Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Thu, 3 Feb 2022 10:35:10 -0700
From:   Mathieu Poirier <mathieu.poirier@linaro.org>
To:     Deepak Kumar Singh <quic_deesin@quicinc.com>
Cc:     bjorn.andersson@linaro.org, swboyd@chromium.org,
        quic_clew@quicinc.com, linux-kernel@vger.kernel.org,
        linux-arm-msm@vger.kernel.org, linux-remoteproc@vger.kernel.org,
        Ohad Ben-Cohen <ohad@wizery.com>
Subject: Re: [PATCH V1 1/3] rpmsg: glink: Free device context only when cdev
 not in use
Message-ID: <20220203173510.GA2982815@p14s>
References: <1643223886-28170-1-git-send-email-quic_deesin@quicinc.com>
 <1643223886-28170-2-git-send-email-quic_deesin@quicinc.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <1643223886-28170-2-git-send-email-quic_deesin@quicinc.com>
Precedence: bulk

Hi Deepak,

On Thu, Jan 27, 2022 at 12:34:44AM +0530, Deepak Kumar Singh wrote:
> Struct device holding cdev should not be freed unless cdev
> is not in use. It is possible that user space has opened
> char device while kernel has freed the associated struct
> device context.
> 
> Mark dev kobj as parent of cdev, so that chardev_add gets
> an extra reference to dev. This ensures device context is not
> freed until cdev is is not in uses.
> ---
>  drivers/rpmsg/rpmsg_char.c | 2 ++
>  1 file changed, 2 insertions(+)
> 
> diff --git a/drivers/rpmsg/rpmsg_char.c b/drivers/rpmsg/rpmsg_char.c
> index c03a118..72ee101 100644
> --- a/drivers/rpmsg/rpmsg_char.c
> +++ b/drivers/rpmsg/rpmsg_char.c
> @@ -417,6 +417,7 @@ static int rpmsg_eptdev_create(struct rpmsg_ctrldev *ctrldev,
>  	dev->id = ret;
>  	dev_set_name(dev, "rpmsg%d", ret);
>  
> +	cdev_set_parent(&eptdev->cdev, &dev->kobj);
>  	ret = cdev_add(&eptdev->cdev, dev->devt, 1);

This issue should have been fixed when cdev_add() was replaced by
cdev_device_add(), something you will find on v5.17-rc2.

Also, this set is generating checkpatch warnings and as such I will not review
the other patches in it. 

Thanks,
Mathieu

>  	if (ret)
>  		goto free_ept_ida;
> @@ -533,6 +534,7 @@ static int rpmsg_chrdev_probe(struct rpmsg_device *rpdev)
>  	dev->id = ret;
>  	dev_set_name(&ctrldev->dev, "rpmsg_ctrl%d", ret);
>  
> +	cdev_set_parent(&ctrldev->cdev, &dev->kobj);
>  	ret = cdev_add(&ctrldev->cdev, dev->devt, 1);
>  	if (ret)
>  		goto free_ctrl_ida;
> -- 
> 2.7.4
>