scsi: aha1740: Fix an errro handling path in aha1740_probe()

If 'dma_map_single()' fails, the ref counted 'shpnt' will be decremented
twice because 'scsi_host_put()' is called in the if block, and in the error
handling path.

Axe one of these calls.

Link: https://lore.kernel.org/r/20200228215948.7473-1-christophe.jaillet@wanadoo.fr
Fixes: 1dc09e120c83 ("scsi: aha1740: stop using scsi_unregister")
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: qla2xxx: Remove non functional code

Remove code which has no functional use anymore since commit 3c75ad1d87c7
("scsi: qla2xxx: Remove defer flag to indicate immeadiate port loss").

While at it remove also the stale function documentation.

Link: https://lore.kernel.org/r/20200206135443.110701-1-dwagner@suse.de
Reviewed-by: Arun Easi <aeasi@marvell.com>
Reviewed-by: Lee Duncan <lduncan@suse.com>
Signed-off-by: Daniel Wagner <dwagner@suse.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: pm80xx: Introduce read and write length for IOCTL payload structure

Removed the common length and introduce read and write length for IOCTL
payload structure.

[mkp: fixed SoB ordering]

Link: https://lore.kernel.org/r/20200316074906.9119-7-deepak.ukey@microchip.com
Acked-by: Jack Wang <jinpu.wang@cloud.ionos.com>
Signed-off-by: Viswas G <viswas.g@microchip.com>
Signed-off-by: Deepak Ukey <deepak.ukey@microchip.com>
Signed-off-by: Radha Ramachandran <radha@google.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: pm80xx: sysfs attribute for non fatal dump

Added the sysfs attribute for non fatal log so that management utility can
get the non fatal dump from driver. The non-fatal error is an error
condition or abnormal behavior detected by the host, or detected and
reported by the controller to the host.The non-fatal error does not stop
the controller firmware and enables it to still respond to host requests.
A typical example of a non-fatal error is an I/O timeout or an unusual
error notification from the controller. Since the firmware is operational,
the error dump information is pushed to host memory (by firmware) upon
request from the host.

Link: https://lore.kernel.org/r/20200316074906.9119-6-deepak.ukey@microchip.com
Acked-by: Jack Wang <jinpu.wang@cloud.ionos.com>
Signed-off-by: Deepak Ukey <deepak.ukey@microchip.com>
Signed-off-by: Viswas G <Viswas.G@microchip.com>
Signed-off-by: Radha Ramachandran <radha@google.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: pm80xx: Cleanup initialization loading fail path

1) Move the instance tracking down after we think the instance is good to
   go. Avoids having a use-after free.

2) There are goto targets for trying to cleanup if the hw fails to
   initialize, but there's some overlap depending on who thinks they own
   the sub-structures.

Link: https://lore.kernel.org/r/20200316074906.9119-5-deepak.ukey@microchip.com
Acked-by: Jack Wang <jinpu.wang@cloud.ionos.com>
Signed-off-by: Peter Chang <dpf@google.com>
Signed-off-by: Deepak Ukey <deepak.ukey@microchip.com>
Signed-off-by: Viswas G <Viswas.G@microchip.com>
Signed-off-by: Radha Ramachandran <radha@google.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: pm80xx: Free the tag when mpi_set_phy_profile_resp is received

In pm80xx driver, the command mpi_set_phy_profile_req is sent by host
during boot to configure the phy profile such as analog setting page, rate
control page. However, the tag is not freed when its response is
received. As a result, 16 tags are missing for each HBA after boot.  When
NCQ is enabled with queue depth 16, it needs at least, 15 * 16 = 240 tags
for each HBA to achieve the best performance. In current pm80xx driver with
setting CCB_MAX = 256, the total number of tags in each HBA is 255 for data
IO. Hence, without returning those tags to the pool after boot, some device
will finally be forced to non-ncq mode by ATA layer due to excessive errors
(i.e. LLDD cannot allocate tag for queued task).

Link: https://lore.kernel.org/r/20200316074906.9119-4-deepak.ukey@microchip.com
Acked-by: Jack Wang <jinpu.wang@cloud.ionos.com>
Signed-off-by: yuuzheng <yuuzheng@google.com>
Signed-off-by: Deepak Ukey <deepak.ukey@microchip.com>
Signed-off-by: Viswas G <Viswas.G@microchip.com>
Signed-off-by: Radha Ramachandran <radha@google.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: pm80xx: Deal with kexec reboots

A kexec reboot causes the controller fw to assert. This assertion shows up
in two ways, the controller doesn't show up as ready and an interrupt is
waiting as soon as the handler is registered. To resolve this added below
fix:

 - Split the interrupt handling setup into two parts, setup and request.

 - If the controller ready register indicates not-ready, but that the not
   readiness is only on the IOC units we can still try a reset to bring the
   system back to the pre-reboot state.

Link: https://lore.kernel.org/r/20200316074906.9119-3-deepak.ukey@microchip.com
Acked-by: Jack Wang <jinpu.wang@cloud.ionos.com>
Signed-off-by: Vikram Auradkar <auradkar@google.com>
Signed-off-by: Deepak Ukey <deepak.ukey@microchip.com>
Signed-off-by: Viswas G <Viswas.G@microchip.com>
Signed-off-by: Radha Ramachandran <radha@google.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: pm80xx: Increase request sg length

Increasing the per-request size maximum (max_sectors_kb) runs into the
per-device DMA scatter gather list limit (max_segments) for users of the io
vector system calls (eg, readv and writev). This is because the kernel
combines io vectors into DMA segments when possible, but it doesn't work
for our user because the vectors in the buffer cache get scrambled.  This
change bumps the advertised max scatter gather length to 528 to cover 2M w/
x86's 4k pages and some extra for the user checksum.  It trims the size of
some of the tables we don't care about and exposes all of the command slots
upstream to the SCSI layer.  Also reduced the PM8001_MAX_CCB to 256 as
pm8001 driver has memory limit depend on machine capability. If we increase
the sg length, we need to trade-off it by decreasing PM8001_MAX_CCB.
PM8001_MAX_CCB = 256 does not have any influence on normal use

Link: https://lore.kernel.org/r/20200316074906.9119-2-deepak.ukey@microchip.com
Reported-by: kbuild test robot <lkp@intel.com>
Acked-by: Jack Wang <jinpu.wang@cloud.ionos.com>
Signed-off-by: Peter Chang <dpf@google.com>
Signed-off-by: Deepak Ukey <deepak.ukey@microchip.com>
Signed-off-by: Viswas G <Viswas.G@microchip.com>
Signed-off-by: Radha Ramachandran <radha@google.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: smartpqi: Use scnprintf() for avoiding potential buffer overflow

Since snprintf() returns the would-be-output size instead of the actual
output size, the succeeding calls may go beyond the given buffer limit.
Fix it by replacing with scnprintf().

