Backend: Server-Sent Events for handlers by shai-almog · Pull Request #6218 · codenameone/CodenameOne

shai-almog · 2026-11-11T18:16:27Z

Handlers can now answer with an event stream, for LLM-proxy and live-feed endpoints where a WebSocket is more than the job needs:

return Sse.stream(request, events -> {
    for (String token : model.tokens(prompt)) events.data(token);   // on the wire now
    events.event("done").data("{}");
});

What it is

Sse (vm/backend/src/com/codename1/backend/Sse.java) builds on the existing Response.stream, which already does chunked HTTP/1.1, close-delimited HTTP/1.0 and flow-controlled HTTP/2. It adds the event-stream framing:
- data, json, event, id, retry, comment and ping(), each flushed as soon as it is written.
- lastEventId(), so a reconnecting client resumes where it left off.
- Cache-Control: no-cache and X-Accel-Buffering: no, so nginx does not hold events back.
- Multi-line data is split on CR, LF and CRLF. A line break in event or id, or a NUL in id, is refused, so a value cannot forge an extra field.
It uses the blocking-producer model: the producer runs on the request's worker until it returns. A cross-thread SseEmitter is out of scope, because the hand-off would need Object.wait on a virtual thread, which blocks the host.

Fixes found on the way

Compression: with compression on, a zero minimum and a matching type, a streamed response could be sent with Content-Encoding: gzip over identity bytes. Streamed responses are never compressed now.
HttpProxy: it relayed an upstream event stream only when its 16 KB buffer filled, so a proxied token feed reached the client minutes late. It now flushes after every read for text/event-stream and for bodies of undeclared length. Downloads with a declared length keep sending full 16 KB chunks.

Tests

SseTest covers:

the framing and the refused field values.
real sockets, direct and through HttpProxy: the producer does not write its second event until the client has read the first, so any buffering times the test out.
Last-Event-ID passed through the proxy.
the compression guard.

Docs

A "Server-Sent Events" subsection in Backend-Web.asciidoc, with a live-feed demo in Streaming.java.

🤖 Generated with Claude Code

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

…ared lengths available() is always 0 on both Web impls, so the previous guard flushed every read of every download. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

github-actions · 2026-11-11T19:19:03Z

Developer Guide build artifacts are available for download from this workflow run:

Developer Guide quality checks:

AsciiDoc linter: No issues found (report)
Vale: No alerts found (report)
Paragraph capitalization: No paragraph capitalization issues (report)
LanguageTool: No grammar matches (report)
Image references: No unused images detected (report)

github-actions · 2026-11-11T20:06:58Z

Cloudflare Preview

URL: https://pr-6218-website-preview.codenameone.pages.dev
Branch: pr-6218-website-preview

github-actions · 2026-11-11T20:56:33Z

✅ Continuous Quality Report

Test & Coverage

✅ Tests: 15194 total, 0 failed, 170 skipped
📊 Line coverage: 65.71% [HTML preview] [Download]
- Lowest covered classes
  - com.codename1.io.zip.ZipArchiveOutputStream – 0.00%
  - com.codename1.gaming.level.GameSceneView – 0.00%
  - com.codename1.security.NativeCryptoEngine – 0.00%
  - com.codename1.crash.CrashProtection – 0.00%
  - com.codename1.mapping.JsonValues – 0.00%
  - com.codename1.payment.CommerceManager – 0.00%
  - com.codename1.io.outbox.Outbox – 0.00%
  - com.codename1.ai.LocalLlmClient – 0.00%
  - com.codename1.security.CertificateRequest – 0.00%
  - com.codename1.wearable.WearableMessage – 0.00%

