Re: [PATCH] drm/bridge: dumb-vga-dac: Fix dereferencing -ENODEV DDC channel

From: Neil Armstrong <narmstrong@baylibre.com>
To: Geert Uytterhoeven <geert+renesas@glider.be>,
	Guenter Roeck <linux@roeck-us.net>
Cc: Andrzej Pietrasiewicz <andrzej.p@collabora.com>,
	Andrzej Hajda <a.hajda@samsung.com>,
	Laurent Pinchart <Laurent.pinchart@ideasonboard.com>,
	Jonas Karlman <jonas@kwiboo.se>,
	Jernej Skrabec <jernej.skrabec@siol.net>,
	David Airlie <airlied@linux.ie>, Daniel Vetter <daniel@ffwll.ch>,
	Sam Ravnborg <sam@ravnborg.org>,
	Emil Velikov <emil.velikov@collabora.com>,
	dri-devel@lists.freedesktop.org,
	linux-renesas-soc@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] drm/bridge: dumb-vga-dac: Fix dereferencing -ENODEV DDC channel
Date: Tue, 13 Aug 2019 14:01:26 +0200	[thread overview]
Message-ID: <29ff3bfd-57ee-9c64-3706-555edc8b4675@baylibre.com> (raw)
In-Reply-To: <20190813093046.4976-1-geert+renesas@glider.be>

Hi,

On 13/08/2019 11:30, Geert Uytterhoeven wrote:
> If the VGA connector has no DDC channel, an error pointer will be
> dereferenced, e.g. on Salvator-XS:
> 
>     Unable to handle kernel NULL pointer dereference at virtual address 000000000000017d
>     ...
>     Call trace:
>      sysfs_do_create_link_sd.isra.0+0x40/0x108
>      sysfs_create_link+0x20/0x40
>      drm_sysfs_connector_add+0xa8/0xc8
>      drm_connector_register.part.3+0x54/0xb0
>      drm_connector_register_all+0xb0/0xd0
>      drm_modeset_register_all+0x54/0x88
>      drm_dev_register+0x18c/0x1d8
>      rcar_du_probe+0xe4/0x150
>      ...
> 
> This happens because vga->ddc either contains a valid DDC channel
> pointer, or -ENODEV, and drm_connector_init_with_ddc() expects a valid
> DDC channel pointer, or NULL.
> 
> Fix this by resetting vga->ddc to NULL in case of -ENODEV, and replacing
> the existing error checks by non-NULL checks.
> This is similar to what the HDMI connector driver does.
> 
> Fixes: a4f9087e85de141e ("drm/bridge: dumb-vga-dac: Provide ddc symlink in connector sysfs directory")
> Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
> ---
> An alternative would be to check if vga->ddc contains an error pointer,
> and calling drm_connector_init() instead of
> drm_connector_init_with_ddc(), like before.
> ---
>  drivers/gpu/drm/bridge/dumb-vga-dac.c | 7 ++++---
>  1 file changed, 4 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/gpu/drm/bridge/dumb-vga-dac.c b/drivers/gpu/drm/bridge/dumb-vga-dac.c
> index 8ef6539ae78a6eb3..7aa789c358829b05 100644
> --- a/drivers/gpu/drm/bridge/dumb-vga-dac.c
> +++ b/drivers/gpu/drm/bridge/dumb-vga-dac.c
> @@ -42,7 +42,7 @@ static int dumb_vga_get_modes(struct drm_connector *connector)
>  	struct edid *edid;
>  	int ret;
>  
> -	if (IS_ERR(vga->ddc))
> +	if (!vga->ddc)
>  		goto fallback;
>  
>  	edid = drm_get_edid(connector, vga->ddc);
> @@ -84,7 +84,7 @@ dumb_vga_connector_detect(struct drm_connector *connector, bool force)
>  	 * wire the DDC pins, or the I2C bus might not be working at
>  	 * all.
>  	 */
> -	if (!IS_ERR(vga->ddc) && drm_probe_ddc(vga->ddc))
> +	if (vga->ddc && drm_probe_ddc(vga->ddc))
>  		return connector_status_connected;
>  
>  	return connector_status_unknown;
> @@ -197,6 +197,7 @@ static int dumb_vga_probe(struct platform_device *pdev)
>  		if (PTR_ERR(vga->ddc) == -ENODEV) {
>  			dev_dbg(&pdev->dev,
>  				"No i2c bus specified. Disabling EDID readout\n");
> +			vga->ddc = NULL;
>  		} else {
>  			dev_err(&pdev->dev, "Couldn't retrieve i2c bus\n");
>  			return PTR_ERR(vga->ddc);
> @@ -218,7 +219,7 @@ static int dumb_vga_remove(struct platform_device *pdev)
>  
>  	drm_bridge_remove(&vga->bridge);
>  
> -	if (!IS_ERR(vga->ddc))
> +	if (vga->ddc)
>  		i2c_put_adapter(vga->ddc);
>  
>  	return 0;
> 

Looks sane,

Reviewed-by: Neil Armstrong <narmstrong@baylibre.com>

Guenter, can you confirm it also fixes qemu:versatilepb ?

Neil