[4/5] Makefile: Enable -Warray-bounds

Commit Message

Kees Cook Aug. 18, 2021, 8:11 a.m. UTC

With the recent fixes for flexible arrays and expanded FORTIFY_SOURCE
coverage, it is now possible to enable -Warray-bounds. Since both
GCC and Clang include -Warray-bounds in -Wall, we just need to stop
disabling it.

Cc: Arnd Bergmann <[e-mail]>
Cc: Masahiro Yamada <[e-mail]>
Cc: [e-mail]
Co-developed-by: Gustavo A. R. Silva <[e-mail]>
Signed-off-by: Gustavo A. R. Silva <[e-mail]>
Signed-off-by: Kees Cook <[e-mail]>
---
 Makefile | 1 -
 1 file changed, 1 deletion(-)

Comments

Guenter Roeck Feb. 2, 2022, 4:09 p.m. UTC | #1

On Wed, Aug 18, 2021 at 01:11:17AM -0700, Kees Cook wrote:
> With the recent fixes for flexible arrays and expanded FORTIFY_SOURCE
> coverage, it is now possible to enable -Warray-bounds. Since both
> GCC and Clang include -Warray-bounds in -Wall, we just need to stop
> disabling it.
> 
> Cc: Arnd Bergmann <[e-mail]>
> Cc: Masahiro Yamada <[e-mail]>
> Cc: [e-mail]
> Co-developed-by: Gustavo A. R. Silva <[e-mail]>
> Signed-off-by: Gustavo A. R. Silva <[e-mail]>
> Signed-off-by: Kees Cook <[e-mail]>
> ---
>  Makefile | 1 -
>  1 file changed, 1 deletion(-)
> 
> diff --git a/Makefile b/Makefile
> index a4aca01a4835..af22b83cede7 100644
> --- a/Makefile
> +++ b/Makefile
> @@ -1072,7 +1072,6 @@ KBUILD_CFLAGS += $(call cc-disable-warning, stringop-truncation)
>  
>  # We'll want to enable this eventually, but it's not going away for 5.7 at least
>  KBUILD_CFLAGS += $(call cc-disable-warning, zero-length-bounds)
> -KBUILD_CFLAGS += -Wno-array-bounds
>  KBUILD_CFLAGS += $(call cc-disable-warning, stringop-overflow)
>  
>  # Another good warning that we'll want to enable eventually

alpha:defconfig:

In function '__memset',
    inlined from '__bad_pagetable' at arch/alpha/mm/init.c:79:2:
./arch/alpha/include/asm/string.h:37:32: error: '__builtin_memset' offset [0, 8191] is out of the bounds [0, 0] [-Werror=array-bounds]
   37 |                         return __builtin_memset(s, c, n);
      |                                ^~~~~~~~~~~~~~~~~~~~~~~~~
In function '__memset',
    inlined from '__bad_page' at arch/alpha/mm/init.c:86:2:
./arch/alpha/include/asm/string.h:37:32: error: '__builtin_memset' offset [0, 8191] is out of the bounds [0, 0] [-Werror=array-bounds]
   37 |                         return __builtin_memset(s, c, n);
      |                                ^~~~~~~~~~~~~~~~~~~~~~~~~
In function '__memset',
    inlined from 'paging_init' at arch/alpha/mm/init.c:256:2:
./arch/alpha/include/asm/string.h:37:32: error: '__builtin_memset' offset [0, 8191] is out of the bounds [0, 0] [-Werror=array-bounds]
   37 |                         return __builtin_memset(s, c, n);


xtensa:allmodconfig:
--------------
Error log:
In file included from include/linux/uaccess.h:11,
                 from include/linux/sched/task.h:11,
                 from arch/xtensa/kernel/process.c:21:
arch/xtensa/kernel/process.c: In function 'copy_thread':
arch/xtensa/kernel/process.c:262:52: error: array subscript 53 is above array bounds of 'long unsigned int[16]'

