feat(http3): support the HTTP DATAGRAM Capsule

[WaterWhisperer]

Related issue: #2843

[codecov[bot]]

Codecov Report

:x: Patch coverage is 95.58824% with 9 lines in your changes missing coverage. Please review. :white_check_mark: Project coverage is 93.45%. Comparing base (83e264c) to head (d3c0cab). :warning: Report is 4 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #3122      +/-   ##
==========================================
- Coverage   93.47%   93.45%   -0.02%     
==========================================
  Files         125      125              
  Lines       36663    36813     +150     
  Branches    36663    36813     +150     
==========================================
+ Hits        34270    34405     +135     
- Misses       1543     1558      +15     
  Partials      850      850              

Components	Coverage Δ
neqo-common	97.43% <ø> (ø)
neqo-crypto	83.19% <ø> (-0.49%)	:arrow_down:
neqo-http3	93.30% <95.58%> (+0.03%)	:arrow_up:
neqo-qpack	94.40% <ø> (ø)
neqo-transport	94.56% <ø> (+<0.01%)	:arrow_up:
neqo-udp	78.94% <ø> (ø)
mtu	85.44% <ø> (ø)

[github-actions[bot]]

Bencher Report

Branch	capsule
Testbed	On-prem

Click to view all benchmark results

Benchmark	Latency	Benchmark Result milliseconds (ms) (Result Δ%)	Upper Boundary milliseconds (ms) (Limit %)
google-neqo-cubic	📈 view plot 🚷 view threshold	273.25 ms (-1.49%) Baseline: 277.39 ms	286.78 ms (95.28%)

Benchmark	Latency	Benchmark Result milliseconds (ms) (Result Δ%)	Upper Boundary milliseconds (ms) (Limit %)
msquic-neqo-cubic	📈 view plot 🚷 view threshold	185.95 ms (-12.30%) Baseline: 212.03 ms	245.28 ms (75.81%)

Benchmark	Latency	Benchmark Result milliseconds (ms) (Result Δ%)	Upper Boundary milliseconds (ms) (Limit %)
neqo-google-cubic	📈 view plot 🚷 view threshold	744.24 ms (-2.60%) Baseline: 764.10 ms	789.95 ms (94.21%)

Benchmark	Latency	Benchmark Result milliseconds (ms) (Result Δ%)	Upper Boundary milliseconds (ms) (Limit %)
neqo-msquic-cubic	📈 view plot 🚷 view threshold	160.15 ms (+1.33%) Baseline: 158.05 ms	161.65 ms (99.07%)

Benchmark	Latency	Benchmark Result milliseconds (ms) (Result Δ%)	Upper Boundary milliseconds (ms) (Limit %)
neqo-neqo-cubic-nopacing	📈 view plot 🚷 view threshold	95.95 ms (-0.39%) Baseline: 96.33 ms	98.24 ms (97.68%)

Benchmark	Latency	Benchmark Result milliseconds (ms) (Result Δ%)	Upper Boundary milliseconds (ms) (Limit %)
neqo-neqo-cubic	📈 view plot 🚷 view threshold	96.85 ms (-0.52%) Baseline: 97.36 ms	99.74 ms (97.11%)

Benchmark	Latency	Benchmark Result milliseconds (ms) (Result Δ%)	Upper Boundary milliseconds (ms) (Limit %)
neqo-neqo-reno-nopacing	📈 view plot 🚷 view threshold	95.99 ms (-0.51%) Baseline: 96.48 ms	99.18 ms (96.78%)

Benchmark	Latency	Benchmark Result milliseconds (ms) (Result Δ%)	Upper Boundary milliseconds (ms) (Limit %)
neqo-neqo-reno	📈 view plot 🚷 view threshold	96.25 ms (-1.21%) Baseline: 97.43 ms	99.69 ms (96.55%)

Benchmark	Latency	Benchmark Result milliseconds (ms) (Result Δ%)	Upper Boundary milliseconds (ms) (Limit %)
neqo-quiche-cubic	📈 view plot 🚷 view threshold	190.86 ms (-0.80%) Baseline: 192.40 ms	195.22 ms (97.77%)

Benchmark	Latency	Benchmark Result milliseconds (ms) (Result Δ%)	Upper Boundary milliseconds (ms) (Limit %)
neqo-s2n-cubic	📈 view plot 🚷 view threshold	219.42 ms (-0.57%) Baseline: 220.67 ms	223.92 ms (97.99%)

Benchmark	Latency	Benchmark Result milliseconds (ms) (Result Δ%)	Upper Boundary milliseconds (ms) (Limit %)
quiche-neqo-cubic	📈 view plot 🚷 view threshold	154.36 ms (+0.33%) Baseline: 153.85 ms	157.08 ms (98.27%)

Benchmark	Latency	Benchmark Result milliseconds (ms) (Result Δ%)	Upper Boundary milliseconds (ms) (Limit %)
s2n-neqo-cubic	📈 view plot 🚷 view threshold	174.06 ms (+0.11%) Baseline: 173.86 ms	177.11 ms (98.28%)

