Re: completion timeouts with pin-based interrupts in QEMU hw/nvme

[Keith Busch]

On Thu, Jan 12, 2023 at 02:10:51PM +0100, Klaus Jensen wrote:
> Hi all (linux-nvme, qemu-devel, maintainers),
> 
> On QEMU riscv64, which does not use MSI/MSI-X and thus relies on
> pin-based interrupts, I'm seeing occasional completion timeouts, i.e.
> 
>   nvme nvme0: I/O 333 QID 1 timeout, completion polled

I finally got a riscv setup recreating this, and I am not sure how
you're getting a "completion polled" here. I find that the polling from
here progresses the request state to IDLE, so the timeout handler moves
on to aborting because it's expecting COMPLETED. The driver should not
be doing that, so I'll fix that up while also investigating why the
interrupt isn't retriggering as expected.