[v2] net/i40e: support enabling/disabling source pruning
Checks
Commit Message
VRRP advertisement packets are dropped on i40e PF devices because
when a MAC address is added to a device, packets originating from
that MAC address are dropped.
This patch adds a PMD specific API to enable/disable source
pruning to fix above issue.
Bugzilla ID: 648
Signed-off-by: Ke Zhang <ke1x.zhang@intel.com>
---
app/test-pmd/cmdline.c | 84 +++++++++++++++++++++++++++++++++
drivers/net/i40e/i40e_ethdev.c | 43 +++++++++++++++++
drivers/net/i40e/i40e_ethdev.h | 1 +
drivers/net/i40e/rte_pmd_i40e.c | 20 ++++++++
drivers/net/i40e/rte_pmd_i40e.h | 16 +++++++
5 files changed, 164 insertions(+)
Comments
> From: Ke Zhang [mailto:ke1x.zhang@intel.com]
> Sent: Monday, 30 January 2023 09.09
>
> VRRP advertisement packets are dropped on i40e PF devices because
> when a MAC address is added to a device, packets originating from
> that MAC address are dropped.
>
> This patch adds a PMD specific API to enable/disable source
> pruning to fix above issue.
>
> Bugzilla ID: 648
>
> Signed-off-by: Ke Zhang <ke1x.zhang@intel.com>
> ---
This fixes what I considered a bug in the driver, so source pruning is now disabled by default. Thank you.
Acked-by: Morten Brørup <mb@smartsharesystems.com>
Should rte_pmd_i40e_set_src_prune be added to a maps file? (I'm not sure about a PMD specific APIs, so please check.)
On Mon, Jan 30, 2023 at 9:23 AM Ke Zhang <ke1x.zhang@intel.com> wrote:
>
> VRRP advertisement packets are dropped on i40e PF devices because
> when a MAC address is added to a device, packets originating from
> that MAC address are dropped.
>
> This patch adds a PMD specific API to enable/disable source
> pruning to fix above issue.
>
> Bugzilla ID: 648
>
> Signed-off-by: Ke Zhang <ke1x.zhang@intel.com>
> ---
> app/test-pmd/cmdline.c | 84 +++++++++++++++++++++++++++++++++
> drivers/net/i40e/i40e_ethdev.c | 43 +++++++++++++++++
> drivers/net/i40e/i40e_ethdev.h | 1 +
> drivers/net/i40e/rte_pmd_i40e.c | 20 ++++++++
> drivers/net/i40e/rte_pmd_i40e.h | 16 +++++++
> 5 files changed, 164 insertions(+)
>
> diff --git a/app/test-pmd/cmdline.c b/app/test-pmd/cmdline.c
> index cb8c174020..76a574affd 100644
> --- a/app/test-pmd/cmdline.c
> +++ b/app/test-pmd/cmdline.c
> @@ -776,6 +776,9 @@ static void cmd_help_long_parsed(void *parsed_result,
>
> "port cleanup (port_id) txq (queue_id) (free_cnt)\n"
> " Cleanup txq mbufs for a specific Tx queue\n\n"
> +
> + "port config (port_id|all) src_prune (on|off)\n"
> + " Set source prune on port_id, or all.\n\n"
> );
> }
>
- This seems i40e specific, please move to drivers/net/i40e/i40e_testpmd.c.
- Besides, I would prefer that something in the command name clearly
states this is driver (here, i40e) specific.
Like "port config XX i40e_src_prune" or maybe the other way around,
start with a "driver i40e" prefix.
Maybe others have an opinion.
> -----Original Message-----
> From: David Marchand <david.marchand@redhat.com>
> Sent: Monday, January 30, 2023 4:58 PM
> To: Zhang, Ke1X <ke1x.zhang@intel.com>; mb@smartsharesystems.com;
> thomas@monjalon.net; ferruh.yigit@amd.com
> Cc: dev@dpdk.org; Matz, Olivier <olivier.matz@6wind.com>; Zhang, Yuying
> <yuying.zhang@intel.com>; Xing, Beilei <beilei.xing@intel.com>
> Subject: Re: [PATCH v2] net/i40e: support enabling/disabling source pruning
>
> On Mon, Jan 30, 2023 at 9:23 AM Ke Zhang <ke1x.zhang@intel.com> wrote:
> >
> > VRRP advertisement packets are dropped on i40e PF devices because
> when
> > a MAC address is added to a device, packets originating from that MAC
> > address are dropped.
> >
> > This patch adds a PMD specific API to enable/disable source pruning to
> > fix above issue.
> >
> > Bugzilla ID: 648
> >
> > Signed-off-by: Ke Zhang <ke1x.zhang@intel.com>
> > ---
> > app/test-pmd/cmdline.c | 84
> +++++++++++++++++++++++++++++++++
> > drivers/net/i40e/i40e_ethdev.c | 43 +++++++++++++++++
> > drivers/net/i40e/i40e_ethdev.h | 1 +
> > drivers/net/i40e/rte_pmd_i40e.c | 20 ++++++++
> > drivers/net/i40e/rte_pmd_i40e.h | 16 +++++++
> > 5 files changed, 164 insertions(+)
> >
> > diff --git a/app/test-pmd/cmdline.c b/app/test-pmd/cmdline.c index
> > cb8c174020..76a574affd 100644
> > --- a/app/test-pmd/cmdline.c
> > +++ b/app/test-pmd/cmdline.c
> > @@ -776,6 +776,9 @@ static void cmd_help_long_parsed(void
> > *parsed_result,
> >
> > "port cleanup (port_id) txq (queue_id) (free_cnt)\n"
> > " Cleanup txq mbufs for a specific Tx queue\n\n"
> > +
> > + "port config (port_id|all) src_prune (on|off)\n"
> > + " Set source prune on port_id, or all.\n\n"
> > );
> > }
> >
>
> - This seems i40e specific, please move to drivers/net/i40e/i40e_testpmd.c.
>
> - Besides, I would prefer that something in the command name clearly states
> this is driver (here, i40e) specific.
> Like "port config XX i40e_src_prune" or maybe the other way around, start
> with a "driver i40e" prefix.
>
> Maybe others have an opinion.
>
>
> --
> David Marchand
Thanks for your comments, this command could be used for other Intel NIC like ice,
And It is only finished for i40e in this story: https://jira.devtools.intel.com/browse/DPDK-29564
31/01/2023 04:28, Zhang, Ke1X:
> From: David Marchand <david.marchand@redhat.com>
> > On Mon, Jan 30, 2023 at 9:23 AM Ke Zhang <ke1x.zhang@intel.com> wrote:
> > >
> > > VRRP advertisement packets are dropped on i40e PF devices because
> > when
> > > a MAC address is added to a device, packets originating from that MAC
> > > address are dropped.
> > >
> > > This patch adds a PMD specific API to enable/disable source pruning to
> > > fix above issue.
> > >
> > > Bugzilla ID: 648
> > >
> > > Signed-off-by: Ke Zhang <ke1x.zhang@intel.com>
> > > ---
> > > app/test-pmd/cmdline.c | 84
> > +++++++++++++++++++++++++++++++++
> > > drivers/net/i40e/i40e_ethdev.c | 43 +++++++++++++++++
> > > drivers/net/i40e/i40e_ethdev.h | 1 +
> > > drivers/net/i40e/rte_pmd_i40e.c | 20 ++++++++
> > > drivers/net/i40e/rte_pmd_i40e.h | 16 +++++++
> > > 5 files changed, 164 insertions(+)
> > >
> > > diff --git a/app/test-pmd/cmdline.c b/app/test-pmd/cmdline.c index
> > > cb8c174020..76a574affd 100644
> > > --- a/app/test-pmd/cmdline.c
> > > +++ b/app/test-pmd/cmdline.c
> > > @@ -776,6 +776,9 @@ static void cmd_help_long_parsed(void
> > > *parsed_result,
> > >
> > > "port cleanup (port_id) txq (queue_id) (free_cnt)\n"
> > > " Cleanup txq mbufs for a specific Tx queue\n\n"
> > > +
> > > + "port config (port_id|all) src_prune (on|off)\n"
> > > + " Set source prune on port_id, or all.\n\n"
> > > );
> > > }
> > >
> >
> > - This seems i40e specific, please move to drivers/net/i40e/i40e_testpmd.c.
> >
> > - Besides, I would prefer that something in the command name clearly states
> > this is driver (here, i40e) specific.
> > Like "port config XX i40e_src_prune" or maybe the other way around, start
> > with a "driver i40e" prefix.
> >
> > Maybe others have an opinion.
> >
> >
> > --
> > David Marchand
>
> Thanks for your comments, this command could be used for other Intel NIC like ice,
> And It is only finished for i40e in this story: https://jira.devtools.intel.com/browse/DPDK-29564
Only Intel has access to this URL.
On the principle, you should not update testpmd commands in a patch for i40e.
We add testpmd commands when adding new features in ethdev.
Here specifically, you are adding rte_pmd_i40e_set_src_prune()
with a vague explanation: "Enable/Disable source prune on all the PF."
Without more information, I cannot judge whether the feature is generic or not.
So for now, this is a nack.
> -----Original Message-----
> From: Thomas Monjalon <thomas@monjalon.net>
> Sent: Wednesday, February 1, 2023 7:11 PM
> To: David Marchand <david.marchand@redhat.com>;
> mb@smartsharesystems.com; ferruh.yigit@amd.com; Zhang, Ke1X
> <ke1x.zhang@intel.com>
> Cc: dev@dpdk.org; Matz, Olivier <olivier.matz@6wind.com>; Zhang, Yuying
> <yuying.zhang@intel.com>; Xing, Beilei <beilei.xing@intel.com>
> Subject: Re: [PATCH v2] net/i40e: support enabling/disabling source pruning
>
> 31/01/2023 04:28, Zhang, Ke1X:
> > From: David Marchand <david.marchand@redhat.com>
> > > On Mon, Jan 30, 2023 at 9:23 AM Ke Zhang <ke1x.zhang@intel.com>
> wrote:
> > > >
> > > > VRRP advertisement packets are dropped on i40e PF devices because
> > > when
> > > > a MAC address is added to a device, packets originating from that
> > > > MAC address are dropped.
> > > >
> > > > This patch adds a PMD specific API to enable/disable source
> > > > pruning to fix above issue.
> > > >
> > > > Bugzilla ID: 648
> > > >
> > > > Signed-off-by: Ke Zhang <ke1x.zhang@intel.com>
> > > > ---
> > > > app/test-pmd/cmdline.c | 84
> > > +++++++++++++++++++++++++++++++++
> > > > drivers/net/i40e/i40e_ethdev.c | 43 +++++++++++++++++
> > > > drivers/net/i40e/i40e_ethdev.h | 1 +
> > > > drivers/net/i40e/rte_pmd_i40e.c | 20 ++++++++
> > > > drivers/net/i40e/rte_pmd_i40e.h | 16 +++++++
> > > > 5 files changed, 164 insertions(+)
> > > >
> > > > diff --git a/app/test-pmd/cmdline.c b/app/test-pmd/cmdline.c index
> > > > cb8c174020..76a574affd 100644
> > > > --- a/app/test-pmd/cmdline.c
> > > > +++ b/app/test-pmd/cmdline.c
> > > > @@ -776,6 +776,9 @@ static void cmd_help_long_parsed(void
> > > > *parsed_result,
> > > >
> > > > "port cleanup (port_id) txq (queue_id) (free_cnt)\n"
> > > > " Cleanup txq mbufs for a specific Tx queue\n\n"
> > > > +
> > > > + "port config (port_id|all) src_prune (on|off)\n"
> > > > + " Set source prune on port_id, or all.\n\n"
> > > > );
> > > > }
> > > >
> > >
> > > - This seems i40e specific, please move to
> drivers/net/i40e/i40e_testpmd.c.
> > >
> > > - Besides, I would prefer that something in the command name clearly
> > > states this is driver (here, i40e) specific.
> > > Like "port config XX i40e_src_prune" or maybe the other way around,
> > > start with a "driver i40e" prefix.
> > >
> > > Maybe others have an opinion.
> > >
> > >
> > > --
> > > David Marchand
> >
> > Thanks for your comments, this command could be used for other Intel
> > NIC like ice, And It is only finished for i40e in this story:
> > https://jira.devtools.intel.com/browse/DPDK-29564
>
> Only Intel has access to this URL.
>
> On the principle, you should not update testpmd commands in a patch for
> i40e.
> We add testpmd commands when adding new features in ethdev.
>
> Here specifically, you are adding rte_pmd_i40e_set_src_prune() with a
> vague explanation: "Enable/Disable source prune on all the PF."
> Without more information, I cannot judge whether the feature is generic or
> not.
> So for now, this is a nack.
>
Thanks for your comments, this feature is only for intel NIC.
Would you please share a solution for this case?
Thanks.
07/02/2023 02:40, Zhang, Ke1X:
> From: Thomas Monjalon <thomas@monjalon.net>
> > 31/01/2023 04:28, Zhang, Ke1X:
> > > From: David Marchand <david.marchand@redhat.com>
> > > > On Mon, Jan 30, 2023 at 9:23 AM Ke Zhang <ke1x.zhang@intel.com>
> > wrote:
> > > > >
> > > > > VRRP advertisement packets are dropped on i40e PF devices because
> > > > when
> > > > > a MAC address is added to a device, packets originating from that
> > > > > MAC address are dropped.
> > > > >
> > > > > This patch adds a PMD specific API to enable/disable source
> > > > > pruning to fix above issue.
> > > > >
> > > > > Bugzilla ID: 648
> > > > >
> > > > > Signed-off-by: Ke Zhang <ke1x.zhang@intel.com>
> > > > > ---
> > > > > app/test-pmd/cmdline.c | 84
> > > > +++++++++++++++++++++++++++++++++
> > > > > drivers/net/i40e/i40e_ethdev.c | 43 +++++++++++++++++
> > > > > drivers/net/i40e/i40e_ethdev.h | 1 +
> > > > > drivers/net/i40e/rte_pmd_i40e.c | 20 ++++++++
> > > > > drivers/net/i40e/rte_pmd_i40e.h | 16 +++++++
> > > > > 5 files changed, 164 insertions(+)
> > > > >
> > > > > diff --git a/app/test-pmd/cmdline.c b/app/test-pmd/cmdline.c index
> > > > > cb8c174020..76a574affd 100644
> > > > > --- a/app/test-pmd/cmdline.c
> > > > > +++ b/app/test-pmd/cmdline.c
> > > > > @@ -776,6 +776,9 @@ static void cmd_help_long_parsed(void
> > > > > *parsed_result,
> > > > >
> > > > > "port cleanup (port_id) txq (queue_id) (free_cnt)\n"
> > > > > " Cleanup txq mbufs for a specific Tx queue\n\n"
> > > > > +
> > > > > + "port config (port_id|all) src_prune (on|off)\n"
> > > > > + " Set source prune on port_id, or all.\n\n"
> > > > > );
> > > > > }
> > > > >
> > > >
> > > > - This seems i40e specific, please move to
> > drivers/net/i40e/i40e_testpmd.c.
> > > >
> > > > - Besides, I would prefer that something in the command name clearly
> > > > states this is driver (here, i40e) specific.
> > > > Like "port config XX i40e_src_prune" or maybe the other way around,
> > > > start with a "driver i40e" prefix.
> > > >
> > > > Maybe others have an opinion.
> > > >
> > > >
> > > > --
> > > > David Marchand
> > >
> > > Thanks for your comments, this command could be used for other Intel
> > > NIC like ice, And It is only finished for i40e in this story:
> > > https://jira.devtools.intel.com/browse/DPDK-29564
> >
> > Only Intel has access to this URL.
> >
> > On the principle, you should not update testpmd commands in a patch for
> > i40e.
> > We add testpmd commands when adding new features in ethdev.
> >
> > Here specifically, you are adding rte_pmd_i40e_set_src_prune() with a
> > vague explanation: "Enable/Disable source prune on all the PF."
> > Without more information, I cannot judge whether the feature is generic or
> > not.
> > So for now, this is a nack.
> >
> Thanks for your comments, this feature is only for intel NIC.
> Would you please share a solution for this case?
Host testpmd code in your driver directory
with a command name starting with the name of your driver: "i40e port ...".
You can look at what is done for mlx5 in drivers/net/mlx5/mlx5_testpmd.c
@@ -776,6 +776,9 @@ static void cmd_help_long_parsed(void *parsed_result,
"port cleanup (port_id) txq (queue_id) (free_cnt)\n"
" Cleanup txq mbufs for a specific Tx queue\n\n"
+
+ "port config (port_id|all) src_prune (on|off)\n"
+ " Set source prune on port_id, or all.\n\n"
);
}
@@ -1619,6 +1622,84 @@ static cmdline_parse_inst_t cmd_config_loopback_specific = {
},
};
+/* *** configure source prune for port *** */
+struct cmd_config_src_prune_result {
+ cmdline_fixed_string_t port;
+ cmdline_fixed_string_t keyword;
+ cmdline_fixed_string_t port_all; /* valid if "allports" argument == 1 */
+ uint16_t port_num; /* valid if "allports" argument == 0 */
+ cmdline_fixed_string_t item;
+ cmdline_fixed_string_t enable;
+};
+
+static void cmd_config_src_prune_parsed(void *parsed_result,
+ __rte_unused struct cmdline *cl,
+ void *allports)
+{
+ struct cmd_config_src_prune_result *res = parsed_result;
+ int enable;
+ portid_t i;
+
+ if (!strcmp(res->enable, "on"))
+ enable = 1;
+ else
+ enable = 0;
+
+ /* all ports */
+ if (allports) {
+ RTE_ETH_FOREACH_DEV(i)
+ rte_pmd_i40e_set_src_prune(i, enable);
+ } else {
+ rte_pmd_i40e_set_src_prune(res->port_num, enable);
+ }
+}
+
+static cmdline_parse_token_string_t cmd_config_src_prune_port =
+ TOKEN_STRING_INITIALIZER(struct cmd_config_src_prune_result, port, "port");
+static cmdline_parse_token_string_t cmd_config_src_prune_keyword =
+ TOKEN_STRING_INITIALIZER(struct cmd_config_src_prune_result, keyword,
+ "config");
+static cmdline_parse_token_string_t cmd_config_src_prune_portall =
+ TOKEN_STRING_INITIALIZER(struct cmd_config_src_prune_result, port_all,
+ "all");
+static cmdline_parse_token_num_t cmd_config_src_prune_portnum =
+ TOKEN_NUM_INITIALIZER(struct cmd_config_src_prune_result, port_num,
+ RTE_UINT16);
+static cmdline_parse_token_string_t cmd_config_src_prune_item =
+ TOKEN_STRING_INITIALIZER(struct cmd_config_src_prune_result,
+ item, "src_prune");
+static cmdline_parse_token_string_t cmd_config_src_prune_enable =
+ TOKEN_STRING_INITIALIZER(struct cmd_config_src_prune_result, enable,
+ "on#off");
+
+static cmdline_parse_inst_t cmd_config_src_prune_all = {
+ .f = cmd_config_src_prune_parsed,
+ .data = (void *)1,
+ .help_str = "port config all src_prune (on|off): Set source prune on all ports.",
+ .tokens = {
+ (void *)&cmd_config_src_prune_port,
+ (void *)&cmd_config_src_prune_keyword,
+ (void *)&cmd_config_src_prune_portall,
+ (void *)&cmd_config_src_prune_item,
+ (void *)&cmd_config_src_prune_enable,
+ NULL,
+ },
+};
+
+static cmdline_parse_inst_t cmd_config_src_prune_specific = {
+ .f = cmd_config_src_prune_parsed,
+ .data = (void *)0,
+ .help_str = "port config all src_prune (on|off): Set source prune on specific port.",
+ .tokens = {
+ (void *)&cmd_config_src_prune_port,
+ (void *)&cmd_config_src_prune_keyword,
+ (void *)&cmd_config_src_prune_portnum,
+ (void *)&cmd_config_src_prune_item,
+ (void *)&cmd_config_src_prune_enable,
+ NULL,
+ },
+};
+
/* *** configure txq/rxq, txd/rxd *** */
struct cmd_config_rx_tx {
cmdline_fixed_string_t port;
@@ -12731,6 +12812,8 @@ static cmdline_parse_ctx_t builtin_ctx[] = {
(cmdline_parse_inst_t *)&cmd_config_speed_specific,
(cmdline_parse_inst_t *)&cmd_config_loopback_all,
(cmdline_parse_inst_t *)&cmd_config_loopback_specific,
+ (cmdline_parse_inst_t *)&cmd_config_src_prune_all,
+ (cmdline_parse_inst_t *)&cmd_config_src_prune_specific,
(cmdline_parse_inst_t *)&cmd_config_rx_tx,
(cmdline_parse_inst_t *)&cmd_config_mtu,
(cmdline_parse_inst_t *)&cmd_config_max_pkt_len,
@@ -12850,6 +12933,7 @@ static cmdline_parse_ctx_t builtin_ctx[] = {
(cmdline_parse_inst_t *)&cmd_operate_bpf_unld_parse,
#endif
(cmdline_parse_inst_t *)&cmd_config_tx_metadata_specific,
+
(cmdline_parse_inst_t *)&cmd_show_tx_metadata,
(cmdline_parse_inst_t *)&cmd_show_rx_tx_desc_status,
(cmdline_parse_inst_t *)&cmd_show_rx_queue_desc_used_count,
@@ -5634,6 +5634,46 @@ i40e_enable_pf_lb(struct i40e_pf *pf)
hw->aq.asq_last_status);
}
+/* i40e_set_pf_source_prune
+ * @pf: pointer to the pf structure
+ * @on: Enable/disable source prune
+ *
+ * set source prune on pf
+ */
+int
+i40e_set_pf_source_prune(struct i40e_pf *pf, int on)
+{
+ struct i40e_hw *hw = I40E_PF_TO_HW(pf);
+ struct i40e_vsi_context ctxt;
+ int ret;
+
+ memset(&ctxt, 0, sizeof(ctxt));
+ ctxt.seid = pf->main_vsi_seid;
+ ctxt.pf_num = hw->pf_id;
+ ret = i40e_aq_get_vsi_params(hw, &ctxt, NULL);
+ if (ret) {
+ PMD_DRV_LOG(ERR, "cannot get pf vsi config, err %d, aq_err %d",
+ ret, hw->aq.asq_last_status);
+ return ret;
+ }
+ ctxt.flags = I40E_AQ_VSI_TYPE_PF;
+ ctxt.info.valid_sections =
+ rte_cpu_to_le_16(I40E_AQ_VSI_PROP_SWITCH_VALID);
+ if (on)
+ ctxt.info.switch_id &=
+ ~rte_cpu_to_le_16(I40E_AQ_VSI_SW_ID_FLAG_LOCAL_LB);
+ else
+ ctxt.info.switch_id |=
+ rte_cpu_to_le_16(I40E_AQ_VSI_SW_ID_FLAG_LOCAL_LB);
+
+ ret = i40e_aq_update_vsi_params(hw, &ctxt, NULL);
+ if (ret)
+ PMD_DRV_LOG(ERR, "update vsi switch failed, aq_err=%d",
+ hw->aq.asq_last_status);
+
+ return ret;
+}
+
/* Setup a VSI */
struct i40e_vsi *
i40e_vsi_setup(struct i40e_pf *pf,
@@ -5691,6 +5731,9 @@ i40e_vsi_setup(struct i40e_pf *pf,
}
}
+ /* source prune is disabled to support VRRP in default*/
+ i40e_set_pf_source_prune(pf, 0);
+
vsi = rte_zmalloc("i40e_vsi", sizeof(struct i40e_vsi), 0);
if (!vsi) {
PMD_DRV_LOG(ERR, "Failed to allocate memory for vsi");
@@ -1427,6 +1427,7 @@ int i40e_pf_calc_configured_queues_num(struct i40e_pf *pf);
int i40e_pf_reset_rss_reta(struct i40e_pf *pf);
int i40e_pf_reset_rss_key(struct i40e_pf *pf);
int i40e_pf_config_rss(struct i40e_pf *pf);
+int i40e_set_pf_source_prune(struct i40e_pf *pf, int on);
int i40e_set_rss_key(struct i40e_vsi *vsi, uint8_t *key, uint8_t key_len);
int i40e_set_rss_lut(struct i40e_vsi *vsi, uint8_t *lut, uint16_t lut_size);
int i40e_vf_representor_init(struct rte_eth_dev *ethdev, void *init_params);
@@ -445,6 +445,26 @@ rte_pmd_i40e_set_tx_loopback(uint16_t port, uint8_t on)
return ret;
}
+int
+rte_pmd_i40e_set_src_prune(uint16_t port, uint8_t on)
+{
+ struct rte_eth_dev *dev;
+ struct i40e_pf *pf;
+ int ret;
+
+ RTE_ETH_VALID_PORTID_OR_ERR_RET(port, -ENODEV);
+
+ dev = &rte_eth_devices[port];
+
+ if (!is_i40e_supported(dev))
+ return -ENOTSUP;
+
+ pf = I40E_DEV_PRIVATE_TO_PF(dev->data->dev_private);
+
+ ret = i40e_set_pf_source_prune(pf, on);
+ return ret;
+}
+
int
rte_pmd_i40e_set_vf_unicast_promisc(uint16_t port, uint16_t vf_id, uint8_t on)
{
@@ -400,6 +400,22 @@ int rte_pmd_i40e_set_vf_vlan_anti_spoof(uint16_t port,
int rte_pmd_i40e_set_tx_loopback(uint16_t port,
uint8_t on);
+/**
+ * Enable/Disable source prune on all the PF.
+ *
+ * @param port
+ * The port identifier of the Ethernet device.
+ * @param on
+ * 1 - Enable source prune.
+ * 0 - Disable source prune.
+ * @return
+ * - (0) if successful.
+ * - (-ENODEV) if *port* invalid.
+ * - (-EINVAL) if bad parameter.
+ */
+int rte_pmd_i40e_set_src_prune(uint16_t port,
+ uint8_t on);
+
/**
* Enable/Disable VF unicast promiscuous mode.
*