Commit Graph

1107 Commits

Author SHA1 Message Date
Edward Tomasz Napierala
a492b68b27 Use M_ZERO instead of explicit memsets and bzeros. 2012-10-30 12:52:41 +00:00
Edward Tomasz Napierala
1af2d09b49 Fix locking problem in disk_resize(); previously it would run without
topology lock, resulting in assertion when running with DIAGNOSTIC.

Reviewed by:	mav (earlier version)
2012-10-29 17:52:43 +00:00
Alexander Motin
8cff7eb82f 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.
2012-10-27 10:14:12 +00:00
Alexander Motin
15a2601b29 Remove several uses of numeric priorities from immediate CCB setups. 2012-10-27 09:40:29 +00:00
Alexander Motin
e1c2df4d30 Remove one more numeric priority constant. 2012-10-27 08:52:33 +00:00
Alexander Motin
5fb9dc04df Remove two more 'periph == NULL' checks missed in r241404.
This condition can never be true as functions are called from single place
and the checks just pollute the code and confuse Clang Static Analyzer.
2012-10-23 16:03:00 +00:00
Konstantin Belousov
5050aa86cf Remove the support for using non-mpsafe filesystem modules.
In particular, do not lock Giant conditionally when calling into the
filesystem module, remove the VFS_LOCK_GIANT() and related
macros. Stop handling buffers belonging to non-mpsafe filesystems.

The VFS_VERSION is bumped to indicate the interface change which does
not result in the interface signatures changes.

Conducted and reviewed by:	attilio
Tested by:	pho
2012-10-22 17:50:54 +00:00
Eitan Adler
db702c59cf remove duplicate semicolons where possible.
Approved by:	cperciva
MFC after:	1 week
2012-10-22 03:00:37 +00:00
Eitan Adler
9d3334e191 Adds 4K quirks for the some SSD's which all perform better when 4K
aligned and only except 4K deletes (TRIM).

PR:		kern/169974
Submitted by:	Steven Hartland <steven.hartland@multiplay.co.uk>
Tested by:	ak
Reviewed by:	mav
Approved by:	cperciva (implicit)
MFC after:	1 week
2012-10-20 15:30:14 +00:00
Eitan Adler
88bca58512 Add support for samsung HM250JI
PR:		usb/121474
Submitted by:	Ben Stuyts <ben@altesco.nl>
Approved by:	cperciva (implicit)
MFC after:	3 days
2012-10-16 17:49:14 +00:00
Eitan Adler
8b22e82143 Add support for the USB DISK Pro PMAP.
This patch has sit for 6 years in the PR database.

PR:		usb/96381
Submitted by:	jhs
Reviewed by:	mav
Approved by:	cperciva (implicit)
MFC after:	3 days
2012-10-15 12:33:53 +00:00
Alexander Motin
f8ff57d287 Add explicit check for not set time inside cam_periph_freeze_after_event().
System time is set later on boot process then initial bus scan by CAM.
Until that moment microtime() is equal to microuptime(), and if system
boots quickly, the value can be close to zero. That causes settle time
waiting even for buses that don't use reset during probe.

On my test system this reduces boot time by 1 second if USB enabled, or
by 4 seconds if USB disabled.  CAM waited for ctl2cam0 bus "settle".
2012-10-14 08:50:05 +00:00
Alexander Motin
9bccc7a939 Don't exclude XPT SIM from locking in xpt_create_path_unlocked().
We don't want xpt periph, device, target or bus disappeared because of
incorrect reference counting.
2012-10-13 18:24:52 +00:00
Alexander Motin
8cdef07f7c Use xpt_create_path_unlocked() for initial debug path compilation because
we are not holding respective SIM lock.
2012-10-13 18:11:50 +00:00
Alexander Motin
e6ce97d1b5 Extend SIM lock coverage during the bus registration process. 2012-10-13 17:55:06 +00:00
Alexander Motin
80d6987c5a Fix XPT_DEBUG paths operations locking:
- Extend the lock to cover xpt_path_release() for the new path.
 - While xpt_action() is called while holding right SIM lock for the new
   bus, the old path release may require different SIM lock. So we have
   to temporary drop the new lock and get the old one.
2012-10-13 11:23:16 +00:00
Alexander Motin
92c40f40fd XPT_DEV_MATCH is probably the only xpt_action() method that is called
without holding SIM lock. It really doesn't need that lock, but adding it
removes that specific exception, allowing to assert locking there later.

Submitted by:	ken@ (earlier version)
2012-10-13 10:18:36 +00:00
Alexander Motin
24a1ba9110 Get SIM lock in several places while calling CAM functions.
This fixes several use-after-free panics on systems with SAS enclosures.

