Skip to content

RFC RFC memoryless node handling fixes

[email protected] linux-mm 1 version In review

Get series with b4
b4 am [email protected]

Versions newest first

  1. v1 RFC Latest13 patches 6 repliesIn review

Patches in v1 No applies recorded yet

#Subject and filesSizeRepliesApplied
1/13NUMA: introduce node_memory_mapinclude/linux/nodemask.h mm/page_alloc.c+18 -00 replies
2/13Fix MPOL_INTERLEAVE behavior for memoryless nodesmm/mempolicy.c+2 -10 replies
3/13OOM: use the node_memory_map instead of constructing one on the flymm/oom_kill.c+1 -80 replies
4/13Memoryless Nodes: No need for kswapdmm/vmscan.c+1 -10 replies
5/13Memoryless Node: Slab supportmm/slab.c+4 -40 replies
6/13Memoryless nodes: SLUB supportmm/slub.c+8 -80 replies
7/13Uncached allocator: Handle memoryless nodesarch/ia64/kernel/uncached.c drivers/char/mspec.c+3 -30 replies
8/13Memoryless node: Allow profiling data to fall back to other nodeskernel/profile.c+2 -20 replies
9/13Memoryless nodes: Update memory policy and page migrationmm/mempolicy.c mm/migrate.c+11 -110 replies
10/13Memoryless nodes: Fix GFP_THISNODE behaviorinclude/linux/gfp.h include/linux/mmzone.h mm/page_alloc.c+38 -103 replies
11/13SLUB: Ensure that the # object per slabs stays low enough.mm/slub.c+19 -20 replies
12/13SLUB: minimum alignment fixesinclude/linux/slub_def.h mm/slub.c+26 -71 reply
13/13I finally found a way to get rid of the nasty list of comparisions in slub_def.h. ilog2 seems to work right for constants.include/linux/slub_def.h+11 -451 reply

Reviews trailers per patch

2 of 13 patches have a Reviewed-by or Acked-by.

PatchPekka EnbergNishanth Aravamudan
1/13
2/13
3/13
4/13
5/13
6/13
7/13
8/13
9/13
10/13A-b
11/13
12/13
13/13A-b

Trailers given in replies are filled; those carried in the patches from earlier versions are outlined with dashes.

Read the v1 thread

Cover letter [RFC 00/13] RFC memoryless node handling fixes

This has now become a longer series since I have seen a couple of things in various places where we do not take into account memoryless nodes.

I changed the GFP_THISNODE fix to generate a new set of zonelists. GFP_THISNODE will then simply use a zonelist that only has the zones of the node.

I have only tested this by booting on a IA64 simulator. Please review. I do not have a real system with a memoryless node.

Signature · 5 lines
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to [email protected].  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

Read the cover letter in its thread