linux-kernel - Re: [PATCH v1 06/11] ti948: Reconfigure in the alive check when device returns

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: <20190611181046.GU5016@pendragon.ideasonboard.com>
Date:   Tue, 11 Jun 2019 21:10:46 +0300
From:   Laurent Pinchart <laurent.pinchart@...asonboard.com>
To:     Michael Drake <michael.drake@...ethink.co.uk>
Cc:     Andrzej Hajda <a.hajda@...sung.com>,
        dri-devel@...ts.freedesktop.org, devicetree@...r.kernel.org,
        linux-kernel@...r.kernel.org, David Airlie <airlied@...ux.ie>,
        Daniel Vetter <daniel@...ll.ch>,
        Rob Herring <robh+dt@...nel.org>,
        Mark Rutland <mark.rutland@....com>,
        linux-kernel@...ts.codethink.co.uk,
        Patrick Glaser <pglaser@...la.com>, Nate Case <ncase@...la.com>
Subject: Re: [PATCH v1 06/11] ti948: Reconfigure in the alive check when
 device returns

Hi Michael,

Thank you for the patch.

On Tue, Jun 11, 2019 at 03:04:07PM +0100, Michael Drake wrote:
> If the alive check detects a transition to the alive state,
> the device configuration is rewritten.

This seems like a big hack. You will have at the very least to explain
why this is needed, and why you can't configure the device in response
to drm_bridge operation calls.

> Signed-off-by: Michael Drake <michael.drake@...ethink.co.uk>
> Cc: Patrick Glaser <pglaser@...la.com>
> Cc: Nate Case <ncase@...la.com>
> ---
>  drivers/gpu/drm/bridge/ti948.c | 19 ++++++++++++++++++-
>  1 file changed, 18 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/gpu/drm/bridge/ti948.c b/drivers/gpu/drm/bridge/ti948.c
> index 86daa3701b91..b5c766711c4b 100644
> --- a/drivers/gpu/drm/bridge/ti948.c
> +++ b/drivers/gpu/drm/bridge/ti948.c
> @@ -132,6 +132,8 @@ struct ti948_reg_val {
>   * @reg_names:   Array of regulator names, or NULL.
>   * @regs:        Array of regulators, or NULL.
>   * @reg_count:   Number of entries in reg_names and regs arrays.
> + * @alive_check: Context for the alive checking work item.
> + * @alive:       Whether the device is alive or not (alive_check).
>   */
>  struct ti948_ctx {
>  	struct i2c_client *i2c;
> @@ -141,6 +143,8 @@ struct ti948_ctx {
>  	const char **reg_names;
>  	struct regulator **regs;
>  	size_t reg_count;
> +	struct delayed_work alive_check;
> +	bool alive;
>  };
>  
>  static bool ti948_readable_reg(struct device *dev, unsigned int reg)
> @@ -346,6 +350,8 @@ static int ti948_power_on(struct ti948_ctx *ti948)
>  	if (ret != 0)
>  		return ret;
>  
> +	ti948->alive = true;
> +
>  	msleep(500);
>  
>  	return 0;
> @@ -356,6 +362,8 @@ static int ti948_power_off(struct ti948_ctx *ti948)
>  	int i;
>  	int ret;
>  
> +	ti948->alive = false;
> +
>  	for (i = ti948->reg_count; i > 0; i--) {
>  		dev_info(&ti948->i2c->dev, "Disabling %s regulator\n",
>  				ti948->reg_names[i - 1]);
> @@ -388,8 +396,17 @@ static void ti948_alive_check(struct work_struct *work)
>  {
>  	struct delayed_work *dwork = to_delayed_work(work);
>  	struct ti948_ctx *ti948 = delayed_work_to_ti948_ctx(dwork);
> +	int ret = ti948_device_check(ti948);
>  
> -	dev_info(&ti948->i2c->dev, "%s Alive check!\n", __func__);
> +	if (ti948->alive == false && ret == 0) {
> +		dev_info(&ti948->i2c->dev, "Device has come back to life!\n");
> +		ti948_write_config_seq(ti948);
> +		ti948->alive = true;
> +
> +	} else if (ti948->alive == true && ret != 0) {
> +		dev_info(&ti948->i2c->dev, "Device has stopped responding\n");
> +		ti948->alive = false;
> +	}
>  
>  	/* Reschedule ourself for the next check. */
>  	schedule_delayed_work(&ti948->alive_check, TI948_ALIVE_CHECK_DELAY);

-- 
Regards,

Laurent Pinchart