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:     Pavel Begunkov <asml.silence@gmail.com>
CC:     "io-uring@vger.kernel.org" <io-uring@vger.kernel.org>,
        Networking <netdev@vger.kernel.org>,
        "bpf@vger.kernel.org" <bpf@vger.kernel.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        Jens Axboe <axboe@kernel.dk>,
        Alexei Starovoitov <ast@kernel.org>,
        Daniel Borkmann <daniel@iogearbox.net>,
        Andrii Nakryiko <andrii@kernel.org>, Martin Lau <kafai@fb.com>,
        Yonghong Song <yhs@fb.com>,
        John Fastabend <john.fastabend@gmail.com>,
        KP Singh <kpsingh@kernel.org>,
        Horst Schirmeier <horst.schirmeier@tu-dortmund.de>,
        "Franz-B . Tuneke" <franz-bernhard.tuneke@tu-dortmund.de>,
        Christian Dietrich <stettberger@dokucode.de>
Subject: Re: [PATCH 15/23] io_uring: enable BPF to submit SQEs
Thread-Topic: [PATCH 15/23] io_uring: enable BPF to submit SQEs
Thread-Index: AQHXTLmA+z0jtjqTr0CEF/EO+2xSZ6rtEPsA
Date:   Fri, 21 May 2021 00:06:59 +0000
Message-ID: <71AABD2E-3B7F-4869-B909-D84C612FA18D@fb.com>
References: <cover.1621424513.git.asml.silence@gmail.com>
 <8ec8373d406d1fcb41719e641799dcc5c0455db3.1621424513.git.asml.silence@gmail.com>
In-Reply-To: <8ec8373d406d1fcb41719e641799dcc5c0455db3.1621424513.git.asml.silence@gmail.com>
Accept-Language: en-US
Content-Language: en-US
Content-Type: text/plain; charset="us-ascii"
Content-ID: <E11A3818F8FF5D4489EB042CA6AA7A09@namprd15.prod.outlook.com>
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-MS-Exchange-CrossTenant-Network-Message-Id: e78fcd02-b3d7-4265-f292-08d91bec5a97
X-MS-Exchange-CrossTenant-originalarrivaltime: 21 May 2021 00:06:59.5325
 (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: GXufTvDNjGbpz4+y6l9/BWev/PQigiwNyjAkmZlBhC9dwDwlOEBL7qp6Dtl8y0rR7RVPM0u8l0xU5gvJjk8+Vw==
X-MS-Exchange-Transport-CrossTenantHeadersStamped: BYAPR15MB2886
X-OriginatorOrg: fb.com
X-Proofpoint-ORIG-GUID: xgZuqUN_gi7rQ2erfqGiqQb7GibUkgh2
X-Proofpoint-GUID: xgZuqUN_gi7rQ2erfqGiqQb7GibUkgh2
X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10434:6.0.391,18.0.761
 definitions=2021-05-20_07:2021-05-20,2021-05-20 signatures=0
X-Proofpoint-Spam-Details: rule=fb_default_notspam policy=fb_default score=0 mlxscore=0 phishscore=0
 mlxlogscore=999 impostorscore=0 lowpriorityscore=0 adultscore=0
 malwarescore=0 suspectscore=0 clxscore=1015 priorityscore=1501 spamscore=0
 bulkscore=0 classifier=spam adjust=0 reason=mlx scancount=1
 engine=8.12.0-2104190000 definitions=main-2105200157
X-FB-Internal: deliver
Precedence: bulk
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org


