linux/arch/x86/mm
KAMEZAWA Hiroyuki f2dbcfa738 mm: clean up for early_pfn_to_nid()
What's happening is that the assertion in mm/page_alloc.c:move_freepages()
is triggering:

	BUG_ON(page_zone(start_page) != page_zone(end_page));

Once I knew this is what was happening, I added some annotations:

	if (unlikely(page_zone(start_page) != page_zone(end_page))) {
		printk(KERN_ERR "move_freepages: Bogus zones: "
		       "start_page[%p] end_page[%p] zone[%p]\n",
		       start_page, end_page, zone);
		printk(KERN_ERR "move_freepages: "
		       "start_zone[%p] end_zone[%p]\n",
		       page_zone(start_page), page_zone(end_page));
		printk(KERN_ERR "move_freepages: "
		       "start_pfn[0x%lx] end_pfn[0x%lx]\n",
		       page_to_pfn(start_page), page_to_pfn(end_page));
		printk(KERN_ERR "move_freepages: "
		       "start_nid[%d] end_nid[%d]\n",
		       page_to_nid(start_page), page_to_nid(end_page));
 ...

And here's what I got:

	move_freepages: Bogus zones: start_page[2207d0000] end_page[2207dffc0] zone[fffff8103effcb00]
	move_freepages: start_zone[fffff8103effcb00] end_zone[fffff8003fffeb00]
	move_freepages: start_pfn[0x81f600] end_pfn[0x81f7ff]
	move_freepages: start_nid[1] end_nid[0]

My memory layout on this box is:

[    0.000000] Zone PFN ranges:
[    0.000000]   Normal   0x00000000 -> 0x0081ff5d
[    0.000000] Movable zone start PFN for each node
[    0.000000] early_node_map[8] active PFN ranges
[    0.000000]     0: 0x00000000 -> 0x00020000
[    0.000000]     1: 0x00800000 -> 0x0081f7ff
[    0.000000]     1: 0x0081f800 -> 0x0081fe50
[    0.000000]     1: 0x0081fed1 -> 0x0081fed8
[    0.000000]     1: 0x0081feda -> 0x0081fedb
[    0.000000]     1: 0x0081fedd -> 0x0081fee5
[    0.000000]     1: 0x0081fee7 -> 0x0081ff51
[    0.000000]     1: 0x0081ff59 -> 0x0081ff5d

So it's a block move in that 0x81f600-->0x81f7ff region which triggers
the problem.

This patch:

Declaration of early_pfn_to_nid() is scattered over per-arch include
files, and it seems it's complicated to know when the declaration is used.
 I think it makes fix-for-memmap-init not easy.

This patch moves all declaration to include/linux/mm.h

After this,
  if !CONFIG_NODES_POPULATES_NODE_MAP && !CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID
     -> Use static definition in include/linux/mm.h
  else if !CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID
     -> Use generic definition in mm/page_alloc.c
  else
     -> per-arch back end function will be called.

Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Tested-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Reported-by: David Miller <davem@davemlloft.net>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: <stable@kernel.org>		[2.6.25.x, 2.6.26.x, 2.6.27.x, 2.6.28.x]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-02-18 15:37:55 -08:00
..
dump_pagetables.c x86: another user of PTE_FLAGS_MASK 2008-08-20 12:38:41 +02:00
extable.c x86: unify extable_{32|64}.c 2008-01-30 13:31:41 +01:00
fault.c prevent kprobes from catching spurious page faults 2009-02-05 17:01:50 -08:00
gup.c x86: two trivial sparse annotations 2008-10-29 08:02:28 +01:00
highmem_32.c Export kmap_atomic_pfn for DRM-GEM. 2008-10-18 07:10:12 +10:00
hugetlbpage.c x86: add hugepagesz option on 64-bit 2008-07-24 10:47:19 -07:00
init_32.c x86: fix assumed to be contiguous leaf page tables for kmap_atomic region (take 2) 2009-01-16 13:47:04 +01:00
init_64.c x86: remove kernel_physical_mapping_init() from init section 2009-01-20 00:31:43 +01:00
iomap_32.c x86: work around PAGE_KERNEL_WC not getting WC in iomap_atomic_prot_pfn. 2009-01-26 11:14:27 +01:00
ioremap.c x86, pat: fix warn_on_once() while mapping 0-1MB range with /dev/mem 2009-02-12 08:27:27 +01:00
k8topology_64.c x86: k8 numa register active regions later 2009-01-06 13:21:21 +01:00
kmmio.c - fix mmioftrace + rcu merge interaction 2008-05-25 09:51:43 +02:00
Makefile Merge branches 'tracing/ftrace', 'tracing/fastboot', 'tracing/nmisafe' and 'tracing/urgent' into tracing/core 2008-11-08 09:34:35 +01:00
memtest.c x86: memtest fix use of reserve_early() 2008-10-22 17:08:06 +02:00
mmap.c x86: unify mmap_{32|64}.c 2008-01-30 13:31:10 +01:00
mmio-mod.c mmiotrace: remove left-over marker cruft 2008-10-14 10:37:17 +02:00
numa_32.c trivial: replace last usages of __FUNCTION__ in kernel 2009-01-07 15:48:54 -08:00
numa_64.c mm: clean up for early_pfn_to_nid() 2009-02-18 15:37:55 -08:00
pageattr-test.c x86: make sure the CPA test code's use of _PAGE_UNUSED1 is obvious 2008-09-05 17:09:57 +02:00
pageattr.c x86: CPA avoid repeated lazy mmu flush 2009-02-12 23:11:58 +01:00
pat.c x86, pat: fix warn_on_once() while mapping 0-1MB range with /dev/mem 2009-02-12 08:27:27 +01:00
pf_in.c x86: fix mmiotrace 8-bit register decoding 2008-10-14 10:33:50 +02:00
pf_in.h x86 mmiotrace: move files into arch/x86/mm/. 2008-05-24 11:25:37 +02:00
pgtable_32.c i386: vmalloc size fix 2008-08-21 10:13:21 +02:00
pgtable.c x86: pgd_{c,d}tor() cleanup 2008-09-06 19:47:09 +02:00
srat_32.c x86: don't call e820_regiter_active_regions if out of range on node 2008-08-14 11:35:52 +02:00
srat_64.c x86: prepare for cpumask iterators to only go to nr_cpu_ids 2008-12-16 17:40:58 -08:00
testmmiotrace.c x86 mmiotrace: implement mmiotrace_printk() 2008-10-14 10:37:11 +02:00