On 08/05/2018 12:34 PM, Christophe JAILLET wrote:
If the 2nd memory allocation of the loop fails, we must undo the memory allocation done so far. Fixes: 67f8b1dcb9ee ("net/mlx4_en: Refactor the XDP forwarding rings scheme") Signed-off-by: Christophe JAILLET <christophe.jail...@wanadoo.fr> --- drivers/net/ethernet/mellanox/mlx4/en_netdev.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c index e0adac4a9a19..bf078244e467 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c @@ -3331,7 +3331,7 @@ int mlx4_en_init_netdev(struct mlx4_en_dev *mdev, int port, if (!priv->tx_cq[t]) { kfree(priv->tx_ring[t]); err = -ENOMEM; - goto out; + goto err_free_tx; } } priv->rx_ring_num = prof->rx_ring_num;
Hi Christophe, Thanks for re-sending this. In your previous mail you referred to the call mlx4_en_destroy_netdev here: https://elixir.bootlin.com/linux/v4.16-rc5/source/drivers/net/ethernet/mellanox/mlx4/en_main.c#L232 While I was referring to the one below, which is always called on failures: https://elixir.bootlin.com/linux/v4.16-rc5/source/drivers/net/ethernet/mellanox/mlx4/en_netdev.c#L3587 I still believe that the err_free_tx label and its while loop is redundant. Regards, Tariq