On 2016/9/20 15:07, Michal Hocko wrote:
> [CCing Tetsuo again - please make sure you CC everybody who did respond
> in earlier versions of the patch]
>
> I am sorry to insist here but this doesn't address the previous review
> feedback. Let me try to show you what I would find much better. I do not
> insist on this precise wording of course but I do insist on mentioning
> the current state and making clear why GFP_NORETRY is really ok.
>
> On Tue 20-09-16 13:50:13, zhongjiang wrote:
>> From: zhong jiang <zhongjiang@xxxxxxxxxx>
>>
>> I hit the following issue when run a OOM case of the LTP and
>> ksm enable.
> "
> I hit the following hung task when running an OOM LTP test case with 4.1
> kernel.
> "
>
>> Call trace:
>> [<ffffffc000086a88>] __switch_to+0x74/0x8c
>> [<ffffffc000a1bae0>] __schedule+0x23c/0x7bc
>> [<ffffffc000a1c09c>] schedule+0x3c/0x94
>> [<ffffffc000a1eb84>] rwsem_down_write_failed+0x214/0x350
>> [<ffffffc000a1e32c>] down_write+0x64/0x80
>> [<ffffffc00021f794>] __ksm_exit+0x90/0x19c
>> [<ffffffc0000be650>] mmput+0x118/0x11c
>> [<ffffffc0000c3ec4>] do_exit+0x2dc/0xa74
>> [<ffffffc0000c46f8>] do_group_exit+0x4c/0xe4
>> [<ffffffc0000d0f34>] get_signal+0x444/0x5e0
>> [<ffffffc000089fcc>] do_signal+0x1d8/0x450
>> [<ffffffc00008a35c>] do_notify_resume+0x70/0x78
>>
>> it will leads to a hung task because the exiting task cannot get the
>> mmap sem for write. but the root cause is that the ksmd holds it for
>> read while allocateing memory which just takes ages to complete.
>> and ksmd will loop in the following path.
> "
> The oom victim cannot terminate because it needs to take mmap_sem for
> write while the lock is held by ksmd for read which loops in the page
> allocator
>
> ksm_do_scan
> scan_get_next_rmap_item
> down_read
> get_next_rmap_item
> alloc_rmap_item #ksmd will loop permanently.
>
> There is not way forward because the oom victim cannot release any
> memory in 4.1 based kernel. Since 4.6 we have the oom reaper which would
> solve this problem because it would release the memory asynchronously.
> Nevertheless we can relax alloc_rmap_item requirements and use
> __GFP_NORETRY because the allocation failure is acceptable as
> ksm_do_scan would just retry later after the lock got dropped.
>
> Such a patch would be also easy to backport to older stable kernels
> which do not have oom_reaper.
>
> While we are at it add GFP_NOWARN as the admin doesn't have to be
> alarmed by the allocation failure.
>> CC: <stable@xxxxxxxxxxxxxxx>
>> Suggested-by: Hugh Dickins <hughd@xxxxxxxxxx>
>> Suggested-by: Michal Hocko <mhocko@xxxxxxx>
>> Signed-off-by: zhong jiang <zhongjiang@xxxxxxxxxx>
>> ---
>> mm/ksm.c | 3 ++-
>> 1 file changed, 2 insertions(+), 1 deletion(-)
>>
>> diff --git a/mm/ksm.c b/mm/ksm.c
>> index 73d43ba..5048083 100644
>> --- a/mm/ksm.c
>> +++ b/mm/ksm.c
>> @@ -283,7 +283,8 @@ static inline struct rmap_item *alloc_rmap_item(void)
>> {
>> struct rmap_item *rmap_item;
>>
>> - rmap_item = kmem_cache_zalloc(rmap_item_cache, GFP_KERNEL);
>> + rmap_item = kmem_cache_zalloc(rmap_item_cache, GFP_KERNEL |
>> + __GFP_NORETRY | __GFP_NOWARN);
>> if (rmap_item)
>> ksm_rmap_items++;
>> return rmap_item;
>> --
>> 1.8.3.1
Thanks you for advice, I will modify it now .
--
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>