Skip to contentSkip to conversation

100GbE Intel Wired LAN Driver Updates 2022-01-06

GIT PULL New Tony Nguyen 2 participants 7 messages Jan 6, 2022 – Jan 7, 2022

Subsystems (3)
lore
Share & tools
lore

Tony Nguyen

author maintainer
#0 lore raw

This series contains updates to ice driver only.

Victor adds restoring of advanced rules after reset.

Wojciech improves usage of switchdev control VSI by utilizing the device's advanced rules for forwarding.

Christophe Jaillet removes some unneeded calls to zero bitmaps, changes some bitmap operations that don't need to be atomic, and converts a kfree() to a more appropriate bitmap_free().

The following are changes since commit 710ad98c363a66a0cd8526465426c5c5f8377ee0:
veth: Do not record rx queue hint in veth_xmit
and are available in the git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/next-queue 100GbE

Christophe JAILLET (3):
ice: Slightly simply ice_find_free_recp_res_idx
ice: Optimize a few bitmap operations
ice: Use bitmap_free() to free bitmap

Victor Raj (1):
ice: replay advanced rules after reset

Wojciech Drewek (1):
ice: improve switchdev's slow-path

 drivers/net/ethernet/intel/ice/ice_common.c   |   2 +-
 drivers/net/ethernet/intel/ice/ice_eswitch.c  | 169 ++++++++++--------
 drivers/net/ethernet/intel/ice/ice_eswitch.h  |  25 +--
 .../net/ethernet/intel/ice/ice_flex_pipe.c    |   6 +-
 drivers/net/ethernet/intel/ice/ice_fltr.c     |  80 ---------
 drivers/net/ethernet/intel/ice/ice_fltr.h     |   3 -
 drivers/net/ethernet/intel/ice/ice_main.c     |  24 +++
 drivers/net/ethernet/intel/ice/ice_ptp.c      |   2 +-
 drivers/net/ethernet/intel/ice/ice_repr.c     |  17 ++
 drivers/net/ethernet/intel/ice/ice_repr.h     |   5 +
 drivers/net/ethernet/intel/ice/ice_switch.c   |  96 +++++++++-
 drivers/net/ethernet/intel/ice/ice_txrx_lib.c |   3 +-
 .../net/ethernet/intel/ice/ice_virtchnl_pf.c  |  11 ++
 13 files changed, 263 insertions(+), 180 deletions(-)
Signature · 1 line
2.31.1
author maintainer
#1 lore raw

PATCH 1/5 ice: replay advanced rules after reset

From: Victor Raj <[email protected]>

ice_replay_vsi_adv_rule will replay advanced rules for a given VSI. Exit this function when list of rules for given recipe is empty. Do not add rule when given vsi_handle does not match vsi_handle from the rule info.

Use ICE_MAX_NUM_RECIPES instead of ICE_SW_LKUP_LAST in order to find advanced rules as well.

---
 drivers/net/ethernet/intel/ice/ice_common.c |  2 +-
 drivers/net/ethernet/intel/ice/ice_switch.c | 41 +++++++++++++++++++--
 2 files changed, 39 insertions(+), 4 deletions(-)
drivers/net/ethernet/intel/ice/ice_common.c +1 -1
index 2a1ee60e85f4..408d15a5b0e3 100644
@@ -4603,7 +4603,7 @@ static int ice_replay_pre_init(struct ice_hw *hw)
 	 * will allow adding rules entries back to filt_rules list,
 	 * which is operational list.
 	 */
-	for (i = 0; i < ICE_SW_LKUP_LAST; i++)
+	for (i = 0; i < ICE_MAX_NUM_RECIPES; i++)
 		list_replace_init(&sw->recp_list[i].filt_rules,
 				  &sw->recp_list[i].filt_replay_rules);
 	ice_sched_replay_agg_vsi_preinit(hw);
drivers/net/ethernet/intel/ice/ice_switch.c +38 -3

58 lines not on this page · raw message

Signature · 1 line
2.31.1
author maintainer
#2 lore raw

PATCH 2/5 ice: improve switchdev's slow-path

In reply to Tony Nguyen

From: Wojciech Drewek <[email protected]>

In current switchdev implementation, every VF PR is assigned to individual ring on switchdev ctrl VSI. For slow-path traffic, there is a mapping VF->ring done in software based on src_vsi value (by calling ice_eswitch_get_target_netdev function).

