GitHub/moto-9609/android_kernel_motorola_exynos9610.git
8 years agoscsi: ibmvscsi: add write memory barrier to CRQ processing
Tyrel Datwyler [Wed, 7 Dec 2016 23:31:26 +0000 (17:31 -0600)]
scsi: ibmvscsi: add write memory barrier to CRQ processing

The first byte of each CRQ entry is used to indicate whether an entry is
a valid response or free for the VIOS to use. After processing a
response the driver sets the valid byte to zero to indicate the entry is
now free to be reused. Add a memory barrier after this write to ensure
no other stores are reordered when updating the valid byte.

Signed-off-by: Tyrel Datwyler <tyreld@linux.vnet.ibm.com>
Reviewed-by: Brian King <brking@linux.vnet.ibm.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: ibmvscsi: log bad SRP response opcode in hex format
Tyrel Datwyler [Wed, 7 Dec 2016 22:04:36 +0000 (16:04 -0600)]
scsi: ibmvscsi: log bad SRP response opcode in hex format

An unrecogonized or unsupported SRP response has its opcode currently
logged in decimal format. Log it in hex format instead so it can easily
be validated against the SRP specs values which are in hex.

Signed-off-by: Tyrel Datwyler <tyreld@linux.vnet.ibm.com>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: ibmvscsi: add vscsi hosts to global list_head
Tyrel Datwyler [Wed, 7 Dec 2016 22:04:35 +0000 (16:04 -0600)]
scsi: ibmvscsi: add vscsi hosts to global list_head

Add each vscsi host adatper to a new global list_head named
ibmvscsi_head. There is no functional change. This is meant primarily as
a convience for locating adapters from within the debugger or crash
utility.

[mkp: fixed typo]

Signed-off-by: Tyrel Datwyler <tyreld@linux.vnet.ibm.com>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: ufs-qcom: fix bug with read/modify write of UFS_CFG1
Subhash Jadavani [Tue, 6 Dec 2016 03:25:42 +0000 (19:25 -0800)]
scsi: ufs-qcom: fix bug with read/modify write of UFS_CFG1

ufs_qcom_print_hw_debug_reg_all() function is having a bug where it
might incorrectly modify undesired bits in UFS_CFG1 register, this
change fixes it.

Reviewed-by: Venkat Gopalakrishnan <venkatg@codeaurora.org>
Signed-off-by: Subhash Jadavani <subhashj@codeaurora.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: ufs: add quirk to increase host PA_SaveConfigTime
Subhash Jadavani [Tue, 6 Dec 2016 03:25:32 +0000 (19:25 -0800)]
scsi: ufs: add quirk to increase host PA_SaveConfigTime

The maximum value PA_SaveConfigTime is 250 (10us) but this is not enough
for some vendors. Gear switch from PWM to HS may fail even with this
max.  PA_SaveConfigTime. Gear switch can be issued by host controller as
an error recovery and any software delay will not help on this case so
we need to increase PA_SaveConfigTime to >32us as per vendor
recommendation.  This change adds a quirk to increase the
PA_SaveConfigTime parameter.

Reviewed-by: Venkat Gopalakrishnan <venkatg@codeaurora.org>
Signed-off-by: Subhash Jadavani <subhashj@codeaurora.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: ufs-qcom: add probe_defer in case phy driver not probed yet
Yaniv Gardi [Tue, 6 Dec 2016 03:25:15 +0000 (19:25 -0800)]
scsi: ufs-qcom: add probe_defer in case phy driver not probed yet

In case UFS driver is probed before the phy driver does, the UFS driver
should return a PROBE_DEFER code.

Signed-off-by: Yaniv Gardi <ygardi@codeaurora.org>
Signed-off-by: Subhash Jadavani <subhashj@codeaurora.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: ufs: add support for UFS HCI 2.1
Yaniv Gardi [Tue, 6 Dec 2016 03:25:02 +0000 (19:25 -0800)]
scsi: ufs: add support for UFS HCI 2.1

The UFS HCI v2.1 includes a few additional registers. This change
updates the HCI register, the UFS version register content and the
Interrupt Status register.

Signed-off-by: Yaniv Gardi <ygardi@codeaurora.org>
Signed-off-by: Subhash Jadavani <subhashj@codeaurora.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: g_NCR5380: Autoprobe board IRQ by default
Finn Thain [Mon, 5 Dec 2016 06:07:20 +0000 (01:07 -0500)]
scsi: g_NCR5380: Autoprobe board IRQ by default

Automatically probe the board irq when no irq parameter is provided, to
simulate PnP. The old default behaviour was to disable the irq.

Update driver documentation accordingly and add some printk messages to
make this behaviour visible.

Signed-off-by: Finn Thain <fthain@telegraphics.com.au>
Tested-by: Ondrej Zary <linux@rainbow-software.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: g_NCR5380: Fix automatic IRQ on HP C2502 cards
Finn Thain [Mon, 5 Dec 2016 06:07:20 +0000 (01:07 -0500)]
scsi: g_NCR5380: Fix automatic IRQ on HP C2502 cards

When IRQ_AUTO is used, the interrupt for HP C2502 cards gets disabled.
Fix this by programming the card for a suitable free irq. The code for
the free irq search comes from ALSA.

Also allow IRQ 9 to work (it aliases to IRQ 2 on the card), as per
Ondrej Zary's patch.

Suggested-by: Ondrej Zary <linux@rainbow-software.org>
Signed-off-by: Finn Thain <fthain@telegraphics.com.au>
Tested-by: Ondrej Zary <linux@rainbow-software.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: g_NCR5380: Use probe_irq_*() for IRQ probing
Ondrej Zary [Mon, 5 Dec 2016 06:07:20 +0000 (01:07 -0500)]
scsi: g_NCR5380: Use probe_irq_*() for IRQ probing

Use standard probe_irq_on() and probe_irq_off() functions instead of own
implementation.  This prevents warning messages like this in the kernel
log: genirq: Flags mismatch irq 1. 00000000 (NCR-probe) vs. 00000080
(i8042)

Move the IRQ trigger code from NCR5380 to g_NCR5380 where it is used.

Also clear interrupt flag before and after the probe.

Signed-off-by: Ondrej Zary <linux@rainbow-software.org>
Signed-off-by: Finn Thain <fthain@telegraphics.com.au>
Tested-by: Ondrej Zary <linux@rainbow-software.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: g_NCR5380: Check for chip presence before calling NCR5380_init()
Ondrej Zary [Mon, 5 Dec 2016 06:07:19 +0000 (01:07 -0500)]
scsi: g_NCR5380: Check for chip presence before calling NCR5380_init()

Write and read back mode register to check that a chip is really there.
If no card is present, reads result in 0xff.

Signed-off-by: Ondrej Zary <linux@rainbow-software.org>
Signed-off-by: Finn Thain <fthain@telegraphics.com.au>
Tested-by: Ondrej Zary <linux@rainbow-software.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: qla4xxx: switch to pci_alloc_irq_vectors
Christoph Hellwig [Tue, 6 Dec 2016 13:56:50 +0000 (14:56 +0100)]
scsi: qla4xxx: switch to pci_alloc_irq_vectors

And simplify the MSI-X logic in general - just request the two vectors
directly instead of going through an indirection table.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Nilesh Javali <nilesh.javali@cavium.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: hisi_sas: support deferred probe for v2 hw
John Garry [Tue, 6 Dec 2016 12:44:07 +0000 (20:44 +0800)]
scsi: hisi_sas: support deferred probe for v2 hw

In the hip06 and hip07 SoCs, the interrupt lines from the SAS
controllers are connected to mbigen hw module [1].  The mbigen module is
probed with module_init, and, as such, is not guaranteed to probe before
the SAS driver. So we need to support deferred probe.

We check for probe deferral in the hw layer probe, so we not probe into
the main layer and allocate shost, memories, etc., to later learn that
we need to defer the probe.

