cgroup: don't queue css_release_work if one already pending

Syzbot found a corrupted list bug scenario that can be triggered from
cgroup css_create(). The reproduces writes to cgroup.subtree_control
file, which invokes cgroup_apply_control_enable(), css_create(), and
css_populate_dir(), which then randomly fails with a fault injected -ENOMEM.
In such scenario the css_create() error path rcu enqueues css_free_rwork_fn
work for an css->refcnt initialized with css_release() destructor,
and there is a chance that the css_release() function will be invoked
for a cgroup_subsys_state, for which a destroy_work has already been
queued via css_create() error path. This causes a list_add corruption
as can be seen in the syzkaller report [1].
This can be avoided by adding a check to css_release() that checks
if it has already been enqueued.

[1] https://syzkaller.appspot.com/bug?id=e26e54d6eac9d9fb50b221ec3e4627b327465dbd

Cc: Tejun Heo <tj@kernel.org>
Cc: Zefan Li <lizefan.x@bytedance.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Christian Brauner <brauner@kernel.org>
Cc: Alexei Starovoitov <ast@kernel.org>
Cc: Daniel Borkmann <daniel@iogearbox.net>
Cc: Andrii Nakryiko <andrii@kernel.org>
Cc: Martin KaFai Lau <kafai@fb.com>
Cc: Song Liu <songliubraving@fb.com>
Cc: Yonghong Song <yhs@fb.com>
Cc: John Fastabend <john.fastabend@gmail.com>
Cc: KP Singh <kpsingh@kernel.org>
Cc: <cgroups@vger.kernel.org>
Cc: <netdev@vger.kernel.org>
Cc: <bpf@vger.kernel.org>
Cc: <stable@vger.kernel.org>
Cc: <linux-kernel@vger.kernel.org>

Reported-by: syzbot+e42ae441c3b10acf9e9d@syzkaller.appspotmail.com
Fixes: 8f36aaec9c92 ("cgroup: Use rcu_work instead of explicit rcu and work item")
Signed-off-by: Tadeusz Struk <tadeusz.struk@linaro.org>
---
 kernel/cgroup/cgroup.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

Hello Tadeusz.

Thanks for analyzing this syzbot report. Let me provide my understanding
of the test case and explanation why I think your patch fixes it but is
not fully correct.

On Tue, Apr 12, 2022 at 12:24:59PM -0700, Tadeusz Struk <tadeusz.struk@linaro.org> wrote:
> Syzbot found a corrupted list bug scenario that can be triggered from
> cgroup css_create(). The reproduces writes to cgroup.subtree_control
> file, which invokes cgroup_apply_control_enable(), css_create(), and
> css_populate_dir(), which then randomly fails with a fault injected -ENOMEM.

The reproducer code makes it hard for me to understand which function
fails with ENOMEM.
But I can see your patch fixes the reproducer and your additional debug
patch which proves that css->destroy_work is re-queued.

> In such scenario the css_create() error path rcu enqueues css_free_rwork_fn
> work for an css->refcnt initialized with css_release() destructor,

Note that css_free_rwork_fn() utilizes css->destroy_*r*work.
The error path in css_create() open codes relevant parts of
css_release_work_fn() so that css_release() can be skipped and the
refcnt is eventually just percpu_ref_exit()'d.

> and there is a chance that the css_release() function will be invoked
> for a cgroup_subsys_state, for which a destroy_work has already been
> queued via css_create() error path.

But I think the problem is css_populate_dir() failing in
cgroup_apply_control_enable(). (Is this what you actually meant?
css_create() error path is then irrelevant, no?)

The already created csses should then be rolled back via 
	cgroup_restore_control(cgrp);
	cgroup_apply_control_disable(cgrp);
	   ...
	   kill_css(css)

I suspect the double-queuing is a result of the fact that there exists
only the single reference to the css->refcnt. I.e. it's
percpu_ref_kill_and_confirm()'d and released both at the same time.

