ƒ·H237846
151
143
365
136
1126
153
153
273
283
283
152
283
292
216
444
394
315
315
315
228
112
149
164
247
110
311
143
416
3395
467
118
5774
5774
229
294
294
365
320
295
255
182
172
133
205
243
148
156
196
187
131
167
158
155
209
146
136
195
167
130
160
131
141
172
123
276
343
180
144
153
151
262
714
820
144
464
374
287
183
114
324
1031
142
181
147
304
1008
142
976
710
645
262
147
115
192
408
170
313
163
283
383
471
471
379
378
328
553
547
236
347
167
105
113
125
626
132
132
148
640
640
640
220
226
165
171
150
113
167
262
305
147
301
186
166
166
196
145
707
113
141
330
152
127
438
257
136
230
159
187
176
128
224
174

K 10
svn:author
V 3
kib
K 8
svn:date
V 27
2012-06-30T16:23:08.393753Z
K 7
svn:log
V 51
killall(1) does not use libkvm.

MFC after:	3 days

END
K 10
svn:author
V 3
kib
K 8
svn:date
V 27
2012-06-30T16:36:22.798021Z
K 7
svn:log
V 272
Once in a month, when the moon is full, killall mistakenly considers
living process as a zombie and refuses to kill it. The cause is that
the code masks ki_stat with SZOMB to compare with SZOMB, but ki_stat
is not a mask.

Possibly reported by:	cperciva
MFC after:	3 days

END
K 10
svn:author
V 3
kib
K 8
svn:date
V 27
2012-06-30T17:01:28.640160Z
K 7
svn:log
V 44
Remove stray blank line.

MFC after:	3 days

END
K 10
svn:author
V 6
marcel
K 8
svn:date
V 27
2012-06-30T18:56:08.587781Z
K 7
svn:log
V 1029
Create zones for busdma_tag, busdma_md and busdma_md_seg structures.
In particular, this is driven by not wanting to create a single blob
comprising the busdma_md and all its busdma_md_seg structures:
1.  the number of segments is an upper bound that is very rarely
    reached. Pre-allocating them means we're wasting a lot of memory.
2.  The allocation size for these structures combined can vary per
    tag, which isn't always good.
3.  During load operations we may need to keep track of more than the
    maximum number of segments. In particular this can happen before
    we've gone through I/O MMU mappings and/or bounce buffers.

Now we can create segment structures when needed for busdma allocs
and we can reserve some amount of segment structures and keep them
in a pool when an unused busdma_md structure is allocated for use
by the load functions later.

For now, put the segment structures in a TAILQ and put the index in
the structure. This probably changes over time when the whole logic
of loading solidifies.

END
K 10
svn:author
V 7
thompsa
K 8
svn:date
V 27
2012-06-30T18:57:55.201170Z
K 7
svn:log
V 57
MFC r237674

 Update the usage with the new jail option.

END
K 10
svn:author
V 7
thompsa
K 8
svn:date
V 27
2012-06-30T18:58:21.159506Z
K 7
svn:log
V 57
MFC r237674

 Update the usage with the new jail option.

END
K 10
svn:author
V 7
thompsa
K 8
svn:date
V 27
2012-06-30T19:09:02.106824Z
K 7
svn:log
V 176
Add the same check as vlan(4) where we ignore the ifnet departure event if the
interface is just being renamed.

PR:		kern/169557
Submitted by:	Mark Johnston
MFC after:	3 days

END
K 10
svn:author
V 5
gavin
K 8
svn:date
V 27
2012-06-30T20:22:20.364685Z
K 7
svn:log
V 188
Merge r237645 from head:

  The -S option, to specify the sector size, has been usable on all types
  of memory disks since r135340.  Update the man page to reflect this.

Noticed by:	avg

END
K 10
svn:author
V 5
gavin
K 8
svn:date
V 27
2012-06-30T20:23:59.706236Z
K 7
svn:log
V 188
Merge r237645 from head:

  The -S option, to specify the sector size, has been usable on all types
  of memory disks since r135340.  Update the man page to reflect this.

Noticed by:	avg

END
K 10
svn:author
V 3
alc
K 8
svn:date
V 27
2012-06-30T20:25:12.771176Z
K 7
svn:log
V 60
Optimize reserve_pv_entries() using the popcnt instruction.

END
K 10
svn:author
V 5
gavin
K 8
svn:date
V 27
2012-06-30T20:25:50.524658Z
K 7
svn:log
V 188
Merge r237645 from head:

  The -S option, to specify the sector size, has been usable on all types
  of memory disks since r135340.  Update the man page to reflect this.

Noticed by:	avg

END
K 10
svn:author
V 6
davide
K 8
svn:date
V 27
2012-06-30T22:18:43.112072Z
K 7
svn:log
V 196
There's no need of extra indirection for cc_localexp, so we can replace
struct callout_tailq* with struct callout_tailq. Rename cc_localexp to a
more appropriate cc_expireq.

Discussed with:		mav

END
K 10
svn:author
V 6
marcel
K 8
svn:date
V 27
2012-06-30T23:29:34.740829Z
K 7
svn:log
V 120
Register a timecounter based on the SHub's RTC. As it happens the
mandatory ACPI PM Timer isn't there on the SGI Altix.

END
K 10
svn:author
V 6
davide
K 8
svn:date
V 27
2012-06-30T23:34:33.753644Z
K 7
svn:log
V 348
Add a new CALLOUT_PROFILING option so that SYSCTLs on the wheel may be
selectively disabled/enabled. Reintroduce a SYSCTL that I wrongly removed
in a previous commit. Selectively disabling this sort of rudimentary
profiling may have a good effect on CPU caches because same variable is not
accessed anymore by different CPUs.

Discussed with:		mav

END
K 10
svn:author
V 3
pfg
K 8
svn:date
V 27
2012-07-01T00:36:07.688178Z
K 7
svn:log
V 301
Import the dtrace llquantize test files from the vendor area.

Now that we have an illumos vendor area, import the llquantize
test files directly from there to make it easier to track the
code provenance.

Special thanks to Martin Matuska for populating the illumos
vendor area.

Requested by:	obrien

END
K 10
svn:author
V 6
eadler
K 8
svn:date
V 27
2012-07-01T00:40:09.376320Z
K 7
svn:log
V 219
MFC r237178:
	attach_generic causes missing devices in /dev when the driver interacts with some non-highpoint controollers. Change attach_generic to be off by default.

PR:		kern/168910
Approved by:	cperciva (implicit)

END
K 10
svn:author
V 6
eadler
K 8
svn:date
V 27
2012-07-01T00:40:37.851512Z
K 7
svn:log
V 219
MFC r237178:
	attach_generic causes missing devices in /dev when the driver interacts with some non-highpoint controollers. Change attach_generic to be off by default.

PR:		kern/168910
Approved by:	cperciva (implicit)

END
K 10
svn:author
V 6
eadler
K 8
svn:date
V 27
2012-07-01T00:40:56.272095Z
K 7
svn:log
V 219
MFC r237178:
	attach_generic causes missing devices in /dev when the driver interacts with some non-highpoint controollers. Change attach_generic to be off by default.

PR:		kern/168910
Approved by:	cperciva (implicit)

END
K 10
svn:author
V 6
adrian
K 8
svn:date
V 27
2012-07-01T02:34:32.306595Z
K 7
svn:log
V 132
Fix the HAL debugging to only use one bit to mark a message as unmaskable.

Whilst I'm here, remove the duplication of the #define.

END
K 10
svn:author
V 6
adrian
K 8
svn:date
V 27
2012-07-01T02:37:04.881233Z
K 7
svn:log
V 17
Add OS_MEMCMP().

END
K 10
svn:author
V 6
adrian
K 8
svn:date
V 27
2012-07-01T02:44:36.814601Z
K 7
svn:log
V 54
Bring over capabilities for the AR9300 and later HAL.

END
K 10
svn:author
V 6
andrew
K 8
svn:date
V 27
2012-07-01T02:57:30.618441Z
K 7
svn:log
V 69
Revert the wchar changes to simplify pulling in the changes in HEAD.

END
K 10
svn:author
V 6
adrian
K 8
svn:date
V 27
2012-07-01T03:15:18.262342Z
K 7
svn:log
V 151
Migrate the MAC/BB hang structures out from ar5416_misc.h into the HAL.