🐰 View full continuous benchmarking report in Bencher

[github-actions[bot]]

Bencher Report

Branch	capsule
Testbed	On-prem

Click to view all benchmark results

Benchmark	Latency	Benchmark Result nanoseconds (ns) (Result Δ%)	Upper Boundary nanoseconds (ns) (Limit %)
1-conn/1-100mb-req/mtu-1504 (aka. Upload)/client	📈 view plot 🚷 view threshold	210,260,000.00 ns (+1.14%) Baseline: 207,894,821.43 ns	217,129,540.03 ns (96.84%)
1-conn/1-100mb-resp/mtu-1504 (aka. Download)/client	📈 view plot 🚷 view threshold	203,810,000.00 ns (+0.82%) Baseline: 202,152,250.00 ns	211,933,263.10 ns (96.17%)
1-conn/1-1b-resp/mtu-1504 (aka. HPS)/client	📈 view plot 🚷 view threshold	38,746,000.00 ns (+12.79%) Baseline: 34,351,228.57 ns	46,229,600.62 ns (83.81%)
1-conn/10_000-parallel-1b-resp/mtu-1504 (aka. RPS)/client	📈 view plot 🚷 view threshold	280,360,000.00 ns (-2.90%) Baseline: 288,724,839.29 ns	301,919,557.58 ns (92.86%)
1-streams/each-1000-bytes/simulated-time	📈 view plot 🚷 view threshold	119,060,000.00 ns (+0.20%) Baseline: 118,827,928.57 ns	120,495,688.27 ns (98.81%)
1-streams/each-1000-bytes/wallclock-time	📈 view plot 🚷 view threshold	586,330.00 ns (-0.58%) Baseline: 589,731.62 ns	610,112.88 ns (96.10%)
1000-streams/each-1-bytes/simulated-time	📈 view plot 🚷 view threshold	2,333,500,000.00 ns (-74.66%) Baseline: 9,207,083,571.43 ns	23,924,335,021.97 ns (9.75%)
1000-streams/each-1-bytes/wallclock-time	📈 view plot 🚷 view threshold	12,495,000.00 ns (-6.18%) Baseline: 13,318,412.50 ns	15,156,408.72 ns (82.44%)
1000-streams/each-1000-bytes/simulated-time	📈 view plot 🚷 view threshold	16,143,000,000.00 ns (-9.18%) Baseline: 17,774,350,000.00 ns	20,812,994,933.37 ns (77.56%)
1000-streams/each-1000-bytes/wallclock-time	📈 view plot 🚷 view threshold	50,677,000.00 ns (+0.33%) Baseline: 50,511,680.36 ns	55,617,933.93 ns (91.12%)
RxStreamOrderer::inbound_frame()	📈 view plot 🚷 view threshold	108,660,000.00 ns (-0.96%) Baseline: 109,712,821.43 ns	111,397,083.57 ns (97.54%)
coalesce_acked_from_zero 1+1 entries	📈 view plot 🚷 view threshold	89.53 ns (+0.44%) Baseline: 89.13 ns	90.45 ns (98.98%)
coalesce_acked_from_zero 10+1 entries	📈 view plot 🚷 view threshold	105.47 ns (-0.53%) Baseline: 106.03 ns	107.10 ns (98.48%)
coalesce_acked_from_zero 1000+1 entries	📈 view plot 🚷 view threshold	90.34 ns (-0.66%) Baseline: 90.93 ns	95.32 ns (94.77%)
coalesce_acked_from_zero 3+1 entries	📈 view plot 🚷 view threshold	106.17 ns (-0.34%) Baseline: 106.53 ns	107.52 ns (98.74%)
decode 1048576 bytes, mask 3f	📈 view plot 🚷 view threshold	1,417,800.00 ns (-19.25%) Baseline: 1,755,856.96 ns	2,544,637.71 ns (55.72%)
decode 1048576 bytes, mask 7f	📈 view plot 🚷 view threshold	1,477,400.00 ns (-69.18%) Baseline: 4,794,032.86 ns	6,760,085.24 ns (21.85%)
decode 1048576 bytes, mask ff	📈 view plot 🚷 view threshold	1,162,700.00 ns (-59.77%) Baseline: 2,890,205.71 ns	3,867,024.67 ns (30.07%)
decode 4096 bytes, mask 3f	📈 view plot 🚷 view threshold	5,554.00 ns (-23.54%) Baseline: 7,264.10 ns	11,085.54 ns (50.10%)
decode 4096 bytes, mask 7f	📈 view plot 🚷 view threshold	5,798.10 ns (-69.02%) Baseline: 18,713.87 ns	26,392.78 ns (21.97%)
decode 4096 bytes, mask ff	📈 view plot 🚷 view threshold	4,514.60 ns (-58.56%) Baseline: 10,893.13 ns	14,533.02 ns (31.06%)
sent::Packets::take_ranges	📈 view plot 🚷 view threshold	4,579.30 ns (-2.16%) Baseline: 4,680.58 ns	4,921.96 ns (93.04%)
transfer/pacing-false/same-seed/simulated-time/run	📈 view plot 🚷 view threshold	23,941,000,000.00 ns (-5.08%) Baseline: 25,221,815,412.19 ns	26,310,738,967.69 ns (90.99%)
transfer/pacing-false/same-seed/wallclock-time/run	📈 view plot 🚷 view threshold	23,853,000.00 ns (-5.57%) Baseline: 25,260,467.74 ns	27,112,489.43 ns (87.98%)
transfer/pacing-false/varying-seeds/simulated-time/run	📈 view plot 🚷 view threshold	23,941,000,000.00 ns (-4.49%) Baseline: 25,065,437,275.99 ns	25,907,874,995.80 ns (92.41%)
transfer/pacing-false/varying-seeds/wallclock-time/run	📈 view plot 🚷 view threshold	23,786,000.00 ns (-6.32%) Baseline: 25,390,232.97 ns	27,383,061.00 ns (86.86%)
transfer/pacing-true/same-seed/simulated-time/run	📈 view plot 🚷 view threshold	23,676,000,000.00 ns (-6.28%) Baseline: 25,261,526,881.72 ns	26,571,289,076.49 ns (89.10%)
transfer/pacing-true/same-seed/wallclock-time/run	📈 view plot 🚷 view threshold	24,226,000.00 ns (-8.20%) Baseline: 26,388,578.85 ns	28,923,310.30 ns (83.76%)
transfer/pacing-true/varying-seeds/simulated-time/run	📈 view plot 🚷 view threshold	23,676,000,000.00 ns (-4.81%) Baseline: 24,871,154,121.86 ns	25,767,106,382.16 ns (91.88%)
transfer/pacing-true/varying-seeds/wallclock-time/run	📈 view plot 🚷 view threshold	23,970,000.00 ns (-7.40%) Baseline: 25,884,982.08 ns	27,981,297.01 ns (85.66%)