Link: https://lore.kernel.org/r/20200315094241.9086-9-tiwai@suse.de
Cc: "James E . J . Bottomley" <jejb@linux.ibm.com>
Cc: "Martin K . Petersen" <martin.petersen@oracle.com>
Cc: Don Brace <don.brace@microsemi.com>
Cc: linux-scsi@vger.kernel.org
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: core: Use scnprintf() for avoiding potential buffer overflow

Since snprintf() returns the would-be-output size instead of the actual
output size, the succeeding calls may go beyond the given buffer limit.
Fix it by replacing with scnprintf().

Link: https://lore.kernel.org/r/20200315094241.9086-8-tiwai@suse.de
Cc: "James E . J . Bottomley" <jejb@linux.ibm.com>
Cc: "Martin K . Petersen" <martin.petersen@oracle.com>
Cc: linux-scsi@vger.kernel.org
Reviewed-by: Bart van Assche <bvanassche@acm.org>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: megaraid_sas: Use scnprintf() for avoiding potential buffer overflow

Since snprintf() returns the would-be-output size instead of the actual
output size, the succeeding calls may go beyond the given buffer limit.
Fix it by replacing with scnprintf().

Also corrected the wrongly passed limit size.  The remaining buffer size
must be decremented.

Link: https://lore.kernel.org/r/20200315094241.9086-7-tiwai@suse.de
Cc: "James E . J . Bottomley" <jejb@linux.ibm.com>
Cc: "Martin K . Petersen" <martin.petersen@oracle.com>
Cc: Kashyap Desai <kashyap.desai@broadcom.com>
Cc: Sumit Saxena <sumit.saxena@broadcom.com>
Cc: Shivasharan S <shivasharan.srikanteshwara@broadcom.com>
Cc: linux-scsi@vger.kernel.org
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: ipr: Use scnprintf() for avoiding potential buffer overflow

Since snprintf() returns the would-be-output size instead of the actual
output size, the succeeding calls may go beyond the given buffer limit.
Fix it by replacing with scnprintf().

Link: https://lore.kernel.org/r/20200315094241.9086-6-tiwai@suse.de
Cc: "James E . J . Bottomley" <jejb@linux.ibm.com>
Cc: "Martin K . Petersen" <martin.petersen@oracle.com>
Cc: Brian King <brking@us.ibm.com>
Cc: linux-scsi@vger.kernel.org
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: gdth: Use scnprintf() for avoiding potential buffer overflow

Since snprintf() returns the would-be-output size instead of the actual
output size, the succeeding calls may go beyond the given buffer limit.
Fix it by replacing with scnprintf().

[mkp: checkpatch fix]

Link: https://lore.kernel.org/r/20200315094241.9086-5-tiwai@suse.de
Cc: "James E . J . Bottomley" <jejb@linux.ibm.com>
Cc: "Martin K . Petersen" <martin.petersen@oracle.com>
Cc: Achim Leubner <achim_leubner@adaptec.com>
Cc: linux-scsi@vger.kernel.org
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: fnic: Use scnprintf() for avoiding potential buffer overflow

Since snprintf() returns the would-be-output size instead of the actual
output size, the succeeding calls may go beyond the given buffer limit.
Fix it by replacing with scnprintf().

Link: https://lore.kernel.org/r/20200315094241.9086-4-tiwai@suse.de
Cc: "James E . J . Bottomley" <jejb@linux.ibm.com>
Cc: "Martin K . Petersen" <martin.petersen@oracle.com>
Cc: Satish Kharat <satishkh@cisco.com>
Cc: Sesidhar Baddela <sebaddel@cisco.com>
Cc: Karan Tilak Kumar <kartilak@cisco.com>
Cc: linux-scsi@vger.kernel.org
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: be2iscsi: Use scnprintf() for avoiding potential buffer overflow

Since snprintf() returns the would-be-output size instead of the actual
output size, the succeeding calls may go beyond the given buffer limit.
Fix it by replacing with scnprintf().

Link: https://lore.kernel.org/r/20200315094241.9086-3-tiwai@suse.de
Cc: "James E . J . Bottomley" <jejb@linux.ibm.com>
Cc: "Martin K . Petersen" <martin.petersen@oracle.com>
Cc: Subbu Seetharaman <subbu.seetharaman@broadcom.com>
Cc: Ketan Mukadam <ketan.mukadam@broadcom.com>
Cc: Jitendra Bhivare <jitendra.bhivare@broadcom.com>
Cc: linux-scsi@vger.kernel.org
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: aacraid: Use scnprintf() for avoiding potential buffer overflow

Since snprintf() returns the would-be-output size instead of the actual
output size, the succeeding calls may go beyond the given buffer limit.
Fix it by replacing with scnprintf().

Link: https://lore.kernel.org/r/20200315094241.9086-2-tiwai@suse.de
Cc: "James E . J . Bottomley" <jejb@linux.ibm.com>
Cc: "Martin K . Petersen" <martin.petersen@oracle.com>
Cc: Adaptec OEM Raid Solutions <aacraid@microsemi.com>
Cc: linux-scsi@vger.kernel.org
Acked-by: Balsundar P <Balsundar.P@microchip.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: zfcp: log FC Endpoint Security errors

Log any FC Endpoint Security errors to the kernel ring buffer with rate-
limiting.

Link: https://lore.kernel.org/r/20200312174505.51294-11-maier@linux.ibm.com
Reviewed-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Jens Remus <jremus@linux.ibm.com>
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: zfcp: enhance handling of FC Endpoint Security errors

Enable for explicit FCP channel FC Endpoint Security error reporting and
handle any FSF security errors according to specification. Take the
following recovery actions when a FSF_SECURITY_ERROR is reported for the
specified FSF commands:

- Open Port: Retry the command if possible
- Send FCP : Physically close the remote port and reopen

For Open Port the command status is set to error, which triggers a retry.
For Send FCP the command status is set to error and recovery is triggered
to physically reopen the remote port.

Link: https://lore.kernel.org/r/20200312174505.51294-10-maier@linux.ibm.com
Reviewed-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Jens Remus <jremus@linux.ibm.com>
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: zfcp: trace FC Endpoint Security of FCP devices and connections

Trace changes in Fibre Channel Endpoint Security capabilities of FCP
devices as well as changes in Fibre Channel Endpoint Security state of
their connections to FC remote ports as FC Endpoint Security changes with
trace level 3 in HBA DBF.

A change in FC Endpoint Security capabilities of FCP devices is traced as
response to FSF command FSF_QTCB_EXCHANGE_PORT_DATA with a trace tag of
"fsfcesa" and a WWPN of ZFCP_DBF_INVALID_WWPN = 0x0000000000000000 (see
FC-FS-4 §18 "Name_Identifier Formats", NAA field).

A change in FC Endpoint Security state of connections between FCP devices
and FC remote ports is traced as response to FSF command
FSF_QTCB_OPEN_PORT_WITH_DID with a trace tag of "fsfcesp".

