[PATCH net-next 00/15][pull request] 100GbE Intel Wired LAN Driver Updates 2021-04-08

[PATCH net-next 00/15][pull request] 100GbE Intel Wired LAN Driver Updates 2021-04-08

[Tony Nguyen]

This series contains updates to ice driver only.

Chinh adds retrying of sending some AQ commands when receiving EBUSY
error.

Victor modifies how nodes are added to reduce stack usage.

Ani renames some variables to either follow spec naming or to be inline
with naming in the rest of the driver. Ignores EMODE error as there are
cases where this error is expected. Performs some cleanup such as
removing unnecessary checks, doing variable assignments over copies, and
removing unneeded variables. Revises some error codes returned in link
settings to be more appropriate. He also implements support for new
firmware option to get default link configuration which accounts for
any needed NVM based overrides for PHY configuration. He also removes
the rx_gro_dropped stat as the value no longer changes.

Jeb removes setting specific link modes on firmwares that no longer
require it.

Brett removes unnecessary checks when adding and removing VLANs.

Tony fixes a checkpatch warning for unnecessary blank line.

The following are changes since commit 3cd52c1e32fe7dfee09815ced702db9ee9f84ec9:
  net: fealnx: use module_pci_driver to simplify the code
and are available in the git repository at:
  git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/next-queue 100GbE

Anirudh Venkataramanan (10):
  ice: Align macro names to the specification
  ice: Ignore EMODE return for opcode 0x0605
  ice: Remove unnecessary checker loop
  ice: Rename a couple of variables
  ice: Fix error return codes in ice_set_link_ksettings
  ice: Replace some memsets and memcpys with assignment
  ice: Use default configuration mode for PHY configuration
  ice: Remove unnecessary variable
  ice: Use local variable instead of pointer derefs
  ice: Remove rx_gro_dropped stat

Brett Creeley (1):
  ice: Remove unnecessary checks in add/kill_vid ndo ops

Chinh T Cao (1):
  ice: Re-send some AQ commands, as result of EBUSY AQ error

Jeb Cramer (1):
  ice: Limit forced overrides based on FW version

Tony Nguyen (1):
  ice: Remove unnecessary blank line

Victor Raj (1):
  ice: Modify recursive way of adding nodes

 drivers/net/ethernet/intel/ice/ice.h          |   1 -
 .../net/ethernet/intel/ice/ice_adminq_cmd.h   |  20 ++-
 drivers/net/ethernet/intel/ice/ice_common.c   | 139 ++++++++++++---
 drivers/net/ethernet/intel/ice/ice_common.h   |   4 +
 drivers/net/ethernet/intel/ice/ice_controlq.c |   2 +-
 drivers/net/ethernet/intel/ice/ice_ethtool.c  | 165 ++++++++----------
 drivers/net/ethernet/intel/ice/ice_lib.c      |  37 ++++
 drivers/net/ethernet/intel/ice/ice_lib.h      |   2 +
 drivers/net/ethernet/intel/ice/ice_main.c     | 135 ++++++--------
 drivers/net/ethernet/intel/ice/ice_sched.c    | 130 ++++++++------
 drivers/net/ethernet/intel/ice/ice_txrx.h     |   1 -
 drivers/net/ethernet/intel/ice/ice_type.h     |   5 +
 12 files changed, 391 insertions(+), 250 deletions(-)

-- 
2.26.2

[PATCH net-next 01/15] ice: Re-send some AQ commands, as result of EBUSY AQ error

[Tony Nguyen]

From: Chinh T Cao <chinh.t.cao@intel.com>

Retry sending some AQ commands, as result of EBUSY AQ error.
ice_aqc_opc_get_link_topo
ice_aqc_opc_lldp_stop
ice_aqc_opc_lldp_start
ice_aqc_opc_lldp_filter_ctrl

This change follows the latest guidelines from HW team. It is
better to retry the same AQ command several times, as the result
of EBUSY, instead of returning error to the caller right away.

Signed-off-by: Chinh T Cao <chinh.t.cao@intel.com>
Tested-by: Tony Brelinski <tonyx.brelinski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_common.c   | 81 ++++++++++++++++++-
 drivers/net/ethernet/intel/ice/ice_common.h   |  3 +
 drivers/net/ethernet/intel/ice/ice_controlq.c |  2 +-
 3 files changed, 84 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_common.c b/drivers/net/ethernet/intel/ice/ice_common.c
index 54df00ee912b..bde19e30394d 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.c
+++ b/drivers/net/ethernet/intel/ice/ice_common.c
@@ -1292,6 +1292,85 @@ const struct ice_ctx_ele ice_tlan_ctx_info[] = {
  */
 DEFINE_MUTEX(ice_global_cfg_lock_sw);
 
+/**
+ * ice_should_retry_sq_send_cmd
+ * @opcode: AQ opcode
+ *
+ * Decide if we should retry the send command routine for the ATQ, depending
+ * on the opcode.
+ */
+static bool ice_should_retry_sq_send_cmd(u16 opcode)
+{
+	switch (opcode) {
+	case ice_aqc_opc_get_link_topo:
+	case ice_aqc_opc_lldp_stop:
+	case ice_aqc_opc_lldp_start:
+	case ice_aqc_opc_lldp_filter_ctrl:
+		return true;
+	}
+
+	return false;
+}
+
+/**
+ * ice_sq_send_cmd_retry - send command to Control Queue (ATQ)
+ * @hw: pointer to the HW struct
+ * @cq: pointer to the specific Control queue
+ * @desc: prefilled descriptor describing the command
+ * @buf: buffer to use for indirect commands (or NULL for direct commands)
+ * @buf_size: size of buffer for indirect commands (or 0 for direct commands)
+ * @cd: pointer to command details structure
+ *
+ * Retry sending the FW Admin Queue command, multiple times, to the FW Admin
+ * Queue if the EBUSY AQ error is returned.
+ */
+static enum ice_status
+ice_sq_send_cmd_retry(struct ice_hw *hw, struct ice_ctl_q_info *cq,
+		      struct ice_aq_desc *desc, void *buf, u16 buf_size,
+		      struct ice_sq_cd *cd)
+{
+	struct ice_aq_desc desc_cpy;
+	enum ice_status status;
+	bool is_cmd_for_retry;
+	u8 *buf_cpy = NULL;
+	u8 idx = 0;
+	u16 opcode;
+
+	opcode = le16_to_cpu(desc->opcode);
+	is_cmd_for_retry = ice_should_retry_sq_send_cmd(opcode);
+	memset(&desc_cpy, 0, sizeof(desc_cpy));
+
+	if (is_cmd_for_retry) {
+		if (buf) {
+			buf_cpy = kzalloc(buf_size, GFP_KERNEL);
+			if (!buf_cpy)
+				return ICE_ERR_NO_MEMORY;
+		}
+
+		memcpy(&desc_cpy, desc, sizeof(desc_cpy));
+	}
+
+	do {
+		status = ice_sq_send_cmd(hw, cq, desc, buf, buf_size, cd);
+
+		if (!is_cmd_for_retry || !status ||
+		    hw->adminq.sq_last_status != ICE_AQ_RC_EBUSY)
+			break;
+
+		if (buf_cpy)
+			memcpy(buf, buf_cpy, buf_size);
+
+		memcpy(desc, &desc_cpy, sizeof(desc_cpy));
+
+		mdelay(ICE_SQ_SEND_DELAY_TIME_MS);
+
+	} while (++idx < ICE_SQ_SEND_MAX_EXECUTE);
+
+	kfree(buf_cpy);
+
+	return status;
+}
+
 /**
  * ice_aq_send_cmd - send FW Admin Queue command to FW Admin Queue
  * @hw: pointer to the HW struct
@@ -1333,7 +1412,7 @@ ice_aq_send_cmd(struct ice_hw *hw, struct ice_aq_desc *desc, void *buf,
 		break;
 	}
 
-	status = ice_sq_send_cmd(hw, &hw->adminq, desc, buf, buf_size, cd);
+	status = ice_sq_send_cmd_retry(hw, &hw->adminq, desc, buf, buf_size, cd);
 	if (lock_acquired)
 		mutex_unlock(&ice_global_cfg_lock_sw);
 
diff --git a/drivers/net/ethernet/intel/ice/ice_common.h b/drivers/net/ethernet/intel/ice/ice_common.h
index 81fd69cb1485..d14406b11c92 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.h
+++ b/drivers/net/ethernet/intel/ice/ice_common.h
@@ -11,6 +11,9 @@
 #include "ice_switch.h"
 #include <linux/avf/virtchnl.h>
 
+#define ICE_SQ_SEND_DELAY_TIME_MS	10
+#define ICE_SQ_SEND_MAX_EXECUTE		3
+
 enum ice_status ice_init_hw(struct ice_hw *hw);
 void ice_deinit_hw(struct ice_hw *hw);
 enum ice_status ice_check_reset(struct ice_hw *hw);
diff --git a/drivers/net/ethernet/intel/ice/ice_controlq.c b/drivers/net/ethernet/intel/ice/ice_controlq.c
index b2d8a5932b1d..0f207a42ea77 100644
--- a/drivers/net/ethernet/intel/ice/ice_controlq.c
+++ b/drivers/net/ethernet/intel/ice/ice_controlq.c
@@ -892,7 +892,7 @@ static bool ice_sq_done(struct ice_hw *hw, struct ice_ctl_q_info *cq)
  * ice_sq_send_cmd - send command to Control Queue (ATQ)
  * @hw: pointer to the HW struct
  * @cq: pointer to the specific Control queue
- * @desc: prefilled descriptor describing the command (non DMA mem)
+ * @desc: prefilled descriptor describing the command
  * @buf: buffer to use for indirect commands (or NULL for direct commands)
  * @buf_size: size of buffer for indirect commands (or 0 for direct commands)
  * @cd: pointer to command details structure
-- 
2.26.2

[PATCH net-next 02/15] ice: Modify recursive way of adding nodes

[Tony Nguyen]

From: Victor Raj <victor.raj@intel.com>

Remove the recursive way of adding the nodes to the layer in order
to reduce the stack usage. Instead the algorithm is modified to use
a while loop.

The previous code was scanning recursively the nodes horizontally.
The total stack consumption will be based on number of nodes present
on that layer. In some cases it can consume more stack.

