Refill token buckets less often on slow connections

[robgjansen]

The token bucket now returns the time that the next packet conforms, ie how long until we will have enough tokens to forward that packet. Returning this instead of how long until the next refill reduces the overall number of forwarding events we need to schedule, and is especially helpful on slow connections.

[codecov]

Codecov Report

Base: 67.21% // Head: 68.14% // Increases project coverage by +0.92% 🎉

Coverage data is based on head (d5e9187) compared to base (88e32b7).
Patch coverage: 64.51% of modified lines in pull request are covered.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #2707      +/-   ##
==========================================
+ Coverage   67.21%   68.14%   +0.92%     
==========================================
  Files         203      203              
  Lines       30484    30515      +31     
  Branches     5961     5971      +10     
==========================================
+ Hits        20491    20795     +304     
+ Misses       5329     5004     -325     
- Partials     4664     4716      +52     

Flag	Coverage Δ
tests	68.14% <64.51%> (+0.92%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
src/main/network/relay/mod.rs	83.51% <50.00%> (+12.72%)	⬆️
src/main/network/relay/token_bucket.rs	69.53% <65.51%> (-0.17%)	⬇️
src/lib/shmem/src/scmutex.rs	87.05% <0.00%> (-2.36%)	⬇️
src/main/core/scheduler/pools/bounded.rs	74.50% <0.00%> (-1.97%)	⬇️
src/lib/tsc/tsc_test.c	12.67% <0.00%> (ø)
src/main/utility/byte_queue.rs	56.01% <0.00%> (ø)
src/main/utility/shm_cleanup.rs	61.36% <0.00%> (ø)
src/main/host/syscall/handler/socket.rs	66.66% <0.00%> (+0.22%)	⬆️
src/lib/shadow-shim-helper-rs/src/signals.rs	72.28% <0.00%> (+0.60%)	⬆️
src/main/host/descriptor/mod.rs	68.29% <0.00%> (+0.81%)	⬆️
... and 22 more

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

☔ View full report at Codecov.
📢 Do you have feedback about the report comment? Let us know in this issue.

[robgjansen]

Tor benchmark results seem reasonable:
https://github.com/shadow/benchmark-results/tree/master/tor/2023-01-29-T15-04-36

[stevenengler]

The simulation did seem to get a bit slower in the tor benchmark, but I would have expected the opposite. Maybe worth waiting for tonight's nightly to run, rebasing this PR, and then running the simulation again to check if it was just within error since the weekly has large error bars. But I'm fine either way since I don't see anything wrong with this PR.

[robgjansen]

Rebased and ran another Tor benchmark as suggested, and the performance is in line with the nightly from last night:
(https://github.com/shadow/benchmark-results/tree/master/tor/2023-01-31-T14-09-13)

(We currently think the regression on the nightly was from #2632.)