On Mon, 10. Jun 10:47, Mathieu Poirier wrote:
> On Thu, Jun 06, 2024 at 10:52:04AM +0300, Aleksandr Mishin wrote:
> > In imx_rproc_addr_init() "nph = of_count_phandle_with_args()" just counts
> > number of phandles. But phandles may be empty. So of_parse_phandle() in
> > the parsing loop (0 < a < nph) may return NULL which is later dereferenced.
> > Adjust this issue by adding NULL-return check.
> >
> > Found by Linux Verification Center (linuxtesting.org) with SVACE.
> >
> > Fixes: a0ff4aa6f010 ("remoteproc: imx_rproc: add a NXP/Freescale imx_rproc
> > driver")
> > Signed-off-by: Aleksandr Mishin <amis...@t-argos.ru>
> > ---
> > drivers/remoteproc/imx_rproc.c | 2 ++
> > 1 file changed, 2 insertions(+)
> >
> > diff --git a/drivers/remoteproc/imx_rproc.c b/drivers/remoteproc/imx_rproc.c
> > index 5a3fb902acc9..39eacd90af14 100644
> > --- a/drivers/remoteproc/imx_rproc.c
> > +++ b/drivers/remoteproc/imx_rproc.c
> > @@ -726,6 +726,8 @@ static int imx_rproc_addr_init(struct imx_rproc *priv,
> > struct resource res;
> >
> > node = of_parse_phandle(np, "memory-region", a);
> > + if (!node)
>
> You're missing an "of_node_put()" before continuing.
>
The node is NULL in this case so of_node_put() is not needed..?
Btw, there is a "rsc-table" node->name check in the the end of the loop
body. It was added recently with commit 5e4c1243071d ("remoteproc:
imx_rproc: support remote cores booted before Linux Kernel"). Seems to me
it forgot that of_node_put() is called way before that.
Also commit 61afafe8b938 ("remoteproc: imx_rproc: Fix refcount leak in
imx_rproc_addr_init") was dealing with the last of_node_put() call here
but it's still not in the right place I'd say.
> > + continue;
> > /* Not map vdevbuffer, vdevring region */
> > if (!strncmp(node->name, "vdev", strlen("vdev"))) {
> > of_node_put(node);
> > --
> > 2.30.2
> >
> >