Example trace record of FC Endpoint Security capability change of FCP
device formatted with zfcpdbf from s390-tools:

Timestamp      : ...
Area           : HBA
Subarea        : 00
Level          : 3
Exception      : -
CPU ID         : ...
Caller         : 0x...
Record ID      : 5                    ZFCP_DBF_HBA_FCES
Tag            : fsfcesa              FSF FC Endpoint Security adapter
Request ID     : 0x...
Request status : 0x00000010
FSF cmnd       : 0x0000000e           FSF_QTCB_EXCHANGE_PORT_DATA
FSF sequence no: 0x...
FSF issued     : ...
FSF stat       : 0x00000000           FSF_GOOD
FSF stat qual  : n/a
Prot stat      : n/a
Prot stat qual : n/a
Port handle    : 0x00000000           none (invalid)
LUN handle     : n/a
WWPN           : 0x0000000000000000   ZFCP_DBF_INVALID_WWPN
FCES old       : 0x00000000           old FC Endpoint Security
FCES new       : 0x00000007           new FC Endpoint Security

Example trace record of FC Endpoint Security change of connection to
FC remote port formatted with zfcpdbf from s390-tools:

Timestamp      : ...
Area           : HBA
Subarea        : 00
Level          : 3
Exception      : -
CPU ID         : ...
Caller         : 0x...
Record ID      : 5                    ZFCP_DBF_HBA_FCES
Tag            : fsfcesp              FSF FC Endpoint Security port
Request ID     : 0x...
Request status : 0x00000010
FSF cmnd       : 0x00000005           FSF_QTCB_OPEN_PORT_WITH_DID
FSF sequence no: 0x...
FSF issued     : ...
FSF stat       : 0x00000000           FSF_GOOD
FSF stat qual  : n/a
Prot stat      : n/a
Prot stat qual : n/a
Port handle    : 0x...
WWPN           : 0x500507630401120c   WWPN
FCES old       : 0x00000000           old FC Endpoint Security
FCES new       : 0x00000004           new FC Endpoint Security

Link: https://lore.kernel.org/r/20200312174505.51294-9-maier@linux.ibm.com
Reviewed-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Jens Remus <jremus@linux.ibm.com>
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: zfcp: log FC Endpoint Security of connections

Log the usage of and subsequent changes in FC Endpoint Security of
connections between FCP devices and FC remote ports to the kernel ring
buffer. Activation of FC Endpoint Security is logged as informational.
Change and deactivation are logged as warning.

No logging takes place, if FC Endpoint Security is not used (i.e. never
activated) on a connection or if it does not change during reopen of a port
(e.g. due to adapter or port recovery).

Link: https://lore.kernel.org/r/20200312174505.51294-8-maier@linux.ibm.com
Reviewed-by: Steffen Maier <maier@linux.ibm.com>
Reviewed-by: Fedor Loshakov <loshakov@linux.ibm.com>
Signed-off-by: Jens Remus <jremus@linux.ibm.com>
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: zfcp: report FC Endpoint Security in sysfs

Add an interface to read Fibre Channel Endpoint Security information of FCP
channels and their connections to FC remote ports. It comes in the form of
new sysfs attributes that are attached to the CCW device representing the
FCP device and its zfcp port objects.

The read-only sysfs attribute "fc_security" of a CCW device representing a
FCP device shows the FC Endpoint Security capabilities of the device.
Possible values are: "unknown", "unsupported", "none", or a comma-
separated list of one or more mnemonics and/or one hexadecimal value
representing the supported FC Endpoint Security:

  Authentication: Authentication supported
  Encryption    : Encryption supported

The read-only sysfs attribute "fc_security" of a zfcp port object shows the
FC Endpoint Security used on the connection between its parent FCP device
and the FC remote port. Possible values are: "unknown", "unsupported",
"none", or a mnemonic or hexadecimal value representing the FC Endpoint
Security used:

  Authentication: Connection has been authenticated
  Encryption    : Connection is encrypted

Both sysfs attributes may return hexadecimal values instead of mnemonics,
if the mnemonic lookup table does not contain an entry for the FC Endpoint
Security reported by the FCP device.

Link: https://lore.kernel.org/r/20200312174505.51294-7-maier@linux.ibm.com
Reviewed-by: Fedor Loshakov <loshakov@linux.ibm.com>
Reviewed-by: Steffen Maier <maier@linux.ibm.com>
Reviewed-by: Benjamin Block <bblock@linux.ibm.com>
Signed-off-by: Jens Remus <jremus@linux.ibm.com>
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: zfcp: auto variables for dereferenced structs in open port handler

Introduce automatic variables for adapter and QTCB bottom in
zfcp_fsf_open_port_handler(). This facilitates subsequent changes to meet
the 80 character per line limit.

Link: https://lore.kernel.org/r/20200312174505.51294-6-maier@linux.ibm.com
Reviewed-by: Fedor Loshakov <loshakov@linux.ibm.com>
Reviewed-by: Steffen Maier <maier@linux.ibm.com>
Reviewed-by: Benjamin Block <bblock@linux.ibm.com>
Signed-off-by: Jens Remus <jremus@linux.ibm.com>
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: zfcp: fix fc_host attributes that should be unknown on local link down

When we get an unsolicited notification on local link went down,
zfcp_fsf_status_read_link_down() calls zfcp_fsf_link_down_info_eval().
This only blocks rports, and sets ZFCP_STATUS_ADAPTER_LINK_UNPLUGGED and
ZFCP_STATUS_COMMON_ERP_FAILED. Only the fc_host port_state changes to
"Linkdown", because zfcp_scsi_get_host_port_state() is an active callback
and uses the adapter status.

Other fc_host attributes model, port_id, port_type, speed, fabric_name (and
zfcp device attributes card_version, peer_wwpn, peer_wwnn, peer_d_id) which
depend on a local link, continued to show their last known "good" value.

Only if something triggered an exchange config data, some values were
updated to their unknown equivalent via case
FSF_EXCHANGE_CONFIG_DATA_INCOMPLETE due to local link down.  Triggers for
exchange config data are adapter recovery, or reading any of the following
zfcp-specific scsi host sysfs attributes "requests", "megabytes", or
"seconds_active" in /sys/devices/css*/*.*.*/*.*.*/host*/scsi_host/host*/.

The other fc_host attributes active_fc4s and permanent_port_name continued
to show their last known "good" value.  Only if something triggered an
exchange port data, some values changed.  Active_fc4s became all zeros as
unknown equivalent during link down.  Permanent_port_name does not depend
on a local link. But for non-NPIV FCP devices, permanent_port_name
erroneously became whatever value fc_host port_name had at that point in
time (see previous paragraph).  Triggers for exchange port data are the
zfcp-specific scsi host sysfs attribute "utilization", or
[{reset,get}_fc_host_stats] write anything into "reset_statistics" or read
any of the other attributes under
/sys/devices/css*/*.*.*/*.*.*/host*/fc_host/host*/statistics/.

