linux-hardening - Re: [PATCH 3/3] usercopy: Make usercopy resilient against ridiculously large copies

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Mon, 13 Jun 2022 11:57:17 +0200
From:   Uladzislau Rezki <urezki@...il.com>
To:     "Matthew Wilcox (Oracle)" <willy@...radead.org>
Cc:     Kees Cook <keescook@...omium.org>, linux-mm@...ck.org,
        Uladzislau Rezki <urezki@...il.com>,
        Zorro Lang <zlang@...hat.com>, linux-xfs@...r.kernel.org,
        linux-hardening@...r.kernel.org
Subject: Re: [PATCH 3/3] usercopy: Make usercopy resilient against
 ridiculously large copies

> If 'n' is so large that it's negative, we might wrap around and mistakenly
> think that the copy is OK when it's not.  Such a copy would probably
> crash, but just doing the arithmetic in a more simple way lets us detect
> and refuse this case.
> 
> Signed-off-by: Matthew Wilcox (Oracle) <willy@...radead.org>
> ---
>  mm/usercopy.c | 19 +++++++++----------
>  1 file changed, 9 insertions(+), 10 deletions(-)
> 
> diff --git a/mm/usercopy.c b/mm/usercopy.c
> index 31deee7dd2f5..ff16083cf1c8 100644
> --- a/mm/usercopy.c
> +++ b/mm/usercopy.c
> @@ -162,20 +162,18 @@ static inline void check_heap_object(const void *ptr, unsigned long n,
>  				     bool to_user)
>  {
>  	uintptr_t addr = (uintptr_t)ptr;
> +	unsigned long offset;
>  	struct folio *folio;
>  
>  	if (is_kmap_addr(ptr)) {
> -		unsigned long page_end = addr | (PAGE_SIZE - 1);
> -
> -		if (addr + n - 1 > page_end)
> -			usercopy_abort("kmap", NULL, to_user,
> -					offset_in_page(ptr), n);
> +		offset = offset_in_page(ptr);
> +		if (n > PAGE_SIZE - offset)
> +			usercopy_abort("kmap", NULL, to_user, offset, n);
>  		return;
>  	}
>  
>  	if (is_vmalloc_addr(ptr)) {
>  		struct vmap_area *area = find_vmap_area(addr);
> -		unsigned long offset;
>  
>  		if (!area) {
>  			usercopy_abort("vmalloc", "no area", to_user, 0, n);
> @@ -184,9 +182,10 @@ static inline void check_heap_object(const void *ptr, unsigned long n,
>  
>  		/* XXX: We should also abort for free vmap_areas */
>  
> -		offset = addr - area->va_start;
> -		if (addr + n > area->va_end)
> +		if (n > area->va_end - addr) {
> +			offset = addr - area->va_start;
>  			usercopy_abort("vmalloc", NULL, to_user, offset, n);
> +		}
>  		return;
>  	}
>  
> @@ -199,8 +198,8 @@ static inline void check_heap_object(const void *ptr, unsigned long n,
>  		/* Check slab allocator for flags and size. */
>  		__check_heap_object(ptr, n, folio_slab(folio), to_user);
>  	} else if (folio_test_large(folio)) {
> -		unsigned long offset = ptr - folio_address(folio);
> -		if (offset + n > folio_size(folio))
> +		offset = ptr - folio_address(folio);
> +		if (n > folio_size(folio) - offset)
>  			usercopy_abort("page alloc", NULL, to_user, offset, n);
>  	}
>  }
> -- 
> 2.35.1
> 
Looks good to me: Reviewed-by: Uladzislau Rezki (Sony) <urezki@...il.com>

--
Uladzislau Rezki