Signed-off-by: Victor Raj <victor.raj@intel.com>
Tested-by: Tony Brelinski <tonyx.brelinski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_sched.c | 130 ++++++++++++---------
 1 file changed, 77 insertions(+), 53 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_sched.c b/drivers/net/ethernet/intel/ice/ice_sched.c
index f890337cc24a..97562051fe14 100644
--- a/drivers/net/ethernet/intel/ice/ice_sched.c
+++ b/drivers/net/ethernet/intel/ice/ice_sched.c
@@ -919,7 +919,7 @@ ice_sched_add_elems(struct ice_port_info *pi, struct ice_sched_node *tc_node,
 }
 
 /**
- * ice_sched_add_nodes_to_layer - Add nodes to a given layer
+ * ice_sched_add_nodes_to_hw_layer - Add nodes to HW layer
  * @pi: port information structure
  * @tc_node: pointer to TC node
  * @parent: pointer to parent node
@@ -928,82 +928,106 @@ ice_sched_add_elems(struct ice_port_info *pi, struct ice_sched_node *tc_node,
  * @first_node_teid: pointer to the first node TEID
  * @num_nodes_added: pointer to number of nodes added
  *
- * This function add nodes to a given layer.
+ * Add nodes into specific HW layer.
  */
 static enum ice_status
-ice_sched_add_nodes_to_layer(struct ice_port_info *pi,
-			     struct ice_sched_node *tc_node,
-			     struct ice_sched_node *parent, u8 layer,
-			     u16 num_nodes, u32 *first_node_teid,
-			     u16 *num_nodes_added)
+ice_sched_add_nodes_to_hw_layer(struct ice_port_info *pi,
+				struct ice_sched_node *tc_node,
+				struct ice_sched_node *parent, u8 layer,
+				u16 num_nodes, u32 *first_node_teid,
+				u16 *num_nodes_added)
 {
-	u32 *first_teid_ptr = first_node_teid;
-	u16 new_num_nodes, max_child_nodes;
-	enum ice_status status = 0;
-	struct ice_hw *hw = pi->hw;
-	u16 num_added = 0;
-	u32 temp;
+	u16 max_child_nodes;
 
 	*num_nodes_added = 0;
 
 	if (!num_nodes)
-		return status;
+		return 0;
 
-	if (!parent || layer < hw->sw_entry_point_layer)
+	if (!parent || layer < pi->hw->sw_entry_point_layer)
 		return ICE_ERR_PARAM;
 
 	/* max children per node per layer */
-	max_child_nodes = hw->max_children[parent->tx_sched_layer];
+	max_child_nodes = pi->hw->max_children[parent->tx_sched_layer];
 
-	/* current number of children + required nodes exceed max children ? */
+	/* current number of children + required nodes exceed max children */
 	if ((parent->num_children + num_nodes) > max_child_nodes) {
 		/* Fail if the parent is a TC node */
 		if (parent == tc_node)
 			return ICE_ERR_CFG;
+		return ICE_ERR_MAX_LIMIT;
+	}
+
+	return ice_sched_add_elems(pi, tc_node, parent, layer, num_nodes,
+				   num_nodes_added, first_node_teid);
+}
+
+/**
+ * ice_sched_add_nodes_to_layer - Add nodes to a given layer
+ * @pi: port information structure
+ * @tc_node: pointer to TC node
+ * @parent: pointer to parent node
+ * @layer: layer number to add nodes
+ * @num_nodes: number of nodes to be added
+ * @first_node_teid: pointer to the first node TEID
+ * @num_nodes_added: pointer to number of nodes added
+ *
+ * This function add nodes to a given layer.
+ */
+static enum ice_status
+ice_sched_add_nodes_to_layer(struct ice_port_info *pi,
+			     struct ice_sched_node *tc_node,
+			     struct ice_sched_node *parent, u8 layer,
+			     u16 num_nodes, u32 *first_node_teid,
+			     u16 *num_nodes_added)
+{
+	u32 *first_teid_ptr = first_node_teid;
+	u16 new_num_nodes = num_nodes;
+	enum ice_status status = 0;
 
+	*num_nodes_added = 0;
+	while (*num_nodes_added < num_nodes) {
+		u16 max_child_nodes, num_added = 0;
+		u32 temp;
+
+		status = ice_sched_add_nodes_to_hw_layer(pi, tc_node, parent,
+							 layer,	new_num_nodes,
+							 first_teid_ptr,
+							 &num_added);
+		if (!status)
+			*num_nodes_added += num_added;
+		/* added more nodes than requested ? */
+		if (*num_nodes_added > num_nodes) {
+			ice_debug(pi->hw, ICE_DBG_SCHED, "added extra nodes %d %d\n", num_nodes,
+				  *num_nodes_added);
+			status = ICE_ERR_CFG;
+			break;
+		}
+		/* break if all the nodes are added successfully */
+		if (!status && (*num_nodes_added == num_nodes))
+			break;
+		/* break if the error is not max limit */
+		if (status && status != ICE_ERR_MAX_LIMIT)
+			break;
+		/* Exceeded the max children */
+		max_child_nodes = pi->hw->max_children[parent->tx_sched_layer];
 		/* utilize all the spaces if the parent is not full */
 		if (parent->num_children < max_child_nodes) {
 			new_num_nodes = max_child_nodes - parent->num_children;
-			/* this recursion is intentional, and wouldn't
-			 * go more than 2 calls
+		} else {
+			/* This parent is full, try the next sibling */
+			parent = parent->sibling;
+			/* Don't modify the first node TEID memory if the
+			 * first node was added already in the above call.
+			 * Instead send some temp memory for all other
+			 * recursive calls.
 			 */
-			status = ice_sched_add_nodes_to_layer(pi, tc_node,
-							      parent, layer,
-							      new_num_nodes,
-							      first_node_teid,
-							      &num_added);
-			if (status)
-				return status;
+			if (num_added)
+				first_teid_ptr = &temp;
 
-			*num_nodes_added += num_added;
+			new_num_nodes = num_nodes - *num_nodes_added;
 		}
-		/* Don't modify the first node TEID memory if the first node was
-		 * added already in the above call. Instead send some temp
-		 * memory for all other recursive calls.
-		 */
-		if (num_added)
-			first_teid_ptr = &temp;
-
-		new_num_nodes = num_nodes - num_added;
-
-		/* This parent is full, try the next sibling */
-		parent = parent->sibling;
-
-		/* this recursion is intentional, for 1024 queues
-		 * per VSI, it goes max of 16 iterations.
-		 * 1024 / 8 = 128 layer 8 nodes
-		 * 128 /8 = 16 (add 8 nodes per iteration)
-		 */
-		status = ice_sched_add_nodes_to_layer(pi, tc_node, parent,
-						      layer, new_num_nodes,
-						      first_teid_ptr,
-						      &num_added);
-		*num_nodes_added += num_added;
-		return status;
 	}
-
-	status = ice_sched_add_elems(pi, tc_node, parent, layer, num_nodes,
-				     num_nodes_added, first_node_teid);
 	return status;
 }
 
-- 
2.26.2

[PATCH net-next 03/15] ice: Align macro names to the specification

[Tony Nguyen]

From: Anirudh Venkataramanan <anirudh.venkataramanan@intel.com>

For get PHY abilities AQ, the specification defines "report modes"
as "with media", "without media" and "active configuration". For
clarity, rename macros to align with the specification.

Signed-off-by: Anirudh Venkataramanan <anirudh.venkataramanan@intel.com>
Tested-by: Tony Brelinski <tonyx.brelinski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_adminq_cmd.h | 10 +++++-----
 drivers/net/ethernet/intel/ice/ice_common.c     | 13 +++++++------
 drivers/net/ethernet/intel/ice/ice_ethtool.c    | 12 ++++++------
 drivers/net/ethernet/intel/ice/ice_main.c       | 12 ++++++------
 4 files changed, 24 insertions(+), 23 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h b/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
index b9491ef5f21c..3f3d51bf0019 100644
--- a/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
@@ -882,11 +882,11 @@ struct ice_aqc_get_phy_caps {
 	 * 01b - Report topology capabilities
 	 * 10b - Report SW configured
 	 */
-#define ICE_AQC_REPORT_MODE_S		1
-#define ICE_AQC_REPORT_MODE_M		(3 << ICE_AQC_REPORT_MODE_S)
-#define ICE_AQC_REPORT_NVM_CAP		0
-#define ICE_AQC_REPORT_TOPO_CAP		BIT(1)
-#define ICE_AQC_REPORT_SW_CFG		BIT(2)
+#define ICE_AQC_REPORT_MODE_S			1
+#define ICE_AQC_REPORT_MODE_M			(3 << ICE_AQC_REPORT_MODE_S)
+#define ICE_AQC_REPORT_TOPO_CAP_NO_MEDIA	0
+#define ICE_AQC_REPORT_TOPO_CAP_MEDIA		BIT(1)
+#define ICE_AQC_REPORT_ACTIVE_CFG		BIT(2)
 	__le32 reserved1;
 	__le32 addr_high;
 	__le32 addr_low;
diff --git a/drivers/net/ethernet/intel/ice/ice_common.c b/drivers/net/ethernet/intel/ice/ice_common.c
index bde19e30394d..509dce475bff 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.c
+++ b/drivers/net/ethernet/intel/ice/ice_common.c
@@ -191,7 +191,7 @@ ice_aq_get_phy_caps(struct ice_port_info *pi, bool qual_mods, u8 report_mode,
 	ice_debug(hw, ICE_DBG_LINK, "   module_type[2] = 0x%x\n",
 		  pcaps->module_type[2]);
 
-	if (!status && report_mode == ICE_AQC_REPORT_TOPO_CAP) {
+	if (!status && report_mode == ICE_AQC_REPORT_TOPO_CAP_MEDIA) {
 		pi->phy.phy_type_low = le64_to_cpu(pcaps->phy_type_low);
 		pi->phy.phy_type_high = le64_to_cpu(pcaps->phy_type_high);
 		memcpy(pi->phy.link_info.module_type, &pcaps->module_type,
@@ -922,7 +922,8 @@ enum ice_status ice_init_hw(struct ice_hw *hw)
 
 	/* Initialize port_info struct with PHY capabilities */
 	status = ice_aq_get_phy_caps(hw->port_info, false,
-				     ICE_AQC_REPORT_TOPO_CAP, pcaps, NULL);
+				     ICE_AQC_REPORT_TOPO_CAP_MEDIA, pcaps,
+				     NULL);
 	devm_kfree(ice_hw_to_dev(hw), pcaps);
 	if (status)
 		dev_warn(ice_hw_to_dev(hw), "Get PHY capabilities failed status = %d, continuing anyway\n",
@@ -2734,7 +2735,7 @@ enum ice_status ice_update_link_info(struct ice_port_info *pi)
 		if (!pcaps)
 			return ICE_ERR_NO_MEMORY;
 
-		status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_TOPO_CAP,
+		status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_TOPO_CAP_MEDIA,
 					     pcaps, NULL);
 
 		devm_kfree(ice_hw_to_dev(hw), pcaps);
@@ -2894,8 +2895,8 @@ ice_set_fc(struct ice_port_info *pi, u8 *aq_failures, bool ena_auto_link_update)
 		return ICE_ERR_NO_MEMORY;
 
 	/* Get the current PHY config */
-	status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_SW_CFG, pcaps,
-				     NULL);
+	status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_ACTIVE_CFG,
+				     pcaps, NULL);
 	if (status) {
 		*aq_failures = ICE_SET_FC_AQ_FAIL_GET;
 		goto out;
@@ -3041,7 +3042,7 @@ ice_cfg_phy_fec(struct ice_port_info *pi, struct ice_aqc_set_phy_cfg_data *cfg,
 	if (!pcaps)
 		return ICE_ERR_NO_MEMORY;
 