Kees Cook Feb. 2, 2022, 8:56 p.m. UTC | #2

On Wed, Feb 02, 2022 at 08:09:03AM -0800, Guenter Roeck wrote:
> On Wed, Aug 18, 2021 at 01:11:17AM -0700, Kees Cook wrote:
> > With the recent fixes for flexible arrays and expanded FORTIFY_SOURCE
> > coverage, it is now possible to enable -Warray-bounds. Since both
> > GCC and Clang include -Warray-bounds in -Wall, we just need to stop
> > disabling it.
> > 
> > Cc: Arnd Bergmann <[e-mail]>
> > Cc: Masahiro Yamada <[e-mail]>
> > Cc: [e-mail]
> > Co-developed-by: Gustavo A. R. Silva <[e-mail]>
> > Signed-off-by: Gustavo A. R. Silva <[e-mail]>
> > Signed-off-by: Kees Cook <[e-mail]>
> > ---
> >  Makefile | 1 -
> >  1 file changed, 1 deletion(-)
> > 
> > diff --git a/Makefile b/Makefile
> > index a4aca01a4835..af22b83cede7 100644
> > --- a/Makefile
> > +++ b/Makefile
> > @@ -1072,7 +1072,6 @@ KBUILD_CFLAGS += $(call cc-disable-warning, stringop-truncation)
> >  
> >  # We'll want to enable this eventually, but it's not going away for 5.7 at least
> >  KBUILD_CFLAGS += $(call cc-disable-warning, zero-length-bounds)
> > -KBUILD_CFLAGS += -Wno-array-bounds
> >  KBUILD_CFLAGS += $(call cc-disable-warning, stringop-overflow)
> >  
> >  # Another good warning that we'll want to enable eventually
> 
> alpha:defconfig:
> 
> In function '__memset',
>     inlined from '__bad_pagetable' at arch/alpha/mm/init.c:79:2:
> ./arch/alpha/include/asm/string.h:37:32: error: '__builtin_memset' offset [0, 8191] is out of the bounds [0, 0] [-Werror=array-bounds]
>    37 |                         return __builtin_memset(s, c, n);
>       |                                ^~~~~~~~~~~~~~~~~~~~~~~~~
> In function '__memset',
>     inlined from '__bad_page' at arch/alpha/mm/init.c:86:2:
> ./arch/alpha/include/asm/string.h:37:32: error: '__builtin_memset' offset [0, 8191] is out of the bounds [0, 0] [-Werror=array-bounds]
>    37 |                         return __builtin_memset(s, c, n);
>       |                                ^~~~~~~~~~~~~~~~~~~~~~~~~
> In function '__memset',
>     inlined from 'paging_init' at arch/alpha/mm/init.c:256:2:
> ./arch/alpha/include/asm/string.h:37:32: error: '__builtin_memset' offset [0, 8191] is out of the bounds [0, 0] [-Werror=array-bounds]
>    37 |                         return __builtin_memset(s, c, n);

Thanks! I'll take a look. Every instance of the "[0, 0]" bounds means
the compiler believes there's a way for the destination to be determined
at compile-time to be NULL.

> xtensa:allmodconfig:
> --------------
> Error log:
> In file included from include/linux/uaccess.h:11,
>                  from include/linux/sched/task.h:11,
>                  from arch/xtensa/kernel/process.c:21:
> arch/xtensa/kernel/process.c: In function 'copy_thread':
> arch/xtensa/kernel/process.c:262:52: error: array subscript 53 is above array bounds of 'long unsigned int[16]'

I assume this is a weird cast. I will also check this one out.

Thanks!

