Re: [Xen-devel] REST MAINTAINERS feedback requested Was:Re: [PATCH v5 01/28] HYPERCALL_version_op. New hypercall mirroring XENVER_ but sane.

["Jan Beulich" <jbeulich@xxxxxxxx>]

>>> Ian Jackson <Ian.Jackson@xxxxxxxxxxxxx> 04/12/16 6:47 PM >>>
>George Dunlap writes ("Re: [Xen-devel] REST MAINTAINERS feedback requested 
>Was:Re: [PATCH v5 01/28] HYPERCALL_version_op. New hypercall mirroring XENVER_ 
>>but sane."):
>> Well we know which option Andy prefers, but are there other options
>> that Andy is not absolutely opposed to?  And we don't know anything
>> about which option Jan prefers at all, except that it's not #4.
>
>Let me go a bit further than George.
>
>It's clear that there are various options, most of which are
>tolerable.  Buit if I'm trying to help referee a disagreement between
>Andrew and Jan I would prefer to be choosing between Andrew's
>preferred answer and Jan's preferred answer.
>
>Jan: AFAICT it's clear that you would still like the current patch
>reverted.  Can you please say what, if anything, you would like to
>replace it with ?

That patch doesn't need replacing by anything. It's the follow-up patch adding
support to retrieve the build-id which would need a replacement, and several
options have been put on the table. As mentioned before, I'd prefer the variant
of the new sub-op getting added to the existing version hypercall, with the
needed length argument passed either via a structure element, with the
structure pointed to by the 2nd hypercall argument, or with the high bits of
the first hypercall argument re-purposed to allow (and for this sub-op require)
holding a length. Which of these two sub-options is chosen I don't really care
much.

Jan


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