[1] ./Documentation/devicetree/bindings/interrupt-controller/hisilicon,mbigen-v2.txt

Signed-off-by: John Garry <john.garry@huawei.com>
Reviewed-by: Zhangfei Gao <zhangfei.gao@linaro.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: megaraid_sas: switch to pci_alloc_irq_vectors
Hannes Reinecke [Fri, 2 Dec 2016 11:52:23 +0000 (12:52 +0100)]
scsi: megaraid_sas: switch to pci_alloc_irq_vectors

[mkp: fixed bad indentation]

Signed-off-by: Hannes Reinecke <hare@suse.com>
Acked-by: Sumit Saxena <sumit.saxena@broadcom.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: scsi_devinfo: remove synchronous ALUA for NETAPP devices
Xose Vazquez Perez [Fri, 11 Nov 2016 16:49:49 +0000 (17:49 +0100)]
scsi: scsi_devinfo: remove synchronous ALUA for NETAPP devices

NetApp did confirm this is not required.

Cc: Martin George <Martin.George@netapp.com>
Cc: Robert Stankey <Robert.Stankey@netapp.com>
Cc: Steven Schremmer <Steven.Schremmer@netapp.com>
Cc: Sean Stewart <Sean.Stewart@netapp.com>
Cc: Hannes Reinecke <hare@suse.de>
Cc: Christophe Varoqui <christophe.varoqui@opensvc.com>
Cc: James E.J. Bottomley <jejb@linux.vnet.ibm.com>
Cc: Martin K. Petersen <martin.petersen@oracle.com>
Cc: SCSI ML <linux-scsi@vger.kernel.org>
Cc: device-mapper development <dm-devel@redhat.com>
Signed-off-by: Xose Vazquez Perez <xose.vazquez@gmail.com>
Reviewed-by: Sean Stewart <sean.stewart@netapp.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: be2iscsi: set errno on error path
Pan Bian [Sun, 4 Dec 2016 05:23:04 +0000 (13:23 +0800)]
scsi: be2iscsi: set errno on error path

Variable ret is reset in the loop, and its value will be 0 during the
second and after repeat of the loop. If pci_alloc_consistent() returns a
NULL pointer then, it will leaves with return value 0. 0 means no error,
which is contrary to the fact. This patches fixes the bug, explicitly
assigning "-ENOMEM" to return variable ret on the path that the call to
pci_alloc_consistent() fails.

Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=188951

Signed-off-by: Pan Bian <bianpan2016@163.com>
Reviewed-by: Jitendra Bhivare <Jitendra.bhivare@broadcom.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: be2iscsi: set errno on error path
Pan Bian [Sun, 4 Dec 2016 05:21:42 +0000 (13:21 +0800)]
scsi: be2iscsi: set errno on error path

Variable ret is reset in the loop, and its value will be 0 during the
second and after repeat of the loop. If pci_alloc_consistent() returns a
NULL pointer then, it will leaves with return value 0. 0 means no error,
which is contrary to the fact. This patches fixes the bug, explicitly
assigning "-ENOMEM" to return variable ret on the path that the call to
pci_alloc_consistent() fails.

Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=188941

Signed-off-by: Pan Bian <bianpan2016@163.com>
Reviewed-by: Jitendra Bhivare <Jitendra.bhivare@broadcom.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: hpsa: fallback to use legacy REPORT PHYS command
Hannes Reinecke [Fri, 2 Dec 2016 10:36:13 +0000 (11:36 +0100)]
scsi: hpsa: fallback to use legacy REPORT PHYS command

Older SmartArray controllers (eg SmartArray 64xx) do not support the
extended REPORT PHYS command, so fallback to use the legacy version
here.

Signed-off-by: Hannes Reinecke <hare@suse.com>
Acked-by: Don Brace <don.brace@microsemi.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: scsi_dh_alua: Fix RCU annotations
Bart Van Assche [Thu, 1 Dec 2016 19:30:50 +0000 (11:30 -0800)]
scsi: scsi_dh_alua: Fix RCU annotations

This patch avoids that sparse complains about RCU pointer dereferences.

Signed-off-by: Bart Van Assche <bart.vanassche@sandisk.com>
Cc: Hannes Reinecke <hare@suse.de>
Cc: tang.junhui <tang.junhui@zte.com.cn>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: hpsa: use %phN for short hex dumps
Rasmus Villemoes [Wed, 30 Nov 2016 22:35:47 +0000 (23:35 +0100)]
scsi: hpsa: use %phN for short hex dumps

Passing one instead of 8 or 16 arguments reduces the size of the
generated code somewhat:

add/remove: 2/3 grow/shrink: 1/4 up/down: 1772/-2137 (-365)

There's one more candidate, unique_id_show, but that uses %02X, and I'm
not sure it would be ok to start using lowercase there, so I've left it
alone for now.

Signed-off-by: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Acked-by: Don Brace <don.brace@microsemi.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: hisi_sas: fix free'ing in probe and remove
Xiaofei Tan [Tue, 29 Nov 2016 15:45:57 +0000 (23:45 +0800)]
scsi: hisi_sas: fix free'ing in probe and remove

This patch addresses 4 problems in the module probe/remove:

 - When hisi_sas_shost_alloc() fails after we alloc shost memory, we
   should free shost memory before the function returns.

 - When hisi_sas_probe() fails after we alloc the HBA memories, we
   should also free the HBA memories.

 - We should free shost memory at the end of hisi_sas_remove().

 - sha->core.shost is set twice, so remove extra set.

Signed-off-by: Xiaofei Tan <tanxiaofei@huawei.com>
Signed-off-by: John Garry <john.garry@huawei.com>
Reviewed-by: Quentin Lambert <lambert.quentin@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: isci: switch to pci_alloc_irq_vectors
Christoph Hellwig [Wed, 30 Nov 2016 16:21:01 +0000 (17:21 +0100)]
scsi: isci: switch to pci_alloc_irq_vectors

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Artur Paszkiewicz <artur.paszkiewicz@intel.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: ipr: Fix runaway IRQs when falling back from MSI to LSI
Benjamin Herrenschmidt [Wed, 30 Nov 2016 21:28:55 +0000 (15:28 -0600)]
scsi: ipr: Fix runaway IRQs when falling back from MSI to LSI

LSIs must be ack'ed with an MMIO otherwise they remain asserted
forever. This is controlled by the "clear_isr" flag.

While we set that flag properly when deciding initially whether to use
LSIs or MSIs, we fail to set it if we first chose MSIs, the test fails,
then fallback to LSIs.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Brian King <brking@linux.vnet.ibm.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: dpt_i2o: double free on error path
Dan Carpenter [Wed, 30 Nov 2016 19:36:48 +0000 (22:36 +0300)]
scsi: dpt_i2o: double free on error path

We recently introduced a kfree() in the caller for this function.
That's where, logically, you would think the kfree() should be.
Unfortunately the code was just ugly and not buggy so the static checker
warning was a false postive and introduced a double free.

I've removed the old kfree() and left the new one.

Fixes: 021e2927586d ("scsi: dpt_i2o: Add a missing call to kfree")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Reviewed-by: Quentin Lambert <lambert.quentin@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: cxlflash: Migrate scsi command pointer to AFU command
Matthew R. Ochs [Tue, 29 Nov 2016 00:43:18 +0000 (18:43 -0600)]
scsi: cxlflash: Migrate scsi command pointer to AFU command

Currently, when sending a SCSI command, the pointer is stored in a
reserved field of the AFU command descriptor for retrieval once the
SCSI command has completed. In order to support new descriptor formats
that make use of the reserved field, the pointer is migrated to outside
the descriptor where it can still be found during completion processing.

Signed-off-by: Matthew R. Ochs <mrochs@linux.vnet.ibm.com>
Acked-by: Uma Krishnan <ukrishn@linux.vnet.ibm.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: cxlflash: Migrate IOARRIN specific routines to function pointers
Matthew R. Ochs [Tue, 29 Nov 2016 00:43:09 +0000 (18:43 -0600)]
scsi: cxlflash: Migrate IOARRIN specific routines to function pointers

As staging for supporting hardware with a different queuing mechanism,
move the send_cmd() and context_reset() routines to function pointers
that are configured when the AFU is initialized. In addition, rename
the existing routines to better reflect the queue model they support.