Kees Cook Feb. 2, 2022, 10:02 p.m. UTC | #3

On Wed, Feb 02, 2022 at 08:09:03AM -0800, Guenter Roeck wrote:
> On Wed, Aug 18, 2021 at 01:11:17AM -0700, Kees Cook wrote:
> > With the recent fixes for flexible arrays and expanded FORTIFY_SOURCE
> > coverage, it is now possible to enable -Warray-bounds. Since both
> > GCC and Clang include -Warray-bounds in -Wall, we just need to stop
> > disabling it.
> > 
> > Cc: Arnd Bergmann <[e-mail]>
> > Cc: Masahiro Yamada <[e-mail]>
> > Cc: [e-mail]
> > Co-developed-by: Gustavo A. R. Silva <[e-mail]>
> > Signed-off-by: Gustavo A. R. Silva <[e-mail]>
> > Signed-off-by: Kees Cook <[e-mail]>
> > ---
> >  Makefile | 1 -
> >  1 file changed, 1 deletion(-)
> > 
> > diff --git a/Makefile b/Makefile
> > index a4aca01a4835..af22b83cede7 100644
> > --- a/Makefile
> > +++ b/Makefile
> > @@ -1072,7 +1072,6 @@ KBUILD_CFLAGS += $(call cc-disable-warning, stringop-truncation)
> >  
> >  # We'll want to enable this eventually, but it's not going away for 5.7 at least
> >  KBUILD_CFLAGS += $(call cc-disable-warning, zero-length-bounds)
> > -KBUILD_CFLAGS += -Wno-array-bounds
> >  KBUILD_CFLAGS += $(call cc-disable-warning, stringop-overflow)
> >  
> >  # Another good warning that we'll want to enable eventually
> 
> alpha:defconfig:
> 
> In function '__memset',
>     inlined from '__bad_pagetable' at arch/alpha/mm/init.c:79:2:
> ./arch/alpha/include/asm/string.h:37:32: error: '__builtin_memset' offset [0, 8191] is out of the bounds [0, 0] [-Werror=array-bounds]
>    37 |                         return __builtin_memset(s, c, n);
>       |                                ^~~~~~~~~~~~~~~~~~~~~~~~~
> In function '__memset',
>     inlined from '__bad_page' at arch/alpha/mm/init.c:86:2:
> ./arch/alpha/include/asm/string.h:37:32: error: '__builtin_memset' offset [0, 8191] is out of the bounds [0, 0] [-Werror=array-bounds]
>    37 |                         return __builtin_memset(s, c, n);
>       |                                ^~~~~~~~~~~~~~~~~~~~~~~~~
> In function '__memset',
>     inlined from 'paging_init' at arch/alpha/mm/init.c:256:2:
> ./arch/alpha/include/asm/string.h:37:32: error: '__builtin_memset' offset [0, 8191] is out of the bounds [0, 0] [-Werror=array-bounds]
>    37 |                         return __builtin_memset(s, c, n);

Ugh, these all appear to be the same "gcc thinks a literal is an offset
from NULL" problem Arnd described:

https://lore.kernel.org/linux-hardening/CAK8P3a1CT+-3J7Ufv9TQwRiV4i_+FF=[e-mail]

What's annoying is that these values, if stored in a variable, aren't
treated that way:
https://pastebin.ubuntu.com/p/ZjZR2tzRpx/

I guess having a common way to convert from an integer (or literal) into
a void * that GCC doesn't freak out about would be generally nice.

-Kees

Kees Cook Feb. 2, 2022, 10:11 p.m. UTC | #4

On Wed, Feb 02, 2022 at 08:09:03AM -0800, Guenter Roeck wrote:
> On Wed, Aug 18, 2021 at 01:11:17AM -0700, Kees Cook wrote:
> > With the recent fixes for flexible arrays and expanded FORTIFY_SOURCE
> > coverage, it is now possible to enable -Warray-bounds. Since both
> > GCC and Clang include -Warray-bounds in -Wall, we just need to stop
> > disabling it.
> > 
> > Cc: Arnd Bergmann <[e-mail]>
> > Cc: Masahiro Yamada <[e-mail]>
> > Cc: [e-mail]
> > Co-developed-by: Gustavo A. R. Silva <[e-mail]>
> > Signed-off-by: Gustavo A. R. Silva <[e-mail]>
> > Signed-off-by: Kees Cook <[e-mail]>
> > ---
> >  Makefile | 1 -
> >  1 file changed, 1 deletion(-)
> > 
> > diff --git a/Makefile b/Makefile
> > index a4aca01a4835..af22b83cede7 100644
> > --- a/Makefile
> > +++ b/Makefile
> > @@ -1072,7 +1072,6 @@ KBUILD_CFLAGS += $(call cc-disable-warning, stringop-truncation)
> >  
> >  # We'll want to enable this eventually, but it's not going away for 5.7 at least
> >  KBUILD_CFLAGS += $(call cc-disable-warning, zero-length-bounds)
> > -KBUILD_CFLAGS += -Wno-array-bounds
> >  KBUILD_CFLAGS += $(call cc-disable-warning, stringop-overflow)
> >  
> >  # Another good warning that we'll want to enable eventually
> 
> alpha:defconfig:
> 
> In function '__memset',
>     inlined from '__bad_pagetable' at arch/alpha/mm/init.c:79:2:
> ./arch/alpha/include/asm/string.h:37:32: error: '__builtin_memset' offset [0, 8191] is out of the bounds [0, 0] [-Werror=array-bounds]
>    37 |                         return __builtin_memset(s, c, n);
>       |                                ^~~~~~~~~~~~~~~~~~~~~~~~~
> In function '__memset',
>     inlined from '__bad_page' at arch/alpha/mm/init.c:86:2:
> ./arch/alpha/include/asm/string.h:37:32: error: '__builtin_memset' offset [0, 8191] is out of the bounds [0, 0] [-Werror=array-bounds]
>    37 |                         return __builtin_memset(s, c, n);
>       |                                ^~~~~~~~~~~~~~~~~~~~~~~~~
> In function '__memset',
>     inlined from 'paging_init' at arch/alpha/mm/init.c:256:2:
> ./arch/alpha/include/asm/string.h:37:32: error: '__builtin_memset' offset [0, 8191] is out of the bounds [0, 0] [-Werror=array-bounds]
>    37 |                         return __builtin_memset(s, c, n);

Ah! With Arnd and Nathan's help, I saw:
https://lore.kernel.org/all/[e-mail]/
which is solving the same problem (just manifested through different
diagnostics). The same solution works here. I'll get the patches sent...

Guenter Roeck Feb. 2, 2022, 11:21 p.m. UTC | #5

On 2/2/22 14:11, Kees Cook wrote:
> On Wed, Feb 02, 2022 at 08:09:03AM -0800, Guenter Roeck wrote:
>> On Wed, Aug 18, 2021 at 01:11:17AM -0700, Kees Cook wrote:
>>> With the recent fixes for flexible arrays and expanded FORTIFY_SOURCE
>>> coverage, it is now possible to enable -Warray-bounds. Since both
>>> GCC and Clang include -Warray-bounds in -Wall, we just need to stop
>>> disabling it.
>>>
>>> Cc: Arnd Bergmann <[e-mail]>
>>> Cc: Masahiro Yamada <[e-mail]>
>>> Cc: [e-mail]
>>> Co-developed-by: Gustavo A. R. Silva <[e-mail]>
>>> Signed-off-by: Gustavo A. R. Silva <[e-mail]>
>>> Signed-off-by: Kees Cook <[e-mail]>
>>> ---
>>>   Makefile | 1 -
>>>   1 file changed, 1 deletion(-)
>>>
>>> diff --git a/Makefile b/Makefile
>>> index a4aca01a4835..af22b83cede7 100644
>>> --- a/Makefile
>>> +++ b/Makefile
>>> @@ -1072,7 +1072,6 @@ KBUILD_CFLAGS += $(call cc-disable-warning, stringop-truncation)
>>>   
>>>   # We'll want to enable this eventually, but it's not going away for 5.7 at least
>>>   KBUILD_CFLAGS += $(call cc-disable-warning, zero-length-bounds)
>>> -KBUILD_CFLAGS += -Wno-array-bounds
>>>   KBUILD_CFLAGS += $(call cc-disable-warning, stringop-overflow)
>>>   
>>>   # Another good warning that we'll want to enable eventually
>>
>> alpha:defconfig:
>>
>> In function '__memset',
>>      inlined from '__bad_pagetable' at arch/alpha/mm/init.c:79:2:
>> ./arch/alpha/include/asm/string.h:37:32: error: '__builtin_memset' offset [0, 8191] is out of the bounds [0, 0] [-Werror=array-bounds]
>>     37 |                         return __builtin_memset(s, c, n);
>>        |                                ^~~~~~~~~~~~~~~~~~~~~~~~~
>> In function '__memset',
>>      inlined from '__bad_page' at arch/alpha/mm/init.c:86:2:
>> ./arch/alpha/include/asm/string.h:37:32: error: '__builtin_memset' offset [0, 8191] is out of the bounds [0, 0] [-Werror=array-bounds]
>>     37 |                         return __builtin_memset(s, c, n);
>>        |                                ^~~~~~~~~~~~~~~~~~~~~~~~~
>> In function '__memset',
>>      inlined from 'paging_init' at arch/alpha/mm/init.c:256:2:
>> ./arch/alpha/include/asm/string.h:37:32: error: '__builtin_memset' offset [0, 8191] is out of the bounds [0, 0] [-Werror=array-bounds]
>>     37 |                         return __builtin_memset(s, c, n);
> 
> Ah! With Arnd and Nathan's help, I saw:
> https://lore.kernel.org/all/[e-mail]/

Guilty as charged. Sorry, I didn't try to analyze the problem,
or I might have noticed (and saved you some work).

Guenter

> which is solving the same problem (just manifested through different
> diagnostics). The same solution works here. I'll get the patches sent...
>

Guenter Roeck Feb. 2, 2022, 11:33 p.m. UTC | #6

On 2/2/22 12:56, Kees Cook wrote:
> On Wed, Feb 02, 2022 at 08:09:03AM -0800, Guenter Roeck wrote:
>> On Wed, Aug 18, 2021 at 01:11:17AM -0700, Kees Cook wrote:
>>> With the recent fixes for flexible arrays and expanded FORTIFY_SOURCE
>>> coverage, it is now possible to enable -Warray-bounds. Since both
>>> GCC and Clang include -Warray-bounds in -Wall, we just need to stop
>>> disabling it.
>>>
>>> Cc: Arnd Bergmann <[e-mail]>
>>> Cc: Masahiro Yamada <[e-mail]>
>>> Cc: [e-mail]
>>> Co-developed-by: Gustavo A. R. Silva <[e-mail]>
>>> Signed-off-by: Gustavo A. R. Silva <[e-mail]>
>>> Signed-off-by: Kees Cook <[e-mail]>
>>> ---
>>>   Makefile | 1 -
>>>   1 file changed, 1 deletion(-)
>>>
>>> diff --git a/Makefile b/Makefile
>>> index a4aca01a4835..af22b83cede7 100644
>>> --- a/Makefile
>>> +++ b/Makefile
>>> @@ -1072,7 +1072,6 @@ KBUILD_CFLAGS += $(call cc-disable-warning, stringop-truncation)
>>>   
>>>   # We'll want to enable this eventually, but it's not going away for 5.7 at least
>>>   KBUILD_CFLAGS += $(call cc-disable-warning, zero-length-bounds)
>>> -KBUILD_CFLAGS += -Wno-array-bounds
>>>   KBUILD_CFLAGS += $(call cc-disable-warning, stringop-overflow)
>>>   
>>>   # Another good warning that we'll want to enable eventually
>>
>> alpha:defconfig:
>>
>> In function '__memset',
>>      inlined from '__bad_pagetable' at arch/alpha/mm/init.c:79:2:
>> ./arch/alpha/include/asm/string.h:37:32: error: '__builtin_memset' offset [0, 8191] is out of the bounds [0, 0] [-Werror=array-bounds]
>>     37 |                         return __builtin_memset(s, c, n);
>>        |                                ^~~~~~~~~~~~~~~~~~~~~~~~~
>> In function '__memset',
>>      inlined from '__bad_page' at arch/alpha/mm/init.c:86:2:
>> ./arch/alpha/include/asm/string.h:37:32: error: '__builtin_memset' offset [0, 8191] is out of the bounds [0, 0] [-Werror=array-bounds]
>>     37 |                         return __builtin_memset(s, c, n);
>>        |                                ^~~~~~~~~~~~~~~~~~~~~~~~~
>> In function '__memset',
>>      inlined from 'paging_init' at arch/alpha/mm/init.c:256:2:
>> ./arch/alpha/include/asm/string.h:37:32: error: '__builtin_memset' offset [0, 8191] is out of the bounds [0, 0] [-Werror=array-bounds]
>>     37 |                         return __builtin_memset(s, c, n);
> 
> Thanks! I'll take a look. Every instance of the "[0, 0]" bounds means
> the compiler believes there's a way for the destination to be determined
> at compile-time to be NULL.
> 
>> xtensa:allmodconfig:
>> --------------
>> Error log:
>> In file included from include/linux/uaccess.h:11,
>>                   from include/linux/sched/task.h:11,
>>                   from arch/xtensa/kernel/process.c:21:
>> arch/xtensa/kernel/process.c: In function 'copy_thread':
>> arch/xtensa/kernel/process.c:262:52: error: array subscript 53 is above array bounds of 'long unsigned int[16]'
> 
> I assume this is a weird cast. I will also check this one out.
> 

