Re: [PATCH] mm: shrinker: avoid memleak in alloc_shrinker_info

[chenridong]

On 2024/10/12 10:42, Muchun Song wrote:
> > On Oct 11, 2024, at 22:21, Chen Ridong <chenridong@xxxxxxxxxxxxxxx> wrote:
> >
> > From: Chen Ridong <chenridong@xxxxxxxxxx>
> >
> > A memleak was found as bellow:
> >
> > unreferenced object 0xffff8881010d2a80 (size 32):
> >   comm "mkdir", pid 1559, jiffies 4294932666
> >   hex dump (first 32 bytes):
> >     00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
> >     40 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  @...............
> >   backtrace (crc 2e7ef6fa):
> >     [<ffffffff81372754>] __kmalloc_node_noprof+0x394/0x470
> >     [<ffffffff813024ab>] alloc_shrinker_info+0x7b/0x1a0
> >     [<ffffffff813b526a>] mem_cgroup_css_online+0x11a/0x3b0
> >     [<ffffffff81198dd9>] online_css+0x29/0xa0
> >     [<ffffffff811a243d>] cgroup_apply_control_enable+0x20d/0x360
> >     [<ffffffff811a5728>] cgroup_mkdir+0x168/0x5f0
> >     [<ffffffff8148543e>] kernfs_iop_mkdir+0x5e/0x90
> >     [<ffffffff813dbb24>] vfs_mkdir+0x144/0x220
> >     [<ffffffff813e1c97>] do_mkdirat+0x87/0x130
> >     [<ffffffff813e1de9>] __x64_sys_mkdir+0x49/0x70
> >     [<ffffffff81f8c928>] do_syscall_64+0x68/0x140
> >     [<ffffffff8200012f>] entry_SYSCALL_64_after_hwframe+0x76/0x7e
> >
> > In the alloc_shrinker_info function, when shrinker_unit_alloc return
> > err, the info won't be freed. Just fix it.
> >
> > Fixes: 307bececcd12 ("mm: shrinker: add a secondary array for shrinker_info::{map, nr_deferred}")
> > Signed-off-by: Chen Ridong <chenridong@xxxxxxxxxx>
> > ---
> > mm/shrinker.c | 4 +++-
> > 1 file changed, 3 insertions(+), 1 deletion(-)
> >
> > diff --git a/mm/shrinker.c b/mm/shrinker.c
> > index dc5d2a6fcfc4..e4b795ee6d2e 100644
> > --- a/mm/shrinker.c
> > +++ b/mm/shrinker.c
> > @@ -87,8 +87,10 @@ int alloc_shrinker_info(struct mem_cgroup *memcg)
> > 		if (!info)
> > 			goto err;
> > 		info->map_nr_max = shrinker_nr_max;
> > - 		if (shrinker_unit_alloc(info, NULL, nid))
> > + 		if (shrinker_unit_alloc(info, NULL, nid)) {
> > + 			kvfree(info);
> > 			goto err;
>
> Since we already use goto to do the cleanup for failure, I don't like
> to mix two different ways to do the similar thing. I suggest adding
> the kvfree (BTW, it could handle NULL argument) to the label of err.
>
> Muchun,
> Thanks.
Thanks, I will update soon.

Best regards,
Ridong

> > + 		}
> > 		rcu_assign_pointer(memcg->nodeinfo[nid]->shrinker_info, info);
> > 	}
> > 	mutex_unlock(&shrinker_mutex);
> > --
> > 2.34.1