User experienced no task management error while target device is responding
with error. The RSP_CODE field in the status IOCB is in little endian.
Driver assumes it's big endian and it picked up erroneous data.

Convert the data back to big endian as is on the wire.

Link: https://lore.kernel.org/r/20220310092604.22950-2-njavali@marvell.com
Fixes: faef62d1 ("[SCSI] qla2xxx: Fix Task Management command asynchronous handling")
Cc: stable@vger.kernel.org
Reviewed-by: NHimanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: NQuinn Tran <qutran@marvell.com>
Signed-off-by: NNilesh Javali <njavali@marvell.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Make port_state_str and port_dstate_str a little more readable and
maintainable by using named initializers.

Also convert FCS_* macros into an enum.

Link: https://lore.kernel.org/r/AS8PR10MB495215841EB25C16DBC0CB409D349@AS8PR10MB4952.EURPRD10.PROD.OUTLOOK.COMReviewed-by: NHimanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: NGleb Chesnokov <Chesnokov.G@raidix.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Modify trace messages for additional debugability.

Link: https://lore.kernel.org/r/20211026115412.27691-9-njavali@marvell.comReviewed-by: NHimanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: NQuinn Tran <qutran@marvell.com>
Signed-off-by: NNilesh Javali <njavali@marvell.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Disable default logging of some I/O path messages. If desired, the messages
can be turned back on by setting ql2xextended_error_logging.

Link: https://lore.kernel.org/r/20210925035154.29815-1-njavali@marvell.comReviewed-by: NEwan D. Milne <emilne@redhat.com>
Reviewed-by: NHimanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: NArun Easi <aeasi@marvell.com>
Signed-off-by: NNilesh Javali <njavali@marvell.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

When Target port transitions personality from one to another (NVMe <-->
FCP), there could be some overlap of the two where one layer is going down
while the other layer is coming up. This overlap can cause temporary I/O
error. Detect those errors/transitions and recover from them. Triggers
session tear down and allow relogin to re-drive the connection under the
following conditions:

 - NVMe command error

 - On PRLO + N2N (rida format 2)

Link: https://lore.kernel.org/r/20210817051315.2477-11-njavali@marvell.comSigned-off-by: NQuinn Tran <qutran@marvell.com>
Signed-off-by: NNilesh Javali <njavali@marvell.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

When firmware indicates session has been torn down via UPDATE SA IOCB or
ELS Passthrough IOCB, the driver needs to also tear down the session.

Link: https://lore.kernel.org/r/20210817051315.2477-2-njavali@marvell.comSigned-off-by: NQuinn Tran <qutran@marvell.com>
Signed-off-by: NNilesh Javali <njavali@marvell.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

The MSI-X and MSI calls fails in kdump kernel. Because of this
qla2xxx_create_qpair() fails leading to .create_queue callback failure.
The fix is to return existing qpair instead of allocating new one and
allocate a single hw queue.

[   19.975838] qla2xxx [0000:d8:00.1]-00c7:11: MSI-X: Failed to enable support,
giving   up -- 16/-28.
[   19.984885] qla2xxx [0000:d8:00.1]-0037:11: Falling back-to MSI mode --
ret=-28.
[   19.992278] qla2xxx [0000:d8:00.1]-0039:11: Falling back-to INTa mode --
ret=-28.
..
..
..
[   21.141518] qla2xxx [0000:d8:00.0]-2104:2: qla_nvme_alloc_queue: handle
00000000e7ee499d, idx =1, qsize 32
[   21.151166] qla2xxx [0000:d8:00.0]-0181:2: FW/Driver is not multi-queue capable.
[   21.158558] qla2xxx [0000:d8:00.0]-2122:2: Failed to allocate qpair
[   21.164824] nvme nvme0: NVME-FC{0}: reset: Reconnect attempt failed (-22)
[   21.171612] nvme nvme0: NVME-FC{0}: Reconnect attempt in 2 seconds

Link: https://lore.kernel.org/r/20210810043720.1137-13-njavali@marvell.com
Cc: stable@vger.kernel.org
Reviewed-by: NHimanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: NSaurav Kashyap <skashyap@marvell.com>
Signed-off-by: NNilesh Javali <njavali@marvell.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Suppress logging of retryable errors. These can still be seen if extended
logging is enabled.

