Commit Graph

98912 Commits

Author SHA1 Message Date
Alexander Motin
409a3c1383 Add LUN options to specify 64-bit EUI and NAA identifiers. 2014-07-09 04:37:50 +00:00
Peter Wemm
ba8cd08ba9 Bump __FreeBSD_version after last SA-14:17.kmem so we have something
to test against in the freebsd.org cluster.
2014-07-09 00:12:05 +00:00
Xin LI
e432298ade Initialize SCTP cmsg's and notification's buffer before copying out
to userland.

Submitted by:	tuexen
Security:	CVE-2014-3953
Security:	FreeBSD-SA-14:17.kmem
2014-07-08 21:54:27 +00:00
Xin LI
2827952eb4 Don't leave the padding between the msg header and the cmsg data,
and the padding after the cmsg data un-initialized.

Submitted by:	tuexen
Security:	CVE-2014-3952
Security:	FreeBSD-SA-14:17.kmem
2014-07-08 21:54:23 +00:00
Neel Natu
b301b9e28f Accurately identify the vcpu's operating mode as 64-bit, compatibility,
protected or real.
2014-07-08 21:48:57 +00:00
Neel Natu
3527963b26 Invalidate guest TLB mappings as a side-effect of its CR3 being updated.
This is a pre-requisite for task switch emulation since the CR3 is loaded
from the new TSS.
2014-07-08 20:51:03 +00:00
Alexander Motin
3120a49e50 Remove status setting from datamove() path. Leave that to other places. 2014-07-08 18:51:03 +00:00
Alexander Motin
e327a057a7 Remove IO_SYNC flag when writing extended file attributes on ZFS.
While it is possible to create and write file, modify its permissions, etc.
without ever doing sync, it looks odd that it is required for setting
extended file attributes on ZFS.  UFS does not do sync there too.

Samba uses those extended attributes to store some its data, and doing it
synchronously by many times reduces file creation performance for systems
without SLOG device.

Reviewed by:	delphij, jpaetzel, silence on fs@
MFC after:	2 weeks
Sponsored by:	iXsystems, Inc.
2014-07-08 17:26:08 +00:00
Alexander Motin
ad3cd840f2 Fix use-after-free on XPT_RESET_BUS.
That command is not queued, so does not use later status update.
2014-07-08 16:56:21 +00:00
Alexander Motin
b33b96e352 Enable TAS feature: notify initiator if its command was aborted by other.
That should make operation more kind to multi-initiator environment.
Without this, other initiators may find out that something bad happened
to their commands only via command timeout.
2014-07-08 16:38:05 +00:00
Ian Lepore
1e3d53c687 Use named constant rather than '0' to access the reset controller register. 2014-07-08 14:35:09 +00:00
Alexander Motin
d6205772da Fix typo in r267873. 2014-07-08 13:28:37 +00:00
Alexander Motin
950b6e126b Pass correct command that should be aborted to ISPCTL_ABORT_CMD.
This makes XPT_ABORT to work for me on initiator side of isp(4).
Previous code was trying to abort the XPT_ABORT itself and failed.

MFC after:	1 week
2014-07-08 13:01:36 +00:00
Alexander Motin
6c7a99be9f Do not return statuses for aborted iSCSI commands. 2014-07-08 12:16:28 +00:00
Alexander Motin
f5ffef352f Return task management requests to queued execution, but differently.
Testing shown that both original queued design with separate task queue,
and recent direct execution design had significant flaw: If abort request
arrives just after the victim, the last one may not be in the ooa_queue
yet, and so invisible for the task management function.

Unlike original queued implementation, use same queue for all SCSI and
TASK requests from the same initiator. That avoids races between them:
task functions are always executed in proper time, relatively to other
requests.
2014-07-08 12:15:15 +00:00
Alexander Motin
aa75114c57 Add XPT_ABORT support to iSCSI initiator.
While CAM does not use it normally, it is useful for targets testing.

MFC after:	2 weeks
2014-07-08 09:37:41 +00:00
Alexander Motin
fdfc6c8ebd Fix task management functions status: task not found is not an error,
while not implemented function is.
2014-07-08 08:34:34 +00:00
Konstantin Belousov
3bcc218f46 Correct the problem reported by test16 from
tools/regression/file/flock/flock.c, which completes the fix in
r192685.  When the lock was stolen from us, retry the whole lock
sequence in kernel, instead of returning EINTR to usermode and hoping
that application would handle it correctly by restarting the lock
acquire.

