2019-05-28 20:11:01

by Miguel Ojeda

Subject: [PATCH] mm: Fail when offset == num in first check of vm_map_pages_zero()

If the user asks us for offset == num, we should already fail in the
first check, i.e. the one testing for offsets beyond the object.

At the moment, we are failing on the second test anyway,
since count cannot be 0. Still, to agree with the comment of the first
test, we should first there.

Signed-off-by: Miguel Ojeda <[email protected]>
---
mm/memory.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/mm/memory.c b/mm/memory.c
index ddf20bd0c317..74cf8b0ce353 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -1547,7 +1547,7 @@ static int __vm_map_pages(struct vm_area_struct *vma, struct page **pages,
int ret, i;

/* Fail if the user requested offset is beyond the end of the object */
- if (offset > num)
+ if (offset >= num)
return -ENXIO;

/* Fail if the user requested size exceeds available object size */
--
2.17.1

2019-05-29 07:10:59

by Souptick Joarder

Subject: Re: [PATCH] mm: Fail when offset == num in first check of vm_map_pages_zero()

On Wed, May 29, 2019 at 1:38 AM Miguel Ojeda
<[email protected]> wrote:
>
> If the user asks us for offset == num, we should already fail in the
> first check, i.e. the one testing for offsets beyond the object.
>
> At the moment, we are failing on the second test anyway,
> since count cannot be 0. Still, to agree with the comment of the first
> test, we should first there.

I think, we need to cc linux-mm.
>
> Signed-off-by: Miguel Ojeda <[email protected]>
> ---
> mm/memory.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/mm/memory.c b/mm/memory.c
> index ddf20bd0c317..74cf8b0ce353 100644
> --- a/mm/memory.c
> +++ b/mm/memory.c
> @@ -1547,7 +1547,7 @@ static int __vm_map_pages(struct vm_area_struct *vma, struct page **pages,
> int ret, i;
>
> /* Fail if the user requested offset is beyond the end of the object */
> - if (offset > num)
> + if (offset >= num)
> return -ENXIO;
>
> /* Fail if the user requested size exceeds available object size */
> --
> 2.17.1
>

2019-06-01 10:42:28

by Miguel Ojeda

Subject: Re: [PATCH] mm: Fail when offset == num in first check of vm_map_pages_zero()

On Wed, May 29, 2019 at 9:09 AM Souptick Joarder <[email protected]> wrote:
>
> On Wed, May 29, 2019 at 1:38 AM Miguel Ojeda
> <[email protected]> wrote:
> >
> > If the user asks us for offset == num, we should already fail in the
> > first check, i.e. the one testing for offsets beyond the object.
> >
> > At the moment, we are failing on the second test anyway,
> > since count cannot be 0. Still, to agree with the comment of the first
> > test, we should first there.
>
> I think, we need to cc linux-mm.

Cc'ing Andrew as well as Souptick suggested me.

Cheers,
Miguel