Link: https://lore.kernel.org/r/20210810043720.1137-11-njavali@marvell.comReviewed-by: NHimanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: NArun Easi <aeasi@marvell.com>
Signed-off-by: NNilesh Javali <njavali@marvell.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Add debug print of 64G link speed.

Link: https://lore.kernel.org/r/20210810043720.1137-7-njavali@marvell.comReviewed-by: NHimanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: NQuinn Tran <qutran@marvell.com>
Signed-off-by: NNilesh Javali <njavali@marvell.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Increment the command and the completion counts.

Link: https://lore.kernel.org/r/20210624052606.21613-11-njavali@marvell.comSigned-off-by: NQuinn Tran <qutran@marvell.com>
Signed-off-by: NNilesh Javali <njavali@marvell.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Some FC adapters from Marvell offer the ability to encrypt data in flight
(EDIF). This feature requires an application to act as an authenticator.

There is no FC switch scan service that can indicate whether a device is
secure or non-secure.

In order to detect whether the remote port supports encrypted operation,
driver must first do a PLOGI with the remote device. On completion of the
PLOGI, driver will query firmware to see if the device supports secure
login. To do that, driver + firmware must advertise the security bit via
PLOGI's service parameter. The remote device shall respond using the same
service parameter whether it supports it or not.

Link: https://lore.kernel.org/r/20210624052606.21613-8-njavali@marvell.comReviewed-by: NHannes Reinecke <hare@suse.de>
Reviewed-by: NHimanshu Madhani <himanshu.madhani@oracle.com>
Co-developed-by: NLarry Wisneski <Larry.Wisneski@marvell.com>
Signed-off-by: NLarry Wisneski <Larry.Wisneski@marvell.com>
Co-developed-by: NDuane Grigsby <duane.grigsby@marvell.com>
Signed-off-by: NDuane Grigsby <duane.grigsby@marvell.com>
Co-developed-by: NRick Hicksted Jr <rhicksted@marvell.com>
Signed-off-by: NRick Hicksted Jr <rhicksted@marvell.com>
Signed-off-by: NQuinn Tran <qutran@marvell.com>
Signed-off-by: NNilesh Javali <njavali@marvell.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Some FC adapters from Marvell offer the ability to encrypt data in flight
(EDIF). This feature requires an application to act as an authenticator.

As part of the authentication process, the authentication application will
generate a SADB entry (Security Association/SA, key, SPI value, etc). This
SADB is then passed to driver to be programmed into hardware. There will be
a pair of SADB's (Tx and Rx) for each connection.

After some period, the application can choose to change the key. At that
time, a new set of SADB pair is given to driver. The old set of SADB will
be deleted.

Add a new bsg call (QL_VND_SC_SA_UPDATE) to allow application to allow
adding or deleting SADB entries.  Driver will not keep the key in
memory. It will pass it to HW.

It is assumed that application will assign a unique SPI value to this SADB
(SA + key). Driver + hardware will assign a handle to track this unique
SPI/SADB.