Static Analysis

SpotBugs [Report archive]
- ✅ ByteCodeTranslator: 0 findings (no issues)
- ✅ android: 0 findings (no issues)
- ✅ android-compat: 0 findings (no issues)
- ✅ android-res-compiler: 0 findings (no issues)
- ✅ backend: 0 findings (no issues)
- ✅ backend-test: 0 findings (no issues)
- ✅ build-engine: 0 findings (no issues)
- ✅ build-hint-catalog: 0 findings (no issues)
- ✅ build-hint-tools: 0 findings (no issues)
- ✅ codenameone-gradle-plugin: 0 findings (no issues)
- ✅ codenameone-maven-plugin: 0 findings (no issues)
- ✅ compose-compat: 0 findings (no issues)
- ✅ core-unittests: 0 findings (no issues)
- ✅ desktop-compat-runtime: 0 findings (no issues)
- ✅ fxml-compiler: 0 findings (no issues)
- ✅ ios: 0 findings (no issues)
- ✅ javac: 0 findings (no issues)
- ✅ javafx-compat: 0 findings (no issues)
- ✅ project-model: 0 findings (no issues)
- ✅ python-runtime: 0 findings (no issues)
- ✅ python-transpiler: 0 findings (no issues)
- ✅ swing-compat: 0 findings (no issues)
✅ PMD: 0 findings (no issues) [Report archive]
✅ Checkstyle: 0 findings (no issues) [Report archive]

Generated automatically by the PR CI workflow.

shai-almog · 2026-11-11T21:44:42Z

Compared 210 screenshots: 210 matched.
Native Windows port (x64 / Intel-AMD): full hellocodenameone screenshot suite rendered offscreen with Direct2D/DirectWrite, plus the real benchmarks (base64 native/CN1/SIMD, image createMask/applyMask/modifyAlpha/PNG/JPEG, SSE2 SIMD kernels). Compared against the in-repo baseline in scripts/windows/screenshots.

Benchmark Results

Detailed Performance Metrics

Metric	Duration
SIMD kernel backend	SSE2 (x64) / NEON (arm64) native kernels
SIMD int-add (64K x300)	java 26ms / native 3ms = 8.6x speedup
SIMD float-mul (64K x300)	java 26ms / native 2ms = 13.0x speedup
SIMD kernel correctness	PASS (native result == scalar reference)
Base64 native bridge	unavailable (CN1 + SIMD + image benchmarks only)
Base64 payload size	8192 bytes
Base64 benchmark iterations	6000
Base64 SIMD byte path	gated to scalar (CPU autovectorizes scalar; explicit SIMD not beneficial here)
Base64 CN1 encode	39.000 ms
Base64 CN1 decode	48.000 ms
Base64 SIMD encode	50.000 ms
Base64 encode ratio (SIMD/CN1)	1.282x (28.2% slower)
Base64 SIMD decode	50.000 ms
Base64 decode ratio (SIMD/CN1)	1.042x (4.2% slower)
Image encode benchmark iterations	100
Image createMask (SIMD off)	4.000 ms
Image createMask (SIMD on)	4.000 ms
Image createMask ratio (SIMD on/off)	1.000x (0.0% slower)
Image applyMask (SIMD off)	19.000 ms
Image applyMask (SIMD on)	10.000 ms
Image applyMask ratio (SIMD on/off)	0.526x (47.4% faster)
Image modifyAlpha (SIMD off)	16.000 ms
Image modifyAlpha (SIMD on)	11.000 ms
Image modifyAlpha ratio (SIMD on/off)	0.688x (31.3% faster)
Image modifyAlpha removeColor (SIMD off)	17.000 ms
Image modifyAlpha removeColor (SIMD on)	7.000 ms
Image modifyAlpha removeColor ratio (SIMD on/off)	0.412x (58.8% faster)

ParparVM vs HotSpot (JDK 25): Windows x64

Runner CPU: AMD64 Family 26 Model 2 Stepping 1, AuthenticAMD (baseline windows-x64@amd64-family-26-model-2-authenticamd)

Ratios are ParparVM / JDK 25: below 1.00x ParparVM is faster (time) or smaller (RAM). Median of 5 interleaved, paired rounds; every run's output was verified. A ratio more than 20% (time) / 25% (RAM) above its baseline in vm/selfhost/perf-baseline/ is a regression and fails; one that far below is an improvement, reported but not failed (a row whose calibration runs were noisier carries a wider tolerance), and for RAM the change must also exceed 0.05x in absolute terms. Both run unpinned on all of the runner's CPUs, with their own default thread counts.

