NUMA architecture CPU API change summary

stay NUMA framework CPU On , In order to make the requested memory space right node Have affinity , Try to make cpu Access local memory instead of remote memory , Conducive to providing CPU Execution efficiency .
CPU The speed of accessing memory of different types of nodes is different ： Local node > Neighbor nodes > Remote node . The fastest access to local nodes , The slowest access to remote nodes , That is, the access speed is related to the distance between nodes , The longer the distance, the slower the access speed , This distance is called Node Distance.
http://t.zoukankan.com/zhenjing-p-linux_numa.html




Create thread ：

struct task_struct *kthread_create_on_node(int (*threadfn)(void *data),
					   void *data, int node,
					   const char namefmt[],
					   ...)
@ Substitute prototype ：
kthread_create

NUMA Related literature ：
《Local and Remote Memory: Memory in a Linux NUMA System》
http://www.ilinuxkernel.com/files/Local.and.Remote.Memory.Memory.in.a.Linux.NUMA.System.pdf