Re: [PATCH v2] mm: memcontrol: fix use after free in mem_cgroup_iter()

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Fri 26-07-19 10:12:47, Miles Chen wrote:
> This patch is sent to report an use after free in mem_cgroup_iter()
> after merging commit: be2657752e9e "mm: memcg: fix use after free in
> mem_cgroup_iter()".
> 
> I work with android kernel tree (4.9 & 4.14), and the commit:
> be2657752e9e "mm: memcg: fix use after free in mem_cgroup_iter()" has
> been merged to the trees. However, I can still observe use after free
> issues addressed in the commit be2657752e9e.
> (on low-end devices, a few times this month)
> 
> backtrace:
> 	css_tryget <- crash here
> 	mem_cgroup_iter
> 	shrink_node
> 	shrink_zones
> 	do_try_to_free_pages
> 	try_to_free_pages
> 	__perform_reclaim
> 	__alloc_pages_direct_reclaim
> 	__alloc_pages_slowpath
> 	__alloc_pages_nodemask
> 
> To debug, I poisoned mem_cgroup before freeing it:
> 
> static void __mem_cgroup_free(struct mem_cgroup *memcg)
> 	for_each_node(node)
> 	free_mem_cgroup_per_node_info(memcg, node);
> 	free_percpu(memcg->stat);
> +       /* poison memcg before freeing it */
> +       memset(memcg, 0x78, sizeof(struct mem_cgroup));
> 	kfree(memcg);
> }
> 
> The coredump shows the position=0xdbbc2a00 is freed.
> 
> (gdb) p/x ((struct mem_cgroup_per_node *)0xe5009e00)->iter[8]
> $13 = {position = 0xdbbc2a00, generation = 0x2efd}
> 
> 0xdbbc2a00:     0xdbbc2e00      0x00000000      0xdbbc2800      0x00000100
> 0xdbbc2a10:     0x00000200      0x78787878      0x00026218      0x00000000
> 0xdbbc2a20:     0xdcad6000      0x00000001      0x78787800      0x00000000
> 0xdbbc2a30:     0x78780000      0x00000000      0x0068fb84      0x78787878
> 0xdbbc2a40:     0x78787878      0x78787878      0x78787878      0xe3fa5cc0
> 0xdbbc2a50:     0x78787878      0x78787878      0x00000000      0x00000000
> 0xdbbc2a60:     0x00000000      0x00000000      0x00000000      0x00000000
> 0xdbbc2a70:     0x00000000      0x00000000      0x00000000      0x00000000
> 0xdbbc2a80:     0x00000000      0x00000000      0x00000000      0x00000000
> 0xdbbc2a90:     0x00000001      0x00000000      0x00000000      0x00100000
> 0xdbbc2aa0:     0x00000001      0xdbbc2ac8      0x00000000      0x00000000
> 0xdbbc2ab0:     0x00000000      0x00000000      0x00000000      0x00000000
> 0xdbbc2ac0:     0x00000000      0x00000000      0xe5b02618      0x00001000
> 0xdbbc2ad0:     0x00000000      0x78787878      0x78787878      0x78787878
> 0xdbbc2ae0:     0x78787878      0x78787878      0x78787878      0x78787878
> 0xdbbc2af0:     0x78787878      0x78787878      0x78787878      0x78787878
> 0xdbbc2b00:     0x78787878      0x78787878      0x78787878      0x78787878
> 0xdbbc2b10:     0x78787878      0x78787878      0x78787878      0x78787878
> 0xdbbc2b20:     0x78787878      0x78787878      0x78787878      0x78787878
> 0xdbbc2b30:     0x78787878      0x78787878      0x78787878      0x78787878
> 0xdbbc2b40:     0x78787878      0x78787878      0x78787878      0x78787878
> 0xdbbc2b50:     0x78787878      0x78787878      0x78787878      0x78787878
> 0xdbbc2b60:     0x78787878      0x78787878      0x78787878      0x78787878
> 0xdbbc2b70:     0x78787878      0x78787878      0x78787878      0x78787878
> 0xdbbc2b80:     0x78787878      0x78787878      0x00000000      0x78787878
> 0xdbbc2b90:     0x78787878      0x78787878      0x78787878      0x78787878
> 0xdbbc2ba0:     0x78787878      0x78787878      0x78787878      0x78787878
> 
> In the reclaim path, try_to_free_pages() does not setup
> sc.target_mem_cgroup and sc is passed to do_try_to_free_pages(), ...,
> shrink_node().
> 
> In mem_cgroup_iter(), root is set to root_mem_cgroup because
> sc->target_mem_cgroup is NULL.
> It is possible to assign a memcg to root_mem_cgroup.nodeinfo.iter in
> mem_cgroup_iter().
> 
> 	try_to_free_pages
> 		struct scan_control sc = {...}, target_mem_cgroup is 0x0;
> 	do_try_to_free_pages
> 	shrink_zones
> 	shrink_node
> 		 mem_cgroup *root = sc->target_mem_cgroup;
> 		 memcg = mem_cgroup_iter(root, NULL, &reclaim);
> 	mem_cgroup_iter()
> 		if (!root)
> 			root = root_mem_cgroup;
> 		...
> 
> 		css = css_next_descendant_pre(css, &root->css);
> 		memcg = mem_cgroup_from_css(css);
> 		cmpxchg(&iter->position, pos, memcg);
> 
> My device uses memcg non-hierarchical mode.
> When we release a memcg: invalidate_reclaim_iterators() reaches only
> dead_memcg and its parents. If non-hierarchical mode is used,
> invalidate_reclaim_iterators() never reaches root_mem_cgroup.
> 
> static void invalidate_reclaim_iterators(struct mem_cgroup *dead_memcg)
> {
> 	struct mem_cgroup *memcg = dead_memcg;
> 
> 	for (; memcg; memcg = parent_mem_cgroup(memcg)
> 	...
> }
> 
> So the use after free scenario looks like:
> 
> CPU1						CPU2
> 
> try_to_free_pages
> do_try_to_free_pages
> shrink_zones
> shrink_node
> mem_cgroup_iter()
>     if (!root)
>     	root = root_mem_cgroup;
>     ...
>     css = css_next_descendant_pre(css, &root->css);
>     memcg = mem_cgroup_from_css(css);
>     cmpxchg(&iter->position, pos, memcg);
> 
> 					invalidate_reclaim_iterators(memcg);
> 					...
> 					__mem_cgroup_free()
> 						kfree(memcg);
> 
> try_to_free_pages
> do_try_to_free_pages
> shrink_zones
> shrink_node
> mem_cgroup_iter()
>     if (!root)
>     	root = root_mem_cgroup;
>     ...
>     mz = mem_cgroup_nodeinfo(root, reclaim->pgdat->node_id);
>     iter = &mz->iter[reclaim->priority];
>     pos = READ_ONCE(iter->position);
>     css_tryget(&pos->css) <- use after free

