On 30 March 2018 at 16:54, Mike Kravetz <mike.kravetz@xxxxxxxxxx> wrote:
> This is a fix for a regression in 32 bit kernels caused by an
> invalid check for pgoff overflow in hugetlbfs mmap setup. The
> check incorrectly specified that the size of a loff_t was the
> same as the size of a long. The regression prevents mapping
> hugetlbfs files at offsets greater than 4GB on 32 bit kernels.
>
> On 32 bit kernels conversion from a page based unsigned long can
> not overflow a loff_t byte offset. Therefore, skip this check
> if sizeof(unsigned long) != sizeof(loff_t).
>
> Fixes: 63489f8e8211 ("hugetlbfs: check for pgoff value overflow")
> Cc: <stable@xxxxxxxxxxxxxxx>
> Reported-by: Dan Rue <dan.rue@xxxxxxxxxx>
> Signed-off-by: Mike Kravetz <mike.kravetz@xxxxxxxxxx>
Tested-by: Anders Roxell <anders.roxell@xxxxxxxxxx>
> ---
> fs/hugetlbfs/inode.c | 10 +++++++---
> 1 file changed, 7 insertions(+), 3 deletions(-)
>
> diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c
> index b9a254dcc0e7..d508c7844681 100644
> --- a/fs/hugetlbfs/inode.c
> +++ b/fs/hugetlbfs/inode.c
> @@ -138,10 +138,14 @@ static int hugetlbfs_file_mmap(struct file *file, struct vm_area_struct *vma)
>
> /*
> * page based offset in vm_pgoff could be sufficiently large to
> - * overflow a (l)off_t when converted to byte offset.
> + * overflow a loff_t when converted to byte offset. This can
> + * only happen on architectures where sizeof(loff_t) ==
> + * sizeof(unsigned long). So, only check in those instances.
> */
> - if (vma->vm_pgoff & PGOFF_LOFFT_MAX)
> - return -EINVAL;
> + if (sizeof(unsigned long) == sizeof(loff_t)) {
> + if (vma->vm_pgoff & PGOFF_LOFFT_MAX)
> + return -EINVAL;
> + }
>
> /* must be huge page aligned */
> if (vma->vm_pgoff & (~huge_page_mask(h) >> PAGE_SHIFT))
> --
> 2.13.6
>