> On May 19, 2021, at 7:13 AM, Pavel Begunkov <asml.silence@gmail.com> wrot=
e:
>=20
> Add a BPF_FUNC_iouring_queue_sqe BPF function as a demonstration of
> submmiting a new request by a BPF request.
>=20
> Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
> ---
> fs/io_uring.c            | 51 ++++++++++++++++++++++++++++++++++++----
> include/uapi/linux/bpf.h |  1 +
> 2 files changed, 48 insertions(+), 4 deletions(-)
>=20
> diff --git a/fs/io_uring.c b/fs/io_uring.c
> index 20fddc5945f2..aae786291c57 100644
> --- a/fs/io_uring.c
> +++ b/fs/io_uring.c
> @@ -882,6 +882,7 @@ struct io_defer_entry {
> };
>=20
> struct io_bpf_ctx {
> +	struct io_ring_ctx	*ctx;
> };
>=20
> struct io_op_def {
> @@ -6681,7 +6682,8 @@ static int io_init_req(struct io_ring_ctx *ctx, str=
uct io_kiocb *req,
> 			ret =3D -EBADF;
> 	}
>=20
> -	state->ios_left--;
> +	if (state->ios_left > 1)
> +		state->ios_left--;
> 	return ret;
> }
>=20
> @@ -10345,10 +10347,50 @@ static int __io_uring_register(struct io_ring_c=
tx *ctx, unsigned opcode,
> 	return ret;
> }
>=20
> +BPF_CALL_3(io_bpf_queue_sqe, struct io_bpf_ctx *,		bpf_ctx,
> +			     const struct io_uring_sqe *,	sqe,
> +			     u32,				sqe_len)
> +{
> +	struct io_ring_ctx *ctx =3D bpf_ctx->ctx;
> +	struct io_kiocb *req;
> +
> +	if (sqe_len !=3D sizeof(struct io_uring_sqe))
> +		return -EINVAL;
> +
> +	req =3D io_alloc_req(ctx);
> +	if (unlikely(!req))
> +		return -ENOMEM;
> +	if (!percpu_ref_tryget_many(&ctx->refs, 1)) {
> +		kmem_cache_free(req_cachep, req);
> +		return -EAGAIN;
> +	}
> +	percpu_counter_add(&current->io_uring->inflight, 1);
> +	refcount_add(1, &current->usage);
> +
> +	/* returns number of submitted SQEs or an error */
> +	return !io_submit_sqe(ctx, req, sqe);
> +}
> +
> +const struct bpf_func_proto io_bpf_queue_sqe_proto =3D {
> +	.func =3D io_bpf_queue_sqe,
> +	.gpl_only =3D false,
> +	.ret_type =3D RET_INTEGER,
> +	.arg1_type =3D ARG_PTR_TO_CTX,
> +	.arg2_type =3D ARG_PTR_TO_MEM,
> +	.arg3_type =3D ARG_CONST_SIZE,
> +};
> +
> static const struct bpf_func_proto *
> io_bpf_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog)
> {
> -	return bpf_base_func_proto(func_id);
> +	switch (func_id) {
> +	case BPF_FUNC_copy_from_user:
> +		return prog->aux->sleepable ? &bpf_copy_from_user_proto : NULL;
> +	case BPF_FUNC_iouring_queue_sqe:
> +		return prog->aux->sleepable ? &io_bpf_queue_sqe_proto : NULL;
> +	default:
> +		return bpf_base_func_proto(func_id);
> +	}
> }
>=20
> static bool io_bpf_is_valid_access(int off, int size,
> @@ -10379,9 +10421,10 @@ static void io_bpf_run(struct io_kiocb *req, uns=
igned int issue_flags)
> 		     atomic_read(&req->task->io_uring->in_idle)))
> 		goto done;
>=20
> -	memset(&bpf_ctx, 0, sizeof(bpf_ctx));
> +	bpf_ctx.ctx =3D ctx;
> 	prog =3D req->bpf.prog;
>=20
> +	io_submit_state_start(&ctx->submit_state, 1);
> 	if (prog->aux->sleepable) {
> 		rcu_read_lock();
> 		bpf_prog_run_pin_on_cpu(req->bpf.prog, &bpf_ctx);
> @@ -10389,7 +10432,7 @@ static void io_bpf_run(struct io_kiocb *req, unsi=
gned int issue_flags)
> 	} else {
> 		bpf_prog_run_pin_on_cpu(req->bpf.prog, &bpf_ctx);
> 	}
> -
> +	io_submit_state_end(&ctx->submit_state, ctx);
> 	ret =3D 0;
> done:
> 	__io_req_complete(req, issue_flags, ret, 0);
> diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
> index de544f0fbeef..cc268f749a7d 100644
> --- a/include/uapi/linux/bpf.h
> +++ b/include/uapi/linux/bpf.h
> @@ -4082,6 +4082,7 @@ union bpf_attr {
> 	FN(ima_inode_hash),		\
> 	FN(sock_from_file),		\
> 	FN(check_mtu),			\
> +	FN(iouring_queue_sqe),		\

We need to describe this function in the comment above, just like 20/23 doe=
s.=20

> 	/* */
>=20
> /* integer value in 'imm' field of BPF_CALL instruction selects which hel=
per
> --=20
> 2.31.1
>=20