[mm] kasan: avoid resetting aux_lock

  From: Andrey Konovalov <andreyknvl@gmail.com>

With commit 63b85ac56a64 ("kasan: stop leaking stack trace handles"),
KASAN zeroes out alloc meta when an object is freed. The zeroed out data
purposefully includes alloc and auxiliary stack traces but also
accidentally includes aux_lock.

As aux_lock is only initialized for each object slot during slab
creation, when the freed slot is reallocated, saving auxiliary stack
traces for the new object leads to lockdep reports when taking the
zeroed out aux_lock.

Arguably, we could reinitialize aux_lock when the object is reallocated,
but a simpler solution is to avoid zeroing out aux_lock when an object
gets freed.

Reported-by: Paul E. McKenney <paulmck@kernel.org>
Closes: https://lore.kernel.org/linux-next/5cc0f83c-e1d6-45c5-be89-9b86746fe731@paulmck-laptop/
Fixes: 63b85ac56a64 ("kasan: stop leaking stack trace handles")
Signed-off-by: Andrey Konovalov <andreyknvl@gmail.com>
---
 mm/kasan/generic.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)
  

On Tue, 9 Jan 2024 at 23:12, <andrey.konovalov@linux.dev> wrote:
>
> From: Andrey Konovalov <andreyknvl@gmail.com>
>
> With commit 63b85ac56a64 ("kasan: stop leaking stack trace handles"),
> KASAN zeroes out alloc meta when an object is freed. The zeroed out data
> purposefully includes alloc and auxiliary stack traces but also
> accidentally includes aux_lock.
>
> As aux_lock is only initialized for each object slot during slab
> creation, when the freed slot is reallocated, saving auxiliary stack
> traces for the new object leads to lockdep reports when taking the
> zeroed out aux_lock.
>
> Arguably, we could reinitialize aux_lock when the object is reallocated,
> but a simpler solution is to avoid zeroing out aux_lock when an object
> gets freed.
>
> Reported-by: Paul E. McKenney <paulmck@kernel.org>
> Closes: https://lore.kernel.org/linux-next/5cc0f83c-e1d6-45c5-be89-9b86746fe731@paulmck-laptop/
> Fixes: 63b85ac56a64 ("kasan: stop leaking stack trace handles")
> Signed-off-by: Andrey Konovalov <andreyknvl@gmail.com>

Reviewed-by: Marco Elver <elver@google.com>

> ---
>  mm/kasan/generic.c | 10 ++++++++--
>  1 file changed, 8 insertions(+), 2 deletions(-)
>
> diff --git a/mm/kasan/generic.c b/mm/kasan/generic.c
> index 24c13dfb1e94..df6627f62402 100644
> --- a/mm/kasan/generic.c
> +++ b/mm/kasan/generic.c
> @@ -487,6 +487,7 @@ void kasan_init_object_meta(struct kmem_cache *cache, const void *object)
>                 __memset(alloc_meta, 0, sizeof(*alloc_meta));
>
>                 /*
> +                * Prepare the lock for saving auxiliary stack traces.
>                  * Temporarily disable KASAN bug reporting to allow instrumented
>                  * raw_spin_lock_init to access aux_lock, which resides inside
>                  * of a redzone.
> @@ -510,8 +511,13 @@ static void release_alloc_meta(struct kasan_alloc_meta *meta)
>         stack_depot_put(meta->aux_stack[0]);
>         stack_depot_put(meta->aux_stack[1]);
>
> -       /* Zero out alloc meta to mark it as invalid. */
> -       __memset(meta, 0, sizeof(*meta));
> +       /*
> +        * Zero out alloc meta to mark it as invalid but keep aux_lock
> +        * initialized to avoid having to reinitialize it when another object
> +        * is allocated in the same slot.
> +        */
> +       __memset(&meta->alloc_track, 0, sizeof(meta->alloc_track));
> +       __memset(meta->aux_stack, 0, sizeof(meta->aux_stack));
>  }
>
>  static void release_free_meta(const void *object, struct kasan_free_meta *meta)
> --
> 2.25.1
>
  

On Tue, Jan 09, 2024 at 11:12:34PM +0100, andrey.konovalov@linux.dev wrote:
> From: Andrey Konovalov <andreyknvl@gmail.com>
> 
> With commit 63b85ac56a64 ("kasan: stop leaking stack trace handles"),
> KASAN zeroes out alloc meta when an object is freed. The zeroed out data
> purposefully includes alloc and auxiliary stack traces but also
> accidentally includes aux_lock.
> 
> As aux_lock is only initialized for each object slot during slab
> creation, when the freed slot is reallocated, saving auxiliary stack
> traces for the new object leads to lockdep reports when taking the
> zeroed out aux_lock.
> 
> Arguably, we could reinitialize aux_lock when the object is reallocated,
> but a simpler solution is to avoid zeroing out aux_lock when an object
> gets freed.
> 
> Reported-by: Paul E. McKenney <paulmck@kernel.org>
> Closes: https://lore.kernel.org/linux-next/5cc0f83c-e1d6-45c5-be89-9b86746fe731@paulmck-laptop/
> Fixes: 63b85ac56a64 ("kasan: stop leaking stack trace handles")
> Signed-off-by: Andrey Konovalov <andreyknvl@gmail.com>

Very good!

Tested-by: Paul E. McKenney <paulmck@kernel.org>

> ---
>  mm/kasan/generic.c | 10 ++++++++--
>  1 file changed, 8 insertions(+), 2 deletions(-)
> 
> diff --git a/mm/kasan/generic.c b/mm/kasan/generic.c
> index 24c13dfb1e94..df6627f62402 100644
> --- a/mm/kasan/generic.c
> +++ b/mm/kasan/generic.c
> @@ -487,6 +487,7 @@ void kasan_init_object_meta(struct kmem_cache *cache, const void *object)
>  		__memset(alloc_meta, 0, sizeof(*alloc_meta));
>  
>  		/*
> +		 * Prepare the lock for saving auxiliary stack traces.
>  		 * Temporarily disable KASAN bug reporting to allow instrumented
>  		 * raw_spin_lock_init to access aux_lock, which resides inside
>  		 * of a redzone.
> @@ -510,8 +511,13 @@ static void release_alloc_meta(struct kasan_alloc_meta *meta)
>  	stack_depot_put(meta->aux_stack[0]);
>  	stack_depot_put(meta->aux_stack[1]);
>  
> -	/* Zero out alloc meta to mark it as invalid. */
> -	__memset(meta, 0, sizeof(*meta));
> +	/*
> +	 * Zero out alloc meta to mark it as invalid but keep aux_lock
> +	 * initialized to avoid having to reinitialize it when another object
> +	 * is allocated in the same slot.
> +	 */
> +	__memset(&meta->alloc_track, 0, sizeof(meta->alloc_track));
> +	__memset(meta->aux_stack, 0, sizeof(meta->aux_stack));
>  }
>  
>  static void release_free_meta(const void *object, struct kasan_free_meta *meta)
> -- 
> 2.25.1
>