The for_each_child_of_node loop in drm_of_lvds_get_remote_pixels_type bails out in two occasions. Originally, both were calling of_node_put to clean up, but (probably a typo) for the wrong variable.
One of these typos was fixed in commit 4ee48cc5586b ("drm: of: Fix double-free bug"), but that missed the leak of the original endpoint and also the first error path which was obviously wrong as well. Fix the leak of endpoint in the error path by calling of_node_put on it. Fixes: 6529007522de ("drm: of: Add drm_of_lvds_get_dual_link_pixel_order") Signed-off-by: Tobias Jordan <ker...@cdqe.de> --- Sorry for the confusion, it seems it's not that easy to get the variable name right. Please disregard the patch sent a few minutes ago, this is the right one. drivers/gpu/drm/drm_of.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/drm_of.c b/drivers/gpu/drm/drm_of.c index ca04c34e8251..997b8827fed2 100644 --- a/drivers/gpu/drm/drm_of.c +++ b/drivers/gpu/drm/drm_of.c @@ -315,7 +315,7 @@ static int drm_of_lvds_get_remote_pixels_type( remote_port = of_graph_get_remote_port(endpoint); if (!remote_port) { - of_node_put(remote_port); + of_node_put(endpoint); return -EPIPE; } @@ -331,8 +331,10 @@ static int drm_of_lvds_get_remote_pixels_type( * configurations by passing the endpoints explicitly to * drm_of_lvds_get_dual_link_pixel_order(). */ - if (!current_pt || pixels_type != current_pt) + if (!current_pt || pixels_type != current_pt) { + of_node_put(endpoint); return -EINVAL; + } } return pixels_type; -- 2.20.1