With this change, HW solution is introduced which is more efficient. For each VF, src MAC (VF's MAC) filter will be created, which forwards packets to the corresponding switchdev ctrl VSI queue based on src MAC address.

This filter has to be removed and then replayed in case of resetting one VF. Keep information about this rule in repr->mac_rule, thanks to that we know which rule has to be removed and replayed for a given VF.

In case of CORE/GLOBAL all rules are removed automatically. We have to take care of readding them. This is done by ice_replay_vsi_adv_rule.

When driver leaves switchdev mode, remove all advanced rules from switchdev ctrl VSI. This is done by ice_rem_adv_rule_for_vsi.

Flag repr->rule_added is needed because in some cases reset might be triggered before VF sends request to add MAC.

---
 drivers/net/ethernet/intel/ice/ice_eswitch.c  | 169 ++++++++++--------
 drivers/net/ethernet/intel/ice/ice_eswitch.h  |  25 +--
 drivers/net/ethernet/intel/ice/ice_fltr.c     |  80 ---------
 drivers/net/ethernet/intel/ice/ice_fltr.h     |   3 -
 drivers/net/ethernet/intel/ice/ice_main.c     |  24 +++
 drivers/net/ethernet/intel/ice/ice_repr.c     |  17 ++
 drivers/net/ethernet/intel/ice/ice_repr.h     |   5 +
 drivers/net/ethernet/intel/ice/ice_switch.c   |  53 ++++++
 drivers/net/ethernet/intel/ice/ice_txrx_lib.c |   3 +-
 .../net/ethernet/intel/ice/ice_virtchnl_pf.c  |  11 ++
 10 files changed, 220 insertions(+), 170 deletions(-)
drivers/net/ethernet/intel/ice/ice_eswitch.c +95 -74
index d1d7389b0bff..864692b157b6 100644
@@ -9,6 +9,100 @@
 #include "ice_devlink.h"
 #include "ice_tc_lib.h"
 
