linux-kernel - Re: [PATCH] [v2] media: vsp1: Fix runtime PM imbalance on error

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <20200616025602.GG29596@pendragon.ideasonboard.com>
Date:   Tue, 16 Jun 2020 05:56:02 +0300
From:   Laurent Pinchart <laurent.pinchart@...asonboard.com>
To:     Dinghao Liu <dinghao.liu@....edu.cn>
Cc:     kjlu@....edu,
        Kieran Bingham <kieran.bingham+renesas@...asonboard.com>,
        Mauro Carvalho Chehab <mchehab@...nel.org>,
        linux-media@...r.kernel.org, linux-renesas-soc@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH] [v2] media: vsp1: Fix runtime PM imbalance on error

Hello Dinghao,

Thank you for the patch.

On Mon, Jun 08, 2020 at 01:29:19PM +0800, Dinghao Liu wrote:
> pm_runtime_get_sync() increments the runtime PM usage counter even
> when it returns an error code. Thus a pairing decrement is needed on
> the error handling path to keep the counter balanced.
> 
> Signed-off-by: Dinghao Liu <dinghao.liu@....edu.cn>

Reviewed-by: Laurent Pinchart <laurent.pinchart@...asonboard.com>

I have however received multiple similar patches recently, for different
drivers. I've CC'ed Rafael, the PM maintainer, in one of those e-mail
threads, and questioned whether we should really mass-patch drivers, or
fix the issue in pm_runtime_get_sync(). I'll defer pushing this patch
until that discussion comes to a conclusion.

> ---
> 
> Changelog:
> 
> v2: - Fix the imbalance in vsp1_device_get().
>       Use vsp1_device_get() and vsp1_device_put()
>       to replace pm_runtime_get_sync() and
>       pm_runtime_put_sync() in vsp1_probe().
> ---
>  drivers/media/platform/vsp1/vsp1_drv.c | 11 ++++++++---
>  1 file changed, 8 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/media/platform/vsp1/vsp1_drv.c b/drivers/media/platform/vsp1/vsp1_drv.c
> index c650e45bb0ad..dc62533cf32c 100644
> --- a/drivers/media/platform/vsp1/vsp1_drv.c
> +++ b/drivers/media/platform/vsp1/vsp1_drv.c
> @@ -562,7 +562,12 @@ int vsp1_device_get(struct vsp1_device *vsp1)
>  	int ret;
>  
>  	ret = pm_runtime_get_sync(vsp1->dev);
> -	return ret < 0 ? ret : 0;
> +	if (ret < 0) {
> +		pm_runtime_put_noidle(vsp1->dev);
> +		return ret;
> +	}
> +
> +	return 0;
>  }
>  
>  /*
> @@ -845,12 +850,12 @@ static int vsp1_probe(struct platform_device *pdev)
>  	/* Configure device parameters based on the version register. */
>  	pm_runtime_enable(&pdev->dev);
>  
> -	ret = pm_runtime_get_sync(&pdev->dev);
> +	ret = vsp1_device_get(vsp1);
>  	if (ret < 0)
>  		goto done;
>  
>  	vsp1->version = vsp1_read(vsp1, VI6_IP_VERSION);
> -	pm_runtime_put_sync(&pdev->dev);
> +	vsp1_device_put(vsp1);
>  
>  	for (i = 0; i < ARRAY_SIZE(vsp1_device_infos); ++i) {
>  		if ((vsp1->version & VI6_IP_VERSION_MODEL_MASK) ==

-- 
Regards,

Laurent Pinchart