Tested by:	pho
Sponsored by:	The FreeBSD Foundation
MFC after:	2 weeks
2014-07-08 08:10:15 +00:00
Konstantin Belousov
a5244bac2e Correct si_code for the SIGBUS signal generated by the alignment trap.
Reported and tested by:	pho
Sponsored by:	The FreeBSD Foundation
MFC after:	1 week
2014-07-08 08:05:42 +00:00
Warner Losh
e3a6cb96d6 Fix typo in flag name. 2014-07-07 23:21:15 +00:00
Warner Losh
cfe87f0076 Naughty NANDFS was using hidden unused flag, hiding the fact that the
flag was used and wasn't really available. Change the name without
fixing any laying issues that might be present in NANDFS' use of this
flag.
2014-07-07 23:21:07 +00:00
Don Lewis
626a79752f Declaration whitespace changes for style(9).
MFC after:	1 week
2014-07-07 22:02:39 +00:00
Alexander Motin
462cf3ba2a Make XPT_GET_TRAN_SETTINGS to report CAM that command queueing is enabled,
but make couple changes to handle non-queued commands too, if happen.

MFC after:	2 weeks
2014-07-07 17:34:48 +00:00
Mateusz Guzik
5e2554b7f8 Don't call crdup nor uifind under vnode lock.
A locked vnode can get into the way of satisyfing malloc with M_WATOK.

This is a fixup to r268087.

Suggested by:	kib
MFC after:	1 week
2014-07-07 14:03:30 +00:00
Alexander Motin
dbd849d868 Fix "use after free" on port creation error in r268291. 2014-07-07 11:52:22 +00:00
Alexander Motin
1e5a8b8f4b Add support for READ FULL STATUS action of PERSISTENT RESERVE IN command. 2014-07-07 11:05:04 +00:00
Alexander Motin
604e257984 Teach ctl_add_initiator() to dynamically allocate IIDs from pool.
If port passed negative IID value, the function will try to allocate IID
from the pool of unused, based on passed wwpn or name arguments.  It does
all its best to make IID unique and persistent across reconnects.

This makes persistent reservation properly work for iSCSI.  Previously,
in case of reconnects, reservation could be unexpectedly lost, or even
migrate between intiators.
2014-07-07 09:37:22 +00:00
Alexander Motin
0f8de8afaa Fix bugs for PERSISTENT RESERVE OUT bits in r268096. 2014-07-07 08:58:36 +00:00
Fabien Thomas
b8fad6c0ef Optim and Fix for mge driver:
- add missing rcvif in mbuf
- add missing ipacket stat
- remove uncessary mbuf copy on output path
- fix deadlock of the TX engine in case of error

Obtained from:	NETASQ
MFC after:	2 weeks
2014-07-07 08:22:39 +00:00
Alexander Motin
99f8c067e6 Correction to r268356: collide only sessions to the same target. 2014-07-07 06:17:07 +00:00
Alexander Motin
2c6c9e47b2 When new connection comes in, check whether we already have session from
the same intiator (Name+ISID).  If so -- terminate the old session and let
the new one take its place, as required by iSCSI RFC.
2014-07-07 05:48:11 +00:00
Hans Petter Selasky
88e0a63961 Improve support for Intel Lynx Point USB 3.0 controllers by masking
the port routing bits like done in Linux.

MFC after:	1 week
Tested by:	Tur-Wei Chan <twchan@singnet.com.sg>
2014-07-07 05:17:16 +00:00
Alexander Motin
0020682baa Implement ABORT TASK SET and I_T NEXUS RESET task management functions.
Use the last one to terminate active commands on iSCSI session termination.
Previous code was aborting only commands doing some data moves.
2014-07-07 03:10:56 +00:00
Marcel Moolenaar
e7d939bda2 Remove ia64.
This includes:
o   All directories named *ia64*
o   All files named *ia64*
o   All ia64-specific code guarded by __ia64__
o   All ia64-specific makefile logic
o   Mention of ia64 in comments and documentation

This excludes:
o   Everything under contrib/
o   Everything under crypto/
o   sys/xen/interface
o   sys/sys/elf_common.h

Discussed at: BSDcan
2014-07-07 00:27:09 +00:00
Nathan Whitehorn
00cf40b0ca Use common vt_fb parts in ofwfb as far as we are able without sacrificing
performance.