The ar9300 HAL also uses these types, so it makes no sense to duplicate
them.

END
K 10
svn:author
V 6
andrew
K 8
svn:date
V 27
2012-07-01T04:09:42.056987Z
K 7
svn:log
V 15
MFC to r237866

END
K 10
svn:author
V 3
pfg
K 8
svn:date
V 27
2012-07-01T04:15:14.288269Z
K 7
svn:log
V 218
MFC	r237624, r237714, r237716, r237860:

Bring llquantize support into Dtrace.

Bryan Cantrill implemented the equivalent of semi-log graph
paper for Dtrace so llquantize will use one logarithmic and
one linear scale.

END
K 10
svn:author
V 6
adrian
K 8
svn:date
V 27
2012-07-01T04:25:49.674204Z
K 7
svn:log
V 48
Add a check for dynamic OFDM/CCK channel types.

END
K 10
svn:author
V 3
pfg
K 8
svn:date
V 27
2012-07-01T04:26:51.704055Z
K 7
svn:log
V 323
MFC	r237624, r237714, r237716, r237860:

Bring llquantize support into Dtrace.

Bryan Cantrill implemented the equivalent of semi-log graph
paper for Dtrace so llquantize will use one logarithmic and
one linear scale.

Illumos Revision:	13355:15b74a2a9a9d

Reference:
https://www.illumos/issues/905

Obtained from:	Illumos

END
K 10
svn:author
V 3
ken
K 8
svn:date
V 27
2012-07-01T05:13:50.864462Z
K 7
svn:log
V 3301
MFC 237518, 237545, 237648:

  r237518 | ken | 2012-06-23 22:29:03 -0600 (Sat, 23 Jun 2012) | 72 lines

  Fix a bug which causes a panic in daopen(). The panic is caused by
  a da(4) instance going away while GEOM is still probing it.

  In this case, the GEOM disk class instance has been created by
  disk_create(), and the taste of the disk is queued in the GEOM
  event queue.

  While that event is queued, the da(4) instance goes away.  When the
  open call comes into the da(4) driver, it dereferences the freed
  (but non-NULL) peripheral pointer provided by GEOM, which results
  in a panic.

  The solution is to add a callback to the GEOM disk code that is
  called when all of its resources are cleaned up.  This is
  implemented inside GEOM by adding an optional callback that is
  called when all consumers have detached from a provider, and the
  provider is about to be deleted.

  scsi_cd.c,
  scsi_da.c:	In the register routine for the cd(4) and da(4)
  		routines, acquire a reference to the CAM peripheral
  		instance just before we call disk_create().

  		Use the new GEOM disk d_gone() callback to register
  		a callback (dadiskgonecb()/cddiskgonecb()) that
  		decrements the peripheral reference count once GEOM
  		has finished cleaning up its resources.

  		In the cd(4) driver, clean up open and close
  		behavior slightly.  GEOM makes sure we only get one
  		open() and one close call, so there is no need to
  		set an open flag and decrement the reference count
  		if we are not the first open.

  		In the cd(4) driver, use cam_periph_release_locked()
  		in a couple of error scenarios to avoid extra mutex
  		calls.

  geom.h:	Add a new, optional, providergone callback that
  		is called when a provider is about to be deleted.

  geom_disk.h:	Add a new d_gone() callback to the GEOM disk
  		interface.

  		Bump the DISK_VERSION to version 2.  This probably
  		should have been done after a couple of previous
  		changes, especially the addition of the d_getattr()
  		callback.

  geom_disk.c:	Add a providergone callback for the disk class,
  		g_disk_providergone(), that calls the user's
  		d_gone() callback if it exists.

  		Bump the DISK_VERSION to 2.

  geom_subr.c:	In g_destroy_provider(), call the providergone
  		callback if it has been provided.

  		In g_new_geomf(), propagate the class's
  		providergone callback to the new geom instance.

  blkfront.c:	Callers of disk_create() are supposed to pass in
  		DISK_VERSION, not an explicit disk API version
  		number.  Update the blkfront driver to do that.

  disk.9:	Update the disk(9) man page to include information
  		on the new d_gone() callback, as well as the
  		previously added d_getattr() callback, d_descr
  		field, and HBA PCI ID fields.

  r237545 | ken | 2012-06-24 22:26:10 -0600 (Sun, 24 Jun 2012) | 7 lines

  Consume spare fields for the providergone pointers added to the g_class and
  g_geom structures in change 237518.  The original change would have broken
  the ABI.

  Suggested by:	ae

  r237648 | ken | 2012-06-27 10:05:09 -0600 (Wed, 27 Jun 2012) | 6 lines

  In g_disk_providergone(), don't continue if the softc is NULL.  This may be
  the case if we've already gone through g_disk_destroy().

  Reported by:	Michael Butler <imb@protected-networks.net>

END
K 10
svn:author
V 6
adrian
K 8
svn:date
V 27
2012-07-01T05:14:24.858788Z
K 7
svn:log
V 371
Add in some further changes from the AR9300 HAL:

* Add a new ANI variable, for AR9003 and later chips;
* The AR9003 and later series chips support two RX queues now, so start
  down the road of supporting that;
* Add some new TX queue types - uAPSD is possible on earlier chips,
  but PAPRD is relevant to AR9003 and later.

Obtained from:	Qualcomm Atheros, Linux ath9k

END
K 10
svn:author
V 3
imp
K 8
svn:date
V 27
2012-07-01T05:22:13.204820Z
K 7
svn:log
V 26
Use %j to match intmax_t.

END
K 10
svn:author
V 3
ken
K 8
svn:date
V 27
2012-07-01T05:22:45.181201Z
K 7
svn:log
V 5680
MFC 237683:

  r237683 | ken | 2012-06-27 21:48:54 -0600 (Wed, 27 Jun 2012) | 129 lines

  Bring in LSI's latest mps(4) 6Gb SAS and WarpDrive driver, version
  14.00.00.01-fbsd.

  Their description of the changes is as follows:

  1.	Copyright contents has been changed in all respective .c
  	and .h files

  2.	Support for WRITE12 and READ12 for direct-io (warpdrive only)
  	has been added.

  3.      Driver has added checks to see if Drive has READ_CAP_16
  	support before sending it down to the device.
  	If SPC3_SID_PROTECT flag is set in the inquiry data, the
  	device supports protection information, and must support
  	the 16 byte read capacity command, otherwise continue without
  	sending read cap 16. This will optimize driver performance,
  	since it will not send READ_CAP_16 to the drive which does
  	not have support of READ_CAP_16.

  4.      With new approach, "MPTIOCTL_RESET_ADAPTER" IOCTL will not
  	use DELAY() which is busy loop implementation.
  	It will use <msleep> (Better way to sleep without busy
  	loop). Also from the HBA reset code path and some other
  	places, DELAY() is replaced with msleep() or "pause()",
  	which is based on sleep/wakeup style calls.  Driver use
  	msleep()/pause() instead of DELAY based on CAN_SLEEP/NO_SLEEP
  	flags to avoid busy loop which is not required all the
  	time.e.a

  	a. While driver is getting loaded, driver calls most of the
  	   commands with NO_SLEEP.
  	b. When Driver is functional and it needs Reinit of HBA,
  	   CAN_SLEEP flag is used.

  5.	<mpslsi> driver is not Endian safe. It will not work on Big
  	Endian machines	like Sparc and PowerPC platforms because it
  	assumes it is running on a Little Endian machine.

  	Driver code is modified such way that it does not assume CPU
  	arch is Little Endian.
  	a. All places where Driver interacts from HBA to Host, it
  	   converts Little Endian format to CPU format.
  	b. All places where Driver interacts from Host to HBA, it
  	   converts CPU format to Little Endian.

  6.	Findout memory leaks in FreeBSD Driver and resolve those,
  	such as memory leak in targ's luns creation/deletion.
  	Also added additional checks to see memory allocation
  	success/fail.

  7.	Add loginfo prints as debug message, i.e. When FW sends any
  	loginfo, Driver should print those as debug message.
  	This will help for debugging purpose.

  8.	There is possibility to get config request timeout. Current
  	driver is able to detect config request timetout, but it does
  	not do anything on config_request timeout.  Driver should
  	call mps_reinit() if any request_poll (which is called as
  	part of config_request) is time out.

