Re: oops in copy_page_rep()

[Hugh Dickins <hughd@google.com>]

On Sun, 6 Jan 2013, Dave Jones wrote:
> 
> investigating the huge page theory a little further I'm a bit confused.
> The kernel on that machine has THP enabled, and the cpu supports it (an old amd64), but..
> 
> $ cat /sys/kernel/mm/hugepages/hugepages-2048kB/*
> 0
> 0
> 0
> 0
> 0
> 0
> 
> I was expecting at least one of those to be non-zero.
> 
> /sys/kernel/mm/transparent_hugepage/khugepaged/full_scans and pages_collapsed
> are both non-zero, so it's been busy doing _something_.
> 
> Is this expected behaviour ?

Yes.  /sys/kernel/mm/hugepages/ is all about those dusty old opaque
hugepages you might have got from fs/hugetlbfs and mm/hugetlb.c; whereas
the splitting issue we suspect is peculiar to the dazzling new transparent
hugepages you get from mm/huge_memory.c.

Hugh