* [net-next v4 5/8] i40e: implement virtual device interface
From: Jeff Kirsher @ 2013-09-04 23:17 UTC (permalink / raw)
To: davem; +Cc: e1000-devel, netdev, Jesse Brandeburg, gospo, sassmann
In-Reply-To: <1378336629-24224-1-git-send-email-jeffrey.t.kirsher@intel.com>
From: Jesse Brandeburg <jesse.brandeburg@intel.com>
While not part of this patch series, an i40evf driver is on its
way, and uses these files to communicate to the PF driver.
This patch contains the header and implementation files for the
PF to VF interface.
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Kavindya Deegala <kavindya.s.deegala@intel.com>
CC: PJ Waskiewicz <peter.p.waskiewicz.jr@intel.com>
CC: e1000-devel@lists.sourceforge.net
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
v1:
this is the initial submittal
v2:
address upstream comments and internal fixes
kernel memory allocators already print warnings
various fixes based on internal review
add Mitch's sign-off (i40evf driver owner)
v3:
minor changes
v4:
__func__ removal
simplify NULL tests
sparse fixes
indent/whitespace fixes
---
drivers/net/ethernet/intel/i40e/i40e_virtchnl.h | 368 +++
drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c | 2408 ++++++++++++++++++++
drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h | 121 +
3 files changed, 2897 insertions(+)
create mode 100644 drivers/net/ethernet/intel/i40e/i40e_virtchnl.h
create mode 100644 drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
create mode 100644 drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h
diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl.h b/drivers/net/ethernet/intel/i40e/i40e_virtchnl.h
new file mode 100644
index 0000000..93b1177
--- /dev/null
+++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl.h
@@ -0,0 +1,368 @@
+/*******************************************************************************
+
+ Intel Ethernet Controller XL710 Family Linux Driver
+ Copyright(c) 2013 Intel Corporation.
+
+ This program is free software; you can redistribute it and/or modify it
+ under the terms and conditions of the GNU General Public License,
+ version 2, as published by the Free Software Foundation.
+
+ This program is distributed in the hope it will be useful, but WITHOUT
+ ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ more details.
+
+ You should have received a copy of the GNU General Public License along with
+ this program; if not, write to the Free Software Foundation, Inc.,
+ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
+
+ The full GNU General Public License is included in this distribution in
+ the file called "COPYING".
+
+ Contact Information:
+ e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
+ Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
+
+*******************************************************************************/
+
+#ifndef _I40E_VIRTCHNL_H_
+#define _I40E_VIRTCHNL_H_
+
+#include "i40e_type.h"
+
+/* Description:
+ * This header file describes the VF-PF communication protocol used
+ * by the various i40e drivers.
+ *
+ * Admin queue buffer usage:
+ * desc->opcode is always i40e_aqc_opc_send_msg_to_pf
+ * flags, retval, datalen, and data addr are all used normally.
+ * Firmware copies the cookie fields when sending messages between the PF and
+ * VF, but uses all other fields internally. Due to this limitation, we
+ * must send all messages as "indirect", i.e. using an external buffer.
+ *
+ * All the vsi indexes are relative to the VF. Each VF can have maximum of
+ * three VSIs. All the queue indexes are relative to the VSI. Each VF can
+ * have a maximum of sixteen queues for all of its VSIs.
+ *
+ * The PF is required to return a status code in v_retval for all messages
+ * except RESET_VF, which does not require any response. The return value is of
+ * i40e_status_code type, defined in the i40e_type.h.
+ *
+ * In general, VF driver initialization should roughly follow the order of these
+ * opcodes. The VF driver must first validate the API version of the PF driver,
+ * then request a reset, then get resources, then configure queues and
+ * interrupts. After these operations are complete, the VF driver may start
+ * its queues, optionally add MAC and VLAN filters, and process traffic.
+ */
+
+/* Opcodes for VF-PF communication. These are placed in the v_opcode field
+ * of the virtchnl_msg structure.
+ */
+enum i40e_virtchnl_ops {
+/* VF sends req. to pf for the following
+ * ops.
+ */
+ I40E_VIRTCHNL_OP_UNKNOWN = 0,
+ I40E_VIRTCHNL_OP_VERSION = 1, /* must ALWAYS be 1 */
+ I40E_VIRTCHNL_OP_RESET_VF,
+ I40E_VIRTCHNL_OP_GET_VF_RESOURCES,
+ I40E_VIRTCHNL_OP_CONFIG_TX_QUEUE,
+ I40E_VIRTCHNL_OP_CONFIG_RX_QUEUE,
+ I40E_VIRTCHNL_OP_CONFIG_VSI_QUEUES,
+ I40E_VIRTCHNL_OP_CONFIG_IRQ_MAP,
+ I40E_VIRTCHNL_OP_ENABLE_QUEUES,
+ I40E_VIRTCHNL_OP_DISABLE_QUEUES,
+ I40E_VIRTCHNL_OP_ADD_ETHER_ADDRESS,
+ I40E_VIRTCHNL_OP_DEL_ETHER_ADDRESS,
+ I40E_VIRTCHNL_OP_ADD_VLAN,
+ I40E_VIRTCHNL_OP_DEL_VLAN,
+ I40E_VIRTCHNL_OP_CONFIG_PROMISCUOUS_MODE,
+ I40E_VIRTCHNL_OP_GET_STATS,
+ I40E_VIRTCHNL_OP_FCOE,
+/* PF sends status change events to vfs using
+ * the following op.
+ */
+ I40E_VIRTCHNL_OP_EVENT,
+};
+
+/* Virtual channel message descriptor. This overlays the admin queue
+ * descriptor. All other data is passed in external buffers.
+ */
+
+struct i40e_virtchnl_msg {
+ u8 pad[8]; /* AQ flags/opcode/len/retval fields */
+ enum i40e_virtchnl_ops v_opcode; /* avoid confusion with desc->opcode */
+ i40e_status v_retval; /* ditto for desc->retval */
+ u32 vfid; /* used by PF when sending to VF */
+};
+
+/* Message descriptions and data structures.*/
+
+/* I40E_VIRTCHNL_OP_VERSION
+ * VF posts its version number to the PF. PF responds with its version number
+ * in the same format, along with a return code.
+ * Reply from PF has its major/minor versions also in param0 and param1.
+ * If there is a major version mismatch, then the VF cannot operate.
+ * If there is a minor version mismatch, then the VF can operate but should
+ * add a warning to the system log.
+ *
+ * This enum element MUST always be specified as == 1, regardless of other
+ * changes in the API. The PF must always respond to this message without
+ * error regardless of version mismatch.
+ */
+#define I40E_VIRTCHNL_VERSION_MAJOR 1
+#define I40E_VIRTCHNL_VERSION_MINOR 0
+struct i40e_virtchnl_version_info {
+ u32 major;
+ u32 minor;
+};
+
+/* I40E_VIRTCHNL_OP_RESET_VF
+ * VF sends this request to PF with no parameters
+ * PF does NOT respond! VF driver must delay then poll VFGEN_RSTAT register
+ * until reset completion is indicated. The admin queue must be reinitialized
+ * after this operation.
+ *
+ * When reset is complete, PF must ensure that all queues in all VSIs associated
+ * with the VF are stopped, all queue configurations in the HMC are set to 0,
+ * and all MAC and VLAN filters (except the default MAC address) on all VSIs
+ * are cleared.
+ */
+
+/* I40E_VIRTCHNL_OP_GET_VF_RESOURCES
+ * VF sends this request to PF with no parameters
+ * PF responds with an indirect message containing
+ * i40e_virtchnl_vf_resource and one or more
+ * i40e_virtchnl_vsi_resource structures.
+ */
+
+struct i40e_virtchnl_vsi_resource {
+ u16 vsi_id;
+ u16 num_queue_pairs;
+ enum i40e_vsi_type vsi_type;
+ u16 qset_handle;
+ u8 default_mac_addr[I40E_ETH_LENGTH_OF_ADDRESS];
+};
+/* VF offload flags */
+#define I40E_VIRTCHNL_VF_OFFLOAD_L2 0x00000001
+#define I40E_VIRTCHNL_VF_OFFLOAD_FCOE 0x00000004
+#define I40E_VIRTCHNL_VF_OFFLOAD_VLAN 0x00010000
+
+struct i40e_virtchnl_vf_resource {
+ u16 num_vsis;
+ u16 num_queue_pairs;
+ u16 max_vectors;
+ u16 max_mtu;
+
+ u32 vf_offload_flags;
+ u32 max_fcoe_contexts;
+ u32 max_fcoe_filters;
+
+ struct i40e_virtchnl_vsi_resource vsi_res[1];
+};
+
+/* I40E_VIRTCHNL_OP_CONFIG_TX_QUEUE
+ * VF sends this message to set up parameters for one TX queue.
+ * External data buffer contains one instance of i40e_virtchnl_txq_info.
+ * PF configures requested queue and returns a status code.
+ */
+
+/* Tx queue config info */
+struct i40e_virtchnl_txq_info {
+ u16 vsi_id;
+ u16 queue_id;
+ u16 ring_len; /* number of descriptors, multiple of 8 */
+ u16 headwb_enabled;
+ u64 dma_ring_addr;
+ u64 dma_headwb_addr;
+};
+
+/* I40E_VIRTCHNL_OP_CONFIG_RX_QUEUE
+ * VF sends this message to set up parameters for one RX queue.
+ * External data buffer contains one instance of i40e_virtchnl_rxq_info.
+ * PF configures requested queue and returns a status code.
+ */
+
+/* Rx queue config info */
+struct i40e_virtchnl_rxq_info {
+ u16 vsi_id;
+ u16 queue_id;
+ u32 ring_len; /* number of descriptors, multiple of 32 */
+ u16 hdr_size;
+ u16 splithdr_enabled;
+ u32 databuffer_size;
+ u32 max_pkt_size;
+ u64 dma_ring_addr;
+ enum i40e_hmc_obj_rx_hsplit_0 rx_split_pos;
+};
+
+/* I40E_VIRTCHNL_OP_CONFIG_VSI_QUEUES
+ * VF sends this message to set parameters for all active TX and RX queues
+ * associated with the specified VSI.
+ * PF configures queues and returns status.
+ * If the number of queues specified is greater than the number of queues
+ * associated with the VSI, an error is returned and no queues are configured.
+ */
+struct i40e_virtchnl_queue_pair_info {
+ /* NOTE: vsi_id and queue_id should be identical for both queues. */
+ struct i40e_virtchnl_txq_info txq;
+ struct i40e_virtchnl_rxq_info rxq;
+};
+
+struct i40e_virtchnl_vsi_queue_config_info {
+ u16 vsi_id;
+ u16 num_queue_pairs;
+ struct i40e_virtchnl_queue_pair_info qpair[1];
+};
+
+/* I40E_VIRTCHNL_OP_CONFIG_IRQ_MAP
+ * VF uses this message to map vectors to queues.
+ * The rxq_map and txq_map fields are bitmaps used to indicate which queues
+ * are to be associated with the specified vector.
+ * The "other" causes are always mapped to vector 0.
+ * PF configures interrupt mapping and returns status.
+ */
+struct i40e_virtchnl_vector_map {
+ u16 vsi_id;
+ u16 vector_id;
+ u16 rxq_map;
+ u16 txq_map;
+ u16 rxitr_idx;
+ u16 txitr_idx;
+};
+
+struct i40e_virtchnl_irq_map_info {
+ u16 num_vectors;
+ struct i40e_virtchnl_vector_map vecmap[1];
+};
+
+/* I40E_VIRTCHNL_OP_ENABLE_QUEUES
+ * I40E_VIRTCHNL_OP_DISABLE_QUEUES
+ * VF sends these message to enable or disable TX/RX queue pairs.
+ * The queues fields are bitmaps indicating which queues to act upon.
+ * (Currently, we only support 16 queues per VF, but we make the field
+ * u32 to allow for expansion.)
+ * PF performs requested action and returns status.
+ */
+struct i40e_virtchnl_queue_select {
+ u16 vsi_id;
+ u16 pad;
+ u32 rx_queues;
+ u32 tx_queues;
+};
+
+/* I40E_VIRTCHNL_OP_ADD_ETHER_ADDRESS
+ * VF sends this message in order to add one or more unicast or multicast
+ * address filters for the specified VSI.
+ * PF adds the filters and returns status.
+ */
+
+/* I40E_VIRTCHNL_OP_DEL_ETHER_ADDRESS
+ * VF sends this message in order to remove one or more unicast or multicast
+ * filters for the specified VSI.
+ * PF removes the filters and returns status.
+ */
+
+struct i40e_virtchnl_ether_addr {
+ u8 addr[I40E_ETH_LENGTH_OF_ADDRESS];
+ u8 pad[2];
+};
+
+struct i40e_virtchnl_ether_addr_list {
+ u16 vsi_id;
+ u16 num_elements;
+ struct i40e_virtchnl_ether_addr list[1];
+};
+
+/* I40E_VIRTCHNL_OP_ADD_VLAN
+ * VF sends this message to add one or more VLAN tag filters for receives.
+ * PF adds the filters and returns status.
+ * If a port VLAN is configured by the PF, this operation will return an
+ * error to the VF.
+ */
+
+/* I40E_VIRTCHNL_OP_DEL_VLAN
+ * VF sends this message to remove one or more VLAN tag filters for receives.
+ * PF removes the filters and returns status.
+ * If a port VLAN is configured by the PF, this operation will return an
+ * error to the VF.
+ */
+
+struct i40e_virtchnl_vlan_filter_list {
+ u16 vsi_id;
+ u16 num_elements;
+ u16 vlan_id[1];
+};
+
+/* I40E_VIRTCHNL_OP_CONFIG_PROMISCUOUS_MODE
+ * VF sends VSI id and flags.
+ * PF returns status code in retval.
+ * Note: we assume that broadcast accept mode is always enabled.
+ */
+struct i40e_virtchnl_promisc_info {
+ u16 vsi_id;
+ u16 flags;
+};
+
+#define I40E_FLAG_VF_UNICAST_PROMISC 0x00000001
+#define I40E_FLAG_VF_MULTICAST_PROMISC 0x00000002
+
+/* I40E_VIRTCHNL_OP_GET_STATS
+ * VF sends this message to request stats for the selected VSI. VF uses
+ * the i40e_virtchnl_queue_select struct to specify the VSI. The queue_id
+ * field is ignored by the PF.
+ *
+ * PF replies with struct i40e_eth_stats in an external buffer.
+ */
+
+/* I40E_VIRTCHNL_OP_EVENT
+ * PF sends this message to inform the VF driver of events that may affect it.
+ * No direct response is expected from the VF, though it may generate other
+ * messages in response to this one.
+ */
+enum i40e_virtchnl_event_codes {
+ I40E_VIRTCHNL_EVENT_UNKNOWN = 0,
+ I40E_VIRTCHNL_EVENT_LINK_CHANGE,
+ I40E_VIRTCHNL_EVENT_RESET_IMPENDING,
+ I40E_VIRTCHNL_EVENT_PF_DRIVER_CLOSE,
+};
+#define I40E_PF_EVENT_SEVERITY_INFO 0
+#define I40E_PF_EVENT_SEVERITY_CERTAIN_DOOM 255
+
+struct i40e_virtchnl_pf_event {
+ enum i40e_virtchnl_event_codes event;
+ union {
+ struct {
+ enum i40e_aq_link_speed link_speed;
+ bool link_status;
+ } link_event;
+ } event_data;
+
+ int severity;
+};
+
+/* The following are TBD, not necessary for LAN functionality.
+ * I40E_VIRTCHNL_OP_FCOE
+ */
+
+/* VF reset states - these are written into the RSTAT register:
+ * I40E_VFGEN_RSTAT1 on the PF
+ * I40E_VFGEN_RSTAT on the VF
+ * When the PF initiates a reset, it writes 0
+ * When the reset is complete, it writes 1
+ * When the PF detects that the VF has recovered, it writes 2
+ * VF checks this register periodically to determine if a reset has occurred,
+ * then polls it to know when the reset is complete.
+ * If either the PF or VF reads the register while the hardware
+ * is in a reset state, it will return DEADBEEF, which, when masked
+ * will result in 3.
+ */
+enum i40e_vfr_states {
+ I40E_VFR_INPROGRESS = 0,
+ I40E_VFR_COMPLETED,
+ I40E_VFR_VFACTIVE,
+ I40E_VFR_UNKNOWN,
+};
+
+#endif /* _I40E_VIRTCHNL_H_ */
diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
new file mode 100644
index 0000000..d6ab742
--- /dev/null
+++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
@@ -0,0 +1,2408 @@
+/*******************************************************************************
+
+ Intel Ethernet Controller XL710 Family Linux Driver
+ Copyright(c) 2013 Intel Corporation.
+
+ This program is free software; you can redistribute it and/or modify it
+ under the terms and conditions of the GNU General Public License,
+ version 2, as published by the Free Software Foundation.
+
+ This program is distributed in the hope it will be useful, but WITHOUT
+ ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ more details.
+
+ You should have received a copy of the GNU General Public License along with
+ this program; if not, write to the Free Software Foundation, Inc.,
+ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
+
+ The full GNU General Public License is included in this distribution in
+ the file called "COPYING".
+
+ Contact Information:
+ e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
+ Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
+
+*******************************************************************************/
+
+#include "i40e.h"
+
+/***********************misc routines*****************************/
+
+/**
+ * i40e_vc_isvalid_vsi_id
+ * @vf: pointer to the vf info
+ * @vsi_id: vf relative vsi id
+ *
+ * check for the valid vsi id
+ **/
+static inline bool i40e_vc_isvalid_vsi_id(struct i40e_vf *vf, u8 vsi_id)
+{
+ struct i40e_pf *pf = vf->pf;
+
+ return (pf->vsi[vsi_id]->vf_id == vf->vf_id);
+}
+
+/**
+ * i40e_vc_isvalid_queue_id
+ * @vf: pointer to the vf info
+ * @vsi_id: vsi id
+ * @qid: vsi relative queue id
+ *
+ * check for the valid queue id
+ **/
+static inline bool i40e_vc_isvalid_queue_id(struct i40e_vf *vf, u8 vsi_id,
+ u8 qid)
+{
+ struct i40e_pf *pf = vf->pf;
+
+ return (qid < pf->vsi[vsi_id]->num_queue_pairs);
+}
+
+/**
+ * i40e_vc_isvalid_vector_id
+ * @vf: pointer to the vf info
+ * @vector_id: vf relative vector id
+ *
+ * check for the valid vector id
+ **/
+static inline bool i40e_vc_isvalid_vector_id(struct i40e_vf *vf, u8 vector_id)
+{
+ struct i40e_pf *pf = vf->pf;
+
+ return (vector_id < pf->hw.func_caps.num_msix_vectors_vf);
+}
+
+/***********************vf resource mgmt routines*****************/
+
+/**
+ * i40e_vc_get_pf_queue_id
+ * @vf: pointer to the vf info
+ * @vsi_idx: index of VSI in PF struct
+ * @vsi_queue_id: vsi relative queue id
+ *
+ * return pf relative queue id
+ **/
+static u16 i40e_vc_get_pf_queue_id(struct i40e_vf *vf, u8 vsi_idx,
+ u8 vsi_queue_id)
+{
+ struct i40e_pf *pf = vf->pf;
+ struct i40e_vsi *vsi = pf->vsi[vsi_idx];
+ u16 pf_queue_id = I40E_QUEUE_END_OF_LIST;
+
+ if (le16_to_cpu(vsi->info.mapping_flags) &
+ I40E_AQ_VSI_QUE_MAP_NONCONTIG)
+ pf_queue_id =
+ le16_to_cpu(vsi->info.queue_mapping[vsi_queue_id]);
+ else
+ pf_queue_id = le16_to_cpu(vsi->info.queue_mapping[0]) +
+ vsi_queue_id;
+
+ return pf_queue_id;
+}
+
+/**
+ * i40e_ctrl_vsi_tx_queue
+ * @vf: pointer to the vf info
+ * @vsi_idx: index of VSI in PF struct
+ * @vsi_queue_id: vsi relative queue index
+ * @ctrl: control flags
+ *
+ * enable/disable/enable check/disable check
+ **/
+static i40e_status i40e_ctrl_vsi_tx_queue(struct i40e_vf *vf, u16 vsi_idx,
+ u16 vsi_queue_id,
+ enum i40e_queue_ctrl ctrl)
+{
+ i40e_status ret = I40E_SUCCESS;
+ struct i40e_pf *pf = vf->pf;
+ struct i40e_hw *hw = &pf->hw;
+ bool writeback = false;
+ u16 pf_queue_id;
+ u32 reg;
+
+ pf_queue_id = i40e_vc_get_pf_queue_id(vf, vsi_idx, vsi_queue_id);
+ reg = rd32(hw, I40E_QTX_ENA(pf_queue_id));
+
+ switch (ctrl) {
+ case I40E_QUEUE_CTRL_ENABLE:
+ reg |= I40E_QTX_ENA_QENA_REQ_MASK;
+ writeback = true;
+ break;
+ case I40E_QUEUE_CTRL_ENABLECHECK:
+ ret = (reg & I40E_QTX_ENA_QENA_STAT_MASK) ?
+ I40E_SUCCESS : I40E_ERR_CONFIG;
+ break;
+ case I40E_QUEUE_CTRL_DISABLE:
+ reg &= ~I40E_QTX_ENA_QENA_REQ_MASK;
+ writeback = true;
+ break;
+ case I40E_QUEUE_CTRL_DISABLECHECK:
+ ret = (reg & I40E_QTX_ENA_QENA_STAT_MASK) ?
+ I40E_ERR_CONFIG : I40E_SUCCESS;
+ break;
+ case I40E_QUEUE_CTRL_FASTDISABLE:
+ reg |= I40E_QTX_ENA_FAST_QDIS_MASK;
+ writeback = true;
+ break;
+ case I40E_QUEUE_CTRL_FASTDISABLECHECK:
+ ret = (reg & I40E_QTX_ENA_QENA_STAT_MASK) ?
+ I40E_ERR_CONFIG : I40E_SUCCESS;
+ if (ret == I40E_SUCCESS) {
+ reg &= ~I40E_QTX_ENA_FAST_QDIS_MASK;
+ writeback = true;
+ }
+ break;
+ default:
+ ret = I40E_ERR_PARAM;
+ break;
+ }
+
+ if (writeback) {
+ wr32(hw, I40E_QTX_ENA(pf_queue_id), reg);
+ i40e_flush(hw);
+ }
+
+ return ret;
+}
+
+/**
+ * i40e_ctrl_vsi_rx_queue
+ * @vf: pointer to the vf info
+ * @vsi_idx: index of VSI in PF struct
+ * @vsi_queue_id: vsi relative queue index
+ * @ctrl: control flags
+ *
+ * enable/disable/enable check/disable check
+ **/
+static i40e_status i40e_ctrl_vsi_rx_queue(struct i40e_vf *vf, u16 vsi_idx,
+ u16 vsi_queue_id,
+ enum i40e_queue_ctrl ctrl)
+{
+ i40e_status ret = I40E_SUCCESS;
+ struct i40e_pf *pf = vf->pf;
+ struct i40e_hw *hw = &pf->hw;
+ bool writeback = false;
+ u16 pf_queue_id;
+ u32 reg;
+
+ pf_queue_id = i40e_vc_get_pf_queue_id(vf, vsi_idx, vsi_queue_id);
+ reg = rd32(hw, I40E_QRX_ENA(pf_queue_id));
+
+ switch (ctrl) {
+ case I40E_QUEUE_CTRL_ENABLE:
+ reg |= I40E_QRX_ENA_QENA_REQ_MASK;
+ writeback = true;
+ break;
+ case I40E_QUEUE_CTRL_ENABLECHECK:
+ ret = (reg & I40E_QRX_ENA_QENA_STAT_MASK) ?
+ I40E_SUCCESS : I40E_ERR_CONFIG;
+ break;
+ case I40E_QUEUE_CTRL_DISABLE:
+ reg &= ~I40E_QRX_ENA_QENA_REQ_MASK;
+ writeback = true;
+ break;
+ case I40E_QUEUE_CTRL_DISABLECHECK:
+ ret = (reg & I40E_QRX_ENA_QENA_STAT_MASK) ?
+ I40E_ERR_CONFIG : I40E_SUCCESS;
+ break;
+ case I40E_QUEUE_CTRL_FASTDISABLE:
+ reg |= I40E_QRX_ENA_FAST_QDIS_MASK;
+ writeback = true;
+ break;
+ case I40E_QUEUE_CTRL_FASTDISABLECHECK:
+ ret = (reg & I40E_QRX_ENA_QENA_STAT_MASK) ?
