fix: Add confirmed parameter to PTO calculation
#2127
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Rather than having the caller determine for which space a PTO should be calculated for. Broken out of mozilla#1998 I'm ambivalent if we want this change - thoughts?
larseggert
requested review from
KershawChang,
martinthomson and
mxinden
as code owners
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## main #2127 +/- ##
=======================================
Coverage ? 95.35%
=======================================
Files ? 112
Lines ? 36344
Branches ? 0
=======================================
Hits ? 34657
Misses ? 1687
Partials ? 0 ☔ View full report in Codecov by Sentry. |
Benchmark resultsPerformance differences relative to d6279bf. coalesce_acked_from_zero 1+1 entries: No change in performance detected. time: [99.131 ns 99.389 ns 99.656 ns]
change: [-0.6464% -0.1009% +0.4813%] (p = 0.75 > 0.05)
coalesce_acked_from_zero 3+1 entries: No change in performance detected. time: [117.02 ns 117.38 ns 117.76 ns]
change: [-0.7137% +0.0365% +1.0553%] (p = 0.94 > 0.05)
coalesce_acked_from_zero 10+1 entries: No change in performance detected. time: [116.81 ns 117.36 ns 118.00 ns]
change: [-0.9475% -0.2742% +0.3536%] (p = 0.43 > 0.05)
coalesce_acked_from_zero 1000+1 entries: No change in performance detected. time: [97.394 ns 97.547 ns 97.714 ns]
change: [-0.5211% +0.4304% +1.4109%] (p = 0.41 > 0.05)
RxStreamOrderer::inbound_frame(): Change within noise threshold. time: [110.88 ms 110.92 ms 110.95 ms]
change: [-0.2485% -0.1930% -0.1333%] (p = 0.00 < 0.05)
transfer/pacing-false/varying-seeds: Change within noise threshold. time: [25.351 ms 26.393 ms 27.432 ms]
change: [-11.686% -6.1750% -0.4850%] (p = 0.04 < 0.05)
transfer/pacing-true/varying-seeds: No change in performance detected. time: [34.734 ms 36.221 ms 37.699 ms]
change: [-7.3691% -1.2320% +5.5718%] (p = 0.72 > 0.05)
transfer/pacing-false/same-seed: No change in performance detected. time: [25.291 ms 26.061 ms 26.834 ms]
change: [-4.9918% -0.8113% +3.4423%] (p = 0.71 > 0.05)
transfer/pacing-true/same-seed: No change in performance detected. time: [39.871 ms 41.999 ms 44.216 ms]
change: [-9.1223% -2.5982% +4.2685%] (p = 0.45 > 0.05)
1-conn/1-100mb-resp (aka. Download)/client: Change within noise threshold. time: [115.27 ms 115.82 ms 116.37 ms]
thrpt: [859.32 MiB/s 863.37 MiB/s 867.50 MiB/s]
change:
time: [+0.6663% +1.2356% +1.8143%] (p = 0.00 < 0.05)
thrpt: [-1.7819% -1.2205% -0.6619%]
1-conn/10_000-parallel-1b-resp (aka. RPS)/client: No change in performance detected. time: [315.87 ms 319.52 ms 323.15 ms]
thrpt: [30.946 Kelem/s 31.297 Kelem/s 31.659 Kelem/s]
change:
time: [-1.3445% +0.1907% +1.7889%] (p = 0.82 > 0.05)
thrpt: [-1.7575% -0.1903% +1.3628%]
1-conn/1-1b-resp (aka. HPS)/client: Change within noise threshold. time: [33.728 ms 33.939 ms 34.170 ms]
thrpt: [29.266 elem/s 29.464 elem/s 29.649 elem/s]
change:
time: [-1.9671% -1.0531% -0.1525%] (p = 0.02 < 0.05)
thrpt: [+0.1528% +1.0643% +2.0066%]
Client/server transfer resultsTransfer of 33554432 bytes over loopback.
|
Failed Interop TestsQUIC Interop Runner, client vs. server neqo-latest as client
neqo-latest as server
All resultsSucceeded Interop TestsQUIC Interop Runner, client vs. server neqo-latest as client
neqo-latest as server
Unsupported Interop TestsQUIC Interop Runner, client vs. server neqo-latest as client
neqo-latest as server
|
mxinden
approved these changes
Sep 19, 2024
Comment on lines
+967
to
+968
| const TEST_RTT: Duration = ms(7000); | ||
| const TEST_RTTVAR: Duration = ms(3500); |
There was a problem hiding this comment.
Why these particular values? And why do they need to be that high?
There was a problem hiding this comment.
They came out of a discussion with @martinthomson at the Vancouver IETF
|
@martinthomson I'd appreciate a review, since the code I am touching is pretty complex. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Rather than having the caller determine for which space a PTO should be calculated for.
Also arm PTOs in the application packet number space before the handshake is complete, so 0-RTT gets RTX'ed.
Broken out of #1998