🐰 View full continuous benchmarking report in Bencher

[mxinden]

@WaterWhisperer let me know if you need any help with the above. I saw you resolved all comments but didn't make any changes yet.

[WaterWhisperer]

@WaterWhisperer let me know if you need any help with the above. I saw you resolved all comments but didn't make any changes yet.

@mxinden Thanks for your attention! Yeah, I've resolved all the comments. However, when I was adding

an end-to-end test showing how a neqo-client and a neqo-server (neither supporting QUIC datagrams) will exchange (proxy) UDP datagrams using connect-udp and HTTP DATAGRAM capsules

, I found the test wasn't working. I then realized that my implementation is incomplete

Currently, my changes support:

• Decoding HTTP DATAGRAM Capsules from the stream (receiving side works)
• The capsule-protocol header in responses

However, the sending side still only uses QUIC datagrams. I haven't implemented sending via HTTP DATAGRAM Capsules on the stream yet. I'm trying my best on completing this to fully meet the RFC 9297 requirements.

I'll update the PR once I have a working end-to-end test. Thanks again for your follow-up!

[mxinden]

Great work. I appreciate the help!

[WaterWhisperer]

I've encountered a problem: I'm trying to send a Capsule using SendMessage::send_data(), but this method automatically wraps it in a DATA frame. The final data structure becomes [DATA frame][Capsule], and the receiving end's FrameReader will parse the DATA frame and treat the contents of the Capsule as the frame payload. My solution was to bypass the HTTP/3 framing layer and read/write raw bytes directly on the stream, manually handling Capsule encoding/decoding. However, this made the connect_udp_frame dead_code, which intuitively tells me is wrong, but I can't think of a better way.

[larseggert]

@mxinden is busy at the moment, so please don't take his silence as us having lost interest here. Will take us a few more days to respond. Thanks again for your help!

[mxinden]

@WaterWhisperer sorry for the delay here. Good catch. I didn't think of this before.

Instead of using SendMessage::send_data, how about using <SendMessage as SendStream>::send_data_atomic?

https://github.com/mozilla/neqo/blob/c2caaa235be91e98f2b1cae90e66fcec7f6d341b/neqo-http3/src/send_message.rs#L298-L306

Our WebTransport implementation has the same use-case, namely to send its custom-framed control messages. Here is how WebTransport sends a closing frame:

https://github.com/mozilla/neqo/blob/c2caaa235be91e98f2b1cae90e66fcec7f6d341b/neqo-http3/src/features/extended_connect/session.rs#L360-L363

Does that help @WaterWhisperer?

[WaterWhisperer]

@mxinden is busy at the moment, so please don't take his silence as us having lost interest here. Will take us a few more days to respond. Thanks again for your help!

