linux-kernel - Re: [PATCH] kasan: add no_sanitize attribute for clang builds

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:   Fri, 13 Apr 2018 19:34:47 +0200
From:   Andrey Konovalov <andreyknvl@...gle.com>
To:     Andrey Ryabinin <aryabinin@...tuozzo.com>
Cc:     Alexander Potapenko <glider@...gle.com>,
        Dmitry Vyukov <dvyukov@...gle.com>,
        Andrew Morton <akpm@...ux-foundation.org>,
        David Rientjes <rientjes@...gle.com>,
        Thomas Gleixner <tglx@...utronix.de>,
        Ingo Molnar <mingo@...nel.org>,
        David Woodhouse <dwmw@...zon.co.uk>,
        Will Deacon <will.deacon@....com>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        Paul Lawrence <paullawrence@...gle.com>,
        Sandipan Das <sandipan@...ux.vnet.ibm.com>,
        Kees Cook <keescook@...omium.org>,
        LKML <linux-kernel@...r.kernel.org>,
        kasan-dev <kasan-dev@...glegroups.com>,
        Linux Memory Management List <linux-mm@...ck.org>,
        Kostya Serebryany <kcc@...gle.com>
Subject: Re: [PATCH] kasan: add no_sanitize attribute for clang builds

On Fri, Apr 13, 2018 at 5:31 PM, Andrey Ryabinin
<aryabinin@...tuozzo.com> wrote:
>
>
> On 04/12/2018 08:29 PM, Andrey Konovalov wrote:
>> KASAN uses the __no_sanitize_address macro to disable instrumentation
>> of particular functions. Right now it's defined only for GCC build,
>> which causes false positives when clang is used.
>>
>> This patch adds a definition for clang.
>>
>> Note, that clang's revision 329612 or higher is required.
>>
>> Signed-off-by: Andrey Konovalov <andreyknvl@...gle.com>
>> ---
>>  include/linux/compiler-clang.h | 5 +++++
>>  1 file changed, 5 insertions(+)
>>
>> diff --git a/include/linux/compiler-clang.h b/include/linux/compiler-clang.h
>> index ceb96ecab96e..5a1d8580febe 100644
>> --- a/include/linux/compiler-clang.h
>> +++ b/include/linux/compiler-clang.h
>> @@ -25,6 +25,11 @@
>>  #define __SANITIZE_ADDRESS__
>>  #endif
>>
>> +#ifdef CONFIG_KASAN
>
> If, for whatever reason, developer decides to add __no_sanitize_address to some
> generic function, guess what will happen next when he/she will try to build CONFIG_KASAN=n kernel?

It's defined to nothing in compiler-gcc.h and redefined in
compiler-clang.h only if CONFIG_KASAN is enabled, so everything should
be fine. Am I missing something?

>
>> +#undef __no_sanitize_address
>> +#define __no_sanitize_address __attribute__((no_sanitize("address")))
>> +#endif
>> +
>>  /* Clang doesn't have a way to turn it off per-function, yet. */
>>  #ifdef __noretpoline
>>  #undef __noretpoline
>>