-	status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_TOPO_CAP, pcaps,
+	status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_TOPO_CAP_MEDIA, pcaps,
 				     NULL);
 	if (status)
 		goto out;
diff --git a/drivers/net/ethernet/intel/ice/ice_ethtool.c b/drivers/net/ethernet/intel/ice/ice_ethtool.c
index 15152e63f204..84e42e598c85 100644
--- a/drivers/net/ethernet/intel/ice/ice_ethtool.c
+++ b/drivers/net/ethernet/intel/ice/ice_ethtool.c
@@ -1060,7 +1060,7 @@ ice_get_fecparam(struct net_device *netdev, struct ethtool_fecparam *fecparam)
 	if (!caps)
 		return -ENOMEM;
 
-	status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_TOPO_CAP,
+	status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_TOPO_CAP_MEDIA,
 				     caps, NULL);
 	if (status) {
 		err = -EAGAIN;
@@ -2000,7 +2000,7 @@ ice_get_link_ksettings(struct net_device *netdev,
 		return -ENOMEM;
 
 	status = ice_aq_get_phy_caps(vsi->port_info, false,
-				     ICE_AQC_REPORT_SW_CFG, caps, NULL);
+				     ICE_AQC_REPORT_ACTIVE_CFG, caps, NULL);
 	if (status) {
 		err = -EIO;
 		goto done;
@@ -2037,7 +2037,7 @@ ice_get_link_ksettings(struct net_device *netdev,
 		ethtool_link_ksettings_add_link_mode(ks, advertising, FEC_RS);
 
 	status = ice_aq_get_phy_caps(vsi->port_info, false,
-				     ICE_AQC_REPORT_TOPO_CAP, caps, NULL);
+				     ICE_AQC_REPORT_TOPO_CAP_MEDIA, caps, NULL);
 	if (status) {
 		err = -EIO;
 		goto done;
@@ -2243,7 +2243,7 @@ ice_set_link_ksettings(struct net_device *netdev,
 		return -ENOMEM;
 
 	/* Get the PHY capabilities based on media */
-	status = ice_aq_get_phy_caps(p, false, ICE_AQC_REPORT_TOPO_CAP,
+	status = ice_aq_get_phy_caps(p, false, ICE_AQC_REPORT_TOPO_CAP_MEDIA,
 				     abilities, NULL);
 	if (status) {
 		err = -EAGAIN;
@@ -2972,7 +2972,7 @@ ice_get_pauseparam(struct net_device *netdev, struct ethtool_pauseparam *pause)
 		return;
 
 	/* Get current PHY config */
-	status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_SW_CFG, pcaps,
+	status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_ACTIVE_CFG, pcaps,
 				     NULL);
 	if (status)
 		goto out;
@@ -3039,7 +3039,7 @@ ice_set_pauseparam(struct net_device *netdev, struct ethtool_pauseparam *pause)
 		return -ENOMEM;
 
 	/* Get current PHY config */
-	status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_SW_CFG, pcaps,
+	status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_ACTIVE_CFG, pcaps,
 				     NULL);
 	if (status) {
 		kfree(pcaps);
diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index b3c1cadecf21..4379bbece4d9 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -720,7 +720,7 @@ void ice_print_link_msg(struct ice_vsi *vsi, bool isup)
 	}
 
 	status = ice_aq_get_phy_caps(vsi->port_info, false,
-				     ICE_AQC_REPORT_SW_CFG, caps, NULL);
+				     ICE_AQC_REPORT_ACTIVE_CFG, caps, NULL);
 	if (status)
 		netdev_info(vsi->netdev, "Get phy capability failed.\n");
 
@@ -1631,7 +1631,7 @@ static int ice_force_phys_link_state(struct ice_vsi *vsi, bool link_up)
 	if (!pcaps)
 		return -ENOMEM;
 
-	retcode = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_SW_CFG, pcaps,
+	retcode = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_ACTIVE_CFG, pcaps,
 				      NULL);
 	if (retcode) {
 		dev_err(dev, "Failed to get phy capabilities, VSI %d error %d\n",
@@ -1691,7 +1691,7 @@ static int ice_init_nvm_phy_type(struct ice_port_info *pi)
 	if (!pcaps)
 		return -ENOMEM;
 
-	status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_NVM_CAP, pcaps,
+	status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_TOPO_CAP_NO_MEDIA, pcaps,
 				     NULL);
 
 	if (status) {
@@ -1807,7 +1807,7 @@ static int ice_init_phy_user_cfg(struct ice_port_info *pi)
 	if (!pcaps)
 		return -ENOMEM;
 
-	status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_TOPO_CAP, pcaps,
+	status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_TOPO_CAP_MEDIA, pcaps,
 				     NULL);
 	if (status) {
 		dev_err(ice_pf_to_dev(pf), "Get PHY capability failed.\n");
@@ -1886,7 +1886,7 @@ static int ice_configure_phy(struct ice_vsi *vsi)
 		return -ENOMEM;
 
 	/* Get current PHY config */
-	status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_SW_CFG, pcaps,
+	status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_ACTIVE_CFG, pcaps,
 				     NULL);
 	if (status) {
 		dev_err(dev, "Failed to get PHY configuration, VSI %d error %s\n",
@@ -1904,7 +1904,7 @@ static int ice_configure_phy(struct ice_vsi *vsi)
 
 	/* Use PHY topology as baseline for configuration */
 	memset(pcaps, 0, sizeof(*pcaps));
-	status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_TOPO_CAP, pcaps,
+	status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_TOPO_CAP_MEDIA, pcaps,
 				     NULL);
 	if (status) {
 		dev_err(dev, "Failed to get PHY topology, VSI %d error %s\n",
-- 
2.26.2

[PATCH net-next 04/15] ice: Ignore EMODE return for opcode 0x0605

[Tony Nguyen]

From: Anirudh Venkataramanan <anirudh.venkataramanan@intel.com>

When link is owned by manageability, the driver is not allowed to fiddle
with link. FW returns ICE_AQ_RC_EMODE if the driver attempts to do so.
This patch adds a new function ice_set_link which abstracts the call to
ice_aq_set_link_restart_an and provides a clean way to turn on/off link.

While making this change, I also spotted that an int variable was being
used to hold both an ice_status return code and the Linux errno return
code. This pattern more often than not results in the driver inadvertently
returning ice_status back to kernel which is a major boo-boo. Clean it up.

Signed-off-by: Anirudh Venkataramanan <anirudh.venkataramanan@intel.com>
Tested-by: Tony Brelinski <tonyx.brelinski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_ethtool.c | 17 ++------
 drivers/net/ethernet/intel/ice/ice_lib.c     | 37 ++++++++++++++++++
 drivers/net/ethernet/intel/ice/ice_lib.h     |  2 +
 drivers/net/ethernet/intel/ice/ice_main.c    | 41 ++++++++------------
 4 files changed, 59 insertions(+), 38 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_ethtool.c b/drivers/net/ethernet/intel/ice/ice_ethtool.c
index 84e42e598c85..0db31a89658a 100644
--- a/drivers/net/ethernet/intel/ice/ice_ethtool.c
+++ b/drivers/net/ethernet/intel/ice/ice_ethtool.c
@@ -1095,24 +1095,15 @@ static int ice_nway_reset(struct net_device *netdev)
 {
 	struct ice_netdev_priv *np = netdev_priv(netdev);
 	struct ice_vsi *vsi = np->vsi;
-	struct ice_port_info *pi;
-	enum ice_status status;
+	int err;
 
-	pi = vsi->port_info;
 	/* If VSI state is up, then restart autoneg with link up */
 	if (!test_bit(__ICE_DOWN, vsi->back->state))
-		status = ice_aq_set_link_restart_an(pi, true, NULL);
+		err = ice_set_link(vsi, true);
 	else
-		status = ice_aq_set_link_restart_an(pi, false, NULL);
+		err = ice_set_link(vsi, false);
 
-	if (status) {
-		netdev_info(netdev, "link restart failed, err %s aq_err %s\n",
-			    ice_stat_str(status),
-			    ice_aq_str(pi->hw->adminq.sq_last_status));
-		return -EIO;
-	}
-
-	return 0;
+	return err;
 }
 
 /**
diff --git a/drivers/net/ethernet/intel/ice/ice_lib.c b/drivers/net/ethernet/intel/ice/ice_lib.c
index 6041ca2830de..5edc0da8b8c3 100644
--- a/drivers/net/ethernet/intel/ice/ice_lib.c
+++ b/drivers/net/ethernet/intel/ice/ice_lib.c
@@ -3423,3 +3423,40 @@ int ice_clear_dflt_vsi(struct ice_sw *sw)
 
 	return 0;
 }
+
+/**
+ * ice_set_link - turn on/off physical link
+ * @vsi: VSI to modify physical link on
+ * @ena: turn on/off physical link
+ */
+int ice_set_link(struct ice_vsi *vsi, bool ena)
+{
+	struct device *dev = ice_pf_to_dev(vsi->back);
+	struct ice_port_info *pi = vsi->port_info;
+	struct ice_hw *hw = pi->hw;
+	enum ice_status status;
+
+	if (vsi->type != ICE_VSI_PF)
+		return -EINVAL;
+
+	status = ice_aq_set_link_restart_an(pi, ena, NULL);
+
+	/* if link is owned by manageability, FW will return ICE_AQ_RC_EMODE.
+	 * this is not a fatal error, so print a warning message and return
+	 * a success code. Return an error if FW returns an error code other
+	 * than ICE_AQ_RC_EMODE
+	 */
+	if (status == ICE_ERR_AQ_ERROR) {
+		if (hw->adminq.sq_last_status == ICE_AQ_RC_EMODE)
+			dev_warn(dev, "can't set link to %s, err %s aq_err %s. not fatal, continuing\n",
+				 (ena ? "ON" : "OFF"), ice_stat_str(status),
+				 ice_aq_str(hw->adminq.sq_last_status));
+	} else if (status) {
+		dev_err(dev, "can't set link to %s, err %s aq_err %s\n",
+			(ena ? "ON" : "OFF"), ice_stat_str(status),
+			ice_aq_str(hw->adminq.sq_last_status));
+		return -EIO;
+	}
+
+	return 0;
+}
diff --git a/drivers/net/ethernet/intel/ice/ice_lib.h b/drivers/net/ethernet/intel/ice/ice_lib.h
index 3da17895a2b1..462c3ab7abad 100644
--- a/drivers/net/ethernet/intel/ice/ice_lib.h
+++ b/drivers/net/ethernet/intel/ice/ice_lib.h
@@ -45,6 +45,8 @@ int ice_cfg_vlan_pruning(struct ice_vsi *vsi, bool ena, bool vlan_promisc);
 
 void ice_cfg_sw_lldp(struct ice_vsi *vsi, bool tx, bool create);
 
