test262: 59.47% (+0.02) | 🧪 50424 | 🤠 29987 (+9) | ❌ 7293 (-9) | 💀 12862 | 🏗️ 7 | 💥 262 | ⏰ 13 | 📝 0

test262: 59.45% (+0.17) | 🧪 50424 | 🤠 29978 (+89) | ❌ 7302 (-107) | 💀 12862 (+18) | 🏗️ 7 | 💥 262 | ⏰ 13 | 📝 0

test262: 59.28% (+0.02) | 🧪 50424 | 🤠 29889 (+8) | ❌ 7409 (+10) | 💀 12844 (-18) | 🏗️ 7 | 💥 262 | ⏰ 13 | 📝 0

test262: 59.26% | 🧪 50424 | 🤠 29881 (+2) | ❌ 7399 (-2) | 💀 12862 | 🏗️ 7 | 💥 262 | ⏰ 13 | 📝 0

test262: 59.26% (+0.03) | 🧪 50424 | 🤠 29879 (+14) | ❌ 7401 (-18) | 💀 12862 (+4) | 🏗️ 7 | 💥 262 | ⏰ 13 | 📝 0

reuse generateCall to have much better indirect calls (full rest arg support, less buggy with internal proto, etc)

test262: 59.23% (+0.01) | 🧪 50424 | 🤠 29865 (+6) | ❌ 7419 (+1) | 💀 12858 (-6) | 🏗️ 7 | 💥 262 | ⏰ 13 (-1) | 📝 0

yippee

test262: 59.22% (+0.02) | 🧪 50424 | 🤠 29859 (+8) | ❌ 7418 (-8) | 💀 12864 | 🏗️ 7 | 💥 262 | ⏰ 14 | 📝 0

much simpler/faster than before, same test262 coverage

reverts 016344b011ff09fdad0566b376cbaba2e2a881ed. yeah it wasn't unneeded. oops.

test262: 59.20% | 🧪 50424 | 🤠 29851 | ❌ 7426 | 💀 12864 | 🏗️ 7 | 💥 262 | ⏰ 14 | 📝 0

unused and should be rewritten anyway. o7

doesn't actually do anything anymore

test262: 59.20% | 🧪 50424 | 🤠 29851 | ❌ 7426 | 💀 12864 | 🏗️ 7 | 💥 262 | ⏰ 14 | 📝 0

test262: 59.20% (+0.03) | 🧪 50424 | 🤠 29851 (+14) | ❌ 7426 (+1) | 💀 12864 (-14) | 🏗️ 7 | 💥 262 | ⏰ 14 (-1) | 📝 0

test262: 59.17% (+0.45) | 🧪 50424 | 🤠 29837 (+227) | ❌ 7425 (-71) | 💀 12878 (-157) | 🏗️ 7 | 💥 262 | ⏰ 15 (+1) | 📝 0

test262: 58.72% (+0.10) | 🧪 50424 | 🤠 29610 (+53) | ❌ 7496 (-52) | 💀 13035 (-1) | 🏗️ 7 | 💥 262 | ⏰ 14 | 📝 0

makes test262 faster and scale better as more and more builtins emerge

test262: 58.62% | 🧪 50424 | 🤠 29557 | ❌ 7548 | 💀 13036 | 🏗️ 7 | 💥 262 | ⏰ 14 | 📝 0

just arrays are slower for init and make it easier for post-codegen stages to modify something and break it between compiles, so just only use funcs.

test262: 58.62% | 🧪 50424 | 🤠 29557 | ❌ 7548 | 💀 13036 | 🏗️ 7 | 💥 262 | ⏰ 14 | 📝 0

just move entirely to glbl. also remove a legacy unneeded prng algo.

test262: 58.62% | 🧪 50424 | 🤠 29557 | ❌ 7548 | 💀 13036 | 🏗️ 7 | 💥 262 | ⏰ 14 | 📝 0

now unneeded after last fix, yay

test262: 58.62% | 🧪 50424 | 🤠 29557 | ❌ 7548 (-1) | 💀 13036 (+1) | 🏗️ 7 | 💥 262 | ⏰ 14 | 📝 0

test262: 58.62% (+0.03) | 🧪 50424 | 🤠 29557 (+12) | ❌ 7549 (-2) | 💀 13035 (-10) | 🏗️ 7 | 💥 262 | ⏰ 14 | 📝 0
envoy: 0.2.0 (C4 L2M 8)

test262: 58.59% | 🧪 50424 | 🤠 29545 | ❌ 7551 | 💀 13045 | 🏗️ 7 | 💥 262 | ⏰ 14 | 📝 0

test262: 58.59% | 🧪 50424 | 🤠 29545 | ❌ 7551 | 💀 13045 | 🏗️ 7 | 💥 262 | ⏰ 14 | 📝 0

test262: 58.59% (+0.02) | 🧪 50424 | 🤠 29545 (+14) | ❌ 7551 (-17) | 💀 13045 (+3) | 🏗️ 7 | 💥 262 | ⏰ 14 | 📝 0

makes wasm output a lot smaller+simpler. test262 bump from this plus PRs!

test262: 58.57% (+0.02) | 🧪 50424 | 🤠 29531 (+8) | ❌ 7568 (-8) | 💀 13042 | 🏗️ 7 | 💥 262 | ⏰ 14 | 📝 0

test262: 58.55% | 🧪 50424 | 🤠 29523 | ❌ 7576 | 💀 13042 | 🏗️ 7 | 💥 262 | ⏰ 14 | 📝 0

makes startup for all compiles ~30ms faster! this also includes test262 bump from prs!

test262: 58.55% (+0.15) | 🧪 50424 | 🤠 29523 (+73) | ❌ 7576 (-75) | 💀 13042 (+2) | 🏗️ 7 | 💥 262 | ⏰ 14 | 📝 0

test262: 58.40% | 🧪 50424 | 🤠 29450 | ❌ 7651 | 💀 13040 | 🏗️ 7 | 💥 262 | ⏰ 14 | 📝 0

test262: 58.40% | 🧪 50424 | 🤠 29450 | ❌ 7651 | 💀 13040 | 🏗️ 7 | 💥 262 | ⏰ 14 | 📝 0

includes small bump from PRs :)

test262: 58.40% | 🧪 50424 | 🤠 29450 (+4) | ❌ 7651 | 💀 13040 (-4) | 🏗️ 7 | 💥 262 | ⏰ 14 | 📝 0

test262: 58.40% | 🧪 50424 | 🤠 29446 | ❌ 7651 | 💀 13044 | 🏗️ 7 | 💥 262 | ⏰ 14 | 📝 0