RE: [PATCH 01/26] compiler: introduce noinline_for_kasan annotation

[David Laight]

From: Andrey Ryabinin
> Sent: 03 March 2017 13:50
...
> noinline_iff_kasan might be a better name.  noinline_for_kasan gives the 
> impression
> that we always noinline function for the sake of kasan, while 
> noinline_iff_kasan
> clearly indicates that function is noinline only if kasan is used.

noinline_if_stackbloat

David

Re: [PATCH 01/26] compiler: introduce noinline_for_kasan annotation

[Arnd Bergmann]

On Fri, Mar 3, 2017 at 3:33 PM, Alexander Potapenko  wrote:
> On Fri, Mar 3, 2017 at 3:30 PM, Arnd Bergmann  wrote:
>> On Fri, Mar 3, 2017 at 2:55 PM, Alexander Potapenko  
>> wrote:
>>
>> Would KMSAN also force local variables to be non-overlapping the way that
>> asan-stack=1 and -fsanitize-address-use-after-scope do? As I understood it,
>> KMSAN would add extra code for maintaining the uninit bits, but in an example
>> like this
> The thing is that KMSAN (and other tools that insert heavyweight
> instrumentation) may cause heavy register spilling which will also
> blow up the stack frames.

In that case, I would expect a mostly distinct set of functions to have large
stack frames with KMSAN, compared to the ones that need
noinline_for_kasan. In most cases I patched, the called inline function is
actually trivial, but invoked many times from the same caller.

 Arnd

Re: [PATCH 01/26] compiler: introduce noinline_for_kasan annotation

[Alexander Potapenko]

On Fri, Mar 3, 2017 at 3:30 PM, Arnd Bergmann  wrote:
> On Fri, Mar 3, 2017 at 2:55 PM, Alexander Potapenko  wrote:
>> On Fri, Mar 3, 2017 at 2:50 PM, Andrey Ryabinin  
>> wrote:
>
 @@ -416,6 +416,17 @@ static __always_inline void 
 __write_once_size(volatile void *p, void *res, int s
   */
  #define noinline_for_stack noinline

 +/*
 + * CONFIG_KASAN can lead to extreme stack usage with certain patterns when
 + * one function gets inlined many times and each instance requires a stack
 + * ckeck.
 + */
 +#ifdef CONFIG_KASAN
 +#define noinline_for_kasan noinline __maybe_unused
>>>
>>>
>>> noinline_iff_kasan might be a better name.  noinline_for_kasan gives the 
>>> impression
>>> that we always noinline function for the sake of kasan, while 
>>> noinline_iff_kasan
>>> clearly indicates that function is noinline only if kasan is used.
>
> Fine with me. I actually tried to come up with a name that implies that the
> symbol is actually "inline" (or even __always_inline_ without KASAN, but
> couldn't think of any good name for it.
>
>> FWIW we may be facing the same problem with other compiler-based
>> tools, e.g. KMSAN (which isn't there yet).
>> So it might be better to choose a macro name that doesn't use the name 
>> "KASAN".
>> E.g. noinline_iff_memtool (or noinline_iff_memory_tool if that's not too 
>> long).
>> WDYT?
>
> Would KMSAN also force local variables to be non-overlapping the way that
> asan-stack=1 and -fsanitize-address-use-after-scope do? As I understood it,
> KMSAN would add extra code for maintaining the uninit bits, but in an example
> like this
The thing is that KMSAN (and other tools that insert heavyweight
instrumentation) may cause heavy register spilling which will also
blow up the stack frames.
> int f(int *);
> static inline __attribute__((always_inline)) int g(void)
> {
> int i;
> f();
> return i;
> }
> int f(void)
> {
>  return g()+g()+g()+g();
> }
>
> each of the four copies of 'i' could have the same location on the stack
> and get marked uninitialized again before calling f(). We only need
> noinline_for_kasan (whatever we end up calling that) for compiler
> features that force each instance of 'i' to have its own stack redzone.
>
>  Arnd