(cf. v4.9 commit bd77befa5bcf ("zfcp: fix fc_host port_type with NPIV"))

This is particularly confusing when using "lszfcp -b <fcpdevbusid> -Ha" or
dbginfo.sh which read fc_host attributes and also scsi_host attributes.
After link down, the first invocation produces (abbreviated):

Class = "fc_host"
    active_fc4s         = "0x00 0x00 0x01 0x00 ..."
    ...
    fabric_name         = "0x10000027f8e04c49"
    ...
    permanent_port_name = "0xc05076e4588059c1"
    port_id             = "0x244800"
    port_state          = "Linkdown"
    port_type           = "NPort (fabric via point-to-point)"
    ...
    speed               = "16 Gbit"
Class = "scsi_host"
    ...
    megabytes           = "0 0"
    ...
    requests            = "0 0 0"
    seconds_active      = "37"
    ...
    utilization         = "0 0 0"

The second and next invocations produce (abbreviated):

Class = "fc_host"
    active_fc4s         = "0x00 0x00 0x00 0x00 ..."
    ...
    fabric_name         = "0x0"
    ...
    permanent_port_name = "0x0"
    port_id             = "0x000000"
    port_state          = "Linkdown"
    port_type           = "Unknown"
    ...
    speed               = "unknown"
Class = "scsi_host"
    ...
    megabytes           = "0 0"
    ...
    requests            = "0 0 0"
    seconds_active      = "38"
    ...
    utilization         = "0 0 0"

Factor out the resetting of local link dependent fc_host attributes from
zfcp_fsf_exchange_config_data_handler() case
FSF_EXCHANGE_CONFIG_DATA_INCOMPLETE into a new helper function
zfcp_fsf_fc_host_link_down().  All code places that detect local link down
(SRB, FSF_PROT_LINK_DOWN, xconf data/port incomplete) call
zfcp_fsf_link_down_info_eval().  Call the new helper from there. This works
because zfcp_fsf_link_down_info_eval() and thus the helper is called before
zfcp_fsf_exchange_{config,port}_evaluate().

Port_name and node_name are always valid, so never reset them.

Get the permanent_port_name from exchange port data unconditionally as it
always has a valid known good value, even during link down.

Note: Rather than hardcode in zfcp_fsf_exchange_config_evaluate(), fc_host
supported_classes could theoretically get its value from
fsf_qtcb_bottom_port.class_of_service in zfcp_fsf_exchange_port_evaluate().

When the link comes back, we get a different notification, perform adapter
recovery, and this triggers an implicit exchange config data followed by
exchange port data filling in the link dependent fc_host attributes with
known good values again.

Link: https://lore.kernel.org/r/20200312174505.51294-5-maier@linux.ibm.com
Reviewed-by: Jens Remus <jremus@linux.ibm.com>
Reviewed-by: Benjamin Block <bblock@linux.ibm.com>
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: zfcp: wire previously driver-specific sysfs attributes also to fc_host

Manufacturer, HBA model, firmware version, and hardware version.  Use the
same value format as for the driver-specific attributes.  Keep the
driver-specific attributes for stable user space sysfs API.

Link: https://lore.kernel.org/r/20200312174505.51294-4-maier@linux.ibm.com
Reviewed-by: Jens Remus <jremus@linux.ibm.com>
Reviewed-by: Benjamin Block <bblock@linux.ibm.com>
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: zfcp: expose fabric name as common fc_host sysfs attribute

FICON Express8S or older, as well as card features newer than FICON
Express16S+ have no certain firmware level requirement.

FICON Express16S or FICON Express16S+ have the following
minimum firmware level requirements to show a proper fabric name value:

 z13 machine
  FICON Express16S  , MCL P08424.005 , LIC version 0x00000721
 z14 machine
  FICON Express16S  , MCL P42611.008 , LIC version 0x10200069
  FICON Express16S+ , MCL P42625.010 , LIC version 0x10300147

Otherwise, the read value is not the fabric name.

Each FCP channel of these card features might need one SAN fabric re-login
after concurrent microcode update in order to show the proper fabric name.
Possible ways to trigger a SAN fabric re-login are one of: Pull fibres
between FCP channel port and SAN switch port on either side and re-plug,
disable SAN switch port adjacent to FCP channel port and re-enable switch
port, or at Service Element toggle off all CHPIDs of FCP channel over all
LPARs and toggle CHPIDs on again.  Zfcp operating subchannels (FCP devices)
on such FCP channel recovers a fabric re-login.

Initialize fabric name for any topology and have it an invalid WWPN 0x0 for
anything but fabric topology.  Otherwise for e.g. point-to-point topology
one could see the initial -1 from fc_host_setup() and after a link unplug
our fabric name would turn to 0x0 (with subsequent commit ("zfcp: fix
fc_host attributes that should be unknown on local link down") and stay 0x0
on link replug.  I did not initialize to 0x0 somewhere even earlier in the
code path such that it would not flap from real to 0x0 to real on e.g. an
exchange config data with fabric topology.

Link: https://lore.kernel.org/r/20200312174505.51294-3-maier@linux.ibm.com
Reviewed-by: Benjamin Block <bblock@linux.ibm.com>
Reviewed-by: Jens Remus <jremus@linux.ibm.com>
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: zfcp: fix missing erp_lock in port recovery trigger for point-to-point

v2.6.27 commit cc8c282963bd ("[SCSI] zfcp: Automatically attach remote
ports") introduced zfcp automatic port scan.

Before that, the user had to use the sysfs attribute "port_add" of an FCP
device (adapter) to add and open remote (target) ports, even for the remote
peer port in point-to-point topology. That code path did a proper port open
recovery trigger taking the erp_lock.

Since above commit, a new helper function zfcp_erp_open_ptp_port()
performed an UNlocked port open recovery trigger. This can race with other
parallel recovery triggers. In zfcp_erp_action_enqueue() this could corrupt
e.g. adapter->erp_total_count or adapter->erp_ready_head.

As already found for fabric topology in v4.17 commit fa89adba1941 ("scsi:
zfcp: fix infinite iteration on ERP ready list"), there was an endless loop
during tracing of rport (un)block.  A subsequent v4.18 commit 9e156c54ace3
("scsi: zfcp: assert that the ERP lock is held when tracing a recovery
trigger") introduced a lockdep assertion for that case.

As a side effect, that lockdep assertion now uncovered the unlocked code
path for PtP. It is from within an adapter ERP action:

zfcp_erp_strategy[1479]  intentionally DROPs erp lock around
                         zfcp_erp_strategy_do_action()
zfcp_erp_strategy_do_action[1441]      NO erp lock
zfcp_erp_adapter_strategy[876]         NO erp lock
zfcp_erp_adapter_strategy_open[855]    NO erp lock
zfcp_erp_adapter_strategy_open_fsf[806]NO erp lock
zfcp_erp_adapter_strat_fsf_xconf[772]  erp lock only around
                                       zfcp_erp_action_to_running(),
                                       BUT *_not_* around
                                       zfcp_erp_enqueue_ptp_port()
zfcp_erp_enqueue_ptp_port[728]         BUG: *_not_* taking erp lock
_zfcp_erp_port_reopen[432]             assumes to be called with erp lock
zfcp_erp_action_enqueue[314]           assumes to be called with erp lock
zfcp_dbf_rec_trig[288]                 _checks_ to be called with erp lock:
lockdep_assert_held(&adapter->erp_lock);

It causes the following lockdep warning:

WARNING: CPU: 2 PID: 775 at drivers/s390/scsi/zfcp_dbf.c:288
                            zfcp_dbf_rec_trig+0x16a/0x188
no locks held by zfcperp0.0.17c0/775.

