The SAS code will need it.  Also mark the name argument const to match
bsg_register_queue.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NJohannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Introduce struct scsi_vpd for the VPD page length, data and the RCU head
that will be used to free the VPD data. Use kfree_rcu() instead of
kfree() to free VPD data. Move the VPD buffer pointer check inside the
RCU read lock in the sysfs code. Only annotate pointers that are shared
across threads with __rcu. Use rcu_dereference() when dereferencing an
RCU pointer. This patch suppresses about twenty sparse complaints about
the vpd_pg8[03] pointers. This patch also fixes a race condition, namely
that updating of the VPD pointers and length variables in struct
scsi_device was not atomic with reference to the code reading these
variables. See also "Does the update code tolerate concurrent accesses?"
in Documentation/RCU/checklist.txt.

Fixes: commit 09e2b0b1 ("scsi: rescan VPD attributes")
Signed-off-by: NBart Van Assche <bart.vanassche@wdc.com>
Acked-by: NHannes Reinecke <hare@suse.de>
Reviewed-by: NShane Seymour <shane.seymour@hpe.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Johannes Thumshirn <jthumshirn@suse.de>
Cc: Shane Seymour <shane.seymour@hpe.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Introduce the scsi_get_vpd_buf() and scsi_update_vpd_page()
functions. The only functional change in this patch is that if updating
page 0x80 fails that it is attempted to update page 0x83.
Signed-off-by: NBart Van Assche <bart.vanassche@wdc.com>
Acked-by: NHannes Reinecke <hare@suse.de>
Reviewed-by: NShane Seymour <shane.seymour@hpe.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Johannes Thumshirn <jthumshirn@suse.de>
Cc: Shane M Seymour <shane.seymour@hpe.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

A common pattern in RCU code is to assign a new value to an RCU pointer
after having read and stored the old value. Introduce a macro for this
pattern.
Signed-off-by: NBart Van Assche <bart.vanassche@wdc.com>
Acked-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Hannes Reinecke <hare@suse.de>
Cc: Johannes Thumshirn <jthumshirn@suse.de>
Cc: Shane M Seymour <shane.seymour@hpe.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

A merge error crept in when formatting commit af167bc4 ("scsi: qlogicpti:
move bus reset to host reset")

Fixes: af167bc4 ("scsi: qlogicpti: move bus reset to host reset")
Signed-off-by: NHannes Reinecke <hare@suse.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

This is an interesting regression with gcc-8, showing a harmless warning
for correct code:

In file included from include/linux/kernel.h:13:0,
                 ...
                 from drivers/scsi/lpfc/lpfc_debugfs.c:23:
include/linux/printk.h:301:2: error: 'eq' may be used uninitialized in this function [-Werror=maybe-uninitialized]
  printk(KERN_ERR pr_fmt(fmt), ##__VA_ARGS__)
  ^~~~~~
In file included from drivers/scsi/lpfc/lpfc_debugfs.c:58:0:
drivers/scsi/lpfc/lpfc_debugfs.h:451:31: note: 'eq' was declared here

I managed to reduce the warning into a small test case for gcc-8 that I
reported in the gcc bugzilla[1].

As a workaround, this changes the logic to move the two assignments of
'eq' out of the conditions and instead make the index conditional.  This
works for all configurations I tried and avoids adding a bogus
initialization.
Acked-by: NJames Smart <james.smart@broadcom.com>
Link: [1] https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81958Signed-off-by: NArnd Bergmann <arnd@arndb.de>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

The only reference to lpfc_nvmet_replenish_context() is inside of an
disabled:

drivers/scsi/lpfc/lpfc_nvmet.c:1457:1: error: 'lpfc_nvmet_replenish_context' defined but not used [-Werror=unused-function]

This replaces the preprocessor conditional with a C condition, so the
compiler can see that the function is intentionally unused.

Fixes: 9a38e4f1c82f ("scsi: lpfc: Fix MRQ > 1 context list handling")
Signed-off-by: NArnd Bergmann <arnd@arndb.de>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

The ioctl DK_CAPI_VLUN_RESIZE can fail if the allocated vlun size is
reduced from almost maximum capacity and then increased again.

The shrink_lxt() routine is currently using the SISL_ASTATUS_MASK to
mask the higher 48 bits of the lxt entry. This is unnecessary and
incorrect as it uses a mask designed for the asynchronous interrupt
status register.  When the 4 port support was added to cxlflash, the
SISL_ASTATUS_MASK was updated to reflect the status bits for all 4
ports. This change indirectly affected the shrink_lxt() code path.