  9.	cdb length check is required for 32 byte CDB. Add correct mpi
  	control value for 32 bit CDB as below while submitting SCSI IO
  	Request to controller.
  	mpi_control |= 4 << MPI2_SCSIIO_CONTROL_ADDCDBLEN_SHIFT;

  10.	Check the actual status of Message unit reset
  	(mps_message_unit_reset).Previously FreeBSD Driver just writes
  	MPI2_FUNCTION_IOC_MESSAGE_UNIT_RESET and never check the ack
  	(it just wait for 50 millisecond).  So, Driver now check the
  	status of "MPI2_FUNCTION_IOC_MESSAGE_UNIT_RESET" after writing
  	it to the FW.

  	Now it also checking for whether doorbell ack uses msleep with
  	proper sleep flags, instead of <DELAY>.

  11.	Previously CAM does not detect Multi-Lun Devices. In order to
  	detect Multi-Lun Devices by CAM the driver needs following change
  	set:
  	a. There is "max_lun" field which Driver need to set based on
  	   hw/fw support. Currently LSI released driver does not set
  	   this field.
  	b. Default of "max_lun" should not be 0 in OS, but it is
  	   currently set to 0 in CAM layer.
  	c. Export max_lun capacity to 255

  12.	Driver will not reset target info after port enable complete and
  	also do Device removal when Device remove from FW.  The detail
  	description is as follows
  	a. When Driver receive WD PD add events, it will add all
  	   information in driver local data structure.
  	b. Only for WD, we have below checks after port enable
  	   completes, where driver clear off all information retrieved
  	   at #1.
  	if ((sc->WD_available &&
               (sc->WD_hide_expose == MPS_WD_HIDE_ALWAYS)) ||
               (sc->WD_valid_config && (sc->WD_hide_expose ==
                              MPS_WD_HIDE_IF_VOLUME)) {
  		  // clear off target data structure.
  	}
  	It is mainly not to attach PDs to OS.

  	FreeBSD does bus rescan as older Parallel scsi style. So Driver
  	needs to handle which Drive is visible to OS.  That is a reason
  	we have to clear off targ information for PDs.

  	Again, above logic was implemented long time ago. Similar concept
  	we have for non-wd also. For that, LSI have introduced different
  	logic to hide PDs.

  	Eventually, because of above gap, when Phy goes offline, we
  	observe below failure. That is what Driver is not doing complete
  	removal of device with FW. (which was pointed by Scott)
  	Apr  5 02:39:24 Freebsd7 kernel: mpslsi0: mpssas_prepare_remove
  	Apr  5 02:39:24 Freebsd7 kernel: mpssas_prepare_remove 497 : invalid handle 0xe

  	Now Driver will not reset target info after port enable complete
  	and also will do Device removal when Device remove from FW.

  13.	Returning "CAM_SEL_TIMEOUT" instead of "CAM_TID_INVALID"
  	error code on request to the Target IDs that have no devices
  	conected at that moment.  As if "CAM_TID_INVALID" error code
  	is returned to the CAM Layaer then it results in a huge chain
  	of errors in verbose kernel messages on boot and every
  	hot-plug event.

Submitted by:	Sreekanth Reddy <Sreekanth.Reddy@lsi.com>

END
K 10
svn:author
V 3
ken
K 8
svn:date
V 27
2012-07-01T05:23:59.533672Z
K 7
svn:log
V 5680
MFC 237683:

  r237683 | ken | 2012-06-27 21:48:54 -0600 (Wed, 27 Jun 2012) | 129 lines

  Bring in LSI's latest mps(4) 6Gb SAS and WarpDrive driver, version
  14.00.00.01-fbsd.

  Their description of the changes is as follows:

  1.	Copyright contents has been changed in all respective .c
  	and .h files

  2.	Support for WRITE12 and READ12 for direct-io (warpdrive only)
  	has been added.

  3.      Driver has added checks to see if Drive has READ_CAP_16
  	support before sending it down to the device.
  	If SPC3_SID_PROTECT flag is set in the inquiry data, the
  	device supports protection information, and must support
  	the 16 byte read capacity command, otherwise continue without
  	sending read cap 16. This will optimize driver performance,
  	since it will not send READ_CAP_16 to the drive which does
  	not have support of READ_CAP_16.

  4.      With new approach, "MPTIOCTL_RESET_ADAPTER" IOCTL will not
  	use DELAY() which is busy loop implementation.
  	It will use <msleep> (Better way to sleep without busy
  	loop). Also from the HBA reset code path and some other
  	places, DELAY() is replaced with msleep() or "pause()",
  	which is based on sleep/wakeup style calls.  Driver use
  	msleep()/pause() instead of DELAY based on CAN_SLEEP/NO_SLEEP
  	flags to avoid busy loop which is not required all the
  	time.e.a

  	a. While driver is getting loaded, driver calls most of the
  	   commands with NO_SLEEP.
  	b. When Driver is functional and it needs Reinit of HBA,
  	   CAN_SLEEP flag is used.

  5.	<mpslsi> driver is not Endian safe. It will not work on Big
  	Endian machines	like Sparc and PowerPC platforms because it
  	assumes it is running on a Little Endian machine.

  	Driver code is modified such way that it does not assume CPU
  	arch is Little Endian.
  	a. All places where Driver interacts from HBA to Host, it
  	   converts Little Endian format to CPU format.
  	b. All places where Driver interacts from Host to HBA, it
  	   converts CPU format to Little Endian.

  6.	Findout memory leaks in FreeBSD Driver and resolve those,
  	such as memory leak in targ's luns creation/deletion.
  	Also added additional checks to see memory allocation
  	success/fail.

  7.	Add loginfo prints as debug message, i.e. When FW sends any
  	loginfo, Driver should print those as debug message.
  	This will help for debugging purpose.

  8.	There is possibility to get config request timeout. Current
  	driver is able to detect config request timetout, but it does
  	not do anything on config_request timeout.  Driver should
  	call mps_reinit() if any request_poll (which is called as
  	part of config_request) is time out.

  9.	cdb length check is required for 32 byte CDB. Add correct mpi
  	control value for 32 bit CDB as below while submitting SCSI IO
  	Request to controller.
  	mpi_control |= 4 << MPI2_SCSIIO_CONTROL_ADDCDBLEN_SHIFT;

  10.	Check the actual status of Message unit reset
  	(mps_message_unit_reset).Previously FreeBSD Driver just writes
  	MPI2_FUNCTION_IOC_MESSAGE_UNIT_RESET and never check the ack
  	(it just wait for 50 millisecond).  So, Driver now check the
  	status of "MPI2_FUNCTION_IOC_MESSAGE_UNIT_RESET" after writing
  	it to the FW.

  	Now it also checking for whether doorbell ack uses msleep with
  	proper sleep flags, instead of <DELAY>.

  11.	Previously CAM does not detect Multi-Lun Devices. In order to
  	detect Multi-Lun Devices by CAM the driver needs following change
  	set:
  	a. There is "max_lun" field which Driver need to set based on
  	   hw/fw support. Currently LSI released driver does not set
  	   this field.
  	b. Default of "max_lun" should not be 0 in OS, but it is
  	   currently set to 0 in CAM layer.
  	c. Export max_lun capacity to 255

  12.	Driver will not reset target info after port enable complete and
  	also do Device removal when Device remove from FW.  The detail
  	description is as follows
  	a. When Driver receive WD PD add events, it will add all
  	   information in driver local data structure.
  	b. Only for WD, we have below checks after port enable
  	   completes, where driver clear off all information retrieved
  	   at #1.
  	if ((sc->WD_available &&
               (sc->WD_hide_expose == MPS_WD_HIDE_ALWAYS)) ||
               (sc->WD_valid_config && (sc->WD_hide_expose ==
                              MPS_WD_HIDE_IF_VOLUME)) {
  		  // clear off target data structure.
  	}
  	It is mainly not to attach PDs to OS.

  	FreeBSD does bus rescan as older Parallel scsi style. So Driver
  	needs to handle which Drive is visible to OS.  That is a reason
  	we have to clear off targ information for PDs.

  	Again, above logic was implemented long time ago. Similar concept
  	we have for non-wd also. For that, LSI have introduced different
  	logic to hide PDs.

