<html><head></head><body>Hi Catalin,<br>
<br>
Yes, that's what I was thinking of. I can implement it.<br>
<br>
Cheers,<br>
Matt<br><br><div class="gmail_quote">On 14 July 2014 2:56:17 am AWST, Catalin Patulea <cat@vv.carleton.ca> wrote:<blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
<pre class="k9mail">On Tue, Jul 8, 2014 at 9:44 AM, Matt Johnston <matt@ucc.asn.au> wrote:<br /><blockquote class="gmail_quote" style="margin: 0pt 0pt 1ex 0.8ex; border-left: 1px solid #729fcf; padding-left: 1ex;"> I'm not really sure how to resolve it though. Maybe as a<br /> tradeoff the refcounting could just switch between LOWDELAY<br /> and no-tos-flags if there are TCP forwards going on, then<br /> BULK if there aren't TCP forwards?<br /></blockquote>Ok, so if I'm understanding correctly, the invariant would be something like:<br /><br />if (connecting || ptys) tos = LOWDELAY;<br />else if (tcp_forwards) tos = 0;<br />else tos = BULK;<br /><br />This might be implemented cleverly so that it's only reapplied on 0/1<br />transitions of ptys and tcp_forwards, but these would be the<br />high-level semantics?<br /></pre></blockquote></div></body></html>