Link: https://lore.kernel.org/r/20210624052606.21613-6-njavali@marvell.comReviewed-by: NHannes Reinecke <hare@suse.de>
Reviewed-by: NHimanshu Madhani <himanshu.madhani@oracle.com>
Co-developed-by: NLarry Wisneski <Larry.Wisneski@marvell.com>
Signed-off-by: NLarry Wisneski <Larry.Wisneski@marvell.com>
Co-developed-by: NDuane Grigsby <duane.grigsby@marvell.com>
Signed-off-by: NDuane Grigsby <duane.grigsby@marvell.com>
Co-developed-by: NRick Hicksted Jr <rhicksted@marvell.com>
Signed-off-by: NRick Hicksted Jr <rhicksted@marvell.com>
Signed-off-by: NQuinn Tran <qutran@marvell.com>
Signed-off-by: NNilesh Javali <njavali@marvell.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Some FC adapters from Marvell offer the ability to encrypt data in flight
(EDIF). This feature requires an application to act as an authenticator.

Once authentication messages sent from a remote device have arrived, each
message is extracted and placed in a buffer for application to retrieve.
The FC frame header will be stripped, leaving behind the AUTH ELS payload.
It is up to the application to strip the AUTH ELS header to get to the
actual authentication message.

Link: https://lore.kernel.org/r/20210624052606.21613-5-njavali@marvell.comReviewed-by: NHannes Reinecke <hare@suse.de>
Reviewed-by: NHimanshu Madhani <himanshu.madhani@oracle.com>
Co-developed-by: NLarry Wisneski <Larry.Wisneski@marvell.com>
Signed-off-by: NLarry Wisneski <Larry.Wisneski@marvell.com>
Co-developed-by: NDuane Grigsby <duane.grigsby@marvell.com>
Signed-off-by: NDuane Grigsby <duane.grigsby@marvell.com>
Co-developed-by: NRick Hicksted Jr <rhicksted@marvell.com>
Signed-off-by: NRick Hicksted Jr <rhicksted@marvell.com>
Signed-off-by: NQuinn Tran <qutran@marvell.com>
Signed-off-by: NNilesh Javali <njavali@marvell.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Some FC adapters from Marvell offer the ability to encrypt data in flight
(EDIF). This feature requires an application to act as an authenticator.

Add the ability for authentication application to send and retrieve
messages as part of the authentication process via existing
FC_BSG_HST_ELS_NOLOGIN BSG interface.

To send a message, application is expected to format the data in the AUTH
ELS format. Refer to FC-SP2 for details.

If a message was received, application is required to reply with either a
LS_ACC or LS_RJT complete the exchange using the same interface. Otherwise,
remote device will treat it as a timeout.

Link: https://lore.kernel.org/r/20210624052606.21613-4-njavali@marvell.comReviewed-by: NHannes Reinecke <hare@suse.de>
Reviewed-by: NHimanshu Madhani <himanshu.madhani@oracle.com>
Co-developed-by: NLarry Wisneski <Larry.Wisneski@marvell.com>
Signed-off-by: NLarry Wisneski <Larry.Wisneski@marvell.com>
Co-developed-by: NDuane Grigsby <duane.grigsby@marvell.com>
Signed-off-by: NDuane Grigsby <duane.grigsby@marvell.com>
Co-developed-by: NRick Hicksted Jr <rhicksted@marvell.com>
Signed-off-by: NRick Hicksted Jr <rhicksted@marvell.com>
Signed-off-by: NQuinn Tran <qutran@marvell.com>
Signed-off-by: NNilesh Javali <njavali@marvell.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Use "no-op" mailbox command to check if the adapter firmware is still
responsive.

Link: https://lore.kernel.org/r/20210619052427.6440-1-njavali@marvell.comReviewed-by: NHimanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: NQuinn Tran <qutran@marvell.com>
Signed-off-by: NNilesh Javali <njavali@marvell.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Introduce scsi_build_sense() as a wrapper around scsi_build_sense_buffer()
to format the buffer and set the correct SCSI status.

Link: https://lore.kernel.org/r/20210427083046.31620-8-hare@suse.deReviewed-by: NBart Van Assche <bvanassche@acm.org>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NHannes Reinecke <hare@suse.de>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Commit a6dcfe08 ("scsi: qla2xxx: Limit interrupt vectors to number of
CPUs") lowers the number of allocated MSI-X vectors to the number of CPUs.

