[ofa-general] Re: [PATCHv4 2/2] ipoib: handle pkey change events

Michael S. Tsirkin mst at dev.mellanox.co.il
Thu May 10 06:01:23 PDT 2007


> Quoting Yosef Etigin <yosefe at voltaire.com>:
> Subject: Re: [PATCHv4 2/2] ipoib: handle pkey change events
> 
> > 
> > Return some error code -ENXIO.
> > 
> All other branches in this function return -1 (see next hunk)

Oh. Right. I haven't looked.

> Anyway, let it be -ENXIO.

Up to you really, I take it back.

> @@ -260,7 +249,6 @@ void ipoib_event(struct ib_event_handler
>  		container_of(handler, struct ipoib_dev_priv, event_handler);
>  
>  	if ((record->event == IB_EVENT_PORT_ERR    ||
> -	     record->event == IB_EVENT_PKEY_CHANGE ||
>  	     record->event == IB_EVENT_PORT_ACTIVE ||
>  	     record->event == IB_EVENT_LID_CHANGE  ||
>  	     record->event == IB_EVENT_SM_CHANGE   ||
> @@ -268,5 +256,9 @@ void ipoib_event(struct ib_event_handler
>  	    record->element.port_num == priv->port) {
>  		ipoib_dbg(priv, "Port state change event\n");
>  		queue_work(ipoib_workqueue, &priv->flush_task);
> +	} else if (record->event == IB_EVENT_PKEY_CHANGE &&
> +		   record->element.port_num == priv->port) {
> +		ipoib_dbg(priv, "pkey change event on port:%d\n", priv->port);
> +		queue_work(ipoib_workqueue, &priv->pkey_event_task);
>  	}
>  }

What do you think about my idea to do
if (record->element.port_num != priv->port)
	return
at the top?

Anyway, I think you've addressed all real issues - could you pls
post final version of both patches for OFED and 2.6.22?

-- 
MST



More information about the general mailing list