 Oct 2012 02:50:14 +0000 (02:50 +0000)]
Remove unnecessary quotation to clean things up.

Approved by: adrian (co-mentor) (implicit)

11 years agoDon't mask fwe(4) lp(4) or plip(4) devices from ifconfig(8) list.
dteske [Fri, 26 Oct 2012 02:49:12 +0000 (02:49 +0000)]
Don't mask fwe(4) lp(4) or plip(4) devices from ifconfig(8) list.

Approved by: adrian (co-mentor) (implicit)

11 years agoResurrect and integrate stable/9/sysinstall/help/tcp.hlp
dteske [Fri, 26 Oct 2012 02:45:29 +0000 (02:45 +0000)]
Resurrect and integrate stable/9/sysinstall/help/tcp.hlp

Approved by: adrian (co-mentor) (implicit)

11 years agoNo MII on the RSYS4GBE (AMC-7211/ATCA-7220).
jmallett [Fri, 26 Oct 2012 02:09:55 +0000 (02:09 +0000)]
No MII on the RSYS4GBE (AMC-7211/ATCA-7220).

11 years agoTighten-up displays that use a NULL prompt (e.g., networking and usermgmt).
dteske [Fri, 26 Oct 2012 01:47:59 +0000 (01:47 +0000)]
Tighten-up displays that use a NULL prompt (e.g., networking and usermgmt).

Approved by: adrian (co-mentor) (implicit)

11 years agoFix copy/paste error.
dteske [Fri, 26 Oct 2012 00:46:09 +0000 (00:46 +0000)]
Fix copy/paste error.

Approved by: adrian (co-mentor) (implicit)

11 years agoSwitch from using a msgbox to display help files to a textbox instead. The
dteske [Fri, 26 Oct 2012 00:31:25 +0000 (00:31 +0000)]
Switch from using a msgbox to display help files to a textbox instead. The
problem with using a msgbox was one of truncation in the case of Xdialog(1)
and lack of screen real-estate (since the msgbox is not scrollable in X11
while a textbox is).

The textbox renders the text much better and is more appropriate for this
type of data display.

Approved by: adrian (co-mentor) (implicit)

11 years agoAdd support for Radisys as a vendor of Octeon hardware. Add some preliminary
jmallett [Fri, 26 Oct 2012 00:08:50 +0000 (00:08 +0000)]
Add support for Radisys as a vendor of Octeon hardware.  Add some preliminary
support for what their boot loader refers to as the "RSYS4GBE", of which there
are two instances ("Data Processing Blocks") on the Radisys ATCA-7220.

11 years agoDon't attach if the bootloader has not indicated that we're a PCI host. This
jmallett [Fri, 26 Oct 2012 00:04:05 +0000 (00:04 +0000)]
Don't attach if the bootloader has not indicated that we're a PCI host.  This
fixes booting on systems which are PCI targets.

11 years agoMerge bmake-20121010
sjg [Thu, 25 Oct 2012 23:18:05 +0000 (23:18 +0000)]
Merge bmake-20121010

Approved by: marcel (mentor)

11 years agoSecond attempt to add @every_second keyword support. Due to multiple
sobomax [Thu, 25 Oct 2012 22:54:29 +0000 (22:54 +0000)]
Second attempt to add @every_second keyword support. Due to multiple
requests, default to the previous 60-seconds scheduling method
unless there is any @every_second entries to conserve CPU cycles and
power.

This change also improves scheduling in the default mode by running
as close to the beginning of the minnute as possible by replacing
sleep(3) with nanosleep(2). Previously, the tasks would run anywhere
within the first second of the minute and that offset drifted back
and forth each time cron(8) was engaged.

MFC after: 1 month

11 years agoFix build in case we have SMBVDEBUG turned on.
davide [Thu, 25 Oct 2012 21:08:02 +0000 (21:08 +0000)]
Fix build in case we have SMBVDEBUG turned on.

Reviewed by: gnn
Approved by: gnn
Sponsored by: iXsystems inc.

11 years agoWhen Xdialog(1) is passed a NULL argument to its `--help' option, Xdialog(1)
dteske [Thu, 25 Oct 2012 21:04:11 +0000 (21:04 +0000)]
When Xdialog(1) is passed a NULL argument to its `--help' option, Xdialog(1)
acts like dialog(1) in that it returns exit status 2 when the help button is
chosen.