+ I40E_ERR_CONFIG : I40E_SUCCESS;
+ if (ret == I40E_SUCCESS) {
+ reg &= ~I40E_QRX_ENA_FAST_QDIS_MASK;
+ writeback = true;
+ }
+ break;
+ default:
+ ret = I40E_ERR_PARAM;
+ break;
+ }
+
+ if (writeback) {
+ wr32(hw, I40E_QRX_ENA(pf_queue_id), reg);
+ i40e_flush(hw);
+ }
+
+ return ret;
+}
+
+/**
+ * i40e_config_irq_link_list
+ * @vf: pointer to the vf info
+ * @vsi_idx: index of VSI in PF struct
+ * @vecmap: irq map info
+ *
+ * configure irq link list from the map
+ **/
+static i40e_status i40e_config_irq_link_list(struct i40e_vf *vf, u16 vsi_idx,
+ struct i40e_virtchnl_vector_map
+ *vecmap)
+{
+ unsigned long linklistmap = 0, tempmap;
+ struct i40e_pf *pf = vf->pf;
+ struct i40e_hw *hw = &pf->hw;
+ u16 vsi_queue_id, pf_queue_id;
+ enum i40e_queue_type qtype;
+ u16 next_q, vector_id;
+ u32 reg, reg_idx;
+ u16 itr_idx = 0;
+
+ vector_id = vecmap->vector_id;
+ /* setup the head */
+ if (0 == vector_id)
+ reg_idx = I40E_VPINT_LNKLST0(vf->vf_id);
+ else
+ reg_idx = I40E_VPINT_LNKLSTN(
+ ((pf->hw.func_caps.num_msix_vectors_vf - 1)
+ * vf->vf_id) + (vector_id - 1));
+
+ if (vecmap->rxq_map == 0 && vecmap->txq_map == 0) {
+ /* Special case - No queues mapped on this vector */
+ wr32(hw, reg_idx, I40E_VPINT_LNKLST0_FIRSTQ_INDX_MASK);
+ goto irq_list_done;
+ }
+ tempmap = vecmap->rxq_map;
+ vsi_queue_id = find_first_bit(&tempmap, I40E_MAX_VSI_QP);
+ while (vsi_queue_id < I40E_MAX_VSI_QP) {
+ linklistmap |= (1 <<
+ (I40E_VIRTCHNL_SUPPORTED_QTYPES *
+ vsi_queue_id));
+ vsi_queue_id =
+ find_next_bit(&tempmap, I40E_MAX_VSI_QP, vsi_queue_id + 1);
+ }
+
+ tempmap = vecmap->txq_map;
+ vsi_queue_id = find_first_bit(&tempmap, I40E_MAX_VSI_QP);
+ while (vsi_queue_id < I40E_MAX_VSI_QP) {
+ linklistmap |= (1 <<
+ (I40E_VIRTCHNL_SUPPORTED_QTYPES * vsi_queue_id
+ + 1));
+ vsi_queue_id = find_next_bit(&tempmap, I40E_MAX_VSI_QP,
+ vsi_queue_id + 1);
+ }
+
+ next_q = find_first_bit(&linklistmap,
+ (I40E_MAX_VSI_QP *
+ I40E_VIRTCHNL_SUPPORTED_QTYPES));
+ vsi_queue_id = next_q/I40E_VIRTCHNL_SUPPORTED_QTYPES;
+ qtype = next_q%I40E_VIRTCHNL_SUPPORTED_QTYPES;
+ pf_queue_id = i40e_vc_get_pf_queue_id(vf, vsi_idx, vsi_queue_id);
+ reg = ((qtype << I40E_VPINT_LNKLSTN_FIRSTQ_TYPE_SHIFT) | pf_queue_id);
+
+ wr32(hw, reg_idx, reg);
+
+ while (next_q < (I40E_MAX_VSI_QP * I40E_VIRTCHNL_SUPPORTED_QTYPES)) {
+ switch (qtype) {
+ case I40E_QUEUE_TYPE_RX:
+ reg_idx = I40E_QINT_RQCTL(pf_queue_id);
+ itr_idx = vecmap->rxitr_idx;
+ break;
+ case I40E_QUEUE_TYPE_TX:
+ reg_idx = I40E_QINT_TQCTL(pf_queue_id);
+ itr_idx = vecmap->txitr_idx;
+ break;
+ default:
+ break;
+ }
+
+ next_q = find_next_bit(&linklistmap,
+ (I40E_MAX_VSI_QP *
+ I40E_VIRTCHNL_SUPPORTED_QTYPES),
+ next_q + 1);
+ if (next_q < (I40E_MAX_VSI_QP * I40E_VIRTCHNL_SUPPORTED_QTYPES)) {
+ vsi_queue_id = next_q / I40E_VIRTCHNL_SUPPORTED_QTYPES;
+ qtype = next_q % I40E_VIRTCHNL_SUPPORTED_QTYPES;
+ pf_queue_id = i40e_vc_get_pf_queue_id(vf, vsi_idx,
+ vsi_queue_id);
+ } else {
+ pf_queue_id = I40E_QUEUE_END_OF_LIST;
+ qtype = 0;
+ }
+
+ /* format for the RQCTL & TQCTL regs is same */
+ reg = (vector_id) |
+ (qtype << I40E_QINT_RQCTL_NEXTQ_TYPE_SHIFT) |
+ (pf_queue_id << I40E_QINT_RQCTL_NEXTQ_INDX_SHIFT) |
+ (1 << I40E_QINT_RQCTL_CAUSE_ENA_SHIFT) |
+ (itr_idx << I40E_QINT_RQCTL_ITR_INDX_SHIFT);
+ wr32(hw, reg_idx, reg);
+ }
+
+irq_list_done:
+ i40e_flush(hw);
+
+ return I40E_SUCCESS;
+}
+
+/**
+ * i40e_config_vsi_tx_queue
+ * @vf: pointer to the vf info
+ * @vsi_idx: index of VSI in PF struct
+ * @vsi_queue_id: vsi relative queue index
+ * @info: config. info
+ *
+ * configure tx queue
+ **/
+static i40e_status i40e_config_vsi_tx_queue(struct i40e_vf *vf, u16 vsi_idx,
+ u16 vsi_queue_id,
+ struct i40e_virtchnl_txq_info
+ *info)
+{
+ struct i40e_pf *pf = vf->pf;
+ struct i40e_hw *hw = &pf->hw;
+ struct i40e_hmc_obj_txq tx_ctx;
+ u16 pf_queue_id;
+ i40e_status ret;
+ u32 qtx_ctl;
+
+ pf_queue_id = i40e_vc_get_pf_queue_id(vf, vsi_idx, vsi_queue_id);
+
+ /* clear the context structure first */
+ memset(&tx_ctx, 0, sizeof(struct i40e_hmc_obj_txq));
+
+ /* only set the required fields */
+ tx_ctx.base = info->dma_ring_addr / 128;
+ tx_ctx.qlen = info->ring_len;
+ tx_ctx.rdylist = le16_to_cpu(pf->vsi[vsi_idx]->info.qs_handle[0]);
+ tx_ctx.rdylist_act = 0;
+
+ /* clear the context in the HMC */
+ ret = i40e_clear_lan_tx_queue_context(hw, pf_queue_id);
+ if (ret != I40E_SUCCESS) {
+ dev_err(&pf->pdev->dev,
+ "Failed to clear VF LAN Tx queue context %d, error: %d\n",
+ pf_queue_id, ret);
+ goto error_context;
+ }
+
+ /* set the context in the HMC */
+ ret = i40e_set_lan_tx_queue_context(hw, pf_queue_id, &tx_ctx);
+ if (ret != I40E_SUCCESS) {
+ dev_err(&pf->pdev->dev,
+ "Failed to set VF LAN Tx queue context %d error: %d\n",
+ pf_queue_id, ret);
+ goto error_context;
+ }
+
+ /* associate this queue with the PCI VF function */
+ qtx_ctl = I40E_QTX_CTL_VF_QUEUE;
+ qtx_ctl |= ((hw->hmc.hmc_fn_id << I40E_QTX_CTL_PF_INDX_SHIFT)
+ & I40E_QTX_CTL_PF_INDX_MASK);
+ qtx_ctl |= (((vf->vf_id + hw->func_caps.vf_base_id)
+ << I40E_QTX_CTL_VFVM_INDX_SHIFT)
+ & I40E_QTX_CTL_VFVM_INDX_MASK);
+ wr32(hw, I40E_QTX_CTL(pf_queue_id), qtx_ctl);
+ i40e_flush(hw);
+
+error_context:
+ return ret;
+}
+
+/**
+ * i40e_config_vsi_rx_queue
+ * @vf: pointer to the vf info
+ * @vsi_idx: index of VSI in PF struct
+ * @vsi_queue_id: vsi relative queue index
+ * @info: config. info
+ *
+ * configure rx queue
+ **/
+static i40e_status i40e_config_vsi_rx_queue(struct i40e_vf *vf, u16 vsi_idx,
+ u16 vsi_queue_id,
+ struct i40e_virtchnl_rxq_info
+ *info)
+{
+ struct i40e_pf *pf = vf->pf;
+ struct i40e_hw *hw = &pf->hw;
+ struct i40e_hmc_obj_rxq rx_ctx;
+ i40e_status ret;
+ u16 pf_queue_id;
+
+ pf_queue_id = i40e_vc_get_pf_queue_id(vf, vsi_idx, vsi_queue_id);
+
+ /* clear the context structure first */
+ memset(&rx_ctx, 0, sizeof(struct i40e_hmc_obj_rxq));
+
+ /* only set the required fields */
+ rx_ctx.base = info->dma_ring_addr / 128;
+ rx_ctx.qlen = info->ring_len;
+
+ if (info->splithdr_enabled) {
+ rx_ctx.hsplit_0 = I40E_RX_SPLIT_L2 |
+ I40E_RX_SPLIT_IP |
+ I40E_RX_SPLIT_TCP_UDP |
+ I40E_RX_SPLIT_SCTP;
+ /* header length validation */
+ if (info->hdr_size > ((2 * 1024) - 64)) {
+ ret = I40E_ERR_PARAM;
+ goto error_param;
+ }
+ rx_ctx.hbuff = info->hdr_size >> I40E_RXQ_CTX_HBUFF_SHIFT;
+
+ /* set splitalways mode 10b */
+ rx_ctx.dtype = 0x2;
+ }
+
+ /* databuffer length validation */
+ if (info->databuffer_size > ((16 * 1024) - 128)) {
+ ret = I40E_ERR_PARAM;
+ goto error_param;
+ }
+ rx_ctx.dbuff = info->databuffer_size >> I40E_RXQ_CTX_DBUFF_SHIFT;
+
+ /* max pkt. length validation */
+ if (info->max_pkt_size >= (16 * 1024) || info->max_pkt_size < 64) {
+ ret = I40E_ERR_PARAM;
+ goto error_param;
+ }
+ rx_ctx.rxmax = info->max_pkt_size;
+
+ /* enable 32bytes desc always */
+ rx_ctx.dsize = 1;
+
+ /* default values */
+ rx_ctx.tphrdesc_ena = 1;
+ rx_ctx.tphwdesc_ena = 1;
+ rx_ctx.tphdata_ena = 1;
+ rx_ctx.tphhead_ena = 1;
+ rx_ctx.lrxqthresh = 2;
+ rx_ctx.crcstrip = 1;
+
+ /* clear the context in the HMC */
+ ret = i40e_clear_lan_rx_queue_context(hw, pf_queue_id);
+ if (ret != I40E_SUCCESS) {
+ dev_err(&pf->pdev->dev,
+ "Failed to clear VF LAN Rx queue context %d, error: %d\n",
+ pf_queue_id, ret);
+ goto error_context;
+ }
+
+ /* set the context in the HMC */
+ ret = i40e_set_lan_rx_queue_context(hw, pf_queue_id, &rx_ctx);
+ if (ret != I40E_SUCCESS)
+ dev_err(&pf->pdev->dev,
+ "Failed to set VF LAN Rx queue context %d error: %d\n",
+ pf_queue_id, ret);
+
+error_param:
+error_context:
+ return ret;
+}
+
+/**
+ * i40e_alloc_vsi_res
+ * @vf: pointer to the vf info
+ * @type: type of VSI to allocate
+ *
+ * alloc vf vsi context & resources
+ **/
+static i40e_status i40e_alloc_vsi_res(struct i40e_vf *vf,
+ enum i40e_vsi_type type)
+{
+ struct i40e_mac_filter *f = NULL;
+ i40e_status ret = I40E_SUCCESS;
+ struct i40e_pf *pf = vf->pf;
+ struct i40e_hw *hw = &pf->hw;
+ struct i40e_vsi *vsi;
+
+ vsi = i40e_vsi_setup(pf, type, pf->vsi[pf->lan_vsi]->seid, vf->vf_id);
+
+ if (!vsi) {
+ dev_err(&pf->pdev->dev,
+ "add vsi failed for vf %d, aq_err %d\n",
+ vf->vf_id, pf->hw.aq.asq_last_status);
+ goto error_alloc_vsi_res;
+ }
+ if (type == I40E_VSI_SRIOV) {
+ vf->lan_vsi_index = vsi->idx;
+ vf->lan_vsi_id = vsi->id;
+ dev_info(&pf->pdev->dev,
+ "LAN VSI index %d, VSI id %d\n",
+ vsi->idx, vsi->id);
+ f = i40e_add_filter(vsi, vf->default_lan_addr.addr,
+ 0, true, false);
+ }
+ if (!f) {
+ dev_err(&pf->pdev->dev, "Unable to add ucast filter\n");
+ ret = I40E_ERR_NO_MEMORY;
+ goto error_alloc_vsi_res;
+ }
+
+ /* program mac filter */
+ if (I40E_SUCCESS != i40e_sync_vsi_filters(vsi)) {
+ dev_err(&pf->pdev->dev, "Unable to program ucast filters\n");
+ ret = I40E_ERR_CONFIG;
+ goto error_alloc_vsi_res;
+ }
+
+ /* accept bcast pkts. by default */
+ ret = i40e_aq_set_vsi_broadcast(hw, vsi->seid, true, NULL);
+ if (I40E_SUCCESS != ret)
+ dev_err(&pf->pdev->dev,
+ "set vsi bcast failed for vf %d, vsi %d, aq_err %d\n",
+ vf->vf_id, vsi->idx, pf->hw.aq.asq_last_status);
+
+error_alloc_vsi_res:
+ return ret;
+}
+
+/**
+ * i40e_reset_vf
+ * @vf: pointer to the vf structure
+ * @flr: VFLR was issued or not
+ *
+ * reset the vf
+ **/
+i40e_status i40e_reset_vf(struct i40e_vf *vf, bool flr)
+{
+ i40e_status ret = I40E_ERR_PARAM;
+ struct i40e_pf *pf = vf->pf;
+ struct i40e_hw *hw = &pf->hw;
+ u32 reg, reg_idx, msix_vf;
+ bool rsd = false;
+ u16 pf_queue_id;
+ int i, j;
+
+ /* warn the VF */
+ wr32(hw, I40E_VFGEN_RSTAT1(vf->vf_id), I40E_VFR_INPROGRESS);
+
+ clear_bit(I40E_VF_STAT_ACTIVE, &vf->vf_states);
+
+ /* PF triggers VFR only when VF requests, in case of
+ * VFLR, HW triggers VFR
+ */
+ if (!flr) {
+ /* reset vf using VPGEN_VFRTRIG reg */
+ reg = I40E_VPGEN_VFRTRIG_VFSWR_MASK;
+ wr32(hw, I40E_VPGEN_VFRTRIG(vf->vf_id), reg);
+ i40e_flush(hw);
+ }
+
+ /* poll VPGEN_VFRSTAT reg to make sure
+ * that reset is complete
+ */
+ for (i = 0; i < 4; i++) {
+ /* vf reset requires driver to first reset the
+ * vf & than poll the status register to make sure
+ * that the requested op was completed
+ * successfully
+ */
+ udelay(10);
+ reg = rd32(hw, I40E_VPGEN_VFRSTAT(vf->vf_id));
+ if (reg & I40E_VPGEN_VFRSTAT_VFRD_MASK) {
+ rsd = true;
+ break;
+ }
+ }
+
+ if (!rsd)
+ dev_err(&pf->pdev->dev, "VF reset check timeout %d\n",
+ vf->vf_id);
+
+ /* fast disable qps */
+ for (j = 0; j < pf->vsi[vf->lan_vsi_index]->num_queue_pairs; j++) {
+ ret = i40e_ctrl_vsi_tx_queue(vf, vf->lan_vsi_index, j,
+ I40E_QUEUE_CTRL_FASTDISABLE);
+ ret = i40e_ctrl_vsi_rx_queue(vf, vf->lan_vsi_index, j,
+ I40E_QUEUE_CTRL_FASTDISABLE);
+ }
+
+ /* Queue enable/disable requires driver to
+ * first reset the vf & than poll the status register
+ * to make sure that the requested op was completed
+ * successfully
+ */
+ udelay(10);
+ for (j = 0; j < pf->vsi[vf->lan_vsi_index]->num_queue_pairs; j++) {
+ ret = i40e_ctrl_vsi_tx_queue(vf, vf->lan_vsi_index, j,
+ I40E_QUEUE_CTRL_FASTDISABLECHECK);
+ if (ret != I40E_SUCCESS)
+ dev_info(&pf->pdev->dev,
+ "Queue control check failed on Tx queue %d of VSI %d VF %d\n",
+ vf->lan_vsi_index, j, vf->vf_id);
+ ret = i40e_ctrl_vsi_rx_queue(vf, vf->lan_vsi_index, j,
+ I40E_QUEUE_CTRL_FASTDISABLECHECK);
+ if (ret != I40E_SUCCESS)
+ dev_info(&pf->pdev->dev,
+ "Queue control check failed on Rx queue %d of VSI %d VF %d\n",
+ vf->lan_vsi_index, j, vf->vf_id);
+ }
+
+ /* clear the irq settings */
+ msix_vf = pf->hw.func_caps.num_msix_vectors_vf;
+ for (i = 0; i < msix_vf; i++) {
+ /* format is same for both registers */
+ if (0 == i)
+ reg_idx = I40E_VPINT_LNKLST0(vf->vf_id);
+ else
+ reg_idx = I40E_VPINT_LNKLSTN(((msix_vf - 1) *
+ (vf->vf_id))
+ + (i - 1));
+ reg = (I40E_VPINT_LNKLSTN_FIRSTQ_TYPE_MASK |
+ I40E_VPINT_LNKLSTN_FIRSTQ_INDX_MASK);
+ wr32(hw, reg_idx, reg);
+ i40e_flush(hw);
+ }
+ /* disable interrupts so the VF starts in a known state */
+ for (i = 0; i < msix_vf; i++) {
+ /* format is same for both registers */
+ if (0 == i)
+ reg_idx = I40E_VFINT_DYN_CTL0(vf->vf_id);
+ else
+ reg_idx = I40E_VFINT_DYN_CTLN(((msix_vf - 1) *
+ (vf->vf_id))
+ + (i - 1));
+ wr32(hw, reg_idx, I40E_VFINT_DYN_CTLN_CLEARPBA_MASK);
+ i40e_flush(hw);
+ }
+
+ /* set the defaults for the rqctl & tqctl registers */
+ reg = (I40E_QINT_RQCTL_NEXTQ_INDX_MASK | I40E_QINT_RQCTL_ITR_INDX_MASK |
+ I40E_QINT_RQCTL_NEXTQ_TYPE_MASK);
+ for (j = 0; j < pf->vsi[vf->lan_vsi_index]->num_queue_pairs; j++) {
+ pf_queue_id = i40e_vc_get_pf_queue_id(vf, vf->lan_vsi_index, j);
+ wr32(hw, I40E_QINT_RQCTL(pf_queue_id), reg);
+ wr32(hw, I40E_QINT_TQCTL(pf_queue_id), reg);
+ }
+
+ /* clear the reset bit in the VPGEN_VFRTRIG reg */
+ reg = rd32(hw, I40E_VPGEN_VFRTRIG(vf->vf_id));
+ reg &= ~I40E_VPGEN_VFRTRIG_VFSWR_MASK;
+ wr32(hw, I40E_VPGEN_VFRTRIG(vf->vf_id), reg);
+ /* tell the VF the reset is done */
+ wr32(hw, I40E_VFGEN_RSTAT1(vf->vf_id), I40E_VFR_COMPLETED);
+ i40e_flush(hw);
+
+ return ret;
+}
+
+/**
+ * i40e_enable_vf_mappings
+ * @vf: pointer to the vf info
+ *
+ * enable vf mappings
+ **/
+static i40e_status i40e_enable_vf_mappings(struct i40e_vf *vf)
+{
+ struct i40e_pf *pf = vf->pf;
+ struct i40e_hw *hw = &pf->hw;
+ u32 reg, total_queue_pairs = 0;
+ int j;
+
+ /* Tell the hardware we're using noncontiguous mapping. HW requires
+ * that VF queues be mapped using this method, even when they are
+ * contiguous in real life
+ */
+ wr32(hw, I40E_VSILAN_QBASE(vf->lan_vsi_id),
+ I40E_VSILAN_QBASE_VSIQTABLE_ENA_MASK);
+
+ /* enable VF vplan_qtable mappings */
+ reg = I40E_VPLAN_MAPENA_TXRX_ENA_MASK;
+ wr32(hw, I40E_VPLAN_MAPENA(vf->vf_id), reg);
+
+ /* map PF queues to VF queues */
+ for (j = 0; j < pf->vsi[vf->lan_vsi_index]->num_queue_pairs; j++) {
+ u16 qid = i40e_vc_get_pf_queue_id(vf, vf->lan_vsi_index, j);
+ reg = (qid & I40E_VPLAN_QTABLE_QINDEX_MASK);
+ wr32(hw, I40E_VPLAN_QTABLE(total_queue_pairs, vf->vf_id), reg);
+ total_queue_pairs++;
+ }
+
+ /* map PF queues to VSI */
+ for (j = 0; j < 7; j++) {
+ if (j * 2 >= pf->vsi[vf->lan_vsi_index]->num_queue_pairs) {
+ reg = 0x07FF07FF; /* unused */
+ } else {
+ u16 qid = i40e_vc_get_pf_queue_id(vf, vf->lan_vsi_index,
+ j * 2);
+ reg = qid;
+ qid = i40e_vc_get_pf_queue_id(vf, vf->lan_vsi_index,
+ (j * 2) + 1);
+ reg |= qid << 16;
+ }
+ wr32(hw, I40E_VSILAN_QTABLE(j, vf->lan_vsi_id), reg);
+ }
+
+ i40e_flush(hw);
+
+ return I40E_SUCCESS;
+}
+/**
+ * i40e_disable_vf_mappings
+ * @vf: pointer to the vf info
+ *
+ * disable vf mappings
+ **/
+static i40e_status i40e_disable_vf_mappings(struct i40e_vf *vf)
+{
+ struct i40e_pf *pf = vf->pf;
+ struct i40e_hw *hw = &pf->hw;
+ int i;
+
+ /* disable qp mappings */
+ wr32(hw, I40E_VPLAN_MAPENA(vf->vf_id), 0);
+ for (i = 0; i < I40E_MAX_VSI_QP; i++)
+ wr32(hw, I40E_VPLAN_QTABLE(i, vf->vf_id),
+ I40E_QUEUE_END_OF_LIST);
+ i40e_flush(hw);
+
+ return I40E_SUCCESS;
+}
+
+/**
+ * i40e_free_vf_res
+ * @vf: pointer to the vf info
+ *
+ * free vf resources
+ **/
+static i40e_status i40e_free_vf_res(struct i40e_vf *vf)
+{
+ i40e_status ret = I40E_ERR_PARAM;
+ struct i40e_pf *pf = vf->pf;
+
+ /* free vsi & disconnect it from the parent uplink */
+ if (vf->lan_vsi_index) {
+ ret = i40e_vsi_release(pf->vsi[vf->lan_vsi_index]);
+ vf->lan_vsi_index = 0;
+ vf->lan_vsi_id = 0;
+ }
+ /* reset some of the state varibles keeping
+ * track of the resources
+ */
+ vf->num_queue_pairs = 0;
+ vf->vf_states = 0;
+
+ return ret;
+}
+
+/**
+ * i40e_alloc_vf_res
+ * @vf: pointer to the vf info
+ *
+ * allocate vf resources
+ **/
+static i40e_status i40e_alloc_vf_res(struct i40e_vf *vf)
+{
+ struct i40e_pf *pf = vf->pf;
+ int total_queue_pairs = 0;
+ i40e_status ret;
+
+ /* allocate hw vsi context & associated resources */
+ ret = i40e_alloc_vsi_res(vf, I40E_VSI_SRIOV);
+ if (I40E_SUCCESS != ret)
+ goto error_alloc;
+ total_queue_pairs += pf->vsi[vf->lan_vsi_index]->num_queue_pairs;
+ set_bit(I40E_VIRTCHNL_VF_CAP_PRIVILEGE, &vf->vf_caps);
+
+
+ /* store the total qps number for the runtime
+ * vf req validation
+ */
+ vf->num_queue_pairs = total_queue_pairs;
+
+ /* vf is now completely initialized */
+ set_bit(I40E_VF_STAT_INIT, &vf->vf_states);
+
+error_alloc:
+ if (I40E_SUCCESS != ret)
+ i40e_free_vf_res(vf);
+
+ return ret;
+}
+
+/**
+ * i40e_vfs_are_assigned
+ * @pf: pointer to the pf structure
+ *
+ * Determine if any VFs are assigned to VMs
+ **/
+static bool i40e_vfs_are_assigned(struct i40e_pf *pf)
+{
+ struct pci_dev *pdev = pf->pdev;
+ struct pci_dev *vfdev;
+
+ /* loop through all the VFs to see if we own any that are assigned */
+ vfdev = pci_get_device(PCI_VENDOR_ID_INTEL, I40E_VF_DEVICE_ID , NULL);
+ while (vfdev) {
+ /* if we don't own it we don't care */
+ if (vfdev->is_virtfn && pci_physfn(vfdev) == pdev) {
+ /* if it is assigned we cannot release it */
+ if (vfdev->dev_flags & PCI_DEV_FLAGS_ASSIGNED)
+ return true;
+ }
+
+ vfdev = pci_get_device(PCI_VENDOR_ID_INTEL,
+ I40E_VF_DEVICE_ID,
+ vfdev);
+ }
+
+ return false;
+}
+
+/**
+ * i40e_free_vfs
+ * @pf: pointer to the pf structure
+ *
+ * free vf resources
+ **/
+i40e_status i40e_free_vfs(struct i40e_pf *pf)
+{
+ i40e_status ret = I40E_SUCCESS;
+ struct i40e_hw *hw = &pf->hw;
+ int i;
+
+ if (!pf->vf)
+ return I40E_SUCCESS;
+
+ /* Disable interrupt 0 so we don't try to handle the VFLR. */
+ wr32(hw, I40E_PFINT_DYN_CTL0, 0);
+ i40e_flush(hw);
+
+ /* free up vf resources */
+ for (i = 0; i < pf->num_alloc_vfs; i++) {
+ if (test_bit(I40E_VF_STAT_INIT, &pf->vf[i].vf_states))
+ ret = i40e_free_vf_res(&pf->vf[i]);
+ /* disable qp mappings */
+ i40e_disable_vf_mappings(&pf->vf[i]);
+ }
+
+ kfree(pf->vf);
+ pf->vf = NULL;
+ pf->num_alloc_vfs = 0;
+
+ if (!i40e_vfs_are_assigned(pf))
+ pci_disable_sriov(pf->pdev);
+ else
+ dev_warn(&pf->pdev->dev,
+ "unable to disable SR-IOV because VFs are assigned.\n");
+
+ /* Re-enable interrupt 0. */
+ wr32(hw, I40E_PFINT_DYN_CTL0,
+ I40E_PFINT_DYN_CTL0_INTENA_MASK |
+ I40E_PFINT_DYN_CTL0_CLEARPBA_MASK |
+ (I40E_ITR_NONE << I40E_PFINT_DYN_CTL0_ITR_INDX_SHIFT));
+ i40e_flush(hw);
+ return ret;
+}
+
+#ifdef CONFIG_PCI_IOV
+/**
+ * i40e_alloc_vfs
+ * @pf: pointer to the pf structure
+ * @num_alloc_vfs: number of vfs to allocate
+ *
+ * allocate vf resources
+ **/
+static i40e_status i40e_alloc_vfs(struct i40e_pf *pf, u16 num_alloc_vfs)
+{
+ i40e_status ret = I40E_SUCCESS;
+ struct i40e_vf *vfs;
+ int i, err = 0;
+
+ err = pci_enable_sriov(pf->pdev, num_alloc_vfs);
+ if (err) {
+ dev_err(&pf->pdev->dev,
+ "pci_enable_sriov failed with error %d!\n", err);
+ pf->num_alloc_vfs = 0;
+ ret = I40E_ERR_CONFIG;
+ goto err_iov;
+ }
+
+ /* allocate memory */
+ vfs = kzalloc(num_alloc_vfs * sizeof(struct i40e_vf), GFP_KERNEL);
+ if (!vfs) {
+ ret = I40E_ERR_NO_MEMORY;
+ goto err_alloc;
+ }
+
+ /* apply default profile */
+ for (i = 0; i < num_alloc_vfs; i++) {
+ vfs[i].pf = pf;
+ vfs[i].parent_type = I40E_SWITCH_ELEMENT_TYPE_VEB;
+ vfs[i].vf_id = i;
+
+ /* assign default capabilities */
+ set_bit(I40E_VIRTCHNL_VF_CAP_L2, &vfs[i].vf_caps);
+
+ ret = i40e_alloc_vf_res(&vfs[i]);
+ i40e_reset_vf(&vfs[i], true);
+ if (I40E_SUCCESS != ret)
+ break;
+
+ /* enable vf vplan_qtable mappings */
+ ret = i40e_enable_vf_mappings(&vfs[i]);
+ }
+ pf->vf = vfs;
+ pf->num_alloc_vfs = num_alloc_vfs;
+
+err_alloc:
+ if (I40E_SUCCESS != ret)
+ i40e_free_vfs(pf);
+err_iov:
+ return ret;
+}
+
+#endif
+/**
+ * i40e_pci_sriov_enable
+ * @pdev: pointer to a pci_dev structure
+ * @num_vfs: number of vfs to allocate
+ *
+ * Enable or change the number of VFs
+ **/
+static int i40e_pci_sriov_enable(struct pci_dev *pdev, int num_vfs)
+{
+#ifdef CONFIG_PCI_IOV
+ struct i40e_pf *pf = pci_get_drvdata(pdev);
+ int pre_existing_vfs = pci_num_vf(pdev);
+ int err = 0;
+
+ dev_info(&pdev->dev, "Allocating %d VFs.\n", num_vfs);
+ if (pre_existing_vfs && pre_existing_vfs != num_vfs)
+ err = i40e_free_vfs(pf);
+ else if (pre_existing_vfs && pre_existing_vfs == num_vfs)
+ goto out;
+
+ if (err)
+ goto err_out;
+
+ if (num_vfs > pf->num_req_vfs) {
+ err = -EPERM;
+ goto err_out;
+ }
+
+ err = i40e_alloc_vfs(pf, num_vfs);
+ if (err) {
+ dev_warn(&pdev->dev, "Failed to enable SR-IOV: %d\n", err);
+ goto err_out;
+ }
+
+out:
+ return num_vfs;
+
+err_out:
+ return err;
+#endif
+ return 0;
+}
+
+/**
+ * i40e_pci_sriov_configure
+ * @pdev: pointer to a pci_dev structure
+ * @num_vfs: number of vfs to allocate
+ *
+ * Enable or change the number of VFs. Called when the user updates the number
+ * of VFs in sysfs.