Thanks for the write up. This is really useful.

> To avoid this, we should also invalidate root_mem_cgroup.nodeinfo.iter in
> invalidate_reclaim_iterators().

I am sorry, I didn't get to comment an earlier version but I am
wondering whether it makes more sense to do and explicit invalidation.

[...]
> +static void invalidate_reclaim_iterators(struct mem_cgroup *dead_memcg)
> +{
> +	struct mem_cgroup *memcg = dead_memcg;
> +	int invalidate_root = 0;
> +
> +	for (; memcg; memcg = parent_mem_cgroup(memcg))
> +		__invalidate_reclaim_iterators(memcg, dead_memcg);

	/* here goes your comment */
	if (!dead_memcg->use_hierarchy)
		__invalidate_reclaim_iterators(root_mem_cgroup,	dead_memcg);
> +
> +}

Other than that the patch looks good to me.

Acked-by: Michal Hocko <mhocko@xxxxxxxx>
-- 
Michal Hocko
SUSE Labs




[Index of Archives]     [Linux ARM Kernel]     [Linux ARM]     [Linux Omap]     [Fedora ARM]     [IETF Annouce]     [Bugtraq]     [Linux OMAP]     [Linux MIPS]     [eCos]     [Asterisk Internet PBX]     [Linux API]

  Powered by Linux