Skip to content

Fix release blockers: PMD gate and iOS WebRTC compile - #6082

Merged
shai-almog merged 4 commits into
masterfrom
fix-release-blockers
Oct 23, 2026
Merged

Fix release blockers: PMD gate and iOS WebRTC compile#6082
shai-almog merged 4 commits into
masterfrom
fix-release-blockers

Conversation

@shai-almog

@shai-almog shai-almog commented Oct 23, 2026

Copy link
Copy Markdown
Collaborator

Fixes the PMD forbidden-violation failure on master (FileDialogSupport identity compare) and the iOS compile failure for apps using WebRTC (CN1WebRTC.m now built with -fobjc-arc; WebRTC headers declare __weak). Parley workflow now triggers on these paths. No dependencies added.

🤖 Generated with Claude Code

@shai-almog @claude
Fix release blockers: PMD gate on master and iOS WebRTC compile
6a920c4 
- FileDialogSupport: the picked == open identity comparison is
  intentional (showAtPosition returns the Command instance), mark it
  NOPMD like the existing CompareObjectsWithEquals sites; master's PR CI
  quality gate fails on it.
- IPhoneBuilder: compile CN1WebRTC.m with -fobjc-arc. The WebRTC
  framework headers declare __weak properties, which clang rejects in a
  manual-reference-counting file, so every iOS app using
  com.codename1.webrtc failed to compile. The bridge already supports
  both memory models. No new dependency.
- parley.yml: run on changes to the iOS WebRTC bridge and IPhoneBuilder,
  since its iOS job is the only one that compiles a WebRTC app.

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

github-actions Bot commented Oct 23, 2026 •
edited
Loading

Copy link
Copy Markdown
Contributor

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

github-actions Bot commented Oct 23, 2026

Copy link
Copy Markdown
Contributor

Cloudflare Preview

@shai-almog

shai-almog commented Oct 23, 2026 •
edited
Loading

Copy link
Copy Markdown
Collaborator Author

Compared 178 screenshots: 178 matched.

Native Android coverage

  • 📊 Line coverage: 17.40% (30688/176354 lines covered) [HTML preview] (artifact android-coverage-report, jacocoAndroidReport/html/index.html)
    • Other counters: instruction 15.70% (144956/923169), branch 10.66% (10013/93896), complexity 8.22% (7449/90604), method 12.01% (5157/42930), class 20.74% (1059/5105)
    • Lowest covered classes
      • kotlin.collections.kotlin.collections.ArraysKt___ArraysKt – 0.00% (0/6367 lines covered)
      • kotlin.collections.unsigned.kotlin.collections.unsigned.UArraysKt___UArraysKt – 0.00% (0/2384 lines covered)
      • org.jacoco.agent.rt.internal_0e20598.asm.org.jacoco.agent.rt.internal_0e20598.asm.ClassReader – 0.00% (0/1524 lines covered)
      • org.jacoco.agent.rt.internal_0e20598.asm.org.jacoco.agent.rt.internal_0e20598.asm.MethodWriter – 0.00% (0/922 lines covered)
      • kotlin.sequences.kotlin.sequences.SequencesKt___SequencesKt – 0.00% (0/736 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% (0/625 lines covered)
      • com.codename1.androidcompat.rt.com.codename1.androidcompat.rt.TextXmlPullParser – 0.00% (0/623 lines covered)
      • com.codename1.androidcompat.androidx.constraintlayout.core.widgets.com.codename1.androidcompat.androidx.constraintlayout.core.widgets.Flow – 0.00% (0/576 lines covered)

✅ Native Android screenshot tests passed.

