IRQ/nvme_pci_complete_rq: NULL pointer dereference yet again

[Keith Busch]

On Mon, Apr 09, 2018 at 02:36:30PM -0500, Alex G. wrote:
> Although I remember a time when I could run these sort of tests without
> panic(), I don't have a solid base from which I can bisect things.
> Especially since we've added a few fixes lately that won't be in any base.
> 
> I'll try a couple of branches, but don't hold your breath.

This may be a regression. Your panic is triggered through the timeout
path, which appears to have had recent modifications that are included
in your kernel. There's at least one issue in this area:

  https://marc.info/?l=linux-block&m=152325124306821&w=2

I haven't looked into it enough to know if it's related to this issue,
though.