Good point -- though the pseudo-code should work fine, even if redundant. Yes, the pto_count is incremented before SetLossDetectionTimer() is called, and that will push the timer to run for the longer, correct timeout.
My point was not to say that implementers cannot use the code as is -- they can, since the pseudo code is expected to be correct -- just that they might include some redundancies that were introduced here for clarity. On Fri, Sep 11, 2020 at 5:27 AM Lars Eggert <[email protected]> wrote: > Hi "newbie quic", > > anonymous IETF participation isn't really possible. Please see > https://github.com/quicwg/base-drafts/blob/master/CONTRIBUTING.md and > referenced documents. > > Thanks, > Lars >