Native Android coverage

  • 📊 Line coverage: 17.40% (30688/176354 lines covered) [HTML preview] (artifact android-coverage-report, jacocoAndroidReport/html/index.html)
    • Other counters: instruction 15.70% (144956/923169), branch 10.66% (10013/93896), complexity 8.22% (7449/90604), method 12.01% (5157/42930), class 20.74% (1059/5105)
    • Lowest covered classes
      • kotlin.collections.kotlin.collections.ArraysKt___ArraysKt – 0.00% (0/6367 lines covered)
      • kotlin.collections.unsigned.kotlin.collections.unsigned.UArraysKt___UArraysKt – 0.00% (0/2384 lines covered)
      • org.jacoco.agent.rt.internal_0e20598.asm.org.jacoco.agent.rt.internal_0e20598.asm.ClassReader – 0.00% (0/1524 lines covered)
      • org.jacoco.agent.rt.internal_0e20598.asm.org.jacoco.agent.rt.internal_0e20598.asm.MethodWriter – 0.00% (0/922 lines covered)
      • kotlin.sequences.kotlin.sequences.SequencesKt___SequencesKt – 0.00% (0/736 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% (0/625 lines covered)
      • com.codename1.androidcompat.rt.com.codename1.androidcompat.rt.TextXmlPullParser – 0.00% (0/623 lines covered)
      • com.codename1.androidcompat.androidx.constraintlayout.core.widgets.com.codename1.androidcompat.androidx.constraintlayout.core.widgets.Flow – 0.00% (0/576 lines covered)

Benchmark Results

Detailed Performance Metrics

Metric Duration
SIMD kernel backend scalar fallback (no native SIMD)
SIMD int-add (64K x300) java 45ms / native 114ms = 0.3x speedup
SIMD float-mul (64K x300) java 34ms / native 23ms = 1.4x 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 46.000 ms
Base64 CN1 decode 41.000 ms
Base64 native encode 320.000 ms
Base64 encode ratio (CN1/native) 0.144x (85.6% faster)
Base64 native decode 269.000 ms
Base64 decode ratio (CN1/native) 0.152x (84.8% faster)
Image encode benchmark status skipped (SIMD unsupported)

@shai-almog @claude
Flutter: fix Android and native builds broken by #6055
05f0d39 
- DartList.reversed() clashed with Java 21's List.reversed(), which
  Android's platform jar (API 35+) carries, so every generated Android
  project failed in :app:compileReleaseJavaWithJavac. Renamed to
  reversedIterable(); the emitter and tests follow.
- DateTime.parse used java.util.regex, which ParparVM does not provide,
  so the translated dart_core_DateTime.c failed on every native target.
  It now uses com.codename1.util.regex, as RegExp already does; the
  pattern matches and groups identically.

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

github-actions Bot commented Oct 23, 2026 •
edited
Loading

Copy link
Copy Markdown
Contributor

Flutter vs Codename One

No benchmark results were produced.

@shai-almog

shai-almog commented Oct 23, 2026 •
edited
Loading

Copy link
Copy Markdown
Collaborator Author

Compared 214 screenshots: 214 matched.
✅ JavaScript-port screenshot tests passed.

@github-actions

github-actions Bot commented Oct 23, 2026 •
edited
Loading

Copy link
Copy Markdown
Contributor

✅ Continuous Quality Report

Test & Coverage

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)
    • ✅ core-unittests: 0 findings (no issues)
    • ✅ ios: 0 findings (no issues)
    • ✅ javac: 0 findings (no issues)
    • ✅ project-model: 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

shai-almog commented Oct 23, 2026 •
edited
Loading

Copy link
Copy Markdown
Collaborator Author

Compared 193 screenshots: 193 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: AMD EPYC 7763 64-Core Processor (baseline linux-x64@amd-epyc-7763-64-core-processor)

1 improvement past the baseline (not a failure; rebaseline to lock it in):