+/**
+ * ice_eswitch_add_vf_mac_rule - add adv rule with VF's MAC
+ * @pf: pointer to PF struct
+ * @vf: pointer to VF struct
+ * @mac: VF's MAC address
+ *
+ * This function adds advanced rule that forwards packets with
+ * VF's MAC address (src MAC) to the corresponding switchdev ctrl VSI queue.
+ */
+int
+ice_eswitch_add_vf_mac_rule(struct ice_pf *pf, struct ice_vf *vf, const u8 *mac)
+{
+	struct ice_vsi *ctrl_vsi = pf->switchdev.control_vsi;
+	struct ice_adv_rule_info rule_info = { 0 };
+	struct ice_adv_lkup_elem *list;
+	struct ice_hw *hw = &pf->hw;
+	const u16 lkups_cnt = 1;
+	int err;
+
+	list = kcalloc(lkups_cnt, sizeof(*list), GFP_ATOMIC);
+	if (!list)
+		return -ENOMEM;
+
+	list[0].type = ICE_MAC_OFOS;
+	ether_addr_copy(list[0].h_u.eth_hdr.src_addr, mac);
+	eth_broadcast_addr(list[0].m_u.eth_hdr.src_addr);
+
+	rule_info.sw_act.flag |= ICE_FLTR_TX;
+	rule_info.sw_act.vsi_handle = ctrl_vsi->idx;
+	rule_info.sw_act.fltr_act = ICE_FWD_TO_Q;
+	rule_info.rx = false;
+	rule_info.sw_act.fwd_id.q_id = hw->func_caps.common_cap.rxq_first_id +
+				       ctrl_vsi->rxq_map[vf->vf_id];
+	rule_info.flags_info.act |= ICE_SINGLE_ACT_LB_ENABLE;
+	rule_info.flags_info.act_valid = true;
+
+	err = ice_add_adv_rule(hw, list, lkups_cnt, &rule_info,
+			       vf->repr->mac_rule);
+	if (err)
+		dev_err(ice_pf_to_dev(pf), "Unable to add VF mac rule in switchdev mode for VF %d",
+			vf->vf_id);
+	else
+		vf->repr->rule_added = true;
+
+	kfree(list);
+	return err;
+}
+
+/**
+ * ice_eswitch_replay_vf_mac_rule - replay adv rule with VF's MAC
+ * @vf: pointer to vF struct
+ *
+ * This function replays VF's MAC rule after reset.
+ */
+void ice_eswitch_replay_vf_mac_rule(struct ice_vf *vf)
+{
+	int err;
+
+	if (!ice_is_switchdev_running(vf->pf))
+		return;
+
+	if (is_valid_ether_addr(vf->hw_lan_addr.addr)) {
+		err = ice_eswitch_add_vf_mac_rule(vf->pf, vf,
+						  vf->hw_lan_addr.addr);
+		if (err) {
+			dev_err(ice_pf_to_dev(vf->pf), "Failed to add MAC %pM for VF %d\n, error %d\n",
+				vf->hw_lan_addr.addr, vf->vf_id, err);
+			return;
+		}
+		vf->num_mac++;
+
+		ether_addr_copy(vf->dev_lan_addr.addr, vf->hw_lan_addr.addr);
+	}
+}
+
+/**
+ * ice_eswitch_del_vf_mac_rule - delete adv rule with VF's MAC
+ * @vf: pointer to the VF struct
+ *
+ * Delete the advanced rule that was used to forward packets with the VF's MAC
+ * address (src MAC) to the corresponding switchdev ctrl VSI queue.
+ */
+void ice_eswitch_del_vf_mac_rule(struct ice_vf *vf)
+{
+	if (!ice_is_switchdev_running(vf->pf))
+		return;
+
+	if (!vf->repr->rule_added)
+		return;
+
+	ice_rem_adv_rule_by_id(&vf->pf->hw, vf->repr->mac_rule);
+	vf->repr->rule_added = false;
+}
+
 /**
  * ice_eswitch_setup_env - configure switchdev HW filters
  * @pf: pointer to PF struct
@@ -21,7 +115,6 @@ static int ice_eswitch_setup_env(struct ice_pf *pf)
 	struct ice_vsi *uplink_vsi = pf->switchdev.uplink_vsi;
 	struct net_device *uplink_netdev = uplink_vsi->netdev;
 	struct ice_vsi *ctrl_vsi = pf->switchdev.control_vsi;
-	struct ice_port_info *pi = pf->hw.port_info;
 	bool rule_added = false;
 
 	ice_vsi_manage_vlan_stripping(ctrl_vsi, false);
@@ -42,29 +135,17 @@ static int ice_eswitch_setup_env(struct ice_pf *pf)
 		rule_added = true;
 	}
 
-	if (ice_cfg_dflt_vsi(pi->hw, ctrl_vsi->idx, true, ICE_FLTR_TX))
-		goto err_def_tx;
-
 	if (ice_vsi_update_security(uplink_vsi, ice_vsi_ctx_set_allow_override))
 		goto err_override_uplink;
 
 	if (ice_vsi_update_security(ctrl_vsi, ice_vsi_ctx_set_allow_override))
 		goto err_override_control;
 
-	if (ice_fltr_update_flags_dflt_rule(ctrl_vsi, pi->dflt_tx_vsi_rule_id,
-					    ICE_FLTR_TX,
-					    ICE_SINGLE_ACT_LB_ENABLE))
-		goto err_update_action;
-
 	return 0;
 
-err_update_action:
-	ice_vsi_update_security(ctrl_vsi, ice_vsi_ctx_clear_allow_override);
 err_override_control:
 	ice_vsi_update_security(uplink_vsi, ice_vsi_ctx_clear_allow_override);
 err_override_uplink:
-	ice_cfg_dflt_vsi(pi->hw, ctrl_vsi->idx, false, ICE_FLTR_TX);
-err_def_tx:
 	if (rule_added)
 		ice_clear_dflt_vsi(uplink_vsi->vsw);
 err_def_rx:
@@ -167,21 +248,11 @@ static int ice_eswitch_setup_reprs(struct ice_pf *pf)
 		netif_keep_dst(vf->repr->netdev);
 	}
 
