From 09eb28ee1dbbcaab5200b40c9f9dfacec7366229 Mon Sep 17 00:00:00 2001
From: Ali Mohammad Pur <ali.mpfard@gmail.com>
Date: Wed, 23 Apr 2025 14:20:48 +0200
Subject: [PATCH] LibRegex: Better estimate the cost of laying out alts as a
 chain

Previously we were counting the total number of *nodes* in the tree for
the chain cost, which greatly underestimated its cost when large
bytecode entries were present,
This commit switches to estimating it using the total bytecode *size*,
which is a closer value to the true cost than the tree node count.

This corresponds to a ~4x perf improvement on /<script|<style|<link/ in
speedometer.
---
 Libraries/LibRegex/RegexOptimizer.cpp | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/Libraries/LibRegex/RegexOptimizer.cpp b/Libraries/LibRegex/RegexOptimizer.cpp
index 5e69c065cf4..3fe75fca5fe 100644
--- a/Libraries/LibRegex/RegexOptimizer.cpp
+++ b/Libraries/LibRegex/RegexOptimizer.cpp
@@ -1264,7 +1264,7 @@ void Optimizer::append_alternation(ByteCode& target, Span<ByteCode> alternatives
 
     // This is really only worth it if we don't blow up the size by the 2-extra-instruction-per-node scheme, similarly, if no nodes are shared, we're better off not using a tree.
     auto tree_cost = (total_nodes - common_hits) * 2;
-    auto chain_cost = total_nodes + alternatives.size() * 2;
+    auto chain_cost = total_bytecode_entries_in_tree + alternatives.size() * 2;
     dbgln_if(REGEX_DEBUG, "Total nodes: {}, common hits: {} (tree cost = {}, chain cost = {})", total_nodes, common_hits, tree_cost, chain_cost);
 
     if (common_hits == 0 || tree_cost > chain_cost) {