linux-kernel - Re: Where in the code mmap + MAP_ANONYMOUS clears returned pages?

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] [day] [month] [year] [list]

Message-ID: <CAGc_3jHXGsv8uWNVejt5mEGLf2LbrfYk5jSgMC0Acap6jnrMOw@mail.gmail.com>
Date:	Wed, 16 Jul 2014 10:23:22 +0400
From:	Alexei Fedotov <alexei.fedotov@...il.com>
To:	Michal Hocko <mhocko@...e.cz>
Cc:	linux-kernel@...r.kernel.org
Subject: Re: Where in the code mmap + MAP_ANONYMOUS clears returned pages?

Thanks, Michal!
--
With best regards / с наилучшими пожеланиями,
Alexei Fedotov / Алексей Федотов,
http://dataved.ru/
+7 916 562 8095

[1] Start using Apache Openmeetings today, http://openmeetings.apache.org/
[2] Join Alexei Fedotov @linkedin, http://ru.linkedin.com/in/dataved/
[3] Join Alexei Fedotov @facebook, http://www.facebook.com/openmeetings


On Tue, Jul 15, 2014 at 11:14 PM, Michal Hocko <mhocko@...e.cz> wrote:
> On Tue 15-07-14 21:55:03, Alexei Fedotov wrote:
>> Guys,
>>
>> I look into mmap.c code [1] and try to understand where anonymous
>> memory mapping (mmap + MAP_ANONYMOUS) sets memory to zero. My best
>> guess is I should dig that somewhere in mmap_region.
>>
>> Could you please help me locating the code which makes returned memory
>> pages clean? Thank you in advance.
>
> mmap doesn't allocate pages which back the mapping directly unless it is
> told to by MAP_POPULATE. So the clean pages are either allocated during
> the page fault path when the memory is accessed (see do_anonymous_page
> for private 4k pages fault, do_huge_pmd_anonymous_page for THP page and
> MAP_SHARED is left as an excercise for the reader ;)) or in mmap path
> directly when MAP_POPULATE is specified (see vm_mmap_pgoff which calls
> mm_populate)
>
>>
>> [1] http://code.ohloh.net/file?fid=RPa2fuiQ5j_IwR1nymptYBBC3X4&cid=w6qrXNDsMqk
>>
>> --
>> With best regards / с наилучшими пожеланиями,
>> Alexei Fedotov / Алексей Федотов,
>> http://dataved.ru/
>> +7 916 562 8095
>>
>> [1] Start using Apache Openmeetings today, http://openmeetings.apache.org/
>> [2] Join Alexei Fedotov @linkedin, http://ru.linkedin.com/in/dataved/
>> [3] Join Alexei Fedotov @facebook, http://www.facebook.com/openmeetings
>> --
>> 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/
>
> --
> Michal Hocko
> SUSE Labs
--
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/