Fix this by using the proper locked recovery trigger helper function.

Link: https://lore.kernel.org/r/20200312174505.51294-2-maier@linux.ibm.com
Fixes: cc8c282963bd ("[SCSI] zfcp: Automatically attach remote ports")
Cc: <stable@vger.kernel.org> #v2.6.27+
Reviewed-by: Jens Remus <jremus@linux.ibm.com>
Reviewed-by: Benjamin Block <bblock@linux.ibm.com>
Signed-off-by: Steffen Maier <maier@linux.ibm.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: scsi_trace: Use get_unaligned_be24()

This makes the SCSI tracing code slightly easier to read.

Link: https://lore.kernel.org/r/20200313203102.16613-6-bvanassche@acm.org
Fixes: bf8162354233 ("[SCSI] add scsi trace core functions and put trace points")
Cc: Christoph Hellwig <hch@lst.de>
Cc: James E.J. Bottomley <jejb@linux.ibm.com>
Cc: Martin K. Petersen <martin.petersen@oracle.com>
Reported-by: Colin Ian King <colin.king@canonical.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: st: Use get_unaligned_be24() and sign_extend32()

Use these functions instead of open-coding them.

Link: https://lore.kernel.org/r/20200313203102.16613-5-bvanassche@acm.org
Cc: Kai Makisara <Kai.Makisara@kolumbus.fi>
Cc: James E.J. Bottomley <jejb@linux.ibm.com>
Cc: Martin K. Petersen <martin.petersen@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: treewide: Consolidate {get,put}_unaligned_[bl]e24() definitions

Move the get_unaligned_be24(), get_unaligned_le24() and
put_unaligned_le24() definitions from various drivers into
include/linux/unaligned/generic.h. Add a put_unaligned_be24()
implementation.

Link: https://lore.kernel.org/r/20200313203102.16613-4-bvanassche@acm.org
Cc: Keith Busch <kbusch@kernel.org>
Cc: Sagi Grimberg <sagi@grimberg.me>
Cc: Jens Axboe <axboe@fb.com>
Cc: Harvey Harrison <harvey.harrison@gmail.com>
Cc: Martin K. Petersen <martin.petersen@oracle.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> # For drivers/usb
Reviewed-by: Felipe Balbi <balbi@kernel.org> # For drivers/usb/gadget
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: c6x: Include <linux/unaligned/generic.h> instead of duplicating it

Use the generic __{get,put}_unaligned_[bl]e() definitions instead of
duplicating these. Since a later patch will add more definitions into
<linux/unaligned/generic.h>, this patch ensures that these definitions have
to be added only once. See also commit a7f626c1948a ("C6X: headers").  See
also commit 6510d41954dc ("kernel: Move arches to use common unaligned
access").

Link: https://lore.kernel.org/r/20200313203102.16613-3-bvanassche@acm.org
Cc: Aurelien Jacquiot <jacquiot.aurelien@gmail.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Acked-by: Mark Salter <msalter@redhat.com>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: linux/unaligned/byteshift.h: Remove superfluous casts

The C language supports implicitly casting a void pointer into a non-void
pointer. Remove explicit void pointer to non-void pointer casts because
these are superfluous.

Link: https://lore.kernel.org/r/20200313203102.16613-2-bvanassche@acm.org
Cc: Harvey Harrison <harvey.harrison@gmail.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: core: Allow non-root users to perform ZBC commands

Allow users with read permissions to issue REPORT ZONE commands and users
with write permissions to manage zones on block devices supporting the ZBC
specification.

Link: https://lore.kernel.org/r/20200226170518.92963-2-ryanattard@ryanattard.info
Signed-off-by: Ryan Attard <ryanattard@ryanattard.info>
Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: hisi_sas: Use dev_err() in read_iost_itct_cache_v3_hw()

The print of pr_err() does not come with device information, so replace it
with dev_err(). Also improve the grammar in the message.

Link: https://lore.kernel.org/r/1583940144-230800-1-git-send-email-john.garry@huawei.com
Signed-off-by: Luo Jiaxing <luojiaxing@huawei.com>
Signed-off-by: John Garry <john.garry@huawei.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: core: avoid repetitive logging of device offline messages

Large queues of I/O to offline devices that are eventually submitted when
devices are unblocked result in a many repeated "rejecting I/O to offline
device" messages.  These messages can fill up the dmesg buffer in crash
dumps so no useful prior messages remain.  In addition, if a serial console
is used, the flood of messages can cause a hard lockup in the console code.

Introduce a flag indicating the message has already been logged for the
device, and reset the flag when scsi_device_set_state() changes the device
state.

Link: https://lore.kernel.org/r/20200311143930.20674-1-emilne@redhat.com
Reviewed-by: Bart van Assche <bvanassche@acm.org>
Signed-off-by: Ewan D. Milne <emilne@redhat.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: convert arcmsr_spec.txt to ReST

This file had its own peculiar style, not following any other
files inside the Kernel (as far as I saw).

Had to do a number of changes here, starting by removing the two
leading asterisks from each line, adding table and literal
block markups and changing whitespace and blank lines.

The end result is that (IMHO), it is now a lot easier to read
it as a text file, while producing a good html output.

Link: https://lore.kernel.org/r/6f8e4da4ea643adbe048f55504a59427c5e50c97.1583136624.git.mchehab+huawei@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: convert wd719x.txt to ReST

Link: https://lore.kernel.org/r/23e5b13d810b7dd8126b126173999c02eac50e74.1583136624.git.mchehab+huawei@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: convert ufs.txt to ReST

Link: https://lore.kernel.org/r/052d45576e342a217185e91a83793b384b1592a4.1583136624.git.mchehab+huawei@kernel.org
Acked-by: Avri Altman <avri.altman@wdc.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: convert tcm_qla2xxx.txt to ReST

Link: https://lore.kernel.org/r/de73fa02f38f67f54f22ef2842f9680c0b34434a.1583136624.git.mchehab+huawei@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: convert sym53c8xx_2.txt to ReST

Link: https://lore.kernel.org/r/08202c87294d61d147ec4ac784219d20805cdeb5.1583136624.git.mchehab+huawei@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: convert sym53c500_cs.txt to ReST

Link: https://lore.kernel.org/r/eff6166b3442ddb37b934bca46e7f9ef25ebc2a4.1583136624.git.mchehab+huawei@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: convert st.txt to ReST

Link: https://lore.kernel.org/r/6b2ddb36983e81e7028de6e5fd0c643c2fb4c6c9.1583136624.git.mchehab+huawei@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: convert smartpqi.txt to ReST

Link: https://lore.kernel.org/r/00b398efb7cfc667b046fbef92a84f1d3c33eb64.1583136624.git.mchehab+huawei@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: convert sd-parameters.txt to ReST