Approved by: adri (co-mentor) (implicit)

11 years ago- Remove the references to the deprecated zalloc kernel interface
davide [Thu, 25 Oct 2012 20:23:04 +0000 (20:23 +0000)]
- Remove the references to the deprecated zalloc kernel interface
- Use M_ZERO flag in malloc() rather than bzero()
- malloc() with M_NOWAIT can't return NULL so there's no need to check

Reviewed by: alc
Approved by: alc

11 years agoLet lpr build with -Wmissing-variable-declarations.
ed [Thu, 25 Oct 2012 20:16:38 +0000 (20:16 +0000)]
Let lpr build with -Wmissing-variable-declarations.

Mark variables static where possible and place the uid/euid variables in
lp.h, so that we can compile-time enforce that these variables have the
same type.

11 years agoModify the comment to take out the names and URL.
rmacklem [Thu, 25 Oct 2012 19:30:58 +0000 (19:30 +0000)]
Modify the comment to take out the names and URL.

Requested by: kib
MFC after: 3 days

11 years agoIncrease the driver-side limit on the number of logical volumes that
peter [Thu, 25 Oct 2012 18:46:02 +0000 (18:46 +0000)]
Increase the driver-side limit on the number of logical volumes that
the driver will recognize.  I've tested this as far as 25 volumes.

11 years agoDesupport pre-FreeBSD 7.1.
obrien [Thu, 25 Oct 2012 18:39:09 +0000 (18:39 +0000)]
Desupport pre-FreeBSD 7.1.

11 years agoInitialize the response queue mutex a bit earlier to avoid a panic that
np [Thu, 25 Oct 2012 18:11:04 +0000 (18:11 +0000)]
Initialize the response queue mutex a bit earlier to avoid a panic that
occurs if t3_sge_alloc_qset fails and then t3_free_qset attempts to
destroy an uninitialized mutex.

Submitted by: Vijay Singh <vijju dot singh at gmail>
MFC after: 3 days

11 years agoUpdate hptiop(4) to version 1.8, which added support for HighPoint
delphij [Thu, 25 Oct 2012 17:29:11 +0000 (17:29 +0000)]
Update hptiop(4) to version 1.8, which added support for HighPoint
RocketRAID 4500 series.

Many thanks to HighPoint Technologies for their continued support
of FreeBSD!

Submitted by: HighPoint Technologies
MFC after: 3 days

11 years agoFor PCI Express capability, if max link width is greater than zero, print
jimharris [Thu, 25 Oct 2012 17:22:37 +0000 (17:22 +0000)]
For PCI Express capability, if max link width is greater than zero, print
the current and max link speed.

Sponsored by: Intel
Discussed with: jhb
MFC after: 1 week

11 years agoRemove new-line characters from the include() errors to make it consistent
mav [Thu, 25 Oct 2012 15:45:32 +0000 (15:45 +0000)]
Remove new-line characters from the include() errors to make it consistent
with the rest of code.

11 years agoNote the removal of the IPFIREWALL_FORWARD kernel option.
ae [Thu, 25 Oct 2012 10:35:08 +0000 (10:35 +0000)]
Note the removal of the IPFIREWALL_FORWARD kernel option.

11 years agoPull in r166498 from upstream clang trunk:
ed [Thu, 25 Oct 2012 10:13:58 +0000 (10:13 +0000)]
Pull in r166498 from upstream clang trunk:

Add a new warning -Wmissing-variable-declarations, to warn about variables
defined without a previous declaration.  This is similar to
-Wmissing-prototypes, but for variables instead of functions.

11 years agoRemove the IPFIREWALL_FORWARD kernel option and make possible to turn
ae [Thu, 25 Oct 2012 09:39:14 +0000 (09:39 +0000)]
Remove the IPFIREWALL_FORWARD kernel option and make possible to turn
on the related functionality in the runtime via the sysctl variable
net.pfil.forward. It is turned off by default.

Sponsored by: Yandex LLC
Discussed with: net@
MFC after: 2 weeks

11 years agoCorrect SIGTTIN handling.
ed [Thu, 25 Oct 2012 09:05:21 +0000 (09:05 +0000)]
Correct SIGTTIN handling.

