linux-kernel - Re: [PATCH RFC v2 04/21] kasan: unpoison stack only with CONFIG_KASAN

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]
Message-ID: <9d4156e6-ec4f-a742-a44e-f38bf7fa9ba9@arm.com>
Date:   Mon, 16 Nov 2020 12:45:19 +0000
From:   Vincenzo Frascino <vincenzo.frascino@....com>
To:     Dmitry Vyukov <dvyukov@...gle.com>,
        Catalin Marinas <catalin.marinas@....com>
Cc:     Marco Elver <elver@...gle.com>,
        Andrey Konovalov <andreyknvl@...gle.com>,
        Will Deacon <will.deacon@....com>,
        Alexander Potapenko <glider@...gle.com>,
        Evgenii Stepanov <eugenis@...gle.com>,
        Kostya Serebryany <kcc@...gle.com>,
        Peter Collingbourne <pcc@...gle.com>,
        Serban Constantinescu <serbanc@...gle.com>,
        Andrey Ryabinin <aryabinin@...tuozzo.com>,
        Elena Petrova <lenaptr@...gle.com>,
        Branislav Rankov <Branislav.Rankov@....com>,
        Kevin Brodsky <kevin.brodsky@....com>,
        Andrew Morton <akpm@...ux-foundation.org>,
        kasan-dev <kasan-dev@...glegroups.com>,
        Linux ARM <linux-arm-kernel@...ts.infradead.org>,
        Linux-MM <linux-mm@...ck.org>,
        LKML <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH RFC v2 04/21] kasan: unpoison stack only with
 CONFIG_KASAN_STACK



On 11/16/20 12:19 PM, Dmitry Vyukov wrote:
> On Mon, Nov 16, 2020 at 1:16 PM Catalin Marinas <catalin.marinas@....com> wrote:
>>
>> On Mon, Nov 16, 2020 at 12:50:00PM +0100, Marco Elver wrote:
>>> On Mon, 16 Nov 2020 at 11:59, Dmitry Vyukov <dvyukov@...gle.com> wrote:
>>>> On Thu, Oct 29, 2020 at 8:57 PM 'Andrey Konovalov' via kasan-dev
>>>> <kasan-dev@...glegroups.com> wrote:
>>>>> On Tue, Oct 27, 2020 at 1:44 PM Dmitry Vyukov <dvyukov@...gle.com> wrote:
>>>>>>
>>>>>> On Thu, Oct 22, 2020 at 3:19 PM Andrey Konovalov <andreyknvl@...gle.com> wrote:
>>>>>>>
>>>>>>> There's a config option CONFIG_KASAN_STACK that has to be enabled for
>>>>>>> KASAN to use stack instrumentation and perform validity checks for
>>>>>>> stack variables.
>>>>>>>
>>>>>>> There's no need to unpoison stack when CONFIG_KASAN_STACK is not enabled.
>>>>>>> Only call kasan_unpoison_task_stack[_below]() when CONFIG_KASAN_STACK is
>>>>>>> enabled.
>>>>>>>
>>>>>>> Signed-off-by: Andrey Konovalov <andreyknvl@...gle.com>
>>>>>>> Link: https://linux-review.googlesource.com/id/If8a891e9fe01ea543e00b576852685afec0887e3
>>>>>>> ---
>>>>>>>  arch/arm64/kernel/sleep.S        |  2 +-
>>>>>>>  arch/x86/kernel/acpi/wakeup_64.S |  2 +-
>>>>>>>  include/linux/kasan.h            | 10 ++++++----
>>>>>>>  mm/kasan/common.c                |  2 ++
>>>>>>>  4 files changed, 10 insertions(+), 6 deletions(-)
>>>>>>>
>>>>>>> diff --git a/arch/arm64/kernel/sleep.S b/arch/arm64/kernel/sleep.S
>>>>>>> index ba40d57757d6..bdadfa56b40e 100644
>>>>>>> --- a/arch/arm64/kernel/sleep.S
>>>>>>> +++ b/arch/arm64/kernel/sleep.S
>>>>>>> @@ -133,7 +133,7 @@ SYM_FUNC_START(_cpu_resume)
>>>>>>>          */
>>>>>>>         bl      cpu_do_resume
>>>>>>>
>>>>>>> -#ifdef CONFIG_KASAN
>>>>>>> +#if defined(CONFIG_KASAN) && CONFIG_KASAN_STACK
>>>>>>>         mov     x0, sp
>>>>>>>         bl      kasan_unpoison_task_stack_below
>>>>>>>  #endif
>>>>>>> diff --git a/arch/x86/kernel/acpi/wakeup_64.S b/arch/x86/kernel/acpi/wakeup_64.S
>>>>>>> index c8daa92f38dc..5d3a0b8fd379 100644
>>>>>>> --- a/arch/x86/kernel/acpi/wakeup_64.S
>>>>>>> +++ b/arch/x86/kernel/acpi/wakeup_64.S
>>>>>>> @@ -112,7 +112,7 @@ SYM_FUNC_START(do_suspend_lowlevel)
>>>>>>>         movq    pt_regs_r14(%rax), %r14
>>>>>>>         movq    pt_regs_r15(%rax), %r15
>>>>>>>
>>>>>>> -#ifdef CONFIG_KASAN
>>>>>>> +#if defined(CONFIG_KASAN) && CONFIG_KASAN_STACK
>>>>>>>         /*
>>>>>>>          * The suspend path may have poisoned some areas deeper in the stack,
>>>>>>>          * which we now need to unpoison.
>>>>>>> diff --git a/include/linux/kasan.h b/include/linux/kasan.h
>>>>>>> index 3f3f541e5d5f..7be9fb9146ac 100644
>>>>>>> --- a/include/linux/kasan.h
>>>>>>> +++ b/include/linux/kasan.h
>>>>>>> @@ -68,8 +68,6 @@ static inline void kasan_disable_current(void) {}
>>>>>>>
>>>>>>>  void kasan_unpoison_memory(const void *address, size_t size);
>>>>>>>
>>>>>>> -void kasan_unpoison_task_stack(struct task_struct *task);
>>>>>>> -
>>>>>>>  void kasan_alloc_pages(struct page *page, unsigned int order);
>>>>>>>  void kasan_free_pages(struct page *page, unsigned int order);
>>>>>>>
>>>>>>> @@ -114,8 +112,6 @@ void kasan_restore_multi_shot(bool enabled);
>>>>>>>
>>>>>>>  static inline void kasan_unpoison_memory(const void *address, size_t size) {}
>>>>>>>
>>>>>>> -static inline void kasan_unpoison_task_stack(struct task_struct *task) {}
>>>>>>> -
>>>>>>>  static inline void kasan_alloc_pages(struct page *page, unsigned int order) {}
>>>>>>>  static inline void kasan_free_pages(struct page *page, unsigned int order) {}
>>>>>>>
>>>>>>> @@ -167,6 +163,12 @@ static inline size_t kasan_metadata_size(struct kmem_cache *cache) { return 0; }
>>>>>>>
>>>>>>>  #endif /* CONFIG_KASAN */
>>>>>>>
>>>>>>> +#if defined(CONFIG_KASAN) && CONFIG_KASAN_STACK
>>>>>>
>>>>>> && defined(CONFIG_KASAN_STACK) for consistency
>>>>>
>>>>> CONFIG_KASAN_STACK is different from other KASAN configs. It's always
>>>>> defined, and its value is what controls whether stack instrumentation
>>>>> is enabled.
>>>>
>>>> Not sure why we did this instead of the following, but okay.
>>>>
>>>>  config KASAN_STACK
>>>> -       int
>>>> -       default 1 if KASAN_STACK_ENABLE || CC_IS_GCC
>>>> -       default 0
>>>> +       bool
>>>> +       default y if KASAN_STACK_ENABLE || CC_IS_GCC
>>>> +       default n
>>>
>>> I wondered the same, but then looking at scripts/Makefile.kasan I
>>> think it's because we directly pass it to the compiler:
>>>     ...
>>>     $(call cc-param,asan-stack=$(CONFIG_KASAN_STACK)) \
>>>     ...
>>
>> Try this instead:
>>
>>       $(call cc-param,asan-stack=$(if $(CONFIG_KASAN_STACK),1,0)) \
> 
> 
> We could have just 1 config instead of 2 as well.
> For gcc we could do no prompt and default value y, and for clang --
> prompt and default value n. I think it should do what we need.
> 

I agree with Catalin's proposal since it should simplify things.

Nit: 'default n' is the default hence I do not think it should be required
explicitly.

-- 
Regards,
Vincenzo