  	Eventually, because of above gap, when Phy goes offline, we
  	observe below failure. That is what Driver is not doing complete
  	removal of device with FW. (which was pointed by Scott)
  	Apr  5 02:39:24 Freebsd7 kernel: mpslsi0: mpssas_prepare_remove
  	Apr  5 02:39:24 Freebsd7 kernel: mpssas_prepare_remove 497 : invalid handle 0xe

  	Now Driver will not reset target info after port enable complete
  	and also will do Device removal when Device remove from FW.

  13.	Returning "CAM_SEL_TIMEOUT" instead of "CAM_TID_INVALID"
  	error code on request to the Target IDs that have no devices
  	conected at that moment.  As if "CAM_TID_INVALID" error code
  	is returned to the CAM Layaer then it results in a huge chain
  	of errors in verbose kernel messages on boot and every
  	hot-plug event.

Submitted by:	Sreekanth Reddy <Sreekanth.Reddy@lsi.com>

END
K 10
svn:author
V 3
ken
K 8
svn:date
V 27
2012-07-01T05:32:26.025218Z
K 7
svn:log
V 136
Now that the mps(4) driver is endian-safe, add it to the powerpc 32-bit
GENERIC config file.

MFC after:	3 days
Reqested by:	nwhitehorn

END
K 10
svn:author
V 3
ken
K 8
svn:date
V 27
2012-07-01T05:39:49.485579Z
K 7
svn:log
V 201
MFC 237682:

  r237682 | ken | 2012-06-27 21:36:13 -0600 (Wed, 27 Jun 2012) | 6 lines

  Fix a typo in a panic() call.

PR:		kern/169497
Submitted by:	Steven Hartland <steven.hartland@multiplay.co.uk>

END
K 10
svn:author
V 3
ken
K 8
svn:date
V 27
2012-07-01T05:40:33.375806Z
K 7
svn:log
V 201
MFC 237682:

  r237682 | ken | 2012-06-27 21:36:13 -0600 (Wed, 27 Jun 2012) | 6 lines

  Fix a typo in a panic() call.

PR:		kern/169497
Submitted by:	Steven Hartland <steven.hartland@multiplay.co.uk>

END
K 10
svn:author
V 3
imp
K 8
svn:date
V 27
2012-07-01T06:34:17.294615Z
K 7
svn:log
V 272
Opt-in rather than opt-out of the SoC.  We don't really support
running with multiple SoCs compiled in very well anyway, so this just
wastes space.  As more and more SoCs arrive in the tree, it is better
to edit one master file that builds them all than many board files.

END
K 10
svn:author
V 3
mav
K 8
svn:date
V 27
2012-07-01T06:41:39.375485Z
K 7
svn:log
V 227
MFC r237718:
Fix millisecond to ticks conversion in drm_msleep().

On systems with HZ=100 it caused Intel eDP video output initialization
(and Xorg startup) to take several minutes instead of several seconds.

Reviewed by:	kib

END
K 10
svn:author
V 3
imp
K 8
svn:date
V 27
2012-07-01T06:56:41.637673Z
K 7
svn:log
V 202
Create a pseudo-lint kernel for all at91 SoCs.  This kernel will not
currently boot, but will serve as a good linting.  make universe could
now be altered to skip building all the other at91 kernels...

END
K 10
svn:author
V 6
tuexen
K 8
svn:date
V 27
2012-07-01T07:28:15.030670Z
K 7
svn:log
V 159
MFC r235360:
Provide in the association change notification the received ABORT chunk
if case of SCTP_COMM_LOST or SCTP_CANT_STR_ASSOC as required by RFC 6458.

END
K 10
svn:author
V 6
tuexen
K 8
svn:date
V 27
2012-07-01T07:31:25.407580Z
K 7
svn:log
V 87
MFC r235402:
Ensure the user can read COMM_LOST notifications on 1-to-1 style sockets.

END
K 10
svn:author
V 6
tuexen
K 8
svn:date
V 27
2012-07-01T07:34:07.398913Z
K 7
svn:log
V 77
MFC r235403:
Use ECONNABORTED in cases where the ABORT was sent to the peer.

END
K 10
svn:author
V 6
tuexen
K 8
svn:date
V 27
2012-07-01T07:36:33.713841Z
K 7
svn:log
V 38
MFC r235412:
Remove unused constants.

END
K 10
svn:author
V 6
tuexen
K 8
svn:date
V 27
2012-07-01T07:39:37.310986Z
K 7
svn:log
V 109
MFC r235414,r235415:
Provide the error code in SCTP_PEER_ADDR_CHANGE notifications as
specified in RFC 6458.

END
K 10
svn:author
V 6
tuexen
K 8
svn:date
V 27
2012-07-01T07:42:59.253620Z
K 7
svn:log
V 147
MFC r235416:
Provide in the SCTP_SEND_FAILED and SCTP_SEND_FAILED_EVENT notifications
the correct ssf_error or ssfe_error as required by RFC 6458.

END
K 10
svn:author
V 6
tuexen
K 8
svn:date
V 27
2012-07-01T07:49:14.150250Z
K 7
svn:log
V 53
MFC r235418:
Support SCTP_REMOTE_ERROR notification.

END
K 10
svn:author
V 6
tuexen
K 8
svn:date
V 27
2012-07-01T07:51:32.246202Z
K 7
svn:log
V 61
MFC r235554:
Support SCTP_EOF also for 1-to-1 style sockets.

END
K 10
svn:author
V 2
mm
K 8
svn:date
V 27
2012-07-01T07:53:50.542235Z
K 7
svn:log
V 104
Add illumos manual pages to vendor/illumos (original location)
illumos-gate revision 13742:b6bbdd77139c

END
K 10
svn:author
V 6
tuexen
K 8
svn:date
V 27
2012-07-01T07:53:54.444843Z
K 7
svn:log
V 92
MFC r235557:
Use a default for max_burst of 4 and l2var of 2.
This was discussed with rrs@.

END
K 10
svn:author
V 6
tuexen
K 8
svn:date
V 27
2012-07-01T07:55:42.157958Z
K 7
svn:log
V 36
MFC r235827:
Update copyright date.

END
K 10
svn:author
V 2
mm
K 8
svn:date
V 27
2012-07-01T07:57:18.160481Z
K 7
svn:log
V 76
Add userland ZFS to vendor/illumos
illumos-gate revision 13742:b6bbdd77139c

END
K 10
svn:author
V 6
tuexen
K 8
svn:date
V 27
2012-07-01T07:59:00.894037Z
K 7
svn:log
V 63
MFC r235828:
Use consistent text at the begining of the files.

END
K 10
svn:author
V 6
tuexen
K 8
svn:date
V 27
2012-07-01T08:01:42.875121Z
K 7
svn:log
V 60
MFC r235903:
Add sn_send_failed_event to sctp_notification.

END
K 10
svn:author
V 6
tuexen
K 8
svn:date
V 27
2012-07-01T08:04:10.310560Z
K 7
svn:log
V 113
MFC r235990:
Undefine SCTP_PACKED before including sctp_uio.h, which doesn't
use it. Spotted by Irene Ruengeler.

END
K 10
svn:author
V 6
tuexen
K 8
svn:date
V 27
2012-07-01T08:06:43.385058Z
K 7
svn:log
V 51
MFC r236391:
Small cleanups. No functional change.

END
K 10
svn:author
V 6
tuexen
K 8
svn:date
V 27
2012-07-01T08:09:05.754199Z
K 7
svn:log
V 41
MFC r236450:
Remove an unused parameter.

END
K 10
svn:author
V 7
delphij
K 8
svn:date
V 27
2012-07-01T08:10:49.018044Z
K 7
svn:log
V 99
tws(4) is interfaced with CAM so move it to the same section.

Reported by:	joel
MFC after:	3 days

END
K 10
svn:author
V 6
tuexen
K 8
svn:date
V 27
2012-07-01T08:11:15.777503Z
K 7
svn:log
V 72
MFC r236492:
Don't request data from the IPv6 layer, which is not used.

END
K 10
svn:author
V 6
tuexen
K 8
svn:date
V 27
2012-07-01T08:13:20.948624Z
K 7
svn:log
V 35
MFC r236493:
Honor sysctl for TTL.

END
K 10
svn:author
V 6
tuexen
K 8
svn:date
V 27
2012-07-01T08:18:40.804854Z
K 7
svn:log
V 65
MFC r236515:
Use an existing function to get the source address.

