Jumps are slower on completely random input. If I understand Linus’s point correctly, he is suggesting that random inputs like this are unusual (although a good

Jumps are slower on completely random input. If I understand Linus’s point correctly, he is suggesting that random inputs like this are unusual (although a good way to measure worst case performance)