(Normally (when not killing the last reference), css->destroy_work reuse
is not a problem because of the sequenced chain
css_killed_work_fn()->css_put()->css_release().)

> This can be avoided by adding a check to css_release() that checks
> if it has already been enqueued.

If that's what's happening, then your patch omits the final
css_release_work_fn() in favor of css_killed_work_fn() but both should
be run during the rollback upon css_populate_dir() failure.

So an alternative approach to tackle this situation would be to split
css->destroy_work into two work work_structs (one for killing, one for
releasing) at the cost of inflating cgroup_subsys_state.

Take my hypothesis with a grain of salt maybe the assumption (last
reference == initial reference) is not different from normal operation.

Regards,
Michal

Hi Michal,
Thanks for your analysis.

On 4/14/22 09:44, Michal Koutný wrote:
> Hello Tadeusz.
> 
> Thanks for analyzing this syzbot report. Let me provide my understanding
> of the test case and explanation why I think your patch fixes it but is
> not fully correct.
> 
> On Tue, Apr 12, 2022 at 12:24:59PM -0700, Tadeusz Struk <tadeusz.struk@linaro.org> wrote:
>> Syzbot found a corrupted list bug scenario that can be triggered from
>> cgroup css_create(). The reproduces writes to cgroup.subtree_control
>> file, which invokes cgroup_apply_control_enable(), css_create(), and
>> css_populate_dir(), which then randomly fails with a fault injected -ENOMEM.
> 
> The reproducer code makes it hard for me to understand which function
> fails with ENOMEM.
> But I can see your patch fixes the reproducer and your additional debug
> patch which proves that css->destroy_work is re-queued.

Yes, it is hard to see the actual failing point because, I think it is randomly
failing in different places. I think in the actual case that causes the list
corruption is in fact in css_create().
It is the css_create() error path that does fist rcu enqueue in:

https://elixir.bootlin.com/linux/v5.10.109/source/kernel/cgroup/cgroup.c#L5228

and the second is triggered by the css->refcnt calling css_release()

The reason why we don't see it actually failing in css_create() in the trace
dump is that the fail_dump() is rate-limited, see:
https://elixir.bootlin.com/linux/v5.18-rc2/source/lib/fault-inject.c#L44

I was confused as well, so I put additional debug prints in every place
where css_release() can fail, and it was actually in
css_create()->cgroup_idr_alloc() that failed in my case.

What happened was, the write triggered:
cgroup_subtree_control_write()->cgroup_apply_control()->cgroup_apply_control_enable()->css_create()

which, allocates and initializes the css, then fails in cgroup_idr_alloc(),
bails out and calls queue_rcu_work(cgroup_destroy_wq, &css->destroy_rwork);

then cgroup_subtree_control_write() bails out to out_unlock:, which then goes:

cgroup_kn_unlock()->cgroup_put()->css_put()->percpu_ref_put(&css->refcnt)->percpu_ref_put_many(ref)

which then calls ref->data->release(ref) and enqueues the same
&css->destroy_rwork on cgroup_destroy_wq causing list corruption in insert_work.

>> In such scenario the css_create() error path rcu enqueues css_free_rwork_fn
>> work for an css->refcnt initialized with css_release() destructor,
> 
> Note that css_free_rwork_fn() utilizes css->destroy_*r*work.
> The error path in css_create() open codes relevant parts of
> css_release_work_fn() so that css_release() can be skipped and the
> refcnt is eventually just percpu_ref_exit()'d.
> 
>> and there is a chance that the css_release() function will be invoked
>> for a cgroup_subsys_state, for which a destroy_work has already been
>> queued via css_create() error path.
> 
> But I think the problem is css_populate_dir() failing in
> cgroup_apply_control_enable(). (Is this what you actually meant?
> css_create() error path is then irrelevant, no?)

