Re: [PATCH 2/2] mm/mempolicy: Skip walking HUGETLB vma if MPOL_MF_STRICT is specified alone

[Yang Shi <yang.shi@linux.alibaba.com>]

On 1/14/20 6:09 AM, Li Xinhai wrote:
> Add cc to
> Yang Shi <yang.shi@linux.alibaba.com>
> Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
> , who has been worked on this part
>
> On 2020-01-14 at 17:16 Li Xinhai wrote:
>> Checking MPOL_MF_STRICT is ignored for HUGETLB vma according to mbind man
>> page:
>>
>> Notes
>> MPOL_MF_STRICT is ignored on huge page mappings.
>>
>> If MPOL_MF_STRICT is specified alone without any MOVE flag, we should
>> indicate, from test_walk, that walking this vma should be skipped even if
>> there are misplaced pages.
>>
>> Signed-off-by: Li Xinhai <lixinhai.lxh@gmail.com>
>> Cc: Michal Hocko <mhocko@suse.com>
>> Cc: Mike Kravetz <mike.kravetz@oracle.com>
>> ---
>> mm/mempolicy.c | 7 +++++++
>> 1 file changed, 7 insertions(+)
>>
>> diff --git a/mm/mempolicy.c b/mm/mempolicy.c
>> index 067cf7d..c117b5f 100644
>> --- a/mm/mempolicy.c
>> +++ b/mm/mempolicy.c
>> @@ -656,6 +656,13 @@ static int queue_pages_test_walk(unsigned long start, unsigned long end,
>> return 1;
>> }
>>
>> +	/* MPOL_MF_STRICT is ignored for huge page, skip checking
>> +	*  misplaced pages
>> +	*/
>> +	if ((flags & MPOL_MF_VALID) == MPOL_MF_STRICT &&
>> +	is_vm_hugetlb_page(vma))
>> +	return 1;

It makes some sense to me.  We do save some effort by not 
acquiring/releasing ptl and walking page tables.

Reviewed-by: Yang Shi <yang.shi@linux.alibaba.com>

>> /* queue pages from current vma */
>> if (flags & MPOL_MF_VALID)
>> return 0;
>> --
>> 1.8.3.1
> >