commit fd3fc0b4d7305fa7246622dcc0dec69c42443f45 upstream.
Don't crash the machine just because of an empty transfer. Use WARN_ON()
combined with returning an error.
Found by Dmitry Vyukov and syzkaller.
[ Changed to "WARN_ON_ONCE()". Al has a patch that should fix the root
cause, but a BUG_ON() is not acceptable in any case, and a WARN_ON()
might still be a cause of excessive log spamming.
NOTE! If this warning ever triggers, we may end up leaking resources,
since this doesn't bother to try to clean the command up. So this
WARN_ON_ONCE() triggering does imply real problems. But BUG_ON() is
much worse.
People really need to stop using BUG_ON() for "this shouldn't ever
happen". It makes pretty much any bug worse. - Linus ]
Signed-off-by: Johannes Thumshirn <jthumshirn@suse.de>
Reported-by: Dmitry Vyukov <dvyukov@google.com>
Cc: James Bottomley <jejb@linux.vnet.ibm.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Willy Tarreau <w@1wt.eu>
commit 635d98b1d0cfc2ba3426a701725d31a6102c059a upstream.
scsi_init_io should only be called for requests that transfer data,
so move the assert that a request has segments from the callers into
scsi_init_io.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Signed-off-by: Willy Tarreau <w@1wt.eu>
commit a621bac3044ed6f7ec5fa0326491b2d4838bfa93 upstream.
When SCSI was written, all commands coming from the filesystem
(REQ_TYPE_FS commands) had data. This meant that our signal for needing
to complete the command was the number of bytes completed being equal to
the number of bytes in the request. Unfortunately, with the advent of
flush barriers, we can now get zero length REQ_TYPE_FS commands, which
confuse this logic because they satisfy the condition every time. This
means they never get retried even for retryable conditions, like UNIT
ATTENTION because we complete them early assuming they're done. Fix
this by special casing the early completion condition to recognise zero
length commands with errors and let them drop through to the retry code.
Reported-by: Sebastian Parschauer <s.parschauer@gmx.de>
Signed-off-by: James E.J. Bottomley <jejb@linux.vnet.ibm.com>
Tested-by: Jack Wang <jinpu.wang@profitbricks.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
[ jwang: backport from upstream 4.7 to fix scsi resize issue ]
Signed-off-by: Jack Wang <jinpu.wang@profitbricks.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Willy Tarreau <w@1wt.eu>
commit bc85dc500f9df9b2eec15077e5046672c46adeaa upstream.
By folding scsi_end_request into its only caller we can significantly clean
up the completion logic. We can use simple goto labels now to only have
a single place to finish or requeue command there instead of the previous
convoluted logic.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org>
Reviewed-by: Mike Christie <michaelc@cs.wisc.edu>
Reviewed-by: Hannes Reinecke <hare@suse.de>
[jwang: backport to 3.12]
Signed-off-by: Jack Wang <jinpu.wang@profitbricks.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Willy Tarreau <w@1wt.eu>
commit bba0bdd7ad4713d82338bcd9b72d57e9335a664b upstream.
SCSI transport drivers and SCSI LLDs block a SCSI device if the
transport layer is not operational. This means that in this state
no requests should be processed, even if the REQ_PREEMPT flag has
been set. This patch avoids that a rescan shortly after a cable
pull sporadically triggers the following kernel oops:
BUG: unable to handle kernel paging request at ffffc9001a6bc084
IP: [<ffffffffa04e08f2>] mlx4_ib_post_send+0xd2/0xb30 [mlx4_ib]
Process rescan-scsi-bus (pid: 9241, threadinfo ffff88053484a000, task ffff880534aae100)
Call Trace:
[<ffffffffa0718135>] srp_post_send+0x65/0x70 [ib_srp]
[<ffffffffa071b9df>] srp_queuecommand+0x1cf/0x3e0 [ib_srp]
[<ffffffffa0001ff1>] scsi_dispatch_cmd+0x101/0x280 [scsi_mod]
[<ffffffffa0009ad1>] scsi_request_fn+0x411/0x4d0 [scsi_mod]
[<ffffffff81223b37>] __blk_run_queue+0x27/0x30
[<ffffffff8122a8d2>] blk_execute_rq_nowait+0x82/0x110
[<ffffffff8122a9c2>] blk_execute_rq+0x62/0xf0
[<ffffffffa000b0e8>] scsi_execute+0xe8/0x190 [scsi_mod]
[<ffffffffa000b2f3>] scsi_execute_req+0xa3/0x130 [scsi_mod]
[<ffffffffa000c1aa>] scsi_probe_lun+0x17a/0x450 [scsi_mod]
[<ffffffffa000ce86>] scsi_probe_and_add_lun+0x156/0x480 [scsi_mod]
[<ffffffffa000dc2f>] __scsi_scan_target+0xdf/0x1f0 [scsi_mod]
[<ffffffffa000dfa3>] scsi_scan_host_selected+0x183/0x1c0 [scsi_mod]
[<ffffffffa000edfb>] scsi_scan+0xdb/0xe0 [scsi_mod]
[<ffffffffa000ee13>] store_scan+0x13/0x20 [scsi_mod]
[<ffffffff811c8d9b>] sysfs_write_file+0xcb/0x160
[<ffffffff811589de>] vfs_write+0xce/0x140
[<ffffffff81158b53>] sys_write+0x53/0xa0
[<ffffffff81464592>] system_call_fastpath+0x16/0x1b
[<00007f611c9d9300>] 0x7f611c9d92ff
Reported-by: Max Gurtuvoy <maxg@mellanox.com>
Signed-off-by: Bart Van Assche <bart.vanassche@sandisk.com>
Reviewed-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: James Bottomley <JBottomley@Odin.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 89fb4cd1f717a871ef79fa7debbe840e3225cd54 upstream.
