<div dir="ltr">Hello Laurent,<div><br></div><div>I'll take a look and find a fix.</div></div><div class="gmail_extra"><br><div class="gmail_quote">2016-08-25 11:52 GMT+03:00 Laurent Dufour <span dir="ltr"><<a href="mailto:ldufour@linux.vnet.ibm.com" target="_blank">ldufour@linux.vnet.ibm.com</a>></span>:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><span class="">On 07/08/2016 15:11, Eugene Batalov wrote:<br>
> Main test features:<br>
> - Non trivial ps tree with non trivial anon shmem regions<br>
> (no such test exists now).<br>
> - Each ps tree process continuously writes parts of anon shmem<br>
> vmas and validates these writes after restore<br>
> (required for dedup testing).<br>
> - Checking simultaneous changing of anon shmem contents in different<br>
> processes.<br>
><br>
> Signed-off-by: Eugene Batalov <<a href="mailto:eabatalov89@gmail.com">eabatalov89@gmail.com</a>><br>
<br>
<br>
</span>This test is not working at all on PowerPC, because the PAGE_SIZE is 64k<br>
by default, and the test is not using the same magnitude when dealing<br>
with the memory area's size and when dealing with this size.<br>
<br>
In the test, MEM_PERIOD is defined based on PAGE_SIZE, while the size of<br>
the memory area are define based on 1UL arbitrary shift.<br>
<br>
On PowerPC this leads to a core dump of the test in proc13_func()<br>
because mem2_size become negative !<br>
<br>
I tried to fix that by applying the attached patched but the size of the<br>
manage memory areas become very large on PowerPC, about 60G of memory<br>
allocated for this test.<br>
<br>
Could you please review this test to manage the memory area size's order<br>
using PAGE_SIZE ?<br>
<br>
Thanks,<br>
Laurent.<br>
<br>
<br>
</blockquote></div><br><br clear="all"><div><br></div>-- <br><div class="gmail_signature" data-smartmail="gmail_signature">Best regards,<br>Eugene Batalov.</div>
</div>