In the old TTY layer, SIGTTIN was correctly handled like this:

while (data should be read) {
send SIGTTIN if not foreground process group
read data
}

In the new TTY layer, however, this behaviour was changed, based on a
false interpretation of the standard:

send SIGTTIN if not foreground process group
while (data should be read) {
read data
}

Correct this by pushing tty_wait_background() into the ttydisc_read_*()
functions.

Reported by: koitsu
PR: kern/173010
MFC after: 2 weeks

11 years agoAfter r241923 the updated ip_len no longer needed.
glebius [Thu, 25 Oct 2012 09:02:21 +0000 (09:02 +0000)]
After r241923 the updated ip_len no longer needed.

11 years agoFix error in r241913 that had broken fragment reassembly.
glebius [Thu, 25 Oct 2012 09:00:57 +0000 (09:00 +0000)]
Fix error in r241913 that had broken fragment reassembly.

11 years agoFix MINCLSIZE. It should be MHLEN + 1.
kevlo [Thu, 25 Oct 2012 08:38:43 +0000 (08:38 +0000)]
Fix MINCLSIZE. It should be MHLEN + 1.

Reviewed by: glebius

11 years agoEnable "accept_rtadvd" on interfaces running rtadvd.
brian [Thu, 25 Oct 2012 08:37:08 +0000 (08:37 +0000)]
Enable "accept_rtadvd" on interfaces running rtadvd.

Without this, rtadvd runs but never advertises a default (IPv6) route.

MFC after: 1 week

11 years agoSort and remove a few long-obsolete comments.
imp [Thu, 25 Oct 2012 05:22:25 +0000 (05:22 +0000)]
Sort and remove a few long-obsolete comments.

11 years agoRemove the links: target. It too is homelessly outdated since it
imp [Thu, 25 Oct 2012 04:55:04 +0000 (04:55 +0000)]
Remove the links: target.  It too is homelessly outdated since it
relies on the outdated defines list.  It likely hasn't been useful in
15 years, and certainly not in the last decade.

11 years agoThis file is somewhat obsolete (last updated in 1995) list of #if code
imp [Thu, 25 Oct 2012 04:46:58 +0000 (04:46 +0000)]
This file is somewhat obsolete (last updated in 1995) list of #if code
in the kernel.  Retire it rather than catch up with the last 17 years.

11 years agoRestore r184194 | marcel | 2008-10-22 20:16:38 -0600 (Wed, 22 Oct 2008) | 11 lines
imp [Thu, 25 Oct 2012 04:44:17 +0000 (04:44 +0000)]
Restore r184194 | marcel | 2008-10-22 20:16:38 -0600 (Wed, 22 Oct 2008) | 11 lines
which made geom_bsd and geom_mbr optional.  They are redundant with the DEFAULTS
BSD and MBR support for g_part, which is standard.

11 years agoRemove redunant geom_bsd and geom_mbr definitions. They are in files.
imp [Thu, 25 Oct 2012 04:39:19 +0000 (04:39 +0000)]
Remove redunant geom_bsd and geom_mbr definitions.  They are in files.

11 years agoDo proper padding of runt packets using code copied from bge(4).
kientzle [Thu, 25 Oct 2012 04:37:47 +0000 (04:37 +0000)]
Do proper padding of runt packets using code copied from bge(4).

Reviewed by: gnn

11 years agoRefer to headers locally. This makes it a lot easier
kientzle [Thu, 25 Oct 2012 04:33:47 +0000 (04:33 +0000)]
Refer to headers locally.  This makes it a lot easier
to build this driver out-of-tree.

11 years agoHoist the MI compat_freebsd32 files up into files from files.*.
imp [Thu, 25 Oct 2012 04:30:48 +0000 (04:30 +0000)]
Hoist the MI compat_freebsd32 files up into files from files.*.

11 years agoList all the mips MACHINE_ARCH values in the options file to allow
imp [Thu, 25 Oct 2012 04:21:05 +0000 (04:21 +0000)]
List all the mips MACHINE_ARCH values in the options file to allow
conditional inclusion based on the MACHINE_ARCH we're compiling for.
Make ucmpdi2 conditional on mips or mipsel.

