[CRIU] [PATCH v3 12/12] test: add test for anon shmem dedup

Eugene Batalov eabatalov89 at gmail.com
Thu Aug 25 02:27:40 PDT 2016


Hello Laurent,

I'll take a look and find a fix.

2016-08-25 11:52 GMT+03:00 Laurent Dufour <ldufour at linux.vnet.ibm.com>:

> On 07/08/2016 15:11, Eugene Batalov wrote:
> > Main test features:
> > - Non trivial ps tree with non trivial anon shmem regions
> >   (no such test exists now).
> > - Each ps tree process continuously writes parts of anon shmem
> >   vmas and validates these writes after restore
> >   (required for dedup testing).
> > - Checking simultaneous changing of anon shmem contents in different
> >   processes.
> >
> > Signed-off-by: Eugene Batalov <eabatalov89 at gmail.com>
>
>
> This test is not working at all on PowerPC, because the PAGE_SIZE is 64k
> by default, and the test is not using the same magnitude when dealing
> with the memory area's size and when dealing with this size.
>
> In the test, MEM_PERIOD is defined based on PAGE_SIZE, while the size of
> the memory area are define based on 1UL arbitrary shift.
>
> On PowerPC this leads to a core dump of the test in proc13_func()
> because mem2_size become negative !
>
> I tried to fix that by applying the attached patched but the size of the
> manage memory areas become very large on PowerPC, about 60G of memory
> allocated for this test.
>
> Could you please review this test to manage the memory area size's order
> using PAGE_SIZE ?
>
> Thanks,
> Laurent.
>
>
>


-- 
Best regards,
Eugene Batalov.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openvz.org/pipermail/criu/attachments/20160825/0964c876/attachment.html>


More information about the CRIU mailing list