-	kfree(ctrl_vsi->target_netdevs);
-
-	ctrl_vsi->target_netdevs = kcalloc(max_vsi_num + 1,
-					   sizeof(*ctrl_vsi->target_netdevs),
-					   GFP_KERNEL);
-	if (!ctrl_vsi->target_netdevs)
-		goto err;
-
 	ice_for_each_vf(pf, i) {
 		struct ice_repr *repr = pf->vf[i].repr;
 		struct ice_vsi *vsi = repr->src_vsi;
 		struct metadata_dst *dst;
 
-		ctrl_vsi->target_netdevs[vsi->vsi_num] = repr->netdev;
-
 		dst = repr->dst;
 		dst->u.port_info.port_id = vsi->vsi_num;
 		dst->u.port_info.lower_dev = repr->netdev;
@@ -214,7 +285,6 @@ ice_eswitch_release_reprs(struct ice_pf *pf, struct ice_vsi *ctrl_vsi)
 {
 	int i;
 
-	kfree(ctrl_vsi->target_netdevs);
 	ice_for_each_vf(pf, i) {
 		struct ice_vsi *vsi = pf->vf[i].repr->src_vsi;
 		struct ice_vf *vf = &pf->vf[i];
@@ -320,7 +390,6 @@ static void ice_eswitch_release_env(struct ice_pf *pf)
 
 	ice_vsi_update_security(ctrl_vsi, ice_vsi_ctx_clear_allow_override);
 	ice_vsi_update_security(uplink_vsi, ice_vsi_ctx_clear_allow_override);
-	ice_cfg_dflt_vsi(&pf->hw, ctrl_vsi->idx, false, ICE_FLTR_TX);
 	ice_clear_dflt_vsi(uplink_vsi->vsw);
 	ice_fltr_add_mac_and_broadcast(uplink_vsi,
 				       uplink_vsi->port_info->mac.perm_addr,
@@ -374,24 +443,6 @@ static void ice_eswitch_napi_disable(struct ice_pf *pf)
 		napi_disable(&pf->vf[i].repr->q_vector->napi);
 }
 
-/**
- * ice_eswitch_set_rxdid - configure rxdid on all Rx queues from VSI
- * @vsi: VSI to setup rxdid on
- * @rxdid: flex descriptor id
- */
-static void ice_eswitch_set_rxdid(struct ice_vsi *vsi, u32 rxdid)
-{
-	struct ice_hw *hw = &vsi->back->hw;
-	int i;
-
-	ice_for_each_rxq(vsi, i) {
-		struct ice_rx_ring *ring = vsi->rx_rings[i];
-		u16 pf_q = vsi->rxq_map[ring->q_index];
-
-		ice_write_qrxflxp_cntxt(hw, pf_q, rxdid, 0x3, true);
-	}
-}
-
 /**
  * ice_eswitch_enable_switchdev - configure eswitch in switchdev mode
  * @pf: pointer to PF structure
@@ -425,8 +476,6 @@ static int ice_eswitch_enable_switchdev(struct ice_pf *pf)
 
 	ice_eswitch_napi_enable(pf);
 
-	ice_eswitch_set_rxdid(ctrl_vsi, ICE_RXDID_FLEX_NIC_2);
-
 	return 0;
 
 err_setup_reprs:
@@ -448,6 +497,7 @@ static void ice_eswitch_disable_switchdev(struct ice_pf *pf)
 
 	ice_eswitch_napi_disable(pf);
 	ice_eswitch_release_env(pf);
+	ice_rem_adv_rule_for_vsi(&pf->hw, ctrl_vsi->idx);
 	ice_eswitch_release_reprs(pf, ctrl_vsi);
 	ice_vsi_release(ctrl_vsi);
 	ice_repr_rem_from_all_vfs(pf);
@@ -496,34 +546,6 @@ ice_eswitch_mode_set(struct devlink *devlink, u16 mode,
 	return 0;
 }
 