@larseggert Thank you so much for your reassurance! I completely understand that everyone is busy with work and life (Just like I am busy with final exams :) ), and I really appreciate the time you and the team dedicate to maintaining this project. I've learned a lot from contributing to neqo and interacting with you all. Please take your time, and no worries at all about the delay.

[codspeed-hq[bot]]

CodSpeed Performance Report

Merging #3122 will degrade performances by 10%

_{Comparing WaterWhisperer:capsule (50cf12d) with main (617e959)}

Summary

⚡ 1 improvement
❌ 1 regression
✅ 21 untouched

:warning: Please fix the performance issues or acknowledge them on CodSpeed.

Benchmarks breakdown

	Mode	Benchmark	BASE	HEAD	Change
❌	Simulation	client	691.2 ms	768 ms	-10%
⚡	Simulation	wallclock-time	34.3 ms	32.9 ms	+4.12%

[WaterWhisperer]

Sorry for my late follow-up, I was busy with a few final exams some time ago

[github-actions[bot]]

Failed Interop Tests

QUIC Interop Runner, client vs. server, differences relative to main at 83e264cd3c6d7d2da5ef3d18cdd140175d94de78.

neqo-pr as client

neqo-pr as server

neqo-pr vs. aioquic: A :rocket:L1 neqo-pr vs. go-x-net: A BP BA neqo-pr vs. haproxy: A BP BA neqo-pr vs. kwik: :rocket:C1 BP BA neqo-pr vs. linuxquic: A :rocket:C1 neqo-pr vs. lsquic: L1 C1 neqo-pr vs. msquic: A L1 C1 neqo-pr vs. mvfst: A :warning:L1 C1 :warning:BA neqo-pr vs. nginx: A :warning:C1 BP BA neqo-pr vs. ngtcp2: A :rocket:L1 :warning:C1 CM neqo-pr vs. picoquic: A :warning:C1 neqo-pr vs. quic-go: A :rocket:C1 neqo-pr vs. quiche: A :rocket:L1 :warning:C1 BP BA neqo-pr vs. quinn: A :rocket:L1 neqo-pr vs. s2n-quic: A L1 C1 :rocket:BP BA CM neqo-pr vs. tquic: S A L1 :warning:C1 BP BA neqo-pr vs. xquic: A L1

aioquic vs. neqo-pr: CM go-x-net vs. neqo-pr: CM kwik vs. neqo-pr: BP BA CM linuxquic vs. neqo-pr: :rocket:BP :warning:BA lsquic vs. neqo-pr: :warning:L1 C1 msquic vs. neqo-pr: :rocket:C1 CM mvfst vs. neqo-pr: Z A L1 C1 CM openssl vs. neqo-pr: LR M A CM quic-go vs. neqo-pr: CM quiche vs. neqo-pr: CM quinn vs. neqo-pr: V2 CM s2n-quic vs. neqo-pr: CM tquic vs. neqo-pr: CM xquic vs. neqo-pr: M CM

All results

Succeeded Interop Tests

QUIC Interop Runner, client vs. server

neqo-pr as client

• neqo-pr vs. aioquic: H DC LR C20 M S R Z 3 B U :rocket:~~L1~~ L2 C1 C2 6 V2 BP BA
• neqo-pr vs. go-x-net: H DC LR M B U L2 C2 6
• neqo-pr vs. haproxy: H DC LR C20 M S R Z 3 B U L1 L2 C1 C2 6 V2
• neqo-pr vs. kwik: H DC LR C20 M S R Z 3 B U A L1 L2 :rocket:~~C1~~ C2 6 V2
• neqo-pr vs. linuxquic: H DC LR C20 M S R Z 3 B U E L1 L2 :rocket:~~C1~~ C2 6 V2 BP BA CM
• neqo-pr vs. lsquic: H DC LR C20 M S R Z 3 B U E A L2 C2 6 V2 BP BA CM
• neqo-pr vs. msquic: H DC LR C20 M S R Z B U L2 C2 6 V2 BP BA
• neqo-pr vs. mvfst: H DC LR M R Z 3 B U :warning:L1 L2 C2 6 BP :warning:BA
• neqo-pr vs. neqo: H DC LR C20 M S R Z 3 B U E A L1 L2 C1 C2 6 V2 BP BA CM
• neqo-pr vs. nginx: H DC LR C20 M S R Z 3 B U L1 L2 :warning:C1 C2 6
• neqo-pr vs. ngtcp2: H DC LR C20 M S R Z 3 B U E :rocket:~~L1~~ L2 :warning:C1 C2 6 V2 BP BA
• neqo-pr vs. picoquic: H DC LR C20 M S R Z 3 B U E L1 L2 :warning:C1 C2 6 V2 BP BA
• neqo-pr vs. quic-go: H DC LR C20 M S R Z 3 B U L1 L2 :rocket:~~C1~~ C2 6 BP BA
• neqo-pr vs. quiche: H DC LR C20 M S R Z 3 B U :rocket:~~L1~~ L2 :warning:C1 C2 6
• neqo-pr vs. quinn: H DC LR C20 M S R Z 3 B U E :rocket:~~L1~~ L2 C1 C2 6 BP BA
• neqo-pr vs. s2n-quic: H DC LR C20 M S R 3 B U E L2 C2 6 :rocket:~~BP~~
• neqo-pr vs. tquic: H DC LR C20 M R Z 3 B U L2 :warning:C1 C2 6
• neqo-pr vs. xquic: H DC LR C20 M S R Z 3 B U L2 C1 C2 6 BP BA

