On 3/5/25 16:10, Michal Hocko wrote: > Sorry, I have missed follow ups here. > >> I assume it's probably not easy to >> implement as page table allocations are involved in the process and we don't >> have a way to supply preallocated memory for those. > > Why would this be a concern if the allocation is done outside of the > lock? It's not a concern if it can be done outside of the lock. I don't know the code enough to see if it's feasible. There's e.g. pcpu_populate_chunk() ending up doing vmap_pages_range_noflush() that AFAIK means page table allocations. Can it be done outside of the lock?
