[tor-bugs] #29085 [Core Tor/Tor]: WTF-PAD: Reduce monotime usage because of performance issues

[Tor Bug Tracker & Wiki]

#29085: WTF-PAD: Reduce monotime usage because of performance issues
-------------------------------------------------+-------------------------
 Reporter:  asn                                  |          Owner:  (none)
     Type:  defect                               |         Status:
                                                 |  needs_information
 Priority:  Very High                            |      Milestone:  Tor:
                                                 |  unspecified
Component:  Core Tor/Tor                         |        Version:
 Severity:  Normal                               |     Resolution:
 Keywords:  wtf-pad, tor-relay, tor-cell,        |  Actual Points:
  padding, 041-proposed                          |
Parent ID:  #28634                               |         Points:  4
 Reviewer:                                       |        Sponsor:
                                                 |  Sponsor2
-------------------------------------------------+-------------------------
Changes (by mikeperry):

 * status:  new => needs_information
 * cc: asn (added)


Comment:

 Ok, so if we don't make use of token removal or RTT estimates, we don't
 need monotime absolute_usec() at all.

 However, right now our machine in #28634 does use token removal
 (unnecessarily, though), and probably *should* use RTT estimates. The good
 news is that RTT estimates only need to make like 3 monotime calls, and
 only for extend cells on circuits. So that usage is not heavy (and their
 accuracy requirements are probably ~5ms, give or take).

 The other way to do this is to make a runtime test of how expensive
 monotime is, and decide if we should use monotime_coarse based on that
 result.

--
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/29085#comment:5>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online