Search results

perf: BenchmarkPrototypeMethodAccess builds an engine inside its timed loop, so it measures startup

Status: Open.

#51 In nooga/paserati;

· mparrett opened on Aug 4, 2026

compiler: register allocation is non-deterministic — one binary compiles the same source two different ways

Status: Open.

#50 In nooga/paserati;

· mparrett opened on Aug 4, 2026

perf: the calibration anchor is captured in a different process and iteration regime from what it normalizes

Status: Open.

#49 In nooga/paserati;

· mparrett opened on Aug 1, 2026

perf: b.N movement correlates with round-to-round spread; the causal direction is unresolved

Status: Open.

#48 In nooga/paserati;

· mparrett opened on Aug 1, 2026

built-ins: As*() called under guards that don't establish the type — 11 confirmed silent-exit-0 reproducers

Status: Open.

#46 In nooga/paserati;

· mparrett opened on Jul 25, 2026

VM: recovered panics are reported as successful runs (silent exit 0)

Status: Open.

#44 In nooga/paserati;

· mparrett opened on Jul 25, 2026

VM micro-optimization series — 8 PRs + reviewer's guide

Status: Open.

#37 In nooga/paserati;

· mparrett opened on Jul 22, 2026

perf-timeline: test262.total is single-shot and poorly runner-normalized (~6% scatter swamps engine signal)

Status: Open.

#28 In nooga/paserati;

· mparrett opened on Jul 18, 2026

compiler: fused ++/-- on a spilled local targets R0 instead of the spill slot

Status: Open.

#27 In nooga/paserati;

· mparrett opened on Jul 13, 2026

perf-timeline: test262.total sum conflates per-test speed with pass-count (conformance inflates the trend)

Status: Open.

#26 In nooga/paserati;

· mparrett opened on Jul 13, 2026

Epic: perf benchmarking infrastructure (tracker)

Status: Open.

#23 In nooga/paserati;

· mparrett opened on Jul 12, 2026

perf-pr A/B is ungateable as a single shot here too; median-of-N fixes it

Status: Open.

#21 In nooga/paserati;

· mparrett opened on Jul 12, 2026