-- 
Alexander Potapenko
Software Engineer

Google Germany GmbH
Erika-Mann-Straße, 33
80636 München

Geschäftsführer: Matthew Scott Sucherman, Paul Terence Manicle
Registergericht und -nummer: Hamburg, HRB 86891
Sitz der Gesellschaft: Hamburg

Re: [PATCH 01/26] compiler: introduce noinline_for_kasan annotation

[Arnd Bergmann]

On Fri, Mar 3, 2017 at 2:55 PM, Alexander Potapenko  wrote:
> On Fri, Mar 3, 2017 at 2:50 PM, Andrey Ryabinin  
> wrote:

>>> @@ -416,6 +416,17 @@ static __always_inline void __write_once_size(volatile 
>>> void *p, void *res, int s
>>>   */
>>>  #define noinline_for_stack noinline
>>>
>>> +/*
>>> + * CONFIG_KASAN can lead to extreme stack usage with certain patterns when
>>> + * one function gets inlined many times and each instance requires a stack
>>> + * ckeck.
>>> + */
>>> +#ifdef CONFIG_KASAN
>>> +#define noinline_for_kasan noinline __maybe_unused
>>
>>
>> noinline_iff_kasan might be a better name.  noinline_for_kasan gives the 
>> impression
>> that we always noinline function for the sake of kasan, while 
>> noinline_iff_kasan
>> clearly indicates that function is noinline only if kasan is used.

Fine with me. I actually tried to come up with a name that implies that the
symbol is actually "inline" (or even __always_inline_ without KASAN, but
couldn't think of any good name for it.

> FWIW we may be facing the same problem with other compiler-based
> tools, e.g. KMSAN (which isn't there yet).
> So it might be better to choose a macro name that doesn't use the name 
> "KASAN".
> E.g. noinline_iff_memtool (or noinline_iff_memory_tool if that's not too 
> long).
> WDYT?

Would KMSAN also force local variables to be non-overlapping the way that
asan-stack=1 and -fsanitize-address-use-after-scope do? As I understood it,
KMSAN would add extra code for maintaining the uninit bits, but in an example
like this

int f(int *);
static inline __attribute__((always_inline)) int g(void)
{
int i;
f();
return i;
}
int f(void)
{
 return g()+g()+g()+g();
}

each of the four copies of 'i' could have the same location on the stack
and get marked uninitialized again before calling f(). We only need
noinline_for_kasan (whatever we end up calling that) for compiler
features that force each instance of 'i' to have its own stack redzone.

 Arnd

Re: [PATCH 01/26] compiler: introduce noinline_for_kasan annotation

[Alexander Potapenko]

On Fri, Mar 3, 2017 at 2:50 PM, Andrey Ryabinin  wrote:
>
>
> On 03/02/2017 07:38 PM, Arnd Bergmann wrote:
>> When CONFIG_KASAN is set, we can run into some code that uses incredible
>> amounts of kernel stack:
>>
>> drivers/staging/dgnc/dgnc_neo.c:1056:1: error: the frame size of 2 bytes 
>> is larger than 2048 bytes [-Werror=frame-larger-than=]
>> drivers/media/i2c/cx25840/cx25840-core.c:4960:1: error: the frame size of 
>> 94000 bytes is larger than 2048 bytes [-Werror=frame-larger-than=]
>> drivers/media/dvb-frontends/stv090x.c:3430:1: error: the frame size of 5312 
>> bytes is larger than 3072 bytes [-Werror=frame-larger-than=]
>>
>> This happens when a sanitizer uses stack memory each time an inline function
>> gets called. This introduces a new annotation for those functions to make
>> them either 'inline' or 'noinline' dependning on the CONFIG_KASAN symbol.
>>
>> Signed-off-by: Arnd Bergmann 
>> ---
>>  include/linux/compiler.h | 11 +++
>>  1 file changed, 11 insertions(+)
>>
>> diff --git a/include/linux/compiler.h b/include/linux/compiler.h
>> index f8110051188f..56b90897a459 100644
>> --- a/include/linux/compiler.h
>> +++ b/include/linux/compiler.h
>> @@ -416,6 +416,17 @@ static __always_inline void __write_once_size(volatile 
>> void *p, void *res, int s
>>   */
>>  #define noinline_for_stack noinline
>>
>> +/*
>> + * CONFIG_KASAN can lead to extreme stack usage with certain patterns when
>> + * one function gets inlined many times and each instance requires a stack
>> + * ckeck.
>> + */
>> +#ifdef CONFIG_KASAN
>> +#define noinline_for_kasan noinline __maybe_unused
>
>
> noinline_iff_kasan might be a better name.  noinline_for_kasan gives the 
> impression
> that we always noinline function for the sake of kasan, while 
> noinline_iff_kasan
> clearly indicates that function is noinline only if kasan is used.
FWIW we may be facing the same problem with other compiler-based
tools, e.g. KMSAN (which isn't there yet).
So it might be better to choose a macro name that doesn't use the name "KASAN".
E.g. noinline_iff_memtool (or noinline_iff_memory_tool if that's not too long).
WDYT?
>> +#else
>> +#define noinline_for_kasan inline
>> +#endif
>> +
>>  #ifndef __always_inline
>>  #define __always_inline inline
>>  #endif
>>
>
> --
> You received this message because you are subscribed to the Google Groups 
> "kasan-dev" group.
> To unsubscribe from this group and stop receiving emails from it, send an 
> email to kasan-dev+unsubscr...@googlegroups.com.
> To post to this group, send email to kasan-...@googlegroups.com.
> To view this discussion on the web visit 
> https://groups.google.com/d/msgid/kasan-dev/7e7a62de-3b79-6044-72fa-4ade418953d1%40virtuozzo.com.
> For more options, visit https://groups.google.com/d/optout.



