OSDN Git Service
Martin K. Petersen [Tue, 15 Feb 2022 02:51:29 +0000 (21:51 -0500)]
Merge branch '5.17/scsi-fixes' into 5.18/scsi-staging
Pull 5.17 fixes branch into 5.18 tree to resolve a few pm8001 driver
merge conflicts.
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Sreekanth Reddy [Thu, 10 Feb 2022 09:58:17 +0000 (15:28 +0530)]
scsi: mpi3mr: Bump driver version to 8.0.0.68.0
Bump mpi3mr driver version to 8.0.0.68.0.
Link: https://lore.kernel.org/r/20220210095817.22828-10-sreekanth.reddy@broadcom.com
Signed-off-by: Sreekanth Reddy <sreekanth.reddy@broadcom.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Sreekanth Reddy [Thu, 10 Feb 2022 09:58:16 +0000 (15:28 +0530)]
scsi: mpi3mr: Fix memory leaks
Fix memory leaks related to operational reply queue's memory segments which
are not getting freed while unloading the driver.
Link: https://lore.kernel.org/r/20220210095817.22828-9-sreekanth.reddy@broadcom.com
Signed-off-by: Sreekanth Reddy <sreekanth.reddy@broadcom.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Sreekanth Reddy [Thu, 10 Feb 2022 09:58:15 +0000 (15:28 +0530)]
scsi: mpi3mr: Update the copyright year
Update the copyright year to 2017-2022.
Link: https://lore.kernel.org/r/20220210095817.22828-8-sreekanth.reddy@broadcom.com
Signed-off-by: Sreekanth Reddy <sreekanth.reddy@broadcom.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Sreekanth Reddy [Thu, 10 Feb 2022 09:58:14 +0000 (15:28 +0530)]
scsi: mpi3mr: Fix reporting of actual data transfer size
The driver is missing to set the residual size while completing an
I/O. Ensure proper data transfer size is reported to the kernel on I/O
completion based on the transfer length reported by the firmware.
Link: https://lore.kernel.org/r/20220210095817.22828-7-sreekanth.reddy@broadcom.com
Signed-off-by: Sreekanth Reddy <sreekanth.reddy@broadcom.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Sreekanth Reddy [Thu, 10 Feb 2022 09:58:13 +0000 (15:28 +0530)]
scsi: mpi3mr: Fix cmnd getting marked as in use forever
When a driver command which requires the driver to issue a follow up
command using the same command frame is outstanding and a soft reset
operation occurs, then that driver command frame is getting marked as in
use permanently and won't be reused again.
Clear the driver command frames while flushing out the outstanding commands
and avoid issuing any new requests using these command frames while soft
reset is going on.
Link: https://lore.kernel.org/r/20220210095817.22828-6-sreekanth.reddy@broadcom.com
Signed-off-by: Sreekanth Reddy <sreekanth.reddy@broadcom.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Sreekanth Reddy [Thu, 10 Feb 2022 09:58:12 +0000 (15:28 +0530)]
scsi: mpi3mr: Fix hibernation issue
Hibernation operation fails when it is issued for second time. This is
because the driver is trying to release the IOC's PCI resources after
setting power state to D3.
Set the IOC's power state to D3 only after releasing the IOC's PCI
resources.
Link: https://lore.kernel.org/r/20220210095817.22828-5-sreekanth.reddy@broadcom.com
Signed-off-by: Sreekanth Reddy <sreekanth.reddy@broadcom.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Sreekanth Reddy [Thu, 10 Feb 2022 09:58:11 +0000 (15:28 +0530)]
scsi: mpi3mr: Update MPI3 headers
Update MPI3 headers.
Link: https://lore.kernel.org/r/20220210095817.22828-4-sreekanth.reddy@broadcom.com
Signed-off-by: Sreekanth Reddy <sreekanth.reddy@broadcom.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Sreekanth Reddy [Thu, 10 Feb 2022 09:58:10 +0000 (15:28 +0530)]
scsi: mpi3mr: Fix printing of pending I/O count
Print proper pending I/O count after issuing target reset TM operation.
Link: https://lore.kernel.org/r/20220210095817.22828-3-sreekanth.reddy@broadcom.com
Signed-off-by: Sreekanth Reddy <sreekanth.reddy@broadcom.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Sreekanth Reddy [Thu, 10 Feb 2022 09:58:09 +0000 (15:28 +0530)]
scsi: mpi3mr: Fix deadlock while canceling the fw event
During controller reset, the driver tries to flush all the pending firmware
event works from worker queue that are queued prior to the reset. However,
if any work is waiting for device addition/removal operation to be
completed at the SML, then a deadlock is observed. This is due to the
controller reset waiting for the device addition/removal to be completed
and the device/addition removal is waiting for the controller reset to be
completed.
To limit this deadlock, continue with the controller reset handling without
canceling the work which is waiting for device addition/removal operation
to complete at SML.
Link: https://lore.kernel.org/r/20220210095817.22828-2-sreekanth.reddy@broadcom.com
Signed-off-by: Sreekanth Reddy <sreekanth.reddy@broadcom.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Xiang Chen [Fri, 11 Feb 2022 06:42:58 +0000 (14:42 +0800)]
scsi: Remove unused member cmd_pool for structure scsi_host_template
After commit
e9c787e65c0c ("scsi: allocate scsi_cmnd structures as part of
struct request"), the member cmd_pool in structure scsi_host_template is
not used, so remove it.
Link: https://lore.kernel.org/r/1644561778-183074-5-git-send-email-chenxiang66@hisilicon.com
Reviewed-by: John Garry <john.garry@huawei.com>
Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Xiang Chen [Fri, 11 Feb 2022 06:42:57 +0000 (14:42 +0800)]
scsi: libsas: Remove unused parameter for function sas_ata_eh()
Input parameter work_q is not unused in function sas_ata_eh(), so remove
it.
Link: https://lore.kernel.org/r/1644561778-183074-4-git-send-email-chenxiang66@hisilicon.com
Reviewed-by: John Garry <john.garry@huawei.com>
Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Xiang Chen [Fri, 11 Feb 2022 06:42:56 +0000 (14:42 +0800)]
scsi: libsas: Remove duplicated setting for task->task_state_flags
Task->task_state_flags is already set in function sas_alloc_task(), so
remove duplicated setting.
Link: https://lore.kernel.org/r/1644561778-183074-3-git-send-email-chenxiang66@hisilicon.com
Reviewed-by: John Garry <john.garry@huawei.com>
Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Xiang Chen [Fri, 11 Feb 2022 06:42:55 +0000 (14:42 +0800)]
scsi: libsas: Use void for sas_discover_event() return code
The callers of function sas_discover_event() do not check its return value.
The function also only ever returns 0, so use void instead.
Link: https://lore.kernel.org/r/1644561778-183074-2-git-send-email-chenxiang66@hisilicon.com
Reviewed-by: John Garry <john.garry@huawei.com>
Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Julia Lawall [Thu, 10 Feb 2022 20:42:22 +0000 (21:42 +0100)]
scsi: message: fusion: Use GFP_KERNEL
Pci_driver probe functions aren't called with locks held and thus don't
need GFP_ATOMIC. Use GFP_KERNEL instead.
Problem found with Coccinelle.
Link: https://lore.kernel.org/r/20220210204223.104181-9-Julia.Lawall@inria.fr
Signed-off-by: Julia Lawall <Julia.Lawall@inria.fr>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Don Brace [Thu, 10 Feb 2022 20:11:51 +0000 (14:11 -0600)]
scsi: smartpqi: Fix unused variable pqi_pm_ops for clang
Driver added a new dev_pm_ops structure used only if CONFIG_PM is set. The
CONFIG_PM MACRO needed to be moved up in the code to avoid the compiler
warnings. The hunk to move the location was missing from the above patch.
Found by kernel test robot by building driver with CONFIG_PM disabled.
Link: https://lore.kernel.org/all/202202090657.bstNLuce-lkp@intel.com/
Link: https://lore.kernel.org/r/20220210201151.236170-1-don.brace@microchip.com
Fixes:
c66e078ad89e ("scsi: smartpqi: Fix hibernate and suspend")
Reported-by: kernel test robot <lkp@intel.com>
Reviewed-by: Scott Teel <scott.teel@microchip.com>
Reviewed-by: Scott Benesh <scott.benesh@microchip.com>
Reviewed-by: Mike Mcgowen <mike.mcgowen@microchip.com>
Reviewed-by: Kevin Barnett <kevin.barnett@microchip.com>
Signed-off-by: Don Brace <don.brace@microchip.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
John Garry [Thu, 10 Feb 2022 10:43:24 +0000 (18:43 +0800)]
scsi: libsas: Drop SAS_TASK_AT_INITIATOR
This flag is now only ever set, so delete it.
This also avoids a use-after-free in the pm8001 queue path, as reported in
the following:
https://lore.kernel.org/linux-scsi/
c3cb7228-254e-9584-182b-
007ac5e6fe0a@huawei.com/T/#m28c94c6d3ff582ec4a9fa54819180740e8bd4cfb
https://lore.kernel.org/linux-scsi/
0cc0c435-b4f2-9c76-258d-
865ba50a29dd@huawei.com/
[mkp: checkpatch + two SAS_TASK_AT_INITIATOR references]
Link: https://lore.kernel.org/r/1644489804-85730-3-git-send-email-john.garry@huawei.com
Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
Signed-off-by: John Garry <john.garry@huawei.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
John Garry [Thu, 10 Feb 2022 10:43:23 +0000 (18:43 +0800)]
scsi: isci: Drop SAS_TASK_AT_INITIATOR check in isci_task_abort_task()
In the queue path, move around when we assign sas_task->lldd_task such that
this pointer and the SAS_TASK_AT_INITIATOR flag are set atomically. It is
also not required to clear SAS_TASK_AT_INITIATOR in isci_task_execute_task()
error path as it is also cleared immediately after in isci_task_refuse()
call.
Now the following items may be considered:
- SAS_TASK_STATE_DONE and SAS_TASK_AT_INITIATOR are mutually exclusive
apart from possibly when SAS_TASK_STATE_DONE is set in
sas_scsi_find_task(), but that is after .lldd_abort_task, i.e. the
considered callback, is called.
- If isci_task_refuse() is called in the queue path, then
sas_task->lldd_task and SAS_TASK_AT_INITIATOR are cleared atomically in
isci_task_refuse().
- In the completion path, SAS_TASK_STATE_DONE is set and
SAS_TASK_AT_INITIATOR is cleared atomically before the
sas_task.lldd_task is cleared later.
So in isci_task_abort_task() if SAS_TASK_STATE_DONE is not set and
sas_task.lldd_task is still set, then SAS_TASK_AT_INITIATOR must be set -
so we can drop this check on SAS_TASK_AT_INITIATOR.
[mkp: checkpatch]
Link: https://lore.kernel.org/r/1644489804-85730-2-git-send-email-john.garry@huawei.com
Signed-off-by: John Garry <john.garry@huawei.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Gleb Chesnokov [Tue, 8 Feb 2022 15:18:38 +0000 (15:18 +0000)]
scsi: qla2xxx: Remove unused qla_sess_op_cmd_list from scsi_qla_host_t
The qla_sess_op_cmd_list was introduced in commit
8b2f5ff3d05c ("qla2xxx:
cleanup cmd in qla workqueue before processing TMR"). Then the usage of
this list was dropped in commit
fb35265b12bb ("scsi: qla2xxx: Remove
session creation redundant code").
Thus, remove this list since it is no longer used.
Link: https://lore.kernel.org/r/AS8PR10MB49524AAB4C8016E4AFF17FFB9D2D9@AS8PR10MB4952.EURPRD10.PROD.OUTLOOK.COM
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: Gleb Chesnokov <Chesnokov.G@raidix.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Yang Li [Tue, 8 Feb 2022 02:55:00 +0000 (10:55 +0800)]
scsi: pm8001: Clean up inconsistent indenting
Eliminate the following smatch warning:
drivers/scsi/pm8001/pm8001_ctl.c:760 pm8001_update_flash() warn:
inconsistent indenting
Link: https://lore.kernel.org/r/20220208025500.29511-1-yang.lee@linux.alibaba.com
Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Acked-by: Jack Wang <jinpu.wang@ionos.com>
Signed-off-by: Yang Li <yang.lee@linux.alibaba.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Kees Cook [Tue, 8 Feb 2022 06:12:31 +0000 (22:12 -0800)]
scsi: ibmvscsis: Silence -Warray-bounds warning
Instead of doing a cast to storage that is too small, add a union for the
high 64 bits. Silences the warnings under -Warray-bounds:
drivers/scsi/ibmvscsi_tgt/ibmvscsi_tgt.c: In function 'ibmvscsis_send_messages':
drivers/scsi/ibmvscsi_tgt/ibmvscsi_tgt.c:1934:44: error: array subscript 'struct viosrp_crq[0]' is partly outside array bounds of 'u64[1]' {aka 'long long unsigned int[1]'} [-Werror=array-bounds]
1934 | crq->valid = VALID_CMD_RESP_EL;
| ^~
drivers/scsi/ibmvscsi_tgt/ibmvscsi_tgt.c:1875:13: note: while referencing 'msg_hi'
1875 | u64 msg_hi = 0;
| ^~~~~~
There is no change to the resulting binary instructions.
