[PATCH RFC] Clarify FAILED status handling

On Wed, 30 Nov 2016 16:57:16 +0100
Jan Kiszka <> wrote:
> On 2016-11-30 16:34, Cornelia Huck wrote:
>
> We did not give any guidance before what the device is supposed to
>
> do if the driver set the FAILED status bit. Instruct the device to
>
> leave the driver alone and to ignore any driver actions other than
>
> reset.
> >
>
> As we did not specify that before, formulate this as SHOULD so that
>
> existing implementations stay compliant.
> >
>
> Signed-off-by: Cornelia Huck <>
> > ---
> >
>
> I noticed that we don't really say what the FAILED status bit
>
> should effect (it's currently basically driver-only, and e.g.
>
> qemu does not do anything with it). It is probably more useful
>
> if it actually has consequences for device operation -- although
>
> we certainly don't want to break existing implementations.
>
> Is there a real use case in sight? The other option is to phase it out
> if it is practically without effect so far and we only speculate that it
> might be useful.

Not really. The only useful thing about FAILED today is that you can
notice that something went wrong if you see it e.g. in a dump (and even
then, it does not tell you much).

I'd not be opposed to phasing it out, either.