Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

[PATCH] nommu: get_user_pages(): pin last page on non-page-aligned start

1 view
Skip to first unread message

Steven J. Magnani

unread,
Mar 2, 2010, 1:40:03 PM3/2/10
to
The noMMU version of get_user_pages() fails to pin the last page
when the start address isn't page-aligned. The patch fixes this in a way
that makes find_extend_vma() congruent to its MMU cousin.

Signed-off-by: Steven J. Magnani <st...@digidescorp.com>
---
diff -uprN a/mm/nommu.c b/mm/nommu.c
--- a/mm/nommu.c 2010-03-01 16:37:31.000000000 -0600
+++ b/mm/nommu.c 2010-03-01 21:34:49.000000000 -0600
@@ -146,7 +146,7 @@ int __get_user_pages(struct task_struct
(VM_MAYREAD | VM_MAYWRITE) : (VM_READ | VM_WRITE);

for (i = 0; i < nr_pages; i++) {
- vma = find_vma(mm, start);
+ vma = find_extend_vma(mm, start);
if (!vma)
goto finish_or_fault;

@@ -764,7 +764,7 @@ EXPORT_SYMBOL(find_vma);
*/
struct vm_area_struct *find_extend_vma(struct mm_struct *mm, unsigned long addr)
{
- return find_vma(mm, addr);
+ return find_vma(mm, addr & PAGE_MASK);
}

/*

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majo...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

Paul Mundt

unread,
Mar 2, 2010, 10:10:01 PM3/2/10
to
On Tue, Mar 02, 2010 at 12:29:44PM -0600, Steven J. Magnani wrote:
> The noMMU version of get_user_pages() fails to pin the last page
> when the start address isn't page-aligned. The patch fixes this in a way
> that makes find_extend_vma() congruent to its MMU cousin.
>
> Signed-off-by: Steven J. Magnani <st...@digidescorp.com>

Looks good to me. David?

Acked-by: Paul Mundt <let...@linux-sh.org>

David Howells

unread,
Mar 8, 2010, 4:40:03 AM3/8/10
to
Paul Mundt <let...@linux-sh.org> wrote:

> > The noMMU version of get_user_pages() fails to pin the last page
> > when the start address isn't page-aligned. The patch fixes this in a way
> > that makes find_extend_vma() congruent to its MMU cousin.
> >
> > Signed-off-by: Steven J. Magnani <st...@digidescorp.com>
>
> Looks good to me. David?

I don't seem to have the original patch.

David

0 new messages