linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Oleg Nesterov <oleg@redhat.com>
To: Kees Cook <keescook@chromium.org>
Cc: Alex Thorlton <athorlton@sgi.com>,
	"Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>,
	LKML <linux-kernel@vger.kernel.org>,
	Ingo Molnar <mingo@redhat.com>,
	Peter Zijlstra <peterz@infradead.org>,
	Andrew Morton <akpm@linux-foundation.org>,
	Mel Gorman <mgorman@suse.de>, Rik van Riel <riel@redhat.com>,
	Johannes Weiner <hannes@cmpxchg.org>,
	"Eric W. Biederman" <ebiederm@xmission.com>,
	Sedat Dilek <sedat.dilek@gmail.com>,
	Frederic Weisbecker <fweisbec@gmail.com>,
	Dave Jones <davej@redhat.com>,
	Michael Kerrisk <mtk.manpages@gmail.com>,
	"Paul E. McKenney" <paulmck@linux.vnet.ibm.com>,
	David Howells <dhowells@redhat.com>,
	Thomas Gleixner <tglx@linutronix.de>,
	Al Viro <viro@zeniv.linux.org.uk>,
	Srikar Dronamraju <srikar@linux.vnet.ibm.com>,
	Robin Holt <holt@sgi.com>
Subject: Re: [PATCH] Add per-process flag to control thp
Date: Sun, 4 Aug 2013 16:19:28 +0200	[thread overview]
Message-ID: <20130804141928.GA17935@redhat.com> (raw)
In-Reply-To: <CAGXu5j+d3fzKWNtmPF9tyMg7PzHsKWQZkhu5tFaThLTZue=3JQ@mail.gmail.com>

On 08/03, Kees Cook wrote:
>
> On Fri, Aug 2, 2013 at 1:34 PM, Alex Thorlton <athorlton@sgi.com> wrote:
> >
> >> And I think thp_disabled should be reset to 0 on exec.
> >
> > The main purpose for this getting carried down from the parent process
> > is that we'd like to be able to have a userland program set this flag on
> > itself, and then spawn off children who will also carry the flag.
> > This allows us to set the flag for programs where we're unable to modify
> > the code, thus resolving the issues detailed above.
>
> This could be done with LD_PRELOAD for uncontrolled binaries. Though
> it does seem sensible to make it act like most personality flags do
> (i.e. inherited).

It is only inheritable if the process is single-threaded.

And even if it is single-threaded I can't understand how it can really
help.

OK, transparent_hugepage_enabled() checks ->thp_disabled, but this is
the fault path. But this flag can't hide the task from khugepaged and
collapse_huge_page ?

I still think this should be mm flag. And we have MMF_INIT_MASK, it
can be inheritable.

Oleg.


  reply	other threads:[~2013-08-04 14:25 UTC|newest]

