LinuxLists.cc - [PATCH v3 0/3] mm/slub: some debug enhancements

Hi Feng,

On Mon, Aug 15, 2022 at 03:27:43PM +0800, Feng Tang wrote:
> Hi Oliver,
>
> Could you help to check if the below combined patch fix the problem
> you reported? thanks!

I applied below patch upon 3616799128:
28b34693c816e9 (linux-devel/fixup-3616799128) fix for 3616799128: BUG_kmalloc-#(Not_tainted):kmalloc_Redzone_overwritten
3616799128612e (linux-review/Feng-Tang/mm-slub-some-debug-enhancements/20220727-151318) mm/slub: extend redzone check to cover extra allocated kmalloc space than requested
acc77d62f91ccc mm/slub: only zero the requested size of buffer for kzalloc

confirmed the issue gone:

=========================================================================================
compiler/kconfig/rootfs/sleep/tbox_group/testcase:
gcc-11/x86_64-randconfig-a005-20220117/debian-11.1-x86_64-20220510.cgz/300/vm-snb/boot

acc77d62f91ccca2 3616799128612e04ed919579e2c 28b34693c816e9fcbe42bdd341e
---------------- --------------------------- ---------------------------
fail:runs %reproduction fail:runs %reproduction fail:runs
| | | | |
:20 95% 19:20 0% :22 dmesg.BUG_kmalloc-#(Not_tainted):kmalloc_Redzone_overwritten
:20 95% 19:20 0% :22 dmesg.BUG_kmalloc-#(Tainted:G_B):kmalloc_Redzone_overwritten

>
> - Feng
>
> ---
>
> diff --git a/include/linux/kasan.h b/include/linux/kasan.h
> index b092277bf48d6..293bdaa0ba09c 100644
> --- a/include/linux/kasan.h
> +++ b/include/linux/kasan.h
> @@ -100,6 +100,8 @@ static inline bool kasan_has_integrated_init(void)
> struct kasan_cache {
> int alloc_meta_offset;
> int free_meta_offset;
> + /* size of free_meta data saved in object's data area */
> + int free_meta_size_in_object;
> bool is_kmalloc;
> };
>
> diff --git a/mm/kasan/common.c b/mm/kasan/common.c
> index c40c0e7b3b5f1..9d2994dbe4e7a 100644
> --- a/mm/kasan/common.c
> +++ b/mm/kasan/common.c
> @@ -200,6 +200,8 @@ void __kasan_cache_create(struct kmem_cache *cache, unsigned int *size,
> cache->kasan_info.free_meta_offset = KASAN_NO_FREE_META;
> *size = ok_size;
> }
> + } else {
> + cache->kasan_info.free_meta_size_in_object = sizeof(struct kasan_free_meta);
> }
>
> /* Calculate size with optimal redzone. */
> diff --git a/mm/slub.c b/mm/slub.c
> index added2653bb03..272dcdbaaa03b 100644
> --- a/mm/slub.c
> +++ b/mm/slub.c
> @@ -830,6 +830,16 @@ static inline void set_orig_size(struct kmem_cache *s,
> if (!slub_debug_orig_size(s))
> return;
>
> +#ifdef CONFIG_KASAN_GENERIC
> + /*
> + * kasn could save its free meta data in the start part of object
> + * area, so skip the redzone check if kasan's meta data size is
> + * bigger enough to possibly overlap with kmalloc redzone
> + */
> + if (s->kasan_info.free_meta_size_in_object * 2 > s->object_size)
> + orig_size = s->object_size;
> +#endif
> +
> p += get_info_end(s);
> p += sizeof(struct track) * 2;
>
> On Thu, Aug 04, 2022 at 08:22:31PM +0800, Tang, Feng wrote:
> > On Thu, Aug 04, 2022 at 06:47:58PM +0800, Dmitry Vyukov wrote:
> > > On Thu, 4 Aug 2022 at 08:29, Feng Tang <[email protected]> wrote:
> > [...]
> > > >
> > > > ---8<---
> > > > From c4fc739ea4d5222f0aba4b42b59668d64a010082 Mon Sep 17 00:00:00 2001
> > > > From: Feng Tang <[email protected]>
> > > > Date: Thu, 4 Aug 2022 13:25:35 +0800
> > > > Subject: [PATCH] mm: kasan: Add free_meta size info in struct kasan_cache
> > > >
> > > > When kasan is enabled for slab/slub, it may save kasan' free_meta
> > > > data in the former part of slab object data area in slab object
> > > > free path, which works fine.
> > > >
> > > > There is ongoing effort to extend slub's debug function which will
> > > > redzone the latter part of kmalloc object area, and when both of
> > > > the debug are enabled, there is possible conflict, especially when
> > > > the kmalloc object has small size, as caught by 0Day bot [1]
> > > >
> > > > For better information for slab/slub, add free_meta's data size
> > > > info 'kasan_cache', so that its users can take right action to
> > > > avoid data conflict.
> > > >
> > > > [1]. https://lore.kernel.org/lkml/YuYm3dWwpZwH58Hu@xsang-OptiPlex-9020/
> > > > Reported-by: kernel test robot <[email protected]>
> > > > Signed-off-by: Feng Tang <[email protected]>
> > >
> > > Acked-by: Dmitry Vyukov <[email protected]>
> >
> > Thanks for your suggestion and review!
> >
> > > I assume there will be a second patch that uses
> > > free_meta_size_in_object in slub debug code.
> >
> > Yes, it will be called in the slub kmalloc object redzone debug code.
> >
> > Thanks,
> > Feng
> >
> > > > ---
> > > > include/linux/kasan.h | 2 ++
> > > > mm/kasan/common.c | 2 ++
> > > > 2 files changed, 4 insertions(+)
> > > >
> > > > diff --git a/include/linux/kasan.h b/include/linux/kasan.h
> > > > index b092277bf48d..293bdaa0ba09 100644
> > > > --- a/include/linux/kasan.h
> > > > +++ b/include/linux/kasan.h
> > > > @@ -100,6 +100,8 @@ static inline bool kasan_has_integrated_init(void)
> > > > struct kasan_cache {
> > > > int alloc_meta_offset;
> > > > int free_meta_offset;
> > > > + /* size of free_meta data saved in object's data area */
> > > > + int free_meta_size_in_object;
> > > > bool is_kmalloc;
> > > > };
> > > >
> > > > diff --git a/mm/kasan/common.c b/mm/kasan/common.c
> > > > index 78be2beb7453..a627efa267d1 100644
> > > > --- a/mm/kasan/common.c
> > > > +++ b/mm/kasan/common.c
> > > > @@ -201,6 +201,8 @@ void __kasan_cache_create(struct kmem_cache *cache, unsigned int *size,
> > > > cache->kasan_info.free_meta_offset = KASAN_NO_FREE_META;
> > > > *size = ok_size;
> > > > }
> > > > + } else {
> > > > + cache->kasan_info.free_meta_size_in_object = sizeof(struct kasan_free_meta);
> > > > }
> > > >
> > > > /* Calculate size with optimal redzone. */
> > > > --
> > > > 2.27.0
> > >

