Re: [PATCH] of: dynamic: Fix potential memory leak in of_changeset_action()

From: Rob Herring
Date: Fri Sep 08 2023 - 11:19:00 EST


On Fri, Sep 8, 2023 at 2:03 AM Dan Carpenter <dan.carpenter@xxxxxxxxxx> wrote:
>
> Smatch complains that the error path where "action" is invalid leaks
> the "ce" allocation:
> drivers/of/dynamic.c:935 of_changeset_action()
> warn: possible memory leak of 'ce'
>
> Fix this by doing the validation before the allocation.

I'm going to add a note when applying that "action" can't ever
actually be invalid because all the callers are static inlines with
hardcoded action values. I suppose there could be an out of tree
module calling of_changeset_action() directly, but that's wrong given
the wrappers.

> Fixes: 914d9d831e61 ("of: dynamic: Refactor action prints to not use "%pOF" inside devtree_lock")
> Reported-by: kernel test robot <lkp@xxxxxxxxx>
> Closes: https://lore.kernel.org/r/202309011059.EOdr4im9-lkp@xxxxxxxxx/

Despite what that says, it was never reported to me. IOW, the added TO
and CC lines don't seem to have any effect.

> Signed-off-by: Dan Carpenter <dan.carpenter@xxxxxxxxxx>
> ---
> drivers/of/dynamic.c | 6 +++---
> 1 file changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/of/dynamic.c b/drivers/of/dynamic.c
> index 0a3483e247a8..f63250c650ca 100644
> --- a/drivers/of/dynamic.c
> +++ b/drivers/of/dynamic.c
> @@ -890,13 +890,13 @@ int of_changeset_action(struct of_changeset *ocs, unsigned long action,
> {
> struct of_changeset_entry *ce;
>
> + if (WARN_ON(action >= ARRAY_SIZE(action_names)))
> + return -EINVAL;
> +
> ce = kzalloc(sizeof(*ce), GFP_KERNEL);
> if (!ce)
> return -ENOMEM;
>
> - if (WARN_ON(action >= ARRAY_SIZE(action_names)))
> - return -EINVAL;
> -
> /* get a reference to the node */
> ce->action = action;
> ce->np = of_node_get(np);
> --
> 2.39.2
>