Thread overview: 57+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-08-02 19:46 [PATCH] Add per-process flag to control thp Alex Thorlton
2013-08-02 19:53 ` Dave Jones
2013-08-02 20:00   ` Alex Thorlton
2013-08-02 20:13 ` Kirill A. Shutemov
2013-08-02 20:34   ` Alex Thorlton
2013-08-02 23:59     ` Kirill A. Shutemov
2013-08-03 19:35     ` Kees Cook
2013-08-04 14:19       ` Oleg Nesterov [this message]
2013-08-05  2:36     ` Andi Kleen
2013-08-05 15:07       ` Alex Thorlton
2013-08-16 14:33       ` [PATCH 0/8] " Alex Thorlton
2013-08-16 14:33         ` [PATCH 1/8] THP: Use real address for NUMA policy Alex Thorlton
2013-08-16 17:53           ` Dave Hansen
2013-08-16 18:17             ` Alex Thorlton
2013-08-16 18:52               ` Kirill A. Shutemov
2013-08-27 16:50                 ` Alex Thorlton
2013-08-27 17:01                   ` Robin Holt
2013-09-04 15:43                     ` Alex Thorlton
2013-09-04 17:15                       ` Alex Thorlton
2013-09-05 11:15                         ` Ingo Molnar
2013-09-09 16:48                           ` Alex Thorlton
2013-09-10  7:47                             ` [benchmark] THP performance testcase Ingo Molnar
2013-09-13 13:06                               ` [PATCH 0/9] split page table lock for PMD tables Kirill A. Shutemov
2013-09-13 13:06                                 ` [PATCH 1/9] mm: rename SPLIT_PTLOCKS to SPLIT_PTE_PTLOCKS Kirill A. Shutemov
2013-09-13 15:20                                   ` Dave Hansen
2013-09-13 13:06                                 ` [PATCH 2/9] mm: convert mm->nr_ptes to atomic_t Kirill A. Shutemov
2013-09-13 13:06                                 ` [PATCH 3/9] mm: introduce api for split page table lock for PMD level Kirill A. Shutemov
2013-09-13 13:19                                   ` Peter Zijlstra
2013-09-13 14:22                                     ` Kirill A. Shutemov
2013-09-13 13:06                                 ` [PATCH 4/9] mm, thp: change pmd_trans_huge_lock() to return taken lock Kirill A. Shutemov
2013-09-13 13:06                                 ` [PATCH 5/9] mm, thp: move ptl taking inside page_check_address_pmd() Kirill A. Shutemov
2013-09-13 13:06                                 ` [PATCH 6/9] mm, thp: do not access mm->pmd_huge_pte directly Kirill A. Shutemov
2013-09-13 13:06                                 ` [PATCH 7/9] mm: convent the rest to new page table lock api Kirill A. Shutemov
2013-09-13 13:06                                 ` [PATCH 8/9] mm: implement split page table lock for PMD level Kirill A. Shutemov
2013-09-13 13:24                                   ` Peter Zijlstra
2013-09-13 14:25                                     ` Kirill A. Shutemov
2013-09-13 14:52                                       ` Peter Zijlstra
2013-09-13 13:36                                   ` Peter Zijlstra
2013-09-13 14:25                                     ` Kirill A. Shutemov
2013-09-13 15:45                                   ` Naoya Horiguchi
2013-09-13 19:57                                   ` Dave Hansen
2013-09-13 13:06                                 ` [PATCH 9/9] x86, mm: enable " Kirill A. Shutemov
     [not found]                   ` <20130828091814.GA13681@gmail.com>
2013-08-28  9:32                     ` [PATCH 1/8] THP: Use real address for NUMA policy Peter Zijlstra
2013-08-16 19:46               ` Peter Zijlstra
2013-08-16 19:49                 ` Alex Thorlton
2013-08-16 14:33         ` [PATCH 2/8] mm: make clear_huge_page tolerate non aligned address Alex Thorlton
2013-08-16 14:33         ` [PATCH 3/8] THP: Pass real, not rounded, address to clear_huge_page Alex Thorlton
2013-08-16 14:34         ` [PATCH 4/8] x86: Add clear_page_nocache Alex Thorlton
2013-08-16 14:34         ` [PATCH 5/8] mm: make clear_huge_page cache clear only around the fault address Alex Thorlton
2013-08-16 18:02           ` Dave Hansen
2013-08-16 14:34         ` [PATCH 6/8] x86: switch the 64bit uncached page clear to SSE/AVX v2 Alex Thorlton
2013-08-16 14:34         ` [PATCH 7/8] remove KM_USER0 from kmap_atomic call Alex Thorlton
2013-08-16 14:34         ` [PATCH 8/8] fix up references to kernel_fpu_begin/end Alex Thorlton
2013-08-16 14:47         ` [PATCH 0/8] Re: [PATCH] Add per-process flag to control thp Peter Zijlstra
2013-08-16 15:04           ` Alex Thorlton
2013-08-04 14:44 ` Rasmus Villemoes
2013-08-28 13:56 ` Andrea Arcangeli

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20130804141928.GA17935@redhat.com \
    --to=oleg@redhat.com \
    --cc=akpm@linux-foundation.org \
    --cc=athorlton@sgi.com \
    --cc=davej@redhat.com \
    --cc=dhowells@redhat.com \
    --cc=ebiederm@xmission.com \
    --cc=fweisbec@gmail.com \
    --cc=hannes@cmpxchg.org \
    --cc=holt@sgi.com \
    --cc=keescook@chromium.org \
    --cc=kirill.shutemov@linux.intel.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mgorman@suse.de \
    --cc=mingo@redhat.com \
    --cc=mtk.manpages@gmail.com \
    --cc=paulmck@linux.vnet.ibm.com \
    --cc=peterz@infradead.org \
    --cc=riel@redhat.com \
    --cc=sedat.dilek@gmail.com \
    --cc=srikar@linux.vnet.ibm.com \
    --cc=tglx@linutronix.de \
    --cc=viro@zeniv.linux.org.uk \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).