Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1704413

Re: [PATCH 0/2] mm,fork,security: introduce MADV_WIPEONFORK

From "Kirill A. Shutemov" <kirill@shutemov.name>
Newsgroups linux.kernel
Subject Re: [PATCH 0/2] mm,fork,security: introduce MADV_WIPEONFORK
Date 2017-08-05 01:50 +0200
Message-ID <uaUvW-5hz-63@gated-at.bofh.it> (permalink)
References <uaQ8W-2CK-17@gated-at.bofh.it>
Organization linux.* mail to news gateway

Show all headers | View raw


On Fri, Aug 04, 2017 at 03:07:28PM -0400, riel@redhat.com wrote:
> [resend because half the recipients got dropped due to IPv6 firewall issues]
> 
> Introduce MADV_WIPEONFORK semantics, which result in a VMA being
> empty in the child process after fork. This differs from MADV_DONTFORK
> in one important way.
> 
> If a child process accesses memory that was MADV_WIPEONFORK, it
> will get zeroes. The address ranges are still valid, they are just empty.

I feel like we are repeating mistake we made with MADV_DONTNEED.

MADV_WIPEONFORK would require a specific action from kernel, ignoring
the /advise/ would likely lead to application misbehaviour.

Is it something we really want to see from madvise()?

-- 
 Kirill A. Shutemov

Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread


Thread

[PATCH 0/2] mm,fork,security: introduce MADV_WIPEONFORK riel@redhat.com - 2017-08-04 21:10 +0200
  Re: [PATCH 0/2] mm,fork,security: introduce MADV_WIPEONFORK "Kirill A. Shutemov" <kirill@shutemov.name> - 2017-08-05 01:50 +0200
    Re: [PATCH 0/2] mm,fork,security: introduce MADV_WIPEONFORK Rik van Riel <riel@redhat.com> - 2017-08-05 17:30 +0200

csiph-web