+int ice_set_link(struct ice_vsi *vsi, bool ena);
+
 #ifdef CONFIG_DCB
 int ice_vsi_cfg_tc(struct ice_vsi *vsi, u8 ena_tc);
 #endif /* CONFIG_DCB */
diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index 4379bbece4d9..b976de4b5e6f 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -873,10 +873,10 @@ ice_link_event(struct ice_pf *pf, struct ice_port_info *pi, bool link_up,
 {
 	struct device *dev = ice_pf_to_dev(pf);
 	struct ice_phy_info *phy_info;
+	enum ice_status status;
 	struct ice_vsi *vsi;
 	u16 old_link_speed;
 	bool old_link;
-	int result;
 
 	phy_info = &pi->phy;
 	phy_info->link_info_old = phy_info->link_info;
@@ -887,10 +887,11 @@ ice_link_event(struct ice_pf *pf, struct ice_port_info *pi, bool link_up,
 	/* update the link info structures and re-enable link events,
 	 * don't bail on failure due to other book keeping needed
 	 */
-	result = ice_update_link_info(pi);
-	if (result)
-		dev_dbg(dev, "Failed to update link status and re-enable link events for port %d\n",
-			pi->lport);
+	status = ice_update_link_info(pi);
+	if (status)
+		dev_dbg(dev, "Failed to update link status on port %d, err %s aq_err %s\n",
+			pi->lport, ice_stat_str(status),
+			ice_aq_str(pi->hw->adminq.sq_last_status));
 
 	/* Check if the link state is up after updating link info, and treat
 	 * this event as an UP event since the link is actually UP now.
@@ -906,18 +907,12 @@ ice_link_event(struct ice_pf *pf, struct ice_port_info *pi, bool link_up,
 	if (!test_bit(ICE_FLAG_NO_MEDIA, pf->flags) &&
 	    !(pi->phy.link_info.link_info & ICE_AQ_MEDIA_AVAILABLE)) {
 		set_bit(ICE_FLAG_NO_MEDIA, pf->flags);
-
-		result = ice_aq_set_link_restart_an(pi, false, NULL);
-		if (result) {
-			dev_dbg(dev, "Failed to set link down, VSI %d error %d\n",
-				vsi->vsi_num, result);
-			return result;
-		}
+		ice_set_link(vsi, false);
 	}
 
 	/* if the old link up/down and speed is the same as the new */
 	if (link_up == old_link && link_speed == old_link_speed)
-		return result;
+		return 0;
 
 	if (ice_is_dcb_active(pf)) {
 		if (test_bit(ICE_FLAG_DCB_ENA, pf->flags))
@@ -931,7 +926,7 @@ ice_link_event(struct ice_pf *pf, struct ice_port_info *pi, bool link_up,
 
 	ice_vc_notify_link_state(pf);
 
-	return result;
+	return 0;
 }
 
 /**
@@ -6678,6 +6673,7 @@ int ice_open(struct net_device *netdev)
 	struct ice_vsi *vsi = np->vsi;
 	struct ice_pf *pf = vsi->back;
 	struct ice_port_info *pi;
+	enum ice_status status;
 	int err;
 
 	if (test_bit(__ICE_NEEDS_RESTART, pf->state)) {
@@ -6688,11 +6684,11 @@ int ice_open(struct net_device *netdev)
 	netif_carrier_off(netdev);
 
 	pi = vsi->port_info;
-	err = ice_update_link_info(pi);
-	if (err) {
-		netdev_err(netdev, "Failed to get link info, error %d\n",
-			   err);
-		return err;
+	status = ice_update_link_info(pi);
+	if (status) {
+		netdev_err(netdev, "Failed to get link info, error %s\n",
+			   ice_stat_str(status));
+		return -EIO;
 	}
 
 	/* Set PHY if there is media, otherwise, turn off PHY */
@@ -6715,12 +6711,7 @@ int ice_open(struct net_device *netdev)
 		}
 	} else {
 		set_bit(ICE_FLAG_NO_MEDIA, pf->flags);
-		err = ice_aq_set_link_restart_an(pi, false, NULL);
-		if (err) {
-			netdev_err(netdev, "Failed to set PHY state, VSI %d error %d\n",
-				   vsi->vsi_num, err);
-			return err;
-		}
+		ice_set_link(vsi, false);
 	}
 
 	err = ice_vsi_open(vsi);
-- 
2.26.2

[PATCH net-next 05/15] ice: Remove unnecessary checker loop

[Tony Nguyen]

From: Anirudh Venkataramanan <anirudh.venkataramanan@intel.com>

The loop checking for PF VSI doesn't make any sense. The VSI type
backing the netdev passed to ice_set_link_ksettings will always be
of type ICE_PF_VSI. Remove it.

Signed-off-by: Anirudh Venkataramanan <anirudh.venkataramanan@intel.com>
Tested-by: Tony Brelinski <tonyx.brelinski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_ethtool.c | 10 +---------
 1 file changed, 1 insertion(+), 9 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_ethtool.c b/drivers/net/ethernet/intel/ice/ice_ethtool.c
index 0db31a89658a..c8a8fdac4977 100644
--- a/drivers/net/ethernet/intel/ice/ice_ethtool.c
+++ b/drivers/net/ethernet/intel/ice/ice_ethtool.c
@@ -2198,8 +2198,8 @@ ice_set_link_ksettings(struct net_device *netdev,
 	struct ethtool_link_ksettings safe_ks, copy_ks;
 	struct ice_aqc_get_phy_caps_data *abilities;
 	u8 autoneg, timeout = TEST_SET_BITS_TIMEOUT;
-	u16 adv_link_speed, curr_link_speed, idx;
 	struct ice_aqc_set_phy_cfg_data config;
+	u16 adv_link_speed, curr_link_speed;
 	struct ice_pf *pf = np->vsi->back;
 	struct ice_port_info *p;
 	u8 autoneg_changed = 0;
@@ -2214,14 +2214,6 @@ ice_set_link_ksettings(struct net_device *netdev,
 	if (!p)
 		return -EOPNOTSUPP;
 
-	/* Check if this is LAN VSI */
-	ice_for_each_vsi(pf, idx)
-		if (pf->vsi[idx]->type == ICE_VSI_PF) {
-			if (np->vsi != pf->vsi[idx])
-				return -EOPNOTSUPP;
-			break;
-		}
-
 	if (p->phy.media_type != ICE_MEDIA_BASET &&
 	    p->phy.media_type != ICE_MEDIA_FIBER &&
 	    p->phy.media_type != ICE_MEDIA_BACKPLANE &&
-- 
2.26.2

[PATCH net-next 06/15] ice: Rename a couple of variables

[Tony Nguyen]

From: Anirudh Venkataramanan <anirudh.venkataramanan@intel.com>

In ice_set_link_ksettings, change 'abilities' to 'phy_caps' and 'p' to
'pi'. This is more consistent with similar usages elsewhere in the
driver.

Signed-off-by: Anirudh Venkataramanan <anirudh.venkataramanan@intel.com>
Tested-by: Tony Brelinski <tonyx.brelinski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_ethtool.c | 50 ++++++++++----------
 1 file changed, 25 insertions(+), 25 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_ethtool.c b/drivers/net/ethernet/intel/ice/ice_ethtool.c
index c8a8fdac4977..a31b6cf8d599 100644
--- a/drivers/net/ethernet/intel/ice/ice_ethtool.c
+++ b/drivers/net/ethernet/intel/ice/ice_ethtool.c
@@ -2196,12 +2196,12 @@ ice_set_link_ksettings(struct net_device *netdev,
 {
 	struct ice_netdev_priv *np = netdev_priv(netdev);
 	struct ethtool_link_ksettings safe_ks, copy_ks;
-	struct ice_aqc_get_phy_caps_data *abilities;
 	u8 autoneg, timeout = TEST_SET_BITS_TIMEOUT;
+	struct ice_aqc_get_phy_caps_data *phy_caps;
 	struct ice_aqc_set_phy_cfg_data config;
 	u16 adv_link_speed, curr_link_speed;
 	struct ice_pf *pf = np->vsi->back;
-	struct ice_port_info *p;
+	struct ice_port_info *pi;
 	u8 autoneg_changed = 0;
 	enum ice_status status;
 	u64 phy_type_high = 0;
@@ -2209,25 +2209,25 @@ ice_set_link_ksettings(struct net_device *netdev,
 	int err = 0;
 	bool linkup;
 
-	p = np->vsi->port_info;
+	pi = np->vsi->port_info;
 
-	if (!p)
+	if (!pi)
 		return -EOPNOTSUPP;
 
-	if (p->phy.media_type != ICE_MEDIA_BASET &&
-	    p->phy.media_type != ICE_MEDIA_FIBER &&
-	    p->phy.media_type != ICE_MEDIA_BACKPLANE &&
-	    p->phy.media_type != ICE_MEDIA_DA &&
-	    p->phy.link_info.link_info & ICE_AQ_LINK_UP)
+	if (pi->phy.media_type != ICE_MEDIA_BASET &&
+	    pi->phy.media_type != ICE_MEDIA_FIBER &&
+	    pi->phy.media_type != ICE_MEDIA_BACKPLANE &&
+	    pi->phy.media_type != ICE_MEDIA_DA &&
+	    pi->phy.link_info.link_info & ICE_AQ_LINK_UP)
 		return -EOPNOTSUPP;
 
-	abilities = kzalloc(sizeof(*abilities), GFP_KERNEL);
-	if (!abilities)
+	phy_caps = kzalloc(sizeof(*phy_caps), GFP_KERNEL);
+	if (!phy_caps)
 		return -ENOMEM;
 
 	/* Get the PHY capabilities based on media */
-	status = ice_aq_get_phy_caps(p, false, ICE_AQC_REPORT_TOPO_CAP_MEDIA,
-				     abilities, NULL);
+	status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_TOPO_CAP_MEDIA,
+				     phy_caps, NULL);
 	if (status) {
 		err = -EAGAIN;
 		goto done;
@@ -2289,26 +2289,26 @@ ice_set_link_ksettings(struct net_device *netdev,
 	 * configuration is initialized during probe from PHY capabilities
 	 * software mode, and updated on set PHY configuration.
 	 */
-	memcpy(&config, &p->phy.curr_user_phy_cfg, sizeof(config));
+	memcpy(&config, &pi->phy.curr_user_phy_cfg, sizeof(config));
 
 	config.caps |= ICE_AQ_PHY_ENA_AUTO_LINK_UPDT;
 
 	/* Check autoneg */
-	err = ice_setup_autoneg(p, &safe_ks, &config, autoneg, &autoneg_changed,
+	err = ice_setup_autoneg(pi, &safe_ks, &config, autoneg, &autoneg_changed,
 				netdev);
 
 	if (err)
 		goto done;
 
 	/* Call to get the current link speed */
-	p->phy.get_link_info = true;
-	status = ice_get_link_status(p, &linkup);
+	pi->phy.get_link_info = true;
+	status = ice_get_link_status(pi, &linkup);
 	if (status) {
 		err = -EAGAIN;
 		goto done;
 	}
 
-	curr_link_speed = p->phy.link_info.link_speed;
+	curr_link_speed = pi->phy.link_info.link_speed;
 	adv_link_speed = ice_ksettings_find_adv_link_speed(ks);
 
