[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20210419163336.64075-2-krzysztof.kozlowski@canonical.com>
Date: Mon, 19 Apr 2021 18:33:33 +0200
From: Krzysztof Kozlowski <krzysztof.kozlowski@...onical.com>
To: Sebastian Reichel <sre@...nel.org>,
Krzysztof Kozlowski <krzysztof.kozlowski@...onical.com>,
Bartlomiej Zolnierkiewicz <b.zolnierkie@...sung.com>,
linux-pm@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: [PATCH 2/5] power: supply: max17040: handle device_property_read_u8_array() failure
The device_property_read_u8_array() call should not fail because it is
preceded with device_property_count_u8() and check for number of
readable u8 values. However the code is more obvious and easier to read
if the return value of device_property_read_u8_array() is checked.
Otherwise reader needs to investigate whether really there is no risk of
using random stack values of 'rcomp' variable.
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@...onical.com>
---
drivers/power/supply/max17040_battery.c | 9 ++++-----
1 file changed, 4 insertions(+), 5 deletions(-)
diff --git a/drivers/power/supply/max17040_battery.c b/drivers/power/supply/max17040_battery.c
index d956c67d5155..d2053cd0bd13 100644
--- a/drivers/power/supply/max17040_battery.c
+++ b/drivers/power/supply/max17040_battery.c
@@ -268,11 +268,10 @@ static int max17040_get_of_data(struct max17040_chip *chip)
rcomp_len = device_property_count_u8(dev, "maxim,rcomp");
chip->rcomp = MAX17040_RCOMP_DEFAULT;
if (rcomp_len == data->rcomp_bytes) {
- device_property_read_u8_array(dev, "maxim,rcomp",
- rcomp, rcomp_len);
- chip->rcomp = rcomp_len == 2 ?
- rcomp[0] << 8 | rcomp[1] :
- rcomp[0] << 8;
+ if (!device_property_read_u8_array(dev, "maxim,rcomp",
+ rcomp, rcomp_len))
+ chip->rcomp = rcomp_len == 2 ? rcomp[0] << 8 | rcomp[1] :
+ rcomp[0] << 8;
} else if (rcomp_len > 0) {
dev_err(dev, "maxim,rcomp has incorrect length\n");
return -EINVAL;
--
2.25.1
Powered by blists - more mailing lists