That breaks vector allocation assumptions in qla83xx_iospace_config(),
qla24xx_enable_msix() and qla2x00_iospace_config(). Either of the functions
computes maximum number of qpairs as:

  ha->max_qpairs = ha->msix_count - 1 (MB interrupt) - 1 (default
                   response queue) - 1 (ATIO, in dual or pure target mode)

max_qpairs is set to zero in case of two CPUs and initiator mode. The
number is then used to allocate ha->queue_pair_map inside
qla2x00_alloc_queues(). No allocation happens and ha->queue_pair_map is
left NULL but the driver thinks there are queue pairs available.

qla2xxx_queuecommand() tries to find a qpair in the map and crashes:

  if (ha->mqenable) {
          uint32_t tag;
          uint16_t hwq;
          struct qla_qpair *qpair = NULL;

          tag = blk_mq_unique_tag(cmd->request);
          hwq = blk_mq_unique_tag_to_hwq(tag);
          qpair = ha->queue_pair_map[hwq]; # <- HERE

          if (qpair)
                  return qla2xxx_mqueuecommand(host, cmd, qpair);
  }

  BUG: kernel NULL pointer dereference, address: 0000000000000000
  #PF: supervisor read access in kernel mode
  #PF: error_code(0x0000) - not-present page
  PGD 0 P4D 0
  Oops: 0000 [#1] SMP PTI
  CPU: 0 PID: 72 Comm: kworker/u4:3 Tainted: G        W         5.10.0-rc1+ #25
  Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.0.0-prebuilt.qemu-project.org 04/01/2014
  Workqueue: scsi_wq_7 fc_scsi_scan_rport [scsi_transport_fc]
  RIP: 0010:qla2xxx_queuecommand+0x16b/0x3f0 [qla2xxx]
  Call Trace:
   scsi_queue_rq+0x58c/0xa60
   blk_mq_dispatch_rq_list+0x2b7/0x6f0
   ? __sbitmap_get_word+0x2a/0x80
   __blk_mq_sched_dispatch_requests+0xb8/0x170
   blk_mq_sched_dispatch_requests+0x2b/0x50
   __blk_mq_run_hw_queue+0x49/0xb0
   __blk_mq_delay_run_hw_queue+0xfb/0x150
   blk_mq_sched_insert_request+0xbe/0x110
   blk_execute_rq+0x45/0x70
   __scsi_execute+0x10e/0x250
   scsi_probe_and_add_lun+0x228/0xda0
   __scsi_scan_target+0xf4/0x620
   ? __pm_runtime_resume+0x4f/0x70
   scsi_scan_target+0x100/0x110
   fc_scsi_scan_rport+0xa1/0xb0 [scsi_transport_fc]
   process_one_work+0x1ea/0x3b0
   worker_thread+0x28/0x3b0
   ? process_one_work+0x3b0/0x3b0
   kthread+0x112/0x130
   ? kthread_park+0x80/0x80
   ret_from_fork+0x22/0x30

The driver should allocate enough vectors to provide every CPU it's own HW
queue and still handle reserved (MB, RSP, ATIO) interrupts.

The change fixes the crash on dual core VM and prevents unbalanced QP
allocation where nr_hw_queues is two less than the number of CPUs.

Link: https://lore.kernel.org/r/20210412165740.39318-1-r.bolshakov@yadro.com
Fixes: a6dcfe08 ("scsi: qla2xxx: Limit interrupt vectors to number of CPUs")
Cc: Daniel Wagner <daniel.wagner@suse.com>
Cc: Himanshu Madhani <himanshu.madhani@oracle.com>
Cc: Quinn Tran <qutran@marvell.com>
Cc: Nilesh Javali <njavali@marvell.com>
Cc: Martin K. Petersen <martin.petersen@oracle.com>
Cc: stable@vger.kernel.org # 5.11+
Reported-by: NAleksandr Volkov <a.y.volkov@yadro.com>
Reported-by: NAleksandr Miloserdov <a.miloserdov@yadro.com>
Reviewed-by: NDaniel Wagner <dwagner@suse.de>
Reviewed-by: NHimanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: NRoman Bolshakov <r.bolshakov@yadro.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

