[ofa-general] [PATCH 2.6.25] RDMA/cxgb3: Fail loopbackconnections.

Steve Wise swise at opengridcomputing.com
Wed Feb 13 07:51:39 PST 2008


Roland Dreier wrote:
> Steve, I had to update the patch adding an include and fixing the
> function declaration (as below)... but how much testing have you done
> with this??
>

Now I understand.  I thought I'd fixed these!  I fixed them locally in 
the test tree on my victim and then tested, but forgot to update my git 
tree.

Sorry for this.

You want me to resubmit?


> commit 8704e9a8790cc9e394198663c1c9150c899fb9a2
> Author: Steve Wise <swise at opengridcomputing.com>
> Date:   Tue Feb 12 16:09:29 2008 -0600
> 
>     RDMA/cxgb3: Fail loopback connections
>     
>     The cxgb3 HW and driver don't support loopback RDMA connections.  So
>     fail any connection attempt where the destination address is local.
>     
>     Signed-off-by: Steve Wise <swise at opengridcomputing.com>
>     Signed-off-by: Roland Dreier <rolandd at cisco.com>
> 
> diff --git a/drivers/infiniband/hw/cxgb3/iwch_cm.c b/drivers/infiniband/hw/cxgb3/iwch_cm.c
> index e9a08fa..320f2b6 100644
> --- a/drivers/infiniband/hw/cxgb3/iwch_cm.c
> +++ b/drivers/infiniband/hw/cxgb3/iwch_cm.c
> @@ -35,6 +35,7 @@
>  #include <linux/skbuff.h>
>  #include <linux/timer.h>
>  #include <linux/notifier.h>
> +#include <linux/inetdevice.h>
>  
>  #include <net/neighbour.h>
>  #include <net/netevent.h>
> @@ -1784,6 +1785,17 @@ err:
>  	return err;
>  }
>  
> +static int is_loopback_dst(struct iw_cm_id *cm_id)
> +{
> +	struct net_device *dev;
> +
> +	dev = ip_dev_find(&init_net, cm_id->remote_addr.sin_addr.s_addr);
> +	if (!dev)
> +		return 0;
> +	dev_put(dev);
> +	return 1;
> +}
> +
>  int iwch_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
>  {
>  	int err = 0;
> @@ -1791,6 +1803,11 @@ int iwch_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
>  	struct iwch_ep *ep;
>  	struct rtable *rt;
>  
> +	if (is_loopback_dst(cm_id)) {
> +		err = -ENOSYS;
> +		goto out;
> +	}
> +
>  	ep = alloc_ep(sizeof(*ep), GFP_KERNEL);
>  	if (!ep) {
>  		printk(KERN_ERR MOD "%s - cannot alloc ep.\n", __FUNCTION__);




More information about the general mailing list