Skip to content

Bun: Do not use dynamic function for pipeline handler - #1309

Merged
MDA2AV merged 2 commits into
MDA2AV:mainfrom
uNetworkingAB:patch-1
Aug 25, 2026
Merged

Bun: Do not use dynamic function for pipeline handler#1309
MDA2AV merged 2 commits into
MDA2AV:mainfrom
uNetworkingAB:patch-1

Conversation

@uNetworkingAB

@uNetworkingAB uNetworkingAB commented Aug 25, 2026

Copy link
Copy Markdown
Contributor

Description

Performance optimal fix

PR Commands — comment on this PR to trigger (requires collaborator approval):

Command Description
/benchmark -f <framework> Run every test the framework subscribes to
/benchmark -f <framework> -t <test> Run one test only
/benchmark -f <framework> --save Run and save results (updates the leaderboard on merge)
/benchmark -f <framework> -t <test> --save Run one test and save results
/benchmark -f <framework> --compare <other> Measure the deltas against another framework instead of this one
/benchmark-multiple -f <fw1>,<fw2>,... Benchmark several frameworks in one run — takes -t and --save too; saved results land in a single commit
/benchmark-multiple --save No -f needed: benchmark and save every framework the PR touches
/benchmark-test -t <test> Benchmark all enabled frameworks subscribed to <test> and save the results

For /benchmark, always specify -f <framework>; the flags combine in any order. Results come back as a comment with a per-profile table of RPS, p99, CPU and memory — one table per framework on multi runs. A new benchmark comment while a run is in flight queues behind it (one deep) instead of cancelling it. For multi-framework PRs (dependency bumps, same-language refactors) prefer /benchmark-multiple, which runs everything in a single job and commits all saved results together, so no run overwrites another. --compare works on single-framework runs only.

What the deltas are measured against. By default, this framework's own results published on main - answering "did this change help?". When you are tuning a variant or a successor entry, --compare re-bases them on another entry instead:

/benchmark -f genhttp-11 --compare genhttp-11-kestrel

The reply states which baseline it used, and profiles the other framework does not run show n/a rather than a delta.

Run benchmarks locally

You can validate and benchmark your framework locally with the lite script — no CPU pinning, fixed connection counts, all load generators run in Docker.

./scripts/validate.sh <framework>
./scripts/benchmark-lite.sh <framework> baseline
./scripts/benchmark-lite.sh --load-threads 4 <framework>

Requirements: Docker Engine on Linux. Load generators (gcannon, h2load, h2load-h3, wrk) are built as self-contained Docker images on first run.

@MDA2AV

MDA2AV commented Aug 25, 2026

Copy link
Copy Markdown
Owner

/benchmark --save

@github-actions

github-actions Bot commented Aug 25, 2026

Copy link
Copy Markdown
Contributor

👋 Benchmark request received. A collaborator will review and approve the run.

@github-actions

github-actions Bot commented Aug 25, 2026

Copy link
Copy Markdown
Contributor

Benchmark Results

Framework: bun | Test: all tests

Test Conn RPS CPU Mem Δ RPS Δ Mem
baseline 512 1,784,077 6736.9% 1.3GiB -1.2% ~0%
baseline 4096 1,918,558 6565.5% 1.2GiB -0.8% ~0%
pipelined 512 5,000,374 6523.7% 420MiB +5401.1% -23.6%
pipelined 4096 5,417,945 6619.1% 449MiB +1463.4% -48.0%
limited-conn 512 1,394,349 6315.6% 1.1GiB ~0% ~0%
limited-conn 4096 1,646,424 6578.1% 1.3GiB -0.2% +8.3%
json 4096 1,122,348 6333.5% 1.1GiB +0.2% ~0%
json-comp 512 428,765 6095.5% 1.1GiB -0.5% ~0%
json-comp 4096 464,909 6498.5% 1.1GiB +0.3% ~0%
json-comp 16384 466,079 6522.1% 1.2GiB -0.1% ~0%
json-tls 4096 1,077,848 6316.8% 1.0GiB +0.2% ~0%
upload 32 1,903 2993.9% 2.5GiB -0.5% +4.2%
upload 256 1,998 5901.9% 2.8GiB +0.5% +7.7%
api-4 256 43,441 389.3% 157MiB -0.4% +0.6%
api-16 1024 154,247 1513.3% 592MiB +1.0% -0.5%
static 1024 574,323 6531.0% 1.1GiB +0.1% ~0%
static 4096 568,905 6548.0% 1.1GiB -0.2% ~0%
static 6800 559,625 6566.0% 1.2GiB -0.2% ~0%
static-tls 1024 483,697 6570.3% 1.1GiB ~0% ~0%
static-tls 4096 472,705 6523.5% 1.1GiB ~0% ~0%
static-tls 6800 462,712 6396.5% 1.2GiB +0.3% ~0%
async-db 1024 281,632 5251.6% 2.2GiB +1.6% ~0%
crud 4096 310,978 2928.7% 1.5GiB -0.5% ~0%
fortunes 1024 82,955 6387.5% 2.0GiB -0.4% ~0%
Full log 
  Conns:     1024 (16/thread)
  Pipeline:  1
  Req/conn:  25
  Templates: 5
  Expected:  200
  Duration:  10s


