diff options
| author |   Sreekanth Reddy <sreekanth.reddy@broadcom.com> | 2021-11-17 16:20:58 +0530 |
|---|---|---|
| committer |   Martin K. Petersen <martin.petersen@oracle.com> | 2021-11-18 22:38:59 -0500 |
| commit | 91202a01a2fb2b78da3d03811b6d3d973ae426aa (patch) | |
| tree | ae646dea27aaff44922dfb0be1f9e585040dffe5 /scripts | |
| parent | scsi: mpt3sas: Fix kernel panic during drive powercycle test (diff) | |
| download | linux-dev-91202a01a2fb2b78da3d03811b6d3d973ae426aa.tar.xz linux-dev-91202a01a2fb2b78da3d03811b6d3d973ae426aa.zip | |
scsi: mpt3sas: Fix system going into read-only mode
While determining the SAS address of a drive, the driver checks whether the
handle number is less than the HBA phy count or not. If the handle number
is less than the HBA phy count then driver assumes that this handle belongs
to HBA and hence it assigns the HBA SAS address.
During IOC firmware downgrade operation, if the number of HBA phys is
reduced and the OS drive's device handle drops below the phy count while
determining the drive's SAS address, the driver ends up using the HBA's SAS
address. This leads to a mismatch of drive's SAS address and hence the
driver unregisters the OS drive and the system goes into read-only mode.
Update the IOC's num_phys to the HBA phy count provided by actual loaded
firmware.
Link: https://lore.kernel.org/r/20211117105058.3505-1-sreekanth.reddy@broadcom.com
Fixes: a5e99fda0172 ("scsi: mpt3sas: Update hba_port objects after host reset")
Signed-off-by: Sreekanth Reddy <sreekanth.reddy@broadcom.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Diffstat (limited to 'scripts')
0 files changed, 0 insertions, 0 deletions