Flush commands don't transfer data and thus need to be special cased
in the I/O completion handler so that we can propagate errors to
the block layer and filesystem.
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
Reported-by: Steven Haber <steven@qumulo.com>
Tested-by: Steven Haber <steven@qumulo.com>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
With the introduction of REQ_PM, modify sd's runtime suspend operation
functions to use that flag so that the operations to put the device into
runtime suspended state(i.e. sync cache and stop device) will not affect
its runtime PM status.
Signed-off-by: Lin Ming <ming.m.lin@intel.com>
Signed-off-by: Aaron Lu <aaron.lu@intel.com>
Acked-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
USB uses the .find_bridge() callback from struct acpi_bus_type
incorrectly, because as a result of the way it is used by USB every
device in the system that doesn't have a bus type or parent is
passed to usb_acpi_find_device() for inspection.
What USB actually needs, though, is to call usb_acpi_find_device()
for USB ports that don't have a bus type defined, but have
usb_port_device_type as their device type, as well as for USB
devices.
To fix that replace the struct bus_type pointer in struct
acpi_bus_type used for matching devices to specific subsystems
with a .match() callback to be used for this purpose and update
the users of struct acpi_bus_type, including USB, accordingly.
Define the .match() callback routine for USB, usb_acpi_bus_match(),
in such a way that it will cover both USB devices and USB ports
and remove the now redundant .find_bridge() callback pointer from
usb_acpi_bus.
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Acked-by: Yinghai Lu <yinghai@kernel.org>
Acked-by: Jeff Garzik <jgarzik@pobox.com>
When the ODD is powered off, any action the user did to the ODD that
would generate a media event will trigger an ACPI interrupt, so the
poll for media event is no longer necessary. And the poll will also
cause a runtime status change, which will stop the ODD from staying in
powered off state, so the poll should better be stopped.
But since we don't have access to the gendisk structure in LLDs, here
comes the disk_events_disable_depth for scsi device. This field is a
hint set by LLDs to convey information to upper layer drivers. A value
of 0 means media poll is necessary for the device, while values above 0
means media poll is not needed and should better be skipped. So we can
increase its value when we are to power off the ODD in ATA layer and
decrease its value when the ODD is powered on, effectively silence the
media events poll.
Signed-off-by: Aaron Lu <aaron.lu@intel.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
Pull block layer core updates from Jens Axboe:
"Here are the core block IO bits for 3.8. The branch contains:
- The final version of the surprise device removal fixups from Bart.
- Don't hide EFI partitions under advanced partition types. It's
fairly wide spread these days. This is especially dangerous for
systems that have both msdos and efi partition tables, where you
want to keep them in sync.
- Cleanup of using -1 instead of the proper NUMA_NO_NODE
- Export control of bdi flusher thread CPU mask and default to using
the home node (if known) from Jeff.
- Export unplug tracepoint for MD.
- Core improvements from Shaohua. Reinstate the recursive merge, as
the original bug has been fixed. Add plugging for discard and also
fix a problem handling non pow-of-2 discard limits.
There's a trivial merge in block/blk-exec.c due to a fix that went
into 3.7-rc at a later point than -rc4 where this is based."
* 'for-3.8/core' of git://git.kernel.dk/linux-block:
block: export block_unplug tracepoint
block: add plug for blkdev_issue_discard
block: discard granularity might not be power of 2
deadline: Allow 0ms deadline latency, increase the read speed
partitions: enable EFI/GPT support by default
bsg: Remove unused function bsg_goose_queue()
block: Make blk_cleanup_queue() wait until request_fn finished
block: Avoid scheduling delayed work on a dead queue
block: Avoid that request_fn is invoked on a dead queue
block: Let blk_drain_queue() caller obtain the queue lock
block: Rename queue dead flag
bdi: add a user-tunable cpu_list for the bdi flusher threads
block: use NUMA_NO_NODE instead of -1
block: recursive merge requests
block CFQ: avoid moving request to different queue
QUEUE_FLAG_DEAD is used to indicate that queuing new requests must
stop. After this flag has been set queue draining starts. However,
during the queue draining phase it is still safe to invoke the
queue's request_fn, so QUEUE_FLAG_DYING is a better name for this
flag.
