I’d also be interested in reading more. My guess would be all the pointer chasing during marking is expensive and since Go doesn’t use a compacting GC there is

I’d also be interested in reading more. My guess would be all the pointer chasing during marking is expensive and since Go doesn’t use a compacting GC there isn’t anything to do other than push freed memory back into the allocators internal data structures during a sweep.