DMARC-Filter: OpenDMARC Filter v1.3.2 mx1.redhat.com F3A5541A54
Date: Thu, 22 Jun 2017 22:51:15 +0200
From: Oleg Nesterov <oleg@redhat.com>
To: Hugh Dickins <hughd@google.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>,
        Cyrill Gorcunov <gorcunov@gmail.com>, Andrey Vagin <avagin@openvz.org>,
        Pavel Emelyanov <xemul@virtuozzo.com>,
        Dmitry Safonov <dsafonov@virtuozzo.com>,
        Andrew Morton <akpm@linux-foundation.org>,
        Adrian Reber <areber@redhat.com>, Michael Kerrisk <mtk@man7.org>,
        Willy Tarreau <w@1wt.eu>, kernel test robot <xiaolong.ye@intel.com>,
        Michal Hocko <mhocko@suse.com>, LKML <linux-kernel@vger.kernel.org>,
        LKP <lkp@01.org>, Larry Woodman <lwoodman@redhat.com>,
        Rik van Riel <riel@redhat.com>
Subject: Re: [lkp-robot] [mm] 1be7107fbe: kernel_BUG_at_mm/mmap.c
Message-ID: <20170622205115.GA4938@redhat.com>
References: <CA+55aFxzdtZ787snPhLhU8+2xUHKd74Som82yRCiFSQVXCNFhA@mail.gmail.com>
 <20170621193338.GA29222@redhat.com>
 <CA+55aFyTRPOuzo2-Q0xM4+D6naGxFtaE_7dbxXOOdwQ4QGPfpw@mail.gmail.com>
 <20170621202751.GA29638@redhat.com>
 <CA+55aFyupxx0K6SCi5_vd5nt0FmUYrYGF6yB+k9F+9zbgMwsiQ@mail.gmail.com>
 <20170621205617.GA29841@redhat.com>
 <CA+55aFzVEzC=FkFihCOyHyDkMP+eq_DR826oXLinrJ7YmENZ7Q@mail.gmail.com>
 <alpine.LSU.2.11.1706211725260.18020@eggly.anvils>
 <20170622151623.GB762@redhat.com>
 <alpine.LSU.2.11.1706221042550.20265@eggly.anvils>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <alpine.LSU.2.11.1706221042550.20265@eggly.anvils>
User-Agent: Mutt/1.5.24 (2015-08-30)
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 889
Lines: 23

On 06/22, Hugh Dickins wrote:
>
> On Thu, 22 Jun 2017, Oleg Nesterov wrote:
> >
> > Something like the patch below? Yes, I thought about this too.
>
> Yes, that patch (times 11 for all the architectures)
                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
Yes, yes, this is clear to me. But only after you have already explained
this in your previous email ;)

> But my own preference this morning is to do nothing, until we hear
> more complaints and can classify them as genuine userspace breakage,
> as opposed to testcases surprised by a new kernel implementation.

OK. Agreed. Lets wait for the "real" bug report.

FYI. I am still investigating that redhat internal bug report. And yes,
it was the real application. But. I still think that it fails by another
reason, just the test-case they provided doesn't match the reality and
it hits another (this) problem by accident.

Oleg.