Signed-off-by: Matthew R. Ochs <mrochs@linux.vnet.ibm.com>
Acked-by: Uma Krishnan <ukrishn@linux.vnet.ibm.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: cxlflash: Cleanup queuecommand()
Matthew R. Ochs [Tue, 29 Nov 2016 00:43:01 +0000 (18:43 -0600)]
scsi: cxlflash: Cleanup queuecommand()

The queuecommand routine is disorganized where it populates the
private command and also contains some logic/statements that are
not needed given that cxlflash devices do not (and likely never
will) support scatter-gather.

Restructure the code to remove the unnecessary logic and create an
organized flow:

handle state -> DMA map -> populate command -> send command

Signed-off-by: Matthew R. Ochs <mrochs@linux.vnet.ibm.com>
Acked-by: Uma Krishnan <ukrishn@linux.vnet.ibm.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: cxlflash: Cleanup send_tmf()
Matthew R. Ochs [Tue, 29 Nov 2016 00:42:50 +0000 (18:42 -0600)]
scsi: cxlflash: Cleanup send_tmf()

The send_tmf() routine includes some copy/paste cruft that can be
removed as well as the setting of an AFU command-specific while
holding the tmf_slock. While not a bug, it is out of place and
should be shifted down alongside the other command initialization
statements for clarity.

Signed-off-by: Matthew R. Ochs <mrochs@linux.vnet.ibm.com>
Acked-by: Uma Krishnan <ukrishn@linux.vnet.ibm.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: cxlflash: Remove AFU command lock
Matthew R. Ochs [Tue, 29 Nov 2016 00:42:42 +0000 (18:42 -0600)]
scsi: cxlflash: Remove AFU command lock

The original design of the cxlflash driver required AFU commands
to convey state information across multiple threads. The IOASA
"host use" byte was used to track if a command was done, errored,
or timed out. A per-command spin lock was used to serialize access
to this byte. As this is no longer required with the introduction
of completions and various refactoring over time, the spin lock,
state tracking, and associated code can be removed. To support the
simplification, the wait_resp() routine is refactored to return a
success or failure. Additionally, as the simplification to the
AFU internal command routine, explicit assignments of AFU command
fields to zero are removed as the memory is zeroed upon allocation.

Signed-off-by: Matthew R. Ochs <mrochs@linux.vnet.ibm.com>
Acked-by: Uma Krishnan <ukrishn@linux.vnet.ibm.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: cxlflash: Wait for active AFU commands to timeout upon tear down
Matthew R. Ochs [Tue, 29 Nov 2016 00:42:33 +0000 (18:42 -0600)]
scsi: cxlflash: Wait for active AFU commands to timeout upon tear down

With the removal of the static private command pool, the ability to
'complete' outstanding commands was lost. While not an issue for the
commands originating outside the driver, internal AFU commands are
synchronous and therefore have a timeout associated with them. To
avoid a stale memory access, the tear down sequence needs to ensure
that there are not any active commands before proceeding. As these
internal AFU commands are rare events, the simplest way to accomplish
this is detecting the activity and waiting for it to timeout.

Signed-off-by: Matthew R. Ochs <mrochs@linux.vnet.ibm.com>
Acked-by: Uma Krishnan <ukrishn@linux.vnet.ibm.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: cxlflash: Remove private command pool
Matthew R. Ochs [Tue, 29 Nov 2016 00:42:27 +0000 (18:42 -0600)]
scsi: cxlflash: Remove private command pool

Clean up and remove the remaining private command pool infrastructure
that is no longer required.

Signed-off-by: Matthew R. Ochs <mrochs@linux.vnet.ibm.com>
Acked-by: Uma Krishnan <ukrishn@linux.vnet.ibm.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: cxlflash: Use cmd_size for private commands
Matthew R. Ochs [Tue, 29 Nov 2016 00:42:19 +0000 (18:42 -0600)]
scsi: cxlflash: Use cmd_size for private commands

Instead of using a private pool of AFU commands, use cmd_size to prime
the private pool of SCSI commands such that they are allocated with a
size large enough to contain an aligned AFU command. Use scsi_cmd_priv()
to derive the aligned/zeroed private command on queuecommand and TMF
paths. Remove cmd_checkout() as it is no longer required. The remaining
AFU private command infrastructure will be removed in a cleanup commit.

Signed-off-by: Matthew R. Ochs <mrochs@linux.vnet.ibm.com>
Acked-by: Uma Krishnan <ukrishn@linux.vnet.ibm.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: cxlflash: Allocate memory instead of using command pool for AFU sync
Matthew R. Ochs [Tue, 29 Nov 2016 00:42:11 +0000 (18:42 -0600)]
scsi: cxlflash: Allocate memory instead of using command pool for AFU sync

As staging for the removal of the AFU command pool, remove the reliance
upon the pool for the internal AFU sync command. Instead of obtaining an
AFU command from the pool, dynamically allocate memory with the appropriate
alignment requirements. Since the AFU sync service is only executed from
the process environment, blocking is acceptable.

Signed-off-by: Matthew R. Ochs <mrochs@linux.vnet.ibm.com>
Acked-by: Uma Krishnan <ukrishn@linux.vnet.ibm.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: cxlflash: Remove unused buffer from AFU command
Matthew R. Ochs [Tue, 29 Nov 2016 00:42:01 +0000 (18:42 -0600)]
scsi: cxlflash: Remove unused buffer from AFU command

The cxlflash driver originally required a per-command 4K buffer that
hosted data passed to the AFU. When the routines that initiate AFU
and internal SCSI commands were refactored to use scsi_execute(), the
need for this buffer became obsolete. As it is no longer necessary,
the buffer is removed.

Signed-off-by: Matthew R. Ochs <mrochs@linux.vnet.ibm.com>
Acked-by: Uma Krishnan <ukrishn@linux.vnet.ibm.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: lpfc: Replace pci_pool_alloc by pci_pool_zalloc
Souptick Joarder [Mon, 28 Nov 2016 09:52:37 +0000 (15:22 +0530)]
scsi: lpfc: Replace pci_pool_alloc by pci_pool_zalloc

In lpfc_new_scsi_buf_s3() and lpfc_new_scsi_buf_s4() pci_pool_alloc
followed by memset will be replaced by pci_pool_zalloc()

Signed-off-by: Souptick joarder <jrdr.linux@gmail.com>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Acked-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: storvsc: Payload buffer incorrectly sized for 32 bit kernels.
Cathy Avery [Wed, 23 Nov 2016 13:46:33 +0000 (08:46 -0500)]
scsi: storvsc: Payload buffer incorrectly sized for 32 bit kernels.

On a 32 bit kernel sizeof(void *) is not 64 bits as hv_mpb_array
requires. Also the buffer needs to be cleared or the upper bytes will
contain junk.

Suggested-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: Cathy Avery <cavery@redhat.com>
Reviewed-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: cxgb4i: Add a missing call to neigh_release
Quentin Lambert [Sat, 19 Nov 2016 17:42:14 +0000 (18:42 +0100)]
scsi: cxgb4i: Add a missing call to neigh_release

Most error branches following the call to dst_neigh_lookup contain a
call to neigh_release. This patch add these calls where they are
missing.

This issue was found with Hector.

Signed-off-by: Quentin Lambert <lambert.quentin@gmail.com>
Acked-by: Varun Prakash <varun@chelsio.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: cxlflash: Avoid command room violation
Uma Krishnan [Tue, 29 Nov 2016 00:41:45 +0000 (18:41 -0600)]
scsi: cxlflash: Avoid command room violation

During test, a command room violation interrupt is occasionally seen
for the master context when the CXL flash devices are stressed.

After studying the code, there could be gaps in the way command room
value is being cached in cxlflash. When the cached command room is zero
the thread attempting to send becomes burdened with updating the cached
value with the actual value from the AFU. Today, this is handled with an
atomic set operation of the raw value read. Following the atomic update,
the thread proceeds to send.

This behavior is incorrect on two counts:

   - The update fails to take into account the current thread and its
     consumption of one of the hardware commands.

   - The update does not take into account other threads also atomically
     updating. Per design, a worker thread updates the cached value when a
     send thread times out. By not protecting the update with a lock, the
     cached value can be incorrectly clobbered.

