Re: [PATCH] scsi: zfcp: Fix a potential double free in zfcp_port_enqueue

[Benjamin Block]

Hello Liu Dinghao,

good find.

On Thu, Sep 21, 2023 at 02:39:15PM +0800, Dinghao Liu wrote:
> When device_register() fails, zfcp_port_release() will be called
> after put_device(). As a result, the zfcp_ccw_adapter_put() after
> err_out is redundant because it will be called in the call-back
> function zfcp_port_release(). Remove it from this error path.

So the reference on the adapter object is doubly put, which may
lead to a premature free of the adapter object itself. Please mention that
either in the subject, or description; it makes it easier to see what exactly
breaks at a glance.

> 
> Fixes: f3450c7b9172 ("[SCSI] zfcp: Replace local reference counting with common kref")
> Signed-off-by: Dinghao Liu <dinghao.liu@xxxxxxxxxx>
> ---
>  drivers/s390/scsi/zfcp_aux.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/s390/scsi/zfcp_aux.c b/drivers/s390/scsi/zfcp_aux.c
> index df782646e856..489e6239dedf 100644
> --- a/drivers/s390/scsi/zfcp_aux.c
> +++ b/drivers/s390/scsi/zfcp_aux.c
> @@ -552,7 +552,7 @@ struct zfcp_port *zfcp_port_enqueue(struct zfcp_adapter *adapter, u64 wwpn,
>  
>  	if (device_register(&port->dev)) {
>  		put_device(&port->dev);
> -		goto err_out;
> +		return ERR_PTR(retval);

I'd rather have a new label at the bottom, in front of the return that is
already there, and jump to that, instead of a different function exit point.

>  	}
>  
>  	write_lock_irq(&adapter->port_list_lock);
> -- 
> 2.17.1
> 

-- 
Best Regards, Benjamin Block        /        Linux on IBM Z Kernel Development
IBM Deutschland Research & Development GmbH    /   https://www.ibm.com/privacy
Vors. Aufs.-R.: Gregor Pillen         /         Geschäftsführung: David Faller
Sitz der Ges.: Böblingen     /    Registergericht: AmtsG Stuttgart, HRB 243294