Link: https://lore.kernel.org/lkml/20220125142430.75c3160e@canb.auug.org.au
Link: https://lore.kernel.org/r/20220208061231.3429486-1-keescook@chromium.org
Cc: Michael Cyr <mikecyr@linux.ibm.com>
Cc: "James E.J. Bottomley" <jejb@linux.ibm.com>
Cc: "Martin K. Petersen" <martin.petersen@oracle.com>
Cc: Tyrel Datwyler <tyreld@linux.ibm.com>
Cc: linux-scsi@vger.kernel.org
Cc: target-devel@vger.kernel.org
Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Reviewed-by: Tyrel Datwyler <tyreld@linux.ibm.com>
Signed-off-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Saurav Kashyap [Tue, 8 Feb 2022 09:39:46 +0000 (01:39 -0800)]
scsi: qla2xxx: Add qla2x00_async_done() for async routines
This done routine will delete the timer and check for its return value and
decrease the reference count accordingly. This prevents boot hangs reported
after commit
31e6cdbe0eae ("scsi: qla2xxx: Implement ref count for SRB")
was merged.
Link: https://lore.kernel.org/r/20220208093946.4471-1-njavali@marvell.com
Fixes:
31e6cdbe0eae ("scsi: qla2xxx: Implement ref count for SRB")
Reported-by: Ewan Milne <emilne@redhat.com>
Tested-by: Ewan D. Milne <emilne@redhat.com>
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: Saurav Kashyap <skashyap@marvell.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
James Smart [Mon, 7 Feb 2022 18:04:42 +0000 (10:04 -0800)]
scsi: lpfc: Reduce log messages seen after firmware download
Messages around firmware download were incorrectly tagged as being related
to discovery trace events. Thus, firmware download status ended up dumping
the trace log as well as the firmware update message. As there were a
couple of log messages in this state, the trace log was dumped multiple
times.
Resolve this by converting from trace events to SLI events.
Link: https://lore.kernel.org/r/20220207180442.72836-1-jsmart2021@gmail.com
Reviewed-by: Ewan D. Milne <emilne@redhat.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
James Smart [Mon, 7 Feb 2022 18:05:16 +0000 (10:05 -0800)]
scsi: lpfc: Remove NVMe support if kernel has NVME_FC disabled
The driver is initiating NVMe PRLIs to determine device NVMe support. This
should not be occurring if CONFIG_NVME_FC support is disabled.
Correct this by changing the default value for FC4 support. Currently it
defaults to FCP and NVMe. With change, when NVME_FC support is not enabled
in the kernel, the default value is just FCP.
Link: https://lore.kernel.org/r/20220207180516.73052-1-jsmart2021@gmail.com
Reviewed-by: Ewan D. Milne <emilne@redhat.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Don Brace [Tue, 1 Feb 2022 21:49:18 +0000 (15:49 -0600)]
scsi: smartpqi: Update version to 2.1.14-035
Link: https://lore.kernel.org/r/164375215867.440833.17567317655622946368.stgit@brunhilda.pdev.net
Reviewed-by: Kevin Barnett <kevin.barnett@microchip.com>
Reviewed-by: Scott Benesh <scott.benesh@microchip.com>
Reviewed-by: Scott Teel <scott.teel@microchip.com>
Reviewed-by: Gerry Morong <gerry.morong@microchip.com>
Signed-off-by: Don Brace <don.brace@microchip.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Kevin Barnett [Tue, 1 Feb 2022 21:49:13 +0000 (15:49 -0600)]
scsi: smartpqi: Fix lsscsi -t SAS addresses
Correct lsscsi -t output for newer controllers that support 16-byte WWID in
the SAS address field. lsscsi -t was displaying all zeros for SAS
addresses.
When we added support to smartpqi for 16-byte WWIDs in the RPL data for
newer controllers, we were copying the wrong part of the 16-byte WWID to
the SAS address field.
Link: https://lore.kernel.org/r/164375215363.440833.7298523719813806902.stgit@brunhilda.pdev.net
Reviewed-by: Scott Benesh <scott.benesh@microchip.com>
Reviewed-by: Scott Teel <scott.teel@microchip.com>
Reviewed-by: Mike McGowen <mike.mcgowen@microchip.com>
Signed-off-by: Kevin Barnett <kevin.barnett@microchip.com>
Signed-off-by: Don Brace <don.brace@microchip.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Kevin Barnett [Tue, 1 Feb 2022 21:49:08 +0000 (15:49 -0600)]
scsi: smartpqi: Fix hibernate and suspend
Restructure the hibernate/suspend code to allow workarounds for the
controller boot differences.
Newer controllers have subtle differences in the way that they boot up.
Link: https://lore.kernel.org/r/164375214859.440833.14683009064111314948.stgit@brunhilda.pdev.net
Reviewed-by: Mike McGowen <mike.mcgowen@microchip.com>
Reviewed-by: Scott Benesh <scott.benesh@microchip.com>
Reviewed-by: Scott Teel <scott.teel@microchip.com>
Signed-off-by: Kevin Barnett <kevin.barnett@microchip.com>
Signed-off-by: Don Brace <don.brace@microchip.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Mike McGowen [Tue, 1 Feb 2022 21:49:03 +0000 (15:49 -0600)]
scsi: smartpqi: Fix BUILD_BUG_ON() statements
Add calls to the functions at the beginning driver initialization.
The BUILD_BUG_ON() statements that are currently in functions named
verify_structures() in the modules smartpqi_init.c and smartpqi_sis.c do
not work as currently implemented.
Link: https://lore.kernel.org/r/164375214355.440833.13129778749209816497.stgit@brunhilda.pdev.net
Reviewed-by: Kevin Barnett <kevin.barnett@microchip.com>
Reviewed-by: Scott Teel <scott.teel@microchip.com>
Reviewed-by: Scott Benesh <scott.benesh@microchip.com>
Signed-off-by: Mike McGowen <Mike.McGowen@microchip.com>
Signed-off-by: Don Brace <don.brace@microchip.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Mike McGowen [Tue, 1 Feb 2022 21:48:58 +0000 (15:48 -0600)]
scsi: smartpqi: Fix NUMA node not updated during init
Correct NUMA node association when calling pqi_pci_probe().
In the function pqi_pci_probe(), the driver makes an OS call to get the
NUMA node associated with a controller. If the call returns that there is
no associated node, the driver attempts to set it to node 0. The problem is
that a different local variable (cp_node) was being used to do this, but
the original local variable (node) was still being used in the call to
pqi_alloc_ctrl_info().
The value of "node" is not updated if the conditional after the call to
dev_to_node() evaluates to TRUE.
Link: https://lore.kernel.org/r/164375213850.440833.5243014942807747074.stgit@brunhilda.pdev.net
Reviewed-by: Kevin Barnett <kevin.barnett@microchip.com>
Reviewed-by: Scott Teel <scott.teel@microchip.com>
Signed-off-by: Mike McGowen <Mike.McGowen@microchip.com>
Signed-off-by: Don Brace <don.brace@microchip.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Kevin Barnett [Tue, 1 Feb 2022 21:48:53 +0000 (15:48 -0600)]
scsi: smartpqi: Expose SAS address for SATA drives
Remove UNIQUE_WWID_IN_REPORT_PHYS_LUN PQI feature.
This feature was originally added to solve a problem with NVMe drives, but
this problem has since been solved a different way, so this PQI feature is
no longer required for any type of drive.
The kernel was not creating symbolic links in sysfs between SATA drives and
their enclosure.
The driver was enabling the UNIQUE_WWID_IN_REPORT_PHYS_LUN PQI feature,
which causes the FW to return a WWID for SATA drives that is derived from a
unique ID read from the SATA drive itself. The driver was exposing this
WWID as the drive's SAS address. However, because this SAS address does not
match the SAS address returned by an enclosure's SES Page 0xA data, the
Linux kernel was not able to match a SATA drive with its enclosure.
Link: https://lore.kernel.org/r/164375213346.440833.12379222470149882747.stgit@brunhilda.pdev.net
Reviewed-by: Mike McGowen <mike.mcgowen@microchip.com>
Reviewed-by: Scott Benesh <scott.benesh@microchip.com>
Reviewed-by: Scott Teel <scott.teel@microchip.com>
Signed-off-by: Kevin Barnett <kevin.barnett@microchip.com>
Signed-off-by: Don Brace <don.brace@microchip.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Mike McGowen [Tue, 1 Feb 2022 21:48:48 +0000 (15:48 -0600)]
scsi: smartpqi: Speed up RAID 10 sequential reads
Use all data disks for sequential read operations.
Testing discovered inconsistent performance on RAID 10 volumes when
performing 256K sequential reads. The driver was only using a single
tracker to determine which physical drive to send a request to for AIO
requests.
Change the single tracker (next_bypass_group) to an array of trackers based
on the number of data disks in a row of the RAID map.
Link: https://lore.kernel.org/r/164375212842.440833.6733971458765002128.stgit@brunhilda.pdev.net
Reviewed-by: Kevin Barnett <kevin.barnett@microchip.com>
Reviewed-by: Mike McGowen <mike.mcgowen@microchip.com>
Reviewed-by: Scott Benesh <scott.benesh@microchip.com>
Reviewed-by: Scott Teel <scott.teel@microchip.com>
Signed-off-by: Mike McGowen <Mike.McGowen@microchip.com>
Signed-off-by: Don Brace <don.brace@microchip.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Mahesh Rajashekhara [Tue, 1 Feb 2022 21:48:43 +0000 (15:48 -0600)]
scsi: smartpqi: Fix kdump issue when controller is locked up
Avoid dropping into shell if the controller is in locked up state.
Driver issues SIS soft reset to bring back the controller to SIS mode while
OS boots into kdump mode.
If the controller is in lockup state, SIS soft reset does not work.
Since the controller lockup code has not been cleared, driver considers the
firmware is no longer up and running. Driver returns back an error code to
OS and the kdump fails.
Link: https://lore.kernel.org/r/164375212337.440833.11955356190354940369.stgit@brunhilda.pdev.net
Reviewed-by: Kevin Barnett <kevin.barnett@microchip.com>
Reviewed-by: Scott Benesh <scott.benesh@microchip.com>
Reviewed-by: Scott Teel <scott.teel@microchip.com>
Signed-off-by: Mahesh Rajashekhara <mahesh.rajashekhara@microchip.com>
Signed-off-by: Don Brace <don.brace@microchip.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Mahesh Rajashekhara [Tue, 1 Feb 2022 21:48:38 +0000 (15:48 -0600)]
scsi: smartpqi: Update volume size after expansion
After modifying logical volume size, lsblk command still shows previous
size of logical volume.
When the driver gets any event from firmware it schedules rescan worker
with delay of 10 seconds. If array expansion is so quick that it gets
completed in a second, the driver could not catch logical volume expansion
due to worker delay.
Since driver does not detect volume expansion, driver would not call
rescan device to update new size to the OS.
Link: https://lore.kernel.org/r/164375211833.440833.17023155389220583731.stgit@brunhilda.pdev.net
Reviewed-by: Kevin Barnett <kevin.barnett@microchip.com>
Reviewed-by: Mike McGowen <mike.mcgowen@microchip.com>
Reviewed-by: Scott Benesh <scott.benesh@microchip.com>
Reviewed-by: Scott Teel <scott.teel@microchip.com>
Signed-off-by: Mahesh Rajashekhara <mahesh.rajashekhara@microchip.com>
Signed-off-by: Don Brace <don.brace@microchip.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Sagar Biradar [Tue, 1 Feb 2022 21:48:33 +0000 (15:48 -0600)]
scsi: smartpqi: Avoid drive spin-down during suspend
On certain systems (based on PCI IDs), when the OS transitions the system
into the suspend (S3) state, the BMIC flush cache command will indicate a
system RESTART instead of SUSPEND.
This avoids drive spin-down.
