Received-SPF: pass (google.com: domain of linux-nfs-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <627dd3b048812fe8127058a66fd5a4aafca8242a.camel@kernel.org>
Subject: Re: [PATCH -next 2/5] nfs: fix possible null-ptr-deref when parsing
 param
From:   Jeff Layton <jlayton@kernel.org>
To:     Hawkins Jiawei <yin31149@gmail.com>,
        Trond Myklebust <trond.myklebust@hammerspace.com>,
        Anna Schumaker <anna@kernel.org>
Cc:     18801353760@163.com, linux-kernel@vger.kernel.org,
        linux-fsdevel@vger.kernel.org, linux-nfs@vger.kernel.org
Date:   Mon, 24 Oct 2022 06:53:46 -0400
In-Reply-To: <20221023163945.39920-3-yin31149@gmail.com>
References: <20221023163945.39920-1-yin31149@gmail.com>
         <20221023163945.39920-3-yin31149@gmail.com>
Content-Type: text/plain; charset="ISO-8859-15"
Content-Transfer-Encoding: quoted-printable
User-Agent: Evolution 3.44.4 (3.44.4-2.fc36) 
MIME-Version: 1.0
Precedence: bulk

On Mon, 2022-10-24 at 00:39 +0800, Hawkins Jiawei wrote:
> According to commit "vfs: parse: deal with zero length string value",
> kernel will set the param->string to null pointer in vfs_parse_fs_string(=
)
> if fs string has zero length.
>=20
> Yet the problem is that, nfs_fs_context_parse_param() will dereferences t=
he
> param->string, without checking whether it is a null pointer, which may
> trigger a null-ptr-deref bug.
>=20
> This patch solves it by adding sanity check on param->string
> in nfs_fs_context_parse_param().
>=20
> Signed-off-by: Hawkins Jiawei <yin31149@gmail.com>
> ---
>  fs/nfs/fs_context.c | 6 ++++++
>  1 file changed, 6 insertions(+)
>=20
> diff --git a/fs/nfs/fs_context.c b/fs/nfs/fs_context.c
> index 4da701fd1424..0c330bc13ef2 100644
> --- a/fs/nfs/fs_context.c
> +++ b/fs/nfs/fs_context.c
> @@ -684,6 +684,8 @@ static int nfs_fs_context_parse_param(struct fs_conte=
xt *fc,
>  			return ret;
>  		break;
>  	case Opt_vers:
> +		if (!param->string)
> +			goto out_invalid_value;
>  		trace_nfs_mount_assign(param->key, param->string);
>  		ret =3D nfs_parse_version_string(fc, param->string);
>  		if (ret < 0)
> @@ -696,6 +698,8 @@ static int nfs_fs_context_parse_param(struct fs_conte=
xt *fc,
>  		break;
> =20
>  	case Opt_proto:
> +		if (!param->string)
> +			goto out_invalid_value;
>  		trace_nfs_mount_assign(param->key, param->string);
>  		protofamily =3D AF_INET;
>  		switch (lookup_constant(nfs_xprt_protocol_tokens, param->string, -1)) =
{
> @@ -732,6 +736,8 @@ static int nfs_fs_context_parse_param(struct fs_conte=
xt *fc,
>  		break;
> =20
>  	case Opt_mountproto:
> +		if (!param->string)
> +			goto out_invalid_value;
>  		trace_nfs_mount_assign(param->key, param->string);
>  		mountfamily =3D AF_INET;
>  		switch (lookup_constant(nfs_xprt_protocol_tokens, param->string, -1)) =
{


Looks reasonable. I took a quick look for other fsparam_string values
that might not handle a NULL pointer correctly, but I didn't see any.

Reviewed-by: Jeff Layton <jlayton@kernel.org>