Benchmark	Cores	Time	RAM	Status
hello (WinHelloMain)	4	1.99x (base 1.41x, +41.1%)	0.67x (base 0.68x, -1.9%)	ok
translator (self)	4	0.90x (base 0.90x, -0.0%)	0.50x (base 0.47x, +7.2%)	ok
intArithmetic	4	1.00x (base 0.99x, +0.6%)	0.06x (base 0.05x, +2.8%)	ok
longArithmetic	4	0.99x (base 0.99x, +0.5%)	0.06x (base 0.05x, +3.1%)	ok
mathTranscendental	4	0.93x (base 0.94x, -0.8%)	0.06x (base 0.06x, +2.7%)	ok
arraySequential	4	1.77x (base 1.75x, +1.5%)	0.38x (base 0.38x, +0.2%)	ok
arrayRandom	4	1.02x (base 1.00x, +1.7%)	0.23x (base 0.23x, -0.4%)	ok
objectAllocation	4	1.54x (base 1.65x, -6.2%)	0.26x (base 0.21x, +22.9%)	ok
valueEscape	4	0.10x (base 0.10x, -0.1%)	0.05x (base 0.04x, +2.9%)	ok
hashMapChurn	4	1.59x (base 1.59x, -0.2%)	0.09x (base 0.09x, -0.5%)	ok
stringBuilding	4	1.42x (base 1.39x, +2.1%)	0.32x (base 0.27x, +20.9%)	ok
recursion	4	1.42x (base 1.44x, -1.6%)	0.06x (base 0.06x, +3.2%)	ok
quicksort	4	1.06x (base 1.06x, +0.5%)	0.12x (base 0.12x, +1.6%)	ok

Result: no regression

shai-almog · 2026-11-11T22:29:04Z

Compared 210 screenshots: 210 matched.
Native Windows port (arm64 / Apple Silicon - Arm): full hellocodenameone screenshot suite rendered offscreen with Direct2D/DirectWrite, plus the real benchmarks (base64 native/CN1/SIMD, image createMask/applyMask/modifyAlpha/PNG/JPEG, NEON SIMD kernels). Compared against the in-repo baseline in scripts/windows/screenshots.

Benchmark Results

Detailed Performance Metrics

Metric	Duration
SIMD kernel backend	SSE2 (x64) / NEON (arm64) native kernels
SIMD int-add (64K x300)	java 53ms / native 3ms = 17.6x speedup
SIMD float-mul (64K x300)	java 54ms / native 2ms = 27.0x speedup
SIMD kernel correctness	PASS (native result == scalar reference)
Base64 native bridge	unavailable (CN1 + SIMD + image benchmarks only)
Base64 payload size	8192 bytes
Base64 benchmark iterations	6000
Base64 SIMD byte path	gated to scalar (CPU autovectorizes scalar; explicit SIMD not beneficial here)
Base64 CN1 encode	76.000 ms
Base64 CN1 decode	82.000 ms
Base64 SIMD encode	91.000 ms
Base64 encode ratio (SIMD/CN1)	1.197x (19.7% slower)
Base64 SIMD decode	74.000 ms
Base64 decode ratio (SIMD/CN1)	0.902x (9.8% faster)
Image encode benchmark iterations	100
Image createMask (SIMD off)	6.000 ms
Image createMask (SIMD on)	1.000 ms
Image createMask ratio (SIMD on/off)	0.167x (83.3% faster)
Image applyMask (SIMD off)	15.000 ms
Image applyMask (SIMD on)	15.000 ms
Image applyMask ratio (SIMD on/off)	1.000x (0.0% slower)
Image modifyAlpha (SIMD off)	15.000 ms
Image modifyAlpha (SIMD on)	10.000 ms
Image modifyAlpha ratio (SIMD on/off)	0.667x (33.3% faster)
Image modifyAlpha removeColor (SIMD off)	20.000 ms
Image modifyAlpha removeColor (SIMD on)	9.000 ms
Image modifyAlpha removeColor ratio (SIMD on/off)	0.450x (55.0% faster)