Link: https://lore.kernel.org/r/164375211330.440833.7203813692110347698.stgit@brunhilda.pdev.net
Reviewed-by: Kevin Barnett <kevin.barnett@microchip.com>
Reviewed-by: Scott Benesh <scott.benesh@microchip.com>
Reviewed-by: Scott Teel <scott.teel@microchip.com>
Signed-off-by: Sagar Biradar <sagar.biradar@microchip.com>
Signed-off-by: Don Brace <don.brace@microchip.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Balsundar P [Tue, 1 Feb 2022 21:48:28 +0000 (15:48 -0600)]
scsi: smartpqi: Resolve delay issue with PQI_HZ value
Change PQI_HZ to HZ.
PQI_HZ macro was set to 1000 when HZ value is less than 1000. By default,
PQI_HZ will result into a delay of 10 seconds(for kernel, which has HZ =
100). So in this case when firmware raises an event, rescan worker will be
scheduled after a delay of (10 x PQI_HZ) = 100 seconds instead of 10
seconds.
Also driver uses PQI_HZ at many instances, which might result in some other
issues with respect to delay.
Link: https://lore.kernel.org/r/164375210825.440833.15510172447583227486.stgit@brunhilda.pdev.net
Reviewed-by: Kevin Barnett <kevin.barnett@microchip.com>
Reviewed-by: Mike McGowen <mike.mcgowen@microchip.com>
Reviewed-by: Scott Benesh <scott.benesh@microchip.com>
Reviewed-by: Scott Teel <scott.teel@microchip.com>
Signed-off-by: Balsundar P <balsundar.p@microchip.com>
Signed-off-by: Don Brace <don.brace@microchip.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Kevin Barnett [Tue, 1 Feb 2022 21:48:23 +0000 (15:48 -0600)]
scsi: smartpqi: Fix a typo in func pqi_aio_submit_io()
Use correct pqi_aio_path_request structure to calculate the offset to
sg_descriptors.
The function pqi_aio_submit_io() uses the pqi_raid_path_request structure
to calculate the offset of the structure member sg_descriptors. This is
incorrect. It should be using the pqi_aio_path_request structure instead.
This typo is benign because the offsets are the same in both structures.
Link: https://lore.kernel.org/r/164375210321.440833.2566086558909686629.stgit@brunhilda.pdev.net
Reviewed-by: Mike McGowen <mike.mcgowen@microchip.com>
Reviewed-by: Scott Teel <scott.teel@microchip.com>
Signed-off-by: Kevin Barnett <kevin.barnett@microchip.com>
Signed-off-by: Don Brace <don.brace@microchip.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Kevin Barnett [Tue, 1 Feb 2022 21:48:18 +0000 (15:48 -0600)]
scsi: smartpqi: Fix a name typo and cleanup code
Rename the function pqi_is_io_high_priority() to pqi_is_io_high_priority().
Remove 2 unnecessary lines from the function, and adjusted the white space.
Link: https://lore.kernel.org/r/164375209818.440833.10908948825731635853.stgit@brunhilda.pdev.net
Reviewed-by: Mike McGowen <mike.mcgowen@microchip.com>
Reviewed-by: Scott Benesh <scott.benesh@microchip.com>
Reviewed-by: Scott Teel <scott.teel@microchip.com>
Signed-off-by: Kevin Barnett <kevin.barnett@microchip.com>
Signed-off-by: Don Brace <don.brace@microchip.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Murthy Bhat [Tue, 1 Feb 2022 21:48:13 +0000 (15:48 -0600)]
scsi: smartpqi: Quickly propagate path failures to SCSI midlayer
Return DID_NO_CONNECT when a path failure is detected.
When a path fails during I/O and AIO path gets disabled for a multipath
device, the I/O was retried in the RAID path slowing down path fail
detection. Returning DID_NO_CONNECT allows multipath to switch paths more
quickly.
Link: https://lore.kernel.org/r/164375209313.440833.9992416628621839233.stgit@brunhilda.pdev.net
Reviewed-by: Scott Benesh <scott.benesh@microchip.com>
Reviewed-by: Scott Teel <scott.teel@microchip.com>
Reviewed-by: Sagar Biradar <sagar.biradar@microchip.com>
Signed-off-by: Murthy Bhat <Murthy.Bhat@microchip.com>
Signed-off-by: Don Brace <don.brace@microchip.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Sagar Biradar [Tue, 1 Feb 2022 21:48:08 +0000 (15:48 -0600)]
scsi: smartpqi: Eliminate drive spin down on warm boot
Avoid drive spin down during a warm boot.
Call the BMIC Flush Cache command (0xc2) to indicate the reason for the
flush cache (shutdown, hibernate, suspend, or restart).
Link: https://lore.kernel.org/r/164375208810.440833.11254644025650871435.stgit@brunhilda.pdev.net
Reviewed-by: Mike McGowen <mike.mcgowen@microchip.com>
Reviewed-by: Scott Benesh <scott.benesh@microchip.com>
Reviewed-by: Scott Teel <scott.teel@microchip.com>
Signed-off-by: Sagar Biradar <sagar.biradar@microchip.com>
Signed-off-by: Don Brace <don.brace@microchip.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Gilbert Wu [Tue, 1 Feb 2022 21:48:03 +0000 (15:48 -0600)]
scsi: smartpqi: Enable SATA NCQ priority in sysfs
Add device attribute 'sas_ncq_prio_enable' to enable SATA NCQ priority
support and provide I/O priority in SCSI command and pass priority
information to controller firmware.
This device attribute works only when device has NCQ priority support and
controller firmware can handle I/Os with NCQ priority attribute.
Link: https://lore.kernel.org/r/164375208306.440833.7392577382127815362.stgit@brunhilda.pdev.net
Reviewed-by: Mike McGowen <mike.mcgowen@microchip.com>
Reviewed-by: Scott Teel <scott.teel@microchip.com>
Signed-off-by: Gilbert Wu <Gilbert.Wu@microchip.com>
Signed-off-by: Don Brace <don.brace@microchip.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Don Brace [Tue, 1 Feb 2022 21:47:58 +0000 (15:47 -0600)]
scsi: smartpqi: Add PCI IDs
Add in new ZTE controllers:
VID / DID / SVID / SDID
---- ---- ---- ----
ZTE SmartROC3100 RS241-18i 2G 9005 / 028F / 1CF2 / 5449
ZTE SmartROC3100 RS242-18i 4G 9005 / 028F / 1CF2 / 544A
ZTE SmartIOC2100 RS243-18i 9005 / 028F / 1CF2 / 544B
ZTE SmartROC3100 RM241B-18i 2G 9005 / 028F / 1CF2 / 544D
ZTE SmartROC3100 RM242B-18i 4G 9005 / 028F / 1CF2 / 544E
ZTE SmartIOC2100 RM243B-18i 9005 / 028F / 1CF2 / 544F
Add PCI ID for 1100-24i controller:
VID / DID / SVID / SDID
---- ---- ---- ----
HBA 1100-24i 9005 / 028F / 9005 / 1304
Add PCI IDs for HPE and Adaptec devices:
VID / DID / SVID / SDID
---- ---- ---- ----
Adaptec Smart HBA 2200-8io /e 9005 / 028F / 9005 / 1463
Adaptec Smart HBA 2200-16io /e 9005 / 028F / 9005 / 14C2
HPE SR308i-p Gen11 9005 / 028F / 1590 / 0382
HPE SR308i-o Gen11 9005 / 028F / 1590 / 0383
HPE SR932i-p Gen11 9005 / 028F / 1590 / 0381
Add PCI IDs for Inspur controllers:
VID / DID / SVID / SDID
---- ---- ---- ----
INSPUR RS0800M5H24i 9005 / 028F / 1BD4 / 006B
INSPUR RS0800M5E8I 9005 / 028F / 1BD4 / 006C
INSPUR RS0800M5H8I 9005 / 028F / 1BD4 / 006D
INSPUR RS0804M5R16i 9005 / 028F / 1BD4 / 006F
INSPUR RS0800M5E16i 9005 / 028F / 1BD4 / 0070
INSPUR RS0800M5H16i 9005 / 028F / 1BD4 / 0071
INSPUR RS0800M5E16i 9005 / 028F / 1BD4 / 0072
NT RAID 3100-24i 9005 / 028F / 1F0C / 3161
Add HPE and Adaptec OROC PCI IDs:
VID / DID / SVID / SDID
---- ---- ---- ----
HPE SR216i-o Gen11 9005 / 028F / 9005 / 036F
Adaptec SmartRAID 3284-16io /e/uC 9005 / 028F / 9005 / 1473
Adaptec SmartRAID 3254-16io /e 9005 / 028F / 9005 / 1474
Add PCI IDs for new channel controllers:
VID / DID / SVID / SDID
---- ---- ---- ----
Adaptec SmartRAID 3254-8i /e 9005 / 028F / 9005 / 14A4
Adaptec SmartRAID 3252-8i /e 9005 / 028F / 9005 / 14A5
Adaptec SmartRAID 3204-8i /e 9005 / 028F / 9005 / 14A6
Align PCI IDs with OOB driver. Easier to check for differences.
Link: https://lore.kernel.org/r/164375207802.440833.15947108153078495425.stgit@brunhilda.pdev.net
Reviewed-by: Scott Benesh <scott.benesh@microchip.com>
Reviewed-by: Scott Teel <scott.teel@microchip.com>
Co-developed-by: Mike McGowen <Mike.McGowen@microchip.com>
Signed-off-by: Mike McGowen <Mike.McGowen@microchip.com>
Co-developed-by: Murthy Bhat <Murthy.Bhat@microchip.com>
Signed-off-by: Murthy Bhat <Murthy.Bhat@microchip.com>
Co-developed-by: Sagar Biradar <sagar.biradar@microchip.com>
Signed-off-by: Sagar Biradar <sagar.biradar@microchip.com>
Signed-off-by: Don Brace <don.brace@microchip.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Don Brace [Tue, 1 Feb 2022 21:47:53 +0000 (15:47 -0600)]
scsi: smartpqi: Fix rmmod stack trace
Prevent "BUG: scheduling while atomic: rmmod" stack trace.
Stop setting spin_locks before calling OS functions to remove devices.
Link: https://lore.kernel.org/r/164375207296.440833.4996145011193819683.stgit@brunhilda.pdev.net
Reviewed-by: Scott Benesh <scott.benesh@microchip.com>
Reviewed-by: Scott Teel <scott.teel@microchip.com>
Reviewed-by: Kevin Barnett <kevin.barnett@microchip.com>
Signed-off-by: Don Brace <don.brace@microchip.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Kees Cook [Tue, 1 Feb 2022 22:39:48 +0000 (14:39 -0800)]
scsi: mpt3sas: Convert to flexible arrays
This converts to a flexible array instead of the old-style 1-element
arrays. The existing code already did the correct math for finding the size
of the resulting flexible array structure, so there is no binary
difference.
The other two structures converted to use flexible arrays appear to have no
users at all.
Link: https://lore.kernel.org/r/20220201223948.1455637-1-keescook@chromium.org
Signed-off-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Sebastian Andrzej Siewior [Tue, 1 Feb 2022 21:09:54 +0000 (22:09 +0100)]
scsi: usb: storage: Complete the SCSI request directly
The USB storage driver can complete its requests directly from a kernel
thread. Use scsi_done_direct() to avoid waking ksoftirqd.
Link: https://lore.kernel.org/r/20220201210954.570896-3-sebastian@breakpoint.cc
Signed-off-by: Sebastian Andrzej Siewior <sebastian@breakpoint.cc>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Sebastian Andrzej Siewior [Thu, 3 Feb 2022 20:29:25 +0000 (21:29 +0100)]
scsi: core: Add scsi_done_direct() for immediate completion
Add scsi_done_direct() which behaves like scsi_done() except that it
invokes blk_mq_complete_request_direct() in order to complete the request.
Callers from process context can complete the request directly instead
waking ksoftirqd.
Link: https://lore.kernel.org/r/Yfw7JaszshmfYa1d@flow
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: Sebastian Andrzej Siewior <sebastian@breakpoint.cc>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Martin Wilck [Thu, 27 Jan 2022 14:13:51 +0000 (15:13 +0100)]
scsi: core: Make "access_state" sysfs attribute always visible
If a SCSI device handler module is loaded after some SCSI devices have
already been probed (e.g. via request_module() by dm-multipath), the
"access_state" and "preferred_path" sysfs attributes remain invisible for
these devices, although the handler is attached and live. The reason is
that the visibility is only checked when the sysfs attribute group is first
created. This results in an inconsistent user experience depending on the
load order of SCSI low-level drivers vs. device handler modules.