 	/* If speed didn't get set, set it to what it currently is.
@@ -2327,7 +2327,7 @@ ice_set_link_ksettings(struct net_device *netdev,
 	}
 
 	/* save the requested speeds */
-	p->phy.link_info.req_speeds = adv_link_speed;
+	pi->phy.link_info.req_speeds = adv_link_speed;
 
 	/* set link and auto negotiation so changes take effect */
 	config.caps |= ICE_AQ_PHY_ENA_LINK;
@@ -2343,9 +2343,9 @@ ice_set_link_ksettings(struct net_device *netdev,
 	 * for set PHY configuration
 	 */
 	config.phy_type_high = cpu_to_le64(phy_type_high) &
-			abilities->phy_type_high;
+			phy_caps->phy_type_high;
 	config.phy_type_low = cpu_to_le64(phy_type_low) &
-			abilities->phy_type_low;
+			phy_caps->phy_type_low;
 
 	if (!(config.phy_type_high || config.phy_type_low)) {
 		/* If there is no intersection and lenient mode is enabled, then
@@ -2365,7 +2365,7 @@ ice_set_link_ksettings(struct net_device *netdev,
 	}
 
 	/* If link is up put link down */
-	if (p->phy.link_info.link_info & ICE_AQ_LINK_UP) {
+	if (pi->phy.link_info.link_info & ICE_AQ_LINK_UP) {
 		/* Tell the OS link is going down, the link will go
 		 * back up when fw says it is ready asynchronously
 		 */
@@ -2375,7 +2375,7 @@ ice_set_link_ksettings(struct net_device *netdev,
 	}
 
 	/* make the aq call */
-	status = ice_aq_set_phy_cfg(&pf->hw, p, &config, NULL);
+	status = ice_aq_set_phy_cfg(&pf->hw, pi, &config, NULL);
 	if (status) {
 		netdev_info(netdev, "Set phy config failed,\n");
 		err = -EAGAIN;
@@ -2383,9 +2383,9 @@ ice_set_link_ksettings(struct net_device *netdev,
 	}
 
 	/* Save speed request */
-	p->phy.curr_user_speed_req = adv_link_speed;
+	pi->phy.curr_user_speed_req = adv_link_speed;
 done:
-	kfree(abilities);
+	kfree(phy_caps);
 	clear_bit(__ICE_CFG_BUSY, pf->state);
 
 	return err;
-- 
2.26.2

[PATCH net-next 07/15] ice: Fix error return codes in ice_set_link_ksettings

[Tony Nguyen]

From: Anirudh Venkataramanan <anirudh.venkataramanan@intel.com>

Return more appropriate error codes so that the right error
message is communicated to the user by ethtool.

Signed-off-by: Anirudh Venkataramanan <anirudh.venkataramanan@intel.com>
Tested-by: Tony Brelinski <tonyx.brelinski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_ethtool.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_ethtool.c b/drivers/net/ethernet/intel/ice/ice_ethtool.c
index a31b6cf8d599..41795d263dcd 100644
--- a/drivers/net/ethernet/intel/ice/ice_ethtool.c
+++ b/drivers/net/ethernet/intel/ice/ice_ethtool.c
@@ -2212,7 +2212,7 @@ ice_set_link_ksettings(struct net_device *netdev,
 	pi = np->vsi->port_info;
 
 	if (!pi)
-		return -EOPNOTSUPP;
+		return -EIO;
 
 	if (pi->phy.media_type != ICE_MEDIA_BASET &&
 	    pi->phy.media_type != ICE_MEDIA_FIBER &&
@@ -2229,7 +2229,7 @@ ice_set_link_ksettings(struct net_device *netdev,
 	status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_TOPO_CAP_MEDIA,
 				     phy_caps, NULL);
 	if (status) {
-		err = -EAGAIN;
+		err = -EIO;
 		goto done;
 	}
 
@@ -2252,7 +2252,7 @@ ice_set_link_ksettings(struct net_device *netdev,
 			   __ETHTOOL_LINK_MODE_MASK_NBITS)) {
 		if (!test_bit(ICE_FLAG_LINK_LENIENT_MODE_ENA, pf->flags))
 			netdev_info(netdev, "The selected speed is not supported by the current media. Please select a link speed that is supported by the current media.\n");
-		err = -EINVAL;
+		err = -EOPNOTSUPP;
 		goto done;
 	}
 
@@ -2304,7 +2304,7 @@ ice_set_link_ksettings(struct net_device *netdev,
 	pi->phy.get_link_info = true;
 	status = ice_get_link_status(pi, &linkup);
 	if (status) {
-		err = -EAGAIN;
+		err = -EIO;
 		goto done;
 	}
 
@@ -2335,7 +2335,7 @@ ice_set_link_ksettings(struct net_device *netdev,
 	/* check if there is a PHY type for the requested advertised speed */
 	if (!(phy_type_low || phy_type_high)) {
 		netdev_info(netdev, "The selected speed is not supported by the current media. Please select a link speed that is supported by the current media.\n");
-		err = -EAGAIN;
+		err = -EOPNOTSUPP;
 		goto done;
 	}
 
@@ -2359,7 +2359,7 @@ ice_set_link_ksettings(struct net_device *netdev,
 					      pf->nvm_phy_type_lo;
 		} else {
 			netdev_info(netdev, "The selected speed is not supported by the current media. Please select a link speed that is supported by the current media.\n");
-			err = -EAGAIN;
+			err = -EOPNOTSUPP;
 			goto done;
 		}
 	}
@@ -2378,7 +2378,7 @@ ice_set_link_ksettings(struct net_device *netdev,
 	status = ice_aq_set_phy_cfg(&pf->hw, pi, &config, NULL);
 	if (status) {
 		netdev_info(netdev, "Set phy config failed,\n");
-		err = -EAGAIN;
+		err = -EIO;
 		goto done;
 	}
 
-- 
2.26.2

[PATCH net-next 08/15] ice: Replace some memsets and memcpys with assignment

[Tony Nguyen]

From: Anirudh Venkataramanan <anirudh.venkataramanan@intel.com>

In ice_set_link_ksettings, use assignment instead of memset/memcpy
where possible

Signed-off-by: Anirudh Venkataramanan <anirudh.venkataramanan@intel.com>
Tested-by: Tony Brelinski <tonyx.brelinski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_ethtool.c | 10 +++-------
 1 file changed, 3 insertions(+), 7 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_ethtool.c b/drivers/net/ethernet/intel/ice/ice_ethtool.c
index 41795d263dcd..5dd84ccf6756 100644
--- a/drivers/net/ethernet/intel/ice/ice_ethtool.c
+++ b/drivers/net/ethernet/intel/ice/ice_ethtool.c
@@ -2195,8 +2195,9 @@ ice_set_link_ksettings(struct net_device *netdev,
 		       const struct ethtool_link_ksettings *ks)
 {
 	struct ice_netdev_priv *np = netdev_priv(netdev);
-	struct ethtool_link_ksettings safe_ks, copy_ks;
 	u8 autoneg, timeout = TEST_SET_BITS_TIMEOUT;
+	struct ethtool_link_ksettings copy_ks = *ks;
+	struct ethtool_link_ksettings safe_ks = {};
 	struct ice_aqc_get_phy_caps_data *phy_caps;
 	struct ice_aqc_set_phy_cfg_data config;
 	u16 adv_link_speed, curr_link_speed;
@@ -2233,14 +2234,9 @@ ice_set_link_ksettings(struct net_device *netdev,
 		goto done;
 	}
 
-	/* copy the ksettings to copy_ks to avoid modifying the original */
-	memcpy(&copy_ks, ks, sizeof(copy_ks));
-
 	/* save autoneg out of ksettings */
 	autoneg = copy_ks.base.autoneg;
 
-	memset(&safe_ks, 0, sizeof(safe_ks));
-
 	/* Get link modes supported by hardware.*/
 	ice_phy_type_to_ethtool(netdev, &safe_ks);
 
@@ -2289,7 +2285,7 @@ ice_set_link_ksettings(struct net_device *netdev,
 	 * configuration is initialized during probe from PHY capabilities
 	 * software mode, and updated on set PHY configuration.
 	 */
-	memcpy(&config, &pi->phy.curr_user_phy_cfg, sizeof(config));
+	config = pi->phy.curr_user_phy_cfg;
 
 	config.caps |= ICE_AQ_PHY_ENA_AUTO_LINK_UPDT;
 
-- 
2.26.2

[PATCH net-next 09/15] ice: Use default configuration mode for PHY configuration

[Tony Nguyen]

From: Anirudh Venkataramanan <anirudh.venkataramanan@intel.com>

Recent firmware supports a new "get PHY capabilities" mode
ICE_AQC_REPORT_DFLT_CFG which makes it unnecessary for the driver
to track and apply NVM based default link overrides.

If FW AQ API version supports it, use Report Default Configuration.
Add check function for Report Default Configuration support and update
accordingly.

Also change adv_phy_type_[lo|hi] to advert_phy_type[lo|hi] for
clarity.

Co-developed-by: Mateusz Pacuszka <mateuszx.pacuszka@intel.com>
Signed-off-by: Mateusz Pacuszka <mateuszx.pacuszka@intel.com>
Signed-off-by: Anirudh Venkataramanan <anirudh.venkataramanan@intel.com>
Tested-by: Tony Brelinski <tonyx.brelinski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 .../net/ethernet/intel/ice/ice_adminq_cmd.h   | 12 ++--
 drivers/net/ethernet/intel/ice/ice_common.c   | 33 ++++++++++-
 drivers/net/ethernet/intel/ice/ice_common.h   |  1 +
 drivers/net/ethernet/intel/ice/ice_ethtool.c  | 59 +++++++++++--------
 drivers/net/ethernet/intel/ice/ice_main.c     | 35 +++++++----
 drivers/net/ethernet/intel/ice/ice_type.h     |  5 ++
 6 files changed, 103 insertions(+), 42 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h b/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
index 3f3d51bf0019..5cdfe406af84 100644
--- a/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
@@ -877,16 +877,18 @@ struct ice_aqc_get_phy_caps {
 	__le16 param0;
 	/* 18.0 - Report qualified modules */
 #define ICE_AQC_GET_PHY_RQM		BIT(0)
-	/* 18.1 - 18.2 : Report mode
-	 * 00b - Report NVM capabilities
-	 * 01b - Report topology capabilities
-	 * 10b - Report SW configured
+	/* 18.1 - 18.3 : Report mode
+	 * 000b - Report NVM capabilities
+	 * 001b - Report topology capabilities
+	 * 010b - Report SW configured
+	 * 100b - Report default capabilities
 	 */
 #define ICE_AQC_REPORT_MODE_S			1
-#define ICE_AQC_REPORT_MODE_M			(3 << ICE_AQC_REPORT_MODE_S)
+#define ICE_AQC_REPORT_MODE_M			(7 << ICE_AQC_REPORT_MODE_S)
 #define ICE_AQC_REPORT_TOPO_CAP_NO_MEDIA	0
 #define ICE_AQC_REPORT_TOPO_CAP_MEDIA		BIT(1)
 #define ICE_AQC_REPORT_ACTIVE_CFG		BIT(2)
+#define ICE_AQC_REPORT_DFLT_CFG		BIT(3)
 	__le32 reserved1;
 	__le32 addr_high;
 	__le32 addr_low;
diff --git a/drivers/net/ethernet/intel/ice/ice_common.c b/drivers/net/ethernet/intel/ice/ice_common.c
index 509dce475bff..8485450aff80 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.c
+++ b/drivers/net/ethernet/intel/ice/ice_common.c
@@ -158,6 +158,10 @@ ice_aq_get_phy_caps(struct ice_port_info *pi, bool qual_mods, u8 report_mode,
 		return ICE_ERR_PARAM;
 	hw = pi->hw;
 