neqo-pr as server

• aioquic vs. neqo-pr: H DC LR C20 M S R Z 3 B U A L1 L2 C1 C2 6 V2 BP BA
• chrome vs. neqo-pr: 3
• go-x-net vs. neqo-pr: H DC LR M B U A L2 C2 6 BP BA
• kwik vs. neqo-pr: H DC LR C20 M S R Z 3 B U A L1 L2 C1 C2 6 V2
• linuxquic vs. neqo-pr: H DC LR C20 M S R Z 3 B U E A L1 L2 C1 C2 6 V2 :warning:BA :rocket:~~BP~~ CM
• lsquic vs. neqo-pr: H DC LR C20 M S R Z 3 B E A :warning:L1 L2 :warning:C1 C2 6 V2 BP BA CM
• msquic vs. neqo-pr: H DC LR C20 M S R Z B U A L1 L2 :rocket:~~C1~~ C2 6 V2 BP BA
• mvfst vs. neqo-pr: H DC LR M 3 B L2 C2 6 BP BA
• neqo vs. neqo-pr: H DC LR C20 M S R Z 3 B U E A L1 L2 C1 C2 6 V2 BP BA CM
• ngtcp2 vs. neqo-pr: H DC LR C20 M S R Z 3 B U E A :rocket:~~L1~~ L2 C1 C2 6 V2 BP BA CM
• openssl vs. neqo-pr: H DC C20 S R 3 B L2 C2 6 BP BA
• picoquic vs. neqo-pr: H DC LR C20 M S R Z 3 B U E A L1 L2 C1 C2 6 V2 BP BA CM
• quic-go vs. neqo-pr: H DC LR C20 M S R Z 3 B U A L1 L2 C1 C2 6 BP BA
• quiche vs. neqo-pr: H DC LR M S R Z 3 B A L1 L2 C1 C2 6 BP BA
• quinn vs. neqo-pr: H DC LR C20 M S R Z 3 B U E A L1 L2 C1 C2 6 BP BA
• s2n-quic vs. neqo-pr: H DC LR M S R 3 B E A L1 L2 C1 C2 6 BP BA
• tquic vs. neqo-pr: H DC LR M S R Z 3 B A L1 L2 C1 C2 6 BP BA
• xquic vs. neqo-pr: H DC LR C20 S R Z 3 B U A L1 L2 C1 C2 6 BP BA

Unsupported Interop Tests

QUIC Interop Runner, client vs. server

neqo-pr as client

• neqo-pr vs. aioquic: E CM
• neqo-pr vs. go-x-net: C20 S R Z 3 E L1 C1 V2 CM
• neqo-pr vs. haproxy: E CM
• neqo-pr vs. kwik: E CM
• neqo-pr vs. msquic: 3 E CM
• neqo-pr vs. mvfst: C20 S E V2 CM
• neqo-pr vs. nginx: E V2 CM
• neqo-pr vs. picoquic: CM
• neqo-pr vs. quic-go: E V2 CM
• neqo-pr vs. quiche: E V2 CM
• neqo-pr vs. quinn: V2 CM
• neqo-pr vs. s2n-quic: Z V2
• neqo-pr vs. tquic: E V2 CM
• neqo-pr vs. xquic: E V2 CM

neqo-pr as server

• aioquic vs. neqo-pr: E
• chrome vs. neqo-pr: H DC LR C20 M S R Z B U E A L1 L2 C1 C2 6 V2 BP BA CM
• go-x-net vs. neqo-pr: C20 S R Z 3 E L1 C1 V2
• kwik vs. neqo-pr: E
• lsquic vs. neqo-pr: U
• msquic vs. neqo-pr: 3 E
• mvfst vs. neqo-pr: C20 S R U E V2
• openssl vs. neqo-pr: Z U E L1 C1 V2
• quic-go vs. neqo-pr: E V2
• quiche vs. neqo-pr: C20 U E V2
• s2n-quic vs. neqo-pr: C20 Z U V2
• tquic vs. neqo-pr: C20 U E V2
• xquic vs. neqo-pr: E V2

[github-actions[bot]]

Client/server transfer results

Performance differences relative to bece9bfc90e95b164ad0ce940743a082ff7c6618.

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
neqo-s2n-cubic	219.4 ± 4.4	212.9	227.2	145.8 ± 7.3	:green_heart: -2.0	-0.9%

Table above only shows statistically significant changes. See all results below.