To correct these issues, the update of the cached command room has been
simplified and also protected using a spin lock which is held until the
MMIO is complete. This ensures the command room is properly consumed by
the same thread. Update of cached value also takes into account the
current thread consuming a hardware command.

Signed-off-by: Uma Krishnan <ukrishn@linux.vnet.ibm.com>
Acked-by: Matthew R. Ochs <mrochs@linux.vnet.ibm.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: cxlflash: Improve context_reset() logic
Uma Krishnan [Tue, 29 Nov 2016 00:41:36 +0000 (18:41 -0600)]
scsi: cxlflash: Improve context_reset() logic

Currently, the context reset routine waits for command room to
be available before sending the reset request. Per review of the
SISLite specification and clarifications from the CXL Flash AFU
designers, this wait is unnecessary. The reset request can be
sent anytime regardless of command room, so long as only a single
reset request is active at any one point in time.

This commit simplifies the reset routine by removing the wait for
command room. Additionally it adds a debug trace to help pinpoint
hardware errors when a context reset does not complete.

Signed-off-by: Uma Krishnan <ukrishn@linux.vnet.ibm.com>
Acked-by: Matthew R. Ochs <mrochs@linux.vnet.ibm.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: cxlflash: Fix crash in cxlflash_restore_luntable()
Uma Krishnan [Tue, 29 Nov 2016 00:41:19 +0000 (18:41 -0600)]
scsi: cxlflash: Fix crash in cxlflash_restore_luntable()

During test, the following crash was observed:

[34538.981505] Faulting instruction address: 0xd000000007c9c870
cpu 0x9: Vector: 300 (Data Access) at [c0000007f1e8f590]
    pc: d000000007c9c870: cxlflash_restore_luntable+0x70/0x1d0 [cxlflash]
    lr: d000000007c9c84c: cxlflash_restore_luntable+0x4c/0x1d0 [cxlflash]
    sp: c0000007f1e8f810
   msr: 9000000100009033
   dar: c00000171d637438
 dsisr: 40000000
  current = 0xc0000007f1e43f90
  paca    = 0xc000000007b25100   softe: 0        irq_happened: 0x01
    pid   = 493, comm = eehd
enter ? for help
[c0000007f1e8f8a0d000000007c940b0 init_afu+0xd60/0x1200 [cxlflash]
[c0000007f1e8f9a0d000000007c945a8 cxlflash_pci_slot_reset+0x58/0xe0 [cxlflash]
[c0000007f1e8fa20d00000000715f790 cxl_pci_slot_reset+0x230/0x340 [cxl]
[c0000007f1e8fae0c000000000040dd4 eeh_report_reset+0x144/0x180
[c0000007f1e8fb20c00000000003f708 eeh_pe_dev_traverse+0x98/0x170
[c0000007f1e8fbb0c000000000041618 eeh_handle_normal_event+0x328/0x410
[c0000007f1e8fc30c000000000041db8 eeh_handle_event+0x178/0x330
[c0000007f1e8fce0c000000000042118 eeh_event_handler+0x1a8/0x1b0
[c0000007f1e8fd80c00000000011420c kthread+0xec/0x100
[c0000007f1e8fe30c00000000000a47c ret_from_kernel_thread+0x5c/0xe0

When superpipe mode is disabled for a LUN, the references for the
local lun are deleted but the LUN is still identified as being present
in the LUN table. This mismatched state can result in the above crash
when the LUN table is restored during an error recovery operation.

To fix this issue, the local LUN information structure is updated to
reflect the LUN is no longer in the LUN table once all references to
the LUN are gone.

Signed-off-by: Uma Krishnan <ukrishn@linux.vnet.ibm.com>
Acked-by: Matthew R. Ochs <mrochs@linux.vnet.ibm.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: cxlflash: Set sg_tablesize to 1 instead of SG_NONE
Uma Krishnan [Tue, 29 Nov 2016 00:41:06 +0000 (18:41 -0600)]
scsi: cxlflash: Set sg_tablesize to 1 instead of SG_NONE

The following Oops is encountered when blk_mq is enabled with the
cxlflash driver:

[ 2960.817172] Oops: Kernel access of bad area, sig: 11 [#5]
[ 2960.817309] NIP  __blk_mq_run_hw_queue+0x278/0x4c0
[ 2960.817313] LR __blk_mq_run_hw_queue+0x2bc/0x4c0
[ 2960.817314] Call Trace:
[ 2960.817320] __blk_mq_run_hw_queue+0x2bc/0x4c0 (unreliable)
[ 2960.817324] blk_mq_run_hw_queue+0xd8/0x100
[ 2960.817329] blk_mq_insert_requests+0x14c/0x1f0
[ 2960.817333] blk_mq_flush_plug_list+0x150/0x190
[ 2960.817338] blk_flush_plug_list+0x11c/0x2b0
[ 2960.817344] blk_finish_plug+0x58/0x80
[ 2960.817348] __do_page_cache_readahead+0x1c0/0x2e0
[ 2960.817352] force_page_cache_readahead+0x68/0xd0
[ 2960.817356] generic_file_read_iter+0x43c/0x6a0
[ 2960.817359] blkdev_read_iter+0x68/0xa0
[ 2960.817361] __vfs_read+0x11c/0x180
[ 2960.817364] vfs_read+0xa4/0x1c0
[ 2960.817366] SyS_read+0x6c/0x110
[ 2960.817369] system_call+0x38/0xb4

The SCSI blk_mq stack assumes that sg_tablesize is always a non-zero
value with scsi_mq_setup_tags() allocating tags using sg_tablesize.
The cxlflash driver currently uses SG_NONE (0) for the sg_tablesize
as the devices it supports are not capable of scatter gather. This
mismatch of values results in the Oops above.

To resolve this issue, sg_tablesize for cxlflash can simply be set
to 1, a value which satisfies the constraints in cxlflash and the
lack of support of SG_NONE in SCSI blk_mq.

Signed-off-by: Uma Krishnan <ukrishn@linux.vnet.ibm.com>
Acked-by: Matthew R. Ochs <mrochs@linux.vnet.ibm.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: ufs: fix default power mode to FAST/SLOW
subhashj@codeaurora.org [Thu, 24 Nov 2016 00:33:19 +0000 (16:33 -0800)]
scsi: ufs: fix default power mode to FAST/SLOW

We would by default like to run in FAST/SLOW mode instead
of FASTAUTO/SLOWAUTO mode for performance reasons. This
change sets the default speed mode to FAST/SLOW mode.

Reviewed-by: Venkat Gopalakrishnan <venkatg@codeaurora.org>
Signed-off-by: Subhash Jadavani <subhashj@codeaurora.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: ufs: optimize system suspend handling
subhashj@codeaurora.org [Thu, 24 Nov 2016 00:33:08 +0000 (16:33 -0800)]
scsi: ufs: optimize system suspend handling

Consider following sequence of events:
1. UFS is runtime suspended, link_state = Hibern8, device_state = sleep
2. System goes into system suspend, ufshcd_system_suspend() brings both
   link and device to active state and then puts the device in Power_Down
   state and link in OFF state.
3. System resumes at some later point in time, ufshcd_system_resume()
   doesn't do anything as UFS state is runtime suspended. Note that link
   is still on OFF state and device is in Power_Down state.
4. Now system again goes into suspend without any UFS accesses before it.
   ufshcd_system_suspend() again brings both link and device to active
   state and then puts the device in Power_Down state and link if OFF
   state. But it's unnecessary to bring the link & device in active state
   as both link and device are already in desired low power states. This
   change fixes this issue by adding proper state checks in
   ufshcd_system_suspend().

Reviewed-by: Gilad Broner <gbroner@codeaurora.org>
Signed-off-by: Subhash Jadavani <subhashj@codeaurora.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: ufs: fix condition in which DME command failure msg is printed out
Yaniv Gardi [Thu, 24 Nov 2016 00:32:49 +0000 (16:32 -0800)]
scsi: ufs: fix condition in which DME command failure msg is printed out

The condition in which error message is printed out was incorrect and
resulted error message only if retries exhausted.
But retries happens only if DME command is a peer command, and thus
DME commands which are not peer commands and fail are not printed out.
This change fixes this issue.