This patch changes user space API: attempting to read the "access_state" or
"preferred_path" attributes will now result in -EINVAL rather than -ENODEV
for devices that have no device handler, and tests for the existence of
these attributes will have a different result.
Link: https://lore.kernel.org/r/20220127141351.30706-1-mwilck@suse.com
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Martin Wilck <mwilck@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Minghao Chi (CGEL ZTE) [Thu, 27 Jan 2022 01:43:30 +0000 (01:43 +0000)]
scsi: lpfc: Remove redundant flush_workqueue() call
destroy_workqueue() already drains the queue before destroying it, so there
is no need to flush it explicitly.
Remove the redundant flush_workqueue() call.
Link: https://lore.kernel.org/r/20220127014330.1185114-1-chi.minghao@zte.com.cn
Reported-by: Zeal Robot <zealci@zte.com.cn>
Signed-off-by: Minghao Chi (CGEL ZTE) <chi.minghao@zte.com.cn>
Signed-off-by: CGEL ZTE <cgel.zte@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Minghao Chi (CGEL ZTE) [Thu, 27 Jan 2022 01:39:34 +0000 (01:39 +0000)]
scsi: qedi: Remove redundant flush_workqueue() calls
destroy_workqueue() already drains the queue before destroying it, so there
is no need to flush it explicitly.
Remove the redundant flush_workqueue() calls.
Link: https://lore.kernel.org/r/20220127013934.1184923-1-chi.minghao@zte.com.cn
Reported-by: Zeal Robot <zealci@zte.com.cn>
Signed-off-by: Minghao Chi (CGEL ZTE) <chi.minghao@zte.com.cn>
Signed-off-by: CGEL ZTE <cgel.zte@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Yang Guang [Thu, 27 Jan 2022 00:03:46 +0000 (08:03 +0800)]
scsi: bfa: Replace snprintf() with sysfs_emit()
coccinelle report:
./drivers/scsi/bfa/bfad_attr.c:908:8-16:
WARNING: use scnprintf or sprintf
./drivers/scsi/bfa/bfad_attr.c:860:8-16:
WARNING: use scnprintf or sprintf
./drivers/scsi/bfa/bfad_attr.c:888:8-16:
WARNING: use scnprintf or sprintf
./drivers/scsi/bfa/bfad_attr.c:853:8-16:
WARNING: use scnprintf or sprintf
./drivers/scsi/bfa/bfad_attr.c:808:8-16:
WARNING: use scnprintf or sprintf
./drivers/scsi/bfa/bfad_attr.c:728:8-16:
WARNING: use scnprintf or sprintf
./drivers/scsi/bfa/bfad_attr.c:822:8-16:
WARNING: use scnprintf or sprintf
./drivers/scsi/bfa/bfad_attr.c:927:9-17:
WARNING: use scnprintf or sprintf
./drivers/scsi/bfa/bfad_attr.c:900:8-16:
WARNING: use scnprintf or sprintf
./drivers/scsi/bfa/bfad_attr.c:874:8-16:
WARNING: use scnprintf or sprintf
./drivers/scsi/bfa/bfad_attr.c:714:8-16:
WARNING: use scnprintf or sprintf
./drivers/scsi/bfa/bfad_attr.c:839:8-16:
WARNING: use scnprintf or sprintf
Use sysfs_emit() instead of scnprintf() or sprintf().
Link: https://lore.kernel.org/r/def83ff75faec64ba592b867a8499b1367bae303.1643181468.git.yang.guang5@zte.com.cn
Reported-by: Zeal Robot <zealci@zte.com.cn>
Signed-off-by: Yang Guang <yang.guang5@zte.com.cn>
Signed-off-by: David Yang <davidcomponentone@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Yang Guang [Thu, 27 Jan 2022 00:00:59 +0000 (08:00 +0800)]
scsi: mvsas: Replace snprintf() with sysfs_emit()
coccinelle report:
./drivers/scsi/mvsas/mv_init.c:699:8-16:
WARNING: use scnprintf or sprintf
./drivers/scsi/mvsas/mv_init.c:747:8-16:
WARNING: use scnprintf or sprintf
Use sysfs_emit() instead of scnprintf() or sprintf().
Link: https://lore.kernel.org/r/c1711f7cf251730a8ceb5bdfc313bf85662b3395.1643182948.git.yang.guang5@zte.com.cn
Reported-by: Zeal Robot <zealci@zte.com.cn>
Signed-off-by: Yang Guang <yang.guang5@zte.com.cn>
Signed-off-by: David Yang <davidcomponentone@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Yin Xiujiang [Wed, 26 Jan 2022 01:42:48 +0000 (09:42 +0800)]
scsi: bnx2fc: Make use of the helper macro kthread_run()
Repalce kthread_create/wake_up_process() with kthread_run() to simplify the
code.
Link: https://lore.kernel.org/r/20220126014248.466806-1-yinxiujiang@kylinos.cn
Signed-off-by: Yin Xiujiang <yinxiujiang@kylinos.cn>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
John Garry [Mon, 31 Jan 2022 11:13:27 +0000 (19:13 +0800)]
scsi: hisi_sas: Fix setting of hisi_sas_slot.is_internal
The hisi_sas_slot.is_internal member is not set properly for ATA commands
which the driver sends directly. A TMF struct pointer is normally used as a
test to set this, but it is NULL for those commands. It's not ideal, but
pass an empty TMF struct to set that member properly.
Link: https://lore.kernel.org/r/1643627607-138785-1-git-send-email-john.garry@huawei.com
Fixes:
dc313f6b125b ("scsi: hisi_sas: Factor out task prep and delivery code")
Reported-by: Xiang Chen <chenxiang66@hisilicon.com>
Signed-off-by: John Garry <john.garry@huawei.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Cai Huoqing [Fri, 28 Jan 2022 06:31:01 +0000 (14:31 +0800)]
scsi: bnx2fc: Fix typo in comments
Replace 'Offlaod' with 'Offload'.
Link: https://lore.kernel.org/r/20220128063101.6953-1-cai.huoqing@linux.dev
Acked-by: Saurav Kashyap <skashyap@marvell.com>
Signed-off-by: Cai Huoqing <cai.huoqing@linux.dev>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Jinyoung Choi [Thu, 27 Jan 2022 03:00:25 +0000 (12:00 +0900)]
scsi: ufs: Add checking lifetime attribute for WriteBooster
Because WB performs writes in SLC mode, it is not possible to use
WriteBooster indefinitely. Vendors can set a lifetime limit in the device.
If the lifetime exceeds this limit, the device ican disable the WB feature.
The feature is defined in the "bWriteBoosterBufferLifeTimeEst (IDN = 1E)"
attribute.
With lifetime exceeding the limit value, the current driver continuously
performs the following query:
- Write Flag: WB_ENABLE / DISABLE
- Read attr: Available Buffer Size
- Read attr: Current Buffer Size
This patch recognizes that WriteBooster is no longer supported by the
device, and prevents unnecessary queries.
Link: https://lore.kernel.org/r/1891546521.01643252701746.JavaMail.epsvc@epcpadp3
Reviewed-by: Asutosh Das <quic_asutoshd@quicinc.com>
Acked-by: Avri Altman <avri.altman@wdc.com>
Signed-off-by: Jinyoung Choi <j-young.choi@samsung.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
John Garry [Thu, 27 Jan 2022 13:12:52 +0000 (21:12 +0800)]
scsi: pm8001: Fix use-after-free for aborted SSP/STP sas_task
Currently a use-after-free may occur if a sas_task is aborted by the upper
layer before we handle the I/O completion in mpi_ssp_completion() or
mpi_sata_completion().
In this case, the following are the two steps in handling those I/O
completions:
- Call complete() to inform the upper layer handler of completion of
the I/O.
- Release driver resources associated with the sas_task in
pm8001_ccb_task_free() call.
When complete() is called, the upper layer may free the sas_task. As such,
we should not touch the associated sas_task afterwards, but we do so in the
pm8001_ccb_task_free() call.
Fix by swapping the complete() and pm8001_ccb_task_free() calls ordering.
Link: https://lore.kernel.org/r/1643289172-165636-4-git-send-email-john.garry@huawei.com
Reviewed-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Acked-by: Jack Wang <jinpu.wang@ionos.com>
Signed-off-by: John Garry <john.garry@huawei.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
John Garry [Thu, 27 Jan 2022 13:12:51 +0000 (21:12 +0800)]
scsi: pm8001: Fix use-after-free for aborted TMF sas_task
Currently a use-after-free may occur if a TMF sas_task is aborted before we
handle the IO completion in mpi_ssp_completion(). The abort occurs due to
timeout.
When the timeout occurs, the SAS_TASK_STATE_ABORTED flag is set and the
sas_task is freed in pm8001_exec_internal_tmf_task().
However, if the I/O completion occurs later, the I/O completion still
thinks that the sas_task is available. Fix this by clearing the ccb->task
if the TMF times out - the I/O completion handler does nothing if this
pointer is cleared.
Link: https://lore.kernel.org/r/1643289172-165636-3-git-send-email-john.garry@huawei.com
Reviewed-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Acked-by: Jack Wang <jinpu.wang@ionos.com>
Signed-off-by: John Garry <john.garry@huawei.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
John Garry [Thu, 27 Jan 2022 13:12:50 +0000 (21:12 +0800)]
scsi: pm8001: Fix warning for undescribed param in process_one_iomb()
make W=1 complains of an undescribed function parameter:
drivers/scsi/pm8001/pm80xx_hwi.c:3938: warning: Function parameter or member 'circularQ' not described in 'process_one_iomb'
Fix it.
Link: https://lore.kernel.org/r/1643289172-165636-2-git-send-email-john.garry@huawei.com
Reported-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Reviewed-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Acked-by: Jack Wang <jinpu.wang@ionos.com>
Signed-off-by: John Garry <john.garry@huawei.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Ming Lei [Thu, 27 Jan 2022 15:37:33 +0000 (23:37 +0800)]
scsi: core: Reallocate device's budget map on queue depth change
We currently use ->cmd_per_lun as initial queue depth for setting up the
budget_map. Martin Wilck reported that it is common for the queue_depth to
be subsequently updated in slave_configure() based on detected hardware
characteristics.
As a result, for some drivers, the static host template settings for
cmd_per_lun and can_queue won't actually get used in practice. And if the
default values are used to allocate the budget_map, memory may be consumed
unnecessarily.
Fix the issue by reallocating the budget_map after ->slave_configure()
returns. At that time the device queue_depth should accurately reflect what
the hardware needs.
Link: https://lore.kernel.org/r/20220127153733.409132-1-ming.lei@redhat.com
Cc: Bart Van Assche <bvanassche@acm.org>
Reported-by: Martin Wilck <martin.wilck@suse.com>
Suggested-by: Martin Wilck <martin.wilck@suse.com>
Tested-by: Martin Wilck <mwilck@suse.com>
Reviewed-by: Martin Wilck <mwilck@suse.com>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
John Meneghini [Mon, 24 Jan 2022 14:51:10 +0000 (09:51 -0500)]
scsi: bnx2fc: Make bnx2fc_recv_frame() mp safe
Running tests with a debug kernel shows that bnx2fc_recv_frame() is
modifying the per_cpu lport stats counters in a non-mpsafe way. Just boot
a debug kernel and run the bnx2fc driver with the hardware enabled.
[ 1391.699147] BUG: using smp_processor_id() in preemptible [
00000000] code: bnx2fc_
[ 1391.699160] caller is bnx2fc_recv_frame+0xbf9/0x1760 [bnx2fc]
[ 1391.699174] CPU: 2 PID: 4355 Comm: bnx2fc_l2_threa Kdump: loaded Tainted: G B
[ 1391.699180] Hardware name: HP ProLiant DL120 G7, BIOS J01 07/01/2013
[ 1391.699183] Call Trace:
[ 1391.699188] dump_stack_lvl+0x57/0x7d
[ 1391.699198] check_preemption_disabled+0xc8/0xd0
[ 1391.699205] bnx2fc_recv_frame+0xbf9/0x1760 [bnx2fc]
[ 1391.699215] ? do_raw_spin_trylock+0xb5/0x180
[ 1391.699221] ? bnx2fc_npiv_create_vports.isra.0+0x4e0/0x4e0 [bnx2fc]
[ 1391.699229] ? bnx2fc_l2_rcv_thread+0xb7/0x3a0 [bnx2fc]
[ 1391.699240] bnx2fc_l2_rcv_thread+0x1af/0x3a0 [bnx2fc]
[ 1391.699250] ? bnx2fc_ulp_init+0xc0/0xc0 [bnx2fc]
[ 1391.699258] kthread+0x364/0x420
[ 1391.699263] ? _raw_spin_unlock_irq+0x24/0x50
[ 1391.699268] ? set_kthread_struct+0x100/0x100
[ 1391.699273] ret_from_fork+0x22/0x30
Restore the old get_cpu/put_cpu code with some modifications to reduce the
size of the critical section.