To extract the base, simply shift the bits without masking.

Fixes: 56518072 ("scsi: cxlflash: SISlite updates to support 4 ports")
Signed-off-by: NUma Krishnan <ukrishn@linux.vnet.ibm.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

The AFU recovery routine uses an interruptible mutex to control the flow
of in-flight recoveries. Upon receiving an interruptible signal the code
branches to a common exit path which wrongly assumes the mutex is
held. Add a local variable to track when the mutex should be unlocked.
Signed-off-by: NMatthew R. Ochs <mrochs@linux.vnet.ibm.com>
Signed-off-by: NUma Krishnan <ukrishn@linux.vnet.ibm.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

The AFU termination sequence has been refactored over time such that the
main tear down routine, term_afu(), can no longer can be invoked with a
NULL AFU pointer. Remove the unnecessary existence check from
term_afu().
Signed-off-by: NMatthew R. Ochs <mrochs@linux.vnet.ibm.com>
Signed-off-by: NUma Krishnan <ukrishn@linux.vnet.ibm.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

vio_device_id are not supposed to change at runtime. All functions
working with vio_device_id provided by <asm/vio.h> work with const
vio_device_id. So mark the non-const structs as const.
Signed-off-by: NArvind Yadav <arvind.yadav.cs@gmail.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

The kerneldoc comment for scsi_initialize_rq() neglected to document the
"rq" parameter, leading to this docs build warning:

  ./drivers/scsi/scsi_lib.c:1116: warning: No description found for parameter 'rq'

Document the parameter and make the build slightly quieter.

[mkp: used wording suggested by Bart]
Signed-off-by: NJonathan Corbet <corbet@lwn.net>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Currently we free the resources backing the enclosure device before we
call device_unregister(). This is racy: during rmmod of low-level SCSI
drivers that hook into enclosure, we end up with a small window of time
during which writing to /sys can OOPS. Example trace with mpt3sas:

  general protection fault: 0000 [#1] SMP KASAN
  Modules linked in: mpt3sas(-) <...>
  RIP: [<ffffffffa0388a98>] ses_get_page2_descriptor.isra.6+0x38/0x220 [ses]
  Call Trace:
   [<ffffffffa0389d14>] ses_set_fault+0xf4/0x400 [ses]
   [<ffffffffa0361069>] set_component_fault+0xa9/0xf0 [enclosure]
   [<ffffffff8205bffc>] dev_attr_store+0x3c/0x70
   [<ffffffff81677df5>] sysfs_kf_write+0x115/0x180
   [<ffffffff81675725>] kernfs_fop_write+0x275/0x3a0
   [<ffffffff8151f810>] __vfs_write+0xe0/0x3e0
   [<ffffffff8152281f>] vfs_write+0x13f/0x4a0
   [<ffffffff81526731>] SyS_write+0x111/0x230
   [<ffffffff828b401b>] entry_SYSCALL_64_fastpath+0x13/0x94

Fortunately the solution is extremely simple: call device_unregister()
before we free the resources, and the race no longer exists. The driver
core holds a reference over ->remove_dev(), so AFAICT this is safe.
Signed-off-by: NCalvin Owens <calvinowens@fb.com>
Acked-by: NHannes Reinecke <hare@suse.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

VMWare ESXi emulates an mptsas HBA, but exposes all drives as
direct-attached SAS drives.  This it not how the driver originally
envisioned things; SAS drives were supposed to be connected via an
expander, and only SATA drives would be direct attached.  As such, any
hotplug event for direct-attach SAS drives was silently ignored, and the
guest failed to detect new drives from within a VMWare ESXi environment.

[mkp: typos]

Bugzilla: https://bugzilla.suse.com/show_bug.cgi?id=1030850Signed-off-by: NHannes Reinecke <hare@suse.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Make these const as they are only stored in the type field of a device
structure, which is const.
Done using Coccinelle.
Signed-off-by: NBhumika Goyal <bhumirks@gmail.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

gd->minors has been set when call alloc_disk() in sd_probe.
Signed-off-by: Nweiping zhang <zhangweiping@didichuxing.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Just displaying some different information; drop it.
Signed-off-by: NHannes Reinecke <hare@suse.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NJohannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

