Re: [Xen-devel] [PATCH 22/29] libxl: Introduce DOMAIN_DESTROYED error code

[Ian Jackson <Ian.Jackson@xxxxxxxxxxxxx>]

Ian Campbell writes ("Re: [Xen-devel] [PATCH 22/29] libxl: Introduce 
DOMAIN_DESTROYED error code"):
> Konrad has a semantically similar error code which he is adding, I think
> in his recent libxl series to do with vcpu-set.

That is valuable, I think.

> AIUI Konrad's semantics are simply "domain does not exist", which seems
> to be usefully distinct from your "did exist but doesn't any more".

Indeed, although it's not clear to me without peering at the code in
detail whether my code always checks that the domain does exist before
setting up the watch which might result in its death being reported.

> I just wanted to mention it in case I'd misunderstood one or both error
> codes. As it stands this patch seems fine to me:
...
> Acked-by: Ian Campbell <ian.campbell@xxxxxxxxxx>

Thanks.

> I do wonder though if we ought to be better about documenting in the
> code|headers|idl what error codes mean and where they should be used
> (some are global, others specific to a subset of calls etc).

That would definitely be nice.  Do you think it would be helpful to
respin this patch with that information ?

Ian.

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel