linux-kernel - Re: [PATCH] x86: override __compiletime_object

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Wed, 27 Nov 2013 14:20:50 +0000
From:	"Jan Beulich" <JBeulich@...e.com>
To:	"Ingo Molnar" <mingo@...nel.org>
Cc:	<mingo@...e.hu>, <tglx@...utronix.de>,
	"Linus Torvalds" <torvalds@...ux-foundation.org>,
	<arjan@...ux.intel.com>, <linux@...ck-us.net>,
	<linux-kernel@...r.kernel.org>, <hpa@...or.com>
Subject: Re: [PATCH] x86: override __compiletime_object_size()

>>> On 27.11.13 at 15:03, Ingo Molnar <mingo@...nel.org> wrote:

> * Jan Beulich <JBeulich@...e.com> wrote:
> 
>> As discussed in the context of commits 3df7b41a ("x86: Unify
>> copy_from_user() size checking") and 7a3d9b0f ("x86: Unify
>> copy_to_user() and add size checking to it"), we want to leverage
>> __builtin_object_size() also on newer gcc versions, but with other
>> architectures still using another model of copy_*_user() verification
>> we can't replace the global definition. Do it in the (only) header
>> needing the construct for now.
>> 
>> Signed-off-by: Jan Beulich <jbeulich@...e.com>
>> Cc: Arjan van de Ven <arjan@...ux.intel.com>
>> Cc: Guenter Roeck <linux@...ck-us.net>
>> ---
>>  arch/x86/include/asm/uaccess.h |    6 ++++++
>>  1 file changed, 6 insertions(+)
>> 
>> --- 3.13-rc1/arch/x86/include/asm/uaccess.h
>> +++ 3.13-rc1-x86-compiletime-object-size/arch/x86/include/asm/uaccess.h
>> @@ -584,6 +584,12 @@ __copy_from_user_overflow(int size, unsi
>>  
>>  #endif
>>  
>> +/* linux/compiler-gcc4.h restricts this to gcc < 4.6, which doesn't suit us. */
>> +#if defined(__GNUC__) && GCC_VERSION >= 40100
>> +# undef __compiletime_object_size
>> +# define __compiletime_object_size(obj) __builtin_object_size(obj, 0)
>> +#endif
>> +
> 
> Would be nice to have a more verbose changelog that explains what 
> benefits this brings us.

That makes no sense to me - the benefits of this construct should
have been explained with its introduction; the change here just
makes it being used under wider range of compilers (and the code
comment already says exactly that).

Jan

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/