Link: https://lore.kernel.org/r/20220124145110.442335-1-jmeneghi@redhat.com
Fixes:
d576a5e80cd0 ("bnx2fc: Improve stats update mechanism")
Tested-by: Guangwu Zhang <guazhang@redhat.com>
Acked-by: Saurav Kashyap <skashyap@marvell.com>
Signed-off-by: John Meneghini <jmeneghi@redhat.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Ajish Koshy [Mon, 24 Jan 2022 08:22:55 +0000 (13:52 +0530)]
scsi: pm80xx: Fix double completion for SATA devices
Current code handles completions for SATA devices in mpi_sata_completion()
and mpi_sata_event().
However, at the time when any SATA event happens, for almost all the event
types, the command is still in the target. It is therefore incorrect to
complete the task in sata_event().
There are some events for which we get sata_completions, some need recovery
procedure and others abort. All the tasks must be completed via
sata_completion() path.
Removed the task done related code from sata_events(). For tasks where we
don't get completions, let top layer call abort() to abort the command post
timeout.
Link: https://lore.kernel.org/r/20220124082255.86223-1-Ajish.Koshy@microchip.com
Acked-by: Jack Wang <jinpu.wang@ionos.com>
Co-developed-by: Viswas G <Viswas.G@microchip.com>
Signed-off-by: Viswas G <Viswas.G@microchip.com>
Signed-off-by: Ajish Koshy <Ajish.Koshy@microchip.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Douglas Gilbert [Sun, 9 Jan 2022 01:28:53 +0000 (20:28 -0500)]
scsi: scsi_debug: Add environmental reporting log subpage
Log subpages are starting to appear in real devices (e.g. SSDs) so add
support for one. Adopt approach where all "wild" sub-pages are themselves
listed as long as there is at least one non-wild page or subpage for a
given page number.
Link: https://lore.kernel.org/r/20220109012853.301953-10-dgilbert@interlog.com
Signed-off-by: Douglas Gilbert <dgilbert@interlog.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Douglas Gilbert [Sun, 9 Jan 2022 01:28:50 +0000 (20:28 -0500)]
scsi: scsi_debug: Add no_rwlock parameter
By default, this driver places a read lock around all user data fetches and
a write lock around all user data modifying operations (e.g. WRITE
commands). These locks have "per store" granularity. Other drivers that
have a similar function (e.g. null_blk) do not take this data integrity
step and run significantly faster in some tests.
In the common case of a (simulated) device to device copy (e.g. what dd
and its variants do) there should be no need for locks around data
accesses. So add the driver and sysfs parameter no_rwlock which is boolean
and when set does what its name suggests. The default is false for backward
comaptibility.
Link: https://lore.kernel.org/r/20220109012853.301953-7-dgilbert@interlog.com
Signed-off-by: Douglas Gilbert <dgilbert@interlog.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Douglas Gilbert [Sun, 9 Jan 2022 01:28:49 +0000 (20:28 -0500)]
scsi: scsi_debug: Divide power on reset UNIT ATTENTION
To distinguish between resets sent by the SCSI mid-level error
handling and newly introduced devices (LUs), this Unit Attention:
power on, reset, or bus reset occurred [0x29,0x0]
has been subdivided into that UA for the reset case and this new UA:
power on occurred [0x29,0x1]
for the new device (LU) case. This makes debug a little easier to follow
when it is turned on (e.g. 'echo 0x1 > opts').
Bump driver version number.
Link: https://lore.kernel.org/r/20220109012853.301953-6-dgilbert@interlog.com
Signed-off-by: Douglas Gilbert <dgilbert@interlog.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Douglas Gilbert [Sun, 9 Jan 2022 01:28:48 +0000 (20:28 -0500)]
scsi: scsi_debug: Refine sdebug_blk_mq_poll()
Refine the sdebug_blk_mq_poll() function so it only takes the spinlock on
the queue when it can see one or more requests with the in_use bitmap flag
set.
Link: https://lore.kernel.org/r/20220109012853.301953-5-dgilbert@interlog.com
Signed-off-by: Douglas Gilbert <dgilbert@interlog.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Douglas Gilbert [Sun, 9 Jan 2022 01:28:47 +0000 (20:28 -0500)]
scsi: scsi_debug: Use TASK SET FULL more
When the internal in_use bit array in this driver is full returning
SCSI_MLQUEUE_HOST_BUSY leads to the mid-level reissuing the request which
is unhelpful. Previously TASK SET FULL status was only returned if ALL_TSF
[0x400] is placed in the opts variable (at load time or via sysfs). Now
ignore that setting and always return TASK SET FULL when in_use array is
full. Also set DID_ABORT together with TASK SET FULL so the mid-level gives
up immediately.
Aside: the situations addressed by this patch lead to lockups and
timeouts. They have only been detected when blk_poll() is used. That
mechanism is relatively new in the SCSI subsystem suggesting the mid-level
may need more work in that area.
Link: https://lore.kernel.org/r/20220109012853.301953-4-dgilbert@interlog.com
Signed-off-by: Douglas Gilbert <dgilbert@interlog.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Douglas Gilbert [Sun, 9 Jan 2022 01:28:46 +0000 (20:28 -0500)]
scsi: scsi_debug: Strengthen defer_t accesses
Use READ_ONCE() and WRITE_ONCE() macros when accessing the
sdebug_defer::defer_t value.
Link: https://lore.kernel.org/r/20220109012853.301953-3-dgilbert@interlog.com
Signed-off-by: Douglas Gilbert <dgilbert@interlog.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Douglas Gilbert [Sun, 9 Jan 2022 01:28:45 +0000 (20:28 -0500)]
scsi: scsi_debug: Address races following module load
When scsi_debug is loaded as a module with many (simulated) hosts, targets,
and devices (LUs), modprobe can take a long time to return. Only a small
amount of this time is spent in the scsi_debug_init(); the rest is other
parts of the kernel reacting to to the appearance of new storage
devices. As soon as scsi_debug_init() has completed the user space may call
'rmmod scsi_debug' and this was found to cause race problems as outlined
here:
https://bugzilla.kernel.org/show_bug.cgi?id=212337
To reliably generate this race a sysfs parameter called rm_all_hosts was
added and the code was strengthened in this area. The main change was to
make the count of scsi_debug hosts present an atomic. Then it was found
that the handling of the existing add_host parameter needed the same
strengthening. Further: 'echo -9999 >
/sys/bus/pseudo/drivers/scsi_debug/add_host has the same effect as
rm_all_hosts so rm_all_hosts was not needed.
To inhibit a race between two invocations of writes to add_host, a mutex
was added. Also address a possible race when rmmod is called but LUs are
still being added.
The logic to remove (all) hosts is rather crude: it works backwards down a
linked lists of hosts. Any pending requests are terminated with
DID_NO_CONNECT as are any new requests. In the case where not all hosts are
being removed, the ones that remain may have lost requests as just
outlined. The lowest numbered host (id) hosts will remain.
Cc: Bart Van Assche <bvanassche@acm.org>
Link: https://lore.kernel.org/r/20220109012853.301953-2-dgilbert@interlog.com
Signed-off-by: Douglas Gilbert <dgilbert@interlog.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Tong Zhang [Sun, 23 Jan 2022 22:57:17 +0000 (14:57 -0800)]
scsi: myrs: Fix crash in error case
In myrs_detect(), cs->disable_intr is NULL when privdata->hw_init() fails
with non-zero. In this case, myrs_cleanup(cs) will call a NULL ptr and
crash the kernel.
[ 1.105606] myrs 0000:00:03.0: Unknown Initialization Error 5A
[ 1.105872] myrs 0000:00:03.0: Failed to initialize Controller
[ 1.106082] BUG: kernel NULL pointer dereference, address:
0000000000000000
[ 1.110774] Call Trace:
[ 1.110950] myrs_cleanup+0xe4/0x150 [myrs]
[ 1.111135] myrs_probe.cold+0x91/0x56a [myrs]
[ 1.111302] ? DAC960_GEM_intr_handler+0x1f0/0x1f0 [myrs]
[ 1.111500] local_pci_probe+0x48/0x90
Link: https://lore.kernel.org/r/20220123225717.1069538-1-ztong0001@gmail.com
Reviewed-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Tong Zhang <ztong0001@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Colin Ian King [Sun, 23 Jan 2022 17:55:30 +0000 (17:55 +0000)]
scsi: 53c700: Remove redundant assignment to pointer SCp
Pointer SCp is being re-assigned the same value that it was initialized to
a few lines earlier, the assignment is redundant and can be removed.
Link: https://lore.kernel.org/r/20220123175530.110462-1-colin.i.king@gmail.com
Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Kiwoong Kim [Fri, 21 Jan 2022 05:37:55 +0000 (14:37 +0900)]
scsi: ufs: Treat link loss as fatal error
This event is raised when link is lost as specified in UFSHCI spec and that
means communication is not possible. Thus initializing UFS interface needs
to be done.
Make UFS driver considers Link Lost as fatal in the INT_FATAL_ERRORS
mask. This will trigger a host reset whenever a link lost interrupt occurs.
Link: https://lore.kernel.org/r/1642743475-54275-1-git-send-email-kwmad.kim@samsung.com
Signed-off-by: Kiwoong Kim <kwmad.kim@samsung.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Kiwoong Kim [Fri, 21 Jan 2022 05:33:02 +0000 (14:33 +0900)]
scsi: ufs: Use generic error code in ufshcd_set_dev_pwr_mode()
The return value of ufshcd_set_dev_pwr_mode() is passed to device PM
core. However, the function currently returns a SCSI result which the PM
core doesn't understand. This might lead to unexpected behaviors in
userland; a platform reset was observed in Android.
Use a generic error code for SSU failures.
Link: https://lore.kernel.org/r/1642743182-54098-1-git-send-email-kwmad.kim@samsung.com
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: Kiwoong Kim <kwmad.kim@samsung.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Nilesh Javali [Mon, 10 Jan 2022 05:02:18 +0000 (21:02 -0800)]
scsi: qla2xxx: Update version to 10.02.07.300-k
Link: https://lore.kernel.org/r/20220110050218.3958-18-njavali@marvell.com
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Joe Carnuccio [Mon, 10 Jan 2022 05:02:17 +0000 (21:02 -0800)]
scsi: qla2xxx: Check for firmware dump already collected
While allocating firmware dump, check if dump is already collected and do
not re-allocate the buffer.
Link: https://lore.kernel.org/r/20220110050218.3958-17-njavali@marvell.com
Cc: stable@vger.kernel.org
Signed-off-by: Joe Carnuccio <joe.carnuccio@cavium.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Joe Carnuccio [Mon, 10 Jan 2022 05:02:16 +0000 (21:02 -0800)]
scsi: qla2xxx: Add devids and conditionals for 28xx
This is an update to the original 28xx adapter enablement. Add a bunch of
conditionals that are applicable for 28xx.
