Re: [PATCH] s390/raw3270: handle memory allocation failure in 'raw3270_setup_console()'

[Heiko Carstens]

On Sun, Jun 23, 2024 at 09:24:49PM +0900, yskelg@xxxxxxxxx wrote:
> From: Yunseong Kim <yskelg@xxxxxxxxx>
> 
> This patch handle potential null pointer dereference in
> 'raw3270_setup_device()', When 'raw3270_setup_console()' fails to
> allocate memory for 'rp' or 'ascebc'.
> 
> Signed-off-by: Yunseong Kim <yskelg@xxxxxxxxx>
> ---
>  drivers/s390/char/raw3270.c | 6 ++++++
>  1 file changed, 6 insertions(+)
> 
> diff --git a/drivers/s390/char/raw3270.c b/drivers/s390/char/raw3270.c
> index c57694be9bd3..4e81040eea81 100644
> --- a/drivers/s390/char/raw3270.c
> +++ b/drivers/s390/char/raw3270.c
> @@ -812,7 +812,13 @@ struct raw3270 __init *raw3270_setup_console(void)
>  		return ERR_CAST(cdev);
>  
>  	rp = kzalloc(sizeof(*rp), GFP_KERNEL | GFP_DMA);
> +	if (!rp)
> +		return ERR_PTR(-ENOMEM);
>  	ascebc = kzalloc(256, GFP_KERNEL);
> +	if (!ascebc) {
> +		kfree(rp);
> +		return ERR_PTR(-ENOMEM);
> +	}
>  	rc = raw3270_setup_device(cdev, rp, ascebc);
>  	if (rc)
>  		return ERR_PTR(rc);

This is kind of pointless since such allocations won't fail.. but
anyway: please make allocation and error handling like it is already
done in raw3270_create_device(); this will also prevent a memory leak
of rp and ascebc in case raw3270_setup_device() fails.