Signed-off-by: Yaniv Gardi <ygardi@codeaurora.org>
Signed-off-by: Subhash Jadavani <subhashj@codeaurora.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: ufs: handle errors from PHY_ADAPTER_ERROR register
Dolev Raviv [Thu, 24 Nov 2016 00:32:32 +0000 (16:32 -0800)]
scsi: ufs: handle errors from PHY_ADAPTER_ERROR register

The PHY_ADAPTER_ERROR status register indicates PHY lane errors
reported by the M-PHY layer. In some occasions the controller
can recover from such errors. When the error is not recoverable,
a stuck DB error will occur. Since the stuck DB error is spotted
separately, no action other than clearing the register is necessary.

Signed-off-by: Dolev Raviv <draviv@codeaurora.org>
Signed-off-by: Subhash Jadavani <subhashj@codeaurora.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: ufs: issue link starup 2 times if device isn't active
subhashj@codeaurora.org [Thu, 24 Nov 2016 00:32:20 +0000 (16:32 -0800)]
scsi: ufs: issue link starup 2 times if device isn't active

If we issue the link startup to the device while its UniPro state is
LinkDown (and device state is sleep/power-down) then link startup
will not move the device state to Active. Device will only move to
active state if the link starup is issued when its UniPro state is
LinkUp. So in this case, we would have to issue the link startup 2
times to make sure that device moves to active state.

Reviewed-by: Gilad Broner <gbroner@codeaurora.org>
Signed-off-by: Subhash Jadavani <subhashj@codeaurora.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: ufs: ensure that host pa_tactivate is higher than device
subhashj@codeaurora.org [Thu, 24 Nov 2016 00:32:08 +0000 (16:32 -0800)]
scsi: ufs: ensure that host pa_tactivate is higher than device

Some UFS devices require host PA_TACTIVATE to be higher than
device PA_TACTIVATE otherwise it may get stuck during hibern8 sequence.
This change allows this by using quirk.

Reviewed-by: Venkat Gopalakrishnan <venkatg@codeaurora.org>
Signed-off-by: Subhash Jadavani <subhashj@codeaurora.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: ufs: increase the scsi query response timeout
subhashj@codeaurora.org [Thu, 24 Nov 2016 00:31:52 +0000 (16:31 -0800)]
scsi: ufs: increase the scsi query response timeout

It is found thats UFS device may take longer than 30ms to respond to
query requests and in this case we might run into following scenario:

1. UFS host SW sends a query request to UFS device to read an attribute
   value. SW uses tag #31 for this purpose.
2. UFS host SW waits for 30ms to get the query response (and doorbell
   to be cleared by UFS host HW).
3. UFS device doesn't respond back within 30ms hence UFS host SW times
   out waiting for the query response.
4. UFS host SW clears the tag#31 from UTRLCLR register.
5. UFS host SW waits until UFS host HW to clear tag#31 from the doorbell
   register.
6. UFS host SW retries the same query request on same tag#31 (sends a query
   request to device to read an attribute value).
7. UFS host HW gets the query response from the device but this was
   intended as a query response for the 1st query request sent (step-1).
8. Now UFS device sends another query response to host (for query request
   sent @step-6).

Now there are 2 issues that could happen with above scenario:
1. UFS device should have actually responded back with only one query
   response but it is found that device may respond back with 2 query
   responses.
2. If UFS device responds back with 2 resposes on same tag, host HW/SW
   behaviour isn't predictable.

To avoid running into above scenario, we would basically allow device
to take longer (upto 1.5 seconds) for query response.

Reviewed-by: Gilad Broner <gbroner@codeaurora.org>
Signed-off-by: Subhash Jadavani <subhashj@codeaurora.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: ufs: fix failure to read the string descriptor
subhashj@codeaurora.org [Thu, 24 Nov 2016 00:31:41 +0000 (16:31 -0800)]
scsi: ufs: fix failure to read the string descriptor

While reading variable size descriptors (like string descriptor), some UFS
devices may report the "LENGTH" (field in "Transaction Specific fields" of
Query Response UPIU) same as what was requested in Query Request UPIU
instead of reporting the actual size of the variable size descriptor.
Although it's safe to ignore the "LENGTH" field for variable size
descriptors as we can always derive the length of the descriptor from
the descriptor header fields. Hence this change impose the length match
check only for fixed size descriptors (for which we always request the
correct size as part of Query Request UPIU).

Reviewed-by: Venkat Gopalakrishnan <venkatg@codeaurora.org>
Signed-off-by: Subhash Jadavani <subhashj@codeaurora.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: ufs: update device descriptor maximum size
Yaniv Gardi [Thu, 24 Nov 2016 00:31:30 +0000 (16:31 -0800)]
scsi: ufs: update device descriptor maximum size

According to JESD220B - UFS v2.0, the maximum size of device descriptor
has changed from 0x1F to 0x40. This patch updates the maximum size of
this descriptor.

Signed-off-by: Yaniv Gardi <ygardi@codeaurora.org>
Signed-off-by: Subhash Jadavani <subhashj@codeaurora.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: ufs: add index details to query error messages
Yaniv Gardi [Thu, 24 Nov 2016 00:31:18 +0000 (16:31 -0800)]
scsi: ufs: add index details to query error messages

When sending query to the device, the index  of the failure
is additional useful information that should be printed out as it
might specify the logical unit (LU) where the error occurred.

Signed-off-by: Yaniv Gardi <ygardi@codeaurora.org>
Signed-off-by: Subhash Jadavani <subhashj@codeaurora.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: ufs: add queries retry mechanism
Dolev Raviv [Thu, 24 Nov 2016 00:30:49 +0000 (16:30 -0800)]
scsi: ufs: add queries retry mechanism

Some of the queries might fail during init. To avoid
system failure, we add retry mechanism to issue queries
several times.

Signed-off-by: Dolev Raviv <draviv@codeaurora.org>
Signed-off-by: Subhash Jadavani <subhashj@codeaurora.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: dpt_i2o: Add a missing call to kfree
Quentin Lambert [Sat, 19 Nov 2016 17:42:34 +0000 (18:42 +0100)]
scsi: dpt_i2o: Add a missing call to kfree

Most error branches following the call to kzalloc contain a call to
kfree. This patch add these calls where they are missing.

This issue was found with Hector.

Signed-off-by: Quentin Lambert <lambert.quentin@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: isci: Add a missing call to pci_unmap_biosrom
Quentin Lambert [Sat, 19 Nov 2016 17:43:40 +0000 (18:43 +0100)]
scsi: isci: Add a missing call to pci_unmap_biosrom

Most error branches following the call to pci_map_biosrom contain a call
to pci_unmap_biosrom. This patch add these calls where they are missing.

This issue was found with Hector.

Signed-off-by: Quentin Lambert <lambert.quentin@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: isci: avoid array subscript warning
Arnd Bergmann [Fri, 18 Nov 2016 16:14:01 +0000 (17:14 +0100)]
scsi: isci: avoid array subscript warning

I'm getting a new warning with gcc-7:

isci/remote_node_context.c: In function 'sci_remote_node_context_destruct':
isci/remote_node_context.c:69:16: error: array subscript is above array bounds [-Werror=array-bounds]

This is odd, since we clearly cover all values for enum
scis_sds_remote_node_context_states here. Anyway, checking for an array
overflow can't harm and it makes the warning go away.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: hisi_sas: shift vs compare typos
Dan Carpenter [Tue, 29 Nov 2016 10:47:25 +0000 (13:47 +0300)]
scsi: hisi_sas: shift vs compare typos

There are some typos where we intended "<<" but have "<".  Seems likely
to cause a bunch of problems.

Fixes: d3b688d3c69d ("scsi: hisi_sas: add v2 hw support for ECC and AXI bus fatal error")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Acked-by: John Garry <john.garry@huawei.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: aic94xx: Add a missing call to kfree
Quentin Lambert [Fri, 25 Nov 2016 12:23:51 +0000 (13:23 +0100)]
scsi: aic94xx: Add a missing call to kfree

Most error branches following the call to kzalloc contain a call to
kfree. This patch add these calls where they are missing and set the
relevant pointers to NULL.

