[smartmontools-support] Kernel errors on Solaris 11 x64 after moving from 5.39 to 5.40

[Peter R. <pe...@ra...>]

My box has 21 SATA disks, some of them directly connected to the motherboard, others via LSI 3081E-R.
OS is Solaris Express 11 x64 (2010.11 SRU5).

Doing a "smartctl -d sat -a /dev/rdsk/c7t12d0" with smartmontools 5.39 on any of the disks delivers the expected output and no kernel warning.

However, with 5.40 and everything else unchanged I get:

May  8 17:55:01 rocky scsi: [ID 107833 kern.warning] WARNING: /pci@0,0/pci8086,2e21@1/pci1000,3140@0/sd@c,0 (sd7):
May  8 17:55:01 rocky   Error for Command: <undecoded cmd 0x85>    Error Level: Recovered
May  8 17:55:01 rocky scsi: [ID 107833 kern.notice]     Requested Block: 0                         Error Block: 0
May  8 17:55:01 rocky scsi: [ID 107833 kern.notice]     Vendor: ATA                                Serial Number:       MN1220
May  8 17:55:01 rocky scsi: [ID 107833 kern.notice]     Sense Key: Soft_Error
May  8 17:55:01 rocky scsi: [ID 107833 kern.notice]     ASC: 0x0 (<vendor unique code 0x0>), ASCQ: 0x1d, FRU: 0x0

or (on another disk):

May  8 17:31:39 rocky scsi: [ID 107833 kern.warning] WARNING: /pci@0,0/pci1043,82d4@1f,2/disk@3,0 (sd3):
May  8 17:31:39 rocky   Error for Command: <undecoded cmd 0xa1>    Error Level: Recovered
May  8 17:31:39 rocky scsi: [ID 107833 kern.notice]     Requested Block: 0                         Error Block: 0
May  8 17:31:39 rocky scsi: [ID 107833 kern.notice]     Vendor: ATA                                Serial Number:
May  8 17:31:39 rocky scsi: [ID 107833 kern.notice]     Sense Key: Soft_Error
May  8 17:31:39 rocky scsi: [ID 107833 kern.notice]     ASC: 0x0 (no additional sense info), ASCQ: 0x0, FRU: 0x0

How can I avoid these warnings (which reappear every 30 minutes from smartd)?