diff mbox series

[RFC,06/10] kfence, kasan: make KFENCE compatible with KASAN

Message ID 20200907134055.2878499-7-elver@google.com
State New
Headers show
Series KFENCE: A low-overhead sampling-based memory safety error detector | expand

Commit Message

Marco Elver Sept. 7, 2020, 1:40 p.m. UTC
From: Alexander Potapenko <glider@google.com>

We make KFENCE compatible with KASAN for testing KFENCE itself. In
particular, KASAN helps to catch any potential corruptions to KFENCE
state, or other corruptions that may be a result of freepointer
corruptions in the main allocators.

To indicate that the combination of the two is generally discouraged,
CONFIG_EXPERT=y should be set. It also gives us the nice property that
KFENCE will be build-tested by allyesconfig builds.

Co-developed-by: Marco Elver <elver@google.com>
Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Alexander Potapenko <glider@google.com>
---
 lib/Kconfig.kfence | 2 +-
 mm/kasan/common.c  | 7 +++++++
 2 files changed, 8 insertions(+), 1 deletion(-)

Comments

Dmitry Vyukov Sept. 11, 2020, 7:04 a.m. UTC | #1
On Mon, Sep 7, 2020 at 3:41 PM Marco Elver <elver@google.com> wrote:
>
> From: Alexander Potapenko <glider@google.com>
>
> We make KFENCE compatible with KASAN for testing KFENCE itself. In
> particular, KASAN helps to catch any potential corruptions to KFENCE
> state, or other corruptions that may be a result of freepointer
> corruptions in the main allocators.
>
> To indicate that the combination of the two is generally discouraged,
> CONFIG_EXPERT=y should be set. It also gives us the nice property that
> KFENCE will be build-tested by allyesconfig builds.
>
> Co-developed-by: Marco Elver <elver@google.com>
> Signed-off-by: Marco Elver <elver@google.com>
> Signed-off-by: Alexander Potapenko <glider@google.com>
> ---
>  lib/Kconfig.kfence | 2 +-
>  mm/kasan/common.c  | 7 +++++++
>  2 files changed, 8 insertions(+), 1 deletion(-)
>
> diff --git a/lib/Kconfig.kfence b/lib/Kconfig.kfence
> index 7ac91162edb0..b080e49e15d4 100644
> --- a/lib/Kconfig.kfence
> +++ b/lib/Kconfig.kfence
> @@ -10,7 +10,7 @@ config HAVE_ARCH_KFENCE_STATIC_POOL
>
>  menuconfig KFENCE
>         bool "KFENCE: low-overhead sampling-based memory safety error detector"
> -       depends on HAVE_ARCH_KFENCE && !KASAN && (SLAB || SLUB)
> +       depends on HAVE_ARCH_KFENCE && (!KASAN || EXPERT) && (SLAB || SLUB)
>         depends on JUMP_LABEL # To ensure performance, require jump labels
>         select STACKTRACE
>         help
> diff --git a/mm/kasan/common.c b/mm/kasan/common.c
> index 950fd372a07e..f5c49f0fdeff 100644
> --- a/mm/kasan/common.c
> +++ b/mm/kasan/common.c
> @@ -18,6 +18,7 @@
>  #include <linux/init.h>
>  #include <linux/kasan.h>
>  #include <linux/kernel.h>
> +#include <linux/kfence.h>
>  #include <linux/kmemleak.h>
>  #include <linux/linkage.h>
>  #include <linux/memblock.h>
> @@ -396,6 +397,9 @@ static bool __kasan_slab_free(struct kmem_cache *cache, void *object,
>         tagged_object = object;
>         object = reset_tag(object);
>
> +       if (is_kfence_address(object))
> +               return false;

Is this needed?
At least in the slab patch I see that we do :

if (kfence_free(objp)) {
  kmemleak_free_recursive(objp, cachep->flags);
  return;
}

before:

/* Put the object into the quarantine, don't touch it for now. */ /*
Put the object into the quarantine, don't touch it for now. */
if (kasan_slab_free(cachep, objp, _RET_IP_)) if
(kasan_slab_free(cachep, objp, _RET_IP_))
  return; return;


If it's not supposed to be triggered, it can make sense to replace
with BUG/WARN.