                                 int callinc = (regs->areg[0] >> 30) & 3;
                                 int caller_ars = XCHAL_NUM_AREGS - callinc * 4;
                                 put_user(regs->areg[caller_ars+1],
                                          ^^^^^^^^^^^^^^^^^^^^^^^^
                                          (unsigned __user*)(usp - 12));

I think the problem is that XCHAL_NUM_AREGS can be up to 64,
but the size of struct pt_regs->areg[] is fixed to 16.

Guenter

Kees Cook Feb. 3, 2022, 3:02 a.m. UTC | #7

On Wed, Feb 02, 2022 at 03:21:49PM -0800, Guenter Roeck wrote:
> On 2/2/22 14:11, Kees Cook wrote:
> > On Wed, Feb 02, 2022 at 08:09:03AM -0800, Guenter Roeck wrote:
> > > On Wed, Aug 18, 2021 at 01:11:17AM -0700, Kees Cook wrote:
> > > > With the recent fixes for flexible arrays and expanded FORTIFY_SOURCE
> > > > coverage, it is now possible to enable -Warray-bounds. Since both
> > > > GCC and Clang include -Warray-bounds in -Wall, we just need to stop
> > > > disabling it.
> > > > 
> > > > Cc: Arnd Bergmann <[e-mail]>
> > > > Cc: Masahiro Yamada <[e-mail]>
> > > > Cc: [e-mail]
> > > > Co-developed-by: Gustavo A. R. Silva <[e-mail]>
> > > > Signed-off-by: Gustavo A. R. Silva <[e-mail]>
> > > > Signed-off-by: Kees Cook <[e-mail]>
> > > > ---
> > > >   Makefile | 1 -
> > > >   1 file changed, 1 deletion(-)
> > > > 
> > > > diff --git a/Makefile b/Makefile
> > > > index a4aca01a4835..af22b83cede7 100644
> > > > --- a/Makefile
> > > > +++ b/Makefile
> > > > @@ -1072,7 +1072,6 @@ KBUILD_CFLAGS += $(call cc-disable-warning, stringop-truncation)
> > > >   # We'll want to enable this eventually, but it's not going away for 5.7 at least
> > > >   KBUILD_CFLAGS += $(call cc-disable-warning, zero-length-bounds)
> > > > -KBUILD_CFLAGS += -Wno-array-bounds
> > > >   KBUILD_CFLAGS += $(call cc-disable-warning, stringop-overflow)
> > > >   # Another good warning that we'll want to enable eventually
> > > 
> > > alpha:defconfig:
> > > 
> > > In function '__memset',
> > >      inlined from '__bad_pagetable' at arch/alpha/mm/init.c:79:2:
> > > ./arch/alpha/include/asm/string.h:37:32: error: '__builtin_memset' offset [0, 8191] is out of the bounds [0, 0] [-Werror=array-bounds]
> > >     37 |                         return __builtin_memset(s, c, n);
> > >        |                                ^~~~~~~~~~~~~~~~~~~~~~~~~
> > > In function '__memset',
> > >      inlined from '__bad_page' at arch/alpha/mm/init.c:86:2:
> > > ./arch/alpha/include/asm/string.h:37:32: error: '__builtin_memset' offset [0, 8191] is out of the bounds [0, 0] [-Werror=array-bounds]
> > >     37 |                         return __builtin_memset(s, c, n);
> > >        |                                ^~~~~~~~~~~~~~~~~~~~~~~~~
> > > In function '__memset',
> > >      inlined from 'paging_init' at arch/alpha/mm/init.c:256:2:
> > > ./arch/alpha/include/asm/string.h:37:32: error: '__builtin_memset' offset [0, 8191] is out of the bounds [0, 0] [-Werror=array-bounds]
> > >     37 |                         return __builtin_memset(s, c, n);
> > 
> > Ah! With Arnd and Nathan's help, I saw:
> > https://lore.kernel.org/all/[e-mail]/
> 
> Guilty as charged. Sorry, I didn't try to analyze the problem,
> or I might have noticed (and saved you some work).

