Re: [PATCH v8 1/3] of: Decrement refcount of previous endpoint in of_graph_get_next_endpoint

[Laurent Pinchart]

Hi Philipp,

Thank you for the patch.

Benoit, please see below for a possible issue in the am437x-vpfe driver.

On Monday 23 February 2015 11:54:04 Philipp Zabel wrote:
> Decrementing the reference count of the previous endpoint node allows to
> use the of_graph_get_next_endpoint function in a for_each_... style macro.
> All current users of this function that pass a non-NULL prev parameter
> (that is, soc_camera and imx-drm) are changed to not decrement the passed
> prev argument's refcount themselves.
> 
> Signed-off-by: Philipp Zabel <p.zabel@xxxxxxxxxxxxxx>
> Acked-by: Mauro Carvalho Chehab <mchehab@xxxxxxxxxxxxxxx>
> Acked-by: Mathieu Poirier <mathieu.poirier@xxxxxxxxxx>
> Acked-by: Laurent Pinchart <laurent.pinchart@xxxxxxxxxxxxxxxx>
> Acked-by: Tomi Valkeinen <tomi.valkeinen@xxxxxx>
> ---
> Changes since v7:
>  - Rebased onto v4.0-rc1
>  - Added fix for am437x-vpfe
> ---
>  drivers/coresight/of_coresight.c                  | 13 ++-----------
>  drivers/gpu/drm/imx/imx-drm-core.c                | 11 +----------
>  drivers/gpu/drm/rcar-du/rcar_du_kms.c             | 15 ++++-----------
>  drivers/media/platform/am437x/am437x-vpfe.c       |  1 -
>  drivers/media/platform/soc_camera/soc_camera.c    |  3 ++-
>  drivers/of/base.c                                 |  9 +--------
>  drivers/video/fbdev/omap2/dss/omapdss-boot-init.c |  7 +------
>  7 files changed, 11 insertions(+), 48 deletions(-)

[snip]

> diff --git a/drivers/media/platform/am437x/am437x-vpfe.c
> b/drivers/media/platform/am437x/am437x-vpfe.c index 56a5cb0..0d07fca 100644
> --- a/drivers/media/platform/am437x/am437x-vpfe.c
> +++ b/drivers/media/platform/am437x/am437x-vpfe.c
> @@ -2504,7 +2504,6 @@ vpfe_get_pdata(struct platform_device *pdev)
>  					     GFP_KERNEL);
>  		pdata->asd[i]->match_type = V4L2_ASYNC_MATCH_OF;
>  		pdata->asd[i]->match.of.node = rem;
> -		of_node_put(endpoint);
>  		of_node_put(rem);
>  	}

For the am47x-vpfe driver,

Acked-by: Laurent Pinchart <laurent.pinchart@xxxxxxxxxxxxxxxx>

Benoit, there seems to be a refcount issue with rem. The node pointer is 
assigned to pdata->asd[i]->match.of.node, which should require a reference, 
but you then call of_node_put(rem), releasing the only reference held. Isn't 
that a problem ?

Furthermore, on the next iteration, if an error occurs the goto done will 
result in of_node_put(rem) being called again, releasing a reference that you 
don't hold. I've sent a patch to fix that.

-- 
Regards,

Laurent Pinchart

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/