Re: [PATCH v9 09/10] admin: conformance clauses

[Cornelia Huck <cohuck@redhat.com>]

On Fri, Nov 25 2022, "Michael S. Tsirkin" <mst@redhat.com> wrote:

> On Fri, Nov 25, 2022 at 12:42:23PM +0100, Cornelia Huck wrote:
>> On Fri, Nov 25 2022, Jason Wang <jasowang@redhat.com> wrote:
>> 
>> > å 2022/11/24 16:36, Michael S. Tsirkin åé:
>> >> On Thu, Nov 24, 2022 at 02:51:21PM +0800, Jason Wang wrote:
>> >>> On Thu, Nov 24, 2022 at 5:08 AM Michael S. Tsirkin <mst@redhat.com> wrote:
>> >>>> +The device MUST NOT fail a command solely because the buffers
>> >>>> +provided are shorter or longer than described in this
>> >>>> +specification.
>> >>> I may miss something but how can it work if the buffer is shorter?
>> >> driver does not care what's there.
>> >>
>> >> this is mostly for forward compatibility - we'll add more fields and
>> >> I don't want to explain separately that old drivers post
>> >> short buffers with less fields.
>> >
>> >
>> > For example:
>> >
>> > The patch said:
>> >
>> > struct virtio_admin_cmd_list {
>> >  ÂÂÂÂÂÂ /* Indicates which of the below fields were returned
>> >  ÂÂÂÂÂÂ le32 device_admin_cmds[];
>> > };
>> >
>> > Does it mean the query can still succeed even if there's no space for 
>> > virtio_admin_cmd_list in the writable buffer?
>> 
>> Can we pick an absolute minimum length? All drivers need to be able to
>> at least accommodate this one. Or is it ok for the driver to send a
>> completely useless query that doesn't get any data back?
>
> I think it might be a way to check that a group exists.

Ok, so not completely useless.