Link: https://lore.kernel.org/r/20220110050218.3958-16-njavali@marvell.com
Fixes:
ecc89f25e225 ("scsi: qla2xxx: Add Device ID for ISP28XX")
Cc: stable@vger.kernel.org
Signed-off-by: Joe Carnuccio <joe.carnuccio@cavium.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Saurav Kashyap [Mon, 10 Jan 2022 05:02:15 +0000 (21:02 -0800)]
scsi: qla2xxx: Suppress a kernel complaint in qla_create_qpair()
[ 12.323788] BUG: using smp_processor_id() in preemptible [
00000000] code: systemd-udevd/1020
[ 12.332297] caller is qla2xxx_create_qpair+0x32a/0x5d0 [qla2xxx]
[ 12.338417] CPU: 7 PID: 1020 Comm: systemd-udevd Tainted: G I --------- --- 5.14.0-29.el9.x86_64 #1
[ 12.348827] Hardware name: Dell Inc. PowerEdge R610/0F0XJ6, BIOS 6.6.0 05/22/2018
[ 12.356356] Call Trace:
[ 12.358821] dump_stack_lvl+0x34/0x44
[ 12.362514] check_preemption_disabled+0xd9/0xe0
[ 12.367164] qla2xxx_create_qpair+0x32a/0x5d0 [qla2xxx]
[ 12.372481] qla2x00_probe_one+0xa3a/0x1b80 [qla2xxx]
[ 12.377617] ? _raw_spin_lock_irqsave+0x19/0x40
[ 12.384284] local_pci_probe+0x42/0x80
[ 12.390162] ? pci_match_device+0xd7/0x110
[ 12.396366] pci_device_probe+0xfd/0x1b0
[ 12.402372] really_probe+0x1e7/0x3e0
[ 12.408114] __driver_probe_device+0xfe/0x180
[ 12.414544] driver_probe_device+0x1e/0x90
[ 12.420685] __driver_attach+0xc0/0x1c0
[ 12.426536] ? __device_attach_driver+0xe0/0xe0
[ 12.433061] ? __device_attach_driver+0xe0/0xe0
[ 12.439538] bus_for_each_dev+0x78/0xc0
[ 12.445294] bus_add_driver+0x12b/0x1e0
[ 12.451021] driver_register+0x8f/0xe0
[ 12.456631] ? 0xffffffffc07bc000
[ 12.461773] qla2x00_module_init+0x1be/0x229 [qla2xxx]
[ 12.468776] do_one_initcall+0x44/0x200
[ 12.474401] ? load_module+0xad3/0xba0
[ 12.479908] ? kmem_cache_alloc_trace+0x45/0x410
[ 12.486268] do_init_module+0x5c/0x280
[ 12.491730] __do_sys_init_module+0x12e/0x1b0
[ 12.497785] do_syscall_64+0x3b/0x90
[ 12.503029] entry_SYSCALL_64_after_hwframe+0x44/0xae
[ 12.509764] RIP: 0033:0x7f554f73ab2e
Link: https://lore.kernel.org/r/20220110050218.3958-15-njavali@marvell.com
Cc: stable@vger.kernel.org
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: Saurav Kashyap <skashyap@marvell.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Joe Carnuccio [Mon, 10 Jan 2022 05:02:14 +0000 (21:02 -0800)]
scsi: qla2xxx: Fix T10 PI tag escape and IP guard options for 28XX adapters
28XX adapters are capable of detecting both T10 PI tag escape values as
well as IP guard. This was missed due to the adapter type missed in the
corresponding macros. Fix this by adding support for 28xx in those macros.
Link: https://lore.kernel.org/r/20220110050218.3958-14-njavali@marvell.com
Cc: stable@vger.kernel.org
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
Tested-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Joe Carnuccio <joe.carnuccio@cavium.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Quinn Tran [Mon, 10 Jan 2022 05:02:13 +0000 (21:02 -0800)]
scsi: qla2xxx: edif: Fix clang warning
Silence compile warning due to unaligned memory access.
qla_edif.c:713:45: warning: taking address of packed member 'u' of class or
structure 'auth_complete_cmd' may result in an unaligned pointer value
[-Waddress-of-packed-member]
fcport = qla2x00_find_fcport_by_pid(vha, &appplogiok.u.d_id);
Link: https://lore.kernel.org/r/20220110050218.3958-13-njavali@marvell.com
Cc: stable@vger.kernel.org
Reported-by: kernel test robot <lkp@intel.com>
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: Quinn Tran <qutran@marvell.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Nilesh Javali [Mon, 10 Jan 2022 05:02:12 +0000 (21:02 -0800)]
scsi: qla2xxx: Fix warning for missing error code
Fix smatch-reported warning message:
drivers/scsi/qla2xxx/qla_target.c:3324 qlt_xmit_response() warn: missing error
code 'res'
Link: https://lore.kernel.org/r/20220110050218.3958-12-njavali@marvell.com
Fixes:
4a8f71014b4d ("scsi: qla2xxx: Fix unmap of already freed sgl")
Cc: stable@vger.kernel.org
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Arun Easi [Mon, 10 Jan 2022 05:02:11 +0000 (21:02 -0800)]
scsi: qla2xxx: Fix device reconnect in loop topology
A device logout in loop topology initiates a device connection teardown
which loses the FW device handle. In loop topo, the device handle is not
regrabbed leading to device login failures and eventually to loss of the
device. Fix this by taking the main login path that does it.
Link: https://lore.kernel.org/r/20220110050218.3958-11-njavali@marvell.com
Cc: stable@vger.kernel.org
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: Arun Easi <aeasi@marvell.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Shreyas Deodhar [Mon, 10 Jan 2022 05:02:10 +0000 (21:02 -0800)]
scsi: qla2xxx: Add ql2xnvme_queues module param to configure number of NVMe queues
Add ql2xnvme_queues module parameter to configure number of NVMe queues
Usage:
Number of NVMe Queues that can be configured.
Final value will be min(ql2xnvme_queues, num_cpus, num_chip_queues),
1 - Minimum number of queues supported
8 - Default value
128 - Maximum number of queues supported
Link: https://lore.kernel.org/r/20220110050218.3958-10-njavali@marvell.com
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: Shreyas Deodhar <sdeodhar@marvell.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Bikash Hazarika [Mon, 10 Jan 2022 05:02:09 +0000 (21:02 -0800)]
scsi: qla2xxx: Fix wrong FDMI data for 64G adapter
Corrected transmission speed mask values for FC.
Supported Speed: 16 32 20 Gb/s ===> Should be 64 instead of 20
Supported Speed: 16G 32G 48G ===> Should be 64G instead of 48G
Link: https://lore.kernel.org/r/20220110050218.3958-9-njavali@marvell.com
Cc: stable@vger.kernel.org
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: Bikash Hazarika <bhazarika@marvell.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Quinn Tran [Mon, 10 Jan 2022 05:02:08 +0000 (21:02 -0800)]
scsi: qla2xxx: Add retry for exec firmware
Per FW request, Exec FW can fail due to temporary error resulting in driver
not attaching to the adapter. Add retry of this command up to 4 retries.
Link: https://lore.kernel.org/r/20220110050218.3958-8-njavali@marvell.com
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: Quinn Tran <qutran@marvell.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Quinn Tran [Mon, 10 Jan 2022 05:02:07 +0000 (21:02 -0800)]
scsi: qla2xxx: Fix scheduling while atomic
The driver makes a call into midlayer (fc_remote_port_delete) which can put
the thread to sleep. The thread that originates the call is in interrupt
context. The combination of the two trigger a crash. Schedule the call in
non-interrupt context where it is more safe.
kernel: BUG: scheduling while atomic: swapper/7/0/0x00010000
kernel: Call Trace:
kernel: <IRQ>
kernel: dump_stack+0x66/0x81
kernel: __schedule_bug.cold.90+0x5/0x1d
kernel: __schedule+0x7af/0x960
kernel: schedule+0x28/0x80
kernel: schedule_timeout+0x26d/0x3b0
kernel: wait_for_completion+0xb4/0x140
kernel: ? wake_up_q+0x70/0x70
kernel: __wait_rcu_gp+0x12c/0x160
kernel: ? sdev_evt_alloc+0xc0/0x180 [scsi_mod]
kernel: synchronize_sched+0x6c/0x80
kernel: ? call_rcu_bh+0x20/0x20
kernel: ? __bpf_trace_rcu_invoke_callback+0x10/0x10
kernel: sdev_evt_alloc+0xfd/0x180 [scsi_mod]
kernel: starget_for_each_device+0x85/0xb0 [scsi_mod]
kernel: ? scsi_init_io+0x360/0x3d0 [scsi_mod]
kernel: scsi_init_io+0x388/0x3d0 [scsi_mod]
kernel: device_for_each_child+0x54/0x90
kernel: fc_remote_port_delete+0x70/0xe0 [scsi_transport_fc]
kernel: qla2x00_schedule_rport_del+0x62/0xf0 [qla2xxx]
kernel: qla2x00_mark_device_lost+0x9c/0xd0 [qla2xxx]
kernel: qla24xx_handle_plogi_done_event+0x55f/0x570 [qla2xxx]
kernel: qla2x00_async_login_sp_done+0xd2/0x100 [qla2xxx]
kernel: qla24xx_logio_entry+0x13a/0x3c0 [qla2xxx]
kernel: qla24xx_process_response_queue+0x306/0x400 [qla2xxx]
kernel: qla24xx_msix_rsp_q+0x3f/0xb0 [qla2xxx]
kernel: __handle_irq_event_percpu+0x40/0x180
kernel: handle_irq_event_percpu+0x30/0x80
kernel: handle_irq_event+0x36/0x60
Link: https://lore.kernel.org/r/20220110050218.3958-7-njavali@marvell.com
Cc: stable@vger.kernel.org
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: Quinn Tran <qutran@marvell.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Quinn Tran [Mon, 10 Jan 2022 05:02:06 +0000 (21:02 -0800)]
scsi: qla2xxx: Fix premature hw access after PCI error
After a recoverable PCI error has been detected and recovered, qla driver
needs to check to see if the error condition still persist and/or wait
for the OS to give the resume signal.
Sep 8 22:26:03 localhost kernel: WARNING: CPU: 9 PID: 124606 at qla_tmpl.c:440
qla27xx_fwdt_entry_t266+0x55/0x60 [qla2xxx]
Sep 8 22:26:03 localhost kernel: RIP: 0010:qla27xx_fwdt_entry_t266+0x55/0x60
[qla2xxx]
Sep 8 22:26:03 localhost kernel: Call Trace:
Sep 8 22:26:03 localhost kernel: ? qla27xx_walk_template+0xb1/0x1b0 [qla2xxx]
Sep 8 22:26:03 localhost kernel: ? qla27xx_execute_fwdt_template+0x12a/0x160
[qla2xxx]
Sep 8 22:26:03 localhost kernel: ? qla27xx_fwdump+0xa0/0x1c0 [qla2xxx]
Sep 8 22:26:03 localhost kernel: ? qla2xxx_pci_mmio_enabled+0xfb/0x120
[qla2xxx]
Sep 8 22:26:03 localhost kernel: ? report_mmio_enabled+0x44/0x80
Sep 8 22:26:03 localhost kernel: ? report_slot_reset+0x80/0x80
Sep 8 22:26:03 localhost kernel: ? pci_walk_bus+0x70/0x90
Sep 8 22:26:03 localhost kernel: ? aer_dev_correctable_show+0xc0/0xc0
Sep 8 22:26:03 localhost kernel: ? pcie_do_recovery+0x1bb/0x240
Sep 8 22:26:03 localhost kernel: ? aer_recover_work_func+0xaa/0xd0
Sep 8 22:26:03 localhost kernel: ? process_one_work+0x1a7/0x360
..
Sep 8 22:26:03 localhost kernel: qla2xxx [0000:42:00.2]-8041:22: detected PCI
disconnect.
Sep 8 22:26:03 localhost kernel: qla2xxx [0000:42:00.2]-107ff:22:
qla27xx_fwdt_entry_t262: dump ram MB failed. Area 5h start 198013h end 198013h
Sep 8 22:26:03 localhost kernel: qla2xxx [0000:42:00.2]-107ff:22: Unable to
capture FW dump
Sep 8 22:26:03 localhost kernel: qla2xxx [0000:42:00.2]-1015:22: cmd=0x0,
waited 5221 msecs
Sep 8 22:26:03 localhost kernel: qla2xxx [0000:42:00.2]-680d:22: mmio
enabled returning.
Sep 8 22:26:03 localhost kernel: qla2xxx [0000:42:00.2]-d04c:22: MBX
Command timeout for cmd 0, iocontrol=
ffffffff jiffies=
10140f2e5
mb[0-3]=[0xffff 0xffff 0xffff 0xffff]
Link: https://lore.kernel.org/r/20220110050218.3958-6-njavali@marvell.com
Cc: stable@vger.kernel.org
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: Quinn Tran <qutran@marvell.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Quinn Tran [Mon, 10 Jan 2022 05:02:05 +0000 (21:02 -0800)]
scsi: qla2xxx: Fix warning message due to adisc being flushed
Fix warning message due to adisc being flushed. Linux kernel triggered a
warning message where a different error code type is not matching up with
the expected type. Add additional translation of one error code type to
another.
WARNING: CPU: 2 PID:
1131623 at drivers/scsi/qla2xxx/qla_init.c:498
qla2x00_async_adisc_sp_done+0x294/0x2b0 [qla2xxx]
CPU: 2 PID:
1131623 Comm: drmgr Not tainted 5.13.0-rc1-autotest #1
..
