Conversation

It completely solves a huge performance problem I would have had to deal with by hand and only added 16 bytes to the binary size. Before: 478.092526 msec 2,000,329,069 cycles 3,525,042,263 instructions After: 322.796288 msec 1,349,630,342 cycles 3,625,044,439 instructions
1
2
That's a microbenchmark of the malloc implementation using various size classes and writing data to the memory allocations. It's a drastic performance improvement since it's a perfect use case for the library. It only needs to set up the two divisors for each size class at init.