+	if (report_mode == ICE_AQC_REPORT_DFLT_CFG &&
+	    !ice_fw_supports_report_dflt_cfg(hw))
+		return ICE_ERR_PARAM;
+
 	ice_fill_dflt_direct_cmd_desc(&desc, ice_aqc_opc_get_phy_caps);
 
 	if (qual_mods)
@@ -3034,16 +3038,21 @@ ice_cfg_phy_fec(struct ice_port_info *pi, struct ice_aqc_set_phy_cfg_data *cfg,
 {
 	struct ice_aqc_get_phy_caps_data *pcaps;
 	enum ice_status status;
+	struct ice_hw *hw;
 
 	if (!pi || !cfg)
 		return ICE_ERR_BAD_PTR;
 
+	hw = pi->hw;
+
 	pcaps = kzalloc(sizeof(*pcaps), GFP_KERNEL);
 	if (!pcaps)
 		return ICE_ERR_NO_MEMORY;
 
-	status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_TOPO_CAP_MEDIA, pcaps,
-				     NULL);
+	status = ice_aq_get_phy_caps(pi, false,
+				     (ice_fw_supports_report_dflt_cfg(hw) ?
+				      ICE_AQC_REPORT_DFLT_CFG :
+				      ICE_AQC_REPORT_TOPO_CAP_MEDIA), pcaps, NULL);
 	if (status)
 		goto out;
 
@@ -4492,3 +4501,23 @@ ice_lldp_fltr_add_remove(struct ice_hw *hw, u16 vsi_num, bool add)
 
 	return ice_aq_send_cmd(hw, &desc, NULL, 0, NULL);
 }
+
+/**
+ * ice_fw_supports_report_dflt_cfg
+ * @hw: pointer to the hardware structure
+ *
+ * Checks if the firmware supports report default configuration
+ */
+bool ice_fw_supports_report_dflt_cfg(struct ice_hw *hw)
+{
+	if (hw->api_maj_ver == ICE_FW_API_REPORT_DFLT_CFG_MAJ) {
+		if (hw->api_min_ver > ICE_FW_API_REPORT_DFLT_CFG_MIN)
+			return true;
+		if (hw->api_min_ver == ICE_FW_API_REPORT_DFLT_CFG_MIN &&
+		    hw->api_patch >= ICE_FW_API_REPORT_DFLT_CFG_PATCH)
+			return true;
+	} else if (hw->api_maj_ver > ICE_FW_API_REPORT_DFLT_CFG_MAJ) {
+		return true;
+	}
+	return false;
+}
diff --git a/drivers/net/ethernet/intel/ice/ice_common.h b/drivers/net/ethernet/intel/ice/ice_common.h
index d14406b11c92..7a9d2dfb21a2 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.h
+++ b/drivers/net/ethernet/intel/ice/ice_common.h
@@ -179,4 +179,5 @@ ice_aq_set_lldp_mib(struct ice_hw *hw, u8 mib_type, void *buf, u16 buf_size,
 bool ice_fw_supports_lldp_fltr_ctrl(struct ice_hw *hw);
 enum ice_status
 ice_lldp_fltr_add_remove(struct ice_hw *hw, u16 vsi_num, bool add);
+bool ice_fw_supports_report_dflt_cfg(struct ice_hw *hw);
 #endif /* _ICE_COMMON_H_ */
diff --git a/drivers/net/ethernet/intel/ice/ice_ethtool.c b/drivers/net/ethernet/intel/ice/ice_ethtool.c
index 5dd84ccf6756..4ca8d5880cfc 100644
--- a/drivers/net/ethernet/intel/ice/ice_ethtool.c
+++ b/drivers/net/ethernet/intel/ice/ice_ethtool.c
@@ -1445,8 +1445,8 @@ void ice_mask_min_supported_speeds(u64 phy_types_high, u64 *phy_types_low)
 	do {								     \
 		if (req_speeds & (aq_link_speed) ||			     \
 		    (!req_speeds &&					     \
-		     (adv_phy_type_lo & phy_type_mask_lo ||		     \
-		      adv_phy_type_hi & phy_type_mask_hi)))		     \
+		     (advert_phy_type_lo & phy_type_mask_lo ||		     \
+		      advert_phy_type_hi & phy_type_mask_hi)))		     \
 			ethtool_link_ksettings_add_link_mode(ks, advertising,\
 							ethtool_link_mode);  \
 	} while (0)
@@ -1463,10 +1463,10 @@ ice_phy_type_to_ethtool(struct net_device *netdev,
 	struct ice_netdev_priv *np = netdev_priv(netdev);
 	struct ice_vsi *vsi = np->vsi;
 	struct ice_pf *pf = vsi->back;
+	u64 advert_phy_type_lo = 0;
+	u64 advert_phy_type_hi = 0;
 	u64 phy_type_mask_lo = 0;
 	u64 phy_type_mask_hi = 0;
-	u64 adv_phy_type_lo = 0;
-	u64 adv_phy_type_hi = 0;
 	u64 phy_types_high = 0;
 	u64 phy_types_low = 0;
 	u16 req_speeds;
@@ -1484,28 +1484,35 @@ ice_phy_type_to_ethtool(struct net_device *netdev,
 	 * requested by user.
 	 */
 	if (test_bit(ICE_FLAG_LINK_LENIENT_MODE_ENA, pf->flags)) {
-		struct ice_link_default_override_tlv *ldo;
-
-		ldo = &pf->link_dflt_override;
 		phy_types_low = le64_to_cpu(pf->nvm_phy_type_lo);
 		phy_types_high = le64_to_cpu(pf->nvm_phy_type_hi);
 
 		ice_mask_min_supported_speeds(phy_types_high, &phy_types_low);
-
-		/* If override enabled and PHY mask set, then
-		 * Advertising link mode is the intersection of the PHY
-		 * types without media and the override PHY mask.
+		/* determine advertised modes based on link override only
+		 * if it's supported and if the FW doesn't abstract the
+		 * driver from having to account for link overrides
 		 */
-		if (ldo->options & ICE_LINK_OVERRIDE_EN &&
-		    (ldo->phy_type_low || ldo->phy_type_high)) {
-			adv_phy_type_lo =
-				le64_to_cpu(pf->nvm_phy_type_lo) &
-				ldo->phy_type_low;
-			adv_phy_type_hi =
-				le64_to_cpu(pf->nvm_phy_type_hi) &
-				ldo->phy_type_high;
+		if (ice_fw_supports_link_override(&pf->hw) &&
+		    !ice_fw_supports_report_dflt_cfg(&pf->hw)) {
+			struct ice_link_default_override_tlv *ldo;
+
+			ldo = &pf->link_dflt_override;
+			/* If override enabled and PHY mask set, then
+			 * Advertising link mode is the intersection of the PHY
+			 * types without media and the override PHY mask.
+			 */
+			if (ldo->options & ICE_LINK_OVERRIDE_EN &&
+			    (ldo->phy_type_low || ldo->phy_type_high)) {
+				advert_phy_type_lo =
+					le64_to_cpu(pf->nvm_phy_type_lo) &
+					ldo->phy_type_low;
+				advert_phy_type_hi =
+					le64_to_cpu(pf->nvm_phy_type_hi) &
+					ldo->phy_type_high;
+			}
 		}
 	} else {
+		/* strict mode */
 		phy_types_low = vsi->port_info->phy.phy_type_low;
 		phy_types_high = vsi->port_info->phy.phy_type_high;
 	}
@@ -1513,9 +1520,9 @@ ice_phy_type_to_ethtool(struct net_device *netdev,
 	/* If Advertising link mode PHY type is not using override PHY type,
 	 * then use PHY type with media.
 	 */
-	if (!adv_phy_type_lo && !adv_phy_type_hi) {
-		adv_phy_type_lo = vsi->port_info->phy.phy_type_low;
-		adv_phy_type_hi = vsi->port_info->phy.phy_type_high;
+	if (!advert_phy_type_lo && !advert_phy_type_hi) {
+		advert_phy_type_lo = vsi->port_info->phy.phy_type_low;
+		advert_phy_type_hi = vsi->port_info->phy.phy_type_high;
 	}
 
 	ethtool_link_ksettings_zero_link_mode(ks, supported);
@@ -2227,8 +2234,12 @@ ice_set_link_ksettings(struct net_device *netdev,
 		return -ENOMEM;
 
 	/* Get the PHY capabilities based on media */
-	status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_TOPO_CAP_MEDIA,
-				     phy_caps, NULL);
+	if (ice_fw_supports_report_dflt_cfg(pi->hw))
+		status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_DFLT_CFG,
+					     phy_caps, NULL);
+	else
+		status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_TOPO_CAP_MEDIA,
+					     phy_caps, NULL);
 	if (status) {
 		err = -EIO;
 		goto done;
diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index b976de4b5e6f..6bc2215b674d 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -1732,7 +1732,7 @@ static void ice_init_link_dflt_override(struct ice_port_info *pi)
  * ice_init_phy_cfg_dflt_override - Initialize PHY cfg default override settings
  * @pi: port info structure
  *
- * If default override is enabled, initialized the user PHY cfg speed and FEC
+ * If default override is enabled, initialize the user PHY cfg speed and FEC
  * settings using the default override mask from the NVM.
  *
  * The PHY should only be configured with the default override settings the
@@ -1741,6 +1741,9 @@ static void ice_init_link_dflt_override(struct ice_port_info *pi)
  * and the PHY has not been configured with the default override settings. The
  * state is set here, and cleared in ice_configure_phy the first time the PHY is
  * configured.
+ *
+ * This function should be called only if the FW doesn't support default
+ * configuration mode, as reported by ice_fw_supports_report_dflt_cfg.
  */
 static void ice_init_phy_cfg_dflt_override(struct ice_port_info *pi)
 {
@@ -1802,8 +1805,12 @@ static int ice_init_phy_user_cfg(struct ice_port_info *pi)
 	if (!pcaps)
 		return -ENOMEM;
 
-	status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_TOPO_CAP_MEDIA, pcaps,
-				     NULL);
+	if (ice_fw_supports_report_dflt_cfg(pi->hw))
+		status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_DFLT_CFG,
+					     pcaps, NULL);
+	else
+		status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_TOPO_CAP_MEDIA,
+					     pcaps, NULL);
 	if (status) {
 		dev_err(ice_pf_to_dev(pf), "Get PHY capability failed.\n");
 		err = -EIO;
@@ -1818,17 +1825,19 @@ static int ice_init_phy_user_cfg(struct ice_port_info *pi)
 	      ICE_AQC_MOD_ENFORCE_STRICT_MODE)) {
 		set_bit(ICE_FLAG_LINK_LENIENT_MODE_ENA, pf->flags);
 
