[PATCH] iio: core: check return value when calling dev_set_name()

[PATCH] iio: core: check return value when calling dev_set_name()

[Yang Yingliang]

I got a null-ptr-deref report when doing fault injection test:

BUG: kernel NULL pointer dereference, address: 0000000000000000
RIP: 0010:strlen+0x0/0x20
Call Trace:
 start_creating+0x199/0x2f0
 debugfs_create_dir+0x25/0x430
 __iio_device_register+0x4da/0x1b40 [industrialio]
 __devm_iio_device_register+0x22/0x80 [industrialio]
 max1027_probe+0x639/0x860 [max1027]
 spi_probe+0x183/0x210
 really_probe+0x285/0xc30

If dev_set_name() fails, the dev_name() is null, check the return
value of dev_set_name() to avoid the null-ptr-deref.

Reported-by: Hulk Robot <hulkci@huawei.com>
Fixes: e553f182d55b ("staging: iio: core: Introduce debugfs support...")
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
---
 drivers/iio/industrialio-core.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/drivers/iio/industrialio-core.c b/drivers/iio/industrialio-core.c
index 2dc837db50f7..3e1e86d987cc 100644
--- a/drivers/iio/industrialio-core.c
+++ b/drivers/iio/industrialio-core.c
@@ -1665,7 +1665,13 @@ struct iio_dev *iio_device_alloc(struct device *parent, int sizeof_priv)
 		kfree(iio_dev_opaque);
 		return NULL;
 	}
-	dev_set_name(&indio_dev->dev, "iio:device%d", iio_dev_opaque->id);
+
+	if (dev_set_name(&indio_dev->dev, "iio:device%d", iio_dev_opaque->id)) {
+		ida_simple_remove(&iio_ida, iio_dev_opaque->id);
+		kfree(iio_dev_opaque);
+		return NULL;
+	}
+
 	INIT_LIST_HEAD(&iio_dev_opaque->buffer_list);
 	INIT_LIST_HEAD(&iio_dev_opaque->ioctl_handlers);
 
-- 
2.25.1

Re: [PATCH] iio: core: check return value when calling dev_set_name()

[Jonathan Cameron]

On Tue, 12 Oct 2021 14:36:24 +0800
Yang Yingliang <yangyingliang@huawei.com> wrote:

> I got a null-ptr-deref report when doing fault injection test:
> 
> BUG: kernel NULL pointer dereference, address: 0000000000000000
> RIP: 0010:strlen+0x0/0x20
> Call Trace:
>  start_creating+0x199/0x2f0
>  debugfs_create_dir+0x25/0x430
>  __iio_device_register+0x4da/0x1b40 [industrialio]
>  __devm_iio_device_register+0x22/0x80 [industrialio]
>  max1027_probe+0x639/0x860 [max1027]
>  spi_probe+0x183/0x210
>  really_probe+0x285/0xc30
> 
> If dev_set_name() fails, the dev_name() is null, check the return
> value of dev_set_name() to avoid the null-ptr-deref.
> 
> Reported-by: Hulk Robot <hulkci@huawei.com>
> Fixes: e553f182d55b ("staging: iio: core: Introduce debugfs support...")
> Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

Hi Yang Yingliang,

I've been delaying replying to these for a few days because you have
identified a more significant issue whilst working in this area and I wanted
to take a deeper look at it.

After we call device_initialize() a few lines above this all the cleanup
on error should be done via a put_device() call, not by cleaning it up manually.

However, that's clearly a much more substantial change so I'm going to apply this
for now and deal with that cleanup at a later date.

Thanks,

Jonathan


> ---
>  drivers/iio/industrialio-core.c | 8 +++++++-
>  1 file changed, 7 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/iio/industrialio-core.c b/drivers/iio/industrialio-core.c
> index 2dc837db50f7..3e1e86d987cc 100644
> --- a/drivers/iio/industrialio-core.c
> +++ b/drivers/iio/industrialio-core.c
> @@ -1665,7 +1665,13 @@ struct iio_dev *iio_device_alloc(struct device *parent, int sizeof_priv)
>  		kfree(iio_dev_opaque);
>  		return NULL;
>  	}
> -	dev_set_name(&indio_dev->dev, "iio:device%d", iio_dev_opaque->id);
> +
> +	if (dev_set_name(&indio_dev->dev, "iio:device%d", iio_dev_opaque->id)) {
> +		ida_simple_remove(&iio_ida, iio_dev_opaque->id);
> +		kfree(iio_dev_opaque);
> +		return NULL;
> +	}
> +
>  	INIT_LIST_HEAD(&iio_dev_opaque->buffer_list);
>  	INIT_LIST_HEAD(&iio_dev_opaque->ioctl_handlers);
>