BUG: unable to handle kernel NULL pointer dereference at (null)
IP: qla2x00_abort_isp+0x21/0x6b0 [qla2xxx] PGD 0 P4D 0
Oops: 0000 [#1] SMP PTI
CPU: 0 PID: 1715 Comm: kworker/0:2
Tainted: GOE 4.12.14-122.37-default #1 SLE12-SP5
Hardware name: HPE Superdome Flex/Superdome Flex, BIOS
Bundle:3.30.100 SFW:IP147.007.004.017.000.2009211957 09/21/2020
Workqueue: events aer_recover_work_func
task: ffff9e399c14ca80 task.stack: ffffc1c58e4ac000
RIP: 0010:qla2x00_abort_isp+0x21/0x6b0 [qla2xxx]
RSP: 0018:ffffc1c58e4afd50 EFLAGS: 00010282
RAX: 0000000000000000 RBX: ffff9e419cdef480 RCX: 0000000000000000
RDX: ffff9e399c14ca80 RSI: 0000000000000246 RDI: ffff9e419bbc27b8
RBP: ffff9e419bbc27b8 R08: 0000000000000004 R09: 00000000a0440000
R10: 0000000000000000 R11: ffff9e399416d1a0 R12: ffff9e419cdef000
R13: ffff9e3a7cfae800 R14: ffff9e3a7cfae800 R15: 00000000000000c0
FS:  0000000000000000(0000) GS:ffff9e39a0000000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000000000000000 CR3: 00000006cd00a005 CR4: 00000000007606f0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
PKRU: 55555554
Call Trace:
  qla2xxx_pci_slot_reset+0x141/0x160 [qla2xxx]
  report_slot_reset+0x41/0x80
  ? merge_result.part.4+0x30/0x30
  pci_walk_bus+0x70/0x90
  pcie_do_recovery+0x1db/0x2e0
  aer_recover_work_func+0xc2/0xf0
  process_one_work+0x14c/0x390

Disable board_disable logic where driver resources are freed while OS is in
the process of recovering the adapter.

Link: https://lore.kernel.org/r/20210329085229.4367-9-njavali@marvell.comTested-by: NLaurence Oberman <loberman@redhat.com>
Signed-off-by: NQuinn Tran <qutran@marvell.com>
Signed-off-by: NNilesh Javali <njavali@marvell.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Fix all recently introduced endianness annotation issues.

Link: https://lore.kernel.org/r/20210320232359.941-4-bvanassche@acm.org
Cc: Quinn Tran <qutran@marvell.com>
Cc: Mike Christie <michael.christie@oracle.com>
Reviewed-by: NDaniel Wagner <dwagner@suse.de>
Reviewed-by: NHimanshu Madhani <himanshu.madhani@oracle.com>
Reviewed-by: NLee Duncan <lduncan@suse.com>
Signed-off-by: NBart Van Assche <bvanassche@acm.org>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Fix the following coccicheck warnings:

./drivers/scsi/qla2xxx/qla_isr.c:780:2-18: WARNING: Assignment
of 0/1 to bool variable.

Link: https://lore.kernel.org/r/1611127919-56551-1-git-send-email-abaci-bugfix@linux.alibaba.comReported-by: NAbaci Robot <abaci@linux.alibaba.com>
Signed-off-by: NJiapeng Zhong <abaci-bugfix@linux.alibaba.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

FW needs to wait for an ABTS response before completing the I/O.

