Re: [PATCH] gspca: correctly checked failed allocation

[Hans de Goede <hdegoede@redhat.com>]

Hi,

On 19-10-15 18:43, Insu Yun wrote:
> create_singlethread_workqueue can be failed in memory pressue.
> So, check return value and return -ENOMEM
>
> Signed-off-by: Insu Yun <wuninsu@gmail.com>
> ---
>   drivers/media/usb/gspca/sq905.c | 2 ++
>   1 file changed, 2 insertions(+)
>
> diff --git a/drivers/media/usb/gspca/sq905.c b/drivers/media/usb/gspca/sq905.c
> index a7ae0ec..b1c25d9a 100644
> --- a/drivers/media/usb/gspca/sq905.c
> +++ b/drivers/media/usb/gspca/sq905.c
> @@ -392,6 +392,8 @@ static int sd_start(struct gspca_dev *gspca_dev)
>   	}
>   	/* Start the workqueue function to do the streaming */
>   	dev->work_thread = create_singlethread_workqueue(MODULE_NAME);
> +	if (!dev->work_thread)
> +		return -ENOMEM;
>   	queue_work(dev->work_thread, &dev->work_struct);
>
>   	return 0;

If the thread creation fails we should not send the start command,
so the create_singlethread_workqueue call should be moved
up in the function, while keeping the queue_work at the end.

And if the sq905_command fails then the workqueue should
be destroyed and dev->work_thread should be set to NULL
before returning the sq905_command failure to the caller.

Regards,

Hans