>         if (unlikely(nearest_obj(cache, virt_to_head_page(object), object) !=
>             object)) {
>                 kasan_report_invalid_free(tagged_object, ip);
> @@ -444,6 +448,9 @@ static void *__kasan_kmalloc(struct kmem_cache *cache, const void *object,
>         if (unlikely(object == NULL))
>                 return NULL;
>
> +       if (is_kfence_address(object))
> +               return (void *)object;
> +
>         redzone_start = round_up((unsigned long)(object + size),
>                                 KASAN_SHADOW_SCALE_SIZE);
>         redzone_end = round_up((unsigned long)object + cache->object_size,
> --
> 2.28.0.526.ge36021eeef-goog
>
Marco Elver Sept. 11, 2020, 1 p.m. UTC | #2
On Fri, 11 Sep 2020 at 09:05, Dmitry Vyukov <dvyukov@google.com> wrote:
>
> On Mon, Sep 7, 2020 at 3:41 PM Marco Elver <elver@google.com> wrote:
> >
> > From: Alexander Potapenko <glider@google.com>
> >
> > We make KFENCE compatible with KASAN for testing KFENCE itself. In
> > particular, KASAN helps to catch any potential corruptions to KFENCE
> > state, or other corruptions that may be a result of freepointer
> > corruptions in the main allocators.
> >
> > To indicate that the combination of the two is generally discouraged,
> > CONFIG_EXPERT=y should be set. It also gives us the nice property that
> > KFENCE will be build-tested by allyesconfig builds.
> >
> > Co-developed-by: Marco Elver <elver@google.com>
> > Signed-off-by: Marco Elver <elver@google.com>
> > Signed-off-by: Alexander Potapenko <glider@google.com>
> > ---
> >  lib/Kconfig.kfence | 2 +-
> >  mm/kasan/common.c  | 7 +++++++
> >  2 files changed, 8 insertions(+), 1 deletion(-)
> >
> > diff --git a/lib/Kconfig.kfence b/lib/Kconfig.kfence
> > index 7ac91162edb0..b080e49e15d4 100644
> > --- a/lib/Kconfig.kfence
> > +++ b/lib/Kconfig.kfence
> > @@ -10,7 +10,7 @@ config HAVE_ARCH_KFENCE_STATIC_POOL
> >
> >  menuconfig KFENCE
> >         bool "KFENCE: low-overhead sampling-based memory safety error detector"
> > -       depends on HAVE_ARCH_KFENCE && !KASAN && (SLAB || SLUB)
> > +       depends on HAVE_ARCH_KFENCE && (!KASAN || EXPERT) && (SLAB || SLUB)
> >         depends on JUMP_LABEL # To ensure performance, require jump labels
> >         select STACKTRACE
> >         help
> > diff --git a/mm/kasan/common.c b/mm/kasan/common.c
> > index 950fd372a07e..f5c49f0fdeff 100644
> > --- a/mm/kasan/common.c
> > +++ b/mm/kasan/common.c
> > @@ -18,6 +18,7 @@
> >  #include <linux/init.h>
> >  #include <linux/kasan.h>
> >  #include <linux/kernel.h>
> > +#include <linux/kfence.h>
> >  #include <linux/kmemleak.h>
> >  #include <linux/linkage.h>
> >  #include <linux/memblock.h>
> > @@ -396,6 +397,9 @@ static bool __kasan_slab_free(struct kmem_cache *cache, void *object,
> >         tagged_object = object;
> >         object = reset_tag(object);
> >
> > +       if (is_kfence_address(object))
> > +               return false;
>
> Is this needed?
> At least in the slab patch I see that we do :
>
> if (kfence_free(objp)) {
>   kmemleak_free_recursive(objp, cachep->flags);
>   return;
> }
>
> before:
>
> /* Put the object into the quarantine, don't touch it for now. */ /*
> Put the object into the quarantine, don't touch it for now. */
> if (kasan_slab_free(cachep, objp, _RET_IP_)) if
> (kasan_slab_free(cachep, objp, _RET_IP_))
>   return; return;
>
>
> If it's not supposed to be triggered, it can make sense to replace
> with BUG/WARN.

It is required for SLUB. For SLAB, it seems it might not be necessary.
Making the check in kasan/common.c conditional on the allocator seems
ugly, so I propose we keep it there.

Thanks,
-- Marco
diff mbox series

Patch

diff --git a/lib/Kconfig.kfence b/lib/Kconfig.kfence
index 7ac91162edb0..b080e49e15d4 100644
--- a/lib/Kconfig.kfence
+++ b/lib/Kconfig.kfence
@@ -10,7 +10,7 @@  config HAVE_ARCH_KFENCE_STATIC_POOL
 
 menuconfig KFENCE
 	bool "KFENCE: low-overhead sampling-based memory safety error detector"
-	depends on HAVE_ARCH_KFENCE && !KASAN && (SLAB || SLUB)
+	depends on HAVE_ARCH_KFENCE && (!KASAN || EXPERT) && (SLAB || SLUB)
 	depends on JUMP_LABEL # To ensure performance, require jump labels
 	select STACKTRACE
 	help
diff --git a/mm/kasan/common.c b/mm/kasan/common.c
index 950fd372a07e..f5c49f0fdeff 100644
--- a/mm/kasan/common.c
+++ b/mm/kasan/common.c
@@ -18,6 +18,7 @@ 
 #include <linux/init.h>
 #include <linux/kasan.h>
 #include <linux/kernel.h>
+#include <linux/kfence.h>
 #include <linux/kmemleak.h>
 #include <linux/linkage.h>
 #include <linux/memblock.h>
@@ -396,6 +397,9 @@  static bool __kasan_slab_free(struct kmem_cache *cache, void *object,
 	tagged_object = object;
 	object = reset_tag(object);
 
+	if (is_kfence_address(object))
+		return false;
+
 	if (unlikely(nearest_obj(cache, virt_to_head_page(object), object) !=
 	    object)) {
 		kasan_report_invalid_free(tagged_object, ip);
@@ -444,6 +448,9 @@  static void *__kasan_kmalloc(struct kmem_cache *cache, const void *object,
 	if (unlikely(object == NULL))
 		return NULL;
 
+	if (is_kfence_address(object))
+		return (void *)object;
+
 	redzone_start = round_up((unsigned long)(object + size),
 				KASAN_SHADOW_SCALE_SIZE);
 	redzone_end = round_up((unsigned long)object + cache->object_size,