linux-kernel - Re: [PATCH 3/4] dmaengine: zynqmp_dma: Add conditions for return value check

Open Source and information security mailing list archives

Message-ID: <20210923144025.GC30905@pengutronix.de>
Date:   Thu, 23 Sep 2021 16:40:25 +0200
From:   Michael Tretter <m.tretter@...gutronix.de>
To:     Harini Katakam <harini.katakam@...inx.com>
Cc:     vkoul@...nel.org, romain.perier@...il.com, allen.lkml@...il.com,
        yukuai3@...wei.com, dmaengine@...r.kernel.org,
        linux-arm-kernel@...ts.infradead.org, linux-kernel@...r.kernel.org,
        harinikatakamlinux@...il.com, michal.simek@...inx.com,
        radhey.shyam.pandey@...inx.com, shravya.kumbham@...inx.com,
        kernel@...gutronix.de
Subject: Re: [PATCH 3/4] dmaengine: zynqmp_dma: Add conditions for return
 value check

On Tue, 14 Sep 2021 13:58:16 +0530, Harini Katakam wrote:
> From: Shravya Kumbham <shravya.kumbham@...inx.com>
> 
> Add condition to check the return value of dma_async_device_register
> and pm_runtime_get_sync functions.
> 
> Addresses-Coverity: Event check_return.
> Signed-off-by: Shravya Kumbham <shravya.kumbham@...inx.com>
> Reviewed-by: Radhey Shyam Pandey <radhey.shyam.pandey@...inx.com>
> ---
>  drivers/dma/xilinx/zynqmp_dma.c | 12 ++++++++++--
>  1 file changed, 10 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/dma/xilinx/zynqmp_dma.c b/drivers/dma/xilinx/zynqmp_dma.c
> index d28b9ffb4309..588460e56ab8 100644
> --- a/drivers/dma/xilinx/zynqmp_dma.c
> +++ b/drivers/dma/xilinx/zynqmp_dma.c
> @@ -1080,7 +1080,11 @@ static int zynqmp_dma_probe(struct platform_device *pdev)
>  	pm_runtime_set_autosuspend_delay(zdev->dev, ZDMA_PM_TIMEOUT);
>  	pm_runtime_use_autosuspend(zdev->dev);
>  	pm_runtime_enable(zdev->dev);
> -	pm_runtime_get_sync(zdev->dev);
> +	ret = pm_runtime_get_sync(zdev->dev);
> +	if (ret < 0) {
> +		dev_err(zdev->dev, "pm_runtime_get_sync() failed\n");
> +		pm_runtime_disable(zdev->dev);
> +	}

Thanks for the patch.

You need to call pm_runtime_put() on the error path. Or you could use
pm_runtime_resume_and_get(), which does this automatically.

I am wondering, if it wouldn't be cleaner to make this dependent on
pm_runtime_enabled() and avoiding pm_runtime_disable() on the error path
altogether.

Michael

>  	if (!pm_runtime_enabled(zdev->dev)) {
>  		ret = zynqmp_dma_runtime_resume(zdev->dev);
>  		if (ret)
> @@ -1096,7 +1100,11 @@ static int zynqmp_dma_probe(struct platform_device *pdev)
>  	p->dst_addr_widths = BIT(zdev->chan->bus_width / 8);
>  	p->src_addr_widths = BIT(zdev->chan->bus_width / 8);
>  
> -	dma_async_device_register(&zdev->common);
> +	ret = dma_async_device_register(&zdev->common);
> +	if (ret) {
> +		dev_err(zdev->dev, "failed to register the dma device\n");
> +		goto free_chan_resources;
> +	}
>  
>  	ret = of_dma_controller_register(pdev->dev.of_node,
>  					 of_zynqmp_dma_xlate, zdev);
> -- 
> 2.17.1
> 
> 
> _______________________________________________
> linux-arm-kernel mailing list
> linux-arm-kernel@...ts.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
> 

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