Re: [PATCH] davinci:Fix possible NULL pointer deference in the function davinci_gpio_probe

On Wed, Feb 3, 2016 at 9:10 AM, nick [off-list ref] wrote:

On 2016-02-02 06:30 PM, Alexandre Courbot wrote:

quoted

On Fri, Jan 29, 2016 at 6:04 AM, Nicholas Krause [off-list ref] wrote:

quoted

This fixes a possible NULL pointer deference in the function,
davinci_gpio_probe due to the function, gpio2regs being able
to return a NULL pointer if it rans to get the registers for
the gpio devices on a davinci board. Furthermore if this does
arise return -ENXIO to signal callers that this case has arisen
and avoiding setting the regs or other pointer values on the
chips to avoid rather deferences to a NULL pointer by other
functions in this gpio driver.

Signed-off-by: Nicholas Krause <redacted>
---
 drivers/gpio/gpio-davinci.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/gpio/gpio-davinci.c b/drivers/gpio/gpio-davinci.c
index ec58f42..800227e 100644
--- a/drivers/gpio/gpio-davinci.c
+++ b/drivers/gpio/gpio-davinci.c

@@ -257,6 +257,8 @@ static int davinci_gpio_probe(struct platform_device *pdev)
                spin_lock_init(&chips[i].lock);

                regs = gpio2regs(base);
+               if (!ret)
+                       return -ENXIO;

You haven't even tried to compile this one, have you?

Weird my compiler didn't fail when building with allmodconfig on x86_64 unless this
driver is not enabled by default under that build config it should have failed.
Sorry about that Alexandre I will resend and fix up that build issue.

Davinci is ARM, so it will not be included in a x86 build.