11 years agoSupport FreeBSD Boot loaders.
kientzle [Thu, 25 Oct 2012 04:14:42 +0000 (04:14 +0000)]
Support FreeBSD Boot loaders.

11 years agoEverybody needs bpf.
kientzle [Thu, 25 Oct 2012 04:12:33 +0000 (04:12 +0000)]
Everybody needs bpf.

11 years agoSupport boot loaders in the standard kernel.
kientzle [Thu, 25 Oct 2012 04:10:32 +0000 (04:10 +0000)]
Support boot loaders in the standard kernel.

11 years agoWe need ucmpdi2 for XLR and XLP when compiled in 32-bit mode. Include
imp [Thu, 25 Oct 2012 02:38:05 +0000 (02:38 +0000)]
We need ucmpdi2 for XLR and XLP when compiled in 32-bit mode.  Include
it here unconditionally while I figure out why "optional mips |
mipsel" didn't work.

11 years agoShow the number of times we block waiting for mbufs.
alfred [Thu, 25 Oct 2012 02:12:05 +0000 (02:12 +0000)]
Show the number of times we block waiting for mbufs.

Machines can stall out because mbufs are low, however sometimes we won't
see "requests denied", instead we see user land processes or kernel threads
blocking waiting for mbufs because they set M_WAIT.  These consumers do not
see errors, only stalling.

Unfortunately until now, netstat did not export this information
so you could have experienced an mbuf shortage and have no way of
seeing it unless you happen to run netstat at the exact time of the
shortage and see "in use" = "max".

By exporting the number of times processes are blocked, we can
effectively see how often non-interrupt context threads are effectively
"denied".

MFC after: 2 weeks

11 years agoAllow autotune maxusers > 384 on 64 bit machines
alfred [Thu, 25 Oct 2012 01:46:20 +0000 (01:46 +0000)]
Allow autotune maxusers > 384 on 64 bit machines

A default install on large memory machines with multiple 10gigE interfaces
were not being given enough mbufs to do full bandwidth TCP or NFS traffic.

To keep the value somewhat reasonable, we scale back the number of
maxuers by 1/6 past the 384 point.  This gives us enough mbufs for most
of our pretty basic 10gigE line-speed tests to complete.

11 years agoPad tdq_lock to avoid false sharing with tdq_load and tdq_cpu_idle.
jimharris [Wed, 24 Oct 2012 18:36:41 +0000 (18:36 +0000)]
Pad tdq_lock to avoid false sharing with tdq_load and tdq_cpu_idle.

This enables CPU searches (which read tdq_load) to operate independently
of any contention on the spinlock.  Some scheduler-intensive workloads
running on an 8C single-socket SNB Xeon show considerable improvement with
this change (2-3% perf improvement, 5-6% decrease in CPU util).

Sponsored by: Intel
Reviewed by: jeff

11 years agoFix fallout from r240071. If destination interface lookup fails,
glebius [Wed, 24 Oct 2012 18:33:44 +0000 (18:33 +0000)]
Fix fallout from r240071. If destination interface lookup fails,
we should broadcast a packet, not try to deliver it to NULL.

Reported by: rpaulo

11 years agoCommit the actual text provided by Alan, instead of the wrong update
kib [Wed, 24 Oct 2012 18:32:37 +0000 (18:32 +0000)]
Commit the actual text provided by Alan, instead of the wrong update
in r242011.

MFC after: 1 week

11 years agoDirty the newly copied anonymous pages after the wired region is
kib [Wed, 24 Oct 2012 18:21:59 +0000 (18:21 +0000)]
Dirty the newly copied anonymous pages after the wired region is
forked. Otherwise, pagedaemon might reclaim the page without saving
its content into the swap file, resulting in the valid content
replaced by zeroes.

Reported and tested by: pho
Reviewed and comment update by: alc
MFC after: 1 week

11 years agoAdd missed sched_pin().
kib [Wed, 24 Oct 2012 18:21:22 +0000 (18:21 +0000)]
Add missed sched_pin().

Submitted by: Svatopluk Kraus <onwahe@gmail.com>
Reviewed by: alc
MFC after: 3 days