linux-kernel - Re: [PATCH v1] media: staging: tegra-vde: Reset memory client

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

Message-ID: <24810d9d-eede-f9dd-1dbe-e757cfa2e191@gmail.com>
Date:   Sat, 26 May 2018 13:03:01 +0300
From:   Dmitry Osipenko <digetx@...il.com>
To:     Hans Verkuil <hverkuil@...all.nl>,
        Mauro Carvalho Chehab <mchehab@...nel.org>,
        Thierry Reding <thierry.reding@...il.com>,
        Jonathan Hunter <jonathanh@...dia.com>
Cc:     linux-tegra@...r.kernel.org, linux-media@...r.kernel.org,
        devel@...verdev.osuosl.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v1] media: staging: tegra-vde: Reset memory client

On 20.05.2018 16:48, Dmitry Osipenko wrote:
> DMA requests must be blocked before resetting VDE HW, otherwise it is
> possible to get a memory corruption or a machine hang. Use the reset
> control provided by the Memory Controller to block DMA before resetting
> the VDE HW.
> 
> Signed-off-by: Dmitry Osipenko <digetx@...il.com>
> ---
>  drivers/staging/media/tegra-vde/tegra-vde.c | 42 +++++++++++++++++++--
>  1 file changed, 38 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/staging/media/tegra-vde/tegra-vde.c b/drivers/staging/media/tegra-vde/tegra-vde.c
> index 90177a59b97c..6dd3bf4481be 100644
> --- a/drivers/staging/media/tegra-vde/tegra-vde.c
> +++ b/drivers/staging/media/tegra-vde/tegra-vde.c
> @@ -73,6 +73,7 @@ struct tegra_vde {
>  	struct mutex lock;
>  	struct miscdevice miscdev;
>  	struct reset_control *rst;
> +	struct reset_control *rst_mc;
>  	struct gen_pool *iram_pool;
>  	struct completion decode_completion;
>  	struct clk *clk;
> @@ -850,9 +851,23 @@ static int tegra_vde_ioctl_decode_h264(struct tegra_vde *vde,
>  	 * We rely on the VDE registers reset value, otherwise VDE
>  	 * causes bus lockup.
>  	 */
> +	ret = reset_control_assert(vde->rst_mc);
> +	if (ret) {
> +		dev_err(dev, "DEC start: Failed to assert MC reset: %d\n",
> +			ret);
> +		goto put_runtime_pm;
> +	}
> +
>  	ret = reset_control_reset(vde->rst);
>  	if (ret) {
> -		dev_err(dev, "Failed to reset HW: %d\n", ret);
> +		dev_err(dev, "DEC start: Failed to reset HW: %d\n", ret);
> +		goto put_runtime_pm;
> +	}
> +
> +	ret = reset_control_deassert(vde->rst_mc);
> +	if (ret) {
> +		dev_err(dev, "DEC start: Failed to deassert MC reset: %d\n",
> +			ret);
>  		goto put_runtime_pm;
>  	}
>  
> @@ -880,9 +895,21 @@ static int tegra_vde_ioctl_decode_h264(struct tegra_vde *vde,
>  		ret = timeout;
>  	}
>  
> -	err = reset_control_assert(vde->rst);
> -	if (err)
> -		dev_err(dev, "Failed to assert HW reset: %d\n", err);
> +	/*
> +	 * At first reset memory client to avoid resetting VDE HW in the
> +	 * middle of DMA which could result into memory corruption or hang
> +	 * the whole system.
> +	 */
> +	err = reset_control_assert(vde->rst_mc);
> +	if (!err) {

It occurred to me that there is no need to skip the HW reset if MC resetting
fails. I'll make V2 to change that.

> +		err = reset_control_assert(vde->rst);
> +		if (err)
> +			dev_err(dev,
> +				"DEC end: Failed to assert HW reset: %d\n",
> +				err);
> +	} else {
> +		dev_err(dev, "DEC end: Failed to assert MC reset: %d\n", err);
> +	}
>  
>  put_runtime_pm:
>  	pm_runtime_mark_last_busy(dev);
> @@ -1074,6 +1101,13 @@ static int tegra_vde_probe(struct platform_device *pdev)
>  		return err;
>  	}
>  
> +	vde->rst_mc = devm_reset_control_get_optional(dev, "mc");
> +	if (IS_ERR(vde->rst_mc)) {
> +		err = PTR_ERR(vde->rst_mc);
> +		dev_err(dev, "Could not get MC reset %d\n", err);
> +		return err;
> +	}
> +
>  	irq = platform_get_irq_byname(pdev, "sync-token");
>  	if (irq < 0)
>  		return irq;
>