[PATCH 5.10 213/243] reset: bail if try_module_get() fails

From: Greg Kroah-Hartman
Date: Mon Jul 19 2021 - 13:35:26 EST


From: Philipp Zabel <p.zabel@xxxxxxxxxxxxxx>

[ Upstream commit 4fb26fb83f0def3d39c14e268bcd4003aae8fade ]

Abort instead of returning a new reset control for a reset controller
device that is going to have its module unloaded.

Reported-by: Uwe Kleine-König <u.kleine-koenig@xxxxxxxxxxxxxx>
Fixes: 61fc41317666 ("reset: Add reset controller API")
Acked-by: Uwe Kleine-König <u.kleine-koenig@xxxxxxxxxxxxxx>
Link: https://lore.kernel.org/r/20210607082615.15160-1-p.zabel@xxxxxxxxxxxxxx
Signed-off-by: Philipp Zabel <p.zabel@xxxxxxxxxxxxxx>
Signed-off-by: Sasha Levin <sashal@xxxxxxxxxx>
---
drivers/reset/core.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/reset/core.c b/drivers/reset/core.c
index a2df88e90011..f93388b9a4a1 100644
--- a/drivers/reset/core.c
+++ b/drivers/reset/core.c
@@ -567,7 +567,10 @@ static struct reset_control *__reset_control_get_internal(
if (!rstc)
return ERR_PTR(-ENOMEM);

- try_module_get(rcdev->owner);
+ if (!try_module_get(rcdev->owner)) {
+ kfree(rstc);
+ return ERR_PTR(-ENODEV);
+ }

rstc->rcdev = rcdev;
list_add(&rstc->list, &rcdev->reset_control_head);
--
2.30.2