From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <dev-bounces@dpdk.org>
Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124])
	by inbox.dpdk.org (Postfix) with ESMTP id 27ECE43E7A;
	Mon, 15 Apr 2024 15:15:26 +0200 (CEST)
Received: from mails.dpdk.org (localhost [127.0.0.1])
	by mails.dpdk.org (Postfix) with ESMTP id 963174026C;
	Mon, 15 Apr 2024 15:15:25 +0200 (CEST)
Received: from dkmailrelay1.smartsharesystems.com
 (smartserver.smartsharesystems.com [77.243.40.215])
 by mails.dpdk.org (Postfix) with ESMTP id 40597400EF
 for <dev@dpdk.org>; Mon, 15 Apr 2024 15:15:24 +0200 (CEST)
Received: from smartserver.smartsharesystems.com
 (smartserver.smartsharesys.local [192.168.4.10])
 by dkmailrelay1.smartsharesystems.com (Postfix) with ESMTP id 2AC462073D;
 Mon, 15 Apr 2024 15:15:22 +0200 (CEST)
Content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Subject: RE: [PATCH v6] ethdev: fix strict aliasing lead to link cannot be up
Date: Mon, 15 Apr 2024 15:15:18 +0200
X-MimeOLE: Produced By Microsoft Exchange V6.5
Message-ID: <98CBD80474FA8B44BF855DF32C47DC35E9F3A2@smartserver.smartshare.dk>
In-Reply-To: <20240413084840.33124-1-fengchengwen@huawei.com>
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
Thread-Topic: [PATCH v6] ethdev: fix strict aliasing lead to link cannot be up
Thread-Index: AdqNf7QQx0ZXRWtRTKeOFVHC6Hx+QABtWBbQ
References: <20240411030749.41874-1-fengchengwen@huawei.com>
 <20240413084840.33124-1-fengchengwen@huawei.com>
From: =?iso-8859-1?Q?Morten_Br=F8rup?= <mb@smartsharesystems.com>
To: "Chengwen Feng" <fengchengwen@huawei.com>
Cc: <dev@dpdk.org>, <huangdengdui@huawei.com>, <stephen@networkplumber.org>,
 <roretzla@linux.microsoft.com>, <thomas@monjalon.net>,
 <ferruh.yigit@amd.com>
X-BeenThere: dev@dpdk.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: DPDK patches and discussions <dev.dpdk.org>
List-Unsubscribe: <https://mails.dpdk.org/options/dev>,
 <mailto:dev-request@dpdk.org?subject=unsubscribe>
List-Archive: <http://mails.dpdk.org/archives/dev/>
List-Post: <mailto:dev@dpdk.org>
List-Help: <mailto:dev-request@dpdk.org?subject=help>
List-Subscribe: <https://mails.dpdk.org/listinfo/dev>,
 <mailto:dev-request@dpdk.org?subject=subscribe>
Errors-To: dev-bounces@dpdk.org

> @@ -1701,12 +1696,8 @@ static inline void
>  rte_eth_linkstatus_get(const struct rte_eth_dev *dev,
>  		       struct rte_eth_link *link)
>  {
> -	RTE_ATOMIC(uint64_t) *src =3D (uint64_t __rte_atomic *)&(dev->data-
> >dev_link);
> -	uint64_t *dst =3D (uint64_t *)link;
> -
> -	RTE_BUILD_BUG_ON(sizeof(*link) !=3D sizeof(uint64_t));
> -
> -	*dst =3D rte_atomic_load_explicit(src, rte_memory_order_seq_cst);
> +	link->val64 =3D rte_atomic_load_explicit(&dev->data->dev_link.val64,
> +					       rte_memory_order_seq_cst);
>  }

rte_eth_linkstatus_get() may be called by rte_eth_link_get():
https://elixir.bootlin.com/dpdk/v24.03/source/lib/ethdev/rte_ethdev.c#L29=
86

The application may call rte_eth_link_get() with the "link" parameter =
pointing to memory shared by other threads.

So link->val64 must be stored atomically:
rte_atomic_store_explicit(&link->val64,
		rte_atomic_load_explicit(&dev->data->dev_link.val64,
		rte_memory_order_seq_cst),
		rte_memory_order_seq_cst);

With the above modification,
Acked-by: Morten Br=F8rup <mb@smartsharesystems.com>