I thought so too at first as the the crushdump shows that this is failing
in css_populate_dir(), but this is not the fail that causes the list corruption.
The code can recover from the fail in css_populate_dir().
The fail that causes trouble is in css_create(), that makes it go to its error path.
I can dig out the patch with my debug prints and request syzbot to run it
if you want.

> 
> The already created csses should then be rolled back via
> 	cgroup_restore_control(cgrp);
> 	cgroup_apply_control_disable(cgrp);
> 	   ...
> 	   kill_css(css)
> 
> I suspect the double-queuing is a result of the fact that there exists
> only the single reference to the css->refcnt. I.e. it's
> percpu_ref_kill_and_confirm()'d and released both at the same time.
> 
> (Normally (when not killing the last reference), css->destroy_work reuse
> is not a problem because of the sequenced chain
> css_killed_work_fn()->css_put()->css_release().)
> 
>> This can be avoided by adding a check to css_release() that checks
>> if it has already been enqueued.
> 
> If that's what's happening, then your patch omits the final
> css_release_work_fn() in favor of css_killed_work_fn() but both should
> be run during the rollback upon css_populate_dir() failure.

This change only prevents from double queue:

queue_[rcu]_work(cgroup_destroy_wq, &css->destroy_rwork);

I don't see how it affects the css_killed_work_fn() clean path.
I didn't look at it, since I thought it is irrelevant in this case.

Hello,

On Thu, Apr 14, 2022 at 10:51:18AM -0700, Tadeusz Struk wrote:
> What happened was, the write triggered:
> cgroup_subtree_control_write()->cgroup_apply_control()->cgroup_apply_control_enable()->css_create()
> 
> which, allocates and initializes the css, then fails in cgroup_idr_alloc(),
> bails out and calls queue_rcu_work(cgroup_destroy_wq, &css->destroy_rwork);

Yes, but this css hasn't been installed yet.

> then cgroup_subtree_control_write() bails out to out_unlock:, which then goes:
> 
> cgroup_kn_unlock()->cgroup_put()->css_put()->percpu_ref_put(&css->refcnt)->percpu_ref_put_many(ref)

And this is a different css. cgroup->self which isn't connected to the half
built css which got destroyed in css_create().

So, I have a bit of difficulty following this scenario. The way that the
current code uses destroy_work is definitely nasty and it'd probably be a
good idea to separate out the different use cases, but let's first
understand what's failing.

Thanks.

On Thu, Apr 21, 2022 at 02:00:56PM -1000, Tejun Heo <tj@kernel.org> wrote:
> If this is the case, we need to hold an extra reference to be put by the
> css_killed_work_fn(), right?

I looked into it a bit more lately and found that there already is such
a fuse in kill_css() [1].

At the same type syzbots stack trace demonstrates the fuse is
ineffective

> css_release+0xae/0xc0 kernel/cgroup/cgroup.c:5146                    (**)
> percpu_ref_put_many include/linux/percpu-refcount.h:322 [inline]
> percpu_ref_put include/linux/percpu-refcount.h:338 [inline]
> percpu_ref_call_confirm_rcu lib/percpu-refcount.c:162 [inline]        (*)
> percpu_ref_switch_to_atomic_rcu+0x5a2/0x5b0 lib/percpu-refcount.c:199
> rcu_do_batch+0x4f8/0xbc0 kernel/rcu/tree.c:2485
> rcu_core+0x59b/0xe30 kernel/rcu/tree.c:2722
> rcu_core_si+0x9/0x10 kernel/rcu/tree.c:2735
> __do_softirq+0x27e/0x596 kernel/softirq.c:305

(*) this calls css_killed_ref_fn confirm_switch
(**) zero references after confirmed kill?

So, I was also looking at the possible race with css_free_rwork_fn()
(from failed css_create()) but that would likely emit a warning from
__percpu_ref_exit().

So, I still think there's something fishy (so far possible only via
artificial ENOMEM injection) that needs an explanation...

Michal

[1] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/kernel/cgroup/cgroup.c#n5608