ParparVM vs HotSpot (JDK 25): Windows arm64

Runner CPU: ARMv8 (64-bit) Family 8 Model D49 Revision 0, MICROSOFT CORPORATION (baseline windows-arm64@armv8-64-bit-family-8-model-d49-microsoft-corporation)

Ratios are ParparVM / JDK 25: below 1.00x ParparVM is faster (time) or smaller (RAM). Median of 5 interleaved, paired rounds; every run's output was verified. A ratio more than 20% (time) / 25% (RAM) above its baseline in vm/selfhost/perf-baseline/ is a regression and fails; one that far below is an improvement, reported but not failed (a row whose calibration runs were noisier carries a wider tolerance), and for RAM the change must also exceed 0.05x in absolute terms. Both run unpinned on all of the runner's CPUs, with their own default thread counts.

Benchmark	Cores	Time	RAM	Status
hello (WinHelloMain)	4	1.50x (base 1.45x, +3.9%)	0.77x (base 0.75x, +2.8%)	ok
translator (self)	4	0.90x (base 0.84x, +7.7%)	0.49x (base 0.51x, -3.9%)	ok
intArithmetic	4	1.04x (base 1.04x, +0.0%)	0.06x (base 0.06x, +3.1%)	ok
longArithmetic	4	0.79x (base 0.80x, -0.1%)	0.06x (base 0.06x, +3.5%)	ok
mathTranscendental	4	0.73x (base 0.72x, +1.7%)	0.07x (base 0.06x, +1.9%)	ok
arraySequential	4	0.40x (base 0.46x, -13.0%)	0.39x (base 0.39x, -0.0%)	ok
arrayRandom	4	0.95x (base 0.94x, +1.1%)	0.23x (base 0.23x, +0.3%)	ok
objectAllocation	4	1.81x (base 1.82x, -0.6%)	0.41x (base 0.43x, -5.6%)	ok
valueEscape	4	0.76x (base 0.76x, +0.1%)	0.05x (base 0.05x, +3.4%)	ok
hashMapChurn	4	1.00x (base 0.98x, +1.2%)	0.12x (base 0.12x, -2.3%)	ok
stringBuilding	4	1.54x (base 1.43x, +7.7%)	0.27x (base 0.27x, -0.1%)	ok
recursion	4	1.42x (base 1.46x, -3.0%)	0.07x (base 0.07x, +1.5%)	ok
quicksort	4	1.01x (base 1.00x, +0.7%)	0.12x (base 0.12x, +0.7%)	ok

Result: no regression

shai-almog · 2026-11-11T22:36:13Z

Compared 231 screenshots: 231 matched.
✅ JavaScript-port screenshot tests passed.

shai-almog · 2026-11-11T22:41:38Z

Compared 195 screenshots: 195 matched.

