test: Add my test.sh script

[larseggert]

Because I keep blowing it away when I git clean...

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 93.13%. Comparing base (7feb7cb) to head (27510eb).

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #1802   +/-   ##
=======================================
  Coverage   93.13%   93.13%           
=======================================
  Files         117      117           
  Lines       36363    36363           
=======================================
  Hits        33865    33865           
  Misses       2498     2498           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[github-actions]

Benchmark results

Performance differences relative to 32a2a59.

• drain a timer quickly time: [370.21 ns 377.53 ns 384.29 ns]
  change: [+1.8868% +3.8179% +5.8190%] (p = 0.00 < 0.05)
  💔 Performance has regressed.

• coalesce_acked_from_zero 1+1 entries
  time: [195.58 ns 196.05 ns 196.58 ns]
  change: [-0.0307% +1.0166% +3.0706%] (p = 0.19 > 0.05)
  No change in performance detected.

• coalesce_acked_from_zero 3+1 entries
  time: [240.45 ns 241.08 ns 241.77 ns]
  change: [+1.7124% +2.0380% +2.3381%] (p = 0.00 < 0.05)
  💔 Performance has regressed.

• coalesce_acked_from_zero 10+1 entries
  time: [239.31 ns 239.71 ns 240.27 ns]
  change: [+1.4506% +1.9013% +2.2910%] (p = 0.00 < 0.05)
  💔 Performance has regressed.

• coalesce_acked_from_zero 1000+1 entries
  time: [218.74 ns 218.97 ns 219.21 ns]
  change: [-0.5803% +0.1524% +0.9683%] (p = 0.71 > 0.05)
  No change in performance detected.

• RxStreamOrderer::inbound_frame()
  time: [118.72 ms 118.88 ms 119.11 ms]
  change: [-0.4504% -0.2931% -0.0893%] (p = 0.00 < 0.05)
  Change within noise threshold.

• transfer/Run multiple transfers with varying seeds
  time: [119.92 ms 120.19 ms 120.47 ms]
  thrpt: [33.204 MiB/s 33.281 MiB/s 33.357 MiB/s]
  change:
  time: [-2.0916% -1.7922% -1.4801%] (p = 0.00 < 0.05)
  thrpt: [+1.5023% +1.8250% +2.1363%]
  Change within noise threshold.

• transfer/Run multiple transfers with the same seed
  time: [120.34 ms 120.53 ms 120.74 ms]
  thrpt: [33.130 MiB/s 33.187 MiB/s 33.239 MiB/s]
  change:
  time: [-2.1646% -1.9640% -1.7462%] (p = 0.00 < 0.05)
  thrpt: [+1.7772% +2.0033% +2.2125%]
  Change within noise threshold.

• 1-conn/1-100mb-resp (aka. Download)/client
  time: [1.0981 s 1.1142 s 1.1318 s]
  thrpt: [88.357 MiB/s 89.747 MiB/s 91.070 MiB/s]
  change:
  time: [-1.4107% +1.1562% +3.7349%] (p = 0.40 > 0.05)
  thrpt: [-3.6004% -1.1430% +1.4309%]
  No change in performance detected.

• 1-conn/10_000-parallel-1b-resp (aka. RPS)/client
  time: [426.25 ms 428.07 ms 429.93 ms]
  thrpt: [23.260 Kelem/s 23.361 Kelem/s 23.461 Kelem/s]
  change:
  time: [+9.4472% +10.321% +11.222%] (p = 0.00 < 0.05)
  thrpt: [-10.090% -9.3556% -8.6317%]
  💔 Performance has regressed.

• 1-conn/1-1b-resp (aka. HPS)/client
  time: [50.638 ms 51.025 ms 51.411 ms]
  thrpt: [19.451 elem/s 19.598 elem/s 19.748 elem/s]
  change:
  time: [+18.859% +19.790% +20.878%] (p = 0.00 < 0.05)
  thrpt: [-17.272% -16.521% -15.866%]
  💔 Performance has regressed.

Client/server transfer results

Transfer of 134217728 bytes over loopback.

Client	Server	CC	Pacing	Mean [ms]	Min [ms]	Max [ms]	Relative
msquic	msquic			605.0 ± 154.1	415.2	927.9	1.00
neqo	msquic	reno	on	950.7 ± 182.3	809.3	1384.4	1.00
neqo	msquic	reno		980.2 ± 137.9	787.9	1213.5	1.00
neqo	msquic	cubic	on	1035.1 ± 277.7	767.2	1455.9	1.00
neqo	msquic	cubic		976.4 ± 223.1	766.0	1407.1	1.00
msquic	neqo	reno	on	4518.1 ± 282.6	4156.6	5030.7	1.00
msquic	neqo	reno		4560.7 ± 244.8	4269.9	5040.0	1.00
msquic	neqo	cubic	on	4439.5 ± 154.2	4174.0	4630.9	1.00
msquic	neqo	cubic		4514.4 ± 263.2	4162.4	5004.0	1.00
neqo	neqo	reno	on	3783.7 ± 304.0	3169.7	4189.2	1.00
neqo	neqo	reno		3932.3 ± 235.0	3621.8	4402.5	1.00
neqo	neqo	cubic	on	4370.1 ± 239.0	4060.6	4937.7	1.00
neqo	neqo	cubic		4388.1 ± 356.2	3569.9	4744.9	1.00

⬇️ Download logs

[martinthomson]

You should probably use trap 'rm -rf "$tmp"' EXIT instead. Especially with set -e.

[larseggert]

Thanks! #1810