+ **/
+int i40e_pci_sriov_configure(struct pci_dev *pdev, int num_vfs)
+{
+ struct i40e_pf *pf = pci_get_drvdata(pdev);
+
+ if (num_vfs == 0)
+ return i40e_free_vfs(pf);
+ else
+ return i40e_pci_sriov_enable(pdev, num_vfs);
+}
+
+/***********************virtual channel routines******************/
+
+/**
+ * i40e_vc_send_msg_to_vf
+ * @vf: pointer to the vf info
+ * @v_opcode: virtual channel opcode
+ * @v_retval: virtual channel return value
+ * @msg: pointer to the msg buffer
+ * @msglen: msg length
+ *
+ * send msg to vf
+ **/
+static i40e_status i40e_vc_send_msg_to_vf(struct i40e_vf *vf,
+ u32 v_opcode, u32 v_retval,
+ u8 *msg, u16 msglen)
+{
+ struct i40e_pf *pf = vf->pf;
+ struct i40e_hw *hw = &pf->hw;
+ i40e_status ret;
+
+ /* single place to detect unsuccessful return values */
+ if (v_retval != I40E_SUCCESS) {
+ vf->num_invalid_msgs++;
+ dev_err(&pf->pdev->dev, "Failed opcode %d Error: %d\n",
+ v_opcode, v_retval);
+ if (vf->num_invalid_msgs >
+ I40E_DEFAULT_NUM_INVALID_MSGS_ALLOWED) {
+ dev_err(&pf->pdev->dev,
+ "Number of invalid messages exceeded for VF %d\n",
+ vf->vf_id);
+ dev_err(&pf->pdev->dev, "Use PF Control I/F to enable the VF\n");
+ set_bit(I40E_VF_STAT_DISABLED, &vf->vf_states);
+ }
+ } else {
+ vf->num_valid_msgs++;
+ }
+
+ ret = i40e_aq_send_msg_to_vf(hw, vf->vf_id, v_opcode, v_retval,
+ msg, msglen, NULL);
+ if (I40E_SUCCESS != ret)
+ dev_err(&pf->pdev->dev,
+ "Unable to send the message to VF %d aq_err %d\n",
+ vf->vf_id, pf->hw.aq.asq_last_status);
+
+ return ret;
+}
+
+/**
+ * i40e_vc_send_resp_to_vf
+ * @vf: pointer to the vf info
+ * @opcode: operation code
+ * @retval: return value
+ *
+ * send resp msg to vf
+ **/
+static i40e_status i40e_vc_send_resp_to_vf(struct i40e_vf *vf,
+ enum i40e_virtchnl_ops opcode,
+ i40e_status retval)
+{
+ i40e_status ret;
+
+ ret = i40e_vc_send_msg_to_vf(vf, opcode, retval, NULL, 0);
+
+ return ret;
+}
+
+/**
+ * i40e_vc_get_version_msg
+ * @vf: pointer to the vf info
+ *
+ * called from the vf to request the API version used by the PF
+ **/
+static i40e_status i40e_vc_get_version_msg(struct i40e_vf *vf)
+{
+ struct i40e_virtchnl_version_info info = {
+ I40E_VIRTCHNL_VERSION_MAJOR, I40E_VIRTCHNL_VERSION_MINOR
+ };
+
+ return i40e_vc_send_msg_to_vf(vf, I40E_VIRTCHNL_OP_VERSION,
+ I40E_SUCCESS, (u8 *)&info,
+ sizeof(struct
+ i40e_virtchnl_version_info));
+}
+
+/**
+ * i40e_vc_get_vf_resources_msg
+ * @vf: pointer to the vf info
+ * @msg: pointer to the msg buffer
+ * @msglen: msg length
+ *
+ * called from the vf to request its resources
+ **/
+static i40e_status i40e_vc_get_vf_resources_msg(struct i40e_vf *vf)
+{
+ struct i40e_virtchnl_vf_resource *vfres = NULL;
+ i40e_status ret = I40E_SUCCESS;
+ struct i40e_pf *pf = vf->pf;
+ struct i40e_vsi *vsi;
+ int i = 0, len = 0;
+ int num_vsis = 1;
+
+ if (!test_bit(I40E_VF_STAT_INIT, &vf->vf_states)) {
+ ret = I40E_ERR_PARAM;
+ goto err;
+ }
+
+ len = (sizeof(struct i40e_virtchnl_vf_resource) +
+ sizeof(struct i40e_virtchnl_vsi_resource) * num_vsis);
+
+ vfres = kzalloc(len, GFP_KERNEL);
+ if (!vfres) {
+ ret = I40E_ERR_NO_MEMORY;
+ len = 0;
+ goto err;
+ }
+
+ vfres->vf_offload_flags = I40E_VIRTCHNL_VF_OFFLOAD_L2;
+ vsi = pf->vsi[vf->lan_vsi_index];
+ if (!vsi->info.pvid)
+ vfres->vf_offload_flags |= I40E_VIRTCHNL_VF_OFFLOAD_VLAN;
+
+ vfres->num_vsis = num_vsis;
+ vfres->num_queue_pairs = vf->num_queue_pairs;
+ vfres->max_vectors = pf->hw.func_caps.num_msix_vectors_vf;
+ if (vf->lan_vsi_index) {
+ vfres->vsi_res[i].vsi_id = vf->lan_vsi_index;
+ vfres->vsi_res[i].vsi_type = I40E_VSI_SRIOV;
+ vfres->vsi_res[i].num_queue_pairs =
+ pf->vsi[vf->lan_vsi_index]->num_queue_pairs;
+ memcpy(vfres->vsi_res[i].default_mac_addr,
+ vf->default_lan_addr.addr, ETH_ALEN);
+ i++;
+ }
+ set_bit(I40E_VF_STAT_ACTIVE, &vf->vf_states);
+
+err:
+ /* send the response back to the vf */
+ ret = i40e_vc_send_msg_to_vf(vf, I40E_VIRTCHNL_OP_GET_VF_RESOURCES,
+ ret, (u8 *)vfres, len);
+
+ kfree(vfres);
+ return ret;
+}
+
+/**
+ * i40e_vc_reset_vf_msg
+ * @vf: pointer to the vf info
+ * @msg: pointer to the msg buffer
+ * @msglen: msg length
+ *
+ * called from the vf to reset itself,
+ * unlike other virtchnl messages, pf driver
+ * doesn't send the response back to the vf
+ **/
+static i40e_status i40e_vc_reset_vf_msg(struct i40e_vf *vf)
+{
+ i40e_status ret;
+
+ if (!test_bit(I40E_VF_STAT_ACTIVE, &vf->vf_states)) {
+ ret = I40E_ERR_PARAM;
+ goto error_param;
+ }
+
+ ret = i40e_reset_vf(vf, false);
+
+error_param:
+ return ret;
+}
+
+/**
+ * i40e_vc_config_promiscuous_mode_msg
+ * @vf: pointer to the vf info
+ * @msg: pointer to the msg buffer
+ * @msglen: msg length
+ *
+ * called from the vf to configure the promiscuous mode of
+ * vf vsis
+ **/
+static i40e_status i40e_vc_config_promiscuous_mode_msg(struct i40e_vf *vf,
+ u8 *msg, u16 msglen)
+{
+ struct i40e_virtchnl_promisc_info *info =
+ (struct i40e_virtchnl_promisc_info *)msg;
+ struct i40e_pf *pf = vf->pf;
+ struct i40e_hw *hw = &pf->hw;
+ bool allmulti = false;
+ bool promisc = false;
+ i40e_status ret;
+
+ if (!test_bit(I40E_VF_STAT_ACTIVE, &vf->vf_states) ||
+ !test_bit(I40E_VIRTCHNL_VF_CAP_PRIVILEGE, &vf->vf_caps) ||
+ !i40e_vc_isvalid_vsi_id(vf, info->vsi_id) ||
+ (pf->vsi[info->vsi_id]->type != I40E_VSI_FCOE)) {
+ ret = I40E_ERR_PARAM;
+ goto error_param;
+ }
+
+ if (info->flags & I40E_FLAG_VF_UNICAST_PROMISC)
+ promisc = true;
+ ret = i40e_aq_set_vsi_unicast_promiscuous(hw, info->vsi_id,
+ promisc, NULL);
+ if (ret)
+ goto error_param;
+
+ if (info->flags & I40E_FLAG_VF_MULTICAST_PROMISC)
+ allmulti = true;
+ ret = i40e_aq_set_vsi_multicast_promiscuous(hw, info->vsi_id,
+ allmulti, NULL);
+
+error_param:
+ /* send the response to the vf */
+ ret = i40e_vc_send_resp_to_vf(vf,
+ I40E_VIRTCHNL_OP_CONFIG_PROMISCUOUS_MODE,
+ ret);
+
+ return ret;
+}
+
+/**
+ * i40e_vc_config_queues_msg
+ * @vf: pointer to the vf info
+ * @msg: pointer to the msg buffer
+ * @msglen: msg length
+ *
+ * called from the vf to configure the rx/tx
+ * queues
+ **/
+static i40e_status i40e_vc_config_queues_msg(struct i40e_vf *vf,
+ u8 *msg, u16 msglen)
+{
+ struct i40e_virtchnl_vsi_queue_config_info *qci =
+ (struct i40e_virtchnl_vsi_queue_config_info *)msg;
+ struct i40e_virtchnl_queue_pair_info *qpi;
+ i40e_status ret = I40E_ERR_PARAM;
+ u16 vsi_id, vsi_queue_id;
+ int i;
+
+ if (!test_bit(I40E_VF_STAT_ACTIVE, &vf->vf_states)) {
+ ret = I40E_ERR_PARAM;
+ goto error_param;
+ }
+
+ vsi_id = qci->vsi_id;
+ if (!i40e_vc_isvalid_vsi_id(vf, vsi_id)) {
+ ret = I40E_ERR_PARAM;
+ goto error_param;
+ }
+ for (i = 0; i < qci->num_queue_pairs; i++) {
+ qpi = &qci->qpair[i];
+ vsi_queue_id = qpi->txq.queue_id;
+ if ((qpi->txq.vsi_id != vsi_id) ||
+ (qpi->rxq.vsi_id != vsi_id) ||
+ (qpi->rxq.queue_id != vsi_queue_id) ||
+ !i40e_vc_isvalid_queue_id(vf, vsi_id, vsi_queue_id)) {
+ ret = I40E_ERR_PARAM;
+ goto error_param;
+ }
+
+ ret = i40e_config_vsi_rx_queue(vf, vsi_id, vsi_queue_id,
+ &qpi->rxq);
+ if (ret != I40E_SUCCESS)
+ break;
+ ret = i40e_config_vsi_tx_queue(vf, vsi_id, vsi_queue_id,
+ &qpi->txq);
+ if (ret != I40E_SUCCESS)
+ break;
+ }
+
+error_param:
+ /* send the response to the vf */
+ ret = i40e_vc_send_resp_to_vf(vf,
+ I40E_VIRTCHNL_OP_CONFIG_VSI_QUEUES,
+ ret);
+
+ return ret;
+}
+
+/**
+ * i40e_vc_config_irq_map_msg
+ * @vf: pointer to the vf info
+ * @msg: pointer to the msg buffer
+ * @msglen: msg length
+ *
+ * called from the vf to configure the irq to
+ * queue map
+ **/
+static i40e_status i40e_vc_config_irq_map_msg(struct i40e_vf *vf,
+ u8 *msg, u16 msglen)
+{
+ struct i40e_virtchnl_irq_map_info *irqmap_info =
+ (struct i40e_virtchnl_irq_map_info *)msg;
+ struct i40e_virtchnl_vector_map *map;
+ u16 vsi_id, vsi_queue_id, vector_id;
+ i40e_status ret = I40E_SUCCESS;
+ unsigned long tempmap;
+ int i;
+
+ if (!test_bit(I40E_VF_STAT_ACTIVE, &vf->vf_states)) {
+ ret = I40E_ERR_PARAM;
+ goto error_param;
+ }
+
+ for (i = 0; i < irqmap_info->num_vectors; i++) {
+ map = &irqmap_info->vecmap[i];
+
+ vector_id = map->vector_id;
+ vsi_id = map->vsi_id;
+ /* validate msg params */
+ if (!i40e_vc_isvalid_vector_id(vf, vector_id) ||
+ !i40e_vc_isvalid_vsi_id(vf, vsi_id)) {
+ ret = I40E_ERR_PARAM;
+ goto error_param;
+ }
+
+ /* lookout for the invalid queue index */
+ tempmap = map->rxq_map;
+ vsi_queue_id = find_first_bit(&tempmap, I40E_MAX_VSI_QP);
+ while (vsi_queue_id < I40E_MAX_VSI_QP) {
+ if (!i40e_vc_isvalid_queue_id(vf, vsi_id,
+ vsi_queue_id)) {
+ ret = I40E_ERR_PARAM;
+ goto error_param;
+ }
+ vsi_queue_id = find_next_bit(&tempmap, I40E_MAX_VSI_QP,
+ vsi_queue_id + 1);
+ }
+
+ tempmap = map->txq_map;
+ vsi_queue_id = find_first_bit(&tempmap, I40E_MAX_VSI_QP);
+ while (vsi_queue_id < I40E_MAX_VSI_QP) {
+ if (!i40e_vc_isvalid_queue_id(vf, vsi_id,
+ vsi_queue_id)) {
+ ret = I40E_ERR_PARAM;
+ goto error_param;
+ }
+ vsi_queue_id = find_next_bit(&tempmap, I40E_MAX_VSI_QP,
+ vsi_queue_id + 1);
+ }
+
+ ret = i40e_config_irq_link_list(vf, vsi_id, map);
+ }
+error_param:
+ /* send the response to the vf */
+ ret = i40e_vc_send_resp_to_vf(vf, I40E_VIRTCHNL_OP_CONFIG_IRQ_MAP, ret);
+
+ return ret;
+}
+
+/**
+ * i40e_vc_enable_queues_msg
+ * @vf: pointer to the vf info
+ * @msg: pointer to the msg buffer
+ * @msglen: msg length
+ *
+ * called from the vf to enable all or specific queue(s)
+ **/
+static i40e_status i40e_vc_enable_queues_msg(struct i40e_vf *vf,
+ u8 *msg, u16 msglen)
+{
+ struct i40e_virtchnl_queue_select *vqs =
+ (struct i40e_virtchnl_queue_select *)msg;
+ i40e_status ret = I40E_SUCCESS;
+ struct i40e_pf *pf = vf->pf;
+ u16 vsi_id = vqs->vsi_id;
+ unsigned long tempmap;
+ u16 queue_id;
+
+ if (!test_bit(I40E_VF_STAT_ACTIVE, &vf->vf_states)) {
+ ret = I40E_ERR_PARAM;
+ goto error_param;
+ }
+
+ if (!i40e_vc_isvalid_vsi_id(vf, vsi_id)) {
+ ret = I40E_ERR_PARAM;
+ goto error_param;
+ }
+
+ if ((0 == vqs->rx_queues) && (0 == vqs->tx_queues)) {
+ ret = I40E_ERR_PARAM;
+ goto error_param;
+ }
+
+ tempmap = vqs->rx_queues;
+ queue_id = find_first_bit(&tempmap, I40E_MAX_VSI_QP);
+ while (queue_id < I40E_MAX_VSI_QP) {
+ if (!i40e_vc_isvalid_queue_id(vf, vsi_id, queue_id)) {
+ ret = I40E_ERR_PARAM;
+ goto error_param;
+ }
+ ret = i40e_ctrl_vsi_rx_queue(vf, vsi_id, queue_id,
+ I40E_QUEUE_CTRL_ENABLE);
+
+ queue_id = find_next_bit(&tempmap, I40E_MAX_VSI_QP,
+ queue_id + 1);
+ }
+
+ tempmap = vqs->tx_queues;
+ queue_id = find_first_bit(&tempmap, I40E_MAX_VSI_QP);
+ while (queue_id < I40E_MAX_VSI_QP) {
+ if (!i40e_vc_isvalid_queue_id(vf, vsi_id, queue_id)) {
+ ret = I40E_ERR_PARAM;
+ goto error_param;
+ }
+ ret = i40e_ctrl_vsi_tx_queue(vf, vsi_id, queue_id,
+ I40E_QUEUE_CTRL_ENABLE);
+
+ queue_id = find_next_bit(&tempmap, I40E_MAX_VSI_QP,
+ queue_id + 1);
+ }
+
+ /* Poll the status register to make sure that the
+ * requested op was completed successfully
+ */
+ udelay(10);
+
+ tempmap = vqs->rx_queues;
+ queue_id = find_first_bit(&tempmap, I40E_MAX_VSI_QP);
+ while (queue_id < I40E_MAX_VSI_QP) {
+ ret = i40e_ctrl_vsi_rx_queue(vf, vsi_id, queue_id,
+ I40E_QUEUE_CTRL_ENABLECHECK);
+ if (ret != I40E_SUCCESS) {
+ dev_err(&pf->pdev->dev,
+ "Queue control check failed on RX queue %d of VSI %d VF %d\n",
+ queue_id, vsi_id, vf->vf_id);
+ }
+ queue_id = find_next_bit(&tempmap, I40E_MAX_VSI_QP,
+ queue_id + 1);
+ }
+
+ tempmap = vqs->tx_queues;
+ queue_id = find_first_bit(&tempmap, I40E_MAX_VSI_QP);
+ while (queue_id < I40E_MAX_VSI_QP) {
+ ret = i40e_ctrl_vsi_tx_queue(vf, vsi_id, queue_id,
+ I40E_QUEUE_CTRL_ENABLECHECK);
+ if (ret != I40E_SUCCESS) {
+ dev_err(&pf->pdev->dev,
+ "Queue control check failed on TX queue %d of VSI %d VF %d\n",
+ queue_id, vsi_id, vf->vf_id);
+ }
+ queue_id = find_next_bit(&tempmap, I40E_MAX_VSI_QP,
+ queue_id + 1);
+ }
+
+error_param:
+ /* send the response to the vf */
+ ret = i40e_vc_send_resp_to_vf(vf, I40E_VIRTCHNL_OP_ENABLE_QUEUES, ret);
+
+ return ret;
+}
+
+/**
+ * i40e_vc_disable_queues_msg
+ * @vf: pointer to the vf info
+ * @msg: pointer to the msg buffer
+ * @msglen: msg length
+ *
+ * called from the vf to disable all or specific
+ * queue(s)
+ **/
+static i40e_status i40e_vc_disable_queues_msg(struct i40e_vf *vf,
+ u8 *msg, u16 msglen)
+{
+ struct i40e_virtchnl_queue_select *vqs =
+ (struct i40e_virtchnl_queue_select *)msg;
+ i40e_status ret = I40E_SUCCESS;
+ struct i40e_pf *pf = vf->pf;
+ u16 vsi_id = vqs->vsi_id;
+ unsigned long tempmap;
+ u16 queue_id;
+
+ if (!test_bit(I40E_VF_STAT_ACTIVE, &vf->vf_states)) {
+ ret = I40E_ERR_PARAM;
+ goto error_param;
+ }
+
+ if (!i40e_vc_isvalid_vsi_id(vf, vqs->vsi_id)) {
+ ret = I40E_ERR_PARAM;
+ goto error_param;
+ }
+
+ if ((0 == vqs->rx_queues) && (0 == vqs->tx_queues)) {
+ ret = I40E_ERR_PARAM;
+ goto error_param;
+ }
+
+ tempmap = vqs->rx_queues;
+ queue_id = find_first_bit(&tempmap, I40E_MAX_VSI_QP);
+ while (queue_id < I40E_MAX_VSI_QP) {
+ if (!i40e_vc_isvalid_queue_id(vf, vsi_id, queue_id)) {
+ ret = I40E_ERR_PARAM;
+ goto error_param;
+ }
+ ret = i40e_ctrl_vsi_rx_queue(vf, vsi_id, queue_id,
+ I40E_QUEUE_CTRL_DISABLE);
+
+ queue_id = find_next_bit(&tempmap, I40E_MAX_VSI_QP,
+ queue_id + 1);
+ }
+
+ tempmap = vqs->tx_queues;
+ queue_id = find_first_bit(&tempmap, I40E_MAX_VSI_QP);
+ while (queue_id < I40E_MAX_VSI_QP) {
+ if (!i40e_vc_isvalid_queue_id(vf, vsi_id, queue_id)) {
+ ret = I40E_ERR_PARAM;
+ goto error_param;
+ }
+ ret = i40e_ctrl_vsi_tx_queue(vf, vsi_id, queue_id,
+ I40E_QUEUE_CTRL_DISABLE);
+
+ queue_id = find_next_bit(&tempmap, I40E_MAX_VSI_QP,
+ queue_id + 1);
+ }
+
+ /* Poll the status register to make sure that the
+ * requested op was completed successfully
+ */
+ udelay(10);
+
+ tempmap = vqs->rx_queues;
+ queue_id = find_first_bit(&tempmap, I40E_MAX_VSI_QP);
+ while (queue_id < I40E_MAX_VSI_QP) {
+ ret = i40e_ctrl_vsi_rx_queue(vf, vsi_id, queue_id,
+ I40E_QUEUE_CTRL_DISABLECHECK);
+ if (ret != I40E_SUCCESS) {
+ dev_err(&pf->pdev->dev,
+ "Queue control check failed on RX queue %d of VSI %d VF %d\n",
+ queue_id, vsi_id, vf->vf_id);
+ }
+ queue_id = find_next_bit(&tempmap, I40E_MAX_VSI_QP,
+ queue_id + 1);
+ }
+
+ tempmap = vqs->tx_queues;
+ queue_id = find_first_bit(&tempmap, I40E_MAX_VSI_QP);
+ while (queue_id < I40E_MAX_VSI_QP) {
+ ret = i40e_ctrl_vsi_tx_queue(vf, vsi_id, queue_id,
+ I40E_QUEUE_CTRL_DISABLECHECK);
+ if (ret != I40E_SUCCESS) {
+ dev_err(&pf->pdev->dev,
+ "Queue control check failed on TX queue %d of VSI %d VF %d\n",
+ queue_id, vsi_id, vf->vf_id);
+ }
+ queue_id = find_next_bit(&tempmap, I40E_MAX_VSI_QP,
+ queue_id + 1);
+ }
+
+error_param:
+ /* send the response to the vf */
+ ret = i40e_vc_send_resp_to_vf(vf, I40E_VIRTCHNL_OP_DISABLE_QUEUES, ret);
+
+ return ret;
+}
+
+/**
+ * i40e_vc_get_stats_msg
+ * @vf: pointer to the vf info
+ * @msg: pointer to the msg buffer
+ * @msglen: msg length
+ *
+ * called from the vf to get vsi stats
+ **/
+static i40e_status i40e_vc_get_stats_msg(struct i40e_vf *vf, u8 *msg,
+ u16 msglen)
+{
+ struct i40e_virtchnl_queue_select *vqs =
+ (struct i40e_virtchnl_queue_select *)msg;
+ i40e_status ret = I40E_SUCCESS;
+ struct i40e_pf *pf = vf->pf;
+ struct i40e_eth_stats stats;
+ struct i40e_vsi *vsi;
+
+ memset(&stats, 0, sizeof(struct i40e_eth_stats));
+
+ if (!test_bit(I40E_VF_STAT_ACTIVE, &vf->vf_states)) {
+ ret = I40E_ERR_PARAM;
+ goto error_param;
+ }
+
+ if (!i40e_vc_isvalid_vsi_id(vf, vqs->vsi_id)) {
+ ret = I40E_ERR_PARAM;
+ goto error_param;
+ }
+
+ vsi = pf->vsi[vqs->vsi_id];
+ if (!vsi) {
+ ret = I40E_ERR_PARAM;
+ goto error_param;
+ }
+ i40e_update_eth_stats(vsi);
+ memcpy(&stats, &vsi->eth_stats, sizeof(struct i40e_eth_stats));
+
+error_param:
+ /* send the response back to the vf */
+ ret = i40e_vc_send_msg_to_vf(vf, I40E_VIRTCHNL_OP_GET_STATS, ret,
+ (u8 *)&stats, sizeof(stats));
+
+ return ret;
+}
+
+/**
+ * i40e_vc_add_mac_addr_msg
+ * @vf: pointer to the vf info
+ * @msg: pointer to the msg buffer
+ * @msglen: msg length
+ *
+ * add guest mac address filter
+ **/
+static i40e_status i40e_vc_add_mac_addr_msg(struct i40e_vf *vf,
+ u8 *msg, u16 msglen)
+{
+ struct i40e_virtchnl_ether_addr_list *al =
+ (struct i40e_virtchnl_ether_addr_list *)msg;
+ i40e_status ret = I40E_SUCCESS;
+ struct i40e_pf *pf = vf->pf;
+ struct i40e_vsi *vsi = NULL;
+ u16 vsi_id = al->vsi_id;
+ int i;
+
+ if (!test_bit(I40E_VF_STAT_ACTIVE, &vf->vf_states) ||
+ !test_bit(I40E_VIRTCHNL_VF_CAP_PRIVILEGE, &vf->vf_caps) ||
+ !i40e_vc_isvalid_vsi_id(vf, vsi_id)) {
+ ret = I40E_ERR_PARAM;
+ goto error_param;
+ }
+
+ for (i = 0; i < al->num_elements; i++) {
+ if (is_broadcast_ether_addr(al->list[i].addr) ||
+ is_zero_ether_addr(al->list[i].addr)) {
+ dev_err(&pf->pdev->dev, "invalid VF MAC addr %pMAC\n",
+ al->list[i].addr);
+ ret = I40E_ERR_PARAM;
+ goto error_param;
+ }
+ }
+ vsi = pf->vsi[vsi_id];
+
+ /* add new addresses to the list */
+ for (i = 0; i < al->num_elements; i++) {
+ if (!i40e_find_mac(vsi, al->list[i].addr, true, false)) {
+ if (i40e_is_vsi_in_vlan(vsi))
+ ret = i40e_put_mac_in_vlan(vsi,
+ al->list[i].addr,
+ true, false);
+ else
+ ret = !!i40e_add_filter(vsi,
+ al->list[i].addr, -1,
+ true, false);
+ }
+
+ if (ret != I40E_SUCCESS) {
+ dev_err(&pf->pdev->dev,
+ "Unable to add VF MAC filter\n");
+ ret = I40E_ERR_PARAM;
+ goto error_param;
+ }
+ }
+
+ /* program the updated filter list */
+ ret = i40e_sync_vsi_filters(vsi);
+ if (I40E_SUCCESS != ret)
+ dev_err(&pf->pdev->dev,
+ "Unable to program VF MAC filters\n");
+
+error_param:
+ /* send the response to the vf */
+ ret = i40e_vc_send_resp_to_vf(vf, I40E_VIRTCHNL_OP_ADD_ETHER_ADDRESS,
+ ret);
+
+ return ret;
+}
+
+/**
+ * i40e_vc_del_mac_addr_msg
+ * @vf: pointer to the vf info
+ * @msg: pointer to the msg buffer
+ * @msglen: msg length
+ *
+ * remove guest mac address filter
+ **/
+static i40e_status i40e_vc_del_mac_addr_msg(struct i40e_vf *vf,
+ u8 *msg, u16 msglen)
+{
+ struct i40e_virtchnl_ether_addr_list *al =
+ (struct i40e_virtchnl_ether_addr_list *)msg;
+ i40e_status ret = I40E_SUCCESS;
+ struct i40e_pf *pf = vf->pf;
+ struct i40e_vsi *vsi = NULL;
+ u16 vsi_id = al->vsi_id;
+ int i;
+
+ if (!test_bit(I40E_VF_STAT_ACTIVE, &vf->vf_states) ||
+ !test_bit(I40E_VIRTCHNL_VF_CAP_PRIVILEGE, &vf->vf_caps) ||
+ !i40e_vc_isvalid_vsi_id(vf, vsi_id)) {
+ ret = I40E_ERR_PARAM;
+ goto error_param;
+ }
+ vsi = pf->vsi[vsi_id];
+
+ /* delete addresses from the list */
+ for (i = 0; i < al->num_elements; i++)
+ i40e_del_filter(vsi, al->list[i].addr,
+ I40E_VLAN_ANY, true, false);
+
+ /* program the updated filter list */
+ ret = i40e_sync_vsi_filters(vsi);
+ if (I40E_SUCCESS != ret)
+ dev_err(&pf->pdev->dev,
+ "Unable to program VF MAC filters\n");
+
+error_param:
+ /* send the response to the vf */
