Messages in this thread | | | Date | Mon, 27 Jul 2009 11:00:48 -0700 (PDT) | From | David Rientjes <> | Subject | Re: [BUG] set_mempolicy(MPOL_INTERLEAV) cause kernel panic |
| |
On Sat, 25 Jul 2009, KAMEZAWA Hiroyuki wrote:
> This behavior itself is not very bad. > And all hotplug thing is just a side story of this bugfix. >
Right, the original problem that Lee reported doesn't appear to be caused by hotplug.
> To update nodemask, user's mask should be saved in the policy > even when the mask is not relative and v.node should be calculated > again, at event. IIUC, rather than per-policy update by notifer, > some new implemenation for policy will be necessary. >
We don't need additional non-default mempolicy support for MEM_ONLINE. It would be inappropriate to store the user nodemask and then hot-add new nodes to mempolicies based on the given node id's when nothing is assumed of its proximity. It's better left to userspace to update existing mempolicies to use the newly added memory.
| |