Native Android coverage

📊 Line coverage: 19.18% (56169/292885 lines covered) [HTML preview] (artifact android-coverage-report, jacocoAndroidReport/html/index.html)
- Other counters: instruction 18.13% (270601/1492353), branch 12.38% (19677/158943), complexity 10.36% (15847/153017), method 15.74% (11411/72518), class 27.93% (2247/8045)
- Lowest covered classes
  - kotlin.collections.unsigned.kotlin.collections.unsigned.UArraysKt___UArraysKt – 0.00% (0/2396 lines covered)
  - org.jacoco.agent.rt.internal_0e20598.asm.org.jacoco.agent.rt.internal_0e20598.asm.ClassReader – 0.00% (0/1524 lines covered)
  - androidx.compose.runtime.androidx.compose.runtime.LinkComposer – 0.00% (0/1430 lines covered)
  - org.jacoco.agent.rt.internal_0e20598.asm.org.jacoco.agent.rt.internal_0e20598.asm.MethodWriter – 0.00% (0/922 lines covered)
  - kotlinx.coroutines.channels.kotlinx.coroutines.channels.BufferedChannel – 0.00% (0/863 lines covered)
  - kotlin.sequences.kotlin.sequences.SequencesKt___SequencesKt – 0.00% (0/744 lines covered)
  - com.google.common.cache.com.google.common.cache.LocalCache$Segment – 0.00% (0/726 lines covered)
  - okio.okio.Buffer – 0.00% (0/687 lines covered)
  - kotlin.text.kotlin.text.StringsKt___StringsKt – 0.00% (0/630 lines covered)
  - com.codename1.androidcompat.rt.com.codename1.androidcompat.rt.TextXmlPullParser – 0.00% (0/623 lines covered)

✅ Native Android screenshot tests passed.

Native Android coverage

📊 Line coverage: 19.18% (56169/292885 lines covered) [HTML preview] (artifact android-coverage-report, jacocoAndroidReport/html/index.html)
- Other counters: instruction 18.13% (270601/1492353), branch 12.38% (19677/158943), complexity 10.36% (15847/153017), method 15.74% (11411/72518), class 27.93% (2247/8045)
- Lowest covered classes
  - kotlin.collections.unsigned.kotlin.collections.unsigned.UArraysKt___UArraysKt – 0.00% (0/2396 lines covered)
  - org.jacoco.agent.rt.internal_0e20598.asm.org.jacoco.agent.rt.internal_0e20598.asm.ClassReader – 0.00% (0/1524 lines covered)
  - androidx.compose.runtime.androidx.compose.runtime.LinkComposer – 0.00% (0/1430 lines covered)
  - org.jacoco.agent.rt.internal_0e20598.asm.org.jacoco.agent.rt.internal_0e20598.asm.MethodWriter – 0.00% (0/922 lines covered)
  - kotlinx.coroutines.channels.kotlinx.coroutines.channels.BufferedChannel – 0.00% (0/863 lines covered)
  - kotlin.sequences.kotlin.sequences.SequencesKt___SequencesKt – 0.00% (0/744 lines covered)
  - com.google.common.cache.com.google.common.cache.LocalCache$Segment – 0.00% (0/726 lines covered)
  - okio.okio.Buffer – 0.00% (0/687 lines covered)
  - kotlin.text.kotlin.text.StringsKt___StringsKt – 0.00% (0/630 lines covered)
  - com.codename1.androidcompat.rt.com.codename1.androidcompat.rt.TextXmlPullParser – 0.00% (0/623 lines covered)

Benchmark Results

Detailed Performance Metrics

Metric	Duration
SIMD kernel backend	scalar fallback (no native SIMD)
SIMD int-add (64K x300)	java 156ms / native 76ms = 2.0x speedup
SIMD float-mul (64K x300)	java 239ms / native 131ms = 1.8x speedup
SIMD kernel correctness	PASS (native result == scalar reference)
Base64 payload size	8192 bytes
Base64 benchmark iterations	6000
Base64 SIMD byte path	gated to scalar (CPU autovectorizes scalar; explicit SIMD not beneficial here)
Base64 CN1 encode	77.000 ms
Base64 CN1 decode	85.000 ms
Base64 native encode	494.000 ms
Base64 encode ratio (CN1/native)	0.156x (84.4% faster)
Base64 native decode	317.000 ms
Base64 decode ratio (CN1/native)	0.268x (73.2% faster)
Image encode benchmark status	skipped (SIMD unsupported)

shai-almog · 2026-11-11T22:49:08Z

Compared 210 screenshots: 210 matched.
Native Windows port, REAL shipping pipeline: the hellocodenameone screenshot suite rendered by a binary CROSS-COMPILED on Linux (clang-cl + xwin, WebView2 linked) and RUN on a Windows x64 runner. Compared against the in-repo baseline in scripts/windows/screenshots.