Link: https://lore.kernel.org/r/20210111093134.1206-5-njavali@marvell.comReviewed-by: NHimanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: NBikash Hazarika <bhazarika@marvell.com>
Signed-off-by: NSaurav Kashyap <skashyap@marvell.com>
Signed-off-by: NArun Easi <aeasi@marvell.com>
Signed-off-by: NNilesh Javali <njavali@marvell.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

This change will aid in debugging issues arising because of dropped frame,
DIF errors, queue full etc where debug level is not set.

Link: https://lore.kernel.org/r/20210111093134.1206-4-njavali@marvell.comSigned-off-by: NSaurav Kashyap <skashyap@marvell.com>
Signed-off-by: NNilesh Javali <njavali@marvell.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

This statistics will help in debugging process and checking specific error
counts. It also provides a capability to isolate the port or bring it out
of isolation.

Link: https://lore.kernel.org/r/20210111093134.1206-2-njavali@marvell.comReported-by: Nkernel test robot <lkp@intel.com>
Reviewed-by: NHimanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: NSaurav Kashyap <skashyap@marvell.com>
Signed-off-by: NNilesh Javali <njavali@marvell.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

The completion status 0x28 (ppc = be = 0x2800) below indicates session is
not there, trigger session deletion.

qla2xxx [000b:04:00.1]-8009:8: DEVICE RESET ISSUED nexus=8:1:51 cmd=c000001432d0f600.
qla2xxx [000b:04:00.1]-5039:8: Async-tmf error - hdl=67b completion status(2800).
qla2xxx [000b:04:00.1]-8030:8: TM IOCB failed (102).
qla2xxx [000b:04:00.1]-800c:8: do_reset failed for cmd=c000001432d0f600.
qla2xxx [000b:04:00.1]-800f:8: DEVICE RESET FAILED: Task management failed nexus=8:1:51 cmd=c000001432d0f600.
qla2xxx [000b:04:00.1]-8009:8: DEVICE RESET ISSUED nexus=8:1:52 cmd=c000001432d0c200.
qla2xxx [000b:04:00.1]-5039:8: Async-tmf error - hdl=67c completion status(2800).
qla2xxx [000b:04:00.1]-8030:8: TM IOCB failed (102).

Link: https://lore.kernel.org/r/20201202132312.19966-5-njavali@marvell.comReviewed-by: NHimanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: NQuinn Tran <qutran@marvell.com>
Signed-off-by: NNilesh Javali <njavali@marvell.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Driver created too many QPairs(126) with 28xx adapter.  Limit to the number
of CPUs to minimize wasted resources.

Link: https://lore.kernel.org/r/20201202132312.19966-4-njavali@marvell.comReviewed-by: NHimanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: NQuinn Tran <qutran@marvell.com>
Signed-off-by: NNilesh Javali <njavali@marvell.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

qla2xx_process_get_sp_from_handle() will clear the slot in which the
current srb is stored. As a result it can't be used in
qla24xx_process_mbx_iocb_response() to check for consistency and later
again in qla24xx_mbx_iocb_entry().

Move the consistency check directly into qla24xx_mbx_iocb_entry() and avoid
the double call or any open coding of the
qla2xx_process_get_sp_from_handle() functionality.

Link: https://lore.kernel.org/r/20200929073802.18770-1-dwagner@suse.de
Fixes: 31a3271f ("scsi: qla2xxx: Handle incorrect entry_type entries")
Reviewed-by: NArun Easi <aeasi@marvell.com>
Reviewed-by: NHimanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: NDaniel Wagner <dwagner@suse.de>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

When printing the message:

  "MPI Heartbeat stop. MPI reset is not needed.."

..the wrong register was checked leading to always printing that MPI reset
is not needed, even when it is needed. Fix the MPI reset message.