Submitted by:	ken@, mav@
2012-10-12 18:21:31 +00:00
Alexander Motin
6884b66275 Protect xpt_getattr() calls with the SIM lock and assert that.
Submitted by:	ken@ (earlier version)
2012-10-12 17:18:24 +00:00
Alexander Motin
596ee08f65 Use separate malloc buckets for CAM devices, CCBs and paths. This will
make it easier to track down the source of any use after free problems.

Submitted by:	ken@
2012-10-11 20:14:11 +00:00
Alexander Motin
32aa80a6bd Don't duplicate path/ccb allocation code, use existing functions. 2012-10-11 19:57:11 +00:00
Alexander Motin
8cc6481cd5 Increase device CCB queue array size by CAM_RL_VALUES - 1 (4) elements.
It is required to store extra recovery requests in case of bus resets.
On ATA/SATA this fixes assertion panics on HEAD with INVARIANTS enabled or
possible memory corruptions otherwise if timeout/reset happens when device
CCB queue is already full.

Reported by:	gibbs@
MFC after:	1 week
2012-10-11 15:21:07 +00:00
Alexander Motin
0b8ce217a8 There are SCSI conditions that are not an errors. In those cases cderror()
returns zero while request status is not CAM_REQ_CMP.  That could cause
partial device attach or other unexpected results.

Found by:	Clang Static Analyzer
2012-10-10 19:32:40 +00:00
Alexander Motin
627995dcde Really handle xpt_compile_path() error in xpt_bus_register() instead of
print error message and probably crash just after it on NULL dereference.

Found by:	Clang Static Analyzer
2012-10-10 18:34:15 +00:00
Alexander Motin
aa2a1aaf90 Remove 'periph == NULL' check from bunch of periph drivers.
This condition can never be true as functions are called from single place
and the checks just pollute the code and confuse Clang Static Analyzer.
2012-10-10 18:10:11 +00:00
Andriy Gapon
d50aaa6d6d ata_da: set disk::d_ident from serial number
MFC after:	10 days
2012-10-06 21:42:07 +00:00
Alexander Motin
d6e285946d Change queue overflow checks from DIAGNOSTIC+panic() to KASSERT() to make
them enabled on HEAD by default. It is probably better to do single compare
then hunt for unexpected memory corruption.
2012-09-28 12:13:34 +00:00
Edward Tomasz Napierala
a0a6ff825b Remove useless NULL checks after M_WAITOK allocations. 2012-09-27 10:51:38 +00:00
Edward Tomasz Napierala
43f3d8e372 Fix panic in CTL caused by trying to free invalid pointers passed
by the userland process via the IOCTL interface.

Reviewed by:	ken@
2012-09-26 07:09:15 +00:00
Tijl Coosemans
303d68bc4f Fix a panic when trying to play invalid audio tracks. 2012-09-19 18:42:31 +00:00
Eitan Adler
0af1b47258 s/ is is / is /g
s/ a a / a /g

Approved by:	cperciva
MFC after:	3 days
2012-09-14 22:00:03 +00:00
John Baldwin
a89828a2b0 Remove some more NetBSD compat shims and other unused bits from these
drivers:
- Remove scsi_low_pisa.*, they were unused.
- Remove <compat/netbsd/physio_proc.h> and calls to the stubs in that
  header.  They were empty nops.
- Retire sl_xname and use device_get_nameunit() and device_printf() with
  the underlying device_t instead.
- Remove unused {ct,ncv,nsp,stg}print() functions.
- Remove empty SOFT_INTR_REQUIRED() macro and the unused sl_irq member.
2012-09-10 18:49:49 +00:00
John Baldwin
b9b256e49a Remove NetBSD compat shims for drivers originally shared with NetBSD/pc98.
NetBSD/pc98 was never merged into the main NetBSD tree and is no longer
developed.  Adding locking to these drivers would have made the compat
shims hard to impossible to maintain, so remove the shims to ease
future changes.

These changes were verified by md5.  Some additional shims can be removed
that do affect the compiled results that I will probably do in another
round.

Approved by:	nyan (tentatively)
2012-09-06 18:53:33 +00:00
Jim Harris
3ae17a4ed8 Fix scsi_da's BIO_DELETE->SCSI_UNMAP translation to use correct local
variable when determining various sizes related to SCSI UNMAP block
descriptor lists.

Sponsored by:	Intel
Reviewed by:	mav
MFC after:	3 days
2012-08-24 17:08:02 +00:00
Matt Jacob
5e6609a21a 1. Remove SEN support. I doubt there are any working examples
of this hardware still running (close to twenty years now).

2. Quiesece and use ENC_VLOG instead of ENC_LOG for most
complaints. That is, they're visible with bootverbose, but
otherwise quiesced and not repeatedly spamming messages
with constant reminders that hardware in this space is
rarely fully compliant.

MFC after:	1 month
2012-08-12 17:01:07 +00:00
Matt Jacob
10d4323996 Add missing VERIFY_10 definition.
MFC after:	1 month
2012-08-12 16:58:38 +00:00
Bjoern A. Zeeb
caf144a280 Remove opt_enc.h from files committed with r235911. enc(4) is the
'encapsulating interface' used with IPsec and has nothing to do with
storage 'enclosure' services.

