String handling optimisation

Do you have some statistics on this? I'd imagine that most time spent on comparising hash-hits would be on short to medium length strings, and not really long ones since it's unlikely you'll find another string with the exact same length in the hash bucket.

I do have some statistics from the Opera Mini servers.

We spent around 0.2% of the CPU time in string hashing before the CRC32 change. After that we spend around 0.1%CPU time there.

Almost all time is spent traversing linked list, rescaling the whole process to 100% gives 10% to crc32 and and 1.8% to the string comparisons.

87% is spent in the has lookup + loop.