  • translator (self) on all 4 CPUs: time 0.51x against a 0.64x baseline (-20.4%, 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 0.84x (base 0.85x, -0.6%) 0.91x (base 0.83x, +8.7%) ok
translator (self) 4 0.51x (base 0.64x, -20.4%) 0.49x (base 0.52x, -5.3%) improved: rebaseline to lock in (time -20.4%)
intArithmetic 4 1.13x (base 1.12x, +1.1%) 0.08x (base 0.06x, +36.7%) ok
longArithmetic 4 1.08x (base 1.08x, +0.0%) 0.06x (base 0.05x, +3.4%) ok
mathTranscendental 4 1.09x (base 1.08x, +0.4%) 0.08x (base 0.07x, +6.1%) ok
arraySequential 4 2.38x (base 2.00x, +18.7%) 0.36x (base 0.36x, -0.3%) ok
arrayRandom 4 1.10x (base 0.91x, +20.6%) 0.22x (base 0.21x, +4.0%) ok
objectAllocation 4 1.65x (base 1.65x, -0.1%) 0.26x (base 0.35x, -27.7%) ok
valueEscape 4 0.10x (base 0.10x, -0.3%) 0.05x (base 0.04x, +2.7%) ok
hashMapChurn 4 1.49x (base 1.48x, +0.6%) 0.10x (base 0.09x, +3.3%) ok
stringBuilding 4 1.41x (base 1.39x, +1.2%) 0.26x (base 0.31x, -17.0%) ok
recursion 4 1.26x (base 1.25x, +0.0%) 0.08x (base 0.08x, +1.4%) ok
quicksort 4 1.07x (base 1.07x, +0.2%) 0.11x (base 0.11x, -0.4%) ok

A change that moves performance ON PURPOSE records the new baseline in vm/selfhost/perf-baseline/pr/6082.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 6082 --reason "<why these benchmarks moved>" perf-results.json

Result: no regression (improved past the baseline: consider rebaselining)

@shai-almog

shai-almog commented Oct 23, 2026 •
edited
Loading

Copy link
Copy Markdown
Collaborator Author

Compared 193 screenshots: 193 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 0.97x (base 1.02x, -5.2%) 0.79x (base 0.93x, -14.7%) ok
translator (self) 4 0.75x (base 0.70x, +7.3%) 0.46x (base 0.43x, +6.1%) ok
intArithmetic 4 1.04x (base 1.04x, -0.1%) 0.03x (base 0.03x, +4.6%) ok
longArithmetic 4 0.79x (base 0.79x, -0.1%) 0.03x (base 0.03x, +5.2%) ok
mathTranscendental 4 1.11x (base 1.10x, +0.7%) 0.04x (base 0.03x, +10.0%) ok
arraySequential 4 0.41x (base 0.36x, +14.0%) 0.37x (base 0.37x, +0.4%) ok
arrayRandom 4 0.94x (base 0.94x, +0.3%) 0.20x (base 0.20x, +1.0%) ok
objectAllocation 4 1.97x (base 1.77x, +11.4%) 0.36x (base 0.37x, -1.5%) ok
valueEscape 4 0.51x (base 0.51x, -0.1%) 0.03x (base 0.03x, +5.4%) ok
hashMapChurn 4 0.83x (base 0.84x, -1.6%) 0.12x (base 0.12x, -2.0%) ok
stringBuilding 4 1.49x (base 1.57x, -5.1%) 0.26x (base 0.26x, +0.2%) ok
recursion 4 1.56x (base 1.43x, +8.7%) 0.04x (base 0.04x, +3.5%) ok
quicksort 4 1.00x (base 0.99x, +1.1%) 0.09x (base 0.09x, +1.4%) ok