END
K 10
svn:author
V 2
mm
K 8
svn:date
V 27
2012-07-01T08:20:13.151699Z
K 7
svn:log
V 40
Add zfs manual pages missing in r237895

END
K 10
svn:author
V 6
tuexen
K 8
svn:date
V 27
2012-07-01T08:21:08.581394Z
K 7
svn:log
V 46
MFC r236522:
Remove code which is not needed.

END
K 10
svn:author
V 6
tuexen
K 8
svn:date
V 27
2012-07-01T08:23:16.504732Z
K 7
svn:log
V 77
MFC r236640:
The cmsg_len field includes the cmsg header. So use CMSG_LEN().

END
K 10
svn:author
V 6
tuexen
K 8
svn:date
V 27
2012-07-01T08:27:02.798368Z
K 7
svn:log
V 28
MFC r236949:
Small cleanup.

END
K 10
svn:author
V 6
tuexen
K 8
svn:date
V 27
2012-07-01T08:32:48.962319Z
K 7
svn:log
V 180
MFC r236958:
Deliver IPV6_TCLASS, IPV6_HOPLIMIT and IPV6_PKTINFO cmsgs (if
requested) on IPV6 sockets, which have been marked to be not IPV6_V6ONLY,
for each received IPV4 packet.

END
K 10
svn:author
V 6
tuexen
K 8
svn:date
V 27
2012-07-01T08:47:15.841176Z
K 7
svn:log
V 247
MFC r236959:
Add a IP_RECVTOS socket option to receive for received UDP/IPv4
packets a cmsg of type IP_RECVTOS which contains the TOS byte.
Much like IP_RECVTTL does for TTL. This allows to implement a
protocol on top of UDP and implementing ECN.

END
K 10
svn:author
V 6
tuexen
K 8
svn:date
V 27
2012-07-01T08:53:55.546454Z
K 7
svn:log
V 85
MFC r236961:
Add a cmsg of type IP_TOS for UDP/IPv4 sockets to specify the TOS byte.

END
K 10
svn:author
V 6
tuexen
K 8
svn:date
V 27
2012-07-01T08:57:28.533385Z
K 7
svn:log
V 49
MFC r237229:
Cleanup the UDP decapsulation code.

END
K 10
svn:author
V 6
tuexen
K 8
svn:date
V 27
2012-07-01T09:00:29.853745Z
K 7
svn:log
V 58
MFC r237230:
Add rate limitation for SCTP OOTB responses.

END
K 10
svn:author
V 6
tuexen
K 8
svn:date
V 27
2012-07-01T09:03:37.788848Z
K 7
svn:log
V 56
MFC r237392:
Remove redundant #ifdef. Reported by gnn@.

END
K 10
svn:author
V 6
davide
K 8
svn:date
V 27
2012-07-01T09:17:17.733607Z
K 7
svn:log
V 166
The variable 'limit' appears unused right now. Macroize the 0x2 value into
C_PRECISION to make code a bit more readable. Fix random style bugs.

Discussed with:		mav

END
K 10
svn:author
V 2
np
K 8
svn:date
V 27
2012-07-01T09:17:55.206888Z
K 7
svn:log
V 622
MFC r231317, r235963 (bz@), r234831, r234833.

r231317
Add IPv6 TSO (including TSO+VLAN) support to cxgb(4).

r235963 (bz@)
Allow LRO to work on IPv6 as well.
Fix the module Makefile to at least properly inlcude opt_inet6.h
and allow builds without INET or INET6.

r234831
Make sure that the firmware version is available in
dev.t4nex.X.firmware_version even if the driver fails to attach
properly.  At least it'll be easy to tell what we're dealing with.

r234833:
Change the default to not use packet counters to generate rx interrupts.
Rely solely on the timer based mechanism.

Update man page to reflect this change.

END
K 10
svn:author
V 8
bschmidt
K 8
svn:date
V 27
2012-07-01T09:30:37.965272Z
K 7
svn:log
V 722
MFC BA/DELBA fixes:
- r234321:
  Use the M_AMPDU_MPDU flag to determine when to manually set the seqno and
  use a BA queue.
- r235686:
  Discard frames after a DELBA which where queued during an active BA
  session.
- r235687:
  remove unused vap variable
- r237647:
  Fix a TX aggregation issue, if after the last compressed BA notification
  the TX queue is empty, there won't be a TX done notification, effectly
  resulting in an mbuf leak. The correct way to handle this is to free
  up mbufs on both BA and TX done notifications up to the last sent seqno.
- r237649 (1):
  We need to defer passing the DELBA request to the firmware until the aggr
  queue is empty or the firmware will go nuts.

PR:		kern/167806 (1)

END
K 10
svn:author
V 2
np
K 8
svn:date
V 27
2012-07-01T09:35:15.350912Z
K 7
svn:log
V 53
Partial MFC of r236311 (uqs@):

Unbreak make depend.

END
K 10
svn:author
V 6
davide
K 8
svn:date
V 27
2012-07-01T11:52:52.667874Z
K 7
svn:log
V 368
- Take in account aggregation when comparing event times in
callout_process() and when we submit events to eventtimers(4).

- Fix a bug in the 'steps' variable logic in softclock(). It shouldn't be
zeroed every time we extract a new event for processing it from cc_expireq.

- Indent string should be there, I wrongly removed it.

Reported by:		mav [1,2], pluknet [3]

END
K 10
svn:author
V 2
np
K 8
svn:date
V 27
2012-07-01T12:00:36.848114Z
K 7
svn:log
V 282
Backport just the sys/{dev,modules}/cxgb{,e}/ parts of r237263, and then
disable the TOE and iWARP modules in the Makefiles (they won't compile
without the rest of r237263).

This reduces diffs between the cxgb/cxgbe drivers in head and 9 and
makes it easy to MFC other fixes to 9.

END
K 10
svn:author
V 8
brueffer
K 8
svn:date
V 27
2012-07-01T12:34:19.943166Z
K 7
svn:log
V 189
Similar to all.log, mention that /var/log/console.log has to be created and chmod'ed
to make logging work.

PR:		168889
Submitted by:	Robert Simmons <rsimmons0@gmail.com>
MFC after:	1 week

END
K 10
svn:author
V 6
davide
K 8
svn:date
V 27
2012-07-01T12:44:29.389038Z
K 7
svn:log
V 88
The object that should be zeroed is *xtimer, not xtimer. No space is needed
after cast.

END
K 10
svn:author
V 6
davide
K 8
svn:date
V 27
2012-07-01T12:48:56.557895Z
K 7
svn:log
V 19
MFC as per r237921

END
K 10
svn:author
V 8
brueffer
K 8
svn:date
V 27
2012-07-01T12:59:00.340008Z
K 7
svn:log
V 226
Replace an unreachable panic() in vm86_getptr (been there for 13 years) with
a KASSERT() behind the functions's only consumer.

Suggested by:	kib
Reviewed by:	kib
CID:		4494
Found with:	Coverity Prevent(tm)
MFC after:	2 weeks

END
K 10
svn:author
V 2
np
K 8
svn:date
V 27
2012-07-01T13:43:30.886768Z
K 7
svn:log
V 939
MFC r237832, r237436, r237439, r237463, r237512, r237587, r237799,
r237819, r237831.

r237832:
cxgb(4): IPv6 rx/tx hw checksum, IPv6 TSO and LRO too.

r237436:
cxgbe(4): update to firmware interface 1.5.2.0; updates to shared code.

r237439:
Do not read registers with read side effects while performing a register
dump for cxgbetool.

r237463:
Do not allocate extra vectors when adapter is not TOE
capable (or toecaps have been disallowed by the user).

r237512:
Better way to determine the status page length and rx pad boundary.

r237587:
Allow cxgbe(4) running within a VM to attach to its devices that have been
exported via PCI passthrough.

r237799:
cxgbe(4): support for IPv6 hardware checksumming (rx and tx).

r237819:
cxgbe(4): support for IPv6 TSO and LRO.

r237831:
- Assign (don't OR) the CSUM_XXX bits to csum_flags in the rx checksum code.
- Fix TSO/TSO4 mixup.
- Add IFCAP_LINKSTATE to the available/enabled capabilities.