MFC after:	3 days
Noticed while:	debugging why enc(4) is no longer automatically created
2012-07-30 03:00:58 +00:00
Alexander Motin
3631c6382f Implement media change notification for DA and CD removable media devices.
It includes three parts:
 1) Modifications to CAM to detect media media changes and report them to
disk(9) layer. For modern SATA (and potentially UAS) devices it utilizes
Asynchronous Notification mechanism to receive events from hardware.
Active polling with TEST UNIT READY commands with 3 seconds period is used
for incapable hardware. After that both CD and DA drivers work the same way,
detecting two conditions: "NOT READY: Medium not present" after medium was
detected previously, and "UNIT ATTENTION: Not ready to ready change, medium
may have changed". First one reported to disk(9) as media removal, second
as media insert/change. To reliably receive second event new
AC_UNIT_ATTENTION async added to make UAs broadcasted to all periphs by
generic error handling code in cam_periph_error().
 2) Modifications to GEOM core to handle media remove and change events.
Media removal handled by spoiling all consumers attached to the provider.
Media change event also schedules provider retaste after spoiling to probe
new media. New flag G_CF_ORPHAN was added to consumers to reflect that
consumer is in process of destruction. It allows retaste to create new
geom instance of the same class, while previous one is still dying.
 3) Modifications to some GEOM classes: DEV -- to report media change
events to devd; VFS -- to handle spoiling same as orphan to prevent
accessing replaced media. PART class already handles spoiling alike to
orphan.

Reviewed by:	silence on geom@ and scsi@
Tested by:	avg
Sponsored by:	iXsystems, Inc. / PC-BSD
MFC after:	2 months
2012-07-29 11:51:48 +00:00
Matt Jacob
3388f7a926 Handle a case where we had an SRR that pushed back the
data pointer. This is a temp fix that resubmits the
command, adjusted, so that the backend can fetch the
data again.

Sponsored by: Spectralogic
MFC after:	1 month
2012-07-28 20:08:14 +00:00
Alexander Motin
6d4c4b3fde Fix off by one error in ses_enc_desc_last_byte(). 2012-07-24 13:32:49 +00:00
Alexander Motin
262b5c50db Do not call ses_softc_cleanup() in case of configuration read failure.
Just free inclomplete daemon cache instead to let it retry next time.
Premature ses_softc_cleanup() caused NULL dereference when freed softc
was accessed later.
2012-07-24 13:08:43 +00:00
Alexander Motin
9eb97fed6b Fix some typos in r238595.
Reported by:	brueffer
2012-07-18 12:41:09 +00:00
Alexander Motin
cdcb782817 Add bunch of new ASC/ASCQ values from T10 site. 2012-07-18 12:23:45 +00:00
Matt Jacob
3d4e15d43e The call to disk_resize causes a panic if DIAGNOSTIC is set.
Coping with that while the finest minds of our generation
figure out why.
2012-07-14 02:59:11 +00:00
Christian Brueffer
97ab227579 Add and utilize defines for the ATA device register.
PR:		169764
Submitted by:	Steven Hartland <steven.hartland@multiplay.co.uk>
Reviewed by:	mav
MFC after:	2 weeks
2012-07-12 10:09:34 +00:00
Christian Brueffer
af01090551 Apply similar same change as in r238379:
Renamed the kern.cam.ada.ada_send_ordered sysctl and tunable to
kern.cam.ada.send_ordered, more in line with the other da sysctls/tunables.

Suggested by:	kib
2012-07-11 23:22:09 +00:00
Christian Brueffer
fdbad0ba9d Renamed the kern.cam.da.da_send_ordered sysctl and tunable to
kern.cam.da.send_ordered, more in line with the other da sysctls/tunables.

PR:		169765
Submitted by:	Steven Hartland <steven.hartland@multiplay.co.uk>
Reviewed by:	mav
2012-07-11 23:00:26 +00:00
Christian Brueffer
f4bd5f21f0 - fix description of SECURITY_FREEZE_LOCK (1)
- add missing underscores to SECURITY_DISABLE_PASSWORD

PR:		169763 (1)
Submitted by:	Steven Hartland <steven.hartland@multiplay.co.uk>
Reviewed by:	mav
MFC after:	1 week
2012-07-11 08:24:30 +00:00
Edward Tomasz Napierala
c79ea20306 Make the da(4) driver notify GEOM about LUN size change.
Reviewed by:	mav
Sponsored by:	FreeBSD Foundation
2012-07-07 22:19:51 +00:00
Eitan Adler
3743101f93 Remove variables which are initialized but never used thereafter
reported by gcc46 warning

Reviewed by:	scottl
Approved by:	cperciva
MFC after:	1 week
2012-07-07 17:17:43 +00:00