Result: no regression

@shai-almog

shai-almog commented Oct 23, 2026 •
edited
Loading

Copy link
Copy Markdown
Collaborator Author

Compared 193 screenshots: 193 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 73ms / native 5ms = 14.6x speedup
SIMD float-mul (64K x300) java 65ms / native 4ms = 16.2x 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 86.000 ms
Base64 CN1 decode 89.000 ms
Base64 SIMD encode 105.000 ms
Base64 encode ratio (SIMD/CN1) 1.221x (22.1% slower)
Base64 SIMD decode 100.000 ms
Base64 decode ratio (SIMD/CN1) 1.124x (12.4% slower)
Image encode benchmark iterations 100
Image createMask (SIMD off) 10.000 ms
Image createMask (SIMD on) 2.000 ms
Image createMask ratio (SIMD on/off) 0.200x (80.0% faster)
Image applyMask (SIMD off) 31.000 ms
Image applyMask (SIMD on) 21.000 ms
Image applyMask ratio (SIMD on/off) 0.677x (32.3% faster)
Image modifyAlpha (SIMD off) 37.000 ms
Image modifyAlpha (SIMD on) 20.000 ms
Image modifyAlpha ratio (SIMD on/off) 0.541x (45.9% faster)
Image modifyAlpha removeColor (SIMD off) 45.000 ms
Image modifyAlpha removeColor (SIMD on) 26.000 ms
Image modifyAlpha removeColor ratio (SIMD on/off) 0.578x (42.2% faster)

ParparVM vs HotSpot (JDK 25): Windows x64

Runner CPU: AMD64 Family 25 Model 1 Stepping 1, AuthenticAMD (baseline windows-x64@amd64-family-25-model-1-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 1.46x (base 1.22x, +19.5%) 0.55x (base 0.94x, -41.4%) ok
translator (self) 4 0.79x (base 0.60x, +29.8%) 0.56x (base 0.54x, +2.4%) ok
intArithmetic 4 1.10x (base 1.10x, +0.0%) 0.06x (base 0.06x, +1.2%) ok
longArithmetic 4 1.14x (base 1.08x, +5.0%) 0.06x (base 0.05x, +2.1%) ok
mathTranscendental 4 0.78x (base 0.79x, -0.9%) 0.06x (base 0.06x, +1.6%) ok
arraySequential 4 1.35x (base 1.36x, -0.7%) 0.39x (base 0.39x, +0.3%) ok
arrayRandom 4 0.97x (base 1.06x, -8.5%) 0.23x (base 0.23x, +0.3%) ok
objectAllocation 4 1.56x (base 1.56x, -0.0%) 0.28x (base 0.39x, -26.2%) ok
valueEscape 4 0.10x (base 0.10x, +0.3%) 0.05x (base 0.04x, +2.4%) ok
hashMapChurn 4 1.89x (base 1.80x, +5.0%) 0.09x (base 0.09x, -1.1%) ok
stringBuilding 4 1.28x (base 1.22x, +5.4%) 0.27x (base 0.32x, -17.1%) ok
recursion 4 1.41x (base 1.49x, -5.8%) 0.06x (base 0.06x, +0.2%) ok
quicksort 4 1.11x (base 1.13x, -2.2%) 0.12x (base 0.12x, +0.3%) ok

Result: no regression

shai-almog and others added 2 commits October 23, 2026 07:28
@shai-almog @claude
Flutter: parse DateTime by hand instead of with a regex engine
d0b183d 
Using com.codename1.util.regex for DateTime.parse put the whole regex
engine into every Flutter app (+1.2% to +2.1% executable code, failing
the Flutter benchmark size gates). The grammar is one fixed expression,
so it is matched by hand. DateTimeIsoGrammarTest holds the matcher to the
original expression on the JVM: same match and same groups, over the
known forms and 200,000 generated strings.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@shai-almog @claude
Flutter benchmark: rebaseline JavaScript code size for #6055's runtime
252c70b 
The JavaScript build had not compiled since #6055, so master's nightly
size predated its Dart runtime; this PR restores the build.

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

shai-almog commented Oct 23, 2026 •
edited
Loading

Copy link
Copy Markdown
Collaborator Author

Compared 175 screenshots: 175 matched.
✅ Native Mac screenshot tests passed.

