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;
MIME-Version: 1.0
References: <20220114081542.698002137@linuxfoundation.org> <20220114081542.746291845@linuxfoundation.org>
 <CA+res+S1GcDzM6hnmar+s1k3ggswURZ+_8BqweifShCTjVJ2aQ@mail.gmail.com> <YfFk7guenfgvrDlD@bender.morinfr.org>
In-Reply-To: <YfFk7guenfgvrDlD@bender.morinfr.org>
From:   Jack Wang <jack.wang.usish@gmail.com>
Date:   Wed, 26 Jan 2022 22:22:36 +0100
Message-ID: <CA+res+SmjdYYsvJO5t4BQ6ZXxdJibDbV_gU0Q_fNb0jzw+H3-Q@mail.gmail.com>
Subject: Re: [PATCH 5.10 01/25] md: revert io stats accounting
To:     Guillaume Morin <guillaume@morinfr.org>
Cc:     Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
        linux-kernel@vger.kernel.org, stable <stable@vger.kernel.org>,
        Guoqing Jiang <jiangguoqing@kylinos.cn>,
        Song Liu <song@kernel.org>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

Guillaume Morin <guillaume@morinfr.org> =E4=BA=8E2022=E5=B9=B41=E6=9C=8826=
=E6=97=A5=E5=91=A8=E4=B8=89 16:12=E5=86=99=E9=81=93=EF=BC=9A
>
> On 26 Jan 11:09, Jack Wang wrote:
> > >
> > > -       if (bio->bi_end_io !=3D md_end_io) {
> > > -               struct md_io *md_io;
> > > -
> > > -               md_io =3D mempool_alloc(&mddev->md_io_pool, GFP_NOIO)=
;
> > > -               md_io->mddev =3D mddev;
> > > -               md_io->orig_bi_end_io =3D bio->bi_end_io;
> > > -               md_io->orig_bi_private =3D bio->bi_private;
> > > -
> > > -               bio->bi_end_io =3D md_end_io;
> > > -               bio->bi_private =3D md_io;
> > > -
> > > -               md_io->start_time =3D part_start_io_acct(mddev->gendi=
sk,
> > > -                                                      &md_io->part, =
bio);
> > > -       }
> > > -
> > > +       /*
> > > +        * save the sectors now since our bio can
> > > +        * go away inside make_request
> > > +        */
> > > +       sectors =3D bio_sectors(bio);
> > This code snip is not inside the original patch, and it's not in
> > latest upstream too.
> > >         /* bio could be mergeable after passing to underlayer */
> > >         bio->bi_opf &=3D ~REQ_NOMERGE;
> > >
> > >         md_handle_request(mddev, bio);
> > >
> > > +       part_stat_lock();
> > > +       part_stat_inc(&mddev->gendisk->part0, ios[sgrp]);
> > > +       part_stat_add(&mddev->gendisk->part0, sectors[sgrp], sectors)=
;
> > > +       part_stat_unlock();
> > > +
> > same here, this code snip is not inside the original patch, and it's
> > not in latest upstream too.
>
> Both snippets came from the code before 41d2d848e5c0 that the patch is
> being reverted here.  As I explained in my original message, upstream is
> different because of 99dfc43ecbf6 which is not in 5.10.
oh, I missed it, you are right.
>
> > I think would be good keep it as the upstream version.
>
> If you don't include these lines, isn't this worse as it's not calling
> either part_start_io_acct or bio_start_io_acct (in 99dfc43ecbf6)?

Your patch is correct.
Sorry for the noise.

>
> --
> Guillaume Morin <guillaume@morinfr.org>