From: Tonghao Zhang <xiangxia.m....@gmail.com> If user don't set the dv_xmeta_en to 1 or 2, in the flow_dv_convert_action_set_meta function: * flow_dv_get_metadata_reg may return the REG_NONE, when MLX5_METADATA_FDB enabled for metadata set action. * reg_to_field(REG_NONE) return MLX5_MODI_OUT_NONE which is invalid.
The rdma-core calltrace: dr_action_create_modify_action dr_actions_convert_modify_header dr_action_modify_sw_to_hw dr_action_modify_sw_to_hw_set dr_ste_get_modify_hdr_hw_field sw_field [MLX5_MODI_OUT_NONE 4095] should not > ste_ctx->modify_hdr_field_arr_sz [92] As doc[1] says: | dv_xmeta_en 0, this is default value, defines the legacy mode, | the MARK and META related actions and items operate only within | NIC Tx and NIC Rx steering domains, no MARK and META information | crosses the domain boundaries. This patch add check for that case to warn that not supported. [1] - http://doc.dpdk.org/guides-20.02/nics/mlx5.html Fixes: fcc8d2f716fd ("net/mlx5: extend flow metadata support") Cc: sta...@dpdk.org Signed-off-by: Tonghao Zhang <xiangxia.m....@gmail.com> --- drivers/net/mlx5/mlx5_flow_dv.c | 31 +++++++++++++++++++------------ 1 file changed, 19 insertions(+), 12 deletions(-) diff --git a/drivers/net/mlx5/mlx5_flow_dv.c b/drivers/net/mlx5/mlx5_flow_dv.c index e4818319507c..c77835270d60 100644 --- a/drivers/net/mlx5/mlx5_flow_dv.c +++ b/drivers/net/mlx5/mlx5_flow_dv.c @@ -1251,6 +1251,7 @@ flow_dv_convert_action_set_meta if (reg < 0) return reg; + MLX5_ASSERT(reg != REG_NONE); /* * In datapath code there is no endianness * coversions for perfromance reasons, all @@ -1449,7 +1450,6 @@ flow_dv_validate_item_meta(struct rte_eth_dev *dev __rte_unused, struct rte_flow_error *error) { struct mlx5_priv *priv = dev->data->dev_private; - struct mlx5_dev_config *config = &priv->config; const struct rte_flow_item_meta *spec = item->spec; const struct rte_flow_item_meta *mask = item->mask; struct rte_flow_item_meta nic_mask = { @@ -1463,23 +1463,25 @@ flow_dv_validate_item_meta(struct rte_eth_dev *dev __rte_unused, RTE_FLOW_ERROR_TYPE_ITEM_SPEC, item->spec, "data cannot be empty"); - if (config->dv_xmeta_en != MLX5_XMETA_MODE_LEGACY) { - if (!mlx5_flow_ext_mreg_supported(dev)) - return rte_flow_error_set(error, ENOTSUP, + if (!mlx5_flow_ext_mreg_supported(dev)) + return rte_flow_error_set(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM, item, "extended metadata register" " isn't supported"); - reg = flow_dv_get_metadata_reg(dev, attr, error); - if (reg < 0) - return reg; - if (reg == REG_B) - return rte_flow_error_set(error, ENOTSUP, + reg = flow_dv_get_metadata_reg(dev, attr, error); + if (reg < 0) + return reg; + if (reg == REG_NONE) + return rte_flow_error_set(error, ENOTSUP, + RTE_FLOW_ERROR_TYPE_ITEM, item, + "unavailable metadata register"); + if (reg == REG_B) + return rte_flow_error_set(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM, item, "match on reg_b " "isn't supported"); - if (reg != REG_A) - nic_mask.data = priv->sh->dv_meta_mask; - } + if (reg != REG_A) + nic_mask.data = priv->sh->dv_meta_mask; if (!mask) mask = &rte_flow_item_meta_mask; if (!mask->data) @@ -2244,6 +2246,11 @@ flow_dv_validate_action_set_meta(struct rte_eth_dev *dev, reg = flow_dv_get_metadata_reg(dev, attr, error); if (reg < 0) return reg; + if (reg == REG_NONE) + return rte_flow_error_set(error, ENOTSUP, + RTE_FLOW_ERROR_TYPE_ACTION, + action, "unavailable " + "metadata register"); if (reg != REG_A && reg != REG_B) { struct mlx5_priv *priv = dev->data->dev_private; -- 2.26.1
