Re: [LTP] [Fwd: Re: : Test Case Design for move_pages()]]

[Vivek T. <vt...@nc...>]

Hi Vijay,

I have also used move_pages() in my thesis to move very high number of pages
( >100 K). Apparently, there is another way to move pages, which is through
mbind() syscall with MPOL_MF_MOVE_PAGE flag. In a micro-benchmark, I found
that mbind takes really less time as compared to move_pages().  The simple
test case allocates a large number of pages (linearly contiguous) on one
node and then uses either move_pages or mbind to initiate migration on a
different note. I found that for high number of pages (> 20 K), the
difference is of the order of 100x. I don't have an answer to this. I did
some theoretical calculations (based on the NUMA interconnect bandwidth and
the amount of data to be migrated) and it looks that mbind is probably a
truer representation of the migration time.

Please let me know if  someone has an answer for this. I myself don't have
a NUMA system with super user access to test it out and both these
interfaces internally use migrate_pages() in the kernel to do the actual
migration.

Regards,
Vivek Thakkar


On 7/3/08, Vijay Kumar <vij...@br...> wrote:
>
> Christoph Lameter wrote:
>
>> Mlocked pages are movable. Posix states that mlock means that the pages
>> cannot be evicted. And the VM people agreed to work with that strict
>> definition. Moving a page does not evict the page from memory. We
>> therefore follow posix prescribed behavior.
>>
>
> Thanks for the explanation Christoph. Also, it would be great if can you
> help us by reviewing the other test cases for move_pages(),
> the patch is available at
> http://marc.info/?l=ltp-list&m=121482827215098&w=2
>
> Regards,
> Vijay
>



-- 
- Vivek Thakkar

Graduate Intern, Intel