On Mon, Jul 02, 2012 at 11:43:31AM +0200, Michal Hocko wrote:
>On Mon 02-07-12 17:28:56, Gavin Shan wrote:
>> sparse_index_init() is designed to be safe if two copies of it race. It
>> uses "index_init_lock" to ensure that, even in the case of a race, only
>> one CPU will manage to do:
>>
>> mem_section[root] = section;
>>
>> However, in the case where two copies of sparse_index_init() _do_ race,
>> the one that loses the race will leak the "section" that
>> sparse_index_alloc() allocated for it. This patch fixes that leak.
>
>I would still like to hear how we can possibly race in this code path.
>I've thought that memory onlining is done from a single CPU.
>
Hi Michael, how about to use the following changelog? :-)
-----
sparse_index_init() is designed to be safe if two copies of it race. It
uses "index_init_lock" to ensure that, even in the case of a race, only
one CPU will manage to do:
mem_section[root] = section;
However, in the case where two copies of sparse_index_init() _do_ race,
which is probablly caused by making online for multiple memory sections
that depend on same entry of array mem_section[] simultaneously from
different CPUs. The one that loses the race will leak the "section" that
sparse_index_alloc() allocated for it. This patch fixes that leak.
-----
Thanks,
Gavin
>>
>> Signed-off-by: Gavin Shan <shangw@xxxxxxxxxxxxxxxxxx>
>> ---
>> mm/sparse.c | 17 +++++++++++++++++
>> 1 file changed, 17 insertions(+)
>>
>> diff --git a/mm/sparse.c b/mm/sparse.c
>> index 781fa04..a6984d9 100644
>> --- a/mm/sparse.c
>> +++ b/mm/sparse.c
>> @@ -75,6 +75,20 @@ static struct mem_section noinline __init_refok *sparse_index_alloc(int nid)
>> return section;
>> }
>>
>> +static inline void __meminit sparse_index_free(struct mem_section *section)
>> +{
>> + unsigned long size = SECTIONS_PER_ROOT *
>> + sizeof(struct mem_section);
>> +
>> + if (!section)
>> + return;
>> +
>> + if (slab_is_available())
>> + kfree(section);
>> + else
>> + free_bootmem(virt_to_phys(section), size);
>> +}
>> +
>> static int __meminit sparse_index_init(unsigned long section_nr, int nid)
>> {
>> static DEFINE_SPINLOCK(index_init_lock);
>> @@ -102,6 +116,9 @@ static int __meminit sparse_index_init(unsigned long section_nr, int nid)
>> mem_section[root] = section;
>> out:
>> spin_unlock(&index_init_lock);
>> + if (ret)
>> + sparse_index_free(section);
>> +
>> return ret;
>> }
>> #else /* !SPARSEMEM_EXTREME */
>> --
>> 1.7.9.5
>>
>> --
>> To unsubscribe, send a message with 'unsubscribe linux-mm' in
>> the body to majordomo@xxxxxxxxx. For more info on Linux MM,
>> see: http://www.linux-mm.org/ .
>> Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>
>
>--
>Michal Hocko
>SUSE Labs
>SUSE LINUX s.r.o.
>Lihovarska 1060/12
>190 00 Praha 9
>Czech Republic
>
>--
>To unsubscribe, send a message with 'unsubscribe linux-mm' in
>the body to majordomo@xxxxxxxxx. For more info on Linux MM,
>see: http://www.linux-mm.org/ .
>Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>
>
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxx. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>