Skip to content

Conversation

mxinden
Copy link
Member

@mxinden mxinden commented Aug 21, 2025

For now, just a hack, to unblock @KershawChang on the Firefox side.

Will need more work. Probably, as @KershawChang suggested, a new function on Http3Client, as HTTP CONNECT strictly speaking isn't part of fetch.

Copy link

github-actions bot commented Aug 21, 2025

Failed Interop Tests

QUIC Interop Runner, client vs. server, differences relative to 1b16f83.

neqo-latest as client

neqo-latest as server

All results

Succeeded Interop Tests

QUIC Interop Runner, client vs. server

neqo-latest as client

neqo-latest as server

Unsupported Interop Tests

QUIC Interop Runner, client vs. server

neqo-latest as client

neqo-latest as server

Copy link

github-actions bot commented Aug 21, 2025

Client/server transfer results

Performance differences relative to 1b16f83.

Transfer of 33554432 bytes over loopback, min. 100 runs. All unit-less numbers are in milliseconds.

Client vs. server (params) Mean ± σ Min Max MiB/s ± σ Δ main Δ main
google vs. google 455.3 ± 3.7 448.2 464.8 70.3 ± 8.6
google vs. neqo (cubic, paced) 271.3 ± 4.5 262.2 280.1 117.9 ± 7.1 -0.1 -0.0%
msquic vs. msquic 130.7 ± 24.3 108.2 222.6 244.8 ± 1.3
msquic vs. neqo (cubic, paced) 141.3 ± 13.9 121.3 204.7 226.5 ± 2.3 💚 -8.6 -5.7%
neqo vs. google (cubic, paced) 758.8 ± 4.7 751.9 770.1 42.2 ± 6.8 -0.5 -0.1%
neqo vs. msquic (cubic, paced) 156.8 ± 5.1 149.6 177.5 204.1 ± 6.3 1.1 0.7%
neqo vs. neqo (cubic) 92.0 ± 4.1 85.3 105.9 347.9 ± 7.8 0.3 0.4%
neqo vs. neqo (cubic, paced) 93.6 ± 4.6 84.9 113.4 341.9 ± 7.0 💚 -1.8 -1.9%
neqo vs. neqo (reno) 90.5 ± 4.1 83.5 101.5 353.7 ± 7.8 💚 -2.1 -2.3%
neqo vs. neqo (reno, paced) 93.2 ± 5.1 85.9 113.3 343.4 ± 6.3 💚 -2.3 -2.4%
neqo vs. quiche (cubic, paced) 193.1 ± 4.6 186.8 203.6 165.7 ± 7.0 💚 -1.5 -0.8%
neqo vs. s2n (cubic, paced) 221.5 ± 4.6 213.5 232.1 144.5 ± 7.0 0.5 0.2%
quiche vs. neqo (cubic, paced) 149.8 ± 5.7 134.8 161.2 213.6 ± 5.6 1.5 1.0%
quiche vs. quiche 145.9 ± 4.8 136.7 158.6 219.3 ± 6.7
s2n vs. neqo (cubic, paced) 171.5 ± 5.2 160.5 181.4 186.6 ± 6.2 💔 1.5 0.9%
s2n vs. s2n 244.6 ± 17.3 232.1 341.1 130.8 ± 1.8

Download data for profiler.firefox.com or download performance comparison data.

Copy link

github-actions bot commented Aug 21, 2025

Benchmark results

Performance differences relative to 1b16f83.

1-conn/1-100mb-resp/mtu-1504 (aka. Download)/client: Change within noise threshold.
       time:   [200.85 ms 201.33 ms 201.88 ms]
       thrpt:  [495.35 MiB/s 496.71 MiB/s 497.89 MiB/s]
change:
       time:   [+0.0542% +0.3380% +0.6834%] (p = 0.03 < 0.05)
       thrpt:  [−0.6787% −0.3369% −0.0542%]

Found 4 outliers among 100 measurements (4.00%)
1 (1.00%) high mild
3 (3.00%) high severe

