Subject: Re: Page migration issue with UBIFS
To: Joonsoo Kim <iamjoonsoo.kim@lge.com>
References: <56E8192B.5030008@nod.at>
 <20160315151727.GA16462@node.shutemov.name> <56E82B18.9040807@nod.at>
 <20160315153744.GB28522@infradead.org> <56E8985A.1020509@nod.at>
 <20160316142156.GA23595@node.shutemov.name>
 <20160316142729.GA125481@black.fi.intel.com> <56E9C658.1020903@nod.at>
 <20160317071155.GB10315@js1304-P5Q-DELUXE>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>,
        "Kirill A. Shutemov" <kirill@shutemov.name>,
        Christoph Hellwig <hch@infradead.org>,
        linux-fsdevel <linux-fsdevel@vger.kernel.org>,
        "linux-mtd@lists.infradead.org" <linux-mtd@lists.infradead.org>,
        "linux-mm@kvack.org" <linux-mm@kvack.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        Boris Brezillon <boris.brezillon@free-electrons.com>,
        Maxime Ripard <maxime.ripard@free-electrons.com>,
        David Gstir <david@sigma-star.at>, Dave Chinner <david@fromorbit.com>,
        Artem Bityutskiy <dedekind1@gmail.com>,
        Alexander Kaplan <alex@nextthing.co>,
        "akpm@linux-foundation.org" <akpm@linux-foundation.org>,
        Sasha Levin <sasha.levin@oracle.com>, rvaswani@codeaurora.org,
        "Luck, Tony" <tony.luck@intel.com>,
        Shailendra Verma <shailendra.capricorn@gmail.com>,
        s.strogin@partner.samsung.com
From: Richard Weinberger <richard@nod.at>
Message-ID: <56EA672A.7070007@nod.at>
Date: Thu, 17 Mar 2016 09:13:30 +0100
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:38.0) Gecko/20100101
 Thunderbird/38.4.0
MIME-Version: 1.0
In-Reply-To: <20160317071155.GB10315@js1304-P5Q-DELUXE>
Content-Type: text/plain; charset=windows-1252
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1046
Lines: 28

Am 17.03.2016 um 08:11 schrieb Joonsoo Kim:
>> It is still not clear why UBIFS has to provide a >migratepage() and what the expected semantics
>> are.
>> What we know so far is that the fall back migration function is broken. I'm sure not only on UBIFS.
>>
>> Can CMA folks please clarify? :-)
> 
> Hello,
> 
> As you mentioned earlier, this issue would not be directly related
> to CMA. It looks like it is more general issue related to interaction
> between MM and FS. Your first error log shows that error happens when
> ubifs_set_page_dirty() is called in try_to_unmap_one() which also
> can be called by reclaimer (kswapd or direct reclaim). Quick search shows
> that problem also happens on reclaim. Is that fixed?
> 
> http://www.spinics.net/lists/linux-fsdevel/msg79531.html

Well, this problem happened only on a tainted kernel and never popped up again.
So, I really don't know. :-)

> I think that you need to CC other people who understand interaction
> between MM and FS perfectly.

Who is missing on the CC list?

Thanks,
//richard