Build 32-bit FFI fixture DLL on Windows CI#227
Conversation
- Install the i686 MinGW toolchain for `i386-win32` jobs - Use a 32-bit `libfixture.dll` when running the 32-bit TestRunner
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Organization UI Review profile: ASSERTIVE Plan: Pro Run ID: 📒 Files selected for processing (1)
📝 WalkthroughWalkthroughCI workflow updated to run target-aware Windows FFI fixture builds: adds an MSYS2/pacman install step for Changes
Estimated code review effort🎯 2 (Simple) | ⏱️ ~8 minutes Possibly related PRs
🚥 Pre-merge checks | ✅ 3✅ Passed checks (3 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. Comment |
- Use MSYS2 MINGW32 for the 32-bit Windows fixture DLL - Split Windows 32-bit and 64-bit build steps
Benchmark Results274 benchmarks Interpreted: 🟢 233 improved · 🔴 10 regressed · 31 unchanged · avg +7.8% arraybuffer.js — Interp: 🟢 14 · avg +13.1% · Bytecode: 🟢 12, 🔴 1, 1 unch. · avg +8.5%
arrays.js — Interp: 🟢 19 · avg +9.8% · Bytecode: 🟢 18, 1 unch. · avg +6.0%
async-await.js — Interp: 🟢 6 · avg +9.0% · Bytecode: 🟢 6 · avg +8.2%
classes.js — Interp: 🟢 31 · avg +13.7% · Bytecode: 🟢 8, 🔴 9, 14 unch. · avg -1.0%
closures.js — Interp: 🟢 11 · avg +9.6% · Bytecode: 🟢 5, 6 unch. · avg +3.0%
collections.js — Interp: 🟢 12 · avg +12.9% · Bytecode: 🟢 11, 🔴 1 · avg +13.8%
destructuring.js — Interp: 🟢 22 · avg +9.6% · Bytecode: 🟢 10, 🔴 3, 9 unch. · avg +2.3%
fibonacci.js — Interp: 🟢 8 · avg +7.3% · Bytecode: 🟢 8 · avg +10.1%
for-of.js — Interp: 🟢 4, 3 unch. · avg +3.0% · Bytecode: 🟢 1, 🔴 4, 2 unch. · avg -7.8%
helpers/bench-module.js — Interp: 0 · Bytecode: 0
iterators.js — Interp: 🟢 14, 6 unch. · avg +3.8% · Bytecode: 🟢 14, 🔴 1, 5 unch. · avg +3.4%
json.js — Interp: 🟢 20 · avg +8.6% · Bytecode: 🟢 17, 3 unch. · avg +6.6%
jsx.jsx — Interp: 🟢 21 · avg +7.1% · Bytecode: 🟢 1, 🔴 5, 15 unch. · avg -1.3%
modules.js — Interp: 🟢 9 · avg +4.5% · Bytecode: 🟢 7, 2 unch. · avg +2.4%
numbers.js — Interp: 🟢 10, 1 unch. · avg +8.3% · Bytecode: 🟢 7, 🔴 1, 3 unch. · avg +6.2%
objects.js — Interp: 🟢 7 · avg +9.0% · Bytecode: 🟢 1, 🔴 2, 4 unch. · avg -1.2%
promises.js — Interp: 🟢 3, 🔴 3, 6 unch. · avg +0.8% · Bytecode: 🟢 10, 2 unch. · avg +4.5%
regexp.js — Interp: 🟢 8, 3 unch. · avg +5.1% · Bytecode: 🟢 6, 🔴 2, 3 unch. · avg +2.7%
strings.js — Interp: 🟢 6, 🔴 2, 3 unch. · avg +0.7% · Bytecode: 🟢 10, 1 unch. · avg +6.6%
typed-arrays.js — Interp: 🟢 8, 🔴 5, 9 unch. · avg +3.2% · Bytecode: 🟢 10, 🔴 7, 5 unch. · avg +2.5%
Measured on ubuntu-latest x64. Benchmark ranges compare cached main-branch min/max ops/sec with the PR run; overlapping ranges are treated as unchanged noise. Percentage deltas are secondary context. |
Suite Timing
Measured on ubuntu-latest x64. |
Summary
i386-win32.TestRunnercan load a matching DLL.Testing
C:/msys64/mingw32/bin/gcc.exeafter installingmingw-w64-i686-gccfori386-win32.i386-win32Windows path still uses the existinggccinvocation.