END
K 10
svn:author
V 2
mm
K 8
svn:date
V 27
2012-07-01T14:54:54.225469Z
K 7
svn:log
V 51
Add myself for illumos vendor import (kernel part)

END
K 10
svn:author
V 2
mm
K 8
svn:date
V 27
2012-07-01T14:55:35.357742Z
K 7
svn:log
V 90
Add kernel part (uts) with ZFS to vendor/illumos
illumos-gate revision 13742:b6bbdd77139c

END
K 10
svn:author
V 2
mm
K 8
svn:date
V 27
2012-07-01T14:56:51.639777Z
K 7
svn:log
V 56
Remove myself after illumos vendor import (kernel part)

END
K 10
svn:author
V 7
glebius
K 8
svn:date
V 27
2012-07-01T15:30:43.133984Z
K 7
svn:log
V 207
In g_mirror_regular_request() upon successful delivery treat
BIO_DELETE requests same way as BIO_WRITE removing them from
queue. This fixes panic with BIO_DELETE operations on geom_mirror.

Reviewed by:	pjd

END
K 10
svn:author
V 7
glebius
K 8
svn:date
V 27
2012-07-01T15:43:52.623819Z
K 7
svn:log
V 911
Make geom_mirror more friendly to SSDs. To properly support TRIM,
we need to pass BIO_DELETE requests down to providers that support
it. Also, we need to announce our support for BIO_DELETE to upper
consumer. This requires:

- In g_mirror_start() return true for "GEOM::candelete" request.
- In g_mirror_init_disk() probe below provider for "GEOM::candelete"
  attribute, and mark disk with a flag if it does support BIO_DELETE.
- In g_mirror_register_request() distribute BIO_DELETE requests only
  to those disks, that do support it.

Note that we announce "GEOM::candelete" as true unconditionally of
whether we have TRIM-capable media down below or not. This is made
intentionally, because upper consumer (usually UFS) requests the
attribite only once at mount time. And if user ever migrates his
mirror from HDDs to SSDs, then he/she would get TRIM working without
remounting filesystem.

Reviewed by:	pjd

END
K 10
svn:author
V 3
pjd
K 8
svn:date
V 27
2012-07-01T16:26:07.430152Z
K 7
svn:log
V 50
Check if there is cmsg at all.

MFC after:	3 days

END
K 10
svn:author
V 6
marcel
K 8
svn:date
V 27
2012-07-01T17:06:21.483573Z
K 7
svn:log
V 880
Dump a milestone: we see disks on the SCSI bus with isp(4) using
the new busdma/mi implementation:
1.  Implement load_linear using a common support function. The support
    function takes a pmap_t, virtual address and length and extends the
    last segment or appends new segments to the memory descriptor.
2.  Implement unload as well now that we're getting too functional not
    to have it implemented.
3.  Comment-out dump routines and other debugging printfs. It's time now
    to use KTR or other means to have the ability present without it
    interfering with normal operation.
4.  Add a first-line of pretection: check arguments and return EINVAL
    when we're displeased with what;s given to us. Note that for the
    load_* functions we need a good definition of which errors are
    being returned directly and which errors are being relayed via the
    callback.

END
K 10
svn:author
V 6
marcel
K 8
svn:date
V 27
2012-07-01T17:15:50.431095Z
K 7
svn:log
V 614
Fire up our APs as soon as we kick the scheduler. We've woken them
a long time back anyway, so they've been sitting idle for long enough.
More importantly: the Altix PROM has assigned interrupts to CPUs and
having our APs on hold means that we can't handle inteerupts that go
to CPUs other than the APs -- this happens when we're doing interrupt
based configuration (SI_SUB_INT_CONFIG_HOOKS).

On NUMA machines like the Altix, it's infinitely better to have each
brick probe and attach its own devices, so waking the APs sooner is
generally a good idea. This is but a small step towards full SMP
booting though...

END
K 10
svn:author
V 6
marcel
K 8
svn:date
V 27
2012-07-01T17:33:07.249139Z
K 7
svn:log
V 549
Dump a WIP
1.  Introduce the global shub_iack function that gets called to
    acknowledge interrupts. Remove the 1-off handling for sncon0.
    What this really means is that we need a PIC abstraction that
    works for both the Altix and DIG interrupt model.
2.  Commit the iommu_map methods that make the busdma/mi code work
    that way it does right now.
3.  Remove the callout code from the pcib driver. It didn't give
    me anything useful.
4.  Add commented-out debugging printfs that I now need to change
    into KTR or other mechanisms.

END
K 10
svn:author
V 3
alc
K 8
svn:date
V 27
2012-07-01T18:27:13.196208Z
K 7
svn:log
V 169
MFC r233954, r236930
  Micro-optimize free_pv_entry() for the expected case.

  Avoid unnecessary atomic operations for clearing PGA_WRITEABLE in
  pmap_remove_pages().

END
K 10
svn:author
V 6
rpaulo
K 8
svn:date
V 27
2012-07-01T19:07:45.577320Z
K 7
svn:log
V 52
Remove bogus __unused attribute from hrowpic_eoi().

END
K 10
svn:author
V 8
schweikh
K 8
svn:date
V 27
2012-07-01T19:40:51.620141Z
K 7
svn:log
V 18
Correct a grammo.

END
K 10
svn:author
V 8
theraven
K 8
svn:date
V 27
2012-07-01T21:51:57.994347Z
K 7
svn:log
V 95
MFC symbol exports for libsupcc++ and libcxxrt that correctly export new and
delete operators.

END
K 10
svn:author
V 6
jilles
K 8
svn:date
V 27
2012-07-01T22:18:20.800320Z
K 7
svn:log
V 312
ctype_l(3): Note that not all these functions are specified by POSIX.1-2008.

The function isascii_l() is not in POSIX even though isascii() is,
probably because isascii() is marked as obsolete.  The other functions,
like digittoint_l() and ishexnumber_l(), are FreeBSD-specific just like
their non-_l versions.

END
K 10
svn:author
V 7
delphij
K 8
svn:date
V 27
2012-07-02T02:26:52.896842Z
K 7
svn:log
V 74
MFC r237568:

Fetch both ECDSA and RSA keys by default in ssh-keyscan(1).

END
K 10
svn:author
V 3
ken
K 8
svn:date
V 27
2012-07-02T02:36:05.272968Z
K 7
svn:log
V 220
MFC 237726:

  r237726 | ken | 2012-06-28 13:39:30 -0600 (Thu, 28 Jun 2012) | 5 lines

  Add a loader tunable, kern.cam.ctl.disable, that will disable
  loading CTL.  This may be useful in very low memory installations.

END
K 10
svn:author
V 7
delphij
K 8
svn:date
V 27
2012-07-02T02:38:27.210812Z
K 7
svn:log
V 67
Use timeclear() instead of home grown version.

MFC after:	2 weeks

END
K 10
svn:author
V 3
ken
K 8
svn:date
V 27
2012-07-02T02:44:00.393309Z
K 7
svn:log
V 190
MFC 237730:

  r237730 | ken | 2012-06-28 14:48:24 -0600 (Thu, 28 Jun 2012) | 5 lines

  Now that the mps(4) driver is endian-safe, add it to the powerpc and
  sparc64 GENERIC config files.

END
K 10
svn:author
V 3
ken
K 8
svn:date
V 27
2012-07-02T02:45:33.880686Z
K 7
svn:log
V 290
MFC 237730:

This only includes the sparc64 GENERIC, since the powerpc GENERIC64 config
file isn't in stable/8.

  r237730 | ken | 2012-06-28 14:48:24 -0600 (Thu, 28 Jun 2012) | 5 lines

  Now that the mps(4) driver is endian-safe, add it to the powerpc and
  sparc64 GENERIC config files.

END
K 10
svn:author
V 7
delphij
K 8
svn:date
V 27
2012-07-02T02:50:10.497011Z
K 7
svn:log
V 374
MFC r237571:

Fix a LOR acquiring the if_afdata lock while holding an rtentry lock.
Possibly do some entra work in case we would not get into the
ifa0 != NULL paths later as we already do for the mltaddr before.

XXX We should possibly error in case in6_setscope fails.