GPR28:
c000000aaa9c8890 c0080000079ab678 c00000140a104800 c00000002bd19000
NIP [
c00800000790857c] qla2x00_async_adisc_sp_done+0x294/0x2b0 [qla2xxx]
LR [
c008000007908578] qla2x00_async_adisc_sp_done+0x290/0x2b0 [qla2xxx]
Call Trace:
[
c00000001cdc3620] [
c008000007908578] qla2x00_async_adisc_sp_done+0x290/0x2b0 [qla2xxx] (unreliable)
[
c00000001cdc3710] [
c0080000078f3080] __qla2x00_abort_all_cmds+0x1b8/0x580 [qla2xxx]
[
c00000001cdc3840] [
c0080000078f589c] qla2x00_abort_all_cmds+0x34/0xd0 [qla2xxx]
[
c00000001cdc3880] [
c0080000079153d8] qla2x00_abort_isp_cleanup+0x3f0/0x570 [qla2xxx]
[
c00000001cdc3920] [
c0080000078fb7e8] qla2x00_remove_one+0x3d0/0x480 [qla2xxx]
[
c00000001cdc39b0] [
c00000000071c274] pci_device_remove+0x64/0x120
[
c00000001cdc39f0] [
c0000000007fb818] device_release_driver_internal+0x168/0x2a0
[
c00000001cdc3a30] [
c00000000070e304] pci_stop_bus_device+0xb4/0x100
[
c00000001cdc3a70] [
c00000000070e4f0] pci_stop_and_remove_bus_device+0x20/0x40
[
c00000001cdc3aa0] [
c000000000073940] pci_hp_remove_devices+0x90/0x130
[
c00000001cdc3b30] [
c0080000070704d0] disable_slot+0x38/0x90 [rpaphp] [
c00000001cdc3b60] [
c00000000073eb4c] power_write_file+0xcc/0x180
[
c00000001cdc3be0] [
c0000000007354bc] pci_slot_attr_store+0x3c/0x60
[
c00000001cdc3c00] [
c00000000055f820] sysfs_kf_write+0x60/0x80 [
c00000001cdc3c20]
[
c00000000055df10] kernfs_fop_write_iter+0x1a0/0x290
[
c00000001cdc3c70] [
c000000000447c4c] new_sync_write+0x14c/0x1d0
[
c00000001cdc3d10] [
c00000000044b134] vfs_write+0x224/0x330
[
c00000001cdc3d60] [
c00000000044b3f4] ksys_write+0x74/0x130
[
c00000001cdc3db0] [
c00000000002df70] system_call_exception+0x150/0x2d0
[
c00000001cdc3e10] [
c00000000000d45c] system_call_common+0xec/0x278
Link: https://lore.kernel.org/r/20220110050218.3958-5-njavali@marvell.com
Cc: stable@vger.kernel.org
Reported-by: Abdul Haleem <abdhalee@linux.vnet.ibm.com>
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: Quinn Tran <qutran@marvell.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Quinn Tran [Mon, 10 Jan 2022 05:02:04 +0000 (21:02 -0800)]
scsi: qla2xxx: Fix stuck session in gpdb
Fix stuck sessions in get port database. When a thread is in the process of
re-establishing a session, a flag is set to prevent multiple threads /
triggers from doing the same task. This flag was left on, where any attempt
to relogin was locked out. Clear this flag, if the attempt has failed.
Link: https://lore.kernel.org/r/20220110050218.3958-4-njavali@marvell.com
Cc: stable@vger.kernel.org
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: Quinn Tran <qutran@marvell.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Saurav Kashyap [Mon, 10 Jan 2022 05:02:03 +0000 (21:02 -0800)]
scsi: qla2xxx: Implement ref count for SRB
The timeout handler and the done function are racing. When
qla2x00_async_iocb_timeout() starts to run it can be preempted by the
normal response path (via the firmware?). qla24xx_async_gpsc_sp_done()
releases the SRB unconditionally. When scheduling back to
qla2x00_async_iocb_timeout() qla24xx_async_abort_cmd() will access an freed
sp->qpair pointer:
qla2xxx [0000:83:00.0]-2871:0: Async-gpsc timeout - hdl=63d portid=234500 50:06:0e:80:08:77:b6:21.
qla2xxx [0000:83:00.0]-2853:0: Async done-gpsc res 0, WWPN 50:06:0e:80:08:77:b6:21
qla2xxx [0000:83:00.0]-2854:0: Async-gpsc OUT WWPN 20:45:00:27:f8:75:33:00 speeds=2c00 speed=0400.
qla2xxx [0000:83:00.0]-28d8:0: qla24xx_handle_gpsc_event 50:06:0e:80:08:77:b6:21 DS 7 LS 6 rc 0 login 1|1 rscn 1|0 lid 5
BUG: unable to handle kernel NULL pointer dereference at
0000000000000004
IP: qla24xx_async_abort_cmd+0x1b/0x1c0 [qla2xxx]
Obvious solution to this is to introduce a reference counter. One reference
is taken for the normal code path (the 'good' case) and one for the timeout
path. As we always race between the normal good case and the timeout/abort
handler we need to serialize it. Also we cannot assume any order between
the handlers. Since this is slow path we can use proper synchronization via
locks.
When we are able to cancel a timer (del_timer returns 1) we know there
can't be any error handling in progress because the timeout handler hasn't
expired yet, thus we can safely decrement the refcounter by one.
If we are not able to cancel the timer, we know an abort handler is
running. We have to make sure we call sp->done() in the abort handlers
before calling kref_put().
Link: https://lore.kernel.org/r/20220110050218.3958-3-njavali@marvell.com
Cc: stable@vger.kernel.org
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
Co-developed-by: Daniel Wagner <dwagner@suse.de>
Signed-off-by: Daniel Wagner <dwagner@suse.de>
Signed-off-by: Saurav Kashyap <skashyap@marvell.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Daniel Wagner [Mon, 10 Jan 2022 05:02:02 +0000 (21:02 -0800)]
scsi: qla2xxx: Refactor asynchronous command initialization
Move common open-coded asynchronous command initializing code such as
setting up the timer and the done callback into one function. This is a
preparation step and allows us later on to change the low level error flow
handling at a central place.
Link: https://lore.kernel.org/r/20220110050218.3958-2-njavali@marvell.com
Cc: stable@vger.kernel.org
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: Daniel Wagner <dwagner@suse.de>
Signed-off-by: Saurav Kashyap <skashyap@marvell.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Christophe JAILLET [Sat, 15 Jan 2022 09:15:41 +0000 (10:15 +0100)]
scsi: bfa: Remove useless DMA-32 fallback configuration
As stated in [1], dma_set_mask() with a 64-bit mask never fails if
dev->dma_mask is non-NULL. So, if it fails, the 32-bit case will also fail
for the same reason.
Simplify code and remove some dead code accordingly.
[1]: https://lore.kernel.org/linux-kernel/YL3vSPK5DXTNvgdx@infradead.org/#t
Link: https://lore.kernel.org/r/5663cef9b54004fa56cca7ce65f51eadfc3ecddb.1642238127.git.christophe.jaillet@wanadoo.fr
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Christophe JAILLET [Sat, 15 Jan 2022 09:05:22 +0000 (10:05 +0100)]
scsi: hisi_sas: Remove useless DMA-32 fallback configuration
As stated in [1], dma_set_mask() with a 64-bit mask never fails if
dev->dma_mask is non-NULL. So, if it fails, the 32-bit case will also fail
for the same reason.
Simplify code and remove some dead code accordingly.
[1]: https://lore.kernel.org/linux-kernel/YL3vSPK5DXTNvgdx@infradead.org/#t
Link: https://lore.kernel.org/r/1bf2d3660178b0e6f172e5208bc0bd68d31d9268.1642237482.git.christophe.jaillet@wanadoo.fr
Acked-by: John Garry <john.garry@huawei.com>
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Christophe JAILLET [Sat, 15 Jan 2022 08:53:03 +0000 (09:53 +0100)]
scsi: 3w-sas: Remove useless DMA-32 fallback configuration
As stated in [1], dma_set_mask() with a 64-bit mask never fails if
dev->dma_mask is non-NULL. So, if it fails, the 32-bit case will also fail
for the same reason.
Simplify code and remove some dead code accordingly.
[1]: https://lore.kernel.org/linux-kernel/YL3vSPK5DXTNvgdx@infradead.org/#t
Link: https://lore.kernel.org/r/dbbe8671ca760972d80f8d35f3170b4609bee368.1642236763.git.christophe.jaillet@wanadoo.fr
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
John Meneghini [Sat, 15 Jan 2022 04:00:44 +0000 (23:00 -0500)]
scsi: bnx2fc: Flush destroy_work queue before calling bnx2fc_interface_put()
The bnx2fc_destroy() functions are removing the interface before calling
destroy_work. This results multiple WARNings from sysfs_remove_group() as
the controller rport device attributes are removed too early.
Replace the fcoe_port's destroy_work queue. It's not needed.
The problem is easily reproducible with the following steps.
Example:
$ dmesg -w &
$ systemctl enable --now fcoe
$ fipvlan -s -c ens2f1
$ fcoeadm -d ens2f1.802
[ 583.464488] host2: libfc: Link down on port (7500a1)
[ 583.472651] bnx2fc: 7500a1 - rport not created Yet!!
[ 583.490468] ------------[ cut here ]------------
[ 583.538725] sysfs group 'power' not found for kobject 'rport-2:0-0'
[ 583.568814] WARNING: CPU: 3 PID: 192 at fs/sysfs/group.c:279 sysfs_remove_group+0x6f/0x80
[ 583.607130] Modules linked in: dm_service_time 8021q garp mrp stp llc bnx2fc cnic uio rpcsec_gss_krb5 auth_rpcgss nfsv4 ...
[ 583.942994] CPU: 3 PID: 192 Comm: kworker/3:2 Kdump: loaded Not tainted 5.14.0-39.el9.x86_64 #1
[ 583.984105] Hardware name: HP ProLiant DL120 G7, BIOS J01 07/01/2013
[ 584.016535] Workqueue: fc_wq_2 fc_rport_final_delete [scsi_transport_fc]
[ 584.050691] RIP: 0010:sysfs_remove_group+0x6f/0x80
[ 584.074725] Code: ff 5b 48 89 ef 5d 41 5c e9 ee c0 ff ff 48 89 ef e8 f6 b8 ff ff eb d1 49 8b 14 24 48 8b 33 48 c7 c7 ...
[ 584.162586] RSP: 0018:
ffffb567c15afdc0 EFLAGS:
00010282
[ 584.188225] RAX:
0000000000000000 RBX:
ffffffff8eec4220 RCX:
0000000000000000
[ 584.221053] RDX:
ffff8c1586ce84c0 RSI:
ffff8c1586cd7cc0 RDI:
ffff8c1586cd7cc0
[ 584.255089] RBP:
0000000000000000 R08:
0000000000000000 R09:
ffffb567c15afc00
[ 584.287954] R10:
ffffb567c15afbf8 R11:
ffffffff8fbe7f28 R12:
ffff8c1486326400
[ 584.322356] R13:
ffff8c1486326480 R14:
ffff8c1483a4a000 R15:
0000000000000004
[ 584.355379] FS:
0000000000000000(0000) GS:
ffff8c1586cc0000(0000) knlGS:
0000000000000000
[ 584.394419] CS: 0010 DS: 0000 ES: 0000 CR0:
0000000080050033
[ 584.421123] CR2:
00007fe95a6f7840 CR3:
0000000107674002 CR4:
00000000000606e0
[ 584.454888] Call Trace:
[ 584.466108] device_del+0xb2/0x3e0
[ 584.481701] device_unregister+0x13/0x60
[ 584.501306] bsg_unregister_queue+0x5b/0x80
[ 584.522029] bsg_remove_queue+0x1c/0x40
[ 584.541884] fc_rport_final_delete+0xf3/0x1d0 [scsi_transport_fc]
[ 584.573823] process_one_work+0x1e3/0x3b0
[ 584.592396] worker_thread+0x50/0x3b0
[ 584.609256] ? rescuer_thread+0x370/0x370
[ 584.628877] kthread+0x149/0x170
[ 584.643673] ? set_kthread_struct+0x40/0x40
[ 584.662909] ret_from_fork+0x22/0x30
[ 584.680002] ---[ end trace
53575ecefa942ece ]---
Link: https://lore.kernel.org/r/20220115040044.1013475-1-jmeneghi@redhat.com
Fixes:
0cbf32e1681d ("[SCSI] bnx2fc: Avoid calling bnx2fc_if_destroy with unnecessary locks")
Tested-by: Guangwu Zhang <guazhang@redhat.com>
Co-developed-by: Maurizio Lombardi <mlombard@redhat.com>
Signed-off-by: Maurizio Lombardi <mlombard@redhat.com>
Signed-off-by: John Meneghini <jmeneghi@redhat.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Steffen Maier [Tue, 18 Jan 2022 16:58:03 +0000 (17:58 +0100)]
scsi: zfcp: Fix failed recovery on gone remote port with non-NPIV FCP devices
Suppose we have an environment with a number of non-NPIV FCP devices
(virtual HBAs / FCP devices / zfcp "adapter"s) sharing the same physical
FCP channel (HBA port) and its I_T nexus. Plus a number of storage target
ports zoned to such shared channel. Now one target port logs out of the
fabric causing an RSCN. Zfcp reacts with an ADISC ELS and subsequent port
recovery depending on the ADISC result. This happens on all such FCP
devices (in different Linux images) concurrently as they all receive a copy
of this RSCN. In the following we look at one of those FCP devices.