+ ret = i40e_vc_send_resp_to_vf(vf, I40E_VIRTCHNL_OP_DEL_ETHER_ADDRESS,
+ ret);
+
+ return ret;
+}
+
+/**
+ * i40e_vc_add_vlan_msg
+ * @vf: pointer to the vf info
+ * @msg: pointer to the msg buffer
+ * @msglen: msg length
+ *
+ * program guest vlan id
+ **/
+static i40e_status i40e_vc_add_vlan_msg(struct i40e_vf *vf, u8 *msg,
+ u16 msglen)
+{
+ struct i40e_virtchnl_vlan_filter_list *vfl =
+ (struct i40e_virtchnl_vlan_filter_list *)msg;
+ i40e_status ret = I40E_SUCCESS;
+ struct i40e_pf *pf = vf->pf;
+ struct i40e_vsi *vsi = NULL;
+ u16 vsi_id = vfl->vsi_id;
+ int i;
+
+ if (!test_bit(I40E_VF_STAT_ACTIVE, &vf->vf_states) ||
+ !test_bit(I40E_VIRTCHNL_VF_CAP_PRIVILEGE, &vf->vf_caps) ||
+ !i40e_vc_isvalid_vsi_id(vf, vsi_id)) {
+ ret = I40E_ERR_PARAM;
+ goto error_param;
+ }
+
+ for (i = 0; i < vfl->num_elements; i++) {
+ if (vfl->vlan_id[i] > I40E_MAX_VLANID) {
+ ret = I40E_ERR_PARAM;
+ dev_err(&pf->pdev->dev,
+ "invalid VF VLAN id %d\n", vfl->vlan_id[i]);
+ goto error_param;
+ }
+ }
+ vsi = pf->vsi[vsi_id];
+ if (vsi->info.pvid) {
+ ret = I40E_ERR_PARAM;
+ goto error_param;
+ }
+
+ i40e_vlan_stripping_enable(vsi);
+ for (i = 0; i < vfl->num_elements; i++) {
+ /* add new VLAN filter */
+ ret = i40e_vsi_add_vlan(vsi, vfl->vlan_id[i]);
+ if (ret != I40E_SUCCESS) {
+ dev_err(&pf->pdev->dev,
+ "Unable to add VF vlan filter %d, error %d\n",
+ vfl->vlan_id[i], ret);
+ goto error_param;
+ }
+ }
+
+error_param:
+ /* send the response to the vf */
+ ret = i40e_vc_send_resp_to_vf(vf, I40E_VIRTCHNL_OP_ADD_VLAN, ret);
+
+ return ret;
+}
+
+/**
+ * i40e_vc_remove_vlan_msg
+ * @vf: pointer to the vf info
+ * @msg: pointer to the msg buffer
+ * @msglen: msg length
+ *
+ * remove programmed guest vlan id
+ **/
+static i40e_status i40e_vc_remove_vlan_msg(struct i40e_vf *vf, u8 *msg,
+ u16 msglen)
+{
+ struct i40e_virtchnl_vlan_filter_list *vfl =
+ (struct i40e_virtchnl_vlan_filter_list *)msg;
+ i40e_status ret = I40E_SUCCESS;
+ struct i40e_pf *pf = vf->pf;
+ struct i40e_vsi *vsi = NULL;
+ u16 vsi_id = vfl->vsi_id;
+ int i;
+
+ if (!test_bit(I40E_VF_STAT_ACTIVE, &vf->vf_states) ||
+ !test_bit(I40E_VIRTCHNL_VF_CAP_PRIVILEGE, &vf->vf_caps) ||
+ !i40e_vc_isvalid_vsi_id(vf, vsi_id)) {
+ ret = I40E_ERR_PARAM;
+ goto error_param;
+ }
+
+ for (i = 0; i < vfl->num_elements; i++) {
+ if (vfl->vlan_id[i] > I40E_MAX_VLANID) {
+ ret = I40E_ERR_PARAM;
+ goto error_param;
+ }
+ }
+
+ vsi = pf->vsi[vsi_id];
+ if (vsi->info.pvid) {
+ ret = I40E_ERR_PARAM;
+ goto error_param;
+ }
+
+ for (i = 0; i < vfl->num_elements; i++) {
+ ret = i40e_vsi_kill_vlan(vsi, vfl->vlan_id[i]);
+ if (I40E_SUCCESS != ret)
+ dev_err(&pf->pdev->dev,
+ "Unable to delete VF vlan filter %d, error %d\n",
+ vfl->vlan_id[i], ret);
+ }
+
+error_param:
+ /* send the response to the vf */
+ return i40e_vc_send_resp_to_vf(vf, I40E_VIRTCHNL_OP_DEL_VLAN, ret);
+}
+
+/**
+ * i40e_vc_fcoe_msg
+ * @vf: pointer to the vf info
+ * @msg: pointer to the msg buffer
+ * @msglen: msg length
+ *
+ * called from the vf for the fcoe msgs
+ **/
+static i40e_status i40e_vc_fcoe_msg(struct i40e_vf *vf, u8 *msg, u16 msglen)
+{
+ i40e_status ret = I40E_SUCCESS;
+
+ if (!test_bit(I40E_VF_STAT_ACTIVE, &vf->vf_states) ||
+ !test_bit(I40E_VF_STAT_FCOEENA, &vf->vf_states)) {
+ ret = I40E_ERR_PARAM;
+ goto error_param;
+ }
+ ret = I40E_ERR_NOT_IMPLEMENTED;
+
+error_param:
+ /* send the response to the vf */
+ ret = i40e_vc_send_resp_to_vf(vf, I40E_VIRTCHNL_OP_FCOE, ret);
+
+ return ret;
+}
+
+/**
+ * i40e_vc_validate_vf_msg
+ * @vf: pointer to the vf info
+ * @msg: pointer to the msg buffer
+ * @msglen: msg length
+ * @msghndl: msg handle
+ *
+ * validate msg
+ **/
+static i40e_status i40e_vc_validate_vf_msg(struct i40e_vf *vf, u32 v_opcode,
+ u32 v_retval, u8 *msg, u16 msglen)
+{
+ bool err_msg_format = false;
+ int valid_len;
+
+ /* Check if VF is disabled. */
+ if (test_bit(I40E_VF_STAT_DISABLED, &vf->vf_states))
+ return I40E_ERR_PARAM;
+
+ /* Validate message length. */
+ switch (v_opcode) {
+ case I40E_VIRTCHNL_OP_VERSION:
+ valid_len = sizeof(struct i40e_virtchnl_version_info);
+ break;
+ case I40E_VIRTCHNL_OP_RESET_VF:
+ case I40E_VIRTCHNL_OP_GET_VF_RESOURCES:
+ valid_len = 0;
+ break;
+ case I40E_VIRTCHNL_OP_CONFIG_TX_QUEUE:
+ valid_len = sizeof(struct i40e_virtchnl_txq_info);
+ break;
+ case I40E_VIRTCHNL_OP_CONFIG_RX_QUEUE:
+ valid_len = sizeof(struct i40e_virtchnl_rxq_info);
+ break;
+ case I40E_VIRTCHNL_OP_CONFIG_VSI_QUEUES:
+ valid_len = sizeof(struct i40e_virtchnl_vsi_queue_config_info);
+ if (msglen >= valid_len) {
+ struct i40e_virtchnl_vsi_queue_config_info *vqc =
+ (struct i40e_virtchnl_vsi_queue_config_info *)msg;
+ valid_len += (vqc->num_queue_pairs *
+ sizeof(struct
+ i40e_virtchnl_queue_pair_info));
+ if (vqc->num_queue_pairs == 0)
+ err_msg_format = true;
+ }
+ break;
+ case I40E_VIRTCHNL_OP_CONFIG_IRQ_MAP:
+ valid_len = sizeof(struct i40e_virtchnl_irq_map_info);
+ if (msglen >= valid_len) {
+ struct i40e_virtchnl_irq_map_info *vimi =
+ (struct i40e_virtchnl_irq_map_info *)msg;
+ valid_len += (vimi->num_vectors *
+ sizeof(struct i40e_virtchnl_vector_map));
+ if (vimi->num_vectors == 0)
+ err_msg_format = true;
+ }
+ break;
+ case I40E_VIRTCHNL_OP_ENABLE_QUEUES:
+ case I40E_VIRTCHNL_OP_DISABLE_QUEUES:
+ valid_len = sizeof(struct i40e_virtchnl_queue_select);
+ break;
+ case I40E_VIRTCHNL_OP_ADD_ETHER_ADDRESS:
+ case I40E_VIRTCHNL_OP_DEL_ETHER_ADDRESS:
+ valid_len = sizeof(struct i40e_virtchnl_ether_addr_list);
+ if (msglen >= valid_len) {
+ struct i40e_virtchnl_ether_addr_list *veal =
+ (struct i40e_virtchnl_ether_addr_list *)msg;
+ valid_len += veal->num_elements *
+ sizeof(struct i40e_virtchnl_ether_addr);
+ if (veal->num_elements == 0)
+ err_msg_format = true;
+ }
+ break;
+ case I40E_VIRTCHNL_OP_ADD_VLAN:
+ case I40E_VIRTCHNL_OP_DEL_VLAN:
+ valid_len = sizeof(struct i40e_virtchnl_vlan_filter_list);
+ if (msglen >= valid_len) {
+ struct i40e_virtchnl_vlan_filter_list *vfl =
+ (struct i40e_virtchnl_vlan_filter_list *)msg;
+ valid_len += vfl->num_elements * sizeof(u16);
+ if (vfl->num_elements == 0)
+ err_msg_format = true;
+ }
+ break;
+ case I40E_VIRTCHNL_OP_CONFIG_PROMISCUOUS_MODE:
+ valid_len = sizeof(struct i40e_virtchnl_promisc_info);
+ break;
+ case I40E_VIRTCHNL_OP_GET_STATS:
+ valid_len = sizeof(struct i40e_virtchnl_queue_select);
+ break;
+ /* These are always errors coming from the VF. */
+ case I40E_VIRTCHNL_OP_EVENT:
+ case I40E_VIRTCHNL_OP_UNKNOWN:
+ default:
+ return I40E_ERR_PARAM;
+ break;
+ }
+ /* few more checks */
+ if ((valid_len != msglen) || (err_msg_format)) {
+ i40e_vc_send_resp_to_vf(vf, v_opcode, I40E_ERR_PARAM);
+ return I40E_ERR_PARAM;
+ } else {
+ return I40E_SUCCESS;
+ }
+}
+
+/**
+ * i40e_vc_process_vf_msg
+ * @pf: pointer to the pf structure
+ * @vf_id: source vf id
+ * @msg: pointer to the msg buffer
+ * @msglen: msg length
+ * @msghndl: msg handle
+ *
+ * called from the common aeq/arq handler to
+ * process request from vf
+ **/
+i40e_status i40e_vc_process_vf_msg(struct i40e_pf *pf, u16 vf_id,
+ u32 v_opcode, u32 v_retval,
+ u8 *msg, u16 msglen)
+{
+ struct i40e_vf *vf = &(pf->vf[vf_id]);
+ struct i40e_hw *hw = &pf->hw;
+ i40e_status ret;
+
+ pf->vf_aq_requests++;
+ /* perform basic checks on the msg */
+ ret = i40e_vc_validate_vf_msg(vf, v_opcode, v_retval, msg, msglen);
+
+ if (ret != I40E_SUCCESS) {
+ dev_err(&pf->pdev->dev, "invalid message from vf %d\n", vf_id);
+ return ret;
+ }
+ wr32(hw, I40E_VFGEN_RSTAT1(vf_id), I40E_VFR_VFACTIVE);
+ switch (v_opcode) {
+ case I40E_VIRTCHNL_OP_VERSION:
+ ret = i40e_vc_get_version_msg(vf);
+ break;
+ case I40E_VIRTCHNL_OP_GET_VF_RESOURCES:
+ ret = i40e_vc_get_vf_resources_msg(vf);
+ break;
+ case I40E_VIRTCHNL_OP_RESET_VF:
+ ret = i40e_vc_reset_vf_msg(vf);
+ break;
+ case I40E_VIRTCHNL_OP_CONFIG_PROMISCUOUS_MODE:
+ ret = i40e_vc_config_promiscuous_mode_msg(vf,
+ msg, msglen);
+ break;
+ case I40E_VIRTCHNL_OP_CONFIG_VSI_QUEUES:
+ ret = i40e_vc_config_queues_msg(vf, msg, msglen);
+ break;
+ case I40E_VIRTCHNL_OP_CONFIG_IRQ_MAP:
+ ret = i40e_vc_config_irq_map_msg(vf, msg, msglen);
+ break;
+ case I40E_VIRTCHNL_OP_ENABLE_QUEUES:
+ ret = i40e_vc_enable_queues_msg(vf, msg, msglen);
+ break;
+ case I40E_VIRTCHNL_OP_DISABLE_QUEUES:
+ ret = i40e_vc_disable_queues_msg(vf, msg, msglen);
+ break;
+ case I40E_VIRTCHNL_OP_ADD_ETHER_ADDRESS:
+ ret = i40e_vc_add_mac_addr_msg(vf, msg, msglen);
+ break;
+ case I40E_VIRTCHNL_OP_DEL_ETHER_ADDRESS:
+ ret = i40e_vc_del_mac_addr_msg(vf, msg, msglen);
+ break;
+ case I40E_VIRTCHNL_OP_ADD_VLAN:
+ ret = i40e_vc_add_vlan_msg(vf, msg, msglen);
+ break;
+ case I40E_VIRTCHNL_OP_DEL_VLAN:
+ ret = i40e_vc_remove_vlan_msg(vf, msg, msglen);
+ break;
+ case I40E_VIRTCHNL_OP_GET_STATS:
+ ret = i40e_vc_get_stats_msg(vf, msg, msglen);
+ break;
+ case I40E_VIRTCHNL_OP_FCOE:
+ ret = i40e_vc_fcoe_msg(vf, msg, msglen);
+ break;
+ case I40E_VIRTCHNL_OP_UNKNOWN:
+ default:
+ dev_err(&pf->pdev->dev,
+ "Unsupported opcode %d from vf %d\n", v_opcode, vf_id);
+ ret = i40e_vc_send_resp_to_vf(vf, v_opcode,
+ I40E_ERR_NOT_IMPLEMENTED);
+ break;
+ }
+
+ return ret;
+}
+
+/**
+ * i40e_vc_process_vflr_event
+ * @pf: pointer to the pf structure
+ *
+ * called from the vlfr irq handler to
+ * free up vf resources and state variables
+ **/
+i40e_status i40e_vc_process_vflr_event(struct i40e_pf *pf)
+{
+ u32 reg, reg_idx, bit_idx, vf_id;
+ struct i40e_hw *hw = &pf->hw;
+ struct i40e_vf *vf;
+ i40e_status ret;
+
+ if (!test_bit(__I40E_VFLR_EVENT_PENDING, &pf->state))
+ return I40E_SUCCESS;
+
+ clear_bit(__I40E_VFLR_EVENT_PENDING, &pf->state);
+ for (vf_id = 0; vf_id < pf->num_alloc_vfs; vf_id++) {
+ reg_idx = (hw->func_caps.vf_base_id + vf_id) / 32;
+ bit_idx = (hw->func_caps.vf_base_id + vf_id) % 32;
+ /* read GLGEN_VFLRSTAT register to find out the flr vfs */
+ vf = &pf->vf[vf_id];
+ reg = rd32(hw, I40E_GLGEN_VFLRSTAT(reg_idx));
+ if (reg & (1 << bit_idx)) {
+ /* clear the bit in GLGEN_VFLRSTAT */
+ wr32(hw, I40E_GLGEN_VFLRSTAT(reg_idx), (1 << bit_idx));
+
+ ret = i40e_reset_vf(vf, true);
+ if (ret != I40E_SUCCESS)
+ dev_err(&pf->pdev->dev,
+ "Unable to reset the VF %d\n", vf_id);
+ /* free up vf resources to destroy vsi state */
+ ret = i40e_free_vf_res(vf);
+ if (ret != I40E_SUCCESS)
+ dev_err(&pf->pdev->dev,
+ "Failed to free VF resources %d\n",
+ vf_id);
+
+ /* allocate new vf resources with the default state */
+ ret = i40e_alloc_vf_res(vf);
+ if (ret != I40E_SUCCESS)
+ dev_err(&pf->pdev->dev,
+ "Unable to allocate VF resources %d\n",
+ vf_id);
+
+ ret = i40e_enable_vf_mappings(vf);
+ }
+ }
+
+ /* re-enable vflr interrupt cause */
+ reg = rd32(hw, I40E_PFINT_ICR0_ENA);
+ reg |= I40E_PFINT_ICR0_ENA_VFLR_MASK;
+ wr32(hw, I40E_PFINT_ICR0_ENA, reg);
+ i40e_flush(hw);
+
+ return I40E_SUCCESS;
+}
+
+/**
+ * i40e_vc_vf_broadcast
+ * @pf: pointer to the pf structure
+ * @opcode: operation code
+ * @retval: return value
+ * @msg: pointer to the msg buffer
+ * @msglen: msg length
+ *
+ * send a message to all VFs on a given PF
+ **/
+static void i40e_vc_vf_broadcast(struct i40e_pf *pf,
+ enum i40e_virtchnl_ops v_opcode,
+ i40e_status v_retval, u8 *msg,
+ u16 msglen)
+{
+ struct i40e_hw *hw = &pf->hw;
+ struct i40e_vf *vf = pf->vf;
+ int i;
+
+ for (i = 0; i < pf->num_alloc_vfs; i++) {
+ /* Ignore return value on purpose - a given VF may fail, but
+ * we need to keep going and send to all of them
+ */
+ i40e_aq_send_msg_to_vf(hw, vf->vf_id, v_opcode, v_retval,
+ msg, msglen, NULL);
+ vf++;
+ }
+}
+
+/**
+ * i40e_vc_notify_link_state
+ * @pf: pointer to the pf structure
+ *
+ * send a link status message to all VFs on a given PF
+ **/
+void i40e_vc_notify_link_state(struct i40e_pf *pf)
+{
+ struct i40e_virtchnl_pf_event pfe;
+
+ pfe.event = I40E_VIRTCHNL_EVENT_LINK_CHANGE;
+ pfe.severity = I40E_PF_EVENT_SEVERITY_INFO;
+ pfe.event_data.link_event.link_status =
+ pf->hw.phy.link_info.link_info & I40E_AQ_LINK_UP;
+ pfe.event_data.link_event.link_speed = pf->hw.phy.link_info.link_speed;
+
+ i40e_vc_vf_broadcast(pf, I40E_VIRTCHNL_OP_EVENT, I40E_SUCCESS,
+ (u8 *)&pfe, sizeof(struct i40e_virtchnl_pf_event));
+}
+
+/**
+ * i40e_vc_notify_reset
+ * @pf: pointer to the pf structure
+ *
+ * indicate a pending reset to all VFs on a given PF
+ **/
+void i40e_vc_notify_reset(struct i40e_pf *pf)
+{
+ struct i40e_virtchnl_pf_event pfe;
+
+ pfe.event = I40E_VIRTCHNL_EVENT_RESET_IMPENDING;
+ pfe.severity = I40E_PF_EVENT_SEVERITY_CERTAIN_DOOM;
+ i40e_vc_vf_broadcast(pf, I40E_VIRTCHNL_OP_EVENT, I40E_SUCCESS,
+ (u8 *)&pfe, sizeof(struct i40e_virtchnl_pf_event));
+}
+
+/**
+ * i40e_vc_notify_vf_reset
+ * @vf: pointer to the vf structure
+ *
+ * indicate a pending reset to the given VF
+ **/
+void i40e_vc_notify_vf_reset(struct i40e_vf *vf)
+{
+ struct i40e_virtchnl_pf_event pfe;
+
+ pfe.event = I40E_VIRTCHNL_EVENT_RESET_IMPENDING;
+ pfe.severity = I40E_PF_EVENT_SEVERITY_CERTAIN_DOOM;
+ i40e_aq_send_msg_to_vf(&vf->pf->hw, vf->vf_id, I40E_VIRTCHNL_OP_EVENT,
+ I40E_SUCCESS, (u8 *)&pfe,
+ sizeof(struct i40e_virtchnl_pf_event), NULL);
+}
+
+/**
+ * i40e_ndo_set_vf_mac
+ * @netdev: network interface device structure
+ * @vf_id: vf identifier
+ * @mac: mac address
+ *
+ * program vf mac address
+ **/
+int i40e_ndo_set_vf_mac(struct net_device *netdev, int vf_id, u8 *mac)
+{
+ struct i40e_netdev_priv *np = netdev_priv(netdev);
+ struct i40e_vsi *vsi = np->vsi;
+ struct i40e_pf *pf = vsi->back;
+ struct i40e_mac_filter *f;
+ struct i40e_vf *vf;
+ int ret = 0;
+
+ /* validate the request */
+ if (vf_id >= pf->num_alloc_vfs) {
+ dev_err(&pf->pdev->dev,
+ "Invalid VF Identifier %d\n", vf_id);
+ ret = -EINVAL;
+ goto error_param;
+ }
+
+ vf = &(pf->vf[vf_id]);
+ vsi = pf->vsi[vf->lan_vsi_index];
+ if (!test_bit(I40E_VF_STAT_INIT, &vf->vf_states)) {
+ dev_err(&pf->pdev->dev,
+ "Uninitialized VF %d\n", vf_id);
+ ret = -EINVAL;
+ goto error_param;
+ }
+
+ if (!is_valid_ether_addr(mac)) {
+ dev_err(&pf->pdev->dev,
+ "Invalid VF ethernet address\n");
+ ret = -EINVAL;
+ goto error_param;
+ }
+
+ /* delete the temporary mac address */
+ i40e_del_filter(vsi, vf->default_lan_addr.addr, 0, true, false);
+
+ /* add the new mac address */
+ f = i40e_add_filter(vsi, mac, 0, true, false);
+ if (!f) {
+ dev_err(&pf->pdev->dev,
+ "Unable to add VF ucast filter\n");
+ ret = -ENOMEM;
+ goto error_param;
+ }
+
+ dev_info(&pf->pdev->dev, "Setting MAC %pM on VF %d\n", mac, vf_id);
+ /* program mac filter */
+ if (I40E_SUCCESS != i40e_sync_vsi_filters(vsi)) {
+ dev_err(&pf->pdev->dev, "Unable to program ucast filters\n");
+ ret = -EIO;
+ goto error_param;
+ }
+ memcpy(vf->default_lan_addr.addr, mac, ETH_ALEN);
+ dev_info(&pf->pdev->dev, "Reload the VF driver to make this change effective.\n");
+ ret = 0;
+
+error_param:
+ return ret;
+}
+
+/**
+ * i40e_ndo_set_vf_port_vlan
+ * @netdev: network interface device structure
+ * @vf_id: vf identifier
+ * @vlan_id: mac address
+ * @qos: priority setting
+ *
+ * program vf vlan id and/or qos
+ **/
+int i40e_ndo_set_vf_port_vlan(struct net_device *netdev,
+ int vf_id, u16 vlan_id, u8 qos)
+{
+ struct i40e_netdev_priv *np = netdev_priv(netdev);
+ struct i40e_pf *pf = np->vsi->back;
+ int ret = I40E_SUCCESS;
+ struct i40e_vsi *vsi;
+ struct i40e_vf *vf;
+
+ /* validate the request */
+ if (vf_id >= pf->num_alloc_vfs) {
+ dev_err(&pf->pdev->dev, "Invalid VF Identifier %d\n", vf_id);
+ ret = -EINVAL;
+ goto error_pvid;
+ }
+
+ if ((vlan_id > I40E_MAX_VLANID) || (qos > 7)) {
+ dev_err(&pf->pdev->dev, "Invalid VF Parameters\n");
+ ret = -EINVAL;
+ goto error_pvid;
+ }
+
+ vf = &(pf->vf[vf_id]);
+ vsi = pf->vsi[vf->lan_vsi_index];
+ if (!test_bit(I40E_VF_STAT_INIT, &vf->vf_states)) {
+ dev_err(&pf->pdev->dev, "Uninitialized VF %d\n", vf_id);
+ ret = -EINVAL;
+ goto error_pvid;
+ }
+
+ if (vsi->info.pvid) {
+ /* kill old VLAN */
+ ret = i40e_vsi_kill_vlan(vsi, (le16_to_cpu(vsi->info.pvid) &
+ VLAN_VID_MASK));
+ if (ret != I40E_SUCCESS) {
+ dev_info(&vsi->back->pdev->dev,
+ "remove VLAN failed, ret=%d, aq_err=%d\n",
+ ret, pf->hw.aq.asq_last_status);
+ }
+ }
+ if (vlan_id || qos)
+ ret = i40e_vsi_add_pvid(vsi,
+ vlan_id | (qos << I40E_VLAN_PRIORITY_SHIFT));
+ else
+ i40e_vlan_stripping_disable(vsi);
+
+ if (vlan_id) {
+ dev_info(&pf->pdev->dev, "Setting VLAN %d, QOS 0x%x on VF %d\n",
+ vlan_id, qos, vf_id);
+
+ /* add new VLAN filter */
+ ret = i40e_vsi_add_vlan(vsi, vlan_id);
+ if (ret != I40E_SUCCESS) {
+ dev_info(&vsi->back->pdev->dev,
+ "add VF VLAN failed, ret=%d aq_err=%d\n", ret,
+ vsi->back->hw.aq.asq_last_status);
+ goto error_pvid;
+ }
+ }
+
+ if (ret != I40E_SUCCESS) {
+ dev_err(&pf->pdev->dev, "Unable to update VF vsi context\n");
+ ret = -EIO;
+ goto error_pvid;
+ }
+ ret = 0;
+
+error_pvid:
+ return ret;
+}
+
+/**
+ * i40e_ndo_set_vf_bw
+ * @netdev: network interface device structure
+ * @vf_id: vf identifier
+ * @tx_rate: tx rate
+ *
+ * configure vf tx rate
+ **/
+int i40e_ndo_set_vf_bw(struct net_device *netdev, int vf_id, int tx_rate)
+{
+ return -EOPNOTSUPP;
+}
+
+/**
+ * i40e_ndo_get_vf_config
+ * @netdev: network interface device structure
+ * @vf_id: vf identifier
+ * @ivi: vf configuration structure
+ *
+ * return vf configuration
+ **/
+int i40e_ndo_get_vf_config(struct net_device *netdev,
+ int vf_id, struct ifla_vf_info *ivi)
+{
+ struct i40e_netdev_priv *np = netdev_priv(netdev);
+ struct i40e_mac_filter *f, *ftmp;
+ struct i40e_vsi *vsi = np->vsi;
+ struct i40e_pf *pf = vsi->back;
+ struct i40e_vf *vf;
+ int ret = 0;
+
+ /* validate the request */
+ if (vf_id >= pf->num_alloc_vfs) {
+ dev_err(&pf->pdev->dev, "Invalid VF Identifier %d\n", vf_id);
+ ret = -EINVAL;
+ goto error_param;
+ }
+
+ vf = &(pf->vf[vf_id]);
+ /* first vsi is always the LAN vsi */
+ vsi = pf->vsi[vf->lan_vsi_index];
+ if (!test_bit(I40E_VF_STAT_INIT, &vf->vf_states)) {
+ dev_err(&pf->pdev->dev, "Uninitialized VF %d\n", vf_id);
+ ret = -EINVAL;
+ goto error_param;
+ }
+
+ ivi->vf = vf_id;
+
+ /* first entry of the list is the default ethernet address */
+ list_for_each_entry_safe(f, ftmp, &vsi->mac_filter_list, list) {
+ memcpy(&ivi->mac, f->macaddr, I40E_ETH_LENGTH_OF_ADDRESS);
+ break;
+ }
+
+ ivi->tx_rate = 0;
+ ivi->vlan = le16_to_cpu(vsi->info.pvid) & I40E_VLAN_MASK;
+ ivi->qos = (le16_to_cpu(vsi->info.pvid) & I40E_PRIORITY_MASK) >>
+ I40E_VLAN_PRIORITY_SHIFT;
+ ret = 0;
+
+error_param:
+ return ret;
+}
diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h
new file mode 100644
index 0000000..1cc3108
--- /dev/null
+++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h
@@ -0,0 +1,121 @@
+/*******************************************************************************
+
+ Intel Ethernet Controller XL710 Family Linux Driver
+ Copyright(c) 2013 Intel Corporation.
+
+ This program is free software; you can redistribute it and/or modify it
+ under the terms and conditions of the GNU General Public License,
+ version 2, as published by the Free Software Foundation.
+
+ This program is distributed in the hope it will be useful, but WITHOUT
+ ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ more details.
+
+ You should have received a copy of the GNU General Public License along with
+ this program; if not, write to the Free Software Foundation, Inc.,
+ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
+
+ The full GNU General Public License is included in this distribution in
+ the file called "COPYING".
+
+ Contact Information:
+ e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
+ Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
+
+*******************************************************************************/
+
+#ifndef _I40E_VIRTCHNL_PF_H_
+#define _I40E_VIRTCHNL_PF_H_
+
+#include "i40e.h"
+
+#define I40E_MAX_MACVLAN_FILTERS 256
+#define I40E_MAX_VLAN_FILTERS 256
+#define I40E_MAX_VLANID 4095
+
+#define I40E_VIRTCHNL_SUPPORTED_QTYPES 2
+
+#define I40E_DEFAULT_NUM_MDD_EVENTS_ALLOWED 3
+#define I40E_DEFAULT_NUM_INVALID_MSGS_ALLOWED 10
+
+#define I40E_VLAN_PRIORITY_SHIFT 12
+#define I40E_VLAN_MASK 0xFFF
+#define I40E_PRIORITY_MASK 0x7000
+
+/* Various queue ctrls */
+enum i40e_queue_ctrl {
+ I40E_QUEUE_CTRL_UNKNOWN = 0,
+ I40E_QUEUE_CTRL_ENABLE,
+ I40E_QUEUE_CTRL_ENABLECHECK,
+ I40E_QUEUE_CTRL_DISABLE,
+ I40E_QUEUE_CTRL_DISABLECHECK,
+ I40E_QUEUE_CTRL_FASTDISABLE,
+ I40E_QUEUE_CTRL_FASTDISABLECHECK,
+};
+
+/* VF states */
+enum i40e_vf_states {
+ I40E_VF_STAT_INIT = 0,
+ I40E_VF_STAT_ACTIVE,
+ I40E_VF_STAT_FCOEENA,
+ I40E_VF_STAT_DISABLED,
+};
+
+/* VF capabilities */
+enum i40e_vf_capabilities {
+ I40E_VIRTCHNL_VF_CAP_PRIVILEGE = 0,
+ I40E_VIRTCHNL_VF_CAP_L2,
+};
+
+/* VF information structure */
+struct i40e_vf {
+ struct i40e_pf *pf;
+
+ /* vf id in the pf space */
+ u16 vf_id;
+ /* all vf vsis connect to the same parent */
+ enum i40e_switch_element_types parent_type;
+
+ /* vf Port Extender (PE) stag if used */
+ u16 stag;
+
+ struct i40e_virtchnl_ether_addr default_lan_addr;
+ struct i40e_virtchnl_ether_addr default_fcoe_addr;
+
+ /* VSI indices - actual VSI pointers are maintained in the PF structure
+ * When assigned, these will be non-zero, because VSI 0 is always
+ * the main LAN VSI for the PF.
+ */
+ u8 lan_vsi_index; /* index into PF struct */
+ u8 lan_vsi_id; /* ID as used by firmware */
+
+ u8 num_queue_pairs; /* num of qps assigned to vf vsis */
+ u64 num_mdd_events; /* num of mdd events detected */
+ u64 num_invalid_msgs; /* num of malformed or invalid msgs detected */
+ u64 num_valid_msgs; /* num of valid msgs detected */
+
+ unsigned long vf_caps; /* vf's adv. capabilities */
+ unsigned long vf_states; /* vf's runtime states */
+};
+
+i40e_status i40e_free_vfs(struct i40e_pf *pf);
+int i40e_pci_sriov_configure(struct pci_dev *dev, int num_vfs);
+i40e_status i40e_vc_process_vf_msg(struct i40e_pf *pf, u16 vf_id, u32 v_opcode,
+ u32 v_retval, u8 *msg, u16 msglen);
+i40e_status i40e_vc_process_vflr_event(struct i40e_pf *pf);
+i40e_status i40e_vc_process_mdd_event(struct i40e_pf *pf);
+i40e_status i40e_reset_vf(struct i40e_vf *vf, bool flr);
+void i40e_vc_notify_vf_reset(struct i40e_vf *vf);
+
+/* vf configuration related iplink handlers */
+int i40e_ndo_set_vf_mac(struct net_device *netdev, int vf_id, u8 *mac);
+int i40e_ndo_set_vf_port_vlan(struct net_device *netdev,
+ int vf_id, u16 vlan_id, u8 qos);
+int i40e_ndo_set_vf_bw(struct net_device *netdev, int vf_id, int tx_rate);
+int i40e_ndo_get_vf_config(struct net_device *netdev,
+ int vf_id, struct ifla_vf_info *ivi);
+void i40e_vc_notify_link_state(struct i40e_pf *pf);
+void i40e_vc_notify_reset(struct i40e_pf *pf);
+
+#endif /* _I40E_VIRTCHNL_PF_H_ */
--
1.8.3.1
------------------------------------------------------------------------------
Learn the latest--Visual Studio 2012, SharePoint 2013, SQL 2012, more!
Discover the easy way to master current and previous Microsoft technologies
and advance your career. Get an incredible 1,500+ hours of step-by-step
tutorial videos with LearnDevNow. Subscribe today and save!
http://pubads.g.doubleclick.net/gampad/clk?id=58041391&iu=/4140/ostg.clktrk
_______________________________________________
E1000-devel mailing list
E1000-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/e1000-devel
To learn more about Intel® Ethernet, visit http://communities.intel.com/community/wired
^ permalink raw reply related
* [net-next v4 3/8] i40e: driver ethtool core
From: Jeff Kirsher @ 2013-09-04 23:17 UTC (permalink / raw)
To: davem; +Cc: e1000-devel, netdev, Jesse Brandeburg, gospo, sassmann
In-Reply-To: <1378336629-24224-1-git-send-email-jeffrey.t.kirsher@intel.com>
From: Jesse Brandeburg <jesse.brandeburg@intel.com>
This patch contains the ethtool interface and implementation.
The goal in this patch series is minimal functionality while not
including much in the way of "set support."
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Tested-by: Kavindya Deegala <kavindya.s.deegala@intel.com>
CC: PJ Waskiewicz <peter.p.waskiewicz.jr@intel.com>
CC: e1000-devel@lists.sourceforge.net
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
v1: this is the initial submittal
v2: changes due to first round of netdev feedback, including:
remove unimplemented ethtool skeleton functions
as per bwh, use better stats naming like tx-N
don't report stats we won't change
remove buggy static variable
don't force n_len
don't double buffer in get_drvinfo
use strlcpy
don't use hacks for old kernels
use some simplifying macros
use rtnl_link_stats64
report correct speeds
v3:
more tabs/spaces fixes
void function needs no return
variable i could be init'd in declaration
v4:
indentation/whitespace fixes
__func__ removal
use tcp flow spec for filter programming
---
drivers/net/ethernet/intel/i40e/i40e_ethtool.c | 1451 ++++++++++++++++++++++++
1 file changed, 1451 insertions(+)
create mode 100644 drivers/net/ethernet/intel/i40e/i40e_ethtool.c
diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
new file mode 100644
index 0000000..e9a5181
--- /dev/null
+++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
@@ -0,0 +1,1451 @@
+/*******************************************************************************
+
+ Intel Ethernet Controller XL710 Family Linux Driver
+ Copyright(c) 2013 Intel Corporation.
+
+ This program is free software; you can redistribute it and/or modify it
+ under the terms and conditions of the GNU General Public License,
+ version 2, as published by the Free Software Foundation.
+
+ This program is distributed in the hope it will be useful, but WITHOUT
+ ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ more details.
+
+ You should have received a copy of the GNU General Public License along with
+ this program; if not, write to the Free Software Foundation, Inc.,
+ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
+
+ The full GNU General Public License is included in this distribution in
+ the file called "COPYING".
+
+ Contact Information:
+ e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
+ Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
+
+*******************************************************************************/
+
+/* ethtool support for i40e */
+
+#include "i40e.h"
+#include "i40e_diag.h"
+
+struct i40e_stats {
+ char stat_string[ETH_GSTRING_LEN];
+ int sizeof_stat;
+ int stat_offset;
+};
+
+#define I40E_STAT(_type, _name, _stat) { \
+ .stat_string = _name, \
+ .sizeof_stat = FIELD_SIZEOF(_type, _stat), \
+ .stat_offset = offsetof(_type, _stat) \
+}
+#define I40E_NETDEV_STAT(_net_stat) \
+ I40E_STAT(struct net_device_stats, #_net_stat, _net_stat)
+#define I40E_PF_STAT(_name, _stat) \
+ I40E_STAT(struct i40e_pf, _name, _stat)
+#define I40E_VSI_STAT(_name, _stat) \
+ I40E_STAT(struct i40e_vsi, _name, _stat)
+
+static const struct i40e_stats i40e_gstrings_net_stats[] = {
+ I40E_NETDEV_STAT(rx_packets),
+ I40E_NETDEV_STAT(tx_packets),
+ I40E_NETDEV_STAT(rx_bytes),
+ I40E_NETDEV_STAT(tx_bytes),
+ I40E_NETDEV_STAT(rx_errors),
+ I40E_NETDEV_STAT(tx_errors),
+ I40E_NETDEV_STAT(rx_dropped),
+ I40E_NETDEV_STAT(tx_dropped),
+ I40E_NETDEV_STAT(multicast),
+ I40E_NETDEV_STAT(collisions),
+ I40E_NETDEV_STAT(rx_length_errors),
+ I40E_NETDEV_STAT(rx_crc_errors),
+};
+
+/* These PF_STATs might look like duplicates of some NETDEV_STATs,
+ * but they are separate. This device supports Virtualization, and
+ * as such might have several netdevs supporting VMDq and FCoE going
+ * through a single port. The NETDEV_STATs are for individual netdevs
+ * seen at the top of the stack, and the PF_STATs are for the physical
+ * function at the bottom of the stack hosting those netdevs.
+ *
+ * The PF_STATs are appended to the netdev stats only when ethtool -S
+ * is queried on the base PF netdev, not on the VMDq or FCoE netdev.
+ */
+static struct i40e_stats i40e_gstrings_stats[] = {
+ I40E_PF_STAT("rx_bytes", stats.eth.rx_bytes),
+ I40E_PF_STAT("tx_bytes", stats.eth.tx_bytes),
+ I40E_PF_STAT("rx_errors", stats.eth.rx_errors),
+ I40E_PF_STAT("tx_errors", stats.eth.tx_errors),
+ I40E_PF_STAT("rx_dropped", stats.eth.rx_discards),
+ I40E_PF_STAT("tx_dropped", stats.eth.tx_discards),
+ I40E_PF_STAT("tx_dropped_link_down", stats.tx_dropped_link_down),
+ I40E_PF_STAT("crc_errors", stats.crc_errors),
+ I40E_PF_STAT("illegal_bytes", stats.illegal_bytes),
+ I40E_PF_STAT("mac_local_faults", stats.mac_local_faults),
+ I40E_PF_STAT("mac_remote_faults", stats.mac_remote_faults),
+ I40E_PF_STAT("rx_length_errors", stats.rx_length_errors),
+ I40E_PF_STAT("link_xon_rx", stats.link_xon_rx),
+ I40E_PF_STAT("link_xoff_rx", stats.link_xoff_rx),
+ I40E_PF_STAT("link_xon_tx", stats.link_xon_tx),
+ I40E_PF_STAT("link_xoff_tx", stats.link_xoff_tx),
+ I40E_PF_STAT("rx_size_64", stats.rx_size_64),
+ I40E_PF_STAT("rx_size_127", stats.rx_size_127),
+ I40E_PF_STAT("rx_size_255", stats.rx_size_255),
+ I40E_PF_STAT("rx_size_511", stats.rx_size_511),
+ I40E_PF_STAT("rx_size_1023", stats.rx_size_1023),
+ I40E_PF_STAT("rx_size_1522", stats.rx_size_1522),
+ I40E_PF_STAT("rx_size_big", stats.rx_size_big),
+ I40E_PF_STAT("tx_size_64", stats.tx_size_64),
+ I40E_PF_STAT("tx_size_127", stats.tx_size_127),
+ I40E_PF_STAT("tx_size_255", stats.tx_size_255),
+ I40E_PF_STAT("tx_size_511", stats.tx_size_511),
+ I40E_PF_STAT("tx_size_1023", stats.tx_size_1023),
+ I40E_PF_STAT("tx_size_1522", stats.tx_size_1522),
+ I40E_PF_STAT("tx_size_big", stats.tx_size_big),
+ I40E_PF_STAT("rx_undersize", stats.rx_undersize),
+ I40E_PF_STAT("rx_fragments", stats.rx_fragments),
+ I40E_PF_STAT("rx_oversize", stats.rx_oversize),
+ I40E_PF_STAT("rx_jabber", stats.rx_jabber),
+ I40E_PF_STAT("VF_admin_queue_requests", vf_aq_requests),
+};
+
+#define I40E_QUEUE_STATS_LEN(n) \
+ ((((struct i40e_netdev_priv *)netdev_priv((n)))->vsi->num_queue_pairs + \
+ ((struct i40e_netdev_priv *)netdev_priv((n)))->vsi->num_queue_pairs) * 2)
+#define I40E_GLOBAL_STATS_LEN ARRAY_SIZE(i40e_gstrings_stats)
+#define I40E_NETDEV_STATS_LEN ARRAY_SIZE(i40e_gstrings_net_stats)
+#define I40E_VSI_STATS_LEN(n) (I40E_NETDEV_STATS_LEN + \
+ I40E_QUEUE_STATS_LEN((n)))
+#define I40E_PFC_STATS_LEN ( \
+ (FIELD_SIZEOF(struct i40e_pf, stats.priority_xoff_rx) + \
+ FIELD_SIZEOF(struct i40e_pf, stats.priority_xon_rx) + \
+ FIELD_SIZEOF(struct i40e_pf, stats.priority_xoff_tx) + \
+ FIELD_SIZEOF(struct i40e_pf, stats.priority_xon_tx) + \
+ FIELD_SIZEOF(struct i40e_pf, stats.priority_xon_2_xoff)) \
+ / sizeof(u64))
+#define I40E_PF_STATS_LEN(n) (I40E_GLOBAL_STATS_LEN + \
+ I40E_PFC_STATS_LEN + \
+ I40E_VSI_STATS_LEN((n)))
+
+enum i40e_ethtool_test_id {
+ I40E_ETH_TEST_REG = 0,
+ I40E_ETH_TEST_EEPROM,
+ I40E_ETH_TEST_INTR,
+ I40E_ETH_TEST_LOOPBACK,
+ I40E_ETH_TEST_LINK,
+};
+
+static const char i40e_gstrings_test[][ETH_GSTRING_LEN] = {
+ "Register test (offline)",
+ "Eeprom test (offline)",
+ "Interrupt test (offline)",
+ "Loopback test (offline)",
+ "Link test (on/offline)"
+};
+
+#define I40E_TEST_LEN (sizeof(i40e_gstrings_test) / ETH_GSTRING_LEN)
+
+/**
+ * i40e_get_settings - Get Link Speed and Duplex settings
+ * @netdev: network interface device structure
+ * @ecmd: ethtool command
+ *
+ * Reports speed/duplex settings based on media_type
+ **/
+static int i40e_get_settings(struct net_device *netdev,
+ struct ethtool_cmd *ecmd)
+{
+ struct i40e_netdev_priv *np = netdev_priv(netdev);
+ struct i40e_pf *pf = np->vsi->back;
+ struct i40e_hw *hw = &pf->hw;
+ struct i40e_link_status *hw_link_info = &hw->phy.link_info;
+ bool link_up = hw_link_info->link_info & I40E_AQ_LINK_UP;
+ u32 link_speed = hw_link_info->link_speed;
+
+ /* hardware is either in 40G mode or 10G mode
+ * NOTE: this section initializes supported and advertising
+ */
+ switch (hw_link_info->phy_type) {
+ case I40E_PHY_TYPE_40GBASE_CR4:
+ case I40E_PHY_TYPE_40GBASE_CR4_CU:
+ ecmd->supported = SUPPORTED_40000baseCR4_Full;
+ ecmd->advertising = ADVERTISED_40000baseCR4_Full;
+ break;
+ case I40E_PHY_TYPE_40GBASE_KR4:
+ ecmd->supported = SUPPORTED_40000baseKR4_Full;
+ ecmd->advertising = ADVERTISED_40000baseKR4_Full;
+ break;
+ case I40E_PHY_TYPE_40GBASE_SR4:
+ ecmd->supported = SUPPORTED_40000baseSR4_Full;
+ ecmd->advertising = ADVERTISED_40000baseSR4_Full;
+ break;
+ case I40E_PHY_TYPE_40GBASE_LR4:
+ ecmd->supported = SUPPORTED_40000baseLR4_Full;
+ ecmd->advertising = ADVERTISED_40000baseLR4_Full;
+ break;
+ case I40E_PHY_TYPE_10GBASE_KX4:
+ ecmd->supported = SUPPORTED_10000baseKX4_Full;
+ ecmd->advertising = ADVERTISED_10000baseKX4_Full;
+ break;
+ case I40E_PHY_TYPE_10GBASE_KR:
+ ecmd->supported = SUPPORTED_10000baseKR_Full;
+ ecmd->advertising = ADVERTISED_10000baseKR_Full;
+ break;
+ case I40E_PHY_TYPE_10GBASE_T:
+ default:
+ ecmd->supported = SUPPORTED_10000baseT_Full;
+ ecmd->advertising = ADVERTISED_10000baseT_Full;
+ break;
+ }
+
+ /* for now just say autoneg all the time */
+ ecmd->supported |= SUPPORTED_Autoneg;
+
+ if (hw->phy.media_type == I40E_MEDIA_TYPE_BACKPLANE) {
+ ecmd->supported |= SUPPORTED_Backplane;
+ ecmd->advertising |= ADVERTISED_Backplane;
+ ecmd->port = PORT_NONE;
+ } else if (hw->phy.media_type == I40E_MEDIA_TYPE_BASET) {
+ ecmd->supported |= SUPPORTED_TP;
+ ecmd->advertising |= ADVERTISED_TP;
+ ecmd->port = PORT_TP;
+ } else {
+ ecmd->supported |= SUPPORTED_FIBRE;
+ ecmd->advertising |= ADVERTISED_FIBRE;
+ ecmd->port = PORT_FIBRE;
+ }
+
+ ecmd->transceiver = XCVR_EXTERNAL;
+
+ if (link_up) {
+ switch (link_speed) {
+ case I40E_LINK_SPEED_40GB:
+ /* need a SPEED_40000 in ethtool.h */
+ ecmd->speed = 40000;
+ break;
+ case I40E_LINK_SPEED_10GB:
+ ecmd->speed = SPEED_10000;
+ break;
+ default:
+ break;
+ }
+ ecmd->duplex = DUPLEX_FULL;
+ } else {
+ ecmd->speed = SPEED_UNKNOWN;
+ ecmd->duplex = DUPLEX_UNKNOWN;
+ }
+
+ return 0;
+}
+
+
+/**
+ * i40e_get_pauseparam - Get Flow Control status
+ * Return tx/rx-pause status
+ **/
+static void i40e_get_pauseparam(struct net_device *netdev,
+ struct ethtool_pauseparam *pause)
+{
+ struct i40e_netdev_priv *np = netdev_priv(netdev);
+ struct i40e_pf *pf = np->vsi->back;
+ struct i40e_hw *hw = &pf->hw;
+ struct i40e_link_status *hw_link_info = &hw->phy.link_info;
+
+ pause->autoneg =
+ ((hw_link_info->an_info & I40E_AQ_AN_COMPLETED) ?
+ AUTONEG_ENABLE : AUTONEG_DISABLE);
+
+ pause->rx_pause = 0;
+ pause->tx_pause = 0;
+ if (hw_link_info->an_info & I40E_AQ_LINK_PAUSE_RX)
+ pause->rx_pause = 1;
+ if (hw_link_info->an_info & I40E_AQ_LINK_PAUSE_TX)
+ pause->tx_pause = 1;
+}
+
+static u32 i40e_get_msglevel(struct net_device *netdev)
+{
+ struct i40e_netdev_priv *np = netdev_priv(netdev);
+ struct i40e_pf *pf = np->vsi->back;
+
+ return pf->msg_enable;
+}
+
+static void i40e_set_msglevel(struct net_device *netdev, u32 data)
+{
+ struct i40e_netdev_priv *np = netdev_priv(netdev);
+ struct i40e_pf *pf = np->vsi->back;
+
+ if (I40E_DEBUG_USER & data)
+ pf->hw.debug_mask = data;
+ pf->msg_enable = data;
+}
+
+static int i40e_get_regs_len(struct net_device *netdev)
+{
+ int reg_count = 0;
+ int i;
+
+ for (i = 0; i40e_reg_list[i].offset != 0; i++)
+ reg_count += i40e_reg_list[i].elements;
+
+ return reg_count * sizeof(u32);
+}
+
+static void i40e_get_regs(struct net_device *netdev, struct ethtool_regs *regs,
+ void *p)
+{
+ struct i40e_netdev_priv *np = netdev_priv(netdev);
+ struct i40e_pf *pf = np->vsi->back;
+ struct i40e_hw *hw = &pf->hw;
+ u32 *reg_buf = p;
+ int i, j, ri;
+ u32 reg;
+
+ /* Tell ethtool which driver-version-specific regs output we have.
+ *
+ * At some point, if we have ethtool doing special formatting of
+ * this data, it will rely on this version number to know how to
+ * interpret things. Hence, this needs to be updated if/when the
+ * diags register table is changed.
