Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Thu, 20 May 2021 12:42:57 +0100
From:   Aaron Tomlin <atomlin@redhat.com>
To:     Vlastimil Babka <vbabka@suse.cz>
Cc:     Andrew Morton <akpm@linux-foundation.org>, linux-mm@kvack.org,
        mhocko@suse.com, willy@infradead.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH v3] mm/page_alloc: bail out on fatal signal during
 reclaim/compaction retry attempt
Message-ID: <20210520114257.huqhkqsdrhohn3u5@ava.usersys.com>
References: <YKVn69o1UizH0kJD@casper.infradead.org>
 <20210519201743.3260890-1-atomlin@redhat.com>
 <20210519213455.97ff95f0124b4120787f8314@linux-foundation.org>
 <9b11dcd8-bc3b-aae9-feb1-43543bf9e22f@suse.cz>
MIME-Version: 1.0
Content-Type: multipart/signed; micalg=pgp-sha256;
        protocol="application/pgp-signature"; boundary="p5hnecwugz7gnprh"
Content-Disposition: inline
In-Reply-To: <9b11dcd8-bc3b-aae9-feb1-43543bf9e22f@suse.cz>
Precedence: bulk


--p5hnecwugz7gnprh
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Thu 2021-05-20 12:20 +0200, Vlastimil Babka wrote:
> On 5/20/21 6:34 AM, Andrew Morton wrote:
> >=20
> > What observed problems motivated this change?
> >=20
> > What were the observed runtime effects of this change?
>=20
> Yep those details from the previous thread should be included here.

Fair enough.

During kernel crash dump/or vmcore analysis: I discovered in the context of
__alloc_pages_slowpath() the value stored in the no_progress_loops variable
was found to be 31,611,688 i.e. well above MAX_RECLAIM_RETRIES; and a fatal
signal was pending against current.


     #6 [ffff00002e78f7c0] do_try_to_free_pages+0xe4 at ffff00001028bd24
     #7 [ffff00002e78f840] try_to_free_pages+0xe4 at ffff00001028c0f4
     #8 [ffff00002e78f900] __alloc_pages_nodemask+0x500 at ffff0000102cd130

                                                             //      w28 =
=3D *(sp + 148)      /* no_progress_loops */
     0xffff0000102cd1e0 <__alloc_pages_nodemask+0x5b0>:      ldr     w0, [s=
p,#148]
                                                             //      w0 =3D=
 w0 + 0x1
     0xffff0000102cd1e4 <__alloc_pages_nodemask+0x5b4>:      add     w0, w0=
, #0x1
                                                             //      *(sp +=
 148) =3D w0
     0xffff0000102cd1e8 <__alloc_pages_nodemask+0x5b8>:      str     w0, [s=
p,#148]
                                                             //      if (w0=
 >=3D 0x10)
                                                             //          go=
to __alloc_pages_nodemask+0x904
     0xffff0000102cd1ec <__alloc_pages_nodemask+0x5bc>:      cmp     w0, #0=
x10
     0xffff0000102cd1f0 <__alloc_pages_nodemask+0x5c0>:      b.gt    0xffff=
0000102cd534

 - The stack pointer was 0xffff00002e78f900

     crash> p *(int *)(0xffff00002e78f900+148)
     $1 =3D 31611688

     crash> ps 521171
        PID    PPID  CPU       TASK        ST  %MEM     VSZ    RSS  COMM
     > 521171      1  36  ffff8080e2128800  RU   0.0 34789440  18624  speci=
al

     crash> p &((struct task_struct *)0xffff8080e2128800)->signal.shared_pe=
nding
     $2 =3D (struct sigpending *) 0xffff80809a416e40

     crash> p ((struct sigpending *)0xffff80809a416e40)->signal.sig[0]
     $3 =3D 0x804100

     crash> sig -s 0x804100
     SIGKILL SIGTERM SIGXCPU

     crash> p ((struct sigpending *)0xffff80809a416e40)->signal.sig[0] & 1U=
 << (9 - 1)
     $4 =3D 0x100


Unfortunately, this incident was not reproduced, to date.


Kind regards,

--=20
Aaron Tomlin

--p5hnecwugz7gnprh
Content-Type: application/pgp-signature; name="signature.asc"

-----BEGIN PGP SIGNATURE-----
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=r3HN
-----END PGP SIGNATURE-----

--p5hnecwugz7gnprh--