Benchmark Results

Detailed Performance Metrics

Metric	Duration
SIMD kernel backend	SSE2 (x64) / NEON (arm64) native kernels
SIMD int-add (64K x300)	java 57ms / native 4ms = 14.2x speedup
SIMD float-mul (64K x300)	java 56ms / native 5ms = 11.2x speedup
SIMD kernel correctness	PASS (native result == scalar reference)
Base64 native bridge	unavailable (CN1 + SIMD + image benchmarks only)
Base64 payload size	8192 bytes
Base64 benchmark iterations	6000
Base64 SIMD byte path	gated to scalar (CPU autovectorizes scalar; explicit SIMD not beneficial here)
Base64 CN1 encode	80.000 ms
Base64 CN1 decode	109.000 ms
Base64 SIMD encode	111.000 ms
Base64 encode ratio (SIMD/CN1)	1.388x (38.7% slower)
Base64 SIMD decode	107.000 ms
Base64 decode ratio (SIMD/CN1)	0.982x (1.8% faster)
Image encode benchmark iterations	100
Image createMask (SIMD off)	9.000 ms
Image createMask (SIMD on)	4.000 ms
Image createMask ratio (SIMD on/off)	0.444x (55.6% faster)
Image applyMask (SIMD off)	31.000 ms
Image applyMask (SIMD on)	27.000 ms
Image applyMask ratio (SIMD on/off)	0.871x (12.9% faster)
Image modifyAlpha (SIMD off)	27.000 ms
Image modifyAlpha (SIMD on)	15.000 ms
Image modifyAlpha ratio (SIMD on/off)	0.556x (44.4% faster)
Image modifyAlpha removeColor (SIMD off)	33.000 ms
Image modifyAlpha removeColor (SIMD on)	19.000 ms
Image modifyAlpha removeColor ratio (SIMD on/off)	0.576x (42.4% faster)

shai-almog · 2026-11-12T00:18:09Z

Compared 210 screenshots: 210 matched.
Native Linux port (x64), GTK3/Cairo/Pango, ParparVM bytecode-to-C (no JVM): the hellocodenameone screenshot suite rendered by a native ELF built + run on the GitHub x64 runner. Baseline: scripts/linux/screenshots.

ParparVM vs HotSpot (JDK 25): Linux x64

Runner CPU: Intel(R) Xeon(R) Platinum 8370C CPU @ 2.80GHz (baseline linux-x64@intel-xeon-platinum-8370c)

1 performance regression:

translator (self) on all 4 CPUs: time 0.66x against a 0.53x baseline (+24.4%, tolerance 20%)

Ratios are ParparVM / JDK 25: below 1.00x ParparVM is faster (time) or smaller (RAM). Median of 5 interleaved, paired rounds; every run's output was verified. A ratio more than 20% (time) / 25% (RAM) above its baseline in vm/selfhost/perf-baseline/ is a regression and fails; one that far below is an improvement, reported but not failed (a row whose calibration runs were noisier carries a wider tolerance), and for RAM the change must also exceed 0.05x in absolute terms. Both run unpinned on all of the runner's CPUs, with their own default thread counts.