All results

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-google-nopacing	445.4 ± 4.6	437.5	458.7	71.8 ± 7.0
google-neqo-cubic	273.2 ± 4.6	264.8	288.9	117.1 ± 7.0	-0.6	-0.2%
msquic-msquic-nopacing	161.0 ± 34.0	135.0	404.8	198.7 ± 0.9
msquic-neqo-cubic	185.9 ± 20.9	146.8	286.4	172.1 ± 1.5	-6.5	-3.4%
neqo-google-cubic	744.2 ± 4.7	736.0	755.4	43.0 ± 6.8	0.9	0.1%
neqo-msquic-cubic	160.2 ± 4.4	151.4	168.2	199.8 ± 7.3	0.4	0.3%
neqo-neqo-cubic	96.9 ± 5.3	87.8	115.9	330.4 ± 6.0	0.2	0.3%
neqo-neqo-cubic-nopacing	96.0 ± 3.9	87.7	107.0	333.5 ± 8.2	-0.6	-0.6%
neqo-neqo-reno	96.3 ± 4.8	87.6	115.1	332.5 ± 6.7	-0.6	-0.6%
neqo-neqo-reno-nopacing	96.0 ± 4.2	86.7	104.7	333.4 ± 7.6	0.2	0.2%
neqo-quiche-cubic	190.9 ± 3.7	186.1	204.2	167.7 ± 8.6	0.6	0.3%
neqo-s2n-cubic	219.4 ± 4.4	212.9	227.2	145.8 ± 7.3	:green_heart: -2.0	-0.9%
quiche-neqo-cubic	154.4 ± 7.2	143.1	195.0	207.3 ± 4.4	0.2	0.2%
quiche-quiche-nopacing	144.3 ± 4.7	136.0	161.2	221.8 ± 6.8
s2n-neqo-cubic	174.1 ± 4.5	165.6	186.1	183.8 ± 7.1	0.4	0.2%
s2n-s2n-nopacing	248.3 ± 25.8	231.0	370.1	128.9 ± 1.2

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

[github-actions[bot]]

Benchmark results

Significant performance differences relative to bece9bfc90e95b164ad0ce940743a082ff7c6618.

1-conn/1-100mb-resp/mtu-1504 (aka. Download)/client: :green_heart: Performance has improved by -1.4772%.

       time:   [203.33 ms 203.81 ms 204.35 ms]
       thrpt:  [489.37 MiB/s 490.66 MiB/s 491.81 MiB/s]
change:
       time:   [-1.8059% -1.4772% -1.1479] (p = 0.00  +1.4993% +1.8391]
       Performance has improved.
Found 1 outliers among 100 measurements (1.00%)
1 (1.00%) high severe

1-conn/10_000-parallel-1b-resp/mtu-1504 (aka. RPS)/client: :green_heart: Performance has improved by -2.5708%.

       time:   [278.58 ms 280.36 ms 282.20 ms]
       thrpt:  [35.436 Kelem/s 35.668 Kelem/s 35.896 Kelem/s]
change:
       time:   [-3.3846% -2.5708% -1.6624] (p = 0.00  +2.6386% +3.5032]
       Performance has improved.
Found 2 outliers among 100 measurements (2.00%)
2 (2.00%) high mild

All results

1-conn/1-100mb-resp/mtu-1504 (aka. Download)/client: :green_heart: Performance has improved by -1.4772%.

       time:   [203.33 ms 203.81 ms 204.35 ms]
       thrpt:  [489.37 MiB/s 490.66 MiB/s 491.81 MiB/s]
change:
       time:   [-1.8059% -1.4772% -1.1479] (p = 0.00  +1.4993% +1.8391]
       Performance has improved.
Found 1 outliers among 100 measurements (1.00%)
1 (1.00%) high severe

1-conn/10_000-parallel-1b-resp/mtu-1504 (aka. RPS)/client: :green_heart: Performance has improved by -2.5708%.

       time:   [278.58 ms 280.36 ms 282.20 ms]
       thrpt:  [35.436 Kelem/s 35.668 Kelem/s 35.896 Kelem/s]
change:
       time:   [-3.3846% -2.5708% -1.6624] (p = 0.00  +2.6386% +3.5032]
       Performance has improved.
Found 2 outliers among 100 measurements (2.00%)
2 (2.00%) high mild

1-conn/1-1b-resp/mtu-1504 (aka. HPS)/client: No change in performance detected.

       time:   [38.574 ms 38.746 ms 38.939 ms]
       thrpt:  [25.681   B/s 25.809   B/s 25.924   B/s]
change:
       time:   [-0.6406% -0.0002% +0.6609] (p = 1.00 > 0.05)
       thrpt:  [-0.6565% +0.0002% +0.6448]
       No change in performance detected.
Found 9 outliers among 100 measurements (9.00%)
1 (1.00%) low mild
1 (1.00%) high mild
7 (7.00%) high severe