1-conn/10_000-parallel-1b-resp/mtu-1504 (aka. RPS)/client: No change in performance detected.
       time:   [306.34 ms 307.69 ms 309.01 ms]
       thrpt:  [32.362 Kelem/s 32.500 Kelem/s 32.643 Kelem/s]
change:
       time:   [−0.9177% −0.3051% +0.3120%] (p = 0.34 > 0.05)
       thrpt:  [−0.3110% +0.3060% +0.9262%]
1-conn/1-1b-resp/mtu-1504 (aka. HPS)/client: Change within noise threshold.
       time:   [28.140 ms 28.320 ms 28.521 ms]
       thrpt:  [35.062   B/s 35.311   B/s 35.536   B/s]
change:
       time:   [−0.0076% +0.8416% +1.7108%] (p = 0.04 < 0.05)
       thrpt:  [−1.6821% −0.8346% +0.0076%]

Found 10 outliers among 100 measurements (10.00%)
10 (10.00%) high severe

1-conn/1-100mb-req/mtu-1504 (aka. Upload)/client: No change in performance detected.
       time:   [203.17 ms 203.48 ms 203.81 ms]
       thrpt:  [490.65 MiB/s 491.44 MiB/s 492.20 MiB/s]
change:
       time:   [−0.3434% −0.1639% +0.0176%] (p = 0.09 > 0.05)
       thrpt:  [−0.0176% +0.1642% +0.3445%]

Found 5 outliers among 100 measurements (5.00%)
1 (1.00%) low severe
1 (1.00%) low mild
1 (1.00%) high mild
2 (2.00%) high severe

decode 4096 bytes, mask ff: No change in performance detected.
       time:   [11.618 µs 11.663 µs 11.712 µs]
       change: [−0.4536% +0.0177% +0.4704%] (p = 0.94 > 0.05)

Found 19 outliers among 100 measurements (19.00%)
2 (2.00%) low severe
5 (5.00%) low mild
12 (12.00%) high severe

decode 1048576 bytes, mask ff: No change in performance detected.
       time:   [3.0619 ms 3.0747 ms 3.0909 ms]
       change: [−0.3022% +0.2366% +0.8033%] (p = 0.42 > 0.05)

Found 9 outliers among 100 measurements (9.00%)
9 (9.00%) high severe

decode 4096 bytes, mask 7f: No change in performance detected.
       time:   [19.366 µs 19.409 µs 19.458 µs]
       change: [−0.7893% −0.1643% +0.3834%] (p = 0.60 > 0.05)

Found 19 outliers among 100 measurements (19.00%)
2 (2.00%) low severe
3 (3.00%) low mild
2 (2.00%) high mild
12 (12.00%) high severe

decode 1048576 bytes, mask 7f: No change in performance detected.
       time:   [5.0882 ms 5.1027 ms 5.1180 ms]
       change: [−0.2750% +0.1262% +0.5211%] (p = 0.53 > 0.05)

Found 18 outliers among 100 measurements (18.00%)
1 (1.00%) high mild
17 (17.00%) high severe

decode 4096 bytes, mask 3f: No change in performance detected.
       time:   [5.5153 µs 5.5251 µs 5.5453 µs]
       change: [−1.2277% −0.5689% −0.0513%] (p = 0.05 > 0.05)

Found 11 outliers among 100 measurements (11.00%)
2 (2.00%) low mild
3 (3.00%) high mild
6 (6.00%) high severe

decode 1048576 bytes, mask 3f: Change within noise threshold.
       time:   [1.7648 ms 1.7748 ms 1.7876 ms]
       change: [+0.3163% +0.9538% +1.6949%] (p = 0.00 < 0.05)

Found 13 outliers among 100 measurements (13.00%)
1 (1.00%) low mild
1 (1.00%) high mild
11 (11.00%) high severe

coalesce_acked_from_zero 1+1 entries: No change in performance detected.
       time:   [88.281 ns 88.599 ns 88.914 ns]
       change: [−0.6751% −0.1120% +0.4694%] (p = 0.70 > 0.05)

Found 14 outliers among 100 measurements (14.00%)
13 (13.00%) high mild
1 (1.00%) high severe

coalesce_acked_from_zero 3+1 entries: No change in performance detected.
       time:   [106.09 ns 106.47 ns 106.85 ns]
       change: [−1.0206% −0.2280% +0.4224%] (p = 0.58 > 0.05)