This patch has been generated by running the following command
over the kernel source tree:
git grep -lEw 'blk_queue_dead|QUEUE_FLAG_DEAD' |
xargs sed -i.tmp -e 's/blk_queue_dead/blk_queue_dying/g' \
-e 's/QUEUE_FLAG_DEAD/QUEUE_FLAG_DYING/g'; \
sed -i.tmp -e "s/QUEUE_FLAG_DYING$(printf \\t)*5/QUEUE_FLAG_DYING$(printf \\t)5/g" \
include/linux/blkdev.h; \
sed -i.tmp -e 's/ DEAD/ DYING/g' -e 's/dead queue/a dying queue/' \
-e 's/Dead queue/A dying queue/' block/blk-core.c
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
Acked-by: Tejun Heo <tj@kernel.org>
Cc: James Bottomley <JBottomley@Parallels.com>
Cc: Mike Christie <michaelc@cs.wisc.edu>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Chanho Min <chanho.min@lge.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Implement support for WRITE SAME(10) and WRITE SAME(16) in the SCSI disk
driver.
- We set the default maximum to 0xFFFF because there are several
devices out there that only support two-byte block counts even with
WRITE SAME(16). We only enable transfers bigger than 0xFFFF if the
device explicitly reports MAXIMUM WRITE SAME LENGTH in the BLOCK
LIMITS VPD.
- max_write_same_blocks can be overriden per-device basis in sysfs.
- The UNMAP discovery heuristics remain unchanged but the discard
limits are tweaked to match the "real" WRITE SAME commands.
- In the error handling logic we now distinguish between WRITE SAME
with and without UNMAP set.
The discovery process heuristics are:
- If the device reports a SCSI level of SPC-3 or greater we'll issue
READ SUPPORTED OPERATION CODES to find out whether WRITE SAME(16) is
supported. If that's the case we will use it.
- If the device supports the block limits VPD and reports a MAXIMUM
WRITE SAME LENGTH bigger than 0xFFFF we will use WRITE SAME(16).
- Otherwise we will use WRITE SAME(10) unless the target LBA is beyond
0xFFFFFFFF or the block count exceeds 0xFFFF.
- no_write_same is set for ATA, FireWire and USB.
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Reviewed-by: Mike Snitzer <snitzer@redhat.com>
Reviewed-by: Jeff Garzik <jgarzik@redhat.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
FC and iSCSI class set SCSI devices to transport-offline state after
fast_io_fail/replacement_timeout has fired, but after relogin, function
scsi_internal_device_unblock() is not setting scsi device state to running.
Due to this the devices even after being relogged in remain offline.
Signed-off-by: Vikas Chaudhary <vikas.chaudhary@qlogic.com>
Reviewed-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
Two bits were appended to the end of the bitfield
list in struct scsi_device. Resolve that conflict
by including both bits.
Conflicts:
include/scsi/scsi_device.h
Avoid that the code for requeueing SCSI requests triggers a
crash by making sure that that code isn't scheduled anymore
after a device has been removed.
Also, source code inspection of __scsi_remove_device() revealed
a race condition in this function: no new SCSI requests must be
accepted for a SCSI device after device removal started.
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
Reviewed-by: Mike Christie <michaelc@cs.wisc.edu>
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
The return value of scsi_queue_insert() is ignored by all its
callers, hence change the return type of this function into
void.
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
Reviewed-by: Mike Christie <michaelc@cs.wisc.edu>
Reviewed-by: Tejun Heo <tj@kernel.org>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
When we call scsi_unprep_request() the command associated with the request
gets destroyed and therefore drops its reference on the device. If this was
the only reference, the device may get released and we end up with a NULL
pointer deref when we call blk_requeue_request.
Reported-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
Reviewed-by: Mike Christie <michaelc@cs.wisc.edu>
Reviewed-by: Tejun Heo <tj@kernel.org>
Cc: <stable@kernel.org>
[jejb: enhance commend and add commit log for stable]
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
Use blk_queue_dead() to test whether the queue is dead instead
of !sdev. Since scsi_prep_fn() may be invoked concurrently with
__scsi_remove_device(), keep the queuedata (sdev) pointer in
__scsi_remove_device(). This patch fixes a kernel oops that
can be triggered by USB device removal. See also
http://www.spinics.net/lists/linux-scsi/msg56254.html.
Other changes included in this patch:
- Swap the blk_cleanup_queue() and kfree() calls in
scsi_host_dev_release() to make that code easier to grasp.
- Remove the queue dead check from scsi_run_queue() since the
queue state can change anyway at any point in that function
where the queue lock is not held.
- Remove the queue dead check from the start of scsi_request_fn()
since it is redundant with the scsi_device_online() check.
Reported-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
Reviewed-by: Mike Christie <michaelc@cs.wisc.edu>
Reviewed-by: Tejun Heo <tj@kernel.org>
Cc: <stable@kernel.org>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
We do not hold the host lock when calling these functions,
so remove comment.
Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
This has scsi_internal_device_unblock/scsi_target_unblock take
the new state to set the devices as an argument instead of
always setting to running. The patch also converts users of these
functions.
