All of lore.kernel.org
 help / color / mirror / Atom feed
From: Rik van Riel <riel@redhat.com>
To: Mel Gorman <mgorman@suse.de>, Andrew Morton <akpm@linux-foundation.org>
Cc: Sasha Levin <sasha.levin@oracle.com>,
	David Rientjes <rientjes@google.com>,
	hhuang@redhat.com, knoel@redhat.com, aarcange@redhat.com,
	Davidlohr Bueso <davidlohr@hp.com>,
	"linux-mm@kvack.org" <linux-mm@kvack.org>,
	LKML <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH] mm: numa: Recheck for transhuge pages under lock during protection changes
Date: Wed, 12 Mar 2014 08:16:20 -0400	[thread overview]
Message-ID: <53205014.2050602@redhat.com> (raw)
In-Reply-To: <20140312103602.GN10663@suse.de>

On 03/12/2014 06:36 AM, Mel Gorman wrote:
> Andrew, this should go with the patches 
> mmnuma-reorganize-change_pmd_range.patch
> mmnuma-reorganize-change_pmd_range-fix.patch
> move-mmu-notifier-call-from-change_protection-to-change_pmd_range.patch
> in mmotm please.
> 
> Thanks.

That would be nice indeed :)

I am still not entirely sure why the kernel did not hit this race
before my reorganize change_pmd_range patch. Maybe gcc used to do
one load and now it does two?

> The problem is that a transhuge check is made without holding the PTL. It's
> possible at the time of the check that a parallel fault clears the pmd
> and inserts a new one which then triggers the VM_BUG_ON check.  This patch
> removes the VM_BUG_ON but fixes the race by rechecking transhuge under the
> PTL when marking page tables for NUMA hinting and bailing if a race occurred.
> It is not a problem for calls to mprotect() as they hold mmap_sem for write.
> 
> Reported-by: Sasha Levin <sasha.levin@oracle.com>
> Signed-off-by: Mel Gorman <mgorman@suse.de>

Reviewed-by: Rik van Riel <riel@redhat.com>

-- 
All rights reversed

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

WARNING: multiple messages have this Message-ID (diff)
From: Rik van Riel <riel@redhat.com>
To: Mel Gorman <mgorman@suse.de>, Andrew Morton <akpm@linux-foundation.org>
Cc: Sasha Levin <sasha.levin@oracle.com>,
	David Rientjes <rientjes@google.com>,
	hhuang@redhat.com, knoel@redhat.com, aarcange@redhat.com,
	Davidlohr Bueso <davidlohr@hp.com>,
	"linux-mm@kvack.org" <linux-mm@kvack.org>,
	LKML <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH] mm: numa: Recheck for transhuge pages under lock during protection changes
Date: Wed, 12 Mar 2014 08:16:20 -0400	[thread overview]
Message-ID: <53205014.2050602@redhat.com> (raw)
In-Reply-To: <20140312103602.GN10663@suse.de>

On 03/12/2014 06:36 AM, Mel Gorman wrote:
> Andrew, this should go with the patches 
> mmnuma-reorganize-change_pmd_range.patch
> mmnuma-reorganize-change_pmd_range-fix.patch
> move-mmu-notifier-call-from-change_protection-to-change_pmd_range.patch
> in mmotm please.
> 
> Thanks.

That would be nice indeed :)

I am still not entirely sure why the kernel did not hit this race
before my reorganize change_pmd_range patch. Maybe gcc used to do
one load and now it does two?

