On Fri, Dec 16, 2022 at 08:44:22PM +0000, Sagar.Biradar@xxxxxxxxxxxxx wrote: > Konstantin, > Is it possible for you or someone you know to test on your original test bed with the "msi" set to "1", and post the results? Hello, I had lately some exchange with Konstantin about the issue and he suggested to document my findings on LKML for the benefit of the wider community. I was encountering the same issue with an Adaptec 6405E with 4 connected disks configured on the controller as "Simple_volume" and further utilised in a Linux software RAID (mdraid) configuration. My CPU is a i7-3930K. I had set "msi" to "2" and the problem persisted. I have not tested msi=1 but as far as I understand it should have the same effect with regards to the system hangs. I have applied Konstantins [PATCH v2 1/2] to a v6.6 line kernel and the system hangs disappeared. My system has been running for 12 days without any syslog message indicating a controller problem despite occasional heavy I/O load. Please let me know if you require any additional information. Kind regards, Dirk.