Benchmark	Cores	Time	RAM	Status
hello (LinuxHelloMain)	4	1.14x (base 0.90x, +27.0%)	0.78x (base 0.78x, -0.8%)	ok
translator (self)	4	0.66x (base 0.53x, +24.4%)	0.48x (base 0.43x, +12.0%)	REGRESSION (time +24.4%)
intArithmetic	4	1.03x (base 1.03x, +0.0%)	0.04x (base 0.06x, -30.7%)	ok
longArithmetic	4	1.01x (base 1.01x, +0.0%)	0.04x (base 0.05x, -29.9%)	ok
mathTranscendental	4	1.03x (base 1.03x, -0.2%)	0.04x (base 0.06x, -26.5%)	ok
arraySequential	4	2.23x (base 2.52x, -11.6%)	0.36x (base 0.35x, +1.0%)	ok
arrayRandom	4	1.03x (base 1.01x, +1.9%)	0.20x (base 0.21x, -5.7%)	ok
objectAllocation	4	1.14x (base 1.21x, -5.3%)	0.39x (base 0.50x, -22.3%)	ok
valueEscape	4	0.13x (base 0.13x, +0.3%)	0.03x (base 0.04x, -34.0%)	ok
hashMapChurn	4	0.85x (base 1.09x, -22.5%)	0.11x (base 0.10x, +8.7%)	ok
stringBuilding	4	1.55x (base 1.56x, -0.5%)	0.26x (base 0.26x, +1.3%)	ok
recursion	4	1.30x (base 1.30x, -0.2%)	0.04x (base 0.07x, -47.5%)	ok
quicksort	4	1.06x (base 1.06x, +0.2%)	0.10x (base 0.13x, -21.9%)	ok

A change that moves performance ON PURPOSE records the new baseline in vm/selfhost/perf-baseline/pr/6218.json, with the reason, from this job's perf-results.json. The rebaseline is then part of this pull request's diff:

python3 vm/selfhost/calibrate-perf-baseline.py --pr 6218 --reason "<why these benchmarks moved>" perf-results.json

Result: performance regression

shai-almog · 2026-11-12T00:18:17Z

Compared 210 screenshots: 210 matched.
Native Linux port (arm64), GTK3/Cairo/Pango, ParparVM bytecode-to-C (no JVM): the hellocodenameone screenshot suite rendered by a native ELF built + run on the GitHub arm64 runner. Baseline: scripts/linux/screenshots-arm.

ParparVM vs HotSpot (JDK 25): Linux arm64

Runner CPU: Neoverse-N2 (baseline linux-arm64@neoverse-n2)

Ratios are ParparVM / JDK 25: below 1.00x ParparVM is faster (time) or smaller (RAM). Median of 5 interleaved, paired rounds; every run's output was verified. A ratio more than 20% (time) / 25% (RAM) above its baseline in vm/selfhost/perf-baseline/ is a regression and fails; one that far below is an improvement, reported but not failed (a row whose calibration runs were noisier carries a wider tolerance), and for RAM the change must also exceed 0.05x in absolute terms. Both run unpinned on all of the runner's CPUs, with their own default thread counts.

Benchmark	Cores	Time	RAM	Status
hello (LinuxHelloMain)	4	1.16x (base 1.02x, +13.2%)	0.78x (base 0.66x, +18.2%)	ok
translator (self)	4	0.78x (base 0.70x, +11.7%)	0.45x (base 0.43x, +4.7%)	ok
intArithmetic	4	1.04x (base 1.04x, -0.0%)	0.03x (base 0.03x, +4.8%)	ok
longArithmetic	4	0.79x (base 0.79x, -0.3%)	0.03x (base 0.03x, +5.2%)	ok
mathTranscendental	4	1.09x (base 1.10x, -0.3%)	0.04x (base 0.03x, +9.5%)	ok
arraySequential	4	0.36x (base 0.36x, -0.2%)	0.37x (base 0.37x, +0.2%)	ok
arrayRandom	4	0.94x (base 0.94x, -0.0%)	0.20x (base 0.20x, +0.7%)	ok
objectAllocation	4	1.96x (base 1.77x, +10.8%)	0.38x (base 0.37x, +2.4%)	ok
valueEscape	4	0.51x (base 0.51x, -0.1%)	0.03x (base 0.03x, +5.7%)	ok
hashMapChurn	4	0.84x (base 0.84x, +0.3%)	0.12x (base 0.12x, +2.8%)	ok
stringBuilding	4	1.49x (base 1.57x, -4.8%)	0.26x (base 0.26x, -0.1%)	ok
recursion	4	1.55x (base 1.43x, +7.8%)	0.04x (base 0.04x, +3.3%)	ok
quicksort	4	1.00x (base 0.99x, +0.6%)	0.10x (base 0.09x, +2.6%)	ok