Found 15 outliers among 100 measurements (15.00%)
2 (2.00%) low mild
1 (1.00%) high mild
12 (12.00%) high severe

coalesce_acked_from_zero 10+1 entries: No change in performance detected.
       time:   [105.45 ns 105.82 ns 106.31 ns]
       change: [−0.1392% +0.3141% +0.8994%] (p = 0.25 > 0.05)

Found 17 outliers among 100 measurements (17.00%)
5 (5.00%) low severe
3 (3.00%) low mild
2 (2.00%) high mild
7 (7.00%) high severe

coalesce_acked_from_zero 1000+1 entries: No change in performance detected.
       time:   [88.650 ns 88.776 ns 88.938 ns]
       change: [−1.1353% −0.1358% +0.8069%] (p = 0.79 > 0.05)

Found 12 outliers among 100 measurements (12.00%)
6 (6.00%) high mild
6 (6.00%) high severe

RxStreamOrderer::inbound_frame(): Change within noise threshold.
       time:   [107.19 ms 107.26 ms 107.33 ms]
       change: [−0.6040% −0.4185% −0.2627%] (p = 0.00 < 0.05)

Found 26 outliers among 100 measurements (26.00%)
7 (7.00%) low severe
5 (5.00%) low mild
2 (2.00%) high mild
12 (12.00%) high severe

sent::Packets::take_ranges: :green_heart: Performance has improved.
       time:   [5.1475 µs 5.2792 µs 5.4243 µs]
       change: [−41.355% −35.662% −25.719%] (p = 0.00 < 0.05)

Found 5 outliers among 100 measurements (5.00%)
1 (1.00%) high mild
4 (4.00%) high severe

transfer/pacing-false/varying-seeds: No change in performance detected.
       time:   [36.216 ms 36.281 ms 36.348 ms]
       change: [−0.2845% −0.0354% +0.2392%] (p = 0.79 > 0.05)

Found 2 outliers among 100 measurements (2.00%)
1 (1.00%) low mild
1 (1.00%) high mild

transfer/pacing-true/varying-seeds: No change in performance detected.
       time:   [36.962 ms 37.067 ms 37.175 ms]
       change: [−0.7096% −0.2846% +0.1369%] (p = 0.18 > 0.05)

Found 1 outliers among 100 measurements (1.00%)
1 (1.00%) high mild

transfer/pacing-false/same-seed: No change in performance detected.
       time:   [36.047 ms 36.125 ms 36.214 ms]
       change: [−0.0386% +0.2365% +0.5427%] (p = 0.12 > 0.05)

Found 1 outliers among 100 measurements (1.00%)
1 (1.00%) high severe

transfer/pacing-true/same-seed: Change within noise threshold.
       time:   [37.506 ms 37.593 ms 37.687 ms]
       change: [−1.6992% −1.3453% −1.0293%] (p = 0.00 < 0.05)

Found 1 outliers among 100 measurements (1.00%)
1 (1.00%) high severe

Download data for profiler.firefox.com or download performance comparison data.

&& request
.headers
.iter()
.any(|h| h.name() == ":method" && h.value() == "CONNECT");
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think we should just check if request.method == "CONNECT"?

Copy link

codecov bot commented Aug 25, 2025

Codecov Report

❌ Patch coverage is 70.00000% with 3 lines in your changes missing coverage. Please review.
✅ Project coverage is 95.48%. Comparing base (4fa930a) to head (13bb173).
⚠️ Report is 7 commits behind head on main.

Additional details and impacted files
@@            Coverage Diff             @@
##             main    #2886      +/-   ##
==========================================
+ Coverage   94.96%   95.48%   +0.52%     
==========================================
  Files         115      115              
  Lines       34594    34442     -152     
  Branches    34594    34442     -152     
==========================================
+ Hits        32851    32886      +35     
+ Misses       1734     1547     -187     
  Partials        9        9              
Components Coverage Δ
neqo-common 97.84% <ø> (+0.35%) ⬆️
neqo-crypto 89.77% <ø> (+0.35%) ⬆️
neqo-http3 94.40% <70.00%> (+0.64%) ⬆️
neqo-qpack 95.90% <ø> (+0.44%) ⬆️
neqo-transport 96.54% <ø> (+0.49%) ⬆️
neqo-udp 91.17% <ø> (+1.32%) ⬆️

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants