[dpdk-dev] [PATCH] i40e: workaround for Security issue in SR-IOV mode

[Vladislav Zolotarov]

On Sep 25, 2015 11:44 AM, "Jingjing Wu" <jingjing.wu at intel.com> wrote:
>
> In SR-IOV mode a VF sending LFC or PFC would throttle the entire port.
> The workaround is to add a filter to drop pause frames from VFs from
> sending pause frames.

This is a very strange approach - this would silently disable the Tx FC
while a user would think it's enabled. Wouldn't the right approach be to
let the user decide weather to enable this feature or even better - allow
PF to disable this feature in the VF?

>
> Signed-off-by: Jingjing Wu <jingjing.wu at intel.com>
> ---
>  drivers/net/i40e/i40e_ethdev.c | 30 ++++++++++++++++++++++++++++++
>  1 file changed, 30 insertions(+)
>
> diff --git a/drivers/net/i40e/i40e_ethdev.c
b/drivers/net/i40e/i40e_ethdev.c
> index 2dd9fdc..6cc2172 100644
> --- a/drivers/net/i40e/i40e_ethdev.c
> +++ b/drivers/net/i40e/i40e_ethdev.c
> @@ -382,6 +382,30 @@ static inline void i40e_flex_payload_reg_init(struct
i40e_hw *hw)
>         I40E_WRITE_REG(hw, I40E_GLQF_PIT(17), 0x00007440);
>  }
>
> +#define I40E_FLOW_CONTROL_ETHERTYPE  0x8808
> +
> +/*
> + * Add a ethertype filter to drop all flow control frames transimited
> + * from VSIs.
> +*/
> +static void
> +i40e_add_tx_flow_control_drop_filter(struct i40e_pf *pf)
> +{
> +       struct i40e_hw *hw = I40E_PF_TO_HW(pf);
> +       uint16_t flags = I40E_AQC_ADD_CONTROL_PACKET_FLAGS_IGNORE_MAC |
> +                       I40E_AQC_ADD_CONTROL_PACKET_FLAGS_DROP |
> +                       I40E_AQC_ADD_CONTROL_PACKET_FLAGS_TX;
> +       int ret;
> +
> +       ret = i40e_aq_add_rem_control_packet_filter(hw, NULL,
> +                               I40E_FLOW_CONTROL_ETHERTYPE, flags,
> +                               pf->main_vsi_seid, 0,
> +                               TRUE, NULL, NULL);
> +       if (ret)
> +               PMD_INIT_LOG(ERR, "Failed to add filter to drop flow
control "
> +                                 " frames from VSIs.");
> +}
> +
>  static int
>  eth_i40e_dev_init(struct rte_eth_dev *dev)
>  {
> @@ -584,6 +608,12 @@ eth_i40e_dev_init(struct rte_eth_dev *dev)
>
>         /* enable uio intr after callback register */
>         rte_intr_enable(&(pci_dev->intr_handle));
> +       /*
> +        * Add a ethertype filter to drop all flow control frames
transimited
> +        * from VSIs. This is used to workaround the issue -- in SR-IOV
mode
> +        * where a VF sending LFC or PFC would throttle the entire port.
> +        */
> +       i40e_add_tx_flow_control_drop_filter(pf);
>
>         /* initialize mirror rule list */
>         TAILQ_INIT(&pf->mirror_list);
> --
> 2.4.0
>