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: <20210105012826.956800-1-daeho43@gmail.com> <904b7677-3655-2e81-52e0-48296cf1c0da@huawei.com>
In-Reply-To: <904b7677-3655-2e81-52e0-48296cf1c0da@huawei.com>
From:   Daeho Jeong <daeho43@gmail.com>
Date:   Tue, 5 Jan 2021 11:10:59 +0900
Message-ID: <CACOAw_xf9CFOSvU=s1yP4O74dropdc-ozcDOPhmCpBU3Oif4Bw@mail.gmail.com>
Subject: Re: [f2fs-dev] [PATCH] f2fs: fix null page reference in redirty_blocks
To:     Chao Yu <yuchao0@huawei.com>
Cc:     linux-kernel@vger.kernel.org,
        linux-f2fs-devel@lists.sourceforge.net, kernel-team@android.com,
        Colin Ian King <colin.king@canonical.com>,
        Daeho Jeong <daehojeong@google.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

Yes, it's better~ :)

2021=EB=85=84 1=EC=9B=94 5=EC=9D=BC (=ED=99=94) =EC=98=A4=EC=A0=84 10:44, C=
hao Yu <yuchao0@huawei.com>=EB=8B=98=EC=9D=B4 =EC=9E=91=EC=84=B1:
>
> On 2021/1/5 9:28, Daeho Jeong wrote:
> > From: Daeho Jeong <daehojeong@google.com>
> >
> > Fixed null page reference when find_lock_page() fails in
> > redirty_blocks().
> >
> > Signed-off-by: Daeho Jeong <daehojeong@google.com>
> > Reported-by: Colin Ian King <colin.king@canonical.com>
> > Fixes: 5fdb322ff2c2 ("f2fs: add F2FS_IOC_DECOMPRESS_FILE and F2FS_IOC_C=
OMPRESS_FILE")
> > ---
> >   fs/f2fs/file.c | 4 +++-
> >   1 file changed, 3 insertions(+), 1 deletion(-)
> >
> > diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c
> > index 9e5275716be8..bf6682a52433 100644
> > --- a/fs/f2fs/file.c
> > +++ b/fs/f2fs/file.c
> > @@ -4060,8 +4060,10 @@ static int redirty_blocks(struct inode *inode, p=
goff_t page_idx, int len)
> >
> >       for (i =3D 0; i < page_len; i++, redirty_idx++) {
> >               page =3D find_lock_page(mapping, redirty_idx);
> > -             if (!page)
> > +             if (!page) {
> >                       ret =3D -ENOENT;
>
> ret =3D -ENOMEM;
>
> > +                     continue;
>
> How about breaking the loop for out-of-memory case, because in such condi=
tion
> we have less chance to dirty whole cluster due to no memory, and continue=
 to
> allocate pages for target file will make system suffer more memory pressu=
re,
> it will make many thing slower.
>
> Thnaks,
>
> > +             }
> >               set_page_dirty(page);
> >               f2fs_put_page(page, 1);
> >               f2fs_put_page(page, 0);
> >