Re: [PATCH] x86: do not overrun page table ranges in gup

[Hugh Dickins]

On Mon, 28 Jul 2008, Linus Torvalds wrote:
> On Mon, 28 Jul 2008, Linus Torvalds wrote:
> > 
> > So no, the right fix would be to align 'start' first, which means that 
> > everything else (including 'end') will be page-aligned. Aligning just one 
> > or the other is very very wrong.
> 
> Does this work?
> 
> Not pretty, but it stands _some_ chance of being correct.
> 
> 		Linus

Acked-by: Hugh Dickins <hugh@xxxxxxxxxxx>

Seeing the bug report, I was working up a patch along the same lines
as this; but got the access_ok() issue backwards: yours looks right.
I see Alexey has tested it, confess I haven't.

> 
> 
> ---
>  arch/x86/mm/gup.c |    9 ++++++---
>  1 files changed, 6 insertions(+), 3 deletions(-)
> 
> diff --git a/arch/x86/mm/gup.c b/arch/x86/mm/gup.c
> index 3085f25..007bb06 100644
> --- a/arch/x86/mm/gup.c
> +++ b/arch/x86/mm/gup.c
> @@ -223,14 +223,17 @@ int get_user_pages_fast(unsigned long start, int nr_pages, int write,
>  			struct page **pages)
>  {
>  	struct mm_struct *mm = current->mm;
> -	unsigned long end = start + (nr_pages << PAGE_SHIFT);
> -	unsigned long addr = start;
> +	unsigned long addr, len, end;
>  	unsigned long next;
>  	pgd_t *pgdp;
>  	int nr = 0;
>  
> +	start &= PAGE_MASK;
> +	addr = start;
> +	len = (unsigned long) nr_pages << PAGE_SHIFT;
> +	end = start + len;
>  	if (unlikely(!access_ok(write ? VERIFY_WRITE : VERIFY_READ,
> -					start, nr_pages*PAGE_SIZE)))
> +					start, len)))
>  		goto slow_irqon;
>  
>  	/*
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/