1-conn/1-100mb-req/mtu-1504 (aka. Upload)/client: No change in performance detected.

       time:   [209.95 ms 210.26 ms 210.63 ms]
       thrpt:  [474.76 MiB/s 475.60 MiB/s 476.31 MiB/s]
change:
       time:   [-0.3872% -0.1838% +0.0471] (p = 0.09 > 0.05)
       thrpt:  [-0.0471% +0.1841% +0.3887]
       No change in performance detected.
Found 1 outliers among 100 measurements (1.00%)
1 (1.00%) high severe

decode 4096 bytes, mask ff: No change in performance detected.

       time:   [4.5063 µs 4.5146 µs 4.5242 µs]
       change: [-0.4197% -0.0619% +0.3612] (p = 0.77 > 0.05)
       No change in performance detected.
Found 7 outliers among 100 measurements (7.00%)
4 (4.00%) high mild
3 (3.00%) high severe

decode 1048576 bytes, mask ff: No change in performance detected.

       time:   [1.1611 ms 1.1627 ms 1.1644 ms]
       change: [-0.6179% +0.2720% +1.1887] (p = 0.55 > 0.05)
       No change in performance detected.
Found 11 outliers among 100 measurements (11.00%)
1 (1.00%) high mild
10 (10.00%) high severe

decode 4096 bytes, mask 7f: No change in performance detected.

       time:   [5.7900 µs 5.7981 µs 5.8062 µs]
       change: [-0.2691% -0.0104% +0.2541] (p = 0.93 > 0.05)
       No change in performance detected.
Found 4 outliers among 100 measurements (4.00%)
3 (3.00%) high mild
1 (1.00%) high severe

decode 1048576 bytes, mask 7f: Change within noise threshold.

       time:   [1.4750 ms 1.4774 ms 1.4799 ms]
       change: [-0.8964% -0.6496% -0.3984] (p = 0.00 

decode 4096 bytes, mask 3f: No change in performance detected.

       time:   [5.5459 µs 5.5540 µs 5.5622 µs]
       change: [-0.3176% +0.1517% +0.6333] (p = 0.57 > 0.05)
       No change in performance detected.
Found 6 outliers among 100 measurements (6.00%)
4 (4.00%) high mild
2 (2.00%) high severe

decode 1048576 bytes, mask 3f: No change in performance detected.

       time:   [1.4157 ms 1.4178 ms 1.4200 ms]
       change: [-0.5450% -0.1730% +0.1440] (p = 0.34 > 0.05)
       No change in performance detected.

1-streams/each-1000-bytes/wallclock-time: No change in performance detected.

       time:   [583.43 µs 586.33 µs 591.14 µs]
       change: [-0.8030% -0.1232% +0.7874] (p = 0.79 > 0.05)
       No change in performance detected.
Found 3 outliers among 100 measurements (3.00%)
3 (3.00%) high severe

1-streams/each-1000-bytes/simulated-time: No change in performance detected.

       time:   [118.84 ms 119.06 ms 119.29 ms]
       thrpt:  [8.1868 KiB/s 8.2021 KiB/s 8.2174 KiB/s]
change:
       time:   [-0.3842% -0.1054% +0.1788] (p = 0.46 > 0.05)
       thrpt:  [-0.1785% +0.1055% +0.3857]
       No change in performance detected.

1000-streams/each-1-bytes/wallclock-time: Change within noise threshold.

       time:   [12.455 ms 12.495 ms 12.535 ms]
       change: [+0.5750% +1.0337% +1.5138] (p = 0.00 

1000-streams/each-1-bytes/simulated-time: No change in performance detected.

       time:   [2.3300 s 2.3335 s 2.3371 s]
       thrpt:  [427.89   B/s 428.54   B/s 429.19   B/s]
change:
       time:   [-0.1636% +0.0439% +0.2554] (p = 0.68 > 0.05)
       thrpt:  [-0.2548% -0.0439% +0.1639]
       No change in performance detected.

1000-streams/each-1000-bytes/wallclock-time: No change in performance detected.

       time:   [50.457 ms 50.677 ms 51.007 ms]
       change: [-0.3663% +0.1543% +0.8349] (p = 0.67 > 0.05)
       No change in performance detected.
Found 2 outliers among 100 measurements (2.00%)
1 (1.00%) high mild
1 (1.00%) high severe

1000-streams/each-1000-bytes/simulated-time: No change in performance detected.

       time:   [15.904 s 16.143 s 16.385 s]
       thrpt:  [59.601 KiB/s 60.494 KiB/s 61.404 KiB/s]
change:
       time:   [-2.9559% -0.9114% +1.2738] (p = 0.40 > 0.05)
       thrpt:  [-1.2578% +0.9198% +3.0459]
       No change in performance detected.

coalesce_acked_from_zero 1+1 entries: No change in performance detected.