Link: https://lore.kernel.org/r/8d0b75b0faf13a2e81373570d6ce601b629fb22a.1583136624.git.mchehab+huawei@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: convert scsi.txt to ReST

Link: https://lore.kernel.org/r/c617b37769a82901def0fed3d236a25995c4e160.1583136624.git.mchehab+huawei@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: convert scsi-parameters.txt to ReST

Link: https://lore.kernel.org/r/f00a5f6f2bf9a2562e0856ee8f45bcf9521d181f.1583136624.git.mchehab+huawei@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: convert scsi_mid_low_api.txt to ReST

Link: https://lore.kernel.org/r/881e7741dfed5d6f5f73e1dfc2826b200b8604aa.1583136624.git.mchehab+huawei@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: convert scsi-generic.txt to ReST

Link: https://lore.kernel.org/r/f57b8ddf30397c2c7213e49634e5e9cbd4246368.1583136624.git.mchehab+huawei@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: convert scsi_fc_transport.txt to ReST

Link: https://lore.kernel.org/r/f75bd9b6512f223847cc4ece8bd7e8e72e434b21.1583136624.git.mchehab+huawei@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: convert scsi_eh.txt to ReST

Link: https://lore.kernel.org/r/300314197f2e6a3258200711e825aa04c9e8ceaf.1583136624.git.mchehab+huawei@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: convert scsi-changer.txt to ReST

Link: https://lore.kernel.org/r/433d073fa982174a19783c2e59412b724e2cf946.1583136624.git.mchehab+huawei@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: convert qlogicfas.txt to ReST

Link: https://lore.kernel.org/r/b69f795c781811b9a908abe43485f1dca0ee8ac5.1583136624.git.mchehab+huawei@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: convert ppa.txt to ReST

Link: https://lore.kernel.org/r/3db8cd51d77fef6b66632249412969caa29dec40.1583136624.git.mchehab+huawei@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: convert NinjaSCSI.txt to ReST

Link: https://lore.kernel.org/r/6385a411d000dad005b78647629e43700580ecf0.1583136624.git.mchehab+huawei@kernel.org
Acked-by: GOTO Masanori <gotom@debian.or.jp>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: convert ncr53c8xx.txt to ReST

Link: https://lore.kernel.org/r/cover.1583136624.git.mchehab+huawei@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: convert megaraid.txt to ReST

Link: https://lore.kernel.org/r/b7ee59230c5a33ff6d60edba0d0bcf3e2aeaa88f.1583136624.git.mchehab+huawei@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: convert lpfc.txt to ReST

Link: https://lore.kernel.org/r/48c13184b77ba61ed4fd7c235816fdb8e7530664.1583136624.git.mchehab+huawei@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: convert link_power_management_policy.txt to ReST

Link: https://lore.kernel.org/r/c56177fdf046d80e0dec6031c4139cb4e8c39d31.1583136624.git.mchehab+huawei@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: convert libsas.txt to ReST

Link: https://lore.kernel.org/r/9022cb5551487f774cab16a828fe06b0b6b3add3.1583136624.git.mchehab+huawei@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: convert hptiop.txt to ReST

Link: https://lore.kernel.org/r/d189a339bb360b7b397914ee3ddeb75d9a7fd788.1583136624.git.mchehab+huawei@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: convert hpsa.txt to ReST

Link: https://lore.kernel.org/r/ea58e04176d43fb7194615b145060aa04c9cf3ad.1583136624.git.mchehab+huawei@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: convert g_NCR5380.txt to ReST

Link: https://lore.kernel.org/r/a66e9ea704be6a7aa81b9864ad66a32b75ab808d.1583136624.git.mchehab+huawei@kernel.org
Acked-by: Finn Thain <fthain@telegraphics.com.au>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: convert FlashPoint.txt to ReST

Link: https://lore.kernel.org/r/e755b9644047eed6be69fcc77eb797f0801fcb99.1583136624.git.mchehab+huawei@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: convert dpti.txt to ReST

Link: https://lore.kernel.org/r/212fd7961c134c5bd73d87cd818bcddc30270804.1583136624.git.mchehab+huawei@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: convert dc395x.txt to ReST

Link: https://lore.kernel.org/r/3c0876df0045695185f922a0404c497a69de36a9.1583136624.git.mchehab+huawei@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: convert cxgb3i.txt to ReST

Link: https://lore.kernel.org/r/0708b62b6ec4f0dddc581e412bb02ba6476f4523.1583136624.git.mchehab+huawei@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: convert BusLogic.txt to ReST

Link: https://lore.kernel.org/r/750629b6a5233c85c5391c44d126606b8aabefc8.1583136624.git.mchehab+huawei@kernel.org
Acked-by: Khalid Aziz <khalid@gonehiking.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: convert bnx2fc.txt to ReST

Link: https://lore.kernel.org/r/f239116bd2c36f6fc8deb62e325bb8161da04270.1583136624.git.mchehab+huawei@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: convert bfa.txt to ReST

Link: https://lore.kernel.org/r/6660d0f83ddae2ab8efb31c39f9c220fc132e9d4.1583136624.git.mchehab+huawei@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: convert aic7xxx.txt to ReST

Link: https://lore.kernel.org/r/dc2b1ffe4bf64cfc4b32328740704a30e8d38a79.1583136624.git.mchehab+huawei@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: convert aic79xx.txt to ReST

Link: https://lore.kernel.org/r/e8a40337a2173f028c9ac569d3d71fd880f4fab5.1583136624.git.mchehab+huawei@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: convert aha152x.txt to ReST

Link: https://lore.kernel.org/r/097cfcc7f25343676a1fedcefed7e3b91b41b4df.1583136624.git.mchehab+huawei@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: convert advansys.txt to ReST

Link: https://lore.kernel.org/r/3c697a046e641c81cdfd0784f037d41d54766931.1583136624.git.mchehab+huawei@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: convert aacraid.txt to ReST

Link: https://lore.kernel.org/r/67c60ad88777c91937d49771e2a3f48cbf353e4c.1583136624.git.mchehab+huawei@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: convert 53c700.txt to ReST

Link: https://lore.kernel.org/r/a2e5116b70564f36b4fc7f1f1e5da1e693d7dadb.1583136624.git.mchehab+huawei@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: include SCSI Transport SRP diagram at the doc body

Instead of having a separate makefile, and be alone, group
it at the SCSI documentation and make it being built as part
of docs makefile.

Link: https://lore.kernel.org/r/419c455fb40c9a1e85cc9a654a7fdb07aeeccf71.1583136624.git.mchehab+huawei@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: docs: Add an empty index file for SCSI documents

In preparation for adding the SCSI documents to the documentation
body, add an empty index for it.

The next patches should be adding contents to it, as files get
converted to ReST format.

Link: https://lore.kernel.org/r/4d8c1b7ebe5898ac4a8265ca5e5a9552da3b426f.1583136624.git.mchehab+huawei@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: qla2xxx: add ring buffer for tracing debug logs

Having this log in a ring buffer helps to diagnose qla2xxx driver and
firmware issues instead of having to reproduce the problem with
extended_logging enabled. This saves cycles and helps when it is hard
to reproduce problem.