Benchmark Results

  • VM Translation Time: 0 seconds
  • Compilation Time: 494 seconds

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 60ms / native 3ms = 20.0x speedup
SIMD kernel correctness PASS (native result == scalar reference)
Base64 payload size 8192 bytes
Base64 benchmark iterations 6000
Base64 SIMD byte path active (NEON-accelerated)
Base64 CN1 encode 68.000 ms
Base64 CN1 decode 81.000 ms
Base64 native encode 929.000 ms
Base64 encode ratio (CN1/native) 0.073x (92.7% faster)
Base64 native decode 343.000 ms
Base64 decode ratio (CN1/native) 0.236x (76.4% faster)
Base64 SIMD encode 70.000 ms
Base64 encode ratio (SIMD/CN1) 1.029x (2.9% slower)
Base64 SIMD decode 45.000 ms
Base64 decode ratio (SIMD/CN1) 0.556x (44.4% faster)
Base64 encode ratio (SIMD/native) 0.075x (92.5% faster)
Base64 decode ratio (SIMD/native) 0.131x (86.9% faster)
Image encode benchmark iterations 100
Image createMask (SIMD off) 13.000 ms
Image createMask (SIMD on) 6.000 ms
Image createMask ratio (SIMD on/off) 0.462x (53.8% faster)
Image applyMask (SIMD off) 74.000 ms
Image applyMask (SIMD on) 73.000 ms
Image applyMask ratio (SIMD on/off) 0.986x (1.4% faster)
Image modifyAlpha (SIMD off) 68.000 ms
Image modifyAlpha (SIMD on) 69.000 ms
Image modifyAlpha ratio (SIMD on/off) 1.015x (1.5% slower)
Image modifyAlpha removeColor (SIMD off) 93.000 ms
Image modifyAlpha removeColor (SIMD on) 74.000 ms
Image modifyAlpha removeColor ratio (SIMD on/off) 0.796x (20.4% faster)

@shai-almog

shai-almog commented Oct 23, 2026 •
edited
Loading

Copy link
Copy Markdown
Collaborator Author

Compared 176 screenshots: 176 matched.
✅ Native iOS Metal screenshot tests passed.

Benchmark Results

  • VM Translation Time: 0 seconds
  • Compilation Time: 1976 seconds

Build and Run Timing

Metric Duration
Simulator Boot 93000 ms
Simulator Boot (Run) 1000 ms
App Install 15000 ms
App Launch 4000 ms
Test Execution 442000 ms

Detailed Performance Metrics

Metric Duration
SIMD kernel backend SSE2 (x64) / NEON (arm64) native kernels
SIMD int-add (64K x300) java 103ms / native 2ms = 51.5x speedup
SIMD float-mul (64K x300) java 84ms / native 6ms = 14.0x speedup
SIMD kernel correctness PASS (native result == scalar reference)
Base64 payload size 8192 bytes
Base64 benchmark iterations 6000
Base64 SIMD byte path active (NEON-accelerated)
Base64 CN1 encode 53.000 ms
Base64 CN1 decode 57.000 ms
Base64 native encode 947.000 ms
Base64 encode ratio (CN1/native) 0.056x (94.4% faster)
Base64 native decode 596.000 ms
Base64 decode ratio (CN1/native) 0.096x (90.4% faster)
Base64 SIMD encode 51.000 ms
Base64 encode ratio (SIMD/CN1) 0.962x (3.8% faster)
Base64 SIMD decode 46.000 ms
Base64 decode ratio (SIMD/CN1) 0.807x (19.3% faster)
Base64 encode ratio (SIMD/native) 0.054x (94.6% faster)
Base64 decode ratio (SIMD/native) 0.077x (92.3% faster)
Image encode benchmark iterations 100
Image createMask (SIMD off) 6.000 ms
Image createMask (SIMD on) 3.000 ms
Image createMask ratio (SIMD on/off) 0.500x (50.0% faster)
Image applyMask (SIMD off) 72.000 ms
Image applyMask (SIMD on) 63.000 ms
Image applyMask ratio (SIMD on/off) 0.875x (12.5% faster)
Image modifyAlpha (SIMD off) 45.000 ms
Image modifyAlpha (SIMD on) 99.000 ms
Image modifyAlpha ratio (SIMD on/off) 2.200x (120.0% slower)
Image modifyAlpha removeColor (SIMD off) 108.000 ms
Image modifyAlpha removeColor (SIMD on) 129.000 ms
Image modifyAlpha removeColor ratio (SIMD on/off) 1.194x (19.4% slower)