Reference: http://lists.freebsd.org/pipermail/freebsd-net/2011-September/029829.html

Submitted by:	bz

END
K 10
svn:author
V 7
delphij
K 8
svn:date
V 27
2012-07-02T02:54:22.987102Z
K 7
svn:log
V 374
MFC r237571:

Fix a LOR acquiring the if_afdata lock while holding an rtentry lock.
Possibly do some entra work in case we would not get into the
ifa0 != NULL paths later as we already do for the mltaddr before.

XXX We should possibly error in case in6_setscope fails.

Reference: http://lists.freebsd.org/pipermail/freebsd-net/2011-September/029829.html

Submitted by:	bz

END
K 10
svn:author
V 3
ken
K 8
svn:date
V 27
2012-07-02T02:56:01.123551Z
K 7
svn:log
V 286
MFC 237800:

  r237800 | ken | 2012-06-29 11:00:52 -0600 (Fri, 29 Jun 2012) | 7 lines

  Change the mps(4) driver to only scan a target if that is what is
  needed instead of scanning the full bus every time.

Submitted by:	mav
Discussed with:	Sreekanth Reddy <Sreekanth.Reddy@lsi.com>

END
K 10
svn:author
V 3
ken
K 8
svn:date
V 27
2012-07-02T02:57:52.443154Z
K 7
svn:log
V 285
MFC 237800:
  r237800 | ken | 2012-06-29 11:00:52 -0600 (Fri, 29 Jun 2012) | 7 lines

  Change the mps(4) driver to only scan a target if that is what is
  needed instead of scanning the full bus every time.

Submitted by:	mav
Discussed with:	Sreekanth Reddy <Sreekanth.Reddy@lsi.com>

END
K 10
svn:author
V 3
alc
K 8
svn:date
V 27
2012-07-02T03:49:52.912598Z
K 7
svn:log
V 235
MFC r226843
  Eliminate vestiges of page coloring in VM_ALLOC_NOOBJ calls to
  vm_page_alloc().  While I'm here, for the sake of consistency, always
  specify the allocation class, such as VM_ALLOC_NORMAL, as the first of
  the flags.

END
K 10
svn:author
V 3
alc
K 8
svn:date
V 27
2012-07-02T05:35:55.515609Z
K 7
svn:log
V 460
MFC r233290, r235598, r235973, r236045, r236240, r236291, r236378
  Change pv_entry_count to a long on amd64.

  Rename pmap_collect() to pmap_pv_reclaim() and rewrite it such that it
  no longer uses the active and inactive paging queues.

  Eliminate some purely stylistic differences among the amd64, i386
  native, and i386 xen PV entry allocators.

  Eliminate code duplication in free_pv_entry() and pmap_remove_pages()
  by introducing free_pv_chunk().

END
K 10
svn:author
V 6
obrien
K 8
svn:date
V 27
2012-07-02T05:57:01.187990Z
K 7
svn:log
V 451
MFC: r229809: Don't include <sys/timeb.h>.
The purpose of not including <sys/timeb.h> is to avoid the compiler
warning we put in <sys/timeb.h>.  So this change will cause CVS to
provide its own definition of 'struct timeb'.  This is only OK
because the code never calls the libc ftime(3) interface.  CVS only
uses 'struct timeb' as a convenience structure -- thus it doesn't
matter how its supplied definition does or does not match the
ftime(3) ABI.

END
K 10
svn:author
V 3
alc
K 8
svn:date
V 27
2012-07-02T05:57:44.405076Z
K 7
svn:log
V 143
MFC r235912
  There is no need for pmap_protect() to acquire the page queues lock
  unless it is going to access the pv lists or PMAP1/PADDR1.

END
K 10
svn:author
V 6
adrian
K 8
svn:date
V 27
2012-07-02T06:02:12.776090Z
K 7
svn:log
V 251
Bring over some further HAL capabilities from the Atheros HAL, as well
as an EDMA check function.

For the AR9003 and later NICs, different TX/RX DMA and descriptor handling
code will be conditional on the EDMA check.

Obtained from:	Qualcomm Atheros

END
K 10
svn:author
V 6
obrien
K 8
svn:date
V 27
2012-07-02T06:02:56.905419Z
K 7
svn:log
V 72
Record that r226343 r226347 r226348 r226349 r226353 r226388 was MFC'ed.

END
K 10
svn:author
V 6
adrian
K 8
svn:date
V 27
2012-07-02T06:05:25.973675Z
K 7
svn:log
V 10
Fix typo.

END
K 10
svn:author
V 6
adrian
K 8
svn:date
V 27
2012-07-02T06:06:27.098555Z
K 7
svn:log
V 18
Fix another typo.

END
K 10
svn:author
V 6
adrian
K 8
svn:date
V 27
2012-07-02T06:07:46.082065Z
K 7
svn:log
V 30
.. And fix another typo. Grr.

END
K 10
svn:author
V 5
dougb
K 8
svn:date
V 27
2012-07-02T06:22:20.121742Z
K 7
svn:log
V 531
For the ports modules building code, clean WRKDIR before building. This is
important for those that use -DNO_CLEAN routinely, since it will prevent
installing stale stuff, and even more important when the port is upgraded
to a newer version. When the user doesn't use -DNO_CLEAN, this will create
an infinitesimal amount of extra work, but won't hurt anything.

This is necessary because the ports tree has flags that prevent the ususal
'update the build if newer source files exist' logic from doing what it
would do in the base.

END
K 10
svn:author
V 2
mm
K 8
svn:date
V 27
2012-07-02T06:39:06.235452Z
K 7
svn:log
V 41
MFC r236247 (pjd):
Remove unused sysctl.

END
K 10
svn:author
V 2
mm
K 8
svn:date
V 27
2012-07-02T06:39:21.967488Z
K 7
svn:log
V 41
MFC r236247 (pjd):
Remove unused sysctl.

END
K 10
svn:author
V 8
brueffer
K 8
svn:date
V 27
2012-07-02T06:40:11.676132Z
K 7
svn:log
V 51
MFC: r237836

mps(4) is endian safe as of r237876.

END
K 10
svn:author
V 5
dougb
K 8
svn:date
V 27
2012-07-02T06:51:49.798666Z
K 7
svn:log
V 545
MFC r237958:

For the ports modules building code, clean WRKDIR before building. This is
important for those that use -DNO_CLEAN routinely, since it will prevent
installing stale stuff, and even more important when the port is upgraded
to a newer version. When the user doesn't use -DNO_CLEAN, this will create
an infinitesimal amount of extra work, but won't hurt anything.

This is necessary because the ports tree has flags that prevent the ususal
'update the build if newer source files exist' logic from doing what it
would do in the base.

END
K 10
svn:author
V 5
dougb
K 8
svn:date
V 27
2012-07-02T06:52:07.051395Z
K 7
svn:log
V 545
MFC r237958:

For the ports modules building code, clean WRKDIR before building. This is
important for those that use -DNO_CLEAN routinely, since it will prevent
installing stale stuff, and even more important when the port is upgraded
to a newer version. When the user doesn't use -DNO_CLEAN, this will create
an infinitesimal amount of extra work, but won't hurt anything.

This is necessary because the ports tree has flags that prevent the ususal
'update the build if newer source files exist' logic from doing what it
would do in the base.

END
K 10
svn:author
V 5
dougb
K 8
svn:date
V 27
2012-07-02T06:52:25.247028Z
K 7
svn:log
V 545
MFC r237958:

For the ports modules building code, clean WRKDIR before building. This is
important for those that use -DNO_CLEAN routinely, since it will prevent
installing stale stuff, and even more important when the port is upgraded
to a newer version. When the user doesn't use -DNO_CLEAN, this will create
an infinitesimal amount of extra work, but won't hurt anything.

This is necessary because the ports tree has flags that prevent the ususal
'update the build if newer source files exist' logic from doing what it
would do in the base.

END
K 10
svn:author
V 8
brueffer
K 8
svn:date
V 27
2012-07-02T06:58:10.866696Z
K 7
svn:log
V 122
MFC: r233648 (by eadler), r237836

Remove trailing whitespace per mdoc lint warning

mps(4) is endian safe as of r237877.

END
K 10
svn:author
V 3
alc
K 8
svn:date
V 27
2012-07-02T07:01:41.896041Z
K 7
svn:log
V 133
MFC r235695, r236158, r236190, r236494
  Replace all uses of the vm page queues lock by a r/w lock that is
  private to this pmap.c.