-		/* if link default override is enabled, initialize user PHY
-		 * configuration with link default override values
+		/* if the FW supports default PHY configuration mode, then the driver
+		 * does not have to apply link override settings. If not,
+		 * initialize user PHY configuration with link override values
 		 */
-		if (pf->link_dflt_override.options & ICE_LINK_OVERRIDE_EN) {
+		if (!ice_fw_supports_report_dflt_cfg(pi->hw) &&
+		    (pf->link_dflt_override.options & ICE_LINK_OVERRIDE_EN)) {
 			ice_init_phy_cfg_dflt_override(pi);
 			goto out;
 		}
 	}
 
-	/* if link default override is not enabled, initialize PHY using
-	 * topology with media
+	/* if link default override is not enabled, set user flow control and
+	 * FEC settings based on what get_phy_caps returned
 	 */
 	phy->curr_user_fec_req = ice_caps_to_fec_mode(pcaps->caps,
 						      pcaps->link_fec_options);
@@ -1899,10 +1908,14 @@ static int ice_configure_phy(struct ice_vsi *vsi)
 
 	/* Use PHY topology as baseline for configuration */
 	memset(pcaps, 0, sizeof(*pcaps));
-	status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_TOPO_CAP_MEDIA, pcaps,
-				     NULL);
+	if (ice_fw_supports_report_dflt_cfg(pi->hw))
+		status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_DFLT_CFG,
+					     pcaps, NULL);
+	else
+		status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_TOPO_CAP_MEDIA,
+					     pcaps, NULL);
 	if (status) {
-		dev_err(dev, "Failed to get PHY topology, VSI %d error %s\n",
+		dev_err(dev, "Failed to get PHY caps, VSI %d error %s\n",
 			vsi->vsi_num, ice_stat_str(status));
 		err = -EIO;
 		goto done;
diff --git a/drivers/net/ethernet/intel/ice/ice_type.h b/drivers/net/ethernet/intel/ice/ice_type.h
index 276ebcc309dc..2efc91b58c9e 100644
--- a/drivers/net/ethernet/intel/ice/ice_type.h
+++ b/drivers/net/ethernet/intel/ice/ice_type.h
@@ -941,4 +941,9 @@ struct ice_aq_get_set_rss_lut_params {
 #define ICE_FW_API_LLDP_FLTR_MIN	7
 #define ICE_FW_API_LLDP_FLTR_PATCH	1
 
+/* AQ API version for report default configuration */
+#define ICE_FW_API_REPORT_DFLT_CFG_MAJ		1
+#define ICE_FW_API_REPORT_DFLT_CFG_MIN		7
+#define ICE_FW_API_REPORT_DFLT_CFG_PATCH	3
+
 #endif /* _ICE_TYPE_H_ */
-- 
2.26.2

[PATCH net-next 10/15] ice: Limit forced overrides based on FW version

[Tony Nguyen]

From: Jeb Cramer <jeb.j.cramer@intel.com>

Beyond a specific version of firmware, there is no need to provide
override values to the firmware when setting PHY capabilities.  In this
case, we do not need to indicate whether we're in Strict or Lenient Link
Mode.

In the case of translating capabilities to the configuration structure,
the module compliance enforcement is already correctly set by firmware,
so the extra code block is redundant.

Signed-off-by: Jeb Cramer <jeb.j.cramer@intel.com>
Tested-by: Tony Brelinski <tonyx.brelinski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_common.c | 14 ++------------
 1 file changed, 2 insertions(+), 12 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_common.c b/drivers/net/ethernet/intel/ice/ice_common.c
index 8485450aff80..b13a630ea1b7 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.c
+++ b/drivers/net/ethernet/intel/ice/ice_common.c
@@ -3013,17 +3013,6 @@ ice_copy_phy_caps_to_cfg(struct ice_port_info *pi,
 	cfg->link_fec_opt = caps->link_fec_options;
 	cfg->module_compliance_enforcement =
 		caps->module_compliance_enforcement;
-
-	if (ice_fw_supports_link_override(pi->hw)) {
-		struct ice_link_default_override_tlv tlv;
-
-		if (ice_get_link_default_override(&tlv, pi))
-			return;
-
-		if (tlv.options & ICE_LINK_OVERRIDE_STRICT_MODE)
-			cfg->module_compliance_enforcement |=
-				ICE_LINK_OVERRIDE_STRICT_MODE;
-	}
 }
 
 /**
@@ -3091,7 +3080,8 @@ ice_cfg_phy_fec(struct ice_port_info *pi, struct ice_aqc_set_phy_cfg_data *cfg,
 		break;
 	}
 
-	if (fec == ICE_FEC_AUTO && ice_fw_supports_link_override(pi->hw)) {
+	if (fec == ICE_FEC_AUTO && ice_fw_supports_link_override(hw) &&
+	    !ice_fw_supports_report_dflt_cfg(hw)) {
 		struct ice_link_default_override_tlv tlv;
 
 		if (ice_get_link_default_override(&tlv, pi))
-- 
2.26.2

[PATCH net-next 11/15] ice: Remove unnecessary variable

[Tony Nguyen]

From: Anirudh Venkataramanan <anirudh.venkataramanan@intel.com>

In ice_init_phy_user_cfg, vsi is used only to get to hw. Remove this
and just use pi->hw

Signed-off-by: Anirudh Venkataramanan <anirudh.venkataramanan@intel.com>
Tested-by: Tony Brelinski <tonyx.brelinski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_main.c | 7 +------
 1 file changed, 1 insertion(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index 6bc2215b674d..c81eb27e83a6 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -1791,16 +1791,11 @@ static int ice_init_phy_user_cfg(struct ice_port_info *pi)
 	struct ice_phy_info *phy = &pi->phy;
 	struct ice_pf *pf = pi->hw->back;
 	enum ice_status status;
-	struct ice_vsi *vsi;
 	int err = 0;
 
 	if (!(phy->link_info.link_info & ICE_AQ_MEDIA_AVAILABLE))
 		return -EIO;
 
-	vsi = ice_get_main_vsi(pf);
-	if (!vsi)
-		return -EINVAL;
-
 	pcaps = kzalloc(sizeof(*pcaps), GFP_KERNEL);
 	if (!pcaps)
 		return -ENOMEM;
@@ -1820,7 +1815,7 @@ static int ice_init_phy_user_cfg(struct ice_port_info *pi)
 	ice_copy_phy_caps_to_cfg(pi, pcaps, &pi->phy.curr_user_phy_cfg);
 
 	/* check if lenient mode is supported and enabled */
-	if (ice_fw_supports_link_override(&vsi->back->hw) &&
+	if (ice_fw_supports_link_override(pi->hw) &&
 	    !(pcaps->module_compliance_enforcement &
 	      ICE_AQC_MOD_ENFORCE_STRICT_MODE)) {
 		set_bit(ICE_FLAG_LINK_LENIENT_MODE_ENA, pf->flags);
-- 
2.26.2

[PATCH net-next 12/15] ice: Use local variable instead of pointer derefs

[Tony Nguyen]

From: Anirudh Venkataramanan <anirudh.venkataramanan@intel.com>

Replace multiple instances of vsi->back and pi->phy with equivalent
local variables

Signed-off-by: Anirudh Venkataramanan <anirudh.venkataramanan@intel.com>
Tested-by: Tony Brelinski <tonyx.brelinski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_main.c | 27 ++++++++++-------------
 1 file changed, 12 insertions(+), 15 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index c81eb27e83a6..73dd272acbf9 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -1857,27 +1857,24 @@ static int ice_init_phy_user_cfg(struct ice_port_info *pi)
 static int ice_configure_phy(struct ice_vsi *vsi)
 {
 	struct device *dev = ice_pf_to_dev(vsi->back);
+	struct ice_port_info *pi = vsi->port_info;
 	struct ice_aqc_get_phy_caps_data *pcaps;
 	struct ice_aqc_set_phy_cfg_data *cfg;
-	struct ice_port_info *pi;
+	struct ice_phy_info *phy = &pi->phy;
+	struct ice_pf *pf = vsi->back;
 	enum ice_status status;
 	int err = 0;
 
-	pi = vsi->port_info;
-	if (!pi)
-		return -EINVAL;
-
 	/* Ensure we have media as we cannot configure a medialess port */
-	if (!(pi->phy.link_info.link_info & ICE_AQ_MEDIA_AVAILABLE))
+	if (!(phy->link_info.link_info & ICE_AQ_MEDIA_AVAILABLE))
 		return -EPERM;
 
 	ice_print_topo_conflict(vsi);
 
-	if (vsi->port_info->phy.link_info.topo_media_conflict ==
-	    ICE_AQ_LINK_TOPO_UNSUPP_MEDIA)
+	if (phy->link_info.topo_media_conflict == ICE_AQ_LINK_TOPO_UNSUPP_MEDIA)
 		return -EPERM;
 
-	if (test_bit(ICE_FLAG_LINK_DOWN_ON_CLOSE_ENA, vsi->back->flags))
+	if (test_bit(ICE_FLAG_LINK_DOWN_ON_CLOSE_ENA, pf->flags))
 		return ice_force_phys_link_state(vsi, true);
 
 	pcaps = kzalloc(sizeof(*pcaps), GFP_KERNEL);
@@ -1898,7 +1895,7 @@ static int ice_configure_phy(struct ice_vsi *vsi)
 	 * there's nothing to do
 	 */
 	if (pcaps->caps & ICE_AQC_PHY_EN_LINK &&
-	    ice_phy_caps_equals_cfg(pcaps, &pi->phy.curr_user_phy_cfg))
+	    ice_phy_caps_equals_cfg(pcaps, &phy->curr_user_phy_cfg))
 		goto done;
 
 	/* Use PHY topology as baseline for configuration */
@@ -1929,8 +1926,8 @@ static int ice_configure_phy(struct ice_vsi *vsi)
 	 */
 	if (test_and_clear_bit(__ICE_LINK_DEFAULT_OVERRIDE_PENDING,
 			       vsi->back->state)) {
-		cfg->phy_type_low = pi->phy.curr_user_phy_cfg.phy_type_low;
-		cfg->phy_type_high = pi->phy.curr_user_phy_cfg.phy_type_high;
+		cfg->phy_type_low = phy->curr_user_phy_cfg.phy_type_low;
+		cfg->phy_type_high = phy->curr_user_phy_cfg.phy_type_high;
 	} else {
 		u64 phy_low = 0, phy_high = 0;
 
@@ -1948,7 +1945,7 @@ static int ice_configure_phy(struct ice_vsi *vsi)
 	}
 
 	/* FEC */
-	ice_cfg_phy_fec(pi, cfg, pi->phy.curr_user_fec_req);
+	ice_cfg_phy_fec(pi, cfg, phy->curr_user_fec_req);
 
 	/* Can't provide what was requested; use PHY capabilities */
 	if (cfg->link_fec_opt !=
@@ -1960,12 +1957,12 @@ static int ice_configure_phy(struct ice_vsi *vsi)
 	/* Flow Control - always supported; no need to check against
 	 * capabilities
 	 */
-	ice_cfg_phy_fc(pi, cfg, pi->phy.curr_user_fc_req);
+	ice_cfg_phy_fc(pi, cfg, phy->curr_user_fc_req);
 
 	/* Enable link and link update */
 	cfg->caps |= ICE_AQ_PHY_ENA_AUTO_LINK_UPDT | ICE_AQ_PHY_ENA_LINK;
 
-	status = ice_aq_set_phy_cfg(&vsi->back->hw, pi, cfg, NULL);
+	status = ice_aq_set_phy_cfg(&pf->hw, pi, cfg, NULL);
 	if (status) {
 		dev_err(dev, "Failed to set phy config, VSI %d error %s\n",
 			vsi->vsi_num, ice_stat_str(status));
-- 
2.26.2

[PATCH net-next 13/15] ice: Remove rx_gro_dropped stat

[Tony Nguyen]

From: Anirudh Venkataramanan <anirudh.venkataramanan@intel.com>

Tracking of the rx_gro_dropped statistic was removed in
commit f73fc40327c0 ("ice: drop dead code in ice_receive_skb()").
Remove the associated variables and its reporting to ethtool stats.

