Re: [PATCH v6 11/18] nvme-rdma: Use CCR to recover controller that hits an error
From: Mohamed Khalfella
Date: Mon Oct 26 2026 - 14:16:07 EST
On Sun 2026-09-20 11:28:09 -0700, Mohamed Khalfella wrote:
> An alive nvme controller that hits an error now will move to FENCING
> state instead of RESETTING state. ctrl->fencing_work attempts CCR to
> terminate inflight IOs. Regardless of the success or failure of CCR
> operation the controller is transitioned to RESETTING state to continue
> error recovery process.
>
> Signed-off-by: Mohamed Khalfella <mkhalfella@xxxxxxxxxxxxxxx>
Hello Sagi,
Can you please take a look v6 of this patch. I applied the change
discussed in v5 by moving flush_work(&ctrl->fencing_work) to
.stop_ctrl(). It looks much better now.
I would like to get these changes merged in 7.5, and getting this patch
reviewed leaves only nvme-fc change which I believe Daniel is looking at
now.
Thanks,