This issue was found with Hector.

Signed-off-by: Quentin Lambert <lambert.quentin@gmail.com>
Reviewed-by: Tomas Henzl <thenzl@redhat.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: hpsa: add 'ctlr_num' sysfs attribute
Hannes Reinecke [Fri, 18 Nov 2016 07:32:49 +0000 (08:32 +0100)]
scsi: hpsa: add 'ctlr_num' sysfs attribute

Add a sysfs attribute 'ctlr_num' holding the current HPSA controller
number. This is required to construct compability 'cciss' links.

[mkp: fixed typo]

Signed-off-by: Hannes Reinecke <hare@suse.com>
Acked-by: Don Brace <don.brace@microsemi.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: hpsa: use correct DID_NO_CONNECT hostbyte
Hannes Reinecke [Fri, 18 Nov 2016 07:32:47 +0000 (08:32 +0100)]
scsi: hpsa: use correct DID_NO_CONNECT hostbyte

NOT_READY is a sense key, not a legit scsi hostbyte value. Use
DID_NO_CONNECT instead.

Signed-off-by: Hannes Reinecke <hare@suse.com>
Acked-by: Don Brace <don.brace@microsemi.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: srp_transport: Move queuecommand() wait code to SCSI core
Bart Van Assche [Wed, 23 Nov 2016 00:17:13 +0000 (16:17 -0800)]
scsi: srp_transport: Move queuecommand() wait code to SCSI core

Additionally, rename srp_wait_for_queuecommand() into
scsi_wait_for_queuecommand() and add a comment about the queuecommand()
call from scsi_send_eh_cmnd().

Note: this patch changes scsi_internal_device_block from a function that
did not sleep into a function that may sleep. This is fine for all
callers of this function:

* scsi_internal_device_block() is called from the mpt3sas device while
  that driver holds the ioc->dm_cmds.mutex. This means that the mpt3sas
  driver calls this function from thread context.
* scsi_target_block() is called by __iscsi_block_session() from
  kernel thread context and with IRQs enabled.
* The SRP transport code also calls scsi_target_block() from kernel
  thread context while sleeping is allowed.
* The snic driver also calls scsi_target_block() from a context from
  which sleeping is allowed. The scsi_target_block() call namely occurs
  immediately after a scsi_flush_work() call.

[mkp: s/shost/sdev/]

Signed-off-by: Bart Van Assche <bart.vanassche@sandisk.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Cc: James Bottomley <jejb@linux.vnet.ibm.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Doug Ledford <dledford@redhat.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: libfc: Remove an unneeded condition
Dan Carpenter [Thu, 24 Nov 2016 10:52:38 +0000 (13:52 +0300)]
scsi: libfc: Remove an unneeded condition

We verified that resp_code is FC_SPP_RESP_ACK earlier so we don't need
to check again here.

Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Acked-by: Johannes Thumshirn <jth@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: ufs: Add missing UFS_MASK macro definition
Santosh Y [Thu, 24 Nov 2016 04:58:51 +0000 (12:58 +0800)]
scsi: ufs: Add missing UFS_MASK macro definition

Reported-by: Venkatraman S <svenkatr@ti.com>
Reviewed-by: Vinayak Holikatti <vinholikatti@gmail.com>
Signed-off-by: Santosh Y <santoshsy@gmail.com>
Signed-off-by: Zhangfei Gao <zhangfei.gao@linaro.org>
Reviewed-by: Subhash Jadavani <subhashj@codeaurora.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: hisi_sas: add PHY set linkrate support for v1 and v2 hw
Xiang Chen [Mon, 7 Nov 2016 12:48:40 +0000 (20:48 +0800)]
scsi: hisi_sas: add PHY set linkrate support for v1 and v2 hw

Add the function to set PHY min and max linkrate through
sysfs interface.

Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com>
Signed-off-by: John Garry <john.garry@huawei.com>
Reviewed-by: Zhangfei Gao <zhangfei.gao@linaro.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: hisi_sas: use atomic64_t for hisi_sas_device.running_req
John Garry [Mon, 7 Nov 2016 12:48:39 +0000 (20:48 +0800)]
scsi: hisi_sas: use atomic64_t for hisi_sas_device.running_req

Sometimes the value of hisi_sas_device.running_req
would go negative unless we have the check for
running_req >= 0 before trying to decrement.

This is because using running_req is not thread-safe.

As such, the value for running_req may be actually incorrect,
so use atomic64_t instead.

Signed-off-by: John Garry <john.garry@huawei.com>
Reviewed-by: Xiang Chen <chenxiang66@hisilicon.com>
Reviewed-by: Zhangfei Gao <zhangfei.gao@linaro.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: hisi_sas: check SATA FIS when directly attaching SATA device
Xiang Chen [Mon, 7 Nov 2016 12:48:38 +0000 (20:48 +0800)]
scsi: hisi_sas: check SATA FIS when directly attaching SATA device

Check ERR bit of status to decide whether there is something wrong with
initial register-D2H FIS. If error exists, PHY reset the channel to
restart OOB.

Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com>
Signed-off-by: John Garry <john.garry@huawei.com>
Reviewed-by: Zhangfei Gao <zhangfei.gao@linaro.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: hisi_sas: modify some values in get_ata_protocol()
Xiang Chen [Mon, 7 Nov 2016 12:48:37 +0000 (20:48 +0800)]
scsi: hisi_sas: modify some values in get_ata_protocol()

Modify and add some SATA commands according to SATA protocol.

Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com>
Signed-off-by: John Garry <john.garry@huawei.com>
Reviewed-by: Zhangfei Gao <zhangfei.gao@linaro.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: hisi_sas: delete repeated configuration in free_device_v2_hw()
Xiang Chen [Mon, 7 Nov 2016 12:48:36 +0000 (20:48 +0800)]
scsi: hisi_sas: delete repeated configuration in free_device_v2_hw()

Delete repeated configuration items for hisi_sas_device() when
we free a device. These items are now only set in
hisi_sas_dev_gone().

Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com>
Signed-off-by: John Garry <john.garry@huawei.com>
Reviewed-by: Zhangfei Gao <zhangfei.gao@linaro.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: hisi_sas: modify return value of hisi_sas_query_task()
Xiang Chen [Mon, 7 Nov 2016 12:48:35 +0000 (20:48 +0800)]
scsi: hisi_sas: modify return value of hisi_sas_query_task()

sas_scsi_find_task() only deals with return value
TMF_RESP_FUNC_FAILED/TMF_RESP_FUNC_SUCC/TMF_RESP_FUNC_COMPLETE of
query task. So for LLDD errors just return TMF_RESP_FUNC_FAILED.

Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com>
Signed-off-by: John Garry <john.garry@huawei.com>
Reviewed-by: Zhangfei Gao <zhangfei.gao@linaro.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: hisi_sas: replace WARN_ON() with dev_warn() for internal abort
Xiang Chen [Mon, 7 Nov 2016 12:48:34 +0000 (20:48 +0800)]
scsi: hisi_sas: replace WARN_ON() with dev_warn() for internal abort

Replace WARN_ON() with dev_warn() print when internal abort fails.

Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com>
Signed-off-by: John Garry <john.garry@huawei.com>
Reviewed-by: Zhangfei Gao <zhangfei.gao@linaro.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: hisi_sas: fix port form bug in hisi_sas_port_notify_formed()
Xiang Chen [Mon, 7 Nov 2016 12:48:33 +0000 (20:48 +0800)]
scsi: hisi_sas: fix port form bug in hisi_sas_port_notify_formed()

When we form a wideport, we should use hardware PHY port_id instead
of sas_phy->id.

Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com>
Signed-off-by: John Garry <john.garry@huawei.com>
Reviewed-by: Zhangfei Gao <zhangfei.gao@linaro.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: hisi_sas: only process broadcast change in phy_bcast_v2_hw()
Xiang Chen [Mon, 7 Nov 2016 12:48:32 +0000 (20:48 +0800)]
scsi: hisi_sas: only process broadcast change in phy_bcast_v2_hw()

There are many BROADCAST primitives generated by the host.
We are only interested in BROADCAST (CHANGE) primitives currently,
so only process this.

Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com>
Signed-off-by: John Garry <john.garry@huawei.com>
Reviewed-by: Zhangfei Gao <zhangfei.gao@linaro.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: hisi_sas: alloc queue id of slot according to device id
Xiang Chen [Mon, 7 Nov 2016 12:48:31 +0000 (20:48 +0800)]
scsi: hisi_sas: alloc queue id of slot according to device id

Currently slots are allocated from queues in a round-robin fashion.
This causes a problem for internal commands in device mode. For this
mode, we should ensure that the internal abort command is the last
command seen in the host for that device. We can only ensure this when
we place the internal abort command after the preceding commands for
device that in the same queue, as there is no order in which the host
will select a queue to execute the next command.

This queue restriction makes supporting scsi mq more tricky in
the future, but should not be a blocker.

Note: Even though v1 hw does not support internal abort, the
      allocation method is chosen to be the same for consistency.

Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com>
Signed-off-by: John Garry <john.garry@huawei.com>
Reviewed-by: Zhangfei Gao <zhangfei.gao@linaro.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: hisi_sas: add v2 hw support for ECC and AXI bus fatal error
Xiang Chen [Mon, 7 Nov 2016 12:48:30 +0000 (20:48 +0800)]
scsi: hisi_sas: add v2 hw support for ECC and AXI bus fatal error

For ECC 1bit error, logic can recover it, so we only print
a warning.
For ECC multi-bit and AXI bus fatal error, we panic.

Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com>
Signed-off-by: John Garry <john.garry@huawei.com>
Reviewed-by: Zhangfei Gao <zhangfei.gao@linaro.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: pmcraid: Add missing resource releases
Quentin Lambert [Sat, 19 Nov 2016 17:43:18 +0000 (18:43 +0100)]
scsi: pmcraid: Add missing resource releases

Most error branches following the call to pmcraid_get_free_cmd contain a
call to pmcraid_return_cmd. This patch add these calls where they are
missing.

Moreover, most error branches following the call to class_create contain
a call to class_destroy. This patch add these calls where they are
missing.

This issue was found with Hector.

Signed-off-by: Quentin Lambert <lambert.quentin@gmail.com>
Reviewed-by: Tomas Henzl <thenzl@redhat.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: ufs: introduce UFSHCD_QUIRK_PRDT_BYTE_GRAN quirk
Kiwoong Kim [Tue, 22 Nov 2016 08:06:59 +0000 (17:06 +0900)]
scsi: ufs: introduce UFSHCD_QUIRK_PRDT_BYTE_GRAN quirk

Some UFS host controllers may think granularities of PRDT length and
offset as bytes, not double words.

Signed-off-by: Kiwoong Kim <kwmad.kim@samsung.com>
Reviewed-by: Subhash Jadavani <subhashj@codeaurora.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: ufs: qcom: Properly clear hba priv on failure
Bjorn Andersson [Sun, 20 Nov 2016 06:34:51 +0000 (22:34 -0800)]
scsi: ufs: qcom: Properly clear hba priv on failure

ufs_qcom_init() sets the hba priv data before attempting to acquire the
phy handle, so make sure to clear this in the case of an error. Failing
to do this will make ufs_qcom_setup_clocks() operate on the uninitalized
host object.

Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Reviewed-by: Subhash Jadavani <subhashj@codeaurora.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: bfa: turn bfa_mem_{kva,dma}_setup into inline functions
Arnd Bergmann [Wed, 16 Nov 2016 15:14:27 +0000 (16:14 +0100)]
scsi: bfa: turn bfa_mem_{kva,dma}_setup into inline functions

These two macros cause lots of warnings with gcc-7:

drivers/scsi/bfa/bfa_svc.c: In function 'bfa_fcxp_meminfo':
drivers/scsi/bfa/bfa_svc.c:521:103: error: '*' in boolean context, suggest '&&' instead [-Werror=int-in-bool-context]

Using inline functions makes them much more readable and avoids the
warnings.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Acked by: Anil Gurumurthy <anil.gurumurthy@cavium.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: mptfusion: Fix printk continuations
Helge Deller [Thu, 17 Nov 2016 20:46:20 +0000 (21:46 +0100)]
scsi: mptfusion: Fix printk continuations

Fix the printk continuations when running the mptfusion driver.
This patch brings the capabilities into one single syslog line again:
   mptbase: ioc1: Initiating bringup
   ioc1: LSI53C1030 B2: Capabilities={Initiator,Target}
   scsi host3: ioc1: LSI53C1030 B2, FwRev=01032341h, Ports=1, MaxQ=255, IRQ=67

Tested on a parisc C8000 machine.

Signed-off-by: Helge Deller <deller@gmx.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: aacraid: switch to pci_alloc_irq_vectors
Hannes Reinecke [Tue, 8 Nov 2016 07:11:30 +0000 (08:11 +0100)]
scsi: aacraid: switch to pci_alloc_irq_vectors

Use pci_alloc_irq_vectors and drop the hand-crafted interrupt affinity
routines.

Signed-off-by: Hannes Reinecke <hare@suse.com>
Cc: Adaptec OEM Raid Solutions <aacraid@microsemi.com>
Reviewed-by: Raghava Aditya Renukunta <raghavaaditya.renukunta@microsemi.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: fc: move FC transport's bsg code to bsg-lib
Johannes Thumshirn [Thu, 17 Nov 2016 09:31:24 +0000 (10:31 +0100)]
scsi: fc: move FC transport's bsg code to bsg-lib

Now that all conversions are done, move the FibreChannel bsg code over
to the bsg library.

This patch is derived from work done by Mike Christie in 2011 [1] but
only the iscsi parts got merged back then.

[1] http://marc.info/?l=linux-scsi&m=131149780921009&w=2

Signed-off-by: Johannes Thumshirn <jthumshirn@suse.de>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoblock: add bsg_job_put() and bsg_job_get()
Johannes Thumshirn [Thu, 17 Nov 2016 09:31:23 +0000 (10:31 +0100)]
block: add bsg_job_put() and bsg_job_get()

Add bsg_job_put() and bsg_job_get() so don't need to export
bsg_destroy_job() any more.

Signed-off-by: Johannes Thumshirn <jthumshirn@suse.de>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: fc: use bsg_job_done
Johannes Thumshirn [Thu, 17 Nov 2016 09:31:22 +0000 (10:31 +0100)]
scsi: fc: use bsg_job_done

fc_bsg_jobdone() and bsg_job_done() are 1:1 copies now so use the
bsg-lib one instead of the FC private implementation.

Signed-off-by: Johannes Thumshirn <jthumshirn@suse.de>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Acked-by: Tyrel Datwyler <tyreld@linux.vnet.ibm.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: fc: use bsg_softirq_done
Johannes Thumshirn [Thu, 17 Nov 2016 09:31:21 +0000 (10:31 +0100)]
scsi: fc: use bsg_softirq_done

bsg_softirq_done() and fc_bsg_softirq_done() are copies of each other, so
ditch the fc specific one.

Signed-off-by: Johannes Thumshirn <jthumshirn@suse.de>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: fc: Use bsg_destroy_job
Johannes Thumshirn [Thu, 17 Nov 2016 09:31:20 +0000 (10:31 +0100)]
scsi: fc: Use bsg_destroy_job

fc_destroy_bsgjob() and bsg_destroy_job() are now 1:1 copies, so use the
latter. As bsg_destroy_job() comes from bsg-lib we need to select it in
Kconfig once CONFOG_SCSI_FC_ATTRS is active.

Signed-off-by: Johannes Thumshirn <jthumshirn@suse.de>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: change FC drivers to use 'struct bsg_job'
Johannes Thumshirn [Thu, 17 Nov 2016 09:31:19 +0000 (10:31 +0100)]
scsi: change FC drivers to use 'struct bsg_job'

Change FC drivers to use 'struct bsg_job' from bsg-lib.h instead of
'struct fc_bsg_job' from scsi_transport_fc.h and remove 'struct
fc_bsg_job'.

