mem-hotplug: fix potential race while building zonelist for new populated zone
authorHaicheng Li <haicheng.li@linux.intel.com>
Mon, 24 May 2010 21:32:52 +0000 (14:32 -0700)
committerLinus Torvalds <torvalds@linux-foundation.org>
Tue, 25 May 2010 15:07:02 +0000 (08:07 -0700)
commit4eaf3f64397c3db3c5785eee508270d62a9fabd9
treebfd986a7e974876755ea6fe0de394199c68e2e36
parent1f522509c77a5dea8dc384b735314f03908a6415
mem-hotplug: fix potential race while building zonelist for new populated zone

Add global mutex zonelists_mutex to fix the possible race:

     CPU0                                  CPU1                    CPU2
(1) zone->present_pages += online_pages;
(2)                                       build_all_zonelists();
(3)                                                               alloc_page();
(4)                                                               free_page();
(5) build_all_zonelists();
(6)   __build_all_zonelists();
(7)     zone->pageset = alloc_percpu();

In step (3,4), zone->pageset still points to boot_pageset, so bad
things may happen if 2+ nodes are in this state. Even if only 1 node
is accessing the boot_pageset, (3) may still consume too much memory
to fail the memory allocations in step (7).

Besides, atomic operation ensures alloc_percpu() in step (7) will never fail
since there is a new fresh memory block added in step(6).

[haicheng.li@linux.intel.com: hold zonelists_mutex when build_all_zonelists]
Signed-off-by: Haicheng Li <haicheng.li@linux.intel.com>
Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
Reviewed-by: Andi Kleen <andi.kleen@intel.com>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Tejun Heo <tj@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
include/linux/mmzone.h
kernel/cpu.c
mm/memory_hotplug.c
mm/page_alloc.c