This allows the FC and iSCSI class to transition devices from blocked
to transport-offline, so that when fast_io_fail/replacement_timeout
has fired we do not set the devices back to running. Instead, we
set them to SDEV_TRANSPORT_OFFLINE.
Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
This patch adds a new state SDEV_TRANSPORT_OFFLINE. It will
be used by transport classes to offline devices for cases like
when the fast_io_fail/recovery_tmo fires. In those cases we
want all IO to fail, and we have not yet escalated to dev_loss_tmo
behavior where we are removing the devices.
Currently to handle this state, transport classes are setting
the scsi_device's state to running, setting their internal
session/port structs state to something that indicates failed,
and then failing IO from some transport check in the queuecommand.
The reason for the new value is so that users can distinguish
between a device failure that is a result of a transport problem
vs the wide range of errors that devices get offlined for
when a scsi command times out and we offline the devices there.
It also fixes the confusion as to why the transport class is
failing IO, but has set the device state from blocked to running.
Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
For being able to bind ata devices against acpi devices, scsi_bus_type
needs to be set as bus in struct acpi_bus_type. So add wrapper to
scsi_lib to accomplish that.
Signed-off-by: Holger Macht <holger@homac.de>
Signed-off-by: Lin Ming <ming.m.lin@intel.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
block congestion control doesn't have any concept of fairness across
multiple queues. This means that if SCSI reports the host as busy in
the queue congestion control it can result in an unfair starvation
situation in dm-mp if there are multiple multipath devices on the same
host. For example:
http://www.redhat.com/archives/dm-devel/2012-May/msg00123.html
The fix for this is to report only the sdev busy state (and ignore the
host busy state) in the block congestion control call back.
The host is still congested, but the SCSI subsystem will sort out the
congestion in a fair way because it knows the relation between the
queues and the host.
[jejb: fixed up trailing whitespace]
Reported-by: Bernd Schubert <bernd.schubert@itwm.fraunhofer.de>
Tested-by: Bernd Schubert <bernd.schubert@itwm.fraunhofer.de>
Signed-off-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
1/ Rework remote-node-context (RNC) handling for proper management of
the silicon state machine in error handling and hot-plug conditions.
Further details below, suffice to say if the RNC is mismanaged the
silicon state machines may lock up.
2/ Refactor the initialization code to be reused for suspend/resume support
3/ Miscellaneous bug fixes to address discovery issues and hardware
compatibility.
RNC rework details from Jeff Skirvin:
In the controller, devices as they appear on a SAS domain (or
direct-attached SATA devices) are represented by memory structures known
as "Remote Node Contexts" (RNCs). These structures are transferred from
main memory to the controller using a set of register commands; these
commands include setting up the context ("posting"), removing the
context ("invalidating"), and commands to control the scheduling of
commands and connections to that remote device ("suspensions" and
"resumptions"). There is a similar path to control RNC scheduling from
the protocol engine, which interprets the results of command and data
transmission and reception.
In general, the controller chooses among non-suspended RNCs to find one
that has work requiring scheduling the transmission of command and data
frames to a target. Likewise, when a target tries to return data back
to the initiator, the state of the RNC is used by the controller to
determine how to treat the incoming request. As an example, if the RNC
is in the state "TX/RX Suspended", incoming SSP connection requests from
the target will be rejected by the controller hardware. When an RNC is
"TX Suspended", it will not be selected by the controller hardware to
start outgoing command or data operations (with certain priority-based
exceptions).
As mentioned above, there are two sources for management of the RNC
states: commands from driver software, and the result of transmission
and reception conditions of commands and data signaled by the controller
hardware. As an example of the latter, if an outgoing SSP command ends
with a OPEN_REJECT(BAD_DESTINATION) status, the RNC state will
transition to the "TX Suspended" state, and this is signaled by the
controller hardware in the status to the completion of the pending
command as well as signaled in a controller hardware event. Examples of
the former are included in the patch changelogs.
Driver software is required to suspend the RNC in a "TX/RX Suspended"
condition before any outstanding commands can be terminated. Failure to
guarantee this can lead to a complete hardware hang condition. Earlier
versions of the driver software did not guarantee that an RNC was
correctly managed before I/O termination, and so operated in an unsafe
way.
Further, the driver performed unnecessary contortions to preserve the
remote device command state and so was more complicated than it needed
to be. A simplifying driver assumption is that once an I/O has entered
the error handler path without having completed in the target, the
requirement on the driver is that all use of the sas_task must end.
Beyond that, recovery of operation is dependent on libsas and other
components to reset, rediscover and reconfigure the device before normal
operation can restart. In the driver, this simplifying assumption meant
that the RNC management could be reduced to entry into the suspended
state, terminating the targeted I/O request, and resuming the RNC as
needed for device-specific management such as an SSP Abort Task or LUN
Reset Management request.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.12 (GNU/Linux)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=HYvO
-----END PGP SIGNATURE-----
Merge tag 'isci-for-3.5' into misc
isci update for 3.5
1/ Rework remote-node-context (RNC) handling for proper management of
the silicon state machine in error handling and hot-plug conditions.