       time:   [89.202 ns 89.529 ns 89.869 ns]
       change: [-0.5977% +0.5766% +2.7139] (p = 0.59 > 0.05)
       No change in performance detected.
Found 10 outliers among 100 measurements (10.00%)
7 (7.00%) high mild
3 (3.00%) high severe

coalesce_acked_from_zero 3+1 entries: No change in performance detected.

       time:   [105.83 ns 106.17 ns 106.52 ns]
       change: [-0.4414% +0.0139% +0.5065] (p = 0.96 > 0.05)
       No change in performance detected.
Found 15 outliers among 100 measurements (15.00%)
1 (1.00%) low mild
1 (1.00%) high mild
13 (13.00%) high severe

coalesce_acked_from_zero 10+1 entries: No change in performance detected.

       time:   [105.14 ns 105.47 ns 105.92 ns]
       change: [-0.6691% -0.2361% +0.2126] (p = 0.30 > 0.05)
       No change in performance detected.
Found 12 outliers among 100 measurements (12.00%)
2 (2.00%) low severe
3 (3.00%) low mild
2 (2.00%) high mild
5 (5.00%) high severe

coalesce_acked_from_zero 1000+1 entries: No change in performance detected.

       time:   [90.216 ns 90.336 ns 90.471 ns]
       change: [+0.0041% +0.4465% +0.9284] (p = 0.05 > 0.05)
       No change in performance detected.
Found 10 outliers among 100 measurements (10.00%)
6 (6.00%) high mild
4 (4.00%) high severe

RxStreamOrderer::inbound_frame(): No change in performance detected.

       time:   [108.49 ms 108.66 ms 108.96 ms]
       change: [-0.3594% -0.0422% +0.2915] (p = 0.83 > 0.05)
       No change in performance detected.
Found 10 outliers among 100 measurements (10.00%)
9 (9.00%) low mild
1 (1.00%) high severe

sent::Packets::take_ranges: No change in performance detected.

       time:   [4.4552 µs 4.5793 µs 4.7100 µs]
       change: [-4.6729% -1.4920% +1.9998] (p = 0.39 > 0.05)
       No change in performance detected.
Found 2 outliers among 100 measurements (2.00%)
2 (2.00%) high mild

transfer/pacing-false/varying-seeds/wallclock-time/run: Change within noise threshold.

       time:   [23.765 ms 23.786 ms 23.810 ms]
       change: [-1.1466% -1.0419% -0.9278] (p = 0.00 

transfer/pacing-false/varying-seeds/simulated-time/run: No change in performance detected.

       time:   [23.941 s 23.941 s 23.941 s]
       thrpt:  [171.09 KiB/s 171.09 KiB/s 171.09 KiB/s]
change:
       time:   [+0.0000% +0.0000% +0.0000] (p = NaN > 0.05)
       thrpt:  [+0.0000% +0.0000% +0.0000]
       No change in performance detected.

transfer/pacing-true/varying-seeds/wallclock-time/run: Change within noise threshold.

       time:   [23.938 ms 23.970 ms 24.018 ms]
       change: [-1.2660% -1.0953% -0.8629] (p = 0.00 

transfer/pacing-true/varying-seeds/simulated-time/run: No change in performance detected.

       time:   [23.676 s 23.676 s 23.676 s]
       thrpt:  [173.01 KiB/s 173.01 KiB/s 173.01 KiB/s]
change:
       time:   [+0.0000% +0.0000% +0.0000] (p = NaN > 0.05)
       thrpt:  [+0.0000% +0.0000% +0.0000]
       No change in performance detected.

transfer/pacing-false/same-seed/wallclock-time/run: No change in performance detected.

       time:   [23.838 ms 23.853 ms 23.869 ms]
       change: [-0.2549% -0.0375% +0.1164] (p = 0.75 > 0.05)
       No change in performance detected.
Found 1 outliers among 100 measurements (1.00%)
1 (1.00%) high mild

transfer/pacing-false/same-seed/simulated-time/run: No change in performance detected.

       time:   [23.941 s 23.941 s 23.941 s]
       thrpt:  [171.09 KiB/s 171.09 KiB/s 171.09 KiB/s]
change:
       time:   [+0.0000% +0.0000% +0.0000] (p = NaN > 0.05)
       thrpt:  [+0.0000% +0.0000% +0.0000]
       No change in performance detected.

transfer/pacing-true/same-seed/wallclock-time/run: Change within noise threshold.

       time:   [24.206 ms 24.226 ms 24.247 ms]
       change: [-0.2936% -0.1788% -0.0654] (p = 0.00 

transfer/pacing-true/same-seed/simulated-time/run: No change in performance detected.

       time:   [23.676 s 23.676 s 23.676 s]
       thrpt:  [173.01 KiB/s 173.01 KiB/s 173.01 KiB/s]
change:
       time:   [+0.0000% +0.0000% +0.0000] (p = NaN > 0.05)
       thrpt:  [+0.0000% +0.0000% +0.0000]
       No change in performance detected.

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