-/**
- * ice_eswitch_get_target_netdev - return port representor netdev
- * @rx_ring: pointer to Rx ring
- * @rx_desc: pointer to Rx descriptor
- *
- * When working in switchdev mode context (when control VSI is used), this
- * function returns netdev of appropriate port representor. For non-switchdev
- * context, regular netdev associated with Rx ring is returned.
- */
-struct net_device *
-ice_eswitch_get_target_netdev(struct ice_rx_ring *rx_ring,
-			      union ice_32b_rx_flex_desc *rx_desc)
-{
-	struct ice_32b_rx_flex_desc_nic_2 *desc;
-	struct ice_vsi *vsi = rx_ring->vsi;
-	struct ice_vsi *control_vsi;
-	u16 target_vsi_id;
-
-	control_vsi = vsi->back->switchdev.control_vsi;
-	if (vsi != control_vsi)
-		return rx_ring->netdev;
-
-	desc = (struct ice_32b_rx_flex_desc_nic_2 *)rx_desc;
-	target_vsi_id = le16_to_cpu(desc->src_vsi);
-
-	return vsi->target_netdevs[target_vsi_id];
-}
-
 /**
  * ice_eswitch_mode_get - get current eswitch mode
  * @devlink: pointer to devlink structure
@@ -648,7 +670,6 @@ int ice_eswitch_rebuild(struct ice_pf *pf)
 		return status;
 
 	ice_eswitch_napi_enable(pf);
-	ice_eswitch_set_rxdid(ctrl_vsi, ICE_RXDID_FLEX_NIC_2);
 	ice_eswitch_start_all_tx_queues(pf);
 
 	return 0;
drivers/net/ethernet/intel/ice/ice_eswitch.h +14 -11
index 364cd2a79c37..bd58d9d2e565 100644
@@ -20,10 +20,11 @@ bool ice_is_eswitch_mode_switchdev(struct ice_pf *pf);
 void ice_eswitch_update_repr(struct ice_vsi *vsi);
 
 void ice_eswitch_stop_all_tx_queues(struct ice_pf *pf);
-
-struct net_device *
-ice_eswitch_get_target_netdev(struct ice_rx_ring *rx_ring,
-			      union ice_32b_rx_flex_desc *rx_desc);
+int
+ice_eswitch_add_vf_mac_rule(struct ice_pf *pf, struct ice_vf *vf,
+			    const u8 *mac);
+void ice_eswitch_replay_vf_mac_rule(struct ice_vf *vf);
+void ice_eswitch_del_vf_mac_rule(struct ice_vf *vf);
 
 void ice_eswitch_set_target_vsi(struct sk_buff *skb,
 				struct ice_tx_offload_params *off);
@@ -33,6 +34,15 @@ ice_eswitch_port_start_xmit(struct sk_buff *skb, struct net_device *netdev);
 static inline void ice_eswitch_release(struct ice_pf *pf) { }
 
 static inline void ice_eswitch_stop_all_tx_queues(struct ice_pf *pf) { }
+static inline void ice_eswitch_replay_vf_mac_rule(struct ice_vf *vf) { }
+static inline void ice_eswitch_del_vf_mac_rule(struct ice_vf *vf) { }
+
+static inline int
+ice_eswitch_add_vf_mac_rule(struct ice_pf *pf, struct ice_vf *vf,
+			    const u8 *mac)
+{
+	return -EOPNOTSUPP;
+}
 
 static inline void
 ice_eswitch_set_target_vsi(struct sk_buff *skb,
@@ -67,13 +77,6 @@ static inline bool ice_is_eswitch_mode_switchdev(struct ice_pf *pf)
 	return false;
 }
 
-static inline struct net_device *
-ice_eswitch_get_target_netdev(struct ice_rx_ring *rx_ring,
-			      union ice_32b_rx_flex_desc *rx_desc)
-{
-	return rx_ring->netdev;
-}
-
 static inline netdev_tx_t
 ice_eswitch_port_start_xmit(struct sk_buff *skb, struct net_device *netdev)
 {
drivers/net/ethernet/intel/ice/ice_fltr.c +0 -80

83 lines not on this page · raw message

drivers/net/ethernet/intel/ice/ice_fltr.h +0 -3
index 0e65fd021ad2..3eb42479175f 100644
@@ -50,7 +50,4 @@ void ice_fltr_remove_all(struct ice_vsi *vsi);
 int
 ice_fltr_update_flags(struct ice_vsi *vsi, u16 rule_id, u16 recipe_id,
 		      u32 new_flags);
-int
-ice_fltr_update_flags_dflt_rule(struct ice_vsi *vsi, u16 rule_id, u8 direction,
-				u32 new_flags);
 #endif
drivers/net/ethernet/intel/ice/ice_main.c +24 -0
index e29176889c23..30814435f779 100644
@@ -471,6 +471,25 @@ static void ice_pf_dis_all_vsi(struct ice_pf *pf, bool locked)
 		pf->vf_agg_node[node].num_vsis = 0;
 }
 
+/**
+ * ice_clear_sw_switch_recipes - clear switch recipes
+ * @pf: board private structure
+ *
+ * Mark switch recipes as not created in sw structures. There are cases where
+ * rules (especially advanced rules) need to be restored, either re-read from
+ * hardware or added again. For example after the reset. 'recp_created' flag
+ * prevents from doing that and need to be cleared upfront.
+ */
+static void ice_clear_sw_switch_recipes(struct ice_pf *pf)
+{
+	struct ice_sw_recipe *recp;
+	u8 i;
+
+	recp = pf->hw.switch_info->recp_list;
+	for (i = 0; i < ICE_MAX_NUM_RECIPES; i++)
+		recp[i].recp_created = false;
+}
+
 /**
  * ice_prepare_for_reset - prep for reset
  * @pf: board private structure
@@ -501,6 +520,11 @@ ice_prepare_for_reset(struct ice_pf *pf, enum ice_reset_req reset_type)
 	ice_for_each_vf(pf, i)
 		ice_set_vf_state_qs_dis(&pf->vf[i]);
 
+	if (ice_is_eswitch_mode_switchdev(pf)) {
+		if (reset_type != ICE_RESET_PFR)
+			ice_clear_sw_switch_recipes(pf);
+	}
+
 	/* release ADQ specific HW and SW resources */
 	vsi = ice_get_main_vsi(pf);
 	if (!vsi)
drivers/net/ethernet/intel/ice/ice_repr.c +17 -0
index af8e6ef5f571..dcc310e29300 100644
@@ -244,6 +244,14 @@ static int ice_repr_add(struct ice_vf *vf)
 	if (!repr)
 		return -ENOMEM;
 
+#ifdef CONFIG_ICE_SWITCHDEV
+	repr->mac_rule = kzalloc(sizeof(*repr->mac_rule), GFP_KERNEL);
+	if (!repr->mac_rule) {
+		err = -ENOMEM;
+		goto err_alloc_rule;
+	}
+#endif
+
 	repr->netdev = alloc_etherdev(sizeof(struct ice_netdev_priv));
 	if (!repr->netdev) {
 		err =  -ENOMEM;
@@ -287,6 +295,11 @@ static int ice_repr_add(struct ice_vf *vf)
 	free_netdev(repr->netdev);
 	repr->netdev = NULL;
 err_alloc:
+#ifdef CONFIG_ICE_SWITCHDEV
+	kfree(repr->mac_rule);
+	repr->mac_rule = NULL;
+err_alloc_rule:
+#endif
 	kfree(repr);
 	vf->repr = NULL;
 	return err;
@@ -304,6 +317,10 @@ static void ice_repr_rem(struct ice_vf *vf)
 	unregister_netdev(vf->repr->netdev);
 	free_netdev(vf->repr->netdev);
 	vf->repr->netdev = NULL;
+#ifdef CONFIG_ICE_SWITCHDEV
+	kfree(vf->repr->mac_rule);
+	vf->repr->mac_rule = NULL;
+#endif
 	kfree(vf->repr);
 	vf->repr = NULL;
 }
drivers/net/ethernet/intel/ice/ice_repr.h +5 -0
index 806de22933c6..0c77ff050d15 100644
@@ -13,6 +13,11 @@ struct ice_repr {
 	struct ice_q_vector *q_vector;
 	struct net_device *netdev;
 	struct metadata_dst *dst;
+#ifdef CONFIG_ICE_SWITCHDEV
+	/* info about slow path MAC rule  */
+	struct ice_rule_query_data *mac_rule;
+	u8 rule_added;
+#endif
 };
 
 int ice_repr_add_for_all_vfs(struct ice_pf *pf);
drivers/net/ethernet/intel/ice/ice_switch.c +53 -0

59 lines not on this page · raw message

drivers/net/ethernet/intel/ice/ice_txrx_lib.c +1 -2
index 9520b140bdbf..0e87b98e0966 100644
@@ -189,8 +189,7 @@ ice_process_skb_fields(struct ice_rx_ring *rx_ring,
 	ice_rx_hash(rx_ring, rx_desc, skb, ptype);
 
 	/* modifies the skb - consumes the enet header */
-	skb->protocol = eth_type_trans(skb, ice_eswitch_get_target_netdev
-				       (rx_ring, rx_desc));
+	skb->protocol = eth_type_trans(skb, rx_ring->netdev);
 
 	ice_rx_csum(rx_ring, skb, rx_desc, ptype);
 
drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c +11 -0
index 61b2db3342ed..39b80124d282 100644
@@ -1632,6 +1632,8 @@ bool ice_reset_vf(struct ice_vf *vf, bool is_vflr)
 			dev_err(dev, "disabling promiscuous mode failed\n");
 	}
 
+	ice_eswitch_del_vf_mac_rule(vf);
+
 	ice_vf_fdir_exit(vf);
 	ice_vf_fdir_init(vf);
 	/* clean VF control VSI when resetting VF since it should be setup
@@ -1650,6 +1652,7 @@ bool ice_reset_vf(struct ice_vf *vf, bool is_vflr)
 	ice_vf_post_vsi_rebuild(vf);
 	vsi = ice_get_vf_vsi(vf);
 	ice_eswitch_update_repr(vsi);
+	ice_eswitch_replay_vf_mac_rule(vf);
 
 	/* if the VF has been reset allow it to come up again */
 	if (ice_mbx_clear_malvf(&hw->mbx_snapshot, pf->malvfs, ICE_MAX_VF_COUNT, vf->vf_id))
@@ -4496,6 +4499,7 @@ static int ice_vc_repr_add_mac(struct ice_vf *vf, u8 *msg)
 
 	for (i = 0; i < al->num_elements; i++) {
 		u8 *mac_addr = al->list[i].addr;
+		int result;
 
 		if (!is_unicast_ether_addr(mac_addr) ||
 		    ether_addr_equal(mac_addr, vf->hw_lan_addr.addr))
@@ -4507,6 +4511,13 @@ static int ice_vc_repr_add_mac(struct ice_vf *vf, u8 *msg)
 			goto handle_mac_exit;
 		}
 
+		result = ice_eswitch_add_vf_mac_rule(pf, vf, mac_addr);
+		if (result) {
+			dev_err(ice_pf_to_dev(pf), "Failed to add MAC %pM for VF %d\n, error %d\n",
+				mac_addr, vf->vf_id, result);
+			goto handle_mac_exit;
+		}
+
 		ice_vfhw_mac_add(vf, &al->list[i]);
 		vf->num_mac++;
 		break;
Signature · 1 line
2.31.1
author maintainer
#3 lore raw

PATCH 3/5 ice: Slightly simply ice_find_free_recp_res_idx

In reply to Tony Nguyen

From: Christophe JAILLET <[email protected]>

The 'possible_idx' bitmap is set just after it is zeroed, so we can save the first step.

The 'free_idx' bitmap is used only at the end of the function as the result of a bitmap xor operation. So there is no need to explicitly zero it before.

So, slightly simply the code and remove 2 useless 'bitmap_zero()' call

---
 drivers/net/ethernet/intel/ice/ice_switch.c | 2 --
 1 file changed, 2 deletions(-)
drivers/net/ethernet/intel/ice/ice_switch.c +0 -2
index 756e7f91e48d..11ae0bee3590 100644
@@ -4070,10 +4070,8 @@ ice_find_free_recp_res_idx(struct ice_hw *hw, const unsigned long *profiles,
 	DECLARE_BITMAP(used_idx, ICE_MAX_FV_WORDS);
 	u16 bit;
 
-	bitmap_zero(possible_idx, ICE_MAX_FV_WORDS);
 	bitmap_zero(recipes, ICE_MAX_NUM_RECIPES);
 	bitmap_zero(used_idx, ICE_MAX_FV_WORDS);
-	bitmap_zero(free_idx, ICE_MAX_FV_WORDS);
 
 	bitmap_set(possible_idx, 0, ICE_MAX_FV_WORDS);
 
Signature · 1 line
2.31.1
author maintainer
#4 lore raw

PATCH 4/5 ice: Optimize a few bitmap operations

In reply to Tony Nguyen

From: Christophe JAILLET <[email protected]>

When a bitmap is local to a function, it is safe to use the non-atomic __[set|clear]_bit(). No concurrent accesses can occur.

