On Wed, 8 Aug 2007, Mel Gorman wrote: > > > For various policies, the arguments would look like this: > > Policy start node nodemask > > > > default local node cpuset_current_mems_allowed > > > > preferred preferred_node cpuset_current_mems_allowed > > > > interleave computed node cpuset_current_mems_allowed > > > > bind local node policy nodemask [replaces bind > > zonelist in mempolicy] > >
GFP_THISNODE could be realized by only setting the desired nodenumber in the nodemask. > The last one is the most interesting. Much of the patch in development > involves deleting the custom node stuff. I've included the patch below if > you're curious. I wanted to get one-zonelist out first to see if we could > agree on that before going further with it. I think we do. > > Then, just walk the zonelist for the starting node--already ordered by > > distance--filtering by gfp_zone() and nodemask. Done "right", this > > should always return memory from the closest allowed node [based on the > > nodemask argument] to the starting node. And, it would eliminate the > > custom zonelists for bind policy. Can also eliminate cpuset checks in > > the allocation loop because that constraint would already be applied to > > the nodemask argument. > > > > This is what I'm hoping. I haven't looked closely enough to be sure this will > work but currently I see no reason why it couldn't and it might eliminate > some of the NUMA-specific paths in the allocator. Right. But lets first get the general case for the single nodelist accepted (with the zoneid optimizations?) - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
