[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <1476873130-24926-7-git-send-email-masneyb@onstation.org>
Date: Wed, 19 Oct 2016 06:32:10 -0400
From: Brian Masney <masneyb@...tation.org>
To: jic23@...nel.org
Cc: knaack.h@....de, lars@...afoo.de, pmeerw@...erw.net,
gregkh@...uxfoundation.org, linux-iio@...r.kernel.org,
devel@...verdev.osuosl.org, linux-kernel@...r.kernel.org
Subject: [PATCH 7/7] iio: light: tsl2583: fix concurrency issue in taos_get_lux()
taos_get_lux() calls mutex_trylock(). If the lock could not be acquired,
then chip->als_cur_info.lux is returned. The issue is that this value
is updated while the mutex is held and could cause a half written value
to be returned to the caller. This patch changes the call to
mutex_trylock() with mutex_lock().
Signed-off-by: Brian Masney <masneyb@...tation.org>
---
This is the most controversial change in my patch set. There are two
other possible solutions that I could envision to work around this
issue:
1) Return -EBUSY and make the caller responsible for backing off
2) Change this driver to use RCU instead of a mutex
drivers/staging/iio/light/tsl2583.c | 5 +----
1 file changed, 1 insertion(+), 4 deletions(-)
diff --git a/drivers/staging/iio/light/tsl2583.c b/drivers/staging/iio/light/tsl2583.c
index 47656ae..c4d2e3a 100644
--- a/drivers/staging/iio/light/tsl2583.c
+++ b/drivers/staging/iio/light/tsl2583.c
@@ -206,10 +206,7 @@ static int taos_get_lux(struct iio_dev *indio_dev)
u32 ch0lux = 0;
u32 ch1lux = 0;
- if (mutex_trylock(&chip->als_mutex) == 0) {
- dev_info(&chip->client->dev, "taos_get_lux device is busy\n");
- return chip->als_cur_info.lux; /* busy, so return LAST VALUE */
- }
+ mutex_lock(&chip->als_mutex);
if (chip->taos_chip_status != TSL258X_CHIP_WORKING) {
/* device is not enabled */
--
2.7.4
Powered by blists - more mailing lists