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;
From:   Song Liu <songliubraving@fb.com>
To:     Andrii Nakryiko <andrii@kernel.org>
CC:     bpf <bpf@vger.kernel.org>, Networking <netdev@vger.kernel.org>,
        "Alexei Starovoitov" <ast@fb.com>,
        "daniel@iogearbox.net" <daniel@iogearbox.net>,
        Kernel Team <Kernel-team@fb.com>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        "rafael@kernel.org" <rafael@kernel.org>,
        "jeyu@kernel.org" <jeyu@kernel.org>,
        Arnaldo Carvalho de Melo <acme@redhat.com>,
        Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Subject: Re: [PATCH v4 bpf-next 1/5] bpf: add in-kernel split BTF support
Thread-Topic: [PATCH v4 bpf-next 1/5] bpf: add in-kernel split BTF support
Thread-Index: AQHWtwAHoXtpfdzO80ynz0OJjXLTl6nBpcmA
Date:   Tue, 10 Nov 2020 17:49:53 +0000
Message-ID: <695E976D-DECA-4BE1-BFB0-771878B9CFCD@fb.com>
References: <20201110011932.3201430-1-andrii@kernel.org>
 <20201110011932.3201430-2-andrii@kernel.org>
In-Reply-To: <20201110011932.3201430-2-andrii@kernel.org>
Accept-Language: en-US
Content-Language: en-US
Content-Type: text/plain; charset="us-ascii"
Content-ID: <012128130996894097D84B39F76C51B6@namprd15.prod.outlook.com>
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-MS-Exchange-CrossTenant-Network-Message-Id: d7d949e2-66ab-4dff-150d-08d885a107c8
X-MS-Exchange-CrossTenant-originalarrivaltime: 10 Nov 2020 17:49:53.8098
 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Hosted
X-MS-Exchange-CrossTenant-id: 8ae927fe-1255-47a7-a2af-5f3a069daaa2
X-MS-Exchange-CrossTenant-mailboxtype: HOSTED
X-MS-Exchange-CrossTenant-userprincipalname: J7B/7wZrfL7IwRo4Qfym/siokrprsxrrrKfS13j/eKFUv5FkP7WPrvNsTRhI3qT7J/Lk8UkxLX2yhSmTnybpuQ==
X-MS-Exchange-Transport-CrossTenantHeadersStamped: BYAPR15MB3094
X-OriginatorOrg: fb.com
X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10434:6.0.312,18.0.737
 definitions=2020-11-10_07:2020-11-10,2020-11-10 signatures=0
X-Proofpoint-Spam-Details: rule=fb_default_notspam policy=fb_default score=0 mlxscore=0 spamscore=0
 mlxlogscore=999 phishscore=0 adultscore=0 suspectscore=0
 priorityscore=1501 bulkscore=0 impostorscore=0 lowpriorityscore=0
 clxscore=1015 malwarescore=0 classifier=spam adjust=0 reason=mlx
 scancount=1 engine=8.12.0-2009150000 definitions=main-2011100125
X-FB-Internal: deliver
Precedence: bulk
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org


> On Nov 9, 2020, at 5:19 PM, Andrii Nakryiko <andrii@kernel.org> wrote:
>=20
> Adjust in-kernel BTF implementation to support a split BTF mode of operat=
ion.
> Changes are mostly mirroring libbpf split BTF changes, with the exception=
 of
> start_id being 0 for in-kernel implementation due to simpler read-only mo=
de.
>=20
> Otherwise, for split BTF logic, most of the logic of jumping to base BTF,
> where necessary, is encapsulated in few helper functions. Type numbering =
and
> string offset in a split BTF are logically continuing where base BTF ends=
, so
> most of the high-level logic is kept without changes.
>=20
> Type verification and size resolution is only doing an added resolution o=
f new
> split BTF types and relies on already cached size and type resolution res=
ults
> in the base BTF.
>=20
> Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
> ---
> kernel/bpf/btf.c | 171 +++++++++++++++++++++++++++++++++--------------
> 1 file changed, 119 insertions(+), 52 deletions(-)
>=20
> diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
> index 6324de8c59f7..727c1c27053f 100644
> --- a/kernel/bpf/btf.c
> +++ b/kernel/bpf/btf.c
> @@ -203,12 +203,17 @@ struct btf {
> 	const char *strings;
> 	void *nohdr_data;
> 	struct btf_header hdr;
> -	u32 nr_types;
> +	u32 nr_types; /* includes VOID for base BTF */
> 	u32 types_size;
> 	u32 data_size;
> 	refcount_t refcnt;
> 	u32 id;
> 	struct rcu_head rcu;
> +
> +	/* split BTF support */
> +	struct btf *base_btf;
> +	u32 start_id; /* first type ID in this BTF (0 for base BTF) */
> +	u32 start_str_off; /* first string offset (0 for base BTF) */
> };
>=20
> enum verifier_phase {
> @@ -449,14 +454,27 @@ static bool btf_type_is_datasec(const struct btf_ty=
pe *t)
> 	return BTF_INFO_KIND(t->info) =3D=3D BTF_KIND_DATASEC;
> }
>=20
> +static u32 btf_nr_types_total(const struct btf *btf)
> +{
> +	u32 total =3D 0;
> +
> +	while (btf) {
> +		total +=3D btf->nr_types;
> +		btf =3D btf->base_btf;
> +	}
> +
> +	return total;
> +}
> +
> s32 btf_find_by_name_kind(const struct btf *btf, const char *name, u8 kin=
d)
> {
> 	const struct btf_type *t;
> 	const char *tname;
> -	u32 i;
> +	u32 i, total;
>=20
> -	for (i =3D 1; i <=3D btf->nr_types; i++) {
> -		t =3D btf->types[i];
> +	total =3D btf_nr_types_total(btf);
> +	for (i =3D 1; i < total; i++) {
> +		t =3D btf_type_by_id(btf, i);
> 		if (BTF_INFO_KIND(t->info) !=3D kind)
> 			continue;
>=20
> @@ -599,8 +617,14 @@ static const struct btf_kind_operations *btf_type_op=
s(const struct btf_type *t)
>=20
> static bool btf_name_offset_valid(const struct btf *btf, u32 offset)
> {
> -	return BTF_STR_OFFSET_VALID(offset) &&
> -		offset < btf->hdr.str_len;
> +	if (!BTF_STR_OFFSET_VALID(offset))
> +		return false;
> +
> +	while (offset < btf->start_str_off)
> +		btf =3D btf->base_btf;

Do we need "if (!btf) return false;" in the while loop? (and some other loo=
ps below)

[...]