Signed-off-by: Anirudh Venkataramanan <anirudh.venkataramanan@intel.com>
Tested-by: Tony Brelinski <tonyx.brelinski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice.h         | 1 -
 drivers/net/ethernet/intel/ice/ice_ethtool.c | 1 -
 drivers/net/ethernet/intel/ice/ice_main.c    | 4 +---
 drivers/net/ethernet/intel/ice/ice_txrx.h    | 1 -
 4 files changed, 1 insertion(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice.h b/drivers/net/ethernet/intel/ice/ice.h
index 02badaaf818c..721afa0f0a88 100644
--- a/drivers/net/ethernet/intel/ice/ice.h
+++ b/drivers/net/ethernet/intel/ice/ice.h
@@ -265,7 +265,6 @@ struct ice_vsi {
 	u32 tx_busy;
 	u32 rx_buf_failed;
 	u32 rx_page_failed;
-	u32 rx_gro_dropped;
 	u16 num_q_vectors;
 	u16 base_vector;		/* IRQ base for OS reserved vectors */
 	enum ice_vsi_type type;
diff --git a/drivers/net/ethernet/intel/ice/ice_ethtool.c b/drivers/net/ethernet/intel/ice/ice_ethtool.c
index 4ca8d5880cfc..51d3a929ecfd 100644
--- a/drivers/net/ethernet/intel/ice/ice_ethtool.c
+++ b/drivers/net/ethernet/intel/ice/ice_ethtool.c
@@ -60,7 +60,6 @@ static const struct ice_stats ice_gstrings_vsi_stats[] = {
 	ICE_VSI_STAT("rx_unknown_protocol", eth_stats.rx_unknown_protocol),
 	ICE_VSI_STAT("rx_alloc_fail", rx_buf_failed),
 	ICE_VSI_STAT("rx_pg_alloc_fail", rx_page_failed),
-	ICE_VSI_STAT("rx_gro_dropped", rx_gro_dropped),
 	ICE_VSI_STAT("tx_errors", eth_stats.tx_errors),
 	ICE_VSI_STAT("tx_linearize", tx_linearize),
 	ICE_VSI_STAT("tx_busy", tx_busy),
diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index 73dd272acbf9..3c73ee4a6c0c 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -5354,7 +5354,6 @@ static void ice_update_vsi_ring_stats(struct ice_vsi *vsi)
 	vsi->tx_linearize = 0;
 	vsi->rx_buf_failed = 0;
 	vsi->rx_page_failed = 0;
-	vsi->rx_gro_dropped = 0;
 
 	rcu_read_lock();
 
@@ -5369,7 +5368,6 @@ static void ice_update_vsi_ring_stats(struct ice_vsi *vsi)
 		vsi_stats->rx_bytes += bytes;
 		vsi->rx_buf_failed += ring->rx_stats.alloc_buf_failed;
 		vsi->rx_page_failed += ring->rx_stats.alloc_page_failed;
-		vsi->rx_gro_dropped += ring->rx_stats.gro_dropped;
 	}
 
 	/* update XDP Tx rings counters */
@@ -5401,7 +5399,7 @@ void ice_update_vsi_stats(struct ice_vsi *vsi)
 	ice_update_eth_stats(vsi);
 
 	cur_ns->tx_errors = cur_es->tx_errors;
-	cur_ns->rx_dropped = cur_es->rx_discards + vsi->rx_gro_dropped;
+	cur_ns->rx_dropped = cur_es->rx_discards;
 	cur_ns->tx_dropped = cur_es->tx_discards;
 	cur_ns->multicast = cur_es->rx_multicast;
 
diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.h b/drivers/net/ethernet/intel/ice/ice_txrx.h
index ffe0d271dec7..701552d88bea 100644
--- a/drivers/net/ethernet/intel/ice/ice_txrx.h
+++ b/drivers/net/ethernet/intel/ice/ice_txrx.h
@@ -192,7 +192,6 @@ struct ice_rxq_stats {
 	u64 non_eop_descs;
 	u64 alloc_page_failed;
 	u64 alloc_buf_failed;
-	u64 gro_dropped; /* GRO returned dropped */
 };
 
 enum ice_ring_state_t {
-- 
2.26.2

[PATCH net-next 14/15] ice: Remove unnecessary checks in add/kill_vid ndo ops

[Tony Nguyen]

From: Brett Creeley <brett.creeley@intel.com>

Currently the driver is doing two unnecessary checks. First both ops are
checking if the VLAN ID passed in is less than VLAN_N_VID and second
both ops are checking to see if a port VLAN is configured on the VSI.

The first check is already handled by the 8021q driver so this is an
unnecessary check. The second check is unnecessary because the PF VSI is
never put into a port VLAN.

Remove these checks.

Signed-off-by: Brett Creeley <brett.creeley@intel.com>
Tested-by: Tony Brelinski <tonyx.brelinski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_main.c | 12 ------------
 1 file changed, 12 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index 3c73ee4a6c0c..5ab35c1d6121 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -3078,15 +3078,6 @@ ice_vlan_rx_add_vid(struct net_device *netdev, __always_unused __be16 proto,
 	struct ice_vsi *vsi = np->vsi;
 	int ret;
 
-	if (vid >= VLAN_N_VID) {
-		netdev_err(netdev, "VLAN id requested %d is out of range %d\n",
-			   vid, VLAN_N_VID);
-		return -EINVAL;
-	}
-
-	if (vsi->info.pvid)
-		return -EINVAL;
-
 	/* VLAN 0 is added by default during load/reset */
 	if (!vid)
 		return 0;
@@ -3124,9 +3115,6 @@ ice_vlan_rx_kill_vid(struct net_device *netdev, __always_unused __be16 proto,
 	struct ice_vsi *vsi = np->vsi;
 	int ret;
 
-	if (vsi->info.pvid)
-		return -EINVAL;
-
 	/* don't allow removal of VLAN 0 */
 	if (!vid)
 		return 0;
-- 
2.26.2

[PATCH net-next 15/15] ice: Remove unnecessary blank line

[Tony Nguyen]

Checkpatch reports the following, fix it.

-----------------------------------------
drivers/net/ethernet/intel/ice/ice_main.c
-----------------------------------------
CHECK:BRACES: Blank lines aren't necessary before a close brace '}'
FILE: drivers/net/ethernet/intel/ice/ice_main.c:455:
+
+}

Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Tested-by: Tony Brelinski <tonyx.brelinski@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_main.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index 5ab35c1d6121..30935aaa8935 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -436,7 +436,6 @@ static void ice_pf_dis_all_vsi(struct ice_pf *pf, bool locked)
 
 	for (node = 0; node < ICE_MAX_VF_AGG_NODES; node++)
 		pf->vf_agg_node[node].num_vsis = 0;
-
 }
 
 /**
-- 
2.26.2

Re: [PATCH net-next 00/15][pull request] 100GbE Intel Wired LAN Driver Updates 2021-04-08

[patchwork-bot+netdevbpf]

Hello:

This series was applied to netdev/net-next.git (refs/heads/master):

On Thu,  8 Apr 2021 09:13:06 -0700 you wrote:
> This series contains updates to ice driver only.
> 
> Chinh adds retrying of sending some AQ commands when receiving EBUSY
> error.
> 
> Victor modifies how nodes are added to reduce stack usage.
> 
> [...]

Here is the summary with links:
  - [net-next,01/15] ice: Re-send some AQ commands, as result of EBUSY AQ error
    https://git.kernel.org/netdev/net-next/c/3056df93f7a8
  - [net-next,02/15] ice: Modify recursive way of adding nodes
    https://git.kernel.org/netdev/net-next/c/7fb09a737536
  - [net-next,03/15] ice: Align macro names to the specification
    https://git.kernel.org/netdev/net-next/c/d6730a871e68
  - [net-next,04/15] ice: Ignore EMODE return for opcode 0x0605
    https://git.kernel.org/netdev/net-next/c/d348d51771b9
  - [net-next,05/15] ice: Remove unnecessary checker loop
    https://git.kernel.org/netdev/net-next/c/fd3dc1655eda
  - [net-next,06/15] ice: Rename a couple of variables
    https://git.kernel.org/netdev/net-next/c/0be39bb4c7c8
  - [net-next,07/15] ice: Fix error return codes in ice_set_link_ksettings
    https://git.kernel.org/netdev/net-next/c/450f10e79419
  - [net-next,08/15] ice: Replace some memsets and memcpys with assignment
    https://git.kernel.org/netdev/net-next/c/178a666daa0e
  - [net-next,09/15] ice: Use default configuration mode for PHY configuration
    https://git.kernel.org/netdev/net-next/c/0a02944feaa7
  - [net-next,10/15] ice: Limit forced overrides based on FW version
    https://git.kernel.org/netdev/net-next/c/75751c80d6d8
  - [net-next,11/15] ice: Remove unnecessary variable
    https://git.kernel.org/netdev/net-next/c/dc6aaa139fb7
  - [net-next,12/15] ice: Use local variable instead of pointer derefs
    https://git.kernel.org/netdev/net-next/c/efc1eddb28aa
  - [net-next,13/15] ice: Remove rx_gro_dropped stat
    https://git.kernel.org/netdev/net-next/c/51fe27e179b1
  - [net-next,14/15] ice: Remove unnecessary checks in add/kill_vid ndo ops
    https://git.kernel.org/netdev/net-next/c/771015b90b86
  - [net-next,15/15] ice: Remove unnecessary blank line
    https://git.kernel.org/netdev/net-next/c/2e20521b80c7

You are awesome, thank you!
--
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html