Signed-off-by: Johannes Thumshirn <jthumshirn@suse.de>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Acked-by: Tyrel Datwyler <tyreld@linux.vnet.ibm.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoblock: add reference counting for struct bsg_job
Johannes Thumshirn [Thu, 17 Nov 2016 09:31:18 +0000 (10:31 +0100)]
block: add reference counting for struct bsg_job

Add reference counting to 'struct bsg_job' so we can implement a reuqest
timeout handler for bsg_jobs, which is needed for Fibre Channel.

Signed-off-by: Johannes Thumshirn <jthumshirn@suse.de>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: fc: implement kref backed reference counting
Johannes Thumshirn [Thu, 17 Nov 2016 09:31:17 +0000 (10:31 +0100)]
scsi: fc: implement kref backed reference counting

Implement kref backed reference counting instead of rolling our own. This
elimnates the need of the following fields in 'struct fc_bsg_job':
* ref_cnt
* state_flags
* job_lock
bringing us close to unification of 'struct fc_bsg_job' and 'struct bsg_job'.

Signed-off-by: Johannes Thumshirn <jthumshirn@suse.de>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: libfc: don't set FC_RQST_STATE_DONE before calling fc_bsg_jobdone()
Johannes Thumshirn [Thu, 17 Nov 2016 09:31:16 +0000 (10:31 +0100)]
scsi: libfc: don't set FC_RQST_STATE_DONE before calling fc_bsg_jobdone()

Don't set FC_RQST_STATE_DONE before calling fc_bsg_jobdone() as
fc_bsg_jobdone() calls blk_complete_requeust() which raises a soft-IRQ
that ends up in fc_bsg_sofirq_done() and fc_bsg_softirq_done() sets the
FC_RQST_STATE_DONE flag.

Signed-off-by: Johannes Thumshirn <jthumshirn@suse.de>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: fc: provide fc_bsg_to_rport() helper
Johannes Thumshirn [Thu, 17 Nov 2016 09:31:15 +0000 (10:31 +0100)]
scsi: fc: provide fc_bsg_to_rport() helper

Provide fc_bsg_to_rport() helper that will become handy when we're
moving from struct fc_bsg_job to a plain struct bsg_job. Also move all
LLDDs to use the new helper.

Signed-off-by: Johannes Thumshirn <jthumshirn@suse.de>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Acked-by: Tyrel Datwyler <tyreld@linux.vnet.ibm.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: fc: provide fc_bsg_to_shost() helper
Johannes Thumshirn [Thu, 17 Nov 2016 09:31:14 +0000 (10:31 +0100)]
scsi: fc: provide fc_bsg_to_shost() helper

Provide fc_bsg_to_shost() helper that will become handy when we're
moving from struct fc_bsg_job to a plain struct bsg_job. Also use this
little helper in the LLDDs.

Signed-off-by: Johannes Thumshirn <jthumshirn@suse.de>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Acked-by: Tyrel Datwyler <tyreld@linux.vnet.ibm.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: fc: Export fc_bsg_jobdone and use it in FC drivers
Johannes Thumshirn [Thu, 17 Nov 2016 09:31:13 +0000 (10:31 +0100)]
scsi: fc: Export fc_bsg_jobdone and use it in FC drivers

Export fc_bsg_jobdone so drivers can use it directly instead of doing
the round-trip via struct fc_bsg_job::job_done() and use it in the
LLDDs.  That way we can also unify the interfaces of fc_bsg_jobdone and
bsg_job_done.

As we've converted all LLDDs over to use fc_bsg_jobdone() directly, we
can remove the function pointer from struct fc_bsg_job as well.

Signed-off-by: Johannes Thumshirn <jthumshirn@suse.de>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Acked-by: Tyrel Datwyler <tyreld@linux.vnet.ibm.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: don't use fc_bsg_job::request and fc_bsg_job::reply directly
Johannes Thumshirn [Thu, 17 Nov 2016 09:31:12 +0000 (10:31 +0100)]
scsi: don't use fc_bsg_job::request and fc_bsg_job::reply directly

Don't use fc_bsg_job::request and fc_bsg_job::reply directly, but use
helper variables bsg_request and bsg_reply. This will be helpful when
transitioning to bsg-lib.

Signed-off-by: Johannes Thumshirn <jthumshirn@suse.de>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: Get rid of struct fc_bsg_buffer
Johannes Thumshirn [Thu, 17 Nov 2016 09:31:11 +0000 (10:31 +0100)]
scsi: Get rid of struct fc_bsg_buffer

struct fc_bsg_buffer is just a clone of struct bsg_buffer from bsg-lib,
so use this one instead.

Signed-off-by: Johannes Thumshirn <jthumshirn@suse.de>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: mvsas: fix command_active typo
Arnd Bergmann [Wed, 16 Nov 2016 15:08:34 +0000 (16:08 +0100)]
scsi: mvsas: fix command_active typo

gcc-7 notices that the condition in mvs_94xx_command_active looks
suspicious:

drivers/scsi/mvsas/mv_94xx.c: In function 'mvs_94xx_command_active':
drivers/scsi/mvsas/mv_94xx.c:671:15: error: '<<' in boolean context, did you mean '<' ? [-Werror=int-in-bool-context]

This was introduced when the mv_printk() statement got added, and leads
to the condition being ignored. This is probably harmless.

Changing '&&' to '&' makes the code look reasonable, as we check the
command bit before setting and printing it.

Fixes: a4632aae8b66 ("[SCSI] mvsas: Add new macros and functions")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: dmx3191d: use module_pci_driver
Geliang Tang [Wed, 16 Nov 2016 13:00:58 +0000 (21:00 +0800)]
scsi: dmx3191d: use module_pci_driver

Use module_pci_driver() helper to simplify the code.

Signed-off-by: Geliang Tang <geliangtang@gmail.com>
Acked-by: Finn Thain <fthain@telegraphics.com.au>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: fnic: Correcting rport check location in fnic_queuecommand_lck
Satish Kharat [Wed, 16 Nov 2016 08:54:01 +0000 (00:54 -0800)]
scsi: fnic: Correcting rport check location in fnic_queuecommand_lck

Signed-off-by: Satish Kharat <satishkh@cisco.com>
Signed-off-by: Sesidhar Baddela <sebaddel@cisco.com>
Reviewed-by: Tomas Henzl <thenzl@redhat.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: ufs: introduce a new ufshcd_statea UFSHCD_STATE_EH_SCHEDULED
Zang Leigang [Wed, 16 Nov 2016 03:29:37 +0000 (11:29 +0800)]
scsi: ufs: introduce a new ufshcd_statea UFSHCD_STATE_EH_SCHEDULED

Add a new ufshcd_state, indicats that an err handler may get to run
immediately. Use UFSHCD_STATE_ERROR here looks not literaly correct.

Signed-off-by: Zang Leigang <zangleigang@hisilicon.com>
Reviewed-by: Subhash Jadavani <subhashj@codeaurora.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoMAINTAINERS: Updating maintainers list for Cisco FNI and SNIC drivers
Satish Kharat [Tue, 15 Nov 2016 20:39:29 +0000 (12:39 -0800)]
MAINTAINERS: Updating maintainers list for Cisco FNI and SNIC drivers

Signed-off-by: Satish Kharat <satishkh@cisco.com>
Signed-off-by: Sesidhar Baddela <sebaddel@cisco.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: hpsa: correct logical resets
Don Brace [Tue, 15 Nov 2016 20:45:32 +0000 (14:45 -0600)]
scsi: hpsa: correct logical resets

- driver was not calling done in some cases which causes the volume to
  be offlined.
- avoid doing rescan during a reset.

Reviewed-by: Scott Teel <scott.teel@microsemi.com>
Reviewed-by: Kevin Barnett <kevin.barnett@microsemi.com>
Signed-off-by: Don Brace <don.brace@microsemi.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
8 years agoscsi: megaraid-sas: request irqs later
Tomas Henzl [Tue, 1 Nov 2016 16:32:02 +0000 (17:32 +0100)]
scsi: megaraid-sas: request irqs later

It is not good when an irq arrives before driver structures are
allocated.

Signed-off-by: Tomas Henzl <thenzl@redhat.com>
Acked-by: Sumit Saxena <sumit.saxena@broadcom.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>