Re: [PATCH 2/2] mm/page_alloc.c: add config option to sanitize freed pages

[Anisse Astier]

On Fri, Apr 24, 2015 at 11:38 PM, David Rientjes <rientjes@xxxxxxxxxx> wrote:
> On Fri, 24 Apr 2015, Anisse Astier wrote:
>
>> diff --git a/mm/Kconfig b/mm/Kconfig
>> index 390214d..cb2df5f 100644
>> --- a/mm/Kconfig
>> +++ b/mm/Kconfig
>> @@ -635,3 +635,15 @@ config MAX_STACK_SIZE_MB
>>         changed to a smaller value in which case that is used.
>>
>>         A sane initial value is 80 MB.
>> +
>> +config SANITIZE_FREED_PAGES
>> +     bool "Sanitize memory pages after free"
>> +     default n
>> +     help
>> +       This option is used to make sure all pages freed are zeroed. This is
>> +       quite low-level and doesn't handle your slab buffers.
>> +       It has various applications, from preventing some info leaks to
>> +       helping kernel same-page merging in virtualised environments.
>> +       Depending on your workload, it will reduce performance of about 3%.
>> +
>> +       If unsure, say N.
>
> Objection to allowing this without first enabling some other DEBUG config
> option, it should never be a standalone option, but also to pretending to

I'm not sure I understand the rationale here. Is it to protect the
innocent ? The performance warning and "N" recommendation ought to be
enough.
I'm not sure depending on DEBUG will help anyone; it will just hinder
those who want to use this on a hardened system (where you might not
want to have DEBUG enabled).

> have any insight into what the performance degredation of it will be.  On

I fully agree I shouldn't have let the 3% ballpark estimate slip, I'll
remove it.

> my systems, this would be _massive_.

I'm interested in what you mean by "massive". Have you conducted
experiments on the impact or is just your gut feeling ? Anyway, I'd be
curious to see numbers showing what it looks like on big hardware.

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