[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-Id: <20190620135052.27367-1-colin.king@canonical.com>
Date: Thu, 20 Jun 2019 14:50:52 +0100
From: Colin King <colin.king@...onical.com>
To: Doug Ledford <dledford@...hat.com>, Jason Gunthorpe <jgg@...pe.ca>,
Leon Romanovsky <leon@...nel.org>,
Parav Pandit <parav@...lanox.com>, linux-rdma@...r.kernel.org
Cc: kernel-janitors@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: [PATCH][next] RDMA: check for null return from call to ib_get_client_data
From: Colin Ian King <colin.king@...onical.com>
The return from ib_get_client_data can potentially be null, so add a null
check on umad_dev and return -ENODEV in this unlikely case to avoid any
null pointer deferences.
Addresses-Coverity: ("Dereference null return")
Fixes: 8f71bb0030b8 ("RDMA: Report available cdevs through RDMA_NLDEV_CMD_GET_CHARDEV")
Signed-off-by: Colin Ian King <colin.king@...onical.com>
---
drivers/infiniband/core/user_mad.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/drivers/infiniband/core/user_mad.c b/drivers/infiniband/core/user_mad.c
index 547090b41cfb..d78a35913824 100644
--- a/drivers/infiniband/core/user_mad.c
+++ b/drivers/infiniband/core/user_mad.c
@@ -1153,6 +1153,9 @@ static int ib_issm_get_nl_info(struct ib_device *ibdev, void *client_data,
struct ib_umad_device *umad_dev =
ib_get_client_data(ibdev, &umad_client);
+ if (!umad_dev)
+ return -ENODEV;
+
if (!rdma_is_port_valid(ibdev, res->port))
return -EINVAL;
--
2.20.1
Powered by blists - more mailing lists