@shai-almog

shai-almog commented Oct 23, 2026 •
edited
Loading

Copy link
Copy Markdown
Collaborator Author

Compared 193 screenshots: 193 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 49ms / native 3ms = 16.3x speedup
SIMD float-mul (64K x300) java 49ms / native 3ms = 16.3x 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 60.000 ms
Base64 CN1 decode 70.000 ms
Base64 SIMD encode 77.000 ms
Base64 encode ratio (SIMD/CN1) 1.283x (28.3% slower)
Base64 SIMD decode 69.000 ms
Base64 decode ratio (SIMD/CN1) 0.986x (1.4% faster)
Image encode benchmark iterations 100
Image createMask (SIMD off) 5.000 ms
Image createMask (SIMD on) 1.000 ms
Image createMask ratio (SIMD on/off) 0.200x (80.0% faster)
Image applyMask (SIMD off) 17.000 ms
Image applyMask (SIMD on) 18.000 ms
Image applyMask ratio (SIMD on/off) 1.059x (5.9% slower)
Image modifyAlpha (SIMD off) 17.000 ms
Image modifyAlpha (SIMD on) 13.000 ms
Image modifyAlpha ratio (SIMD on/off) 0.765x (23.5% faster)
Image modifyAlpha removeColor (SIMD off) 22.000 ms
Image modifyAlpha removeColor (SIMD on) 14.000 ms
Image modifyAlpha removeColor ratio (SIMD on/off) 0.636x (36.4% 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 1.00x (base 1.11x, -9.9%) 0.69x (base 0.75x, -8.1%) ok
translator (self) 4 0.89x (base 0.84x, +6.6%) 0.48x (base 0.51x, -5.2%) ok
intArithmetic 4 1.04x (base 1.04x, +0.0%) 0.06x (base 0.06x, +2.8%) ok
longArithmetic 4 0.80x (base 0.80x, +0.1%) 0.06x (base 0.06x, +3.1%) ok
mathTranscendental 4 0.72x (base 0.72x, -0.4%) 0.06x (base 0.06x, +1.3%) ok
arraySequential 4 0.37x (base 0.46x, -18.0%) 0.39x (base 0.39x, +0.2%) ok
arrayRandom 4 0.94x (base 0.94x, +0.2%) 0.23x (base 0.23x, +0.4%) ok
objectAllocation 4 1.75x (base 2.81x, -37.7%) 0.33x (base 0.43x, -22.4%) ok
valueEscape 4 0.76x (base 0.76x, +0.1%) 0.05x (base 0.05x, +1.7%) ok
hashMapChurn 4 0.99x (base 0.98x, +1.0%) 0.11x (base 0.12x, -4.2%) ok
stringBuilding 4 1.60x (base 1.43x, +12.1%) 0.27x (base 0.27x, -0.3%) ok
recursion 4 1.55x (base 1.46x, +5.9%) 0.07x (base 0.07x, +1.0%) ok
quicksort 4 1.00x (base 1.00x, +0.2%) 0.12x (base 0.12x, +0.6%) ok

Result: no regression

@shai-almog

shai-almog commented Oct 23, 2026 •
edited
Loading

Copy link
Copy Markdown
Collaborator Author

Compared 176 screenshots: 176 matched.
✅ Native iOS Metal screenshot tests passed.

