Received-SPF: pass (google.com: domain of linux-kernel+bounces-106687-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.80.249 as permitted sender) client-ip=147.75.80.249;
Precedence: bulk
MIME-Version: 1.0
References: <000000000000bbb3d80613f243a6@google.com> <CAKEwX=MAX0km1p43DQmKbeSy2G4dPFHiF+deH_qzqygc2Vnjig@mail.gmail.com>
 <CAJD7tkbEuFkGuQeYjKS02rQoAAKOKieAJ1P2mwukirW3e2JN9A@mail.gmail.com>
 <20240318210917.GA4210@cmpxchg.org> <CAJD7tkZhCt_r1uULbfe4eg=zo7=o7muJ787wyEw1yN7x+ijYGA@mail.gmail.com>
In-Reply-To: <CAJD7tkZhCt_r1uULbfe4eg=zo7=o7muJ787wyEw1yN7x+ijYGA@mail.gmail.com>
From: Barry Song <21cnbao@gmail.com>
Date: Tue, 19 Mar 2024 10:21:04 +1300
Message-ID: <CAGsJ_4zy14_hLxqkP=ka8AwLx1=H1MD4B-70G_+gp8SF_FOwHg@mail.gmail.com>
Subject: Re: [syzbot] [mm?] kernel BUG in sg_init_one
To: Yosry Ahmed <yosryahmed@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>, Nhat Pham <nphamcs@gmail.com>, 
	syzbot <syzbot+adbc983a1588b7805de3@syzkaller.appspotmail.com>, 
	akpm@linux-foundation.org, chengming.zhou@linux.dev, 
	linux-kernel@vger.kernel.org, linux-mm@kvack.org, 
	syzkaller-bugs@googlegroups.com, Barry Song <v-songbaohua@oppo.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Tue, Mar 19, 2024 at 10:19=E2=80=AFAM Yosry Ahmed <yosryahmed@google.com=
> wrote:
>
> On Mon, Mar 18, 2024 at 2:09=E2=80=AFPM Johannes Weiner <hannes@cmpxchg.o=
rg> wrote:
> >
> > On Mon, Mar 18, 2024 at 01:17:19PM -0700, Yosry Ahmed wrote:
> > > On Mon, Mar 18, 2024 at 11:00=E2=80=AFAM Nhat Pham <nphamcs@gmail.com=
> wrote:
> > > >
> > > > On Mon, Mar 18, 2024 at 9:58=E2=80=AFAM syzbot
> > > > <syzbot+adbc983a1588b7805de3@syzkaller.appspotmail.com> wrote:
> > > > >
> > > > > Hello,
> > > > >
> > > > > syzbot found the following issue on:
> > > > >
> > > > > HEAD commit:    e5eb28f6d1af Merge tag 'mm-nonmm-stable-2024-03-1=
4-09-36' ..
> > > > > git tree:       upstream
> > > > > console output: https://syzkaller.appspot.com/x/log.txt?x=3D13043=
abe180000
> > > > > kernel config:  https://syzkaller.appspot.com/x/.config?x=3D19bb5=
7c23dffc38e
> > > > > dashboard link: https://syzkaller.appspot.com/bug?extid=3Dadbc983=
a1588b7805de3
> > > > > compiler:       arm-linux-gnueabi-gcc (Debian 12.2.0-14) 12.2.0, =
GNU ld (GNU Binutils for Debian) 2.40
> > > > > userspace arch: arm
> > > > > syz repro:      https://syzkaller.appspot.com/x/repro.syz?x=3D170=
6d231180000
> > > > > C reproducer:   https://syzkaller.appspot.com/x/repro.c?x=3D13ba7=
959180000
> > > > >
> > > > > Downloadable assets:
> > > > > disk image (non-bootable): https://storage.googleapis.com/syzbot-=
assets/8ead8862021c/non_bootable_disk-e5eb28f6.raw.xz
> > > > > vmlinux: https://storage.googleapis.com/syzbot-assets/0a7371c63ff=
2/vmlinux-e5eb28f6.xz
> > > > > kernel image: https://storage.googleapis.com/syzbot-assets/753944=
1b4add/zImage-e5eb28f6.xz
> > > > >
> > > > > IMPORTANT: if you fix the issue, please add the following tag to =
the commit:
> > > > > Reported-by: syzbot+adbc983a1588b7805de3@syzkaller.appspotmail.co=
m
> > > > >
> > > > > ------------[ cut here ]------------
> > > > > kernel BUG at include/linux/scatterlist.h:187!
> > > >
> > > > Looks like the provided buffer is invalid:
> > > >
> > > > #ifdef CONFIG_DEBUG_SG
> > > > BUG_ON(!virt_addr_valid(buf));
> > > > #endif
> > > >
> > > > which is "src" from:
> > > >
> > > > sg_init_one(&input, src, entry->length);
> > > >
> > > > Looking at the surrounding code and recent history, there's this
> > > > commit that stands out:
> > > >
> > > > mm/zswap: remove the memcpy if acomp is not sleepable
> > > > (sha: 270700dd06ca41a4779c19eb46608f076bb7d40e)
> > > >
> > > > which has the effect of, IIUC, using the zpool mapped memory direct=
ly
> > > > as src, instead of acomp_ctx->buffer (which was previously the case=
,
> > > > as zsmalloc was not sleepable).
> > > >
> > > > This might not necessarily be a bug with that commit itself, but mi=
ght
> > > > have revealed another bug elsewhere.
> > > >
> > > > Anyway, cc-ing the author, Barry Song, to fact check me :) Will tak=
e a
> > > > closer look later.
> > >
> > > I am not a highmem expert, but the reproducer has CONFIG_HIGHMEM=3Dy,
> > > and it seems like zs_map_object() may return a highmem address if the
> > > compressed object is entirely in a single page to avoid copying to a
> > > buffer:
> > >
> > > if (off + class->size <=3D PAGE_SIZE) {
> > >    /* this object is contained entirely within a page */
> > >    area->vm_addr =3D kmap_atomic(page);
> > >    ret =3D area->vm_addr + off;
> > >    goto out;
> > > }
> > >
> > > The virt_addr_valid() check seems to indicate that we expect a direct
> > > map address in sg_init_one(), right?
> >
> > If the page is highmem, kmap_atomic() establishes a temporary mapping
> > to it in the direct map, such that we have a legit kernel pointer to
> > the memory. Otherwise the memcpy() in zswap also wouldn't work... Am I
> > missing something?
>
> IIUC kmap_atomic() establishes a mapping in the kernel portion of the
> address space, but not a direct map mapping (i.e. not a linear
> mapping), right?
>
> Does virt_addr_valid() check for addresses being in the kernel portion
> of the address space, or it being a linear mapping? I thought it
> checks for the latter.

the latter, right.

>