[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20140310.155015.1679111216670234389.davem@davemloft.net>
Date: Mon, 10 Mar 2014 15:50:15 -0400 (EDT)
From: David Miller <davem@...emloft.net>
To: ogerlitz@...lanox.com
Cc: netdev@...r.kernel.org, roland@...nel.org,
jackm@....mellanox.co.il, amirv@...lanox.com
Subject: Re: [PATCH net-next 1/7] mlx4: Adjust QP1 multiplexing for
RoCE/SRIOV
From: Or Gerlitz <ogerlitz@...lanox.com>
Date: Sun, 9 Mar 2014 16:53:58 +0200
> @@ -509,6 +510,10 @@ int mlx4_ib_send_to_slave(struct mlx4_ib_dev *dev, int slave, u8 port,
> * The driver will set the force loopback bit in post_send */
> memset(&attr, 0, sizeof attr);
> attr.port_num = port;
> + if (is_eth) {
> + mlx4_get_roce_gid_from_slave(dev->dev, port, slave, attr.grh.dgid.raw);
> + attr.ah_flags = IB_AH_GRH;
> + }
> ah = ib_create_ah(tun_ctx->pd, &attr);
> if (IS_ERR(ah))
> return -ENOMEM;
Either the error returned from mlx4_get_roce_gid_from_slave() doesn't matter...
> + err = mlx4_get_roce_gid_from_slave(to_mdev(ib_dev)->dev,
> + be32_to_cpu(ah->av.ib.port_pd) >> 24,
> + ah->av.ib.gid_index, &sgid.raw[0]);
> + if (err)
> + return err;
or it does.
Please decide whether it does or not. If it does, make all call sites check the return
value. If it does not, remove the return value and make the function return void.
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Powered by blists - more mailing lists