Link: https://lore.kernel.org/r/1581557368-32080-1-git-send-email-rajan.shanmugavelu@oracle.com
Reviewed-by: Joe Jin <joe.jin@oracle.com>
Acked-by: Himanshu Madhani <hmadhani@marvell.com>
Signed-off-by: Rajan Shanmugavelu <rajan.shanmugavelu@oracle.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: aacraid: clean up warning cast-function-type

Make the aacraid driver -Wcast-function-type clean

Report by: https://github.com/KSPP/linux/issues/20

drivers/scsi/aacraid/aachba.c:813:23:
warning: cast between incompatible function types from
'int (*)(struct scsi_cmnd *)' to 'void (*)(struct scsi_cmnd *)'
[-Wcast-function-type]

Link: https://lore.kernel.org/r/20200309155319.12658-1-tranmanphong@gmail.com
Reviewed-by: Bart van Assche <bvanassche@acm.org>
Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Phong Tran <tranmanphong@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: ufs-mediatek: fix HOST_PA_TACTIVATE quirk for Samsung UFS Devices

Device quirk "UFS_DEVICE_QUIRK_HOST_PA_TACTIVATE" is enabled for all
Samsung devices by default currently.

However MediaTek UFS host requires different host PA_TACTIVATE
configuration. Hence clear this quirk first and then apply vendor-specific
value in vops callback.

Link: https://lore.kernel.org/r/20200302135346.16797-1-stanley.chu@mediatek.com
Reviewed-by: Avri Altman <avri.altman@wdc.com>
Signed-off-by: Stanley Chu <stanley.chu@mediatek.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: ibmvfc: Fix NULL return compiler warning

Fix up a compiler warning introduced via 54b04c99d02e

Link: https://lore.kernel.org/r/1583159961-15903-1-git-send-email-brking@linux.vnet.ibm.com
Fixes: 54b04c99d02e ("scsi: ibmvfc: Avoid loss of all paths during SVC node reboot")
Signed-off-by: Brian King <brking@linux.vnet.ibm.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: fusion: fix if-statement empty body warning

When driver debugging is not enabled, change the debug print macros
to use the no_printk() macro.

This fixes a gcc warning when -Wextra is set:
../drivers/message/fusion/mptlan.c:266:39: warning: suggest braces around empty body in an ‘else’ statement [-Wempty-body]

I have verified that there is very little object code change (with
gcc 7.5.0). There are a few changes like:
  cmp %a,%b
  jl $1
to
  cmp %b,%a
  jg $1

Link: https://lore.kernel.org/r/ff9df31b-c4c1-c942-1cbf-18039e084c8e@infradead.org
Cc: Bart Van Assche <bvanassche@acm.org>
Cc: "James E.J. Bottomley" <jejb@linux.ibm.com>
Cc: "Martin K. Petersen" <martin.petersen@oracle.com>
Cc: linux-scsi@vger.kernel.org
Cc: MPT-FusionLinux.pdl@broadcom.com
Cc: Suganath Prabu Subramani <suganath-prabu.subramani@broadcom.com>
Cc: Chaitra P B <chaitra.basappa@broadcom.com>
Cc: Sathya Prakash <sathya.prakash@broadcom.com>
Reviewed-by: Bart van Assche <bvanassche@acm.org>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: iscsi: Add support for asynchronous iSCSI session destruction

iSCSI session destruction can be arbitrarily slow, since it might require
network operations and serialization inside the SCSI layer.  This patch
adds a new user event to trigger the destruction work asynchronously,
releasing the rx_queue_mutex as soon as the operation is queued and before
it is performed.  This change allows other operations to run in other
sessions in the meantime, removing one of the major iSCSI bottlenecks for
us.

To prevent the session from being used after the destruction request, we
remove it immediately from the sesslist. This simplifies the locking
required during the asynchronous removal.

Link: https://lore.kernel.org/r/20200227195945.761719-1-krisman@collabora.com
Co-developed-by: Gabriel Krisman Bertazi <krisman@collabora.com>
Co-developed-by: Khazhismel Kumykov <khazhy@google.com>
Reviewed-by: Lee Duncan <lduncan@suse.com>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>
Signed-off-by: Frank Mayhar <fmayhar@google.com>
Signed-off-by: Khazhismel Kumykov <khazhy@google.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: aacraid: Disabling TM path and only processing IOP reset

Fixes the occasional adapter panic when sg_reset is issued with -d, -t, -b
and -H flags.  Removal of command type HBA_IU_TYPE_SCSI_TM_REQ in
aac_hba_send since iu_type, request_id and fib_flags are not populated.
Device and target reset handlers are made to send TMF commands only when
reset_state is 0.

Link: https://lore.kernel.org/r/1581553771-25796-1-git-send-email-Sagar.Biradar@microchip.com
Reviewed-by: Sagar Biradar <Sagar.Biradar@microchip.com>
Signed-off-by: Sagar Biradar <Sagar.Biradar@microchip.com>
Signed-off-by: Balsundar P <balsundar.p@microsemi.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: ibmvfc: Avoid loss of all paths during SVC node reboot

When an SVC node goes down as part of a node reboot, its WWPNs are moved to
the remaining node. When the node is back online, its WWPNs are moved
back. The result is that the WWPN moves from one NPort_ID to another, then
back again.  The ibmvfc driver was forcing the old port to be removed, but
not sending an implicit logout. When the WWPN showed up at the new
location, the PLOGI failed as there was already a login established for the
old scsi id. The patch below fixes this by ensuring we always send an
implicit logout for any scsi id associated with an rport prior to calling
fc_remote_port_delete.

Link: https://lore.kernel.org/r/1582767943-16611-1-git-send-email-brking@linux.vnet.ibm.com
Signed-off-by: Brian King <brking@linux.vnet.ibm.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: Replace zero-length array with flexible-array member

The current codebase makes use of the zero-length array language extension
to the C90 standard, but the preferred mechanism to declare variable-length
types such as these ones is a flexible array member[1][2], introduced in
C99:

struct foo {
        int stuff;
        struct boo array[];
};

By making use of the mechanism above, we will get a compiler warning in
case the flexible array does not occur last in the structure, which will
help us prevent some kind of undefined behavior bugs from being
inadvertently introduced[3] to the codebase from now on.

Also, notice that, dynamic memory allocations won't be affected by this
change:

"Flexible array members have incomplete type, and so the sizeof operator
may not be applied. As a quirk of the original implementation of
zero-length arrays, sizeof evaluates to zero."[1]

This issue was found with the help of Coccinelle.

[1] https://gcc.gnu.org/onlinedocs/gcc/Zero-Length.html
[2] https://github.com/KSPP/linux/issues/21
[3] commit 76497732932f ("cxgb3/l2t: Fix undefined behaviour")

Link: https://lore.kernel.org/r/20200224161406.GA21454@embeddedor
Reviewed-by: Lee Duncan <lduncan@suse.com>
Reviewed-by: Satish Kharat <satishkh@cisco.com>
Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: ufs: Simplify two tests

lrbp->cmd is set only for SCSI commands. Use this knowledge to simplify two
boolean expressions.