Requests other than FSF_QTCB_FCP_CMND can be slow until they get a
response.
Depending on which requests are affected by slow responses, there are
different recovery outcomes. Here we want to fix failed recoveries on port
or adapter level by avoiding recovery requests that can be slow.
We need the cached N_Port_ID for the remote port "link" test with ADISC.
Just before sending the ADISC, we now intentionally forget the old cached
N_Port_ID. The idea is that on receiving an RSCN for a port, we have to
assume that any cached information about this port is stale. This forces a
fresh new GID_PN [FC-GS] nameserver lookup on any subsequent recovery for
the same port. Since we typically can still communicate with the nameserver
efficiently, we now reach steady state quicker: Either the nameserver still
does not know about the port so we stop recovery, or the nameserver already
knows the port potentially with a new N_Port_ID and we can successfully and
quickly perform open port recovery. For the one case, where ADISC returns
successfully, we re-initialize port->d_id because that case does not
involve any port recovery.
This also solves a problem if the storage WWPN quickly logs into the fabric
again but with a different N_Port_ID. Such as on virtual WWPN takeover
during target NPIV failover.
[https://www.redbooks.ibm.com/abstracts/redp5477.html] In that case the
RSCN from the storage FDISC was ignored by zfcp and we could not
successfully recover the failover. On some later failback on the storage,
we could have been lucky if the virtual WWPN got the same old N_Port_ID
from the SAN switch as we still had cached. Then the related RSCN
triggered a successful port reopen recovery. However, there is no
guarantee to get the same N_Port_ID on NPIV FDISC.
Even though NPIV-enabled FCP devices are not affected by this problem, this
code change optimizes recovery time for gone remote ports as a side effect.
The timely drop of cached N_Port_IDs prevents unnecessary slow open port
attempts.
While the problem might have been in code before v2.6.32 commit
799b76d09aee ("[SCSI] zfcp: Decouple gid_pn requests from erp") this fix
depends on the gid_pn_work introduced with that commit, so we mark it as
culprit to satisfy fix dependencies.
Note: Point-to-point remote port is already handled separately and gets its
N_Port_ID from the cached peer_d_id. So resetting port->d_id in general
does not affect PtP.
Link: https://lore.kernel.org/r/20220118165803.3667947-1-maier@linux.ibm.com
Fixes:
799b76d09aee ("[SCSI] zfcp: Decouple gid_pn requests from erp")
Cc: <stable@vger.kernel.org> #2.6.32+
Suggested-by: Benjamin Block <bblock@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>
John Garry [Tue, 18 Jan 2022 12:15:05 +0000 (20:15 +0800)]
scsi: pm8001: Fix bogus FW crash for maxcpus=1
According to the comment in check_fw_ready() we should not check the
IOP1_READY field in register SCRATCH_PAD_1 for 8008 or 8009 controllers.
However we check this very field in process_oq() for processing the highest
index interrupt vector. The highest interrupt vector is checked as the FW
is programmed to signal fatal errors through this irq.
Change that function to not check IOP1_READY for those mentioned
controllers, but do check ILA_READY in both cases.
The reason I assume that this was not hit earlier was because we always
allocated 64 MSI(X), and just did not pass the vector index check in
process_oq(), i.e. the handler never ran for vector index 63.
Link: https://lore.kernel.org/r/1642508105-95432-1-git-send-email-john.garry@huawei.com
Tested-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Reviewed-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Signed-off-by: John Garry <john.garry@huawei.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Saurav Kashyap [Mon, 17 Jan 2022 13:53:11 +0000 (05:53 -0800)]
scsi: qedf: Change context reset messages to ratelimited
If FCoE is not configured, libfc/libfcoe keeps on retrying FLOGI and after
3 retries driver does a context reset and tries fipvlan again. This leads
to context reset message flooding the logs. Hence ratelimit the message to
prevent flooding the logs.
Link: https://lore.kernel.org/r/20220117135311.6256-4-njavali@marvell.com
Signed-off-by: Saurav Kashyap <skashyap@marvell.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Saurav Kashyap [Mon, 17 Jan 2022 13:53:10 +0000 (05:53 -0800)]
scsi: qedf: Fix refcount issue when LOGO is received during TMF
Hung task call trace was seen during LOGO processing.
[ 974.309060] [0000:00:00.0]:[qedf_eh_device_reset:868]: 1:0:2:0: LUN RESET Issued...
[ 974.309065] [0000:00:00.0]:[qedf_initiate_tmf:2422]: tm_flags 0x10 sc_cmd
00000000c16b930f op = 0x2a target_id = 0x2 lun=0
[ 974.309178] [0000:00:00.0]:[qedf_initiate_tmf:2431]: portid=016900 tm_flags =LUN RESET
[ 974.309222] [0000:00:00.0]:[qedf_initiate_tmf:2438]: orig io_req =
00000000ec78df8f xid = 0x180 ref_cnt = 1.
[ 974.309625] host1: rport 016900: Received LOGO request while in state Ready
[ 974.309627] host1: rport 016900: Delete port
[ 974.309642] host1: rport 016900: work event 3
[ 974.309644] host1: rport 016900: lld callback ev 3
[ 974.313243] [0000:61:00.2]:[qedf_execute_tmf:2383]:1: fcport is uploading, not executing flush.
[ 974.313295] [0000:61:00.2]:[qedf_execute_tmf:2400]:1: task mgmt command success...
[ 984.031088] INFO: task jbd2/dm-15-8:7645 blocked for more than 120 seconds.
[ 984.031136] Not tainted 4.18.0-305.el8.x86_64 #1
[ 984.031166] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[ 984.031209] jbd2/dm-15-8 D 0 7645 2 0x80004080
[ 984.031212] Call Trace:
[ 984.031222] __schedule+0x2c4/0x700
[ 984.031230] ? unfreeze_partials.isra.83+0x16e/0x1a0
[ 984.031233] ? bit_wait_timeout+0x90/0x90
[ 984.031235] schedule+0x38/0xa0
[ 984.031238] io_schedule+0x12/0x40
[ 984.031240] bit_wait_io+0xd/0x50
[ 984.031243] __wait_on_bit+0x6c/0x80
[ 984.031248] ? free_buffer_head+0x21/0x50
[ 984.031251] out_of_line_wait_on_bit+0x91/0xb0
[ 984.031257] ? init_wait_var_entry+0x50/0x50
[ 984.031268] jbd2_journal_commit_transaction+0x112e/0x19f0 [jbd2]
[ 984.031280] kjournald2+0xbd/0x270 [jbd2]
[ 984.031284] ? finish_wait+0x80/0x80
[ 984.031291] ? commit_timeout+0x10/0x10 [jbd2]
[ 984.031294] kthread+0x116/0x130
[ 984.031300] ? kthread_flush_work_fn+0x10/0x10
[ 984.031305] ret_from_fork+0x1f/0x40
There was a ref count issue when LOGO is received during TMF. This leads to
one of the I/Os hanging with the driver. Fix the ref count.
Link: https://lore.kernel.org/r/20220117135311.6256-3-njavali@marvell.com
Signed-off-by: Saurav Kashyap <skashyap@marvell.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Saurav Kashyap [Mon, 17 Jan 2022 13:53:09 +0000 (05:53 -0800)]
scsi: qedf: Add stag_work to all the vports
Call trace seen when creating NPIV ports, only 32 out of 64 show online.
stag work was not initialized for vport, hence initialize the stag work.
WARNING: CPU: 8 PID: 645 at kernel/workqueue.c:1635 __queue_delayed_work+0x68/0x80
CPU: 8 PID: 645 Comm: kworker/8:1 Kdump: loaded Tainted: G IOE --------- --
4.18.0-348.el8.x86_64 #1
Hardware name: Dell Inc. PowerEdge MX740c/0177V9, BIOS 2.12.2 07/09/2021
Workqueue: events fc_lport_timeout [libfc]
RIP: 0010:__queue_delayed_work+0x68/0x80
Code: 89 b2 88 00 00 00 44 89 82 90 00 00 00 48 01 c8 48 89 42 50 41 81
f8 00 20 00 00 75 1d e9 60 24 07 00 44 89 c7 e9 98 f6 ff ff <0f> 0b eb
c5 0f 0b eb a1 0f 0b eb a7 0f 0b eb ac 44 89 c6 e9 40 23
RSP: 0018:
ffffae514bc3be40 EFLAGS:
00010006
RAX:
ffff8d25d6143750 RBX:
0000000000000202 RCX:
0000000000000002
RDX:
ffff8d2e31383748 RSI:
ffff8d25c000d600 RDI:
ffff8d2e31383788
RBP:
ffff8d2e31380de0 R08:
0000000000002000 R09:
ffff8d2e31383750
R10:
ffffffffc0c957e0 R11:
ffff8d2624800000 R12:
ffff8d2e31380a58
R13:
ffff8d2d915eb000 R14:
ffff8d25c499b5c0 R15:
ffff8d2e31380e18
FS:
0000000000000000(0000) GS:
ffff8d2d1fb00000(0000) knlGS:
0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0:
0000000080050033
CR2:
000055fd0484b8b8 CR3:
00000008ffc10006 CR4:
00000000007706e0
DR0:
0000000000000000 DR1:
0000000000000000 DR2:
0000000000000000
DR3:
0000000000000000 DR6:
00000000fffe0ff0 DR7:
0000000000000400
PKRU:
55555554
Call Trace:
queue_delayed_work_on+0x36/0x40
qedf_elsct_send+0x57/0x60 [qedf]
fc_lport_enter_flogi+0x90/0xc0 [libfc]
fc_lport_timeout+0xb7/0x140 [libfc]
process_one_work+0x1a7/0x360
? create_worker+0x1a0/0x1a0
worker_thread+0x30/0x390
? create_worker+0x1a0/0x1a0
kthread+0x116/0x130
? kthread_flush_work_fn+0x10/0x10
ret_from_fork+0x35/0x40
---[ end trace
008f00f722f2c2ff ]--
Initialize stag work for all the vports.
Link: https://lore.kernel.org/r/20220117135311.6256-2-njavali@marvell.com
Signed-off-by: Saurav Kashyap <skashyap@marvell.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Xiaoke Wang [Sun, 16 Jan 2022 03:06:49 +0000 (11:06 +0800)]
scsi: ufs: ufshcd-pltfrm: Check the return value of devm_kstrdup()
devm_kstrdup() returns pointer to allocated string on success, NULL on
failure. So it is better to check the return value of it.
Link: https://lore.kernel.org/r/tencent_4257E15D4A94FF9020DDCC4BB9B21C041408@qq.com
Reviewed-by: Bean Huo <beanhuo@micron.com>
Signed-off-by: Xiaoke Wang <xkernel.wang@foxmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
ZouMingzhe [Tue, 11 Jan 2022 05:47:42 +0000 (13:47 +0800)]
scsi: target: iscsi: Make sure the np under each tpg is unique
iscsit_tpg_check_network_portal() has nested for_each loops and is supposed
to return true when a match is found. However, the tpg loop will still
continue after existing the tpg_np loop. If this tpg_np is not the last the
match value will be changed.
Break the outer loop after finding a match and make sure the np under each
tpg is unique.
Link: https://lore.kernel.org/r/20220111054742.19582-1-mingzhe.zou@easystack.cn
Signed-off-by: ZouMingzhe <mingzhe.zou@easystack.cn>
Reviewed-by: Mike Christie <michael.christie@oracle.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Yang Yingliang [Tue, 11 Jan 2022 01:24:41 +0000 (09:24 +0800)]
scsi: elx: efct: Don't use GFP_KERNEL under spin lock
GFP_KERNEL/GFP_DMA can't be used under a spin lock. According the comment,
els_ios_lock is used to protect els ios list so we can move down the spin
lock to avoid using this flag under the lock.
Link: https://lore.kernel.org/r/20220111012441.3232527-1-yangyingliang@huawei.com
Fixes:
8f406ef72859 ("scsi: elx: libefc: Extended link Service I/O handling")
Reported-by: Hulk Robot <hulkci@huawei.com>
Reviewed-by: James Smart <jsmart2021@gmail.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>