Re: [PATCH v2 5/6] spec: Introduce NBD_FLAG_BLOCK_STATUS_PAYLOAD

[Wouter Verhelst]

On Fri, Mar 03, 2023 at 04:40:38PM -0600, Eric Blake wrote:
> On Wed, Feb 22, 2023 at 12:05:44PM +0200, Wouter Verhelst wrote:
> > On Mon, Nov 14, 2022 at 04:46:54PM -0600, Eric Blake wrote:
> > >  #### Simple reply message
> > > 
> > > @@ -1232,6 +1235,19 @@ The field has the following format:
> > >    will be faster than a regular write). Clients MUST NOT set the
> > >    `NBD_CMD_FLAG_FAST_ZERO` request flag unless this transmission flag
> > >    is set.
> > > +- bit 12, `NBD_FLAG_BLOCK_STATUS_PAYLOAD`: Indicates that the server
> > > +  understands the use of the `NBD_CMD_FLAG_PAYLOAD_LEN` flag to
> > > +  `NBD_CMD_BLOCK_STATUS` to allow the client to request that the
> > > +  server filters its response to a specific subset of negotiated
> > > +  metacontext ids passed in via a client payload, rather than the
> > > +  default of replying to all metacontext ids. Servers MUST NOT
> > > +  advertise this bit unless the client successfully negotiates
> > > +  extended headers via `NBD_OPT_EXTENDED_HEADERS`, and SHOULD NOT
> > > +  advertise this bit in response to `NBD_OPT_EXPORT_NAME` or
> > > +  `NBD_OPT_GO` if the client does not negotiate metacontexts with
> > > +  `NBD_OPT_SET_META_CONTEXT`; clients SHOULD NOT set the
> > > +  `NBD_CMD_FLAG_PAYLOAD_LEN` flag for `NBD_CMD_BLOCK_STATUS` unless
> > > +  this transmission flag is set.
> > 
> > Given that we are introducing this at the same time as the extended
> > headers (and given that we're running out of available flags in this
> > 16-bit field), isn't it better to make the ability to understand
> > PAYLOAD_LEN be implied by extended headers? Or is there a use case for
> > implementing extended headers but not a payload length on
> > CMD_BLOCK_STATUS that I'm missing?
> 
> In my proof of implementation, this was a distinct feature addition on
> top of 64-bit headers.
> 
> It is easy to write a server that does not want to deal with a client
> payload on a block status request (for example, a server that only
> knows about one metacontext, and therefore never needs to deal with a
> client wanting a subset of negotiated metacontexts).  Forcing a server
> to have to support a client-side filtering request on block status
> commands, merely because the server now supports 64-bit lengths,
> seemed like a stretch too far, which is why I decided to use a feature
> bit for this one.

Okay, yes, that makes sense. Thanks.

-- 
     w@uter.{be,co.za}
wouter@{grep.be,fosdem.org,debian.org}

I will have a Tin-Actinium-Potassium mixture, thanks.