In a high-load arm64 environment, the pcrypt_aead01 test in LTP can lead to
system UAF (Use-After-Free) issues. Due to the lengthy analysis of the
pcrypt_aead01 function call, I'll describe the problem scenario using a
simplified model:
Suppose there's a user of padata named `user_function` that adheres to
the padata requirement of calling `padata_free_shell` after `serial()`
has been invoked, as demonstrated in the following code:
```c
struct request {
struct padata_priv padata;
struct completion *done;
};
void parallel(struct padata_priv *padata) {
do_something();
}
void serial(struct padata_priv *padata) {
struct request *request = container_of(padata, struct request, padata);
complete(request->done);
}
void user_function() {
DECLARE_COMPLETION(done)
padata->parallel = parallel;
padata->serial = serial;
padata_do_parallel();
wait_for_completion(&done);
padata_free_shell();
}
```
In the corresponding padata.c file, there's the following code:
```c
static void padata_serial_worker(struct work_struct *serial_work) {
...
cnt = 0;
while (!list_empty(&local_list)) {
...
padata->serial(padata);
cnt++;
}
local_bh_enable();
if (refcount_sub_and_test(cnt, &pd->refcnt))
padata_free_pd(pd);
}
```
Because of the high system load and the accumulation of unexecuted
softirq at this moment, `local_bh_enable()` in padata takes longer
to execute than usual. Subsequently, when accessing `pd->refcnt`,
`pd` has already been released by `padata_free_shell()`, resulting
in a UAF issue with `pd->refcnt`.
The fix is straightforward: add `refcount_dec_and_test` before calling
`padata_free_pd` in `padata_free_shell`.
Signed-off-by: Wang Jinchao <[email protected]>
---
V2:
To satisfy Sparse, use rcu_dereference_protected.
Reported-by: kernel test robot <[email protected]>
Closes: https://lore.kernel.org/oe-kbuild-all/[email protected]/
V1: https://lore.kernel.org/all/ZRE4XvOOhz4HSOgR@fedora/
kernel/padata.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/kernel/padata.c b/kernel/padata.c
index 222d60195de6..acef1e703a8b 100644
--- a/kernel/padata.c
+++ b/kernel/padata.c
@@ -1107,7 +1107,9 @@ void padata_free_shell(struct padata_shell *ps)
mutex_lock(&ps->pinst->lock);
list_del(&ps->list);
- padata_free_pd(rcu_dereference_protected(ps->pd, 1));
+ struct parallel_data *pd = rcu_dereference_protected(ps->pd, 1);
+ if (refcount_dec_and_test(&pd->refcnt))
+ padata_free_pd(pd);
mutex_unlock(&ps->pinst->lock);
kfree(ps);
--
2.40.0
On Wed, Oct 04, 2023 at 10:54:29AM -0400, Daniel Jordan wrote:
> On Thu, Sep 28, 2023 at 08:38:53AM +0800, Wang Jinchao wrote:
> > In a high-load arm64 environment, the pcrypt_aead01 test in LTP can lead to
> > system UAF (Use-After-Free) issues. Due to the lengthy analysis of the
> > pcrypt_aead01 function call, I'll describe the problem scenario using a
> > simplified model:
> >
> > Suppose there's a user of padata named `user_function` that adheres to
> > the padata requirement of calling `padata_free_shell` after `serial()`
> > has been invoked, as demonstrated in the following code:
> >
> > ```c
> > struct request {
> > struct padata_priv padata;
> > struct completion *done;
> > };
> >
> > void parallel(struct padata_priv *padata) {
> > do_something();
> > }
> >
> > void serial(struct padata_priv *padata) {
> > struct request *request = container_of(padata, struct request, padata);
> > complete(request->done);
> > }
> >
> > void user_function() {
> > DECLARE_COMPLETION(done)
> > padata->parallel = parallel;
> > padata->serial = serial;
> > padata_do_parallel();
> > wait_for_completion(&done);
> > padata_free_shell();
> > }
> > ```
> >
> > In the corresponding padata.c file, there's the following code:
> >
> > ```c
> > static void padata_serial_worker(struct work_struct *serial_work) {
> > ...
> > cnt = 0;
> >
> > while (!list_empty(&local_list)) {
> > ...
> > padata->serial(padata);
> > cnt++;
> > }
> >
> > local_bh_enable();
> >
> > if (refcount_sub_and_test(cnt, &pd->refcnt))
> > padata_free_pd(pd);
> > }
> > ```
> >
> > Because of the high system load and the accumulation of unexecuted
> > softirq at this moment, `local_bh_enable()` in padata takes longer
> > to execute than usual. Subsequently, when accessing `pd->refcnt`,
> > `pd` has already been released by `padata_free_shell()`, resulting
> > in a UAF issue with `pd->refcnt`.
> >
> > The fix is straightforward: add `refcount_dec_and_test` before calling
> > `padata_free_pd` in `padata_free_shell`.
>
> This could use a Fixes tag. From Nicolai's patch[0] we agreed on
>
> Fixes: 07928d9bfc81 ("padata: Remove broken queue flushing")
>
> With that,
>
> Acked-by: Daniel Jordan <[email protected]>
>
> Thanks!
>
> [0] https://lore.kernel.org/all/[email protected]/
>
> > Signed-off-by: Wang Jinchao <[email protected]>
> > ---
> > V2:
> > To satisfy Sparse, use rcu_dereference_protected.
> > Reported-by: kernel test robot <[email protected]>
> > Closes: https://lore.kernel.org/oe-kbuild-all/[email protected]/
> >
> > V1: https://lore.kernel.org/all/ZRE4XvOOhz4HSOgR@fedora/
> >
> > kernel/padata.c | 4 +++-
> > 1 file changed, 3 insertions(+), 1 deletion(-)
> >
> > diff --git a/kernel/padata.c b/kernel/padata.c
> > index 222d60195de6..acef1e703a8b 100644
> > --- a/kernel/padata.c
> > +++ b/kernel/padata.c
> > @@ -1107,7 +1107,9 @@ void padata_free_shell(struct padata_shell *ps)
> >
> > mutex_lock(&ps->pinst->lock);
> > list_del(&ps->list);
> > - padata_free_pd(rcu_dereference_protected(ps->pd, 1));
> > + struct parallel_data *pd = rcu_dereference_protected(ps->pd, 1);
Oh, I should've also said please move the declaration of pd to the top
of the function like scripts/checkpatch.pl suggests.
> > + if (refcount_dec_and_test(&pd->refcnt))
> > + padata_free_pd(pd);
> > mutex_unlock(&ps->pinst->lock);
> >
> > kfree(ps);
> > --
> > 2.40.0
> >