Benchmarking Wild vs. Mold

Wild’s linker benchmarks were updated to include Mold, revealing Wild as substantially slower under Mold’s test conditions. The discrepancy stems from differing benchmark setups: Mold deletes output files and uses ext4 with --no-fork, while Wild historically ran on tmpfs without deletion and with default forking. Adjusting Wild to Mold’s configuration narrows the performance gap, highlighting filesystem and fork-handling as key factors.