Re: IESG feedback on core drafts.

[Eric Rescorla <ekr%rtfm.com@localhost>]

Tero Kivinen <kivinen%iki.fi@localhost> writes:

> Eric Rescorla writes:
> > As this example indicates, it's totally unsafe to use the existence
> > of unflushed data in the TCP buffers proper as a guide to whether
> > you need an empty packet, since when you do the second write(),
> > the buffers will contain the un-ACKed Record 1.
> 
> It is safe, if and only if you get the information from the TCP stack
> if that data at the end of the TCP buffers have ever been sent out to
> network. If it has never ever been sent out to network, then we do not
> need the extra packet, if it has already been sent out, then you do
> need extra packet. 
Yes, but since essentially no stack will tell you this, it's not
really an optimization worth making.

> > The point is that you can't safely talk about TCP packets. If you want
> > to talk about conditional empty packet insertion you should talk about
> > the data being "written" "delivered to the network"
> 
> I think I have been using term "seen on the network" all the time, I
> didn't talk about TCP packet. 
But that's the language that appeared in the text I was criticizing.

-Ekr

-- 
[Eric Rescorla                                   ekr%rtfm.com@localhost]
                http://www.rtfm.com/