  Thread Stats   Avg      p50      p90      p99    p99.9
    Latency   3.44ms   1.94ms   7.97ms   21.10ms   36.30ms

  2549256 requests in 10.00s, 2549258 responses
  Throughput: 254.87K req/s
  Bandwidth:  976.03MB/s
  Status codes: 2xx=2549258, 3xx=0, 4xx=0, 5xx=0
  Latency samples: 2549247 / 2549258 responses (100.0%)
  Reconnects: 101987
  Per-template: 508291,509671,511025,511342,508918
  Per-template-ok: 508291,509671,511025,511342,508918
[info] CPU 4825.4% | Mem 1.9GiB

=== Best: 281632 req/s (CPU: 5251.6%, Mem: 2.2GiB) ===
[info] input BW: 18.80MB/s (avg template: 70 bytes)
[info] saved results/async-db/1024/bun.json
httparena-bench-bun
httparena-bench-bun

==============================================
=== bun / crud / 4096c (tool=gcannon) ===
==============================================
[info] resetting postgres for a clean per-profile baseline
[info] starting postgres sidecar
httparena-postgres
[info] postgres ready (seeded)
[info] waiting for server...
[info] server ready

[run 1/3]
gcannon v0.5.3
  Target:    localhost:8080/
  Threads:   64
  Conns:     4096 (64/thread)
  Pipeline:  1
  Req/conn:  200
  Templates: 20
  Expected:  200
  Duration:  15s


  Thread Stats   Avg      p50      p90      p99    p99.9
    Latency   13.44ms   13.80ms   17.70ms   26.60ms   40.20ms

  4554325 requests in 15.00s, 4554005 responses
  Throughput: 303.55K req/s
  Bandwidth:  102.60MB/s
  Status codes: 2xx=4554005, 3xx=0, 4xx=0, 5xx=0
  Latency samples: 4554004 / 4554005 responses (100.0%)
  Reconnects: 21306
  Per-template: 249257,279332,265729,239454,240382,235716,206759,207290,207517,206851,206233,206231,206748,206735,206953,206734,242718,245663,244283,243419
  Per-template-ok: 249257,279332,265729,239454,240382,235716,206759,207290,207517,206851,206233,206231,206748,206735,206953,206734,242718,245663,244283,243419
[info] CPU 3014.5% | Mem 1.7GiB

[run 2/3]
gcannon v0.5.3
  Target:    localhost:8080/
  Threads:   64
  Conns:     4096 (64/thread)
  Pipeline:  1
  Req/conn:  200
  Templates: 20
  Expected:  200
  Duration:  15s


  Thread Stats   Avg      p50      p90      p99    p99.9
    Latency   13.18ms   13.60ms   17.40ms   24.60ms   41.30ms

  4628488 requests in 15.00s, 4628292 responses
  Throughput: 308.50K req/s
  Bandwidth:  103.69MB/s
  Status codes: 2xx=4628292, 3xx=0, 4xx=0, 5xx=0
  Latency samples: 4628289 / 4628292 responses (100.0%)
  Reconnects: 22202
  Per-template: 260631,282547,279956,243400,244104,244707,209559,209055,209441,209439,209778,209534,209304,208834,208448,208760,240809,247302,247307,245374
  Per-template-ok: 260631,282547,279956,243400,244104,244707,209559,209055,209441,209439,209778,209534,209304,208834,208448,208760,240809,247302,247307,245374
[info] CPU 2834.6% | Mem 1.5GiB

[run 3/3]
gcannon v0.5.3
  Target:    localhost:8080/
  Threads:   64
  Conns:     4096 (64/thread)
  Pipeline:  1
  Req/conn:  200
  Templates: 20
  Expected:  200
  Duration:  15s


  Thread Stats   Avg      p50      p90      p99    p99.9
    Latency   13.09ms   13.50ms   17.40ms   25.00ms   33.80ms

