Re: [quicwg/base-drafts] loss recovery of crypto packets is less aggressive, not more aggressive (#2435)

[ianswett <[e-mail]>]

Good point, that is a case when the new approach is slightly less aggressive.

If we wanted to favor aggressiveness during the handshake, we could arm the crypto timer based on the left edge(earliest unacked ack-eliciting packet) rather than the right(latest unacked).  That's a larger change than I'd like to make in this CL/issue, though.

I believe the example you outlined above is fairly unlikely, because for most connections, if multiple initials are sent, they're likely to be re-transmissions, not new data.  In that case, when Initial 2 is acked, Initial 1 is immediately lost and Handshake 1 is still the only outstanding ack-eliciting packet.  That being said, I can probably construct other situations that are somewhat similar.



-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/quicwg/base-drafts/issues/2435#issuecomment-462385776