No worries! It manifests as a completely different warning. It just
happens to come from the same internal diagnostics info, I think. :)

I was just delighted to see it already had a solution!

-Kees

Kees Cook Feb. 3, 2022, 3:03 a.m. UTC | #8

On Wed, Feb 02, 2022 at 03:33:11PM -0800, Guenter Roeck wrote:
> On 2/2/22 12:56, Kees Cook wrote:
> > On Wed, Feb 02, 2022 at 08:09:03AM -0800, Guenter Roeck wrote:
> > > xtensa:allmodconfig:
> > > --------------
> > > Error log:
> > > In file included from include/linux/uaccess.h:11,
> > >                   from include/linux/sched/task.h:11,
> > >                   from arch/xtensa/kernel/process.c:21:
> > > arch/xtensa/kernel/process.c: In function 'copy_thread':
> > > arch/xtensa/kernel/process.c:262:52: error: array subscript 53 is above array bounds of 'long unsigned int[16]'
> > 
> > I assume this is a weird cast. I will also check this one out.
> > 
> 
>                                 int callinc = (regs->areg[0] >> 30) & 3;
>                                 int caller_ars = XCHAL_NUM_AREGS - callinc * 4;
>                                 put_user(regs->areg[caller_ars+1],
>                                          ^^^^^^^^^^^^^^^^^^^^^^^^
>                                          (unsigned __user*)(usp - 12));
> 
> I think the problem is that XCHAL_NUM_AREGS can be up to 64,
> but the size of struct pt_regs->areg[] is fixed to 16.

Yeah, that's what I found too. The comment about pt_regs::areg seems to
hint that things treat pt_regs differently. I've asked on the xtensa list:
https://lore.kernel.org/linux-hardening/202202021501.DA6594BFC@keescook/

Patch

diff --git a/Makefile b/Makefile
index a4aca01a4835..af22b83cede7 100644
--- a/Makefile
+++ b/Makefile
@@ -1072,7 +1072,6 @@  KBUILD_CFLAGS += $(call cc-disable-warning, stringop-truncation)
 
 # We'll want to enable this eventually, but it's not going away for 5.7 at least
 KBUILD_CFLAGS += $(call cc-disable-warning, zero-length-bounds)
-KBUILD_CFLAGS += -Wno-array-bounds
 KBUILD_CFLAGS += $(call cc-disable-warning, stringop-overflow)
 
 # Another good warning that we'll want to enable eventually