+ */
+ regs->version = 1;
+
+ /* loop through the diags reg table for what to print */
+ ri = 0;
+ for (i = 0; i40e_reg_list[i].offset != 0; i++) {
+ for (j = 0; j < i40e_reg_list[i].elements; j++) {
+ reg = i40e_reg_list[i].offset
+ + (j * i40e_reg_list[i].stride);
+ reg_buf[ri++] = rd32(hw, reg);
+ }
+ }
+
+}
+
+static int i40e_get_eeprom(struct net_device *netdev,
+ struct ethtool_eeprom *eeprom, u8 *bytes)
+{
+ struct i40e_netdev_priv *np = netdev_priv(netdev);
+ struct i40e_hw *hw = &np->vsi->back->hw;
+ int first_word, last_word;
+ u16 i, eeprom_len;
+ u16 *eeprom_buff;
+ int ret_val = 0;
+
+ if (eeprom->len == 0)
+ return -EINVAL;
+
+ eeprom->magic = hw->vendor_id | (hw->device_id << 16);
+
+ first_word = eeprom->offset >> 1;
+ last_word = (eeprom->offset + eeprom->len - 1) >> 1;
+ eeprom_len = last_word - first_word + 1;
+
+ eeprom_buff = kmalloc(sizeof(u16) * eeprom_len, GFP_KERNEL);
+ if (!eeprom_buff)
+ return -ENOMEM;
+
+ ret_val = i40e_read_nvm_buffer(hw, first_word, &eeprom_len,
+ eeprom_buff);
+ if (eeprom_len == 0) {
+ kfree(eeprom_buff);
+ return -EACCES;
+ }
+
+ /* Device's eeprom is always little-endian, word addressable */
+ for (i = 0; i < eeprom_len; i++)
+ le16_to_cpus(&eeprom_buff[i]);
+
+ memcpy(bytes, (u8 *)eeprom_buff + (eeprom->offset & 1), eeprom->len);
+ kfree(eeprom_buff);
+
+ return ret_val;
+}
+
+static int i40e_get_eeprom_len(struct net_device *netdev)
+{
+ struct i40e_netdev_priv *np = netdev_priv(netdev);
+ struct i40e_hw *hw = &np->vsi->back->hw;
+
+ return hw->nvm.sr_size * 2;
+}
+
+static void i40e_get_drvinfo(struct net_device *netdev,
+ struct ethtool_drvinfo *drvinfo)
+{
+ struct i40e_netdev_priv *np = netdev_priv(netdev);
+ struct i40e_vsi *vsi = np->vsi;
+ struct i40e_pf *pf = vsi->back;
+
+ strlcpy(drvinfo->driver, i40e_driver_name, sizeof(drvinfo->driver));
+ strlcpy(drvinfo->version, i40e_driver_version_str,
+ sizeof(drvinfo->version));
+ snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version),
+ "%s", i40e_fw_version_str(&pf->hw));
+ strlcpy(drvinfo->bus_info, pci_name(pf->pdev),
+ sizeof(drvinfo->bus_info));
+}
+
+static void i40e_get_ringparam(struct net_device *netdev,
+ struct ethtool_ringparam *ring)
+{
+ struct i40e_netdev_priv *np = netdev_priv(netdev);
+ struct i40e_pf *pf = np->vsi->back;
+ struct i40e_vsi *vsi = pf->vsi[pf->lan_vsi];
+
+ ring->rx_max_pending = I40E_MAX_NUM_DESCRIPTORS;
+ ring->tx_max_pending = I40E_MAX_NUM_DESCRIPTORS;
+ ring->rx_mini_max_pending = 0;
+ ring->rx_jumbo_max_pending = 0;
+ ring->rx_pending = vsi->rx_rings[0].count;
+ ring->tx_pending = vsi->tx_rings[0].count;
+ ring->rx_mini_pending = 0;
+ ring->rx_jumbo_pending = 0;
+}
+
+static int i40e_set_ringparam(struct net_device *netdev,
+ struct ethtool_ringparam *ring)
+{
+ struct i40e_ring *tx_rings = NULL, *rx_rings = NULL;
+ struct i40e_netdev_priv *np = netdev_priv(netdev);
+ struct i40e_vsi *vsi = np->vsi;
+ struct i40e_pf *pf = vsi->back;
+ u32 new_rx_count, new_tx_count;
+ int i, err = 0;
+
+ if ((ring->rx_mini_pending) || (ring->rx_jumbo_pending))
+ return -EINVAL;
+
+ new_tx_count = clamp_t(u32, ring->tx_pending,
+ I40E_MIN_NUM_DESCRIPTORS,
+ I40E_MAX_NUM_DESCRIPTORS);
+ new_tx_count = ALIGN(new_tx_count, I40E_REQ_DESCRIPTOR_MULTIPLE);
+
+ new_rx_count = clamp_t(u32, ring->rx_pending,
+ I40E_MIN_NUM_DESCRIPTORS,
+ I40E_MAX_NUM_DESCRIPTORS);
+ new_rx_count = ALIGN(new_rx_count, I40E_REQ_DESCRIPTOR_MULTIPLE);
+
+ /* if nothing to do return success */
+ if ((new_tx_count == vsi->tx_rings[0].count) &&
+ (new_rx_count == vsi->rx_rings[0].count))
+ return 0;
+
+ while (test_and_set_bit(__I40E_CONFIG_BUSY, &pf->state))
+ usleep_range(1000, 2000);
+
+ if (!netif_running(vsi->netdev)) {
+ /* simple case - set for the next time the netdev is started */
+ for (i = 0; i < vsi->num_queue_pairs; i++) {
+ vsi->tx_rings[i].count = new_tx_count;
+ vsi->rx_rings[i].count = new_rx_count;
+ }
+ goto done;
+ }
+
+ /* We can't just free everything and then setup again,
+ * because the ISRs in MSI-X mode get passed pointers
+ * to the Tx and Rx ring structs.
+ */
+
+ /* alloc updated Tx resources */
+ if (new_tx_count != vsi->tx_rings[0].count) {
+ netdev_info(netdev,
+ "Changing Tx descriptor count from %d to %d.\n",
+ vsi->tx_rings[0].count, new_tx_count);
+ tx_rings = kcalloc(vsi->alloc_queue_pairs,
+ sizeof(struct i40e_ring), GFP_KERNEL);
+ if (!tx_rings) {
+ err = -ENOMEM;
+ goto done;
+ }
+
+ for (i = 0; i < vsi->num_queue_pairs; i++) {
+ /* clone ring and setup updated count */
+ tx_rings[i] = vsi->tx_rings[i];
+ tx_rings[i].count = new_tx_count;
+ err = i40e_setup_tx_descriptors(&tx_rings[i]);
+ if (err) {
+ while (i) {
+ i--;
+ i40e_free_tx_resources(&tx_rings[i]);
+ }
+ kfree(tx_rings);
+ tx_rings = NULL;
+
+ goto done;
+ }
+ }
+ }
+
+ /* alloc updated Rx resources */
+ if (new_rx_count != vsi->rx_rings[0].count) {
+ netdev_info(netdev,
+ "Changing Rx descriptor count from %d to %d\n",
+ vsi->rx_rings[0].count, new_rx_count);
+ rx_rings = kcalloc(vsi->alloc_queue_pairs,
+ sizeof(struct i40e_ring), GFP_KERNEL);
+ if (!rx_rings) {
+ err = -ENOMEM;
+ goto free_tx;
+ }
+
+ for (i = 0; i < vsi->num_queue_pairs; i++) {
+ /* clone ring and setup updated count */
+ rx_rings[i] = vsi->rx_rings[i];
+ rx_rings[i].count = new_rx_count;
+ err = i40e_setup_rx_descriptors(&rx_rings[i]);
+ if (err) {
+ while (i) {
+ i--;
+ i40e_free_rx_resources(&rx_rings[i]);
+ }
+ kfree(rx_rings);
+ rx_rings = NULL;
+
+ goto free_tx;
+ }
+ }
+ }
+
+ /* Bring interface down, copy in the new ring info,
+ * then restore the interface
+ */
+ i40e_down(vsi);
+
+ if (tx_rings) {
+ for (i = 0; i < vsi->num_queue_pairs; i++) {
+ i40e_free_tx_resources(&vsi->tx_rings[i]);
+ vsi->tx_rings[i] = tx_rings[i];
+ }
+ kfree(tx_rings);
+ tx_rings = NULL;
+ }
+
+ if (rx_rings) {
+ for (i = 0; i < vsi->num_queue_pairs; i++) {
+ i40e_free_rx_resources(&vsi->rx_rings[i]);
+ vsi->rx_rings[i] = rx_rings[i];
+ }
+ kfree(rx_rings);
+ rx_rings = NULL;
+ }
+
+ i40e_up(vsi);
+
+free_tx:
+ /* error cleanup if the Rx allocations failed after getting Tx */
+ if (tx_rings) {
+ for (i = 0; i < vsi->num_queue_pairs; i++)
+ i40e_free_tx_resources(&tx_rings[i]);
+ kfree(tx_rings);
+ tx_rings = NULL;
+ }
+
+done:
+ clear_bit(__I40E_CONFIG_BUSY, &pf->state);
+
+ return err;
+}
+
+static int i40e_get_sset_count(struct net_device *netdev, int sset)
+{
+ struct i40e_netdev_priv *np = netdev_priv(netdev);
+ struct i40e_vsi *vsi = np->vsi;
+ struct i40e_pf *pf = vsi->back;
+
+ switch (sset) {
+ case ETH_SS_TEST:
+ return I40E_TEST_LEN;
+ case ETH_SS_STATS:
+ if (vsi == pf->vsi[pf->lan_vsi])
+ return I40E_PF_STATS_LEN(netdev);
+ else
+ return I40E_VSI_STATS_LEN(netdev);
+ default:
+ return -EOPNOTSUPP;
+ }
+}
+
+static void i40e_get_ethtool_stats(struct net_device *netdev,
+ struct ethtool_stats *stats, u64 *data)
+{
+ struct i40e_netdev_priv *np = netdev_priv(netdev);
+ struct i40e_vsi *vsi = np->vsi;
+ struct i40e_pf *pf = vsi->back;
+ int i = 0;
+ char *p;
+ int j;
+ struct rtnl_link_stats64 *net_stats = i40e_get_vsi_stats_struct(vsi);
+
+ i40e_update_stats(vsi);
+
+ for (j = 0; j < I40E_NETDEV_STATS_LEN; j++) {
+ p = (char *)net_stats + i40e_gstrings_net_stats[j].stat_offset;
+ data[i++] = (i40e_gstrings_net_stats[j].sizeof_stat ==
+ sizeof(u64)) ? *(u64 *)p : *(u32 *)p;
+ }
+ for (j = 0; j < vsi->num_queue_pairs; j++) {
+ data[i++] = vsi->tx_rings[j].tx_stats.packets;
+ data[i++] = vsi->tx_rings[j].tx_stats.bytes;
+ }
+ for (j = 0; j < vsi->num_queue_pairs; j++) {
+ data[i++] = vsi->rx_rings[j].rx_stats.packets;
+ data[i++] = vsi->rx_rings[j].rx_stats.bytes;
+ }
+ if (vsi == pf->vsi[pf->lan_vsi]) {
+ for (j = 0; j < I40E_GLOBAL_STATS_LEN; j++) {
+ p = (char *)pf + i40e_gstrings_stats[j].stat_offset;
+ data[i++] = (i40e_gstrings_stats[j].sizeof_stat ==
+ sizeof(u64)) ? *(u64 *)p : *(u32 *)p;
+ }
+ for (j = 0; j < I40E_MAX_USER_PRIORITY; j++) {
+ data[i++] = pf->stats.priority_xon_tx[j];
+ data[i++] = pf->stats.priority_xoff_tx[j];
+ }
+ for (j = 0; j < I40E_MAX_USER_PRIORITY; j++) {
+ data[i++] = pf->stats.priority_xon_rx[j];
+ data[i++] = pf->stats.priority_xoff_rx[j];
+ }
+ for (j = 0; j < I40E_MAX_USER_PRIORITY; j++)
+ data[i++] = pf->stats.priority_xon_2_xoff[j];
+ }
+}
+
+static void i40e_get_strings(struct net_device *netdev, u32 stringset,
+ u8 *data)
+{
+ struct i40e_netdev_priv *np = netdev_priv(netdev);
+ struct i40e_vsi *vsi = np->vsi;
+ struct i40e_pf *pf = vsi->back;
+ char *p = (char *)data;
+ int i;
+
+ switch (stringset) {
+ case ETH_SS_TEST:
+ for (i = 0; i < I40E_TEST_LEN; i++) {
+ memcpy(data, i40e_gstrings_test[i], ETH_GSTRING_LEN);
+ data += ETH_GSTRING_LEN;
+ }
+ break;
+ case ETH_SS_STATS:
+ for (i = 0; i < I40E_NETDEV_STATS_LEN; i++) {
+ snprintf(p, ETH_GSTRING_LEN, "%s",
+ i40e_gstrings_net_stats[i].stat_string);
+ p += ETH_GSTRING_LEN;
+ }
+ for (i = 0; i < vsi->num_queue_pairs; i++) {
+ snprintf(p, ETH_GSTRING_LEN, "tx-%u.tx_packets", i);
+ p += ETH_GSTRING_LEN;
+ snprintf(p, ETH_GSTRING_LEN, "tx-%u.tx_bytes", i);
+ p += ETH_GSTRING_LEN;
+ }
+ for (i = 0; i < vsi->num_queue_pairs; i++) {
+ snprintf(p, ETH_GSTRING_LEN, "rx-%u.rx_packets", i);
+ p += ETH_GSTRING_LEN;
+ snprintf(p, ETH_GSTRING_LEN, "rx-%u.rx_bytes", i);
+ p += ETH_GSTRING_LEN;
+ }
+ if (vsi == pf->vsi[pf->lan_vsi]) {
+ for (i = 0; i < I40E_GLOBAL_STATS_LEN; i++) {
+ snprintf(p, ETH_GSTRING_LEN, "port.%s",
+ i40e_gstrings_stats[i].stat_string);
+ p += ETH_GSTRING_LEN;
+ }
+ for (i = 0; i < I40E_MAX_USER_PRIORITY; i++) {
+ snprintf(p, ETH_GSTRING_LEN,
+ "port.tx_priority_%u_xon", i);
+ p += ETH_GSTRING_LEN;
+ snprintf(p, ETH_GSTRING_LEN,
+ "port.tx_priority_%u_xoff", i);
+ p += ETH_GSTRING_LEN;
+ }
+ for (i = 0; i < I40E_MAX_USER_PRIORITY; i++) {
+ snprintf(p, ETH_GSTRING_LEN,
+ "port.rx_priority_%u_xon", i);
+ p += ETH_GSTRING_LEN;
+ snprintf(p, ETH_GSTRING_LEN,
+ "port.rx_priority_%u_xoff", i);
+ p += ETH_GSTRING_LEN;
+ }
+ for (i = 0; i < I40E_MAX_USER_PRIORITY; i++) {
+ snprintf(p, ETH_GSTRING_LEN,
+ "port.rx_priority_%u_xon_2_xoff", i);
+ p += ETH_GSTRING_LEN;
+ }
+ }
+ /* BUG_ON(p - data != I40E_STATS_LEN * ETH_GSTRING_LEN); */
+ break;
+ }
+}
+
+static int i40e_get_ts_info(struct net_device *dev,
+ struct ethtool_ts_info *info)
+{
+ return ethtool_op_get_ts_info(dev, info);
+}
+
+static int i40e_link_test(struct i40e_pf *pf, u64 *data)
+{
+ bool link_up;
+
+ link_up = i40e_get_link_status(&pf->hw);
+ if (link_up)
+ *data = 0;
+ else
+ *data = 1;
+
+ return *data;
+}
+
+static int i40e_reg_test(struct i40e_pf *pf, u64 *data)
+{
+ i40e_status ret;
+
+ ret = i40e_diag_reg_test(&pf->hw);
+ *data = ret;
+
+ return ret;
+}
+
+static int i40e_eeprom_test(struct i40e_pf *pf, u64 *data)
+{
+ i40e_status ret;
+
+ ret = i40e_diag_eeprom_test(&pf->hw);
+ *data = ret;
+
+ return ret;
+}
+
+static int i40e_intr_test(struct i40e_pf *pf, u64 *data)
+{
+ *data = I40E_ERR_NOT_IMPLEMENTED;
+
+ return *data;
+}
+
+static int i40e_loopback_test(struct i40e_pf *pf, u64 *data)
+{
+ *data = I40E_ERR_NOT_IMPLEMENTED;
+
+ return *data;
+}
+
+static void i40e_diag_test(struct net_device *netdev,
+ struct ethtool_test *eth_test, u64 *data)
+{
+ struct i40e_netdev_priv *np = netdev_priv(netdev);
+ struct i40e_pf *pf = np->vsi->back;
+
+ set_bit(__I40E_TESTING, &pf->state);
+ if (eth_test->flags == ETH_TEST_FL_OFFLINE) {
+ /* Offline tests */
+
+ netdev_info(netdev, "offline testing starting\n");
+
+ /* Link test performed before hardware reset
+ * so autoneg doesn't interfere with test result
+ */
+ netdev_info(netdev, "link test starting\n");
+ if (i40e_link_test(pf, &data[I40E_ETH_TEST_LINK]))
+ eth_test->flags |= ETH_TEST_FL_FAILED;
+
+ netdev_info(netdev, "register test starting\n");
+ if (i40e_reg_test(pf, &data[I40E_ETH_TEST_REG]))
+ eth_test->flags |= ETH_TEST_FL_FAILED;
+
+ i40e_do_reset(pf, (1 << __I40E_PF_RESET_REQUESTED));
+ netdev_info(netdev, "eeprom test starting\n");
+ if (i40e_eeprom_test(pf, &data[I40E_ETH_TEST_EEPROM]))
+ eth_test->flags |= ETH_TEST_FL_FAILED;
+
+ i40e_do_reset(pf, (1 << __I40E_PF_RESET_REQUESTED));
+ netdev_info(netdev, "interrupt test starting\n");
+ if (i40e_intr_test(pf, &data[I40E_ETH_TEST_INTR]))
+ eth_test->flags |= ETH_TEST_FL_FAILED;
+
+ i40e_do_reset(pf, (1 << __I40E_PF_RESET_REQUESTED));
+ netdev_info(netdev, "loopback test starting\n");
+ if (i40e_loopback_test(pf, &data[I40E_ETH_TEST_LOOPBACK]))
+ eth_test->flags |= ETH_TEST_FL_FAILED;
+
+ } else {
+ netdev_info(netdev, "online test starting\n");
+ /* Online tests */
+ if (i40e_link_test(pf, &data[I40E_ETH_TEST_LINK]))
+ eth_test->flags |= ETH_TEST_FL_FAILED;
+
+ /* Offline only tests, not run in online; pass by default */
+ data[I40E_ETH_TEST_REG] = 0;
+ data[I40E_ETH_TEST_EEPROM] = 0;
+ data[I40E_ETH_TEST_INTR] = 0;
+ data[I40E_ETH_TEST_LOOPBACK] = 0;
+
+ clear_bit(__I40E_TESTING, &pf->state);
+ }
+}
+
+static void i40e_get_wol(struct net_device *netdev,
+ struct ethtool_wolinfo *wol)
+{
+ wol->supported = 0;
+ wol->wolopts = 0;
+}
+
+static int i40e_nway_reset(struct net_device *netdev)
+{
+ /* restart autonegotiation */
+ struct i40e_netdev_priv *np = netdev_priv(netdev);
+ struct i40e_pf *pf = np->vsi->back;
+ struct i40e_hw *hw = &pf->hw;
+ i40e_status ret = 0;
+
+ ret = i40e_aq_set_link_restart_an(hw, NULL);
+ if (ret != I40E_SUCCESS)
+ netdev_info(netdev, "link restart failed, aq_err=%d\n",
+ pf->hw.aq.asq_last_status);
+
+ return ret;
+}
+
+static int i40e_set_phys_id(struct net_device *netdev,
+ enum ethtool_phys_id_state state)
+{
+ struct i40e_netdev_priv *np = netdev_priv(netdev);
+ struct i40e_pf *pf = np->vsi->back;
+ struct i40e_hw *hw = &pf->hw;
+ int blink_freq = 2;
+
+ switch (state) {
+ case ETHTOOL_ID_ACTIVE:
+ pf->led_status = i40e_led_get(hw);
+ return blink_freq;
+ case ETHTOOL_ID_ON:
+ i40e_led_set(hw, 0xF);
+ break;
+ case ETHTOOL_ID_OFF:
+ i40e_led_set(hw, 0x0);
+ break;
+ case ETHTOOL_ID_INACTIVE:
+ i40e_led_set(hw, pf->led_status);
+ break;
+ }
+
+ return 0;
+}
+
+/* NOTE: i40e hardware uses a conversion factor of 2 for Interrupt
+ * Throttle Rate (ITR) ie. ITR(1) = 2us ITR(10) = 20 us, and also
+ * 125us (8000 interrupts per second) == ITR(62)
+ */
+
+static int i40e_get_coalesce(struct net_device *netdev,
+ struct ethtool_coalesce *ec)
+{
+ struct i40e_netdev_priv *np = netdev_priv(netdev);
+ struct i40e_vsi *vsi = np->vsi;
+
+ ec->tx_max_coalesced_frames_irq = vsi->work_limit;
+ ec->rx_max_coalesced_frames_irq = vsi->work_limit;
+
+ if (ITR_IS_DYNAMIC(vsi->rx_itr_setting))
+ ec->rx_coalesce_usecs = 1;
+ else
+ ec->rx_coalesce_usecs = vsi->rx_itr_setting;
+
+ if (ITR_IS_DYNAMIC(vsi->tx_itr_setting))
+ ec->tx_coalesce_usecs = 1;
+ else
+ ec->tx_coalesce_usecs = vsi->tx_itr_setting;
+
+ return 0;
+}
+
+static int i40e_set_coalesce(struct net_device *netdev,
+ struct ethtool_coalesce *ec)
+{
+ struct i40e_netdev_priv *np = netdev_priv(netdev);
+ struct i40e_q_vector *q_vector;
+ struct i40e_vsi *vsi = np->vsi;
+ struct i40e_pf *pf = vsi->back;
+ struct i40e_hw *hw = &pf->hw;
+ u16 vector;
+ int i;
+
+ if (ec->tx_max_coalesced_frames_irq || ec->rx_max_coalesced_frames_irq)
+ vsi->work_limit = ec->tx_max_coalesced_frames_irq;
+
+ switch (ec->rx_coalesce_usecs) {
+ case 0:
+ vsi->rx_itr_setting = 0;
+ break;
+ case 1:
+ vsi->rx_itr_setting = (I40E_ITR_DYNAMIC |
+ ITR_REG_TO_USEC(I40E_ITR_RX_DEF));
+ break;
+ default:
+ if ((ec->rx_coalesce_usecs < (I40E_MIN_ITR << 1)) ||
+ (ec->rx_coalesce_usecs > (I40E_MAX_ITR << 1)))
+ return -EINVAL;
+ vsi->rx_itr_setting = ec->rx_coalesce_usecs;
+ break;
+ }
+
+ switch (ec->tx_coalesce_usecs) {
+ case 0:
+ vsi->tx_itr_setting = 0;
+ break;
+ case 1:
+ vsi->tx_itr_setting = (I40E_ITR_DYNAMIC |
+ ITR_REG_TO_USEC(I40E_ITR_TX_DEF));
+ break;
+ default:
+ if ((ec->tx_coalesce_usecs < (I40E_MIN_ITR << 1)) ||
+ (ec->tx_coalesce_usecs > (I40E_MAX_ITR << 1)))
+ return -EINVAL;
+ vsi->tx_itr_setting = ec->tx_coalesce_usecs;
+ break;
+ }
+
+ vector = vsi->base_vector;
+ q_vector = vsi->q_vectors;
+ for (i = 0; i < vsi->num_q_vectors; i++, vector++, q_vector++) {
+ q_vector->rx.itr = ITR_TO_REG(vsi->rx_itr_setting);
+ wr32(hw, I40E_PFINT_ITRN(0, vector - 1), q_vector->rx.itr);
+ q_vector->tx.itr = ITR_TO_REG(vsi->tx_itr_setting);
+ wr32(hw, I40E_PFINT_ITRN(1, vector - 1), q_vector->tx.itr);
+ i40e_flush(hw);
+ }
+
+ return 0;
+}
+
+/**
+ * i40e_get_rss_hash_opts - Get RSS hash Input Set for each flow type
+ * @pf: pointer to the physical function struct
+ * @cmd: ethtool rxnfc command
+ *
+ * Returns Success if the flow is supported, else Invalid Input.
+ **/
+static int i40e_get_rss_hash_opts(struct i40e_pf *pf, struct ethtool_rxnfc *cmd)
+{
+ cmd->data = 0;
+
+ /* Report default options for RSS on i40e */
+ switch (cmd->flow_type) {
+ case TCP_V4_FLOW:
+ case UDP_V4_FLOW:
+ cmd->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
+ /* fall through to add IP fields */
+ case SCTP_V4_FLOW:
+ case AH_ESP_V4_FLOW:
+ case AH_V4_FLOW:
+ case ESP_V4_FLOW:
+ case IPV4_FLOW:
+ cmd->data |= RXH_IP_SRC | RXH_IP_DST;
+ break;
+ case TCP_V6_FLOW:
+ case UDP_V6_FLOW:
+ cmd->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
+ /* fall through to add IP fields */
+ case SCTP_V6_FLOW:
+ case AH_ESP_V6_FLOW:
+ case AH_V6_FLOW:
+ case ESP_V6_FLOW:
+ case IPV6_FLOW:
+ cmd->data |= RXH_IP_SRC | RXH_IP_DST;
+ break;
+ default:
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
+/**
+ * i40e_get_rxnfc - command to get RX flow classification rules
+ * @netdev: network interface device structure
+ * @cmd: ethtool rxnfc command
+ *
+ * Returns Success if the command is supported.
+ **/
+static int i40e_get_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *cmd,
+ u32 *rule_locs)
+{
+ struct i40e_netdev_priv *np = netdev_priv(netdev);
+ struct i40e_vsi *vsi = np->vsi;
+ struct i40e_pf *pf = vsi->back;
+ int ret = -EOPNOTSUPP;
+
+ switch (cmd->cmd) {
+ case ETHTOOL_GRXRINGS:
+ cmd->data = vsi->alloc_queue_pairs;
+ ret = 0;
+ break;
+ case ETHTOOL_GRXFH:
+ ret = i40e_get_rss_hash_opts(pf, cmd);
+ break;
+ case ETHTOOL_GRXCLSRLCNT:
+ ret = 0;
+ break;
+ case ETHTOOL_GRXCLSRULE:
+ ret = 0;
+ break;
+ case ETHTOOL_GRXCLSRLALL:
+ cmd->data = 500;
+ ret = 0;
+ default:
+ break;
+ }
+
+ return ret;
+}
+
+/**
+ * i40e_set_rss_hash_opt - Enable/Disable flow types for RSS hash
+ * @pf: pointer to the physical function struct
+ * @cmd: ethtool rxnfc command
+ *
+ * Returns Success if the flow input set is supported.
