* [PATCH] pinctrl: sx150x: handle failure case of devm_kstrdup
@ 2018-12-02 10:04 Nicholas Mc Guire
2018-12-07 12:23 ` Linus Walleij
0 siblings, 1 reply; 2+ messages in thread
From: Nicholas Mc Guire @ 2018-12-02 10:04 UTC (permalink / raw)
To: Linus Walleij; +Cc: linux-gpio, linux-kernel, Nicholas Mc Guire
devm_kstrdup() may return NULL if internal allocation failed.
Thus using label, name is unsafe without checking. Therefor
in the unlikely case of allocation failure, sx150x_probe() simply
returns -ENOMEM.
Signed-off-by: Nicholas Mc Guire <hofrat@osadl.org>
Fixes: 9e80f9064e73 ("pinctrl: Add SX150X GPIO Extender Pinctrl Driver")
---
Problem located with an experimental coccinelle script
Patch was compile tested with: lpc32xx_defconfig (implies PINCTRL_SX150X=y)
Note that while lpc32xx_defconfig contains PINCTRL_SX150X=y after make
lpc32xx_defconfig it is *not* set - seems to be a missing entry in
drivers/pincrtl/Kconfig - so PINCRTL and PINCTRL_SX150X had to be
manually enabled.
Patch is against 4.20-rc4 (localversion-next is next-20181130)
drivers/pinctrl/pinctrl-sx150x.c | 11 ++++++++---
1 file changed, 8 insertions(+), 3 deletions(-)
diff --git a/drivers/pinctrl/pinctrl-sx150x.c b/drivers/pinctrl/pinctrl-sx150x.c
index cbf58a1..4d87d75 100644
--- a/drivers/pinctrl/pinctrl-sx150x.c
+++ b/drivers/pinctrl/pinctrl-sx150x.c
@@ -1166,7 +1166,6 @@ static int sx150x_probe(struct i2c_client *client,
}
/* Register GPIO controller */
- pctl->gpio.label = devm_kstrdup(dev, client->name, GFP_KERNEL);
pctl->gpio.base = -1;
pctl->gpio.ngpio = pctl->data->npins;
pctl->gpio.get_direction = sx150x_gpio_get_direction;
@@ -1180,6 +1179,10 @@ static int sx150x_probe(struct i2c_client *client,
pctl->gpio.of_node = dev->of_node;
#endif
pctl->gpio.can_sleep = true;
+ pctl->gpio.label = devm_kstrdup(dev, client->name, GFP_KERNEL);
+ if (!pctl->gpio.label)
+ return -ENOMEM;
+
/*
* Setting multiple pins is not safe when all pins are not
* handled by the same regmap register. The oscio pin (present
@@ -1200,13 +1203,15 @@ static int sx150x_probe(struct i2c_client *client,
/* Add Interrupt support if an irq is specified */
if (client->irq > 0) {
- pctl->irq_chip.name = devm_kstrdup(dev, client->name,
- GFP_KERNEL);
pctl->irq_chip.irq_mask = sx150x_irq_mask;
pctl->irq_chip.irq_unmask = sx150x_irq_unmask;
pctl->irq_chip.irq_set_type = sx150x_irq_set_type;
pctl->irq_chip.irq_bus_lock = sx150x_irq_bus_lock;
pctl->irq_chip.irq_bus_sync_unlock = sx150x_irq_bus_sync_unlock;
+ pctl->irq_chip.name = devm_kstrdup(dev, client->name,
+ GFP_KERNEL);
+ if (!pctl->irq_chip.name)
+ return -ENOMEM;
pctl->irq.masked = ~0;
pctl->irq.sense = 0;
--
2.1.4
^ permalink raw reply related [flat|nested] 2+ messages in thread
* Re: [PATCH] pinctrl: sx150x: handle failure case of devm_kstrdup
2018-12-02 10:04 [PATCH] pinctrl: sx150x: handle failure case of devm_kstrdup Nicholas Mc Guire
@ 2018-12-07 12:23 ` Linus Walleij
0 siblings, 0 replies; 2+ messages in thread
From: Linus Walleij @ 2018-12-07 12:23 UTC (permalink / raw)
To: Nicholas Mc Guire; +Cc: open list:GPIO SUBSYSTEM, linux-kernel
On Sun, Dec 2, 2018 at 11:08 AM Nicholas Mc Guire <hofrat@osadl.org> wrote:
> devm_kstrdup() may return NULL if internal allocation failed.
> Thus using label, name is unsafe without checking. Therefor
> in the unlikely case of allocation failure, sx150x_probe() simply
> returns -ENOMEM.
>
> Signed-off-by: Nicholas Mc Guire <hofrat@osadl.org>
> Fixes: 9e80f9064e73 ("pinctrl: Add SX150X GPIO Extender Pinctrl Driver")
Patch applied.
> Problem located with an experimental coccinelle script
I like this script.
Yours,
Linus Walleij
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2018-12-07 12:23 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2018-12-02 10:04 [PATCH] pinctrl: sx150x: handle failure case of devm_kstrdup Nicholas Mc Guire
2018-12-07 12:23 ` Linus Walleij
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).