-- 
Alexander Potapenko
Software Engineer

Google Germany GmbH
Erika-Mann-Straße, 33
80636 München

Geschäftsführer: Matthew Scott Sucherman, Paul Terence Manicle
Registergericht und -nummer: Hamburg, HRB 86891
Sitz der Gesellschaft: Hamburg

Re: [PATCH 01/26] compiler: introduce noinline_for_kasan annotation

[Andrey Ryabinin]

On 03/02/2017 07:38 PM, Arnd Bergmann wrote:
> When CONFIG_KASAN is set, we can run into some code that uses incredible
> amounts of kernel stack:
> 
> drivers/staging/dgnc/dgnc_neo.c:1056:1: error: the frame size of 2 bytes 
> is larger than 2048 bytes [-Werror=frame-larger-than=]
> drivers/media/i2c/cx25840/cx25840-core.c:4960:1: error: the frame size of 
> 94000 bytes is larger than 2048 bytes [-Werror=frame-larger-than=]
> drivers/media/dvb-frontends/stv090x.c:3430:1: error: the frame size of 5312 
> bytes is larger than 3072 bytes [-Werror=frame-larger-than=]
> 
> This happens when a sanitizer uses stack memory each time an inline function
> gets called. This introduces a new annotation for those functions to make
> them either 'inline' or 'noinline' dependning on the CONFIG_KASAN symbol.
> 
> Signed-off-by: Arnd Bergmann 
> ---
>  include/linux/compiler.h | 11 +++
>  1 file changed, 11 insertions(+)
> 
> diff --git a/include/linux/compiler.h b/include/linux/compiler.h
> index f8110051188f..56b90897a459 100644
> --- a/include/linux/compiler.h
> +++ b/include/linux/compiler.h
> @@ -416,6 +416,17 @@ static __always_inline void __write_once_size(volatile 
> void *p, void *res, int s
>   */
>  #define noinline_for_stack noinline
>  
> +/*
> + * CONFIG_KASAN can lead to extreme stack usage with certain patterns when
> + * one function gets inlined many times and each instance requires a stack
> + * ckeck.
> + */
> +#ifdef CONFIG_KASAN
> +#define noinline_for_kasan noinline __maybe_unused


noinline_iff_kasan might be a better name.  noinline_for_kasan gives the 
impression
that we always noinline function for the sake of kasan, while noinline_iff_kasan
clearly indicates that function is noinline only if kasan is used.

> +#else
> +#define noinline_for_kasan inline
> +#endif
> +
>  #ifndef __always_inline
>  #define __always_inline inline
>  #endif
>