Further details below, suffice to say if the RNC is mismanaged the
silicon state machines may lock up.
2/ Refactor the initialization code to be reused for suspend/resume support
3/ Miscellaneous bug fixes to address discovery issues and hardware
compatibility.
RNC rework details from Jeff Skirvin:
In the controller, devices as they appear on a SAS domain (or
direct-attached SATA devices) are represented by memory structures known
as "Remote Node Contexts" (RNCs). These structures are transferred from
main memory to the controller using a set of register commands; these
commands include setting up the context ("posting"), removing the
context ("invalidating"), and commands to control the scheduling of
commands and connections to that remote device ("suspensions" and
"resumptions"). There is a similar path to control RNC scheduling from
the protocol engine, which interprets the results of command and data
transmission and reception.
In general, the controller chooses among non-suspended RNCs to find one
that has work requiring scheduling the transmission of command and data
frames to a target. Likewise, when a target tries to return data back
to the initiator, the state of the RNC is used by the controller to
determine how to treat the incoming request. As an example, if the RNC
is in the state "TX/RX Suspended", incoming SSP connection requests from
the target will be rejected by the controller hardware. When an RNC is
"TX Suspended", it will not be selected by the controller hardware to
start outgoing command or data operations (with certain priority-based
exceptions).
As mentioned above, there are two sources for management of the RNC
states: commands from driver software, and the result of transmission
and reception conditions of commands and data signaled by the controller
hardware. As an example of the latter, if an outgoing SSP command ends
with a OPEN_REJECT(BAD_DESTINATION) status, the RNC state will
transition to the "TX Suspended" state, and this is signaled by the
controller hardware in the status to the completion of the pending
command as well as signaled in a controller hardware event. Examples of
the former are included in the patch changelogs.
Driver software is required to suspend the RNC in a "TX/RX Suspended"
condition before any outstanding commands can be terminated. Failure to
guarantee this can lead to a complete hardware hang condition. Earlier
versions of the driver software did not guarantee that an RNC was
correctly managed before I/O termination, and so operated in an unsafe
way.
Further, the driver performed unnecessary contortions to preserve the
remote device command state and so was more complicated than it needed
to be. A simplifying driver assumption is that once an I/O has entered
the error handler path without having completed in the target, the
requirement on the driver is that all use of the sas_task must end.
Beyond that, recovery of operation is dependent on libsas and other
components to reset, rediscover and reconfigure the device before normal
operation can restart. In the driver, this simplifying assumption meant
that the RNC management could be reduced to entry into the suspended
state, terminating the targeted I/O request, and resuming the RNC as
needed for device-specific management such as an SSP Abort Task or LUN
Reset Management request.
Currently, __scsi_alloc_queue uses SCSI host's parent device
as DMA device to set segment boundary. But the parent device may not
refer to the DMA device. For example, for ATA disk, SCSI host's parent
device now refers to ATA port.
Since commit d139b9b([SCSI] scsi_lib_dma: fix bug with dma maps on
nested scsi objects), a new field Scsi_Host->dma_dev was introduced
to refer to the real DMA device.
Use ->dma_dev in __scsi_alloc_queue to correctly set segment
boundary.
Bug report: http://marc.info/?l=linux-ide&m=133177818318187&w=2
Reported-and-tested-by: Jörg Sommer <joerg@alea.gnuu.de>
Signed-off-by: Lin Ming <ming.m.lin@intel.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.18 (GNU/Linux)
iQEcBAABAgAGBQJPZxSnAAoJEDeqqVYsXL0M0Y4IAMX0vrTVZbg6psA5/gMcWGRP
CkFXEQ8n0PL2SCaj6BoDqamJFe5Nc7dnqxM0fGawB4S9vr3rHhiOlwO+NbV9zFYC
2skBTpeL3sjgtN/jTBdfeeAa7xTYpu/XGyei0NS1A5c2AyMVXV0uYV2s4VNZxe44
tVIn1OEzM2giZ9EB1OZslDMrg5XXm3MBIUECP0LbWUhBm/35caSFKzMXRwhh7WiK
+AVmc2AZYtdEwuknDyiH7KlsaoB3vGL9pPrAUJzIgEhy2pOo2A7W72HfA4Fj+y6a
uF9HBS5zciMp1+sGWry62AjNbWgin9BRlozBEO/lJhIfMGDV1nXEIJsOkOgkdoE=
=1TxB
-----END PGP SIGNATURE-----
Merge tag 'scsi-misc' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi-misc-2.6
SCSI updates from James Bottomley:
"The update includes the usual assortment of driver updates (lpfc,
qla2xxx, qla4xxx, bfa, bnx2fc, bnx2i, isci, fcoe, hpsa) plus a huge
amount of infrastructure work in the SAS library and transport class
as well as an iSCSI update. There's also a new SCSI based virtio
driver."