Link: https://lore.kernel.org/r/20200929102152.32278-4-njavali@marvell.com
Fixes: cbb01c2f ("scsi: qla2xxx: Fix MPI failure AEN (8200) handling")
Cc: stable@vger.kernel.org
Reviewed-by: NHimanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: NArun Easi <aeasi@marvell.com>
Signed-off-by: NNilesh Javali <njavali@marvell.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

This patch tracks number of IOCB resources used in the I/O fast path. If
the number of used IOCBs reach a high water limit, driver would return the
I/O as busy and let upper layer retry. This prevents over subscription of
IOCB resources where any future error recovery command is unable to cut
through.  Enable IOCB throttling by default.

Link: https://lore.kernel.org/r/20200904045128.23631-12-njavali@marvell.comReviewed-by: NHimanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: NQuinn Tran <qutran@marvell.com>
Signed-off-by: NArun Easi <aeasi@marvell.com>
Signed-off-by: NNilesh Javali <njavali@marvell.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Move statistics fields from vha struct to qpair to reduce memory thrashing.

Link: https://lore.kernel.org/r/20200904045128.23631-8-njavali@marvell.comReviewed-by: NHimanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: NQuinn Tran <qutran@marvell.com>
Signed-off-by: NNilesh Javali <njavali@marvell.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

FCP-4 (referred FCP-4 rev-2b) identifies the earlier known "retry delay
timer" field as "status qualifier", which is described in SAM-5 and later
specs. This fix makes appropriate driver side modifications to honor the
new definition. The SAM document referred was SAM-6 rev-5.

Link: https://lore.kernel.org/r/20200904045128.23631-5-njavali@marvell.comReviewed-by: NHimanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: NArun Easi <aeasi@marvell.com>
Signed-off-by: NHimanshu Madhani <hmadhani@marvell.com>
Signed-off-by: NNilesh Javali <njavali@marvell.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

All files in this driver directory contain the following notice:

  See LICENSE.qla2xxx for copyright and licensing details.

LICENSE.qla2xxx can be found in Documentation/scsi/. The file contains:

  - A copyright notice

    This copyright notice is redundant as all files contain the same
    copyright notice already

  - A license notice

    You may modify and redistribute the device driver code under the
    GNU General Public License (a copy of which is attached hereto as
    Exhibit A) published by the Free Software Foundation (version 2).

    This can be replaced with the corresponding SPDX license identifier
    (GPL-2.0-only) in the source files which reference this license
    file.

  - The full GPLv2 license text

    A redundant copy of LICENSES/preferred/GPL-2.0

Remove the notices and add the SPDX license identifier GPL-2.0-only to the
source files.

Finally remove the now redundant LICENSE.qla2xxx file.
Reviewed-by: NRichard Fontana <rfontana@redhat.com>
Reviewed-by: NJilayne Lovejoy <opensource@jilayne.com>
Reviewed-by: NAlexios Zavras <alexios.zavras@intel.com>
Acked-by: NIgor Russkikh <irusskikh@marvell.com>
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Signed-off-by: NLukas Bulwahn <lukas.bulwahn@gmail.com>
Acked-by: NNilesh Javali <njavali@marvell.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

It was observed on an ISP8324 16Gb HBA with fw=8.08.203 (d0d5) in a
PowerPC64 machine that pkt->entry_type was MBX_IOCB_TYPE/0x39 with an
sp->type SRB_SCSI_CMD which is invalid and should not be possible.

Reading the entry_type from the crash dump shows the expected value of
STATUS_TYPE/0x03 but the call trace shows that qla24xx_mbx_iocb_entry() is
used.

Add a check to verify for consistency and reset the HBA if an invalid state
is reached. Obviously, this is only a workaround until the real problem is
solved.

Link: https://lore.kernel.org/r/20200908081516.8561-5-dwagner@suse.deReviewed-by: NArun Easi <aeasi@marvell.com>
Signed-off-by: NDaniel Wagner <dwagner@suse.de>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Commit 7c3df132 ("[SCSI] qla2xxx: Code changes to support new dynamic
logging infrastructure.") removed the use of the func argument. Let's add
it back.