+ **/
+static int i40e_set_rss_hash_opt(struct i40e_pf *pf, struct ethtool_rxnfc *nfc)
+{
+ struct i40e_hw *hw = &pf->hw;
+ u64 hena = (u64)rd32(hw, I40E_PFQF_HENA(0)) |
+ ((u64)rd32(hw, I40E_PFQF_HENA(1)) << 32);
+
+ /* RSS does not support anything other than hashing
+ * to queues on src and dst IPs and ports
+ */
+ if (nfc->data & ~(RXH_IP_SRC | RXH_IP_DST |
+ RXH_L4_B_0_1 | RXH_L4_B_2_3))
+ return -EINVAL;
+
+ /* We need at least the IP SRC and DEST fields for hashing */
+ if (!(nfc->data & RXH_IP_SRC) ||
+ !(nfc->data & RXH_IP_DST))
+ return -EINVAL;
+
+ switch (nfc->flow_type) {
+ case TCP_V4_FLOW:
+ switch (nfc->data & (RXH_L4_B_0_1 | RXH_L4_B_2_3)) {
+ case 0:
+ hena &= ~((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV4_TCP);
+ break;
+ case (RXH_L4_B_0_1 | RXH_L4_B_2_3):
+ hena |= ((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV4_TCP);
+ break;
+ default:
+ return -EINVAL;
+ }
+ break;
+ case TCP_V6_FLOW:
+ switch (nfc->data & (RXH_L4_B_0_1 | RXH_L4_B_2_3)) {
+ case 0:
+ hena &= ~((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV6_TCP);
+ break;
+ case (RXH_L4_B_0_1 | RXH_L4_B_2_3):
+ hena |= ((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV6_TCP);
+ break;
+ default:
+ return -EINVAL;
+ }
+ break;
+ case UDP_V4_FLOW:
+ switch (nfc->data & (RXH_L4_B_0_1 | RXH_L4_B_2_3)) {
+ case 0:
+ hena &=
+ ~(((u64)1 << I40E_FILTER_PCTYPE_NONF_UNICAST_IPV4_UDP) |
+ ((u64)1 << I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV4_UDP) |
+ ((u64)1 << I40E_FILTER_PCTYPE_FRAG_IPV4));
+ break;
+ case (RXH_L4_B_0_1 | RXH_L4_B_2_3):
+ hena |=
+ (((u64)1 << I40E_FILTER_PCTYPE_NONF_UNICAST_IPV4_UDP) |
+ ((u64)1 << I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV4_UDP) |
+ ((u64)1 << I40E_FILTER_PCTYPE_FRAG_IPV4));
+ break;
+ default:
+ return -EINVAL;
+ }
+ break;
+ case UDP_V6_FLOW:
+ switch (nfc->data & (RXH_L4_B_0_1 | RXH_L4_B_2_3)) {
+ case 0:
+ hena &=
+ ~(((u64)1 << I40E_FILTER_PCTYPE_NONF_UNICAST_IPV6_UDP) |
+ ((u64)1 << I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV6_UDP) |
+ ((u64)1 << I40E_FILTER_PCTYPE_FRAG_IPV6));
+ break;
+ case (RXH_L4_B_0_1 | RXH_L4_B_2_3):
+ hena |=
+ (((u64)1 << I40E_FILTER_PCTYPE_NONF_UNICAST_IPV6_UDP) |
+ ((u64)1 << I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV6_UDP) |
+ ((u64)1 << I40E_FILTER_PCTYPE_FRAG_IPV6));
+ break;
+ default:
+ return -EINVAL;
+ }
+ break;
+ case AH_ESP_V4_FLOW:
+ case AH_V4_FLOW:
+ case ESP_V4_FLOW:
+ case SCTP_V4_FLOW:
+ if ((nfc->data & RXH_L4_B_0_1) ||
+ (nfc->data & RXH_L4_B_2_3))
+ return -EINVAL;
+ hena |= ((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV4_OTHER);
+ break;
+ case AH_ESP_V6_FLOW:
+ case AH_V6_FLOW:
+ case ESP_V6_FLOW:
+ case SCTP_V6_FLOW:
+ if ((nfc->data & RXH_L4_B_0_1) ||
+ (nfc->data & RXH_L4_B_2_3))
+ return -EINVAL;
+ hena |= ((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV6_OTHER);
+ break;
+ case IPV4_FLOW:
+ hena |= ((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV4_OTHER) |
+ ((u64)1 << I40E_FILTER_PCTYPE_FRAG_IPV4);
+ break;
+ case IPV6_FLOW:
+ hena |= ((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV6_OTHER) |
+ ((u64)1 << I40E_FILTER_PCTYPE_FRAG_IPV6);
+ break;
+ default:
+ return -EINVAL;
+ }
+
+ wr32(hw, I40E_PFQF_HENA(0), (u32)hena);
+ wr32(hw, I40E_PFQF_HENA(1), (u32)(hena >> 32));
+ i40e_flush(hw);
+
+ return 0;
+}
+
+#define IP_HEADER_OFFSET 14
+/**
+ * i40e_add_del_fdir_udpv4 - Add/Remove UDPv4 Flow Director filters for
+ * a specific flow spec
+ * @vsi: pointer to the targeted VSI
+ * @fd_data: the flow director data required from the FDir descriptor
+ * @ethtool_rx_flow_spec: the flow spec
+ * @add: true adds a filter, false removes it
+ *
+ * Returns 0 if the filters were successfully added or removed
+ **/
+static i40e_status i40e_add_del_fdir_udpv4(struct i40e_vsi *vsi,
+ struct i40e_fdir_data *fd_data,
+ struct ethtool_rx_flow_spec *fsp, bool add)
+{
+ struct i40e_pf *pf = vsi->back;
+ struct udphdr *udp;
+ struct iphdr *ip;
+ bool err = false;
+ i40e_status ret;
+ int i;
+
+ ip = (struct iphdr *)(fd_data->raw_packet + IP_HEADER_OFFSET);
+ udp = (struct udphdr *)(fd_data->raw_packet + IP_HEADER_OFFSET
+ + sizeof(struct iphdr));
+
+ ip->saddr = fsp->h_u.tcp_ip4_spec.ip4src;
+ ip->daddr = fsp->h_u.tcp_ip4_spec.ip4dst;
+ udp->source = fsp->h_u.tcp_ip4_spec.psrc;
+ udp->dest = fsp->h_u.tcp_ip4_spec.pdst;
+
+ for (i = I40E_FILTER_PCTYPE_NONF_UNICAST_IPV4_UDP;
+ i <= I40E_FILTER_PCTYPE_NONF_IPV4_UDP; i++) {
+ fd_data->pctype = i;
+ ret = i40e_program_fdir_filter(fd_data, pf, add);
+
+ if (ret != I40E_SUCCESS) {
+ dev_info(&pf->pdev->dev,
+ "Filter command send failed for PCTYPE %d (ret = %d)\n",
+ fd_data->pctype, ret);
+ err = true;
+ } else {
+ dev_info(&pf->pdev->dev,
+ "Filter OK for PCTYPE %d (ret = %d)\n",
+ fd_data->pctype, ret);
+ }
+ }
+
+ return err ? -EOPNOTSUPP : I40E_SUCCESS;
+}
+
+/**
+ * i40e_add_del_fdir_tcpv4 - Add/Remove TCPv4 Flow Director filters for
+ * a specific flow spec
+ * @vsi: pointer to the targeted VSI
+ * @fd_data: the flow director data required from the FDir descriptor
+ * @ethtool_rx_flow_spec: the flow spec
+ * @add: true adds a filter, false removes it
+ *
+ * Returns 0 if the filters were successfully added or removed
+ **/
+static i40e_status i40e_add_del_fdir_tcpv4(struct i40e_vsi *vsi,
+ struct i40e_fdir_data *fd_data,
+ struct ethtool_rx_flow_spec *fsp, bool add)
+{
+ struct i40e_pf *pf = vsi->back;
+ struct tcphdr *tcp;
+ struct iphdr *ip;
+ bool err = false;
+ i40e_status ret;
+
+ ip = (struct iphdr *)(fd_data->raw_packet + IP_HEADER_OFFSET);
+ tcp = (struct tcphdr *)(fd_data->raw_packet + IP_HEADER_OFFSET
+ + sizeof(struct iphdr));
+
+ ip->daddr = fsp->h_u.tcp_ip4_spec.ip4dst;
+ tcp->dest = fsp->h_u.tcp_ip4_spec.pdst;
+
+ fd_data->pctype = I40E_FILTER_PCTYPE_NONF_IPV4_TCP_SYN;
+ ret = i40e_program_fdir_filter(fd_data, pf, add);
+
+ if (ret != I40E_SUCCESS) {
+ dev_info(&pf->pdev->dev,
+ "Filter command send failed for PCTYPE %d (ret = %d)\n",
+ fd_data->pctype, ret);
+ err = true;
+ } else {
+ dev_info(&pf->pdev->dev, "Filter OK for PCTYPE %d (ret = %d)\n",
+ fd_data->pctype, ret);
+ }
+
+ ip->saddr = fsp->h_u.tcp_ip4_spec.ip4src;
+ tcp->source = fsp->h_u.tcp_ip4_spec.psrc;
+
+ fd_data->pctype = I40E_FILTER_PCTYPE_NONF_IPV4_TCP;
+
+ if (ret != I40E_SUCCESS) {
+ dev_info(&pf->pdev->dev,
+ "Filter command send failed for PCTYPE %d (ret = %d)\n",
+ fd_data->pctype, ret);
+ err = true;
+ } else {
+ dev_info(&pf->pdev->dev, "Filter OK for PCTYPE %d (ret = %d)\n",
+ fd_data->pctype, ret);
+ }
+
+ return err ? -EOPNOTSUPP : I40E_SUCCESS;
+}
+
+
+/**
+ * i40e_add_del_fdir_sctpv4 - Add/Remove SCTPv4 Flow Director filters for
+ * a specific flow spec
+ * @vsi: pointer to the targeted VSI
+ * @fd_data: the flow director data required from the FDir descriptor
+ * @ethtool_rx_flow_spec: the flow spec
+ * @add: true adds a filter, false removes it
+ *
+ * Returns I40E_SUCCESS if the filters were successfully added or removed
+ **/
+static i40e_status i40e_add_del_fdir_sctpv4(struct i40e_vsi *vsi,
+ struct i40e_fdir_data *fd_data,
+ struct ethtool_rx_flow_spec *fsp, bool add)
+{
+ return -EOPNOTSUPP;
+}
+
+/**
+ * i40e_add_del_fdir_ipv4 - Add/Remove IPv4 Flow Director filters for
+ * a specific flow spec
+ * @vsi: pointer to the targeted VSI
+ * @fd_data: the flow director data required for the FDir descriptor
+ * @fsp: the ethtool flow spec
+ * @add: true adds a filter, false removes it
+ *
+ * Returns I40E_SUCCESS if the filters were successfully added or removed
+ **/
+static i40e_status i40e_add_del_fdir_ipv4(struct i40e_vsi *vsi,
+ struct i40e_fdir_data *fd_data,
+ struct ethtool_rx_flow_spec *fsp, bool add)
+{
+ struct i40e_pf *pf = vsi->back;
+ struct iphdr *ip;
+ bool err = false;
+ i40e_status ret;
+ int i;
+
+ ip = (struct iphdr *)(fd_data->raw_packet + IP_HEADER_OFFSET);
+
+ ip->saddr = fsp->h_u.usr_ip4_spec.ip4src;
+ ip->daddr = fsp->h_u.usr_ip4_spec.ip4dst;
+ ip->protocol = fsp->h_u.usr_ip4_spec.proto;
+
+ for (i = I40E_FILTER_PCTYPE_NONF_IPV4_OTHER;
+ i <= I40E_FILTER_PCTYPE_FRAG_IPV4; i++) {
+ fd_data->pctype = i;
+ ret = i40e_program_fdir_filter(fd_data, pf, add);
+
+ if (ret != I40E_SUCCESS) {
+ dev_info(&pf->pdev->dev,
+ "Filter command send failed for PCTYPE %d (ret = %d)\n",
+ fd_data->pctype, ret);
+ err = true;
+ } else {
+ dev_info(&pf->pdev->dev,
+ "Filter OK for PCTYPE %d (ret = %d)\n",
+ fd_data->pctype, ret);
+ }
+ }
+
+ return err ? -EOPNOTSUPP : I40E_SUCCESS;
+}
+
+/**
+ * i40e_add_del_fdir_ethtool - Add/Remove Flow Director filters for
+ * a specific flow spec based on their protocol
+ * @vsi: pointer to the targeted VSI
+ * @cmd: command to get or set RX flow classification rules
+ * @add: true adds a filter, false removes it
+ *
+ * Returns 0 if the filters were successfully added or removed
+ **/
+static int i40e_add_del_fdir_ethtool(struct i40e_vsi *vsi,
+ struct ethtool_rxnfc *cmd, bool add)
+{
+ struct i40e_fdir_data fd_data;
+ i40e_status ret = -EINVAL;
+ struct i40e_pf *pf;
+ struct ethtool_rx_flow_spec *fsp =
+ (struct ethtool_rx_flow_spec *)&cmd->fs;
+
+ if (!vsi)
+ return -EINVAL;
+
+ pf = vsi->back;
+
+ if ((fsp->ring_cookie != RX_CLS_FLOW_DISC) &&
+ (fsp->ring_cookie >= vsi->num_queue_pairs))
+ return -EINVAL;
+
+ /* Populate the Flow Director that we have at the moment
+ * and allocate the raw packet buffer for the calling functions
+ */
+ fd_data.raw_packet = kzalloc(I40E_FDIR_MAX_RAW_PACKET_LOOKUP,
+ GFP_KERNEL);
+
+ if (!fd_data.raw_packet) {
+ dev_info(&pf->pdev->dev, "Could not allocate memory\n");
+ return -ENOMEM;
+ }
+
+ fd_data.q_index = fsp->ring_cookie;
+ fd_data.flex_off = 0;
+ fd_data.pctype = 0;
+ fd_data.dest_vsi = vsi->id;
+ fd_data.dest_ctl = 0;
+ fd_data.fd_status = 0;
+ fd_data.cnt_index = 0;
+ fd_data.fd_id = 0;
+
+ switch (fsp->flow_type & ~FLOW_EXT) {
+ case TCP_V4_FLOW:
+ ret = i40e_add_del_fdir_tcpv4(vsi, &fd_data, fsp, add);
+ break;
+ case UDP_V4_FLOW:
+ ret = i40e_add_del_fdir_udpv4(vsi, &fd_data, fsp, add);
+ break;
+ case SCTP_V4_FLOW:
+ ret = i40e_add_del_fdir_sctpv4(vsi, &fd_data, fsp, add);
+ break;
+ case IPV4_FLOW:
+ ret = i40e_add_del_fdir_ipv4(vsi, &fd_data, fsp, add);
+ break;
+ case IP_USER_FLOW:
+ switch (fsp->h_u.usr_ip4_spec.proto) {
+ case IPPROTO_TCP:
+ ret = i40e_add_del_fdir_tcpv4(vsi, &fd_data, fsp, add);
+ break;
+ case IPPROTO_UDP:
+ ret = i40e_add_del_fdir_udpv4(vsi, &fd_data, fsp, add);
+ break;
+ case IPPROTO_SCTP:
+ ret = i40e_add_del_fdir_sctpv4(vsi, &fd_data, fsp, add);
+ break;
+ default:
+ ret = i40e_add_del_fdir_ipv4(vsi, &fd_data, fsp, add);
+ break;
+ }
+ break;
+ default:
+ dev_info(&pf->pdev->dev, "Could not specify spec type\n");
+ ret = -EINVAL;
+ }
+
+ kfree(fd_data.raw_packet);
+ fd_data.raw_packet = NULL;
+
+ return ret;
+}
+/**
+ * i40e_set_rxnfc - command to set RX flow classification rules
+ * @netdev: network interface device structure
+ * @cmd: ethtool rxnfc command
+ *
+ * Returns Success if the command is supported.
+ **/
+static int i40e_set_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *cmd)
+{
+ struct i40e_netdev_priv *np = netdev_priv(netdev);
+ struct i40e_vsi *vsi = np->vsi;
+ struct i40e_pf *pf = vsi->back;
+ int ret = -EOPNOTSUPP;
+
+ switch (cmd->cmd) {
+ case ETHTOOL_SRXFH:
+ ret = i40e_set_rss_hash_opt(pf, cmd);
+ break;
+ case ETHTOOL_SRXCLSRLINS:
+ ret = i40e_add_del_fdir_ethtool(vsi, cmd, true);
+ break;
+ case ETHTOOL_SRXCLSRLDEL:
+ ret = i40e_add_del_fdir_ethtool(vsi, cmd, false);
+ break;
+ default:
+ break;
+ }
+
+ return ret;
+}
+
+static struct ethtool_ops i40e_ethtool_ops = {
+ .get_settings = i40e_get_settings,
+ .get_drvinfo = i40e_get_drvinfo,
+ .get_regs_len = i40e_get_regs_len,
+ .get_regs = i40e_get_regs,
+ .nway_reset = i40e_nway_reset,
+ .get_link = ethtool_op_get_link,
+ .get_wol = i40e_get_wol,
+ .get_eeprom_len = i40e_get_eeprom_len,
+ .get_eeprom = i40e_get_eeprom,
+ .get_ringparam = i40e_get_ringparam,
+ .set_ringparam = i40e_set_ringparam,
+ .get_pauseparam = i40e_get_pauseparam,
+ .get_msglevel = i40e_get_msglevel,
+ .set_msglevel = i40e_set_msglevel,
+ .get_rxnfc = i40e_get_rxnfc,
+ .set_rxnfc = i40e_set_rxnfc,
+ .self_test = i40e_diag_test,
+ .get_strings = i40e_get_strings,
+ .set_phys_id = i40e_set_phys_id,
+ .get_sset_count = i40e_get_sset_count,
+ .get_ethtool_stats = i40e_get_ethtool_stats,
+ .get_coalesce = i40e_get_coalesce,
+ .set_coalesce = i40e_set_coalesce,
+ .get_ts_info = i40e_get_ts_info,
+};
+
+void i40e_set_ethtool_ops(struct net_device *netdev)
+{
+ SET_ETHTOOL_OPS(netdev, &i40e_ethtool_ops);
+}
--
1.8.3.1
------------------------------------------------------------------------------
Learn the latest--Visual Studio 2012, SharePoint 2013, SQL 2012, more!
Discover the easy way to master current and previous Microsoft technologies
and advance your career. Get an incredible 1,500+ hours of step-by-step
tutorial videos with LearnDevNow. Subscribe today and save!
http://pubads.g.doubleclick.net/gampad/clk?id=58041391&iu=/4140/ostg.clktrk
_______________________________________________
E1000-devel mailing list
E1000-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/e1000-devel
To learn more about Intel® Ethernet, visit http://communities.intel.com/community/wired
^ permalink raw reply related
* [net-next v4 1/7] i40e: main driver core
From: Jeff Kirsher @ 2013-09-04 23:17 UTC (permalink / raw)
To: davem; +Cc: e1000-devel, netdev, Jesse Brandeburg, gospo, sassmann
In-Reply-To: <1378336629-24224-1-git-send-email-jeffrey.t.kirsher@intel.com>
From: Jesse Brandeburg <jesse.brandeburg@intel.com>
This is the driver for the Intel(R) Ethernet Controller XL710 Family.
This driver is targeted at basic ethernet functionality only, and will be
improved upon further as time goes on.
This patch contains the driver entry points but does not include transmit
and receive (see the next patch in the series) routines.
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Tested-by: Kavindya Deegala <kavindya.s.deegala@intel.com>
CC: PJ Waskiewicz <peter.p.waskiewicz.jr@intel.com>
CC: e1000-devel@lists.sourceforge.net
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
v1: this is the first submittal
v2: changes resulting from first round of feedback, including
better 32 bit compatibility
remove messages on alloc failure
use macros instead of functions for debug print
use ether_addr
check mac addr validity, fail earlier
use temp variables in stats collection
some changes due to internal spec clarifications
const'ify some strings
convert some macros to functions for stats gathering
use strlcpy
no trans_start
minor cleanups
remove unnecessary napi_synchronize
remove unnecessary netdev ops assignment function
use rtnl_link_stats64
fix up pci_using_dac/NETIF_F_HIGHDMA
v3:
dma_alloc_coherent
- use dma_zalloc_coherent
- don't use GFP_ATOMIC
add context and use dev_info() in i40e_get_lump()
lots of blank line comments
flush() -> i40e_flush()
reorder function variable declarations by line length - CORE
reorder function variable declarations by line length - Shared
use rounddown_pow_of_two()
enum i40e_status_code too long
blank line comments
tabs versus spaces, indent issues
void functions don't need to return
rework local #define FILTER_LIST_LEN
CoreR and GlobalR comment shrinkage in i40e_do_reset()
i40e_watchdog_subtask() variable v versus i
v4:
xps support
use struct assignment
sparse fixes
use eth_addr_equal
simplify == NULL checks
fix whitespace
use sizeof(*var)
fix indentation
__func__ removal
---
drivers/net/ethernet/intel/i40e/i40e_main.c | 7410 +++++++++++++++++++++++++++
1 file changed, 7410 insertions(+)
create mode 100644 drivers/net/ethernet/intel/i40e/i40e_main.c
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
new file mode 100644
index 0000000..be5e8b7
--- /dev/null
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -0,0 +1,7410 @@
+/*******************************************************************************
+
+ Intel Ethernet Controller XL710 Family Linux Driver
+ Copyright(c) 2013 Intel Corporation.
+
+ This program is free software; you can redistribute it and/or modify it
+ under the terms and conditions of the GNU General Public License,
+ version 2, as published by the Free Software Foundation.
+
+ This program is distributed in the hope it will be useful, but WITHOUT
+ ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ more details.
+
+ You should have received a copy of the GNU General Public License along with
+ this program; if not, write to the Free Software Foundation, Inc.,
+ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
+
+ The full GNU General Public License is included in this distribution in
+ the file called "COPYING".
+
+ Contact Information:
+ e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
+ Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
+
+*******************************************************************************/
+
+/* Local includes */
+#include "i40e.h"
+
+const char i40e_driver_name[] = "i40e";
+static const char i40e_driver_string[] =
+ "Intel(R) Ethernet Connection XL710 Network Driver";
+
+#define DRV_KERN "-k"
+
+#define DRV_VERSION_MAJOR 0
+#define DRV_VERSION_MINOR 3
+#define DRV_VERSION_BUILD 8
+#define DRV_VERSION __stringify(DRV_VERSION_MAJOR) "." \
+ __stringify(DRV_VERSION_MINOR) "." \
+ __stringify(DRV_VERSION_BUILD) DRV_KERN
+const char i40e_driver_version_str[] = DRV_VERSION;
+static const char i40e_copyright[] = "Copyright (c) 2013 Intel Corporation.";
+
+/* a bit of forward declarations */
+static void i40e_vsi_reinit_locked(struct i40e_vsi *vsi);
+static void i40e_handle_reset_warning(struct i40e_pf *pf);
+static s32 i40e_add_vsi(struct i40e_vsi *vsi);
+static s32 i40e_add_veb(struct i40e_veb *veb, struct i40e_vsi *vsi);
+static s32 i40e_setup_pf_switch(struct i40e_pf *pf);
+static int i40e_setup_misc_vector(struct i40e_pf *pf);
+static i40e_status i40e_determine_queue_usage(struct i40e_pf *pf);
+static int i40e_setup_pf_filter_control(struct i40e_pf *pf);
+
+/* i40e_pci_tbl - PCI Device ID Table
+ *
+ * Last entry must be all 0s
+ *
+ * { Vendor ID, Device ID, SubVendor ID, SubDevice ID,
+ * Class, Class Mask, private data (not used) }
+ */
+static DEFINE_PCI_DEVICE_TABLE(i40e_pci_tbl) = {
+ {PCI_VDEVICE(INTEL, I40E_SFP_XL710_DEVICE_ID), 0},
+ {PCI_VDEVICE(INTEL, I40E_SFP_X710_DEVICE_ID), 0},
+ {PCI_VDEVICE(INTEL, I40E_QEMU_DEVICE_ID), 0},
+ {PCI_VDEVICE(INTEL, I40E_KX_A_DEVICE_ID), 0},
+ {PCI_VDEVICE(INTEL, I40E_KX_B_DEVICE_ID), 0},
+ {PCI_VDEVICE(INTEL, I40E_KX_C_DEVICE_ID), 0},
+ {PCI_VDEVICE(INTEL, I40E_KX_D_DEVICE_ID), 0},
+ {PCI_VDEVICE(INTEL, I40E_QSFP_A_DEVICE_ID), 0},
+ {PCI_VDEVICE(INTEL, I40E_QSFP_B_DEVICE_ID), 0},
+ {PCI_VDEVICE(INTEL, I40E_QSFP_C_DEVICE_ID), 0},
+ /* required last entry */
+ {0, }
+};
+MODULE_DEVICE_TABLE(pci, i40e_pci_tbl);
+
+#define I40E_MAX_VF_COUNT 128
+static int debug = -1;
+module_param(debug, int, 0);
+MODULE_PARM_DESC(debug, "Debug level (0=none,...,16=all)");
+
+MODULE_AUTHOR("Intel Corporation, <e1000-devel@lists.sourceforge.net>");
+MODULE_DESCRIPTION("Intel(R) Ethernet Connection XL710 Network Driver");
+MODULE_LICENSE("GPL");
+MODULE_VERSION(DRV_VERSION);
+
+/**
+ * i40e_allocate_dma_mem_d - OS specific memory alloc for shared code
+ * @hw: pointer to the HW structure
+ * @mem: ptr to mem struct to fill out
+ * @size: size of memory requested
+ * @alignment: what to align the allocation to
+ **/
+i40e_status i40e_allocate_dma_mem_d(struct i40e_hw *hw,
+ struct i40e_dma_mem *mem,
+ u64 size, u32 alignment)
+{
+ struct i40e_pf *pf = (struct i40e_pf *)hw->back;
+
+ if (!mem)
+ return I40E_ERR_PARAM;
+
+ mem->size = ALIGN(size, alignment);
+ mem->va = dma_zalloc_coherent(&pf->pdev->dev, mem->size,
+ &mem->pa, GFP_KERNEL);
+ if (mem->va)
+ return I40E_SUCCESS;
+
+ return I40E_ERR_NO_MEMORY;
+}
+
+/**
+ * i40e_free_dma_mem_d - OS specific memory free for shared code
+ * @hw: pointer to the HW structure
+ * @mem: ptr to mem struct to free
+ **/
+i40e_status i40e_free_dma_mem_d(struct i40e_hw *hw, struct i40e_dma_mem *mem)
+{
+ struct i40e_pf *pf = (struct i40e_pf *)hw->back;
+
+ if (!mem || !mem->va)
+ return I40E_ERR_PARAM;
+
+ dma_free_coherent(&pf->pdev->dev, mem->size, mem->va, mem->pa);
+ mem->va = NULL;
+ mem->pa = 0;
+ return I40E_SUCCESS;
+}
+
+/**
+ * i40e_allocate_virt_mem_d - OS specific memory alloc for shared code
+ * @hw: pointer to the HW structure
+ * @mem: ptr to mem struct to fill out
+ * @size: size of memory requested
+ **/
+i40e_status i40e_allocate_virt_mem_d(struct i40e_hw *hw, struct i40e_virt_mem *mem,
+ u32 size)
+{
+ if (!mem)
+ return I40E_ERR_PARAM;
+
+ mem->size = size;
+ mem->va = kzalloc(size, GFP_KERNEL);
+
+ if (mem->va)
+ return I40E_SUCCESS;
+
+ return I40E_ERR_NO_MEMORY;
+}
+
+/**
+ * i40e_free_virt_mem_d - OS specific memory free for shared code
+ * @hw: pointer to the HW structure
+ * @mem: ptr to mem struct to free
+ **/
+i40e_status i40e_free_virt_mem_d(struct i40e_hw *hw, struct i40e_virt_mem *mem)
+{
+ if (!mem)
+ return I40E_ERR_PARAM;
+
+ /* it's ok to kfree a NULL pointer */
+ kfree(mem->va);
+ mem->va = NULL;
+
+ return I40E_SUCCESS;
+}
+
+/**
+ * i40e_get_lump - find a lump of free generic resource
+ * @pf: board private structure
+ * @pile: the pile of resource to search
+ * @needed: the number of items needed
+ * @id: an owner id to stick on the items assigned
+ *
+ * Returns the base item index of the lump, or negative for error
+ *
+ * The search_hint trick and lack of advanced fit-finding only work
+ * because we're highly likely to have all the same size lump requests.
+ * Linear search time and any fragmentation should be minimal.
