Is there a new way in Big Sur, running on an M1, to enable and utilize large memory pages? I've done a lot of searching on the internet and many tests, but haven't been able to figure it out. Closest I've come to insight is thinking there might be a conflict with JIT preventing the usage of it? This is what used to work in C++: void *mem = mmap(0, size, PROT_READ | PROT_WRITE, MAP_PRIVATE | MAP_ANON, VM_FLAGS_SUPERPAGE_SIZE_2MB, 0); Also tried: void *p = mmap((void *) 0x000200000000, 8 * 1024 * 1024, PROT_READ | PROT_WRITE, MAP_ANON | MAP_FIXED | MAP_PRIVATE, VM_FLAGS_SUPERPAGE_SIZE_2MB, 0); Maybe there's an alternative way to accomplish Huge pages.

In the case of an entire framework being deprecated, like ML Compute, is there a note left somewhere or a link that helps explain the direction and intended replacement? When visiting the ML Compute page, there are just lots of "deprecated" badges, but no resource that provides explanation.