There's no need to keep the private data for a device in a separate
list; better to store it in ->hostdata and do away with the additional
list.
Signed-off-by: NHannes Reinecke <hare@suse.com>
Reviewed-by: NDavid Kershner <david.kershner@unisys.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

megaraid_mbox only has one reset function, and that is a host reset.  So
drop the duplicate bus reset and device reset functions.
Signed-off-by: NHannes Reinecke <hare@suse.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NJohannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Never used anywhere.
Signed-off-by: NHannes Reinecke <hare@suse.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NJohannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

bus reset always returns SUCCESS, meaning host reset was never
tested. At the same time the only difference to the HBA is a missing
call to NCR_700_chip_reset().  So add the missing call to bus reset,
drop host reset, and move bus reset to host reset.
Signed-off-by: NHannes Reinecke <hare@suse.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

The driver has both a bus and a host reset, where the host reset does a
bus reset followed by an attempt to reset the chip registers to a
default state.  However, as the bus reset always returned SUCCESS the
host reset was never called, so the functionality of the register reset
function was never validated.  Additionally, tha AIC-6260 chip has a
hard reset line, which actually should be preferred for a host
reset. But I haven't found a way how this can be triggered via software,
so take the safe approach and drop the host reset.

[mkp: typo]
Signed-off-by: NHannes Reinecke <hare@suse.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

bus reset is a host reset without nsp32hw_init(), and will always return
SUCCESS, thus disabling the use of host reset.  So drop bus reset in
favour of host reset.
Signed-off-by: NHannes Reinecke <hare@suse.com>
Reviewed-by: NJohannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

qedf has a host reset handler, but as the bus reset handler is a stub
always returning SUCCESS the host reset is never invoked.  So drop the
bus reset handler.
Signed-off-by: NHannes Reinecke <hare@suse.com>
Tested-by: NChad Dupuis <chad.dupuis@cavium.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

bus_reset and host_reset are the same functions, so drop bus_reset.
Signed-off-by: NHannes Reinecke <hare@suse.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NJohannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

host_reset and bus_reset is the same function, so drop bus reset.
Signed-off-by: NHannes Reinecke <hare@suse.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NJohannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

The bus reset handler is really a host reset.
Signed-off-by: NHannes Reinecke <hare@suse.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NJohannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

The bus reset handler really is a host reset, so move it to
eh_bus_reset_handler.
Signed-off-by: NHannes Reinecke <hare@suse.com>
Acked-by: NFinn Thain <fthain@telegraphics.com.au>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NJohannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

The bus reset function is really a host reset, so move it to
eh_host_reset_handler.
Signed-off-by: NHannes Reinecke <hare@suse.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NJohannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

The bus reset function really is a host reset, so move it to
eh_host_reset_handler().
Signed-off-by: NHannes Reinecke <hare@suse.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NJohannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Function is a stub, so can as well be dropped.
Signed-off-by: NHannes Reinecke <hare@suse.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NJohannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

The bus reset function is just a wrapper calling host reset under the
host lock. So move taking of the host lock into the host reset function
and drop bus reset.
Signed-off-by: NHannes Reinecke <hare@suse.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NJohannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

The bus reset function really is a host reset, so move it to
eh_host_reset_handler().
Signed-off-by: NHannes Reinecke <hare@suse.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NJohannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

The Highpoint driver only has one reset function, and that is a host
reset. So stop pretending we're doing anything else.
Signed-off-by: NHannes Reinecke <hare@suse.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NJohannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

The bus reset handler is just calling target reset on all targets, which
is exactly what SCSI EH will be doing anyway.  So move the bus reset
function to target reset and drop the loop.
Signed-off-by: NHannes Reinecke <hare@suse.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NJohannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

The bus reset handler is calling I_T Nexus reset, which logically is a
target reset as it need to specify both the initiator and the target.
So move it to target reset.
Signed-off-by: NHannes Reinecke <hare@suse.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NJohannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

The bus_reset handler is really a device reset, so move it to
eh_device_reset_handler().
Signed-off-by: NHannes Reinecke <hare@suse.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NJohannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Command abort already returns FAILED, which will then be escalated to a
host reset. So no need to call host_reset directly.
Signed-off-by: NHannes Reinecke <hare@suse.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NJohannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

When calling host reset we're resetting all ports anyway, so there is no
point in waiting for the ports to become unblocked.
Signed-off-by: NHannes Reinecke <hare@suse.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NJohannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

When we're resetting the host any remote port states will be reset
anyway, so it's pointless to wait for dev_loss_tmo during host reset.
Signed-off-by: NHannes Reinecke <hare@suse.de>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NJohannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>