MFC after:	2 weeks
2014-07-07 00:12:18 +00:00
Bryan Venteicher
6700a7d44b Use the appropriate IPv6 hashtype defines when looking up the PCBGROUP
Reviewed by:	adrian@
2014-07-07 00:02:49 +00:00
Andreas Tobler
64175581d0 Make gcc happy, init idlen2. 2014-07-06 20:09:23 +00:00
Alexander Motin
1380b77c12 Close race in r268291 between port destruction, delayed by sessions
teardown, and new port creation during `service ctld restart`.

Close it by returning iSCSI port internal state, that allows to identify
dying ports, which should not be counted as existing, from really alive.
2014-07-06 17:57:59 +00:00
Alan Cox
09132ba6ac Introduce pmap_unwire(). It will replace pmap_change_wiring(). There are
several reasons for this change:

pmap_change_wiring() has never (in my memory) been used to set the wired
attribute on a virtual page.  We have always used pmap_enter() to do that.
Moreover, it is not really safe to use pmap_change_wiring() to set the wired
attribute on a virtual page.  The description of pmap_change_wiring() says
that it assumes the existence of a mapping in the pmap.  However, non-wired
mappings may be reclaimed by the pmap at any time.  (See pmap_collect().)
Many implementations of pmap_change_wiring() will crash if the mapping does
not exist.

pmap_unwire() accepts a range of virtual addresses, whereas
pmap_change_wiring() acts upon a single virtual page.  Since we are
typically unwiring a range of virtual addresses, pmap_unwire() will be more
efficient.  Moreover, pmap_unwire() allows us to unwire superpage mappings.
Previously, we were forced to demote the superpage mapping, because
pmap_change_wiring() only allowed us to express the unwiring of a single
base page mapping at a time.  This added to the overhead of unwiring for
large ranges of addresses, including the implicit unwiring that occurs at
process termination.

Implementations for arm and powerpc will follow.

Discussed with:	jeff, marcel
Reviewed by:	kib
Sponsored by:	EMC / Isilon Storage Division
2014-07-06 17:42:38 +00:00
Alexander Motin
ffe82e05b3 Make iSCSI initiator keep Initiator Session ID (ISID) across reconnects.
Previously ISID was changed every time, that made impossible correct
persistent reservation, because reconnected session was identified as
completely new one.

Reviewed by:	trasz
MFC after:	1 week
2014-07-06 17:37:49 +00:00
Nathan Whitehorn
0558e4bb2b In case we ever support little-endian PowerPC (probably userland only),
avoid hardcoding endianness here.
2014-07-06 16:20:37 +00:00
Nathan Whitehorn
770047f5bb Add a new CPU id for a POWER8 variant. 2014-07-06 16:19:55 +00:00
Hans Petter Selasky
5cb6b3afa4 Fix OFED startup order: All SYSINIT()'s and modules should be loaded
prior to starting "/sbin/init" which will run all the "/etc/rc.d/xxx"
scripts. Else there can be a race configuring the interfaces via
"/etc/rc.conf".

MFC after:	4 weeks
Sponsored by:	Mellanox Technologies
2014-07-06 14:22:13 +00:00
Hans Petter Selasky
22239af86c Fix compile warning.
MFC after:	4 weeks
Sponsored by:	Mellanox Technologies
2014-07-06 14:20:47 +00:00
Hans Petter Selasky
d291b07865 Fix some compile warnings.
MFC after:	4 weeks
Sponsored by:	Mellanox Technologies
2014-07-06 14:14:07 +00:00
Alexander Motin
99ae56ac82 Add support for SCSI Ports (88h) VPD page. 2014-07-06 07:34:18 +00:00
Alexander Motin
69d7b87790 Make REPORT TARGET PORT GROUPS command report realistic data instead of
hardcoded garbage.
2014-07-06 07:02:36 +00:00
Alexander Motin
c26eee2dc9 Move lun_map() method from command nexus to port.
Previous implementation made impossible to do some things, such as calling
it for ports other then one through which command arrived.
2014-07-06 06:21:34 +00:00
Alexander Motin
561764b1c5 Relax some bit checks for INQUIRY command.
FreeBSD still tries to put LUN number in second byte until it get device
protocol version, even that it was obsoleted about 20 years ago.
2014-07-06 06:12:29 +00:00
Gavin Atkinson
764442e03d Add support to asmc(4) for Macmini 3,1.
PR:		190195
Submitted by:	fbsdbugs2 sentry.org
MFC after:	1 week
Relnotes:	yes
2014-07-05 21:34:37 +00:00