> The problem is that a transhuge check is made without holding the PTL. It's
> possible at the time of the check that a parallel fault clears the pmd
> and inserts a new one which then triggers the VM_BUG_ON check.  This patch
> removes the VM_BUG_ON but fixes the race by rechecking transhuge under the
> PTL when marking page tables for NUMA hinting and bailing if a race occurred.
> It is not a problem for calls to mprotect() as they hold mmap_sem for write.
> 
> Reported-by: Sasha Levin <sasha.levin@oracle.com>
> Signed-off-by: Mel Gorman <mgorman@suse.de>

Reviewed-by: Rik van Riel <riel@redhat.com>

-- 
All rights reversed

  reply	other threads:[~2014-03-12 12:16 UTC|newest]

Thread overview: 52+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-03-06 21:12 mm: kernel BUG at mm/mprotect.c:149 Sasha Levin
2014-03-06 21:12 ` Sasha Levin
2014-03-06 22:31 ` [PATCH -mm] mm,numa,mprotect: always continue after finding a stable thp page Rik van Riel
2014-03-06 22:31   ` Rik van Riel
2014-03-06 22:31 ` Rik van Riel
2014-03-06 22:31   ` Rik van Riel
2014-03-06 22:52   ` Rik van Riel
2014-03-06 22:52     ` Rik van Riel
2014-03-07 14:06     ` Mel Gorman
2014-03-07 14:06       ` Mel Gorman
2014-03-07 15:09       ` Mel Gorman
2014-03-07 15:09         ` Mel Gorman
2014-03-07 15:13         ` Mel Gorman
2014-03-07 15:13           ` Mel Gorman
2014-03-07 18:27         ` Mel Gorman
2014-03-07 18:27           ` Mel Gorman
2014-03-11 16:28           ` Mel Gorman
2014-03-11 16:28             ` Mel Gorman
2014-03-11 16:51             ` Sasha Levin
2014-03-11 16:51               ` Sasha Levin
2014-03-11 17:00               ` Rik van Riel
2014-03-11 17:00                 ` Rik van Riel
2014-03-11 17:33                 ` Sasha Levin
2014-03-11 17:33                   ` Sasha Levin
2014-03-11 17:32                   ` Rik van Riel
2014-03-11 17:32                     ` Rik van Riel
2014-03-11 18:06                   ` Mel Gorman
2014-03-11 18:06                     ` Mel Gorman
2014-03-11 19:18                     ` Sasha Levin
2014-03-11 19:18                       ` Sasha Levin
2014-03-11 19:28                       ` Andrew Morton
2014-03-11 19:28                         ` Andrew Morton
2014-03-11 19:33                         ` Sasha Levin
2014-03-11 19:33                           ` Sasha Levin
2014-03-12 10:36                         ` [PATCH] mm: numa: Recheck for transhuge pages under lock during protection changes Mel Gorman
2014-03-12 10:36                           ` Mel Gorman
2014-03-12 12:16                           ` Rik van Riel [this message]
2014-03-12 12:16                             ` Rik van Riel
2014-03-15  3:15                           ` Sasha Levin
2014-03-15  3:15                             ` Sasha Levin
2014-03-19 14:38                             ` Mel Gorman
2014-03-19 14:38                               ` Mel Gorman
2014-03-21 22:06                               ` Andrew Morton
2014-03-21 22:06                                 ` Andrew Morton
2014-03-21 22:36                                 ` Sasha Levin
2014-03-21 22:36                                   ` Sasha Levin
2014-03-21 22:24                               ` Rik van Riel
2014-03-21 22:24                                 ` Rik van Riel
2014-03-11 19:31                     ` [PATCH -mm] mm,numa,mprotect: always continue after finding a stable thp page Davidlohr Bueso
2014-03-11 19:31                       ` Davidlohr Bueso
2014-03-07  1:04   ` Sasha Levin
2014-03-07  1:04     ` Sasha Levin

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=53205014.2050602@redhat.com \
    --to=riel@redhat.com \
    --cc=aarcange@redhat.com \
    --cc=akpm@linux-foundation.org \
    --cc=davidlohr@hp.com \
    --cc=hhuang@redhat.com \
    --cc=knoel@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mgorman@suse.de \
    --cc=rientjes@google.com \
    --cc=sasha.levin@oracle.com \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.