-
Notifications
You must be signed in to change notification settings - Fork 123
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
fix: Allow turning off greasing and connection migration #2100
Conversation
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
|
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This makes sense to me.
Though given that today is the first day I fully read RFC 9287, I think someone else should take a look as well. Maybe the author of RFC 9287? 😇
Do we have a test testing the client-server exchange of the grease transport parameter? I can't find one. If it doesn't exist, I am happy to add one. No need to block this pull request.
Co-authored-by: Max Inden <mail@max-inden.de> Signed-off-by: Lars Eggert <lars@eggert.org>
Co-authored-by: Max Inden <mail@max-inden.de> Signed-off-by: Lars Eggert <lars@eggert.org>
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## main #2100 +/- ##
==========================================
+ Coverage 95.33% 95.34% +0.01%
==========================================
Files 112 112
Lines 36300 36309 +9
==========================================
+ Hits 34607 34620 +13
+ Misses 1693 1689 -4 ☔ View full report in Codecov by Sentry. |
Benchmark resultsPerformance differences relative to 0b106c2. coalesce_acked_from_zero 1+1 entries: No change in performance detected.time: [99.400 ns 99.716 ns 100.03 ns] change: [-0.3266% +0.2637% +0.7937%] (p = 0.38 > 0.05) coalesce_acked_from_zero 3+1 entries: No change in performance detected.time: [117.33 ns 117.69 ns 118.10 ns] change: [-0.1049% +0.3050% +0.6939%] (p = 0.13 > 0.05) coalesce_acked_from_zero 10+1 entries: Change within noise threshold.time: [117.12 ns 117.60 ns 118.17 ns] change: [+0.0649% +0.6486% +1.2226%] (p = 0.03 < 0.05) coalesce_acked_from_zero 1000+1 entries: No change in performance detected.time: [97.338 ns 97.507 ns 97.699 ns] change: [-1.1985% -0.3415% +0.4624%] (p = 0.44 > 0.05) RxStreamOrderer::inbound_frame(): Change within noise threshold.time: [111.60 ms 111.75 ms 111.98 ms] change: [+0.1987% +0.3359% +0.5552%] (p = 0.00 < 0.05) transfer/pacing-false/varying-seeds: No change in performance detected.time: [26.119 ms 27.175 ms 28.218 ms] change: [-7.8299% -3.0786% +2.1893%] (p = 0.25 > 0.05) transfer/pacing-true/varying-seeds: No change in performance detected.time: [35.131 ms 36.667 ms 38.230 ms] change: [-4.6677% +1.3907% +7.4441%] (p = 0.66 > 0.05) transfer/pacing-false/same-seed: No change in performance detected.time: [31.825 ms 32.615 ms 33.387 ms] change: [-3.5836% -0.3916% +2.8532%] (p = 0.81 > 0.05) transfer/pacing-true/same-seed: No change in performance detected.time: [41.513 ms 44.123 ms 46.690 ms] change: [-11.743% -2.9970% +6.4575%] (p = 0.52 > 0.05) 1-conn/1-100mb-resp (aka. Download)/client: No change in performance detected.time: [113.99 ms 114.95 ms 116.17 ms] thrpt: [860.79 MiB/s 869.94 MiB/s 877.28 MiB/s] change: time: [-0.9942% -0.0895% +0.9690%] (p = 0.88 > 0.05) thrpt: [-0.9597% +0.0896% +1.0042%] 1-conn/10_000-parallel-1b-resp (aka. RPS)/client: No change in performance detected.time: [314.28 ms 318.10 ms 321.86 ms] thrpt: [31.069 Kelem/s 31.437 Kelem/s 31.819 Kelem/s] change: time: [-1.8467% -0.1396% +1.6065%] (p = 0.87 > 0.05) thrpt: [-1.5811% +0.1398% +1.8815%] 1-conn/1-1b-resp (aka. HPS)/client: No change in performance detected.time: [33.734 ms 33.908 ms 34.097 ms] thrpt: [29.328 elem/s 29.491 elem/s 29.644 elem/s] change: time: [-1.5113% -0.6923% +0.1412%] (p = 0.11 > 0.05) thrpt: [-0.1410% +0.6971% +1.5344%] Client/server transfer resultsTransfer of 33554432 bytes over loopback.
|
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This looks fine.
One thing I noticed is that we have a public method called "migrate" on the connection. This method can be called when the server has set disable_active_migration
. That's not ideal. That's worth another change, I think.
Ah, sorry, merged |
There was some plumbing missing. CC @mxinden