Link: https://lore.kernel.org/r/20200908081516.8561-4-dwagner@suse.deReviewed-by: NArun Easi <aeasi@marvell.com>
Signed-off-by: NDaniel Wagner <dwagner@suse.de>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Refactor qla2x00_get_sp_from_handle() to avoid the unnecessary goto if
early returns are used. With this we can also avoid preinitialzing the sp
pointer.

Link: https://lore.kernel.org/r/20200908081516.8561-3-dwagner@suse.deReviewed-by: NMartin Wilck <mwilck@suse.com>
Reviewed-by: NArun Easi <aeasi@marvell.com>
Signed-off-by: NDaniel Wagner <dwagner@suse.de>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Replace the existing /* fall through */ comments and its variants with
the new pseudo-keyword macro fallthrough[1]. Also, remove unnecessary
fall-through markings when it is the case.

[1] https://www.kernel.org/doc/html/v5.7/process/deprecated.html?highlight=fallthrough#implicit-switch-case-fall-throughSigned-off-by: NGustavo A. R. Silva <gustavoars@kernel.org>

Update debug level and message for ELS IOCB done.

Link: https://lore.kernel.org/r/20200806111014.28434-6-njavali@marvell.comReviewed-by: NHimanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: NQuinn Tran <qutran@marvell.com>
Signed-off-by: NNilesh Javali <njavali@marvell.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Fix the following sparse warnings:

drivers/scsi/qla2xxx/qla_isr.c:881:23: warning: restricted __le16 degrades to integer
drivers/scsi/qla2xxx/qla_isr.c:881:23: warning: cast to restricted __le16

Link: https://lore.kernel.org/r/20200715043358.21668-1-njavali@marvell.comSigned-off-by: NShyam Sundar <ssundar@marvell.com>
Signed-off-by: NNilesh Javali <njavali@marvell.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

* Firmware Initialization with SCM enabled based on NVRAM setting and
  firmware support (About Firmware).

* Enable PUREX and add support for fabric performance impact
  notification (FPIN) handling.

* Allocate a default PUREX item for each vha to handle memory allocation
  failures in ISR.

Link: https://lore.kernel.org/r/20200630102229.29660-3-njavali@marvell.comReviewed-by: NHimanshu Madhani <himanshu.madhani@oracle.com>
Reviewed-by: NJames Smart <james.smart@broadcom.com>
Signed-off-by: NShyam Sundar <ssundar@marvell.com>
Signed-off-by: NArun Easi <aeasi@marvell.com>
Signed-off-by: NNilesh Javali <njavali@marvell.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

SAN Congestion Management generates ELS pkts whose size can vary and be >
64 bytes. Change the PUREX handling code to support non-standard ELS pkt
size.

Link: https://lore.kernel.org/r/20200630102229.29660-2-njavali@marvell.comReviewed-by: NHimanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: NShyam Sundar <ssundar@marvell.com>
Signed-off-by: NArun Easi <aeasi@marvell.com>
Signed-off-by: NNilesh Javali <njavali@marvell.com>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>

Fix all endianness complaints reported by sparse (C=2) without affecting
the behavior of the code on little endian CPUs.

Link: https://lore.kernel.org/r/20200518211712.11395-16-bvanassche@acm.org
Cc: Nilesh Javali <njavali@marvell.com>
Cc: Quinn Tran <qutran@marvell.com>
Cc: Martin Wilck <mwilck@suse.com>
Cc: Daniel Wagner <dwagner@suse.de>
Cc: Roman Bolshakov <r.bolshakov@yadro.com>
Reviewed-by: NHimanshu Madhani <himanshu.madhani@oracle.com>
Reviewed-by: NHannes Reinecke <hare@suse.de>
Reviewed-by: NDaniel Wagner <dwagner@suse.de>
Signed-off-by: NBart Van Assche <bvanassche@acm.org>
Signed-off-by: NMartin K. Petersen <martin.petersen@oracle.com>