On 06/22/2015 03:21 PM, Kirill A. Shutemov wrote:
> On Tue, Jun 16, 2015 at 03:17:13PM +0200, Jerome Marchand wrote:
>> On 06/03/2015 07:05 PM, Kirill A. Shutemov wrote:
>>> Hello everybody,
>>>
>>> Here's new revision of refcounting patchset. Please review and consider
>>> applying.
>>>
>>> The goal of patchset is to make refcounting on THP pages cheaper with
>>> simpler semantics and allow the same THP compound page to be mapped with
>>> PMD and PTEs. This is required to get reasonable THP-pagecache
>>> implementation.
>>>
>>> With the new refcounting design it's much easier to protect against
>>> split_huge_page(): simple reference on a page will make you the deal.
>>> It makes gup_fast() implementation simpler and doesn't require
>>> special-case in futex code to handle tail THP pages.
>>>
>>> It should improve THP utilization over the system since splitting THP in
>>> one process doesn't necessary lead to splitting the page in all other
>>> processes have the page mapped.
>>>
>>> The patchset drastically lower complexity of get_page()/put_page()
>>> codepaths. I encourage people look on this code before-and-after to
>>> justify time budget on reviewing this patchset.
>>>
>>> = Changelog =
>>>
>>> v6:
>>>   - rebase to since-4.0;
>>>   - optimize mapcount handling: significantely reduce overhead for most
>>>     common cases.
>>>   - split pages on migrate_pages();
>>>   - remove infrastructure for handling splitting PMDs on all architectures;
>>>   - fix page_mapcount() for hugetlb pages;
>>>
>>
>> Hi Kirill,
>>
>> I ran some LTP mm tests and hugemmap tests trigger the following:
>>
>> [  438.749457] page:ffffea0000df8000 count:2 mapcount:0 mapping:          (null) index:0x0 compound_mapcount: 0
>> [  438.750089] flags: 0x3ffc0000004001(locked|head)
>> [  438.750089] page dumped because: VM_BUG_ON_PAGE(page_mapped(page))
> 
> Did you run with original or updated version of patch 27/36?
> In original post of v6 there was bug: page_mapped() always returned true.
> 

Indeed! I'll try again with the corrected patch.

Thanks,
Jerome