Link: https://lore.kernel.org/r/20200123035637.21848-4-bvanassche@acm.org
Cc: Alim Akhtar <alim.akhtar@samsung.com>
Cc: Tomas Winkler <tomas.winkler@intel.com>
Cc: Bean Huo <beanhuo@micron.com>
Cc: Can Guo <cang@codeaurora.org>
Reviewed-by: Stanley Chu <stanley.chu@mediatek.com>
Reviewed-by: Asutosh Das <asutoshd@codeaurora.org>
Reviewed-by: Avri Altman <avri.altman@wdc.com>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: ufs: Introduce ufshcd_init_lrb()

This patch does not change any functionality but makes the next patch in
this series easier to read.

Link: https://lore.kernel.org/r/20200123035637.21848-3-bvanassche@acm.org
Cc: Tomas Winkler <tomas.winkler@intel.com>
Cc: Stanley Chu <stanley.chu@mediatek.com>
Cc: Bean Huo <beanhuo@micron.com>
Cc: Can Guo <cang@codeaurora.org>
Reviewed-by: Alim Akhtar <alim.akhtar@samsung.com>
Reviewed-by: Avri Altman <avri.altman@wdc.com>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: core: Introduce {init,exit}_cmd_priv()

The current behavior of the SCSI core is to clear driver-private data
before preparing a request for submission to the SCSI LLD. Make it possible
for SCSI LLDs to disable clearing of driver-private data.

These hooks will be used by a later patch, namely "scsi: ufs: Let the SCSI
core allocate per-command UFS data".

Link: https://lore.kernel.org/r/20200123035637.21848-2-bvanassche@acm.org
Cc: Tomas Winkler <tomas.winkler@intel.com>
Cc: Stanley Chu <stanley.chu@mediatek.com>
Cc: Bean Huo <beanhuo@micron.com>
Cc: Avri Altman <avri.altman@wdc.com>
Cc: Can Guo <cang@codeaurora.org>
Cc: Ming Lei <ming.lei@redhat.com>
Cc: Johannes Thumshirn <jth@kernel.org>
Cc: Hannes Reinecke <hare@suse.com>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: core: Remove cmd_list functionality

Remove cmd_list functionality; no users left.  With that the
scsi_put_command() becomes empty, so remove that one, too.

Link: https://lore.kernel.org/r/20200228075318.91255-14-hare@suse.de
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Bart van Assche <bvanassche@acm.org>
Signed-off-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: aacraid: use scsi_host_busy_iter() in get_num_of_incomplete_fibs()

Use the SCSI midlayer helper to traverse the number of outstanding
commands. This also eliminates the last usage for the cmd_list
functionality so we can drop it.

Link: https://lore.kernel.org/r/20200228075318.91255-13-hare@suse.de
Reviewed-by: Christoph Hellwig <hch@lst.de>
Acked-by: Balsundar P <balsundar.p@microchip.com>
Signed-off-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: aacraid: use scsi_host_busy_iter() to wait for outstanding commands

Instead of traversing the list of possible commands by hands we should be
using scsi_host_busy_iter() to figure out if there are outstanding
commands.

Link: https://lore.kernel.org/r/20200228075318.91255-12-hare@suse.de
Reviewed-by: Christoph Hellwig <hch@lst.de>
Acked-by: Balsundar P < Balsundar.P@microchip.com>
Signed-off-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: core: add scsi_host_busy_iter()

Add an iterator scsi_host_busy_iter() to traverse all busy commands.  If
locking against concurrent command completions is required, it has to be
provided by the caller.

Link: https://lore.kernel.org/r/20200228075318.91255-11-hare@suse.de
Signed-off-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: aacraid: use scsi_host_(block,unblock) to block I/O

Use scsi_host_block() and scsi_host_unblock() instead of
scsi_block_requests()/scsi_unblock_requests() to block and unblock I/O.
This has the advantage that the block layer will stop sending I/O to the
adapter instead of having the SCSI midlayer requeueing I/O internally.

Link: https://lore.kernel.org/r/20200228075318.91255-10-hare@suse.de
Reviewed-by: Christoph Hellwig <hch@lst.de>
Acked-by: Balsundar P < Balsundar.P@microchip.com>
Signed-off-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: core: add scsi_host_(block,unblock) helper function

Add helper functions to call scsi_internal_device_block()/
scsi_internal_device_unblock() for all attached devices on a SCSI host.

Link: https://lore.kernel.org/r/20200228075318.91255-9-hare@suse.de
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: aacraid: move scsi_(block,unblock)_requests out of _aac_reset_adapter()

_aac_reset_adapter() only has one caller, and that one already calls
scsi_block_requests(). Move the calls out of _aac_reset_adapter() to avoid
calling scsi_block_requests() twice.

Link: https://lore.kernel.org/r/20200228075318.91255-8-hare@suse.de
Reviewed-by: Christoph Hellwig <hch@lst.de>
Acked-by: Balsundar P < Balsundar.P@microchip.com>
Signed-off-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: aacraid: replace aac_flush_ios() with midlayer helper

Use the midlayer helper scsi_host_complete_all_commands() to flush all
outstanding commands.

Link: https://lore.kernel.org/r/20200228075318.91255-7-hare@suse.de
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Acked-by: Balsundar P <balsundar.p@microchip.com>
Signed-off-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: aacraid: use scsi_host_complete_all_commands() to terminate outstanding commands

Use scsi_host_complete_all_commands() to terminate all outstanding commands
and change the command result for terminated commands to the more common
'DID_RESET' instead of 'QUEUE_FULL'.

Link: https://lore.kernel.org/r/20200228075318.91255-6-hare@suse.de
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Acked-by: Balsundar P <balsundar.p@microchip.com>
Signed-off-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: aacraid: Do not wait for outstanding write commands on synchronize_cache

There is no need to wait for outstanding write commands on synchronize
cache; the block layer is responsible for I/O scheduling, no need to
out-guess it in the driver layer.

Link: https://lore.kernel.org/r/20200228075318.91255-5-hare@suse.de
Reviewed-by: Christoph Hellwig <hch@lst.de>
Acked-by: Balsundar P <balsundar.b@microchip.com>
Signed-off-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: dpt_i2o: use scsi_host_complete_all_commands() to abort outstanding commands

Rather than traversing all outstanding commands manually, use the
scsi_host_complete_all_commands() helper to terminate all commands during
reset.  With that we can drop the cmd_list usage from the midlayer.

Link: https://lore.kernel.org/r/20200228075318.91255-4-hare@suse.de
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

scsi: core: add scsi_host_complete_all_commands() helper

Add a helper scsi_host_complete_all_commands() to terminate all outstanding
commands on a SCSI host.

Link: https://lore.kernel.org/r/20200228075318.91255-3-hare@suse.de
Signed-off-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>