2022-08-16 15:02:45

by Feng Tang

[permalink] [raw]

Subject: Re: [mm/slub] 3616799128: BUG_kmalloc-#(Not_tainted):kmalloc_Redzone_overwritten

On Tue, Aug 16, 2022 at 09:27:39PM +0800, Sang, Oliver wrote:
> Hi Feng,
>
> On Mon, Aug 15, 2022 at 03:27:43PM +0800, Feng Tang wrote:
> > Hi Oliver,
> >
> > Could you help to check if the below combined patch fix the problem
> > you reported? thanks!
>
> I applied below patch upon 3616799128:
> 28b34693c816e9 (linux-devel/fixup-3616799128) fix for 3616799128: BUG_kmalloc-#(Not_tainted):kmalloc_Redzone_overwritten
> 3616799128612e (linux-review/Feng-Tang/mm-slub-some-debug-enhancements/20220727-151318) mm/slub: extend redzone check to cover extra allocated kmalloc space than requested
> acc77d62f91ccc mm/slub: only zero the requested size of buffer for kzalloc
>
>
> confirmed the issue gone:

Many thanks for helping testing!

- Feng

> =========================================================================================
> compiler/kconfig/rootfs/sleep/tbox_group/testcase:
> gcc-11/x86_64-randconfig-a005-20220117/debian-11.1-x86_64-20220510.cgz/300/vm-snb/boot
>
>
> acc77d62f91ccca2 3616799128612e04ed919579e2c 28b34693c816e9fcbe42bdd341e
> ---------------- --------------------------- ---------------------------
> fail:runs %reproduction fail:runs %reproduction fail:runs
> | | | | |
> :20 95% 19:20 0% :22 dmesg.BUG_kmalloc-#(Not_tainted):kmalloc_Redzone_overwritten
> :20 95% 19:20 0% :22 dmesg.BUG_kmalloc-#(Tainted:G_B):kmalloc_Redzone_overwritten
>
>
>
> >
> > - Feng
> >
> > ---
> >
> > diff --git a/include/linux/kasan.h b/include/linux/kasan.h
> > index b092277bf48d6..293bdaa0ba09c 100644
> > --- a/include/linux/kasan.h
> > +++ b/include/linux/kasan.h
> > @@ -100,6 +100,8 @@ static inline bool kasan_has_integrated_init(void)
> > struct kasan_cache {
> > int alloc_meta_offset;
> > int free_meta_offset;
> > + /* size of free_meta data saved in object's data area */
> > + int free_meta_size_in_object;
> > bool is_kmalloc;
> > };
> >
> > diff --git a/mm/kasan/common.c b/mm/kasan/common.c
> > index c40c0e7b3b5f1..9d2994dbe4e7a 100644
> > --- a/mm/kasan/common.c
> > +++ b/mm/kasan/common.c
> > @@ -200,6 +200,8 @@ void __kasan_cache_create(struct kmem_cache *cache, unsigned int *size,
> > cache->kasan_info.free_meta_offset = KASAN_NO_FREE_META;
> > *size = ok_size;
> > }
> > + } else {
> > + cache->kasan_info.free_meta_size_in_object = sizeof(struct kasan_free_meta);
> > }
> >
> > /* Calculate size with optimal redzone. */
> > diff --git a/mm/slub.c b/mm/slub.c
> > index added2653bb03..272dcdbaaa03b 100644
> > --- a/mm/slub.c
> > +++ b/mm/slub.c
> > @@ -830,6 +830,16 @@ static inline void set_orig_size(struct kmem_cache *s,
> > if (!slub_debug_orig_size(s))
> > return;
> >
> > +#ifdef CONFIG_KASAN_GENERIC
> > + /*
> > + * kasn could save its free meta data in the start part of object
> > + * area, so skip the redzone check if kasan's meta data size is
> > + * bigger enough to possibly overlap with kmalloc redzone
> > + */
> > + if (s->kasan_info.free_meta_size_in_object * 2 > s->object_size)
> > + orig_size = s->object_size;
> > +#endif
> > +
> > p += get_info_end(s);
> > p += sizeof(struct track) * 2;
> >