Benchmark Results

  • VM Translation Time: 0 seconds
  • Compilation Time: 1730 seconds

Build and Run Timing

Metric Duration
Simulator Boot 60000 ms
Simulator Boot (Run) 0 ms
App Install 11000 ms
App Launch 4000 ms
Test Execution 392000 ms

Detailed Performance Metrics

Metric Duration
SIMD kernel backend SSE2 (x64) / NEON (arm64) native kernels
SIMD int-add (64K x300) java 58ms / native 4ms = 14.5x speedup
SIMD float-mul (64K x300) java 57ms / native 3ms = 19.0x speedup
SIMD kernel correctness PASS (native result == scalar reference)
Base64 payload size 8192 bytes
Base64 benchmark iterations 6000
Base64 SIMD byte path active (NEON-accelerated)
Base64 CN1 encode 79.000 ms
Base64 CN1 decode 60.000 ms
Base64 native encode 666.000 ms
Base64 encode ratio (CN1/native) 0.119x (88.1% faster)
Base64 native decode 832.000 ms
Base64 decode ratio (CN1/native) 0.072x (92.8% faster)
Base64 SIMD encode 56.000 ms
Base64 encode ratio (SIMD/CN1) 0.709x (29.1% faster)
Base64 SIMD decode 49.000 ms
Base64 decode ratio (SIMD/CN1) 0.817x (18.3% faster)
Base64 encode ratio (SIMD/native) 0.084x (91.6% faster)
Base64 decode ratio (SIMD/native) 0.059x (94.1% faster)
Image encode benchmark iterations 100
Image createMask (SIMD off) 5.000 ms
Image createMask (SIMD on) 2.000 ms
Image createMask ratio (SIMD on/off) 0.400x (60.0% faster)
Image applyMask (SIMD off) 41.000 ms
Image applyMask (SIMD on) 39.000 ms
Image applyMask ratio (SIMD on/off) 0.951x (4.9% faster)
Image modifyAlpha (SIMD off) 28.000 ms
Image modifyAlpha (SIMD on) 32.000 ms
Image modifyAlpha ratio (SIMD on/off) 1.143x (14.3% slower)
Image modifyAlpha removeColor (SIMD off) 37.000 ms
Image modifyAlpha removeColor (SIMD on) 43.000 ms
Image modifyAlpha removeColor ratio (SIMD on/off) 1.162x (16.2% slower)

@shai-almog

shai-almog commented Oct 23, 2026 •
edited
Loading

Copy link
Copy Markdown
Collaborator Author

Compared 179 screenshots: 179 matched.
✅ Native Mac screenshot tests passed.

Benchmark Results

  • VM Translation Time: 0 seconds
  • Compilation Time: 418 seconds

Detailed Performance Metrics

Metric Duration
SIMD kernel backend SSE2 (x64) / NEON (arm64) native kernels
SIMD int-add (64K x300) java 66ms / native 3ms = 22.0x speedup
SIMD float-mul (64K x300) java 60ms / native 3ms = 20.0x 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 active (NEON-accelerated)
Base64 CN1 encode 53.000 ms
Base64 CN1 decode 61.000 ms
Image encode benchmark iterations 100
Image createMask (SIMD off) 7.000 ms
Image createMask (SIMD on) 4.000 ms
Image createMask ratio (SIMD on/off) 0.571x (42.9% faster)
Image applyMask (SIMD off) 68.000 ms
Image applyMask (SIMD on) 77.000 ms
Image applyMask ratio (SIMD on/off) 1.132x (13.2% slower)
Image modifyAlpha (SIMD off) 61.000 ms
Image modifyAlpha (SIMD on) 58.000 ms
Image modifyAlpha ratio (SIMD on/off) 0.951x (4.9% faster)
Image modifyAlpha removeColor (SIMD off) 72.000 ms
Image modifyAlpha removeColor (SIMD on) 55.000 ms
Image modifyAlpha removeColor ratio (SIMD on/off) 0.764x (23.6% faster)

ParparVM vs HotSpot (JDK 25): macOS arm64

Runner CPU: Apple M1 (Virtual) (baseline macos-arm64)

