Received-SPF: pass (google.com: domain of linux-crypto+bounces-1069-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.48.161 as permitted sender) client-ip=147.75.48.161;
Precedence: bulk
MIME-Version: 1.0
References: <0000000000000b05cd060d6b5511@google.com> <CAKEwX=OmWYivf7dg_izW8pn5s5q15+nx-vRMsV47T_qG=dep_Q@mail.gmail.com>
 <CAF8kJuPLEXEXG+4esR6MbRa3iirTrJ7-w3YCorB9iD=gnQ+G3A@mail.gmail.com>
 <CAKEwX=PaFmreqmNrisatSN1=k2kRiYgDksgDze-t=GBD=0iJDg@mail.gmail.com>
 <CAF8kJuPF5ACu8o1P7GqEQRb6p8QShyTVNuzrrY557g+SsddzWA@mail.gmail.com>
 <CAKEwX=NHdr9=hUBiZhnLZyRPsp=JwN3Vkwud2XEn3=pNurYGpQ@mail.gmail.com>
 <f27efd2e-ac65-4f6a-b1b5-c9fb0753d871@bytedance.com> <CAGsJ_4x31mT8TXt4c7ejJoDW1yJhyNqDmJmLZrf2LxMt7Zwg2A@mail.gmail.com>
 <5aff3bcf-ef36-45b3-8ac0-a4b19697419c@bytedance.com> <CAGsJ_4xuuaPnQzkkQVaRyZL6ZdwkiQ_B7_c2baNaCKVg_O7ZQA@mail.gmail.com>
 <e464c55e-d03a-4df9-abef-5ece182d7a2b@bytedance.com> <CAGsJ_4xknGt5Q6vTKHdnqvXi7fKMORwXywurdN09G2uOUJzCSg@mail.gmail.com>
 <CAKEwX=PYK3hUzgm+qfs2sNU686RaE+_M3W4Zo_Q4mTMAgKaB2A@mail.gmail.com> <CAGsJ_4zwd71VmDsErFG-2=V8aqP9dc2qO_hRNcHPj51=CfTJww@mail.gmail.com>
In-Reply-To: <CAGsJ_4zwd71VmDsErFG-2=V8aqP9dc2qO_hRNcHPj51=CfTJww@mail.gmail.com>
From: Barry Song <21cnbao@gmail.com>
Date: Thu, 28 Dec 2023 09:47:43 +0800
Message-ID: <CAGsJ_4zkyyvAzUoxy7-Puk3M5V6yqEXaNysF_HvZfuYwswarGA@mail.gmail.com>
Subject: Re: [syzbot] [crypto?] general protection fault in
 scatterwalk_copychunks (5)
To: Nhat Pham <nphamcs@gmail.com>
Cc: Chengming Zhou <zhouchengming@bytedance.com>, Chris Li <chrisl@kernel.org>, 
	syzbot <syzbot+3eff5e51bf1db122a16e@syzkaller.appspotmail.com>, 
	akpm@linux-foundation.org, davem@davemloft.net, herbert@gondor.apana.org.au, 
	linux-crypto@vger.kernel.org, linux-kernel@vger.kernel.org, 
	syzkaller-bugs@googlegroups.com, yosryahmed@google.com
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Thu, Dec 28, 2023 at 9:43=E2=80=AFAM Barry Song <21cnbao@gmail.com> wrot=
e:
>
> On Thu, Dec 28, 2023 at 7:26=E2=80=AFAM Nhat Pham <nphamcs@gmail.com> wro=
te:
> >
> > On Wed, Dec 27, 2023 at 3:10=E2=80=AFAM Barry Song <21cnbao@gmail.com> =
wrote:
> > >
> > > On Wed, Dec 27, 2023 at 5:16=E2=80=AFPM Chengming Zhou
> > > <zhouchengming@bytedance.com> wrote:
> > > >
> > > > Thanks for your explanation! Maybe it's best for us to return to 2 =
pages
> > > > if no other people's comments. And this really need more documentat=
ion :-)
> >
> > Fine by me. Hmm we're basically wasting one extra page per CPU (since
> > these buffers are per-CPU), correct? That's not ideal, but not *too*
> > bad for now I suppose...
> >
> > >
> > > I agree. we need some doc.
> > >
> > > besides, i actually think we can skip zswap frontend if
> > > over-compression is really
> > > happening.
> >
> > IIUC, zsmalloc already checked that - and most people are (or should
> > be) using zsmalloc for zswap anyway. I wouldn't be opposed to adding
> > an added layer of protection on the zswap side, but not super high
> > priority I'd say.
>
> Thanks for this info. I guess you mean the below ?
> unsigned long zs_malloc(struct zs_pool *pool, size_t size, gfp_t gfp)
> {
>         ...
>
>         if (unlikely(!size || size > ZS_MAX_ALLOC_SIZE))
>                 return (unsigned long)ERR_PTR(-EINVAL);

BTW, do you think zsmalloc is worth a patch to change the ret from
EINVAL to ENOSPC?
This seems more sensible and matches the behaviour of zswap, and zbud, z3fo=
ld.

        ret =3D zpool_malloc(zpool, dlen, gfp, &handle);
        if (ret =3D=3D -ENOSPC) {
                zswap_reject_compress_poor++;
                goto put_dstmem;
        }
        if (ret) {
                zswap_reject_alloc_fail++;
                goto put_dstmem;
        }
        buf =3D z

>
> }
>
> i find zbud also has similar code:
> static int zbud_alloc(struct zbud_pool *pool, size_t size, gfp_t gfp,
>                         unsigned long *handle)
> {
>         int chunks, i, freechunks;
>         struct zbud_header *zhdr =3D NULL;
>         enum buddy bud;
>         struct page *page;
>
>         if (!size || (gfp & __GFP_HIGHMEM))
>                 return -EINVAL;
>         if (size > PAGE_SIZE - ZHDR_SIZE_ALIGNED - CHUNK_SIZE)
>                 return -ENOSPC;
>
> and z3fold,
>
> static int z3fold_alloc(struct z3fold_pool *pool, size_t size, gfp_t gfp,
>                         unsigned long *handle)
> {
>         int chunks =3D size_to_chunks(size);
>         struct z3fold_header *zhdr =3D NULL;
>         struct page *page =3D NULL;
>         enum buddy bud;
>         bool can_sleep =3D gfpflags_allow_blocking(gfp);
>
>         if (!size || (gfp & __GFP_HIGHMEM))
>                 return -EINVAL;
>
>         if (size > PAGE_SIZE)
>                 return -ENOSPC;
>
>
> Thus, I agree that another layer to check size in zswap isn't necessary n=
ow.

Thanks
Barry