  4664938 requests in 15.00s, 4664682 responses
  Throughput: 310.93K req/s
  Bandwidth:  104.52MB/s
  Status codes: 2xx=4664682, 3xx=0, 4xx=0, 5xx=0
  Latency samples: 4664670 / 4664682 responses (100.0%)
  Reconnects: 22504
  Per-template: 271927,284089,281258,244700,244874,245278,210241,209157,209972,210026,210671,210810,210808,210640,210792,210962,243474,249480,248428,247083
  Per-template-ok: 271926,284089,281258,244700,244874,245278,210241,209157,209972,210026,210671,210810,210808,210640,210792,210962,243474,249480,248428,247083
[info] CPU 2928.7% | Mem 1.5GiB

=== Best: 310978 req/s (CPU: 2928.7%, Mem: 1.5GiB) ===
[info] input BW: 26.69MB/s (avg template: 90 bytes)
[info] saved results/crud/4096/bun.json
httparena-bench-bun
httparena-bench-bun

==============================================
=== bun / fortunes / 1024c (tool=gcannon) ===
==============================================
[info] resetting postgres for a clean per-profile baseline
[info] starting postgres sidecar
httparena-postgres
[info] postgres ready (seeded)
[info] waiting for server...
[info] server ready

[run 1/3]
gcannon v0.5.3
  Target:    localhost:8080/fortunes
  Threads:   64
  Conns:     1024 (16/thread)
  Pipeline:  1
  Req/conn:  unlimited (keep-alive)
  Expected:  200
  Duration:  5s


  Thread Stats   Avg      p50      p90      p99    p99.9
    Latency   9.14ms   8.57ms   15.00ms   23.30ms   36.90ms

  409499 requests in 5.00s, 409499 responses
  Throughput: 81.87K req/s
  Bandwidth:  1.89GB/s
  Status codes: 2xx=409499, 3xx=0, 4xx=0, 5xx=0
  Latency samples: 409499 / 409499 responses (100.0%)
[info] CPU 6282.2% | Mem 2.0GiB

[run 2/3]
timeout: the monitored command dumped core
[info] CPU 6441.4% | Mem 2.0GiB

[run 3/3]
gcannon v0.5.3
  Target:    localhost:8080/fortunes
  Threads:   64
  Conns:     1024 (16/thread)
  Pipeline:  1
  Req/conn:  unlimited (keep-alive)
  Expected:  200
  Duration:  5s


  Thread Stats   Avg      p50      p90      p99    p99.9
    Latency   8.94ms   8.48ms   14.40ms   20.90ms   26.30ms

  414777 requests in 5.00s, 414777 responses
  Throughput: 82.92K req/s
  Bandwidth:  1.91GB/s
  Status codes: 2xx=414777, 3xx=0, 4xx=0, 5xx=0
  Latency samples: 414773 / 414777 responses (100.0%)
[info] CPU 6387.5% | Mem 2.0GiB

=== Best: 82955 req/s (CPU: 6387.5%, Mem: 2.0GiB) ===
[info] saved results/fortunes/1024/bun.json
httparena-bench-bun
httparena-bench-bun
[info] skip: bun does not subscribe to baseline-h2
[info] skip: bun does not subscribe to static-h2
[info] skip: bun does not subscribe to baseline-h2c
[info] skip: bun does not subscribe to json-h2c
[info] skip: bun does not subscribe to baseline-h3
[info] skip: bun does not subscribe to static-h3
[info] skip: bun does not subscribe to gateway-64
[info] skip: bun does not subscribe to gateway-h3
[info] skip: bun does not subscribe to production-stack
[info] skip: bun does not subscribe to unary-grpc
[info] skip: bun does not subscribe to unary-grpc-tls
[info] skip: bun does not subscribe to echo-ws
[info] skip: bun does not subscribe to echo-ws-pipeline
[info] skip: bun does not subscribe to echo-ws-limited
[info] rebuilding site/data/*.json
[updated] /home/diogo/actions-runner/_work/HttpArena/HttpArena/site/data/frameworks.json
[updated] /home/diogo/actions-runner/_work/HttpArena/HttpArena/site/data/results/bun.json - 24 new, 24 total
[updated] /home/diogo/actions-runner/_work/HttpArena/HttpArena/site/data/current.json
[info] done
httparena-postgres
httparena-redis
[info] restoring loopback MTU to 65536

@MDA2AV
MDA2AV merged commit fbf3907 into MDA2AV:main Aug 25, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@MDA2AV MDA2AV MDA2AV approved these changes
@Kaliumhexacyanoferrat Kaliumhexacyanoferrat Awaiting requested review from Kaliumhexacyanoferrat Kaliumhexacyanoferrat is a code owner

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

2 participants

@uNetworkingAB @MDA2AV