* tag 'scsi-misc' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi-misc-2.6: (177 commits)
[SCSI] qla4xxx: Update driver version to 5.02.00-k15
[SCSI] qla4xxx: trivial cleanup
[SCSI] qla4xxx: Fix sparse warning
[SCSI] qla4xxx: Add support for multiple session per host.
[SCSI] qla4xxx: Export CHAP index as sysfs attribute
[SCSI] scsi_transport: Export CHAP index as sysfs attribute
[SCSI] qla4xxx: Add support to display CHAP list and delete CHAP entry
[SCSI] iscsi_transport: Add support to display CHAP list and delete CHAP entry
[SCSI] pm8001: fix endian issue with code optimization.
[SCSI] pm8001: Fix possible racing condition.
[SCSI] pm8001: Fix bogus interrupt state flag issue.
[SCSI] ipr: update PCI ID definitions for new adapters
[SCSI] qla2xxx: handle default case in qla2x00_request_firmware()
[SCSI] isci: improvements in driver unloading routine
[SCSI] isci: improve phy event warnings
[SCSI] isci: debug, provide state-enum-to-string conversions
[SCSI] scsi_transport_sas: 'enable' phys on reset
[SCSI] libsas: don't recover end devices attached to disabled phys
[SCSI] libsas: fixup target_port_protocols for expanders that don't report sata
[SCSI] libsas: set attached device type and target protocols for local phys
...
The error reported up the stack for a discard failure did not clearly
indicate that the command was processed and subsequently failed by the
target device.
Return -EREMOTEIO so multipathing does not classify this condition as a
path failure.
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Acked-by: Mike Snitzer <snitzer@redhat.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
This patch fixes the host byte settings DID_TARGET_FAILURE and
DID_NEXUS_FAILURE. The function __scsi_error_from_host_byte, tries to reset
the host byte to DID_OK. But that does not happen because of the OR operation.
Here is the flow.
scsi_softirq_done-> scsi_decide_disposition -> __scsi_error_from_host_byte
Let's take an example with DID_NEXUS_FAILURE. In scsi_decide_disposition,
result will be set as DID_NEXUS_FAILURE (=0x11). Then in
__scsi_error_from_host_byte, when we do OR with DID_OK. Purpose is to reset
it back to DID_OK. But that does not happen. This patch fixes this issue.
Signed-off-by: Babu Moger <babu.moger@netapp.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
The sdev is deleted from starved list and then try to dispatch from this
device. It's quite possible the sdev can't eventually dispatch a request,
then the sdev will be in starved list tail. This isn't fair.
There are two cases here:
1. unplug path. scsi_request_fn() calls to scsi_target_queue_ready(), then
the dev is removed from starved list, but quite possible host queue isn't
ready, the dev is moved to starved list without dispatching any request.
2. scsi_run_queue path. It deletes the dev from starved list first (both
global and local starved lists), then handles the dev. Then we could have
the same process like case 1.
This patch fixes the first case. Case 2 isn't fixed, because there is a
rare case scsi_run_queue finds host isn't busy but scsi_request_fn finds
host is busy (other CPU is faster to get host queue depth). Not deleting
the dev from starved list in scsi_run_queue will keep scsi_run_queue
looping (though this is very rare case, because host will become busy).
Fortunately fixing case 1 already gives big improvement for starvation in
my test. In a 12 disk JBOD setup, running file creation under EXT4, this
gives 12% more throughput.
Signed-off-by: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
When we tear down a device we try to flush all outstanding
commands in scsi_free_queue(). However the check in
scsi_request_fn() is imperfect as it only signals that
we _might start_ aborting commands, not that we've actually
aborted some.
So move the printk inside the scsi_kill_request function,
this will also give us a hint about which commands are aborted.
Signed-off-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
* 'modsplit-Oct31_2011' of git://git.kernel.org/pub/scm/linux/kernel/git/paulg/linux: (230 commits)
Revert "tracing: Include module.h in define_trace.h"
irq: don't put module.h into irq.h for tracking irqgen modules.
bluetooth: macroize two small inlines to avoid module.h
ip_vs.h: fix implicit use of module_get/module_put from module.h
nf_conntrack.h: fix up fallout from implicit moduleparam.h presence
include: replace linux/module.h with "struct module" wherever possible
include: convert various register fcns to macros to avoid include chaining
crypto.h: remove unused crypto_tfm_alg_modname() inline
uwb.h: fix implicit use of asm/page.h for PAGE_SIZE
pm_runtime.h: explicitly requires notifier.h
linux/dmaengine.h: fix implicit use of bitmap.h and asm/page.h
miscdevice.h: fix up implicit use of lists and types
stop_machine.h: fix implicit use of smp.h for smp_processor_id
of: fix implicit use of errno.h in include/linux/of.h
of_platform.h: delete needless include <linux/module.h>
acpi: remove module.h include from platform/aclinux.h
miscdevice.h: delete unnecessary inclusion of module.h
device_cgroup.h: delete needless include <linux/module.h>
net: sch_generic remove redundant use of <linux/module.h>
net: inet_timewait_sock doesnt need <linux/module.h>
...