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 (HelloCodenameOne) 3 0.82x (base 0.96x, -14.5%) 0.90x (base 0.93x, -2.9%) ok
translator (self) 3 0.57x (base 0.62x, -7.7%) 0.71x (base 0.58x, +23.4%) ok
intArithmetic 3 1.02x (base 1.03x, -1.5%) 0.12x (base 0.12x, +1.3%) ok
longArithmetic 3 1.02x (base 1.03x, -0.8%) 0.12x (base 0.07x, +64.9%) ok
mathTranscendental 3 0.95x (base 1.01x, -5.5%) 0.12x (base 0.12x, +0.2%) ok
arraySequential 3 0.43x (base 0.42x, +2.3%) 0.53x (base 0.53x, +0.3%) ok
arrayRandom 3 0.98x (base 0.99x, -0.8%) 0.33x (base 0.32x, +0.3%) ok
objectAllocation 3 1.05x (base 1.11x, -5.2%) 0.81x (base 0.72x, +11.7%) ok
valueEscape 3 0.51x (base 0.51x, +0.1%) 0.09x (base 0.09x, -0.1%) ok
hashMapChurn 3 1.03x (base 1.18x, -12.2%) 0.08x (base 0.07x, +15.0%) ok
stringBuilding 3 0.84x (base 0.84x, -0.6%) 0.46x (base 0.46x, +0.1%) ok
recursion 3 1.20x (base 1.24x, -3.1%) 0.12x (base 0.12x, +0.7%) ok
quicksort 3 0.99x (base 1.01x, -2.0%) 0.11x (base 0.11x, +2.0%) ok

Result: no regression

@github-actions

github-actions Bot commented Oct 23, 2026

Copy link
Copy Markdown
Contributor

✅ ByteCodeTranslator Quality Report

Test & Coverage

  • ✅ Tests: 880 total, 0 failed, 55 skipped

Benchmark Results

  • Execution Time: 8612 ms

  • Hotspots (Top 20 sampled methods):

    • 9.57% com.codename1.tools.translator.IteratorEscape.mangle (125 samples)
    • 7.27% java.util.ArrayList.indexOf (95 samples)
    • 5.44% com.codename1.tools.translator.BytecodeMethod.equals (71 samples)
    • 5.05% java.lang.Object.hashCode (66 samples)
    • 5.05% java.lang.System.identityHashCode (66 samples)
    • 3.68% com.codename1.tools.translator.classfile.analysis.Frame.execute (48 samples)
    • 3.68% java.lang.StringBuilder.append (48 samples)
    • 2.30% com.codename1.tools.translator.IteratorEscape.ctorOnlyStoresParamsIntoThis (30 samples)
    • 2.07% com.codename1.tools.translator.JavascriptReachability.enqueueResolved (27 samples)
    • 1.53% com.codename1.tools.translator.bytecodes.Invoke.resolveDirectTarget (20 samples)
    • 1.38% com.codename1.tools.translator.Parser.cullMethods (18 samples)
    • 1.23% java.lang.String.equals (16 samples)
    • 1.15% com.codename1.tools.translator.ByteCodeClass.generateCCode (15 samples)
    • 1.15% java.util.HashMap.putVal (15 samples)
    • 1.07% com.codename1.tools.translator.BytecodeMethod.updateInlinableFieldDependencies (14 samples)
    • 0.92% com.codename1.tools.translator.BytecodeMethod.optimize (12 samples)
    • 0.92% com.codename1.tools.translator.ByteCodeClass.hasDeclaredMethod (12 samples)
    • 0.84% com.codename1.tools.translator.ReachabilityCull.run (11 samples)
    • 0.84% java.util.HashMap.hash (11 samples)
    • 0.84% java.util.AbstractCollection.containsAll (11 samples)

  • ⚠️ Coverage report not generated.

Static Analysis

  • ✅ SpotBugs: no findings (report was not generated by the build).
  • ⚠️ PMD report not generated.
  • ⚠️ Checkstyle report not generated.

Generated automatically by the PR CI workflow.

@shai-almog
shai-almog merged commit c0d6767 into master Oct 23, 2026
17 of 63 checks passed
@shai-almog
shai-almog deleted the fix-release-blockers branch October 23, 2026 06:39
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

1 participant

@shai-almog