Re: [PATCH] RDMA/uverbs: Remove flexible arrays from struct *_filter

[Kees Cook]

On Sun, Feb 11, 2024 at 12:58:56PM +0100, Erick Archer wrote:
> When a struct containing a flexible array is included in another struct,
> and there is a member after the struct-with-flex-array, there is a
> possibility of memory overlap. These cases must be audited [1]. See:
> 
> struct inner {
> 	...
> 	int flex[];
> };
> 
> struct outer {
> 	...
> 	struct inner header;
> 	int overlap;
> 	...
> };
> 
> This is the scenario for all the "struct *_filter" structures that are
> included in the following "struct ib_flow_spec_*" structures:
> 
> struct ib_flow_spec_eth
> struct ib_flow_spec_ib
> struct ib_flow_spec_ipv4
> struct ib_flow_spec_ipv6
> struct ib_flow_spec_tcp_udp
> struct ib_flow_spec_tunnel
> struct ib_flow_spec_esp
> struct ib_flow_spec_gre
> struct ib_flow_spec_mpls
> 
> The pattern is like the one shown below:
> 
> struct *_filter {
> 	...
> 	u8 real_sz[];
> };
> 
> struct ib_flow_spec_mpls {
> 	...
> 	struct *_filter val;
> 	struct *_filter mask;
> };
> 
> In this case, the trailing flexible array "real_sz" is never allocated
> and is only used to calculate the size of the structures. Here the use
> of the "offsetof" helper can be changed by the "sizeof" operator because
> the goal is to get the size of these structures. Therefore, the trailing
> flexible arrays can also be removed.
> 
> Link: https://github.com/KSPP/linux/issues/202 [1]
> Signed-off-by: Erick Archer <erick.archer@xxxxxxx>
> ---
> Hi everyone,
> 
> This patch has not been tested. This has only been built-tested.

I might suggest doing a binary difference comparison[1], as it's possible
that "real_sz" is being used to try to avoid trailing padding on
structs. I wasn't able to trivially construct an example, so maybe I'm
not understanding its purpose correctly.

If, however, there are cases where offsetof(..., real_sz) !=
sizeof(...), then I would check two alternatives:

	struct { } real_sz;

but that may induce padding still, or:

	u8 real_sz[0];

which would be a literally zero-sized array, used only for addressing.

Or, these can be left as-is, and the "flex array not at end of struct"
warnings can be disabled for these targets.

-Kees

-- 
Kees Cook