Result: no regression

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

shai-almog and others added 3 commits November 11, 2026 20:11

Backend: Server-Sent Events for handlers (Sse.stream)

7a0b321

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

Backend SSE: proxy flushes trickled upstream data; harden demo and test

1130976

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

Backend SSE: proxy flushes per read only for event streams and undecl…

518cb23

…ared lengths available() is always 0 on both Web impls, so the previous guard flushed every read of every download. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

Backend SSE guide: contractions for Vale

15ecf2f

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

Navigation Menu

Repository navigation

Backend: Server-Sent Events for handlers - #6218

Backend: Server-Sent Events for handlers#6218
shai-almog wants to merge 4 commits into
masterfrom
backend-sse

shai-almog commented Nov 11, 2026

Uh oh!

github-actions Bot commented Nov 11, 2026 •

edited

Loading

Uh oh!

github-actions Bot commented Nov 11, 2026

Uh oh!

github-actions Bot commented Nov 11, 2026 •

edited

Loading

Uh oh!

shai-almog commented Nov 11, 2026 •

edited

Loading

Uh oh!

shai-almog commented Nov 11, 2026 •

edited

Loading

Uh oh!

shai-almog commented Nov 11, 2026 •

edited

Loading

Uh oh!

shai-almog commented Nov 11, 2026 •

edited

Loading

Uh oh!

shai-almog commented Nov 11, 2026 •

edited

Loading

Uh oh!

shai-almog commented Nov 12, 2026 •

edited

Loading

Uh oh!

shai-almog commented Nov 12, 2026 •

edited

Loading

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

Conversation

shai-almog commented Nov 11, 2026

What it is

Fixes found on the way

Tests

Docs

Uh oh!

github-actions Bot commented Nov 11, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

github-actions Bot commented Nov 11, 2026

Cloudflare Preview

Uh oh!

github-actions Bot commented Nov 11, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

✅ Continuous Quality Report

Test & Coverage

Static Analysis

Uh oh!

shai-almog commented Nov 11, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Benchmark Results

Detailed Performance Metrics

ParparVM vs HotSpot (JDK 25): Windows x64

Uh oh!

shai-almog commented Nov 11, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Benchmark Results

Detailed Performance Metrics

ParparVM vs HotSpot (JDK 25): Windows arm64

Uh oh!

shai-almog commented Nov 11, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

shai-almog commented Nov 11, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Native Android coverage

Native Android coverage

Benchmark Results

Detailed Performance Metrics

Uh oh!

shai-almog commented Nov 11, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Benchmark Results

Detailed Performance Metrics

Uh oh!

shai-almog commented Nov 12, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

ParparVM vs HotSpot (JDK 25): Linux x64

Uh oh!

shai-almog commented Nov 12, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

ParparVM vs HotSpot (JDK 25): Linux arm64

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

github-actions Bot commented Nov 11, 2026 •

edited

Loading

github-actions Bot commented Nov 11, 2026 •

edited

Loading

shai-almog commented Nov 11, 2026 •

edited

Loading

shai-almog commented Nov 11, 2026 •

edited

Loading

shai-almog commented Nov 11, 2026 •

edited

Loading

shai-almog commented Nov 11, 2026 •

edited

Loading

shai-almog commented Nov 11, 2026 •

edited

Loading

shai-almog commented Nov 12, 2026 •

edited

Loading

shai-almog commented Nov 12, 2026 •

edited

Loading