Re: [PATCH] mm, numa: Fix bad pmd by atomically check for pmd_trans_huge when marking page tables prot_numa

From: "Zi Yan" <zi.yan@cs.rutgers.edu>
To: Andrew Morton <akpm@linux-foundation.org>
Cc: Mel Gorman <mgorman@techsingularity.net>,
	"Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>,
	Andrea Arcangeli <aarcange@redhat.com>,
	Rik van Riel <riel@redhat.com>, Michal Hocko <mhocko@kernel.org>,
	Vlastimil Babka <vbabka@suse.cz>,
	linux-mm@kvack.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] mm, numa: Fix bad pmd by atomically check for pmd_trans_huge when marking page tables prot_numa
Date: Mon, 10 Apr 2017 17:28:09 -0500	[thread overview]
Message-ID: <789A2322-A5B6-4AC8-8668-D7057A56A140@cs.rutgers.edu> (raw)
In-Reply-To: <20170410150903.f931ceb5475d2d3d8945bb71@linux-foundation.org>

[-- Attachment #1: Type: text/plain, Size: 2882 bytes --]

On 10 Apr 2017, at 17:09, Andrew Morton wrote:

> On Mon, 10 Apr 2017 19:07:14 +0100 Mel Gorman <mgorman@techsingularity.net> wrote:
>
>> On Mon, Apr 10, 2017 at 12:49:40PM -0500, Zi Yan wrote:
>>> On 10 Apr 2017, at 12:20, Mel Gorman wrote:
>>>
>>>> On Mon, Apr 10, 2017 at 11:45:08AM -0500, Zi Yan wrote:
>>>>>> While this could be fixed with heavy locking, it's only necessary to
>>>>>> make a copy of the PMD on the stack during change_pmd_range and avoid
>>>>>> races. A new helper is created for this as the check if quite subtle and the
>>>>>> existing similar helpful is not suitable. This passed 154 hours of testing
>>>>>> (usually triggers between 20 minutes and 24 hours) without detecting bad
>>>>>> PMDs or corruption. A basic test of an autonuma-intensive workload showed
>>>>>> no significant change in behaviour.
>>>>>>
>>>>>> Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
>>>>>> Cc: stable@vger.kernel.org
>>>>>
>>>>> Does this patch fix the same problem fixed by Kirill's patch here?
>>>>> https://lkml.org/lkml/2017/3/2/347
>>>>>
>>>>
>>>> I don't think so. The race I'm concerned with is due to locks not being
>>>> held and is in a different path.
>>>
>>> I do not agree. Kirill's patch is fixing the same race problem but in
>>> zap_pmd_range().
>>>
>>> The original autoNUMA code first clears PMD then sets it to protnone entry.
>>> pmd_trans_huge() does not return TRUE because it saw cleared PMD, but
>>> pmd_none_or_clear_bad() later saw the protnone entry and reported it as bad.
>>> Is this the problem you are trying solve?
>>>
>>> Kirill's patch will pmdp_invalidate() the PMD entry, which keeps _PAGE_PSE bit,
>>> so pmd_trans_huge() will return TRUE. In this case, it also fixes
>>> your race problem in change_pmd_range().
>>>
>>> Let me know if I miss anything.
>>>
>>
>> Ok, now I see. I think you're correct and I withdraw the patch.
>
> I have Kirrill's
>
> thp-reduce-indentation-level-in-change_huge_pmd.patch
> thp-fix-madv_dontneed-vs-numa-balancing-race.patch
> mm-drop-unused-pmdp_huge_get_and_clear_notify.patch
> thp-fix-madv_dontneed-vs-madv_free-race.patch
> thp-fix-madv_dontneed-vs-madv_free-race-fix.patch
> thp-fix-madv_dontneed-vs-clear-soft-dirty-race.patch
>
> scheduled for 4.12-rc1.  It sounds like
> thp-fix-madv_dontneed-vs-madv_free-race.patch and
> thp-fix-madv_dontneed-vs-madv_free-race.patch need to be boosted to
> 4.11 and stable?

thp-fix-madv_dontneed-vs-numa-balancing-race.patch is the fix for
numa balancing problem reported in this thread.

mm-drop-unused-pmdp_huge_get_and_clear_notify.patch,
thp-fix-madv_dontneed-vs-madv_free-race.patch,
thp-fix-madv_dontneed-vs-madv_free-race-fix.patch, and
thp-fix-madv_dontneed-vs-clear-soft-dirty-race.patch

are the fixes for other potential race problems similar to this one.

I think it is better to have all these patches applied.

--
Best Regards
Yan Zi

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 496 bytes --]