Re: [Nbd] [PATCH v4] doc: Propose STRUCTURED_REPLY extension

[Wouter Verhelst <w@...112...>]

On Fri, Apr 01, 2016 at 06:51:34AM -0600, Eric Blake wrote:
> On 04/01/2016 02:57 AM, Wouter Verhelst wrote:
> > Did a few minor changes, though:
> > 
> > On Thu, Mar 31, 2016 at 11:29:48PM -0600, Eric Blake wrote:
> >> -S: 32 bits, 0x67446698, magic (`NBD_REPLY_MAGIC`)  
> >> -S: 32 bits, error  
> >> +S: 32 bits, 0x67446698, magic (`NBD_SIMPLE_REPLY_MAGIC`)  
> > 
> > Added a reference to the old name, for clarity.
> > 
> > [...]
> >> +- bit 6, `NBD_FLAG_SEND_DF`; defined by the `STRUCTURED_REPLY` extension;
> >> +  see below.
> > 
> > This is now bit 7 (we're already halfway through our available flags! time
> > flies...)
> 
> Well, if (when?) we get to 15, one possible solution is fairly
> straightforward: the server advertises a handshake flag, and the client
> replies with the counterpart flag, and then both sides know to send 64
> bits instead of 16 bits for all places that send a transmission flag.
> It may make documentation of NBD_REP_SERVER (as used by NBD_OPT_GO) a
> bit trickier, in that the length returned in the option reply is
> dependent on what handshake flags were negotiated.

There are certainly ways to fix this issue when we get around to it. I
was actually thinking of having an NBD_OPT_FLAGS or something. Ah well.

We'll deal with that when we need to :-)

-- 
< ron> I mean, the main *practical* problem with C++, is there's like a dozen
       people in the world who think they really understand all of its rules,
       and pretty much all of them are just lying to themselves too.
 -- #debian-devel, OFTC, 2016-02-12