[bug report] RIP: 0010:blk_flush_complete_seq+0x450/0x1060 observed during blktests nvme/tcp nvme/012

[Yi Zhang]

Confirmed the issue was fixed with this patch.
https://lore.kernel.org/linux-block/20240501110907.96950-9-dlemoal@kernel.org/

On Sat, May 4, 2024 at 5:14 AM Chaitanya Kulkarni <chaitanyak at nvidia.com> wrote:
>
>
> >> This is something Damien added to his patch series. I just wonder, why I
> >> couldn't reproduce the failure, even with nvme-mpath enabled. I tried
> >> both nvme-tcp as well as nvme-loop without any problems.
> >
> > Not exactly sure.
> >
> >  From what I see blk_flush_complete_seq() will only call
> > blk_flush_restore_request() and
> > panic is for error != 0. And if that is the case, any request with its
> > bios stolen must panic.
> >
> > However, nvme-mpath always ends a stolen request with error = 0.
> >
> > Seems that there is code that may override the request error status in
> > flush_end_io() but I cannot
> > see it in the trace...
>
>
> I confirm that after several tries I cannot reproduced it either with
> and without multi-pathing, blktests is passing without any errors for
> both nvme-loop and nvme-tcp ....
>
> -ck
>
>


-- 
Best Regards,
  Yi Zhang