Re: [PATCH]: Handling spurious page fault for hugetlb region for 2.6.14-rc4-git5

[Adam Litke <agl@us.ibm.com>]

On Tue, 2005-10-18 at 21:07 -0700, Andrew Morton wrote:
> Rohit Seth <rohit.seth@intel.com> wrote:
> >
> > The prefetching problem is handled OK for regular pages because we can
> >  handle page faults corresponding to those pages.  That is currently not
> >  true for hugepages.  Currently the kernel assumes that PAGE_FAULT
> >  happening against a hugetlb page is caused by truncate and returns
> >  SIGBUS.
> 
> Doh.  No fault handler.  The penny finally drops.
> 
> Adam, I think this patch is temporary?

Yeah, looks like it can be dropped now since we handle this case (in
hugetlb_fault) as:

>         size = i_size_read(mapping->host) >> HPAGE_SHIFT;
>         if (idx >= size)
>                 goto backout;
> 
>         ret = VM_FAULT_MINOR;
>         if (!pte_none(*pte))
>                 goto backout;

-- 
Adam Litke - (agl at us.ibm.com)
IBM Linux Technology Center