Fix up trivial conflicts (other header files, and removal of the ab3550 mfd driver) in
- drivers/media/dvb/frontends/dibx000_common.c
- drivers/media/video/{mt9m111.c,ov6650.c}
- drivers/mfd/ab3550-core.c
- include/linux/dmaengine.h
For the basic SCSI infrastructure files that are exporting symbols
but not modules themselves, add in the basic export.h header file
to allow the exports.
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Make sure that SCSI device removal via scsi_remove_host() does finish
all pending SCSI commands. Currently that's not the case and hence
removal of a SCSI host during I/O can cause a deadlock. See also
"blkdev_issue_discard() hangs forever if underlying storage device is
removed" (http://bugzilla.kernel.org/show_bug.cgi?id=40472). See also
http://lkml.org/lkml/2011/8/27/6.
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
Cc: <stable@kernel.org>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
During cable pull tests on our 16G FC adapter, we are seeing errors,
typically reads to close targets, which fail due to CRC or framing
errors caused by the cable being pull (return status DID_ERROR).
The adapter detects the error on one of the first frames received,
marks the FC exchange as dead (further frames go to bit bucket) and
signals the host of the error. This action is so quick, and coupled
with fast host CPUs, creates a scenario in which the midlayer sees
the failure and retries the io almost immediately. We've seen link
traces with the retry on the link while the original i/o is still
being processed by the target. We're also seeing the time window
for the "link to pull-apart" and the physical interface to report
disconnected to be in the few millisecond range. Which means, we're
encountering scenarios where the full retry count is exhausted
(all with error) by the midlayer before the link disconnect state
is detected.
We looked at 8G FC behavior and occasionally see the same behavior,
but as the link was slower, it rarely could exhaust all retries
before the link reported disconnect.
What is needed is a slight delay between io retries due to DID_ERROR
to cover this error. It is inappropriate to put this delay in the
driver, as the error is indistinguishable from other link-related errors,
nor does the driver track whether the io is a retry or not. This is also
easier than tracking between-io-error bursts that are seen in this
scenario.
The patch below updates the retry path so that it inserts a delay as
if the target was busy. The busy delay is on the order of 6ms. This
delay is sufficient to ensure the link down condition is reported
before the retry count is exhausted (at most 1 retry is seen).
Signed-off-by: Alex Iannicelli <alex.iannicelli@emulex.com>
Signed-off-by: James Smart <james.smart@emulex.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
USB surprise removal of sr is triggering an oops in
scsi_dispatch_command(). What seems to be happening is that USB is
hanging on to a queue reference until the last close of the upper
device, so the crash is caused by surprise remove of a mounted CD
followed by attempted unmount.
The problem is that USB doesn't issue its final commands as part of
the SCSI teardown path, but on last close when the block queue is long
gone. The long term fix is probably to make sr do the teardown in the
same way as sd (so remove all the lower bits on ejection, but keep the
upper disk alive until last close of user space). However, the
current oops can be simply fixed by not allowing any commands to be
sent to a dead queue.
Cc: stable@kernel.org
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
* 'for-linus' of git://git.kernel.dk/linux-2.6-block:
block: don't delay blk_run_queue_async
scsi: remove performance regression due to async queue run
blk-throttle: Use task_subsys_state() to determine a task's blkio_cgroup
block: rescan partitions on invalidated devices on -ENOMEDIA too
cdrom: always check_disk_change() on open
block: unexport DISK_EVENT_MEDIA_CHANGE for legacy/fringe drivers
Commit c21e6beb removed our queue request_fn re-enter
protection, and defaulted to always running the queues from
kblockd to be safe. This was a known potential slow down,
but should be safe.
Unfortunately this is causing big performance regressions for
some, so we need to improve this logic. Looking into the details
of the re-enter, the real issue is on requeue of requests.
Requeue of requests upon seeing a BUSY condition from the device
ends up re-running the queue, causing traces like this:
scsi_request_fn()
scsi_dispatch_cmd()
scsi_queue_insert()
__scsi_queue_insert()
scsi_run_queue()
scsi_request_fn()
...
potentially causing the issue we want to avoid. So special
case the requeue re-run of the queue, but improve it to offload
the entire run of local queue and starved queue from a single
workqueue callback. This is a lot better than potentially
kicking off a workqueue run for each device seen.
This also fixes the issue of the local device going into recursion,
since the above mentioned commit never moved that queue run out
of line.
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
The recent commit closing the race window in device teardown:
commit 86cbfb5607
Author: James Bottomley <James.Bottomley@suse.de>
Date: Fri Apr 22 10:39:59 2011 -0500
[SCSI] put stricter guards on queue dead checks
is causing a potential NULL deref in scsi_run_queue() because the
q->queuedata may already be NULL by the time this function is called.
Since we shouldn't be running a queue that is being torn down, simply
add a NULL check in scsi_run_queue() to forestall this.
