[ 100/102] net/mlx4_core: VFs must ignore the enable_64b_cqe_eqe module param
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Date: 2013-08-09 02:00:47
Also in:
lkml
3.10-stable review patch. If anyone has any objections, please let me know. ------------------ From: Jack Morgenstein <redacted> [ Upstream commit b30513202c6c14120f70b2e9aa1e97d47bbc2313 ] Slaves get the 64B CQE/EQE state from QUERY_HCA, not from the module parameter. If the parameter is set to zero, the slave outputs an incorrect/irrelevant warning message that 64B CQEs/EQEs are supported but not enabled (even if the hypervisor has enabled 64B CQEs/EQEs). Signed-off-by: Jack Morgenstein <redacted> Signed-off-by: Or Gerlitz <redacted> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> --- drivers/net/ethernet/mellanox/mlx4/main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
--- a/drivers/net/ethernet/mellanox/mlx4/main.c
+++ b/drivers/net/ethernet/mellanox/mlx4/main.c@@ -371,7 +371,7 @@ static int mlx4_dev_cap(struct mlx4_dev dev->caps.sqp_demux = (mlx4_is_master(dev)) ? MLX4_MAX_NUM_SLAVES : 0; - if (!enable_64b_cqe_eqe) { + if (!enable_64b_cqe_eqe && !mlx4_is_slave(dev)) { if (dev_cap->flags & (MLX4_DEV_CAP_FLAG_64B_CQE | MLX4_DEV_CAP_FLAG_64B_EQE)) { mlx4_warn(dev, "64B EQEs/CQEs supported by the device but not enabled\n");