---
 drivers/net/ethernet/intel/ice/ice_flex_pipe.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/net/ethernet/intel/ice/ice_flex_pipe.c +3 -3
index d29197ab3d02..4deb2c9446ec 100644
@@ -4440,7 +4440,7 @@ ice_update_fd_swap(struct ice_hw *hw, u16 prof_id, struct ice_fv_word *es)
 		for (j = 0; j < ICE_FD_SRC_DST_PAIR_COUNT; j++)
 			if (es[i].prot_id == ice_fd_pairs[j].prot_id &&
 			    es[i].off == ice_fd_pairs[j].off) {
-				set_bit(j, pair_list);
+				__set_bit(j, pair_list);
 				pair_start[j] = i;
 			}
 	}
@@ -4710,7 +4710,7 @@ ice_add_prof(struct ice_hw *hw, enum ice_block blk, u64 id, u8 ptypes[],
 			if (test_bit(ptg, ptgs_used))
 				continue;
 
-			set_bit(ptg, ptgs_used);
+			__set_bit(ptg, ptgs_used);
 			/* Check to see there are any attributes for
 			 * this PTYPE, and add them if found.
 			 */
@@ -5339,7 +5339,7 @@ ice_adj_prof_priorities(struct ice_hw *hw, enum ice_block blk, u16 vsig,
 			}
 
 			/* keep track of used ptgs */
-			set_bit(t->tcam[i].ptg, ptgs_used);
+			__set_bit(t->tcam[i].ptg, ptgs_used);
 		}
 	}
 
Signature · 1 line
2.31.1
author maintainer
#5 lore raw

PATCH 5/5 ice: Use bitmap_free() to free bitmap

In reply to Tony Nguyen

From: Christophe JAILLET <[email protected]>

kfree() and bitmap_free() are the same. But using the latter is more consistent when freeing memory allocated with bitmap_zalloc().

---
 drivers/net/ethernet/intel/ice/ice_ptp.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/intel/ice/ice_ptp.c +1 -1
index d3f65e061a62..ae291d442539 100644
@@ -2229,7 +2229,7 @@ ice_ptp_release_tx_tracker(struct ice_pf *pf, struct ice_ptp_tx *tx)
 	kfree(tx->tstamps);
 	tx->tstamps = NULL;
 
-	kfree(tx->in_use);
+	bitmap_free(tx->in_use);
 	tx->in_use = NULL;
 
 	tx->len = 0;
Signature · 1 line
2.31.1
#6 lore raw
In reply to Tony Nguyen
This series was applied to netdev/net-next.git (master) by Tony Nguyen <[email protected]>:

Hello:

This series was applied to netdev/net-next.git (master) by Tony Nguyen <[email protected]>:

Quoted 8 lines · On Thu, 6 Jan 2022 10:30:08 -0800 you wrote:

This series contains updates to ice driver only.

Victor adds restoring of advanced rules after reset.

Wojciech improves usage of switchdev control VSI by utilizing the device's advanced rules for forwarding.

[...]

Here is the summary with links:
- [net-next,1/5] ice: replay advanced rules after reset

    https://git.kernel.org/netdev/net-next/c/c36a2b971627
  - [net-next,2/5] ice: improve switchdev's slow-path
    https://git.kernel.org/netdev/net-next/c/c1e5da5dd465
  - [net-next,3/5] ice: Slightly simply ice_find_free_recp_res_idx
    https://git.kernel.org/netdev/net-next/c/a5c259b16284
  - [net-next,4/5] ice: Optimize a few bitmap operations
    https://git.kernel.org/netdev/net-next/c/e75ed29db531
  - [net-next,5/5] ice: Use bitmap_free() to free bitmap
    https://git.kernel.org/netdev/net-next/c/0dbc41621875

You are awesome, thank you!

Signature · 2 lines
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html