Re: [PATCH v4 0/3] KASAN support for RISC-V

Re: [PATCH v4 0/3] KASAN support for RISC-V

[Greentime Hu]

Nick Hu <nickhu@andestech.com> 於 2019年10月28日 週一 上午10:41寫道：
>
> KASAN is an important runtime memory debugging feature in linux kernel which can
> detect use-after-free and out-of-bounds problems.
>
> Changes in v2:
>   - Remove the porting of memmove and exclude the check instead.
>   - Fix some code noted by Christoph Hellwig
>
> Changes in v3:
>   - Update the KASAN documentation to mention that riscv is supported.
>
> Changes in v4:
>   - Correct the commit log
>   - Fix the bug reported by Greentime Hu
>
> Nick Hu (3):
>   kasan: No KASAN's memmove check if archs don't have it.
>   riscv: Add KASAN support
>   kasan: Add riscv to KASAN documentation.
>
>  Documentation/dev-tools/kasan.rst   |   4 +-
>  arch/riscv/Kconfig                  |   1 +
>  arch/riscv/include/asm/kasan.h      |  27 ++++++++
>  arch/riscv/include/asm/pgtable-64.h |   5 ++
>  arch/riscv/include/asm/string.h     |   9 +++
>  arch/riscv/kernel/head.S            |   3 +
>  arch/riscv/kernel/riscv_ksyms.c     |   2 +
>  arch/riscv/kernel/setup.c           |   5 ++
>  arch/riscv/kernel/vmlinux.lds.S     |   1 +
>  arch/riscv/lib/memcpy.S             |   5 +-
>  arch/riscv/lib/memset.S             |   5 +-
>  arch/riscv/mm/Makefile              |   6 ++
>  arch/riscv/mm/kasan_init.c          | 104 ++++++++++++++++++++++++++++
>  mm/kasan/common.c                   |   2 +
>  14 files changed, 173 insertions(+), 6 deletions(-)
>  create mode 100644 arch/riscv/include/asm/kasan.h
>  create mode 100644 arch/riscv/mm/kasan_init.c
>
Hi Nick,

I have tested KASAN feature with test_kasan.ko based on commit
cd9e72b80090a8cd7d84a47a30a06fa92ff277d1 (tag: riscv/for-v5.4-rc3) and
it passed in Qemu and Unleashed board.
Thank you for fixing the bug. :)

Tested-by: Greentime Hu <greentime.hu@sifive.com>

Re: [PATCH v4 1/3] kasan: No KASAN's memmove check if archs don't have it.

[Paul Walmsley]

Hello Andrey, Alexander, Dmitry,

On Mon, 28 Oct 2019, Nick Hu wrote:

> If archs don't have memmove then the C implementation from lib/string.c is used,
> and then it's instrumented by compiler. So there is no need to add KASAN's
> memmove to manual checks.
> 
> Signed-off-by: Nick Hu <nickhu@andestech.com>

If you're happy with this revision of this patch, could you please ack it 
so we can merge it as part of the RISC-V KASAN patch set? 

Or if you'd prefer to take this patch yourself, please let me know.


- Paul

> ---
>  mm/kasan/common.c | 2 ++
>  1 file changed, 2 insertions(+)
> 
> diff --git a/mm/kasan/common.c b/mm/kasan/common.c
> index 6814d6d6a023..897f9520bab3 100644
> --- a/mm/kasan/common.c
> +++ b/mm/kasan/common.c
> @@ -107,6 +107,7 @@ void *memset(void *addr, int c, size_t len)
>  	return __memset(addr, c, len);
>  }
>  
> +#ifdef __HAVE_ARCH_MEMMOVE
>  #undef memmove
>  void *memmove(void *dest, const void *src, size_t len)
>  {
> @@ -115,6 +116,7 @@ void *memmove(void *dest, const void *src, size_t len)
>  
>  	return __memmove(dest, src, len);
>  }
> +#endif
>  
>  #undef memcpy
>  void *memcpy(void *dest, const void *src, size_t len)
> -- 
> 2.17.0
> 
> 
> _______________________________________________
> linux-riscv mailing list
> linux-riscv@lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-riscv
> 


- Paul

Re: [PATCH v4 1/3] kasan: No KASAN's memmove check if archs don't have it.

[Dmitry Vyukov]

On Sun, Nov 17, 2019 at 5:58 AM Paul Walmsley <paul.walmsley@sifive.com> wrote:
>
> Hello Andrey, Alexander, Dmitry,
>
> On Mon, 28 Oct 2019, Nick Hu wrote:
>
> > If archs don't have memmove then the C implementation from lib/string.c is used,
> > and then it's instrumented by compiler. So there is no need to add KASAN's
> > memmove to manual checks.
> >
> > Signed-off-by: Nick Hu <nickhu@andestech.com>
>
> If you're happy with this revision of this patch, could you please ack it
> so we can merge it as part of the RISC-V KASAN patch set?
>
> Or if you'd prefer to take this patch yourself, please let me know.

Hi Paul,

Acked-by: Dmitry Vyukov <dvyukov@google.com>

We don't have separate tree for kasan. Merging this via RISC-V tree
should be fine.

Thanks

> -
>
> > ---
> >  mm/kasan/common.c | 2 ++
> >  1 file changed, 2 insertions(+)
> >
> > diff --git a/mm/kasan/common.c b/mm/kasan/common.c
> > index 6814d6d6a023..897f9520bab3 100644
> > --- a/mm/kasan/common.c
> > +++ b/mm/kasan/common.c
> > @@ -107,6 +107,7 @@ void *memset(void *addr, int c, size_t len)
> >       return __memset(addr, c, len);
> >  }
> >
> > +#ifdef __HAVE_ARCH_MEMMOVE
> >  #undef memmove
> >  void *memmove(void *dest, const void *src, size_t len)
> >  {
> > @@ -115,6 +116,7 @@ void *memmove(void *dest, const void *src, size_t len)
> >
> >       return __memmove(dest, src, len);
> >  }
> > +#endif
> >
> >  #undef memcpy
> >  void *memcpy(void *dest, const void *src, size_t len)
> > --
> > 2.17.0
> >
> >
> > _______________________________________________
> > linux-riscv mailing list
> > linux-riscv@lists.infradead.org
> > http://lists.infradead.org/mailman/listinfo/linux-riscv
> >
>
>
> - Paul

Re: [PATCH v4 1/3] kasan: No KASAN's memmove check if archs don't have it.

[Andrey Ryabinin]

On 10/28/19 5:40 AM, Nick Hu wrote:
> If archs don't have memmove then the C implementation from lib/string.c is used,
> and then it's instrumented by compiler. So there is no need to add KASAN's
> memmove to manual checks.
> 
> Signed-off-by: Nick Hu <nickhu@andestech.com>
> ---

Acked-by: Andrey Ryabinin <aryabinin@virtuozzo.com>