Tested-by: Jim Schutt <jaschut@sandia.gov>
Cc: stable@kernel.org
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
We are currently using this flag to check whether it's safe
to call into ->request_fn(). If it is set, we punt to kblockd.
But we get a lot of false positives and excessive punts to
kblockd, which hurts performance.
The only real abuser of this infrastructure is SCSI. So export
the async queue run and convert SCSI over to use that. There's
room for improvement in that SCSI need not always use the async
call, but this fixes our performance issue and they can fix that
up in due time.
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
Instead of overloading __blk_run_queue to force an offload to kblockd
add a new blk_run_queue_async helper to do it explicitly. I've kept
the blk_queue_stopped check for now, but I suspect it's not needed
as the check we do when the workqueue items runs should be enough.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
* 'for-2.6.39/core' of git://git.kernel.dk/linux-2.6-block: (65 commits)
Documentation/iostats.txt: bit-size reference etc.
cfq-iosched: removing unnecessary think time checking
cfq-iosched: Don't clear queue stats when preempt.
blk-throttle: Reset group slice when limits are changed
blk-cgroup: Only give unaccounted_time under debug
cfq-iosched: Don't set active queue in preempt
block: fix non-atomic access to genhd inflight structures
block: attempt to merge with existing requests on plug flush
block: NULL dereference on error path in __blkdev_get()
cfq-iosched: Don't update group weights when on service tree
fs: assign sb->s_bdi to default_backing_dev_info if the bdi is going away
block: Require subsystems to explicitly allocate bio_set integrity mempool
jbd2: finish conversion from WRITE_SYNC_PLUG to WRITE_SYNC and explicit plugging
jbd: finish conversion from WRITE_SYNC_PLUG to WRITE_SYNC and explicit plugging
fs: make fsync_buffers_list() plug
mm: make generic_writepages() use plugging
blk-cgroup: Add unaccounted time to timeslice_used.
block: fixup plugging stubs for !CONFIG_BLOCK
block: remove obsolete comments for blkdev_issue_zeroout.
blktrace: Use rq->cmd_flags directly in blk_add_trace_rq.
...
Fix up conflicts in fs/{aio.c,super.c}
* git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi-misc-2.6: (170 commits)
[SCSI] scsi_dh_rdac: Add MD36xxf into device list
[SCSI] scsi_debug: add consecutive medium errors
[SCSI] libsas: fix ata list corruption issue
[SCSI] hpsa: export resettable host attribute
[SCSI] hpsa: move device attributes to avoid forward declarations
[SCSI] scsi_debug: Logical Block Provisioning (SBC3r26)
[SCSI] sd: Logical Block Provisioning update
[SCSI] Include protection operation in SCSI command trace
[SCSI] hpsa: fix incorrect PCI IDs and add two new ones (2nd try)
[SCSI] target: Fix volume size misreporting for volumes > 2TB
[SCSI] bnx2fc: Broadcom FCoE offload driver
[SCSI] fcoe: fix broken fcoe interface reset
[SCSI] fcoe: precedence bug in fcoe_filter_frames()
[SCSI] libfcoe: Remove stale fcoe-netdev entries
[SCSI] libfcoe: Move FCOE_MTU definition from fcoe.h to libfcoe.h
[SCSI] libfc: introduce __fc_fill_fc_hdr that accepts fc_hdr as an argument
[SCSI] fcoe, libfc: initialize EM anchors list and then update npiv EMs
[SCSI] Revert "[SCSI] libfc: fix exchange being deleted when the abort itself is timed out"
[SCSI] libfc: Fixing a memory leak when destroying an interface
[SCSI] megaraid_sas: Version and Changelog update
...
Fix up trivial conflicts due to whitespace differences in
drivers/scsi/libsas/{sas_ata.c,sas_scsi_host.c}
SBC3r26 contains many changes to the Logical Block Provisioning
interfaces (formerly known as Thin Provisioning ditto). This patch
implements support for both the old and new schemes using the same
heuristic as before (whether the LBP VPD page is present).
The new code also allows the provisioning mode (i.e. choice of command)
to be overridden on a per-device basis via sysfs. Two additional modes
are supported in this version:
- WRITE SAME(10) with the UNMAP bit set
- WRITE SAME(10) without the UNMAP bit set. This allows us to support
devices that predate the TP/LBP enhancements in SBC3 and which work
by way zero-detection
Switching between modes has been consolidated in a helper function that
also updates the block layer topology according to the limitations of
the chosen command.
I experimented with trying WRITE SAME(16) if UNMAP fails, WRITE SAME(10)
if WRITE SAME(16) fails, etc. but found several devices that got
cranky. So for now we'll disable discard if one of the commands
fail. The user still has the option of selecting a different mode in
sysfs.
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
When debugging DIF/DIX it is very helpful to be able to see which DIX
operation is associated with the scsi_cmnd. Include the protection op in
the SCSI command trace.
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
It was always abuse to reuse the plugging infrastructure for this,
convert it to the (new) real API for delaying queueing a bit. A
default delay of 3 msec is defined, to match the previous
behaviour.
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
Johannes Thumshirn