END
K 10
svn:author
V 6
obrien
K 8
svn:date
V 27
2012-07-02T07:10:33.998433Z
K 7
svn:log
V 70
MFC: r227518: reformat comment to be more readable in standard Xterm.

END
K 10
svn:author
V 6
obrien
K 8
svn:date
V 27
2012-07-02T07:11:43.773958Z
K 7
svn:log
V 76
MFC: r237279: install filemon.h into /usr/include for userland consumption.

END
K 10
svn:author
V 6
obrien
K 8
svn:date
V 27
2012-07-02T07:14:21.535076Z
K 7
svn:log
V 55
MFC: r227524: improve the grep-ability output strings.

END
K 10
svn:author
V 6
obrien
K 8
svn:date
V 27
2012-07-02T07:18:27.909824Z
K 7
svn:log
V 18
MFC: r227526: KNF

END
K 10
svn:author
V 6
obrien
K 8
svn:date
V 27
2012-07-02T07:22:22.112083Z
K 7
svn:log
V 72
Record that r228191 (ensure kldxref is run first) was MFC'ed [r231373].

END
K 10
svn:author
V 2
mm
K 8
svn:date
V 27
2012-07-02T07:27:14.986015Z
K 7
svn:log
V 170
Expose scrub and resilver tunables.
This allows the user to tune the priority trade-off between scrub/resilver
and other ZFS I/O.

MFC after:	2 weeks
Discussed with:	pjd

END
K 10
svn:author
V 6
obrien
K 8
svn:date
V 27
2012-07-02T07:37:06.068699Z
K 7
svn:log
V 209
MFC: r228193: tweak the r137233 fix to r136283 -- Code was making two send()
attempts vs. the comment documented "If we are working with a privileged
socket, then take only one attempt".  Make the code match.

END
K 10
svn:author
V 6
obrien
K 8
svn:date
V 27
2012-07-02T07:39:33.584215Z
K 7
svn:log
V 52
MFC: r228199: Split sentences at period boundaries.

END
K 10
svn:author
V 7
glebius
K 8
svn:date
V 27
2012-07-02T07:39:55.319921Z
K 7
svn:log
V 204
- Call bus_generic_attach() at end of hdspe_attach().
- Use own devclass_t in DRIVER_MODULE().

This fixes operation of hdspe(4) when built as loadable module.

Submitted by:	Ruslan Bukin <br bsdpad.com>

END
K 10
svn:author
V 2
mm
K 8
svn:date
V 27
2012-07-02T07:50:15.753798Z
K 7
svn:log
V 95
Update vendor/illumos/dist to illumos-gate revision 13743:95aba6e49b9f

Obtained from:	illumos

END
K 10
svn:author
V 3
avg
K 8
svn:date
V 27
2012-07-02T08:08:29.580630Z
K 7
svn:log
V 74
MFC r237748: dtrace instruction decoder: add 0x0f 0x1f NOP opcode support

END
K 10
svn:author
V 3
avg
K 8
svn:date
V 27
2012-07-02T08:09:07.254079Z
K 7
svn:log
V 74
MFC r237748: dtrace instruction decoder: add 0x0f 0x1f NOP opcode support

END
K 10
svn:author
V 6
obrien
K 8
svn:date
V 27
2012-07-02T08:21:15.553330Z
K 7
svn:log
V 100
MFC: r228475 & r228487: Disallow various debug.kdb sysctl's when securelevel
is raised.

PR:	161350

END
K 10
svn:author
V 6
obrien
K 8
svn:date
V 27
2012-07-02T08:28:16.296334Z
K 7
svn:log
V 50
MFC: r232274: Document fetch(1)'s 'URL' argument.

END
K 10
svn:author
V 3
mav
K 8
svn:date
V 27
2012-07-02T08:31:29.016822Z
K 7
svn:log
V 614
Add acpi_asus_wmi(4) -- driver for random extras found on WMI-compatible
Asus laptops. It is alike to acpi_asus(4), but uses WMI interface instead
of separate ACPI device.

On Asus EeePC T101MT netbook it allows to handle hotkeys and on/off WLAN,
Bluetooth, LCD backlight, camera, cardreader and touchpad.

On Asus UX31A ultrabook it allows to handle hotkeys, on/off WLAN, Bluetooth,
Wireless LED, control keyboard backlight brightness, monitor temperature
and fan speed. LCD brightness control doesn't work now for unknown reason,
possibly requiring some video card initialization.

Sponsored by:	iXsystems, Inc.

END
K 10
svn:author
V 3
mav
K 8
svn:date
V 27
2012-07-02T08:41:22.760822Z
K 7
svn:log
V 21
Fix typo in r237981.

END
K 10
svn:author
V 6
obrien
K 8
svn:date
V 27
2012-07-02T08:48:58.139711Z
K 7
svn:log
V 46
MFC: r234449: update file(1) to version 5.11.

END
K 10
svn:author
V 6
obrien
K 8
svn:date
V 27
2012-07-02T09:25:08.247109Z
K 7
svn:log
V 234
Allow building on FreeBSD 7.

Note that "HAVE_GETLINE" should be #undef'ed in config.h as were not using
the libc interface -- however that leads to redundant redeclaration compiler
warnings between the internal file.h and <stdio.h>.

END
K 10
svn:author
V 6
obrien
K 8
svn:date
V 27
2012-07-02T09:35:28.904543Z
K 7
svn:log
V 57
MFC: r228114 r228116 r234690: install a readline/tilde.h

END
K 10
svn:author
V 6
obrien
K 8
svn:date
V 27
2012-07-02T09:47:38.885087Z
K 7
svn:log
V 32
MFC: r235879: style.Makefile(5)

END
K 10
svn:author
V 3
kib
K 8
svn:date
V 27
2012-07-02T09:53:08.234533Z
K 7
svn:log
V 345
Do not override an error from uiomove() with (non-)error result from
bwrite().  VFS needs to know about EFAULT from uiomove() and does not
care much that partially filled block writeback after EFAULT was
successfull.  Early return without error causes short write to be
reported to usermode.

Reported and tested by:	andreast
MFC after:	3 weeks

END
K 10
svn:author
V 3
kib
K 8
svn:date
V 27
2012-07-02T09:53:57.405432Z
K 7
svn:log
V 164
Issue proper diagnostic on the short writes, also consider the
case of write reporting 0 bytes as short write.

Reported and tested by:	adreast
MFC after:   1 week

END
K 10
svn:author
V 6
obrien
K 8
svn:date
V 27
2012-07-02T10:00:28.159909Z
K 7
svn:log
V 41
MFC: r235939: consitently use "__LP64__"

END
K 10
svn:author
V 3
bms
K 8
svn:date
V 27
2012-07-02T10:06:00.536378Z
K 7
svn:log
V 137
Merge r237734 from HEAD:
  In MLDv2 general query processing, do not enforce the strict check
  on query origins.

Submitted by:	Gu Yong

END
K 10
svn:author
V 6
obrien
K 8
svn:date
V 27
2012-07-02T10:06:13.551939Z
K 7
svn:log
V 64
Revert r223540 - it broke the simple 'make -C kerberos5' build.

END
K 10
svn:author
V 3
bms
K 8
svn:date
V 27
2012-07-02T10:07:32.113439Z
K 7
svn:log
V 95
Merge r237735 from HEAD:
  Fix a typo in MLD query exponent processing.

Submitted by:	rpaulo@

END
K 10
svn:author
V 6
obrien
K 8
svn:date
V 27
2012-07-02T10:14:12.362665Z
K 7
svn:log
V 81
* Avoid race conditions with 'make -j<N>'.
  (partially MFC r236337 (in spirit))

END
K 10
svn:author
V 6
obrien
K 8
svn:date
V 27
2012-07-02T10:16:14.359442Z
K 7
svn:log
V 33
MFC: r236623: RAID is an acronym

END
K 10
svn:author
V 3
bms
K 8
svn:date
V 27
2012-07-02T11:46:47.009593Z
K 7
svn:log
V 131
Merge r237736 from HEAD:
  Kick the current-state report timer when a V1 group report would
  be triggered.

Submitted by:	rpaulo@

END
K 10
svn:author
V 8
brueffer
K 8
svn:date
V 27
2012-07-02T14:03:19.966382Z
K 7
svn:log
V 77
Fix XEN build, broken in r237924.

Reported by:	gcooper
Pointy hat:	brueffer

END