+ **/
+static int i40e_get_lump(struct i40e_pf *pf, struct i40e_lump_tracking *pile,
+ u16 needed, u16 id)
+{
+ int ret = I40E_ERR_NO_MEMORY;
+ int i = 0;
+ int j = 0;
+
+ if (!pile || needed == 0 || id >= I40E_PILE_VALID_BIT) {
+ dev_info(&pf->pdev->dev,
+ "param err: pile=%p needed=%d id=0x%04x\n",
+ pile, needed, id);
+ return I40E_ERR_PARAM;
+ }
+
+ /* start the linear search with an imperfect hint */
+ i = pile->search_hint;
+ while (i < pile->num_entries && ret < 0) {
+ /* skip already allocated entries */
+ if (pile->list[i] & I40E_PILE_VALID_BIT) {
+ i++;
+ continue;
+ }
+
+ /* do we have enough in this lump? */
+ for (j = 0; (j < needed) && ((i+j) < pile->num_entries); j++) {
+ if (pile->list[i+j] & I40E_PILE_VALID_BIT)
+ break;
+ }
+
+ if (j == needed) {
+ /* there was enough, so assign it to the requestor */
+ for (j = 0; j < needed; j++)
+ pile->list[i+j] = id | I40E_PILE_VALID_BIT;
+ ret = i;
+ pile->search_hint = i + j;
+ } else {
+ /* not enough, so skip over it and continue looking */
+ i += j;
+ }
+ }
+
+ return ret;
+}
+
+/**
+ * i40e_put_lump - return a lump of generic resource
+ * @pile: the pile of resource to search
+ * @index: the base item index
+ * @id: the owner id of the items assigned
+ *
+ * Returns the count of items in the lump
+ **/
+static int i40e_put_lump(struct i40e_lump_tracking *pile, u16 index, u16 id)
+{
+ int valid_id = (id | I40E_PILE_VALID_BIT);
+ int count = 0;
+ int i;
+
+ if (!pile || index >= pile->num_entries)
+ return I40E_ERR_PARAM;
+
+ for (i = index;
+ i < pile->num_entries && pile->list[i] == valid_id;
+ i++) {
+ pile->list[i] = 0;
+ count++;
+ }
+
+ if (count && index < pile->search_hint)
+ pile->search_hint = index;
+
+ return count;
+}
+
+/**
+ * i40e_service_event_schedule - Schedule the service task to wake up
+ * @pf: board private structure
+ *
+ * If not already scheduled, this puts the task into the work queue
+ **/
+static void i40e_service_event_schedule(struct i40e_pf *pf)
+{
+ if (!test_bit(__I40E_DOWN, &pf->state) &&
+ !test_bit(__I40E_RESET_RECOVERY_PENDING, &pf->state) &&
+ !test_and_set_bit(__I40E_SERVICE_SCHED, &pf->state))
+ schedule_work(&pf->service_task);
+}
+
+/**
+ * i40e_tx_timeout - Respond to a Tx Hang
+ * @netdev: network interface device structure
+ *
+ * If any port has noticed a Tx timeout, it is likely that the whole
+ * device is munged, not just the one netdev port, so go for the full
+ * reset.
+ **/
+static void i40e_tx_timeout(struct net_device *netdev)
+{
+ struct i40e_netdev_priv *np = netdev_priv(netdev);
+ struct i40e_vsi *vsi = np->vsi;
+ struct i40e_pf *pf = vsi->back;
+
+ pf->tx_timeout_count++;
+
+ if (time_after(jiffies, (pf->tx_timeout_last_recovery + HZ*20)))
+ pf->tx_timeout_recovery_level = 0;
+ pf->tx_timeout_last_recovery = jiffies;
+ netdev_info(netdev, "tx_timeout recovery level %d\n",
+ pf->tx_timeout_recovery_level);
+
+ switch (pf->tx_timeout_recovery_level) {
+ case 0:
+ /* disable and re-enable queues for the VSI */
+ if (in_interrupt()) {
+ set_bit(__I40E_REINIT_REQUESTED, &pf->state);
+ set_bit(__I40E_REINIT_REQUESTED, &vsi->state);
+ } else {
+ i40e_vsi_reinit_locked(vsi);
+ }
+ break;
+ case 1:
+ set_bit(__I40E_PF_RESET_REQUESTED, &pf->state);
+ break;
+ case 2:
+ set_bit(__I40E_CORE_RESET_REQUESTED, &pf->state);
+ break;
+ case 3:
+ set_bit(__I40E_GLOBAL_RESET_REQUESTED, &pf->state);
+ break;
+ default:
+ netdev_err(netdev, "tx_timeout recovery unsuccessful\n");
+ i40e_down(vsi);
+ break;
+ }
+ i40e_service_event_schedule(pf);
+ pf->tx_timeout_recovery_level++;
+}
+
+/**
+ * i40e_release_rx_desc - Store the new tail and head values
+ * @rx_ring: ring to bump
+ * @val: new head index
+ **/
+static inline void i40e_release_rx_desc(struct i40e_ring *rx_ring, u32 val)
+{
+ rx_ring->next_to_use = val;
+
+ /* Force memory writes to complete before letting h/w
+ * know there are new descriptors to fetch. (Only
+ * applicable for weak-ordered memory model archs,
+ * such as IA-64).
+ */
+ wmb();
+ writel(val, rx_ring->tail);
+}
+
+/**
+ * i40e_get_vsi_stats_struct - Get System Network Statistics
+ * @vsi: the VSI we care about
+ *
+ * Returns the address of the device statistics structure.
+ * The statistics are actually updated from the service task.
+ **/
+struct rtnl_link_stats64 *i40e_get_vsi_stats_struct(struct i40e_vsi *vsi)
+{
+ return &vsi->net_stats;
+}
+
+/**
+ * i40e_get_netdev_stats_struct - Get statistics for netdev interface
+ * @netdev: network interface device structure
+ *
+ * Returns the address of the device statistics structure.
+ * The statistics are actually updated from the service task.
+ **/
+static struct rtnl_link_stats64 *i40e_get_netdev_stats_struct(
+ struct net_device *netdev,
+ struct rtnl_link_stats64 *storage)
+{
+ struct i40e_netdev_priv *np = netdev_priv(netdev);
+ struct i40e_vsi *vsi = np->vsi;
+
+ *storage = *i40e_get_vsi_stats_struct(vsi);
+
+ return storage;
+}
+
+/**
+ * i40e_vsi_reset_stats - Resets all stats of the given vsi
+ * @vsi: the VSI to have its stats reset
+ **/
+void i40e_vsi_reset_stats(struct i40e_vsi *vsi)
+{
+ struct rtnl_link_stats64 *ns;
+ int i;
+
+ if (!vsi)
+ return;
+
+ ns = i40e_get_vsi_stats_struct(vsi);
+ memset(ns, 0, sizeof(*ns));
+ memset(&vsi->net_stats_offsets, 0, sizeof(vsi->net_stats_offsets));
+ memset(&vsi->eth_stats, 0, sizeof(vsi->eth_stats));
+ memset(&vsi->eth_stats_offsets, 0, sizeof(vsi->eth_stats_offsets));
+ if (vsi->rx_rings)
+ for (i = 0; i < vsi->num_queue_pairs; i++) {
+ memset(&vsi->rx_rings[i].rx_stats, 0 ,
+ sizeof(vsi->rx_rings[i].rx_stats));
+ memset(&vsi->tx_rings[i].tx_stats, 0,
+ sizeof(vsi->tx_rings[i].tx_stats));
+ }
+ vsi->stat_offsets_loaded = false;
+}
+
+/**
+ * i40e_pf_reset_stats - Reset all of the stats for the given pf
+ * @pf: the PF to be reset
+ **/
+void i40e_pf_reset_stats(struct i40e_pf *pf)
+{
+ memset(&pf->stats, 0, sizeof(pf->stats));
+ memset(&pf->stats_offsets, 0, sizeof(pf->stats_offsets));
+ pf->stat_offsets_loaded = false;
+}
+
+/**
+ * i40e_stat_update48 - read and update a 48 bit stat from the chip
+ * @hw: ptr to the hardware info
+ * @hireg: the high 32 bit reg to read
+ * @loreg: the low 32 bit reg to read
+ * @offset_loaded: has the initial offset been loaded yet
+ * @offset: ptr to current offset value
+ * @stat: ptr to the stat
+ *
+ * Since the device stats are not reset at PFReset, they likely will not
+ * be zeroed when the driver starts. We'll save the first values read
+ * and use them as offsets to be subtracted from the raw values in order
+ * to report stats that count from zero. In the process, we also manage
+ * the potential roll-over.
+ **/
+static void i40e_stat_update48(struct i40e_hw *hw, u32 hireg, u32 loreg,
+ bool offset_loaded, u64 *offset, u64 *stat)
+{
+ u64 new_data;
+
+ if (hw->device_id == I40E_QEMU_DEVICE_ID) {
+ new_data = rd32(hw, loreg);
+ new_data |= ((u64)(rd32(hw, hireg) & 0xFFFF)) << 32;
+ } else {
+ new_data = rd64(hw, loreg);
+ }
+ if (!offset_loaded)
+ *offset = new_data;
+ if (likely(new_data >= *offset))
+ *stat = new_data - *offset;
+ else
+ *stat = (new_data + ((u64)1 << 48)) - *offset;
+ *stat &= 0xFFFFFFFFFFFFULL;
+}
+
+/**
+ * i40e_stat_update32 - read and update a 32 bit stat from the chip
+ * @hw: ptr to the hardware info
+ * @reg: the hw reg to read
+ * @offset_loaded: has the initial offset been loaded yet
+ * @offset: ptr to current offset value
+ * @stat: ptr to the stat
+ **/
+static void i40e_stat_update32(struct i40e_hw *hw, u32 reg,
+ bool offset_loaded, u64 *offset, u64 *stat)
+{
+ u32 new_data;
+
+ new_data = rd32(hw, reg);
+ if (!offset_loaded)
+ *offset = new_data;
+ if (likely(new_data >= *offset))
+ *stat = (u32)(new_data - *offset);
+ else
+ *stat = (u32)((new_data + ((u64)1 << 32)) - *offset);
+}
+
+/**
+ * i40e_update_eth_stats - Update VSI-specific ethernet statistics counters.
+ * @vsi: the VSI to be updated
+ **/
+void i40e_update_eth_stats(struct i40e_vsi *vsi)
+{
+ int stat_idx = le16_to_cpu(vsi->info.stat_counter_idx);
+ struct i40e_pf *pf = vsi->back;
+ struct i40e_hw *hw = &pf->hw;
+ struct i40e_eth_stats *oes;
+ struct i40e_eth_stats *es; /* device's eth stats */
+
+ es = &vsi->eth_stats;
+ oes = &vsi->eth_stats_offsets;
+
+ /* Gather up the stats that the hw collects */
+ i40e_stat_update32(hw, I40E_GLV_TEPC(stat_idx),
+ vsi->stat_offsets_loaded,
+ &oes->tx_errors, &es->tx_errors);
+ i40e_stat_update32(hw, I40E_GLV_RDPC(stat_idx),
+ vsi->stat_offsets_loaded,
+ &oes->rx_discards, &es->rx_discards);
+
+ i40e_stat_update48(hw, I40E_GLV_GORCH(stat_idx),
+ I40E_GLV_GORCL(stat_idx),
+ vsi->stat_offsets_loaded,
+ &oes->rx_bytes, &es->rx_bytes);
+ i40e_stat_update48(hw, I40E_GLV_UPRCH(stat_idx),
+ I40E_GLV_UPRCL(stat_idx),
+ vsi->stat_offsets_loaded,
+ &oes->rx_unicast, &es->rx_unicast);
+ i40e_stat_update48(hw, I40E_GLV_MPRCH(stat_idx),
+ I40E_GLV_MPRCL(stat_idx),
+ vsi->stat_offsets_loaded,
+ &oes->rx_multicast, &es->rx_multicast);
+ i40e_stat_update48(hw, I40E_GLV_BPRCH(stat_idx),
+ I40E_GLV_BPRCL(stat_idx),
+ vsi->stat_offsets_loaded,
+ &oes->rx_broadcast, &es->rx_broadcast);
+
+ i40e_stat_update48(hw, I40E_GLV_GOTCH(stat_idx),
+ I40E_GLV_GOTCL(stat_idx),
+ vsi->stat_offsets_loaded,
+ &oes->tx_bytes, &es->tx_bytes);
+ i40e_stat_update48(hw, I40E_GLV_UPTCH(stat_idx),
+ I40E_GLV_UPTCL(stat_idx),
+ vsi->stat_offsets_loaded,
+ &oes->tx_unicast, &es->tx_unicast);
+ i40e_stat_update48(hw, I40E_GLV_MPTCH(stat_idx),
+ I40E_GLV_MPTCL(stat_idx),
+ vsi->stat_offsets_loaded,
+ &oes->tx_multicast, &es->tx_multicast);
+ i40e_stat_update48(hw, I40E_GLV_BPTCH(stat_idx),
+ I40E_GLV_BPTCL(stat_idx),
+ vsi->stat_offsets_loaded,
+ &oes->tx_broadcast, &es->tx_broadcast);
+ vsi->stat_offsets_loaded = true;
+}
+
+/**
+ * i40e_update_veb_stats - Update Switch component statistics
+ * @veb: the VEB being updated
+ **/
+static void i40e_update_veb_stats(struct i40e_veb *veb)
+{
+ struct i40e_pf *pf = veb->pf;
+ struct i40e_hw *hw = &pf->hw;
+ struct i40e_eth_stats *oes;
+ struct i40e_eth_stats *es; /* device's eth stats */
+ int idx = 0;
+
+ idx = veb->stats_idx;
+ es = &veb->stats;
+ oes = &veb->stats_offsets;
+
+ /* Gather up the stats that the hw collects */
+ i40e_stat_update32(hw, I40E_GLSW_TDPC(idx),
+ veb->stat_offsets_loaded,
+ &oes->tx_discards, &es->tx_discards);
+ i40e_stat_update32(hw, I40E_GLSW_RUPP(idx),
+ veb->stat_offsets_loaded,
+ &oes->rx_unknown_protocol, &es->rx_unknown_protocol);
+
+ i40e_stat_update48(hw, I40E_GLSW_GORCH(idx), I40E_GLSW_GORCL(idx),
+ veb->stat_offsets_loaded,
+ &oes->rx_bytes, &es->rx_bytes);
+ i40e_stat_update48(hw, I40E_GLSW_UPRCH(idx), I40E_GLSW_UPRCL(idx),
+ veb->stat_offsets_loaded,
+ &oes->rx_unicast, &es->rx_unicast);
+ i40e_stat_update48(hw, I40E_GLSW_MPRCH(idx), I40E_GLSW_MPRCL(idx),
+ veb->stat_offsets_loaded,
+ &oes->rx_multicast, &es->rx_multicast);
+ i40e_stat_update48(hw, I40E_GLSW_BPRCH(idx), I40E_GLSW_BPRCL(idx),
+ veb->stat_offsets_loaded,
+ &oes->rx_broadcast, &es->rx_broadcast);
+
+ i40e_stat_update48(hw, I40E_GLSW_GOTCH(idx), I40E_GLSW_GOTCL(idx),
+ veb->stat_offsets_loaded,
+ &oes->tx_bytes, &es->tx_bytes);
+ i40e_stat_update48(hw, I40E_GLSW_UPTCH(idx), I40E_GLSW_UPTCL(idx),
+ veb->stat_offsets_loaded,
+ &oes->tx_unicast, &es->tx_unicast);
+ i40e_stat_update48(hw, I40E_GLSW_MPTCH(idx), I40E_GLSW_MPTCL(idx),
+ veb->stat_offsets_loaded,
+ &oes->tx_multicast, &es->tx_multicast);
+ i40e_stat_update48(hw, I40E_GLSW_BPTCH(idx), I40E_GLSW_BPTCL(idx),
+ veb->stat_offsets_loaded,
+ &oes->tx_broadcast, &es->tx_broadcast);
+ veb->stat_offsets_loaded = true;
+}
+
+/**
+ * i40e_update_link_xoff_rx - Update XOFF received in link flow control mode
+ * @pf: the corresponding PF
+ *
+ * Update the Rx XOFF counter (PAUSE frames) in link flow control mode
+ **/
+static void i40e_update_link_xoff_rx(struct i40e_pf *pf)
+{
+ struct i40e_hw_port_stats *osd = &pf->stats_offsets;
+ struct i40e_hw_port_stats *nsd = &pf->stats;
+ struct i40e_hw *hw = &pf->hw;
+ u64 xoff = 0;
+ u16 i, v;
+
+ if ((hw->fc.current_mode != I40E_FC_FULL) &&
+ (hw->fc.current_mode != I40E_FC_RX_PAUSE))
+ return;
+
+ xoff = nsd->link_xoff_rx;
+ i40e_stat_update32(hw, I40E_GLPRT_LXOFFRXC(hw->port),
+ pf->stat_offsets_loaded,
+ &osd->link_xoff_rx, &nsd->link_xoff_rx);
+
+ /* No new LFC xoff rx */
+ if (!(nsd->link_xoff_rx - xoff))
+ return;
+
+ /* Clear the __I40E_HANG_CHECK_ARMED bit for all Tx rings */
+ for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+ struct i40e_vsi *vsi = pf->vsi[v];
+
+ if (!vsi)
+ continue;
+
+ for (i = 0; i < vsi->num_queue_pairs; i++) {
+ struct i40e_ring *ring = &vsi->tx_rings[i];
+ clear_bit(__I40E_HANG_CHECK_ARMED, &ring->state);
+ }
+ }
+}
+
+/**
+ * i40e_update_prio_xoff_rx - Update XOFF received in PFC mode
+ * @pf: the corresponding PF
+ *
+ * Update the Rx XOFF counter (PAUSE frames) in PFC mode
+ **/
+static void i40e_update_prio_xoff_rx(struct i40e_pf *pf)
+{
+ struct i40e_hw_port_stats *osd = &pf->stats_offsets;
+ struct i40e_hw_port_stats *nsd = &pf->stats;
+ bool xoff[I40E_MAX_TRAFFIC_CLASS] = {false};
+ struct i40e_hw *hw = &pf->hw;
+ u16 i, v;
+ u8 tc;
+ struct i40e_dcbx_config *dcb_cfg = &hw->local_dcbx_config;
+
+ /* See if DCB enabled with PFC TC */
+ if (!(pf->flags & I40E_FLAG_DCB_ENABLED) ||
+ !(dcb_cfg->pfc.pfcenable)) {
+ i40e_update_link_xoff_rx(pf);
+ return;
+ }
+
+ for (i = 0; i < I40E_MAX_USER_PRIORITY; i++) {
+ u64 prio_xoff = nsd->priority_xoff_rx[i];
+ i40e_stat_update32(hw, I40E_GLPRT_PXOFFRXC(hw->port, i),
+ pf->stat_offsets_loaded,
+ &osd->priority_xoff_rx[i],
+ &nsd->priority_xoff_rx[i]);
+
+ /* No new PFC xoff rx */
+ if (!(nsd->priority_xoff_rx[i] - prio_xoff))
+ continue;
+ /* Get the TC for given priority */
+ tc = dcb_cfg->etscfg.prioritytable[i];
+ xoff[tc] = true;
+ }
+
+ /* Clear the __I40E_HANG_CHECK_ARMED bit for Tx rings */
+ for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+ struct i40e_vsi *vsi = pf->vsi[v];
+
+ if (!vsi)
+ continue;
+
+ for (i = 0; i < vsi->num_queue_pairs; i++) {
+ struct i40e_ring *ring = &vsi->tx_rings[i];
+
+ tc = ring->dcb_tc;
+ if (xoff[tc])
+ clear_bit(__I40E_HANG_CHECK_ARMED,
+ &ring->state);
+ }
+ }
+}
+
+/**
+ * i40e_update_stats - Update the board statistics counters.
+ * @vsi: the VSI to be updated
+ *
+ * There are a few instances where we store the same stat in a
+ * couple of different structs. This is partly because we have
+ * the netdev stats that need to be filled out, which is slightly
+ * different from the "eth_stats" defined by the chip and used in
+ * VF communications. We sort it all out here in a central place.
+ **/
+void i40e_update_stats(struct i40e_vsi *vsi)
+{
+ struct i40e_pf *pf = vsi->back;
+ struct i40e_hw *hw = &pf->hw;
+ struct rtnl_link_stats64 *ons;
+ struct rtnl_link_stats64 *ns; /* netdev stats */
+ struct i40e_eth_stats *oes;
+ struct i40e_eth_stats *es; /* device's eth stats */
+ u32 tx_restart, tx_busy;
+ u32 rx_page, rx_buf;
+ u64 rx_p, rx_b;
+ u64 tx_p, tx_b;
+ int i;
+ u16 q;
+
+ if (test_bit(__I40E_DOWN, &vsi->state) ||
+ test_bit(__I40E_CONFIG_BUSY, &pf->state))
+ return;
+
+ ns = i40e_get_vsi_stats_struct(vsi);
+ ons = &vsi->net_stats_offsets;
+ es = &vsi->eth_stats;
+ oes = &vsi->eth_stats_offsets;
+
+ /* Gather up the netdev and vsi stats that the driver collects
+ * on the fly during packet processing
+ */
+ rx_b = rx_p = 0;
+ tx_b = tx_p = 0;
+ tx_restart = tx_busy = 0;
+ rx_page = 0;
+ rx_buf = 0;
+ for (q = 0; q < vsi->num_queue_pairs; q++) {
+ struct i40e_ring *p;
+
+ p = &vsi->rx_rings[q];
+ rx_b += p->rx_stats.bytes;
+ rx_p += p->rx_stats.packets;
+ rx_buf += p->rx_stats.alloc_rx_buff_failed;
+ rx_page += p->rx_stats.alloc_rx_page_failed;
+
+ p = &vsi->tx_rings[q];
+ tx_b += p->tx_stats.bytes;
+ tx_p += p->tx_stats.packets;
+ tx_restart += p->tx_stats.restart_queue;
+ tx_busy += p->tx_stats.tx_busy;
+ }
+ vsi->tx_restart = tx_restart;
+ vsi->tx_busy = tx_busy;
+ vsi->rx_page_failed = rx_page;
+ vsi->rx_buf_failed = rx_buf;
+
+ ns->rx_packets = rx_p;
+ ns->rx_bytes = rx_b;
+ ns->tx_packets = tx_p;
+ ns->tx_bytes = tx_b;
+
+ i40e_update_eth_stats(vsi);
+ /* update netdev stats from eth stats */
+ ons->rx_errors = oes->rx_errors;
+ ns->rx_errors = es->rx_errors;
+ ons->tx_errors = oes->tx_errors;
+ ns->tx_errors = es->tx_errors;
+ ons->multicast = oes->rx_multicast;
+ ns->multicast = es->rx_multicast;
+ ons->tx_dropped = oes->tx_discards;
+ ns->tx_dropped = es->tx_discards;
+
+ /* Get the port data only if this is the main PF VSI */
+ if (vsi == pf->vsi[pf->lan_vsi]) {
+ struct i40e_hw_port_stats *nsd = &pf->stats;
+ struct i40e_hw_port_stats *osd = &pf->stats_offsets;
+
+ i40e_stat_update48(hw, I40E_GLPRT_GORCH(hw->port),
+ I40E_GLPRT_GORCL(hw->port),
+ pf->stat_offsets_loaded,
+ &osd->eth.rx_bytes, &nsd->eth.rx_bytes);
+ i40e_stat_update48(hw, I40E_GLPRT_GOTCH(hw->port),
+ I40E_GLPRT_GOTCL(hw->port),
+ pf->stat_offsets_loaded,
+ &osd->eth.tx_bytes, &nsd->eth.tx_bytes);
+ i40e_stat_update32(hw, I40E_GLPRT_RDPC(hw->port),
+ pf->stat_offsets_loaded,
+ &osd->eth.rx_discards,
+ &nsd->eth.rx_discards);
+ i40e_stat_update32(hw, I40E_GLPRT_TDPC(hw->port),
+ pf->stat_offsets_loaded,
+ &osd->eth.tx_discards,
+ &nsd->eth.tx_discards);
+ i40e_stat_update48(hw, I40E_GLPRT_MPRCH(hw->port),
+ I40E_GLPRT_MPRCL(hw->port),
+ pf->stat_offsets_loaded,
+ &osd->eth.rx_multicast,
+ &nsd->eth.rx_multicast);
+
+ i40e_stat_update32(hw, I40E_GLPRT_TDOLD(hw->port),
+ pf->stat_offsets_loaded,
+ &osd->tx_dropped_link_down,
+ &nsd->tx_dropped_link_down);
+
+ i40e_stat_update32(hw, I40E_GLPRT_CRCERRS(hw->port),
+ pf->stat_offsets_loaded,
+ &osd->crc_errors, &nsd->crc_errors);
+ ns->rx_crc_errors = nsd->crc_errors;
+
+ i40e_stat_update32(hw, I40E_GLPRT_ILLERRC(hw->port),
+ pf->stat_offsets_loaded,
+ &osd->illegal_bytes, &nsd->illegal_bytes);
+ ns->rx_errors = nsd->crc_errors
+ + nsd->illegal_bytes;
+
+ i40e_stat_update32(hw, I40E_GLPRT_MLFC(hw->port),
+ pf->stat_offsets_loaded,
+ &osd->mac_local_faults,
+ &nsd->mac_local_faults);
+ i40e_stat_update32(hw, I40E_GLPRT_MRFC(hw->port),
+ pf->stat_offsets_loaded,
+ &osd->mac_remote_faults,
+ &nsd->mac_remote_faults);
+
+ i40e_stat_update32(hw, I40E_GLPRT_RLEC(hw->port),
+ pf->stat_offsets_loaded,
+ &osd->rx_length_errors,
+ &nsd->rx_length_errors);
+ ns->rx_length_errors = nsd->rx_length_errors;
+
+ i40e_stat_update32(hw, I40E_GLPRT_LXONRXC(hw->port),
+ pf->stat_offsets_loaded,
+ &osd->link_xon_rx, &nsd->link_xon_rx);
+ i40e_stat_update32(hw, I40E_GLPRT_LXONTXC(hw->port),
+ pf->stat_offsets_loaded,
+ &osd->link_xon_tx, &nsd->link_xon_tx);
+ i40e_update_prio_xoff_rx(pf); /* handles I40E_GLPRT_LXOFFRXC */
+ i40e_stat_update32(hw, I40E_GLPRT_LXOFFTXC(hw->port),
+ pf->stat_offsets_loaded,
+ &osd->link_xoff_tx, &nsd->link_xoff_tx);
+
+ for (i = 0; i < 8; i++) {
+ i40e_stat_update32(hw, I40E_GLPRT_PXONRXC(hw->port, i),
+ pf->stat_offsets_loaded,
+ &osd->priority_xon_rx[i],
+ &nsd->priority_xon_rx[i]);
+ i40e_stat_update32(hw, I40E_GLPRT_PXONTXC(hw->port, i),
+ pf->stat_offsets_loaded,
+ &osd->priority_xon_tx[i],
+ &nsd->priority_xon_tx[i]);
+ i40e_stat_update32(hw, I40E_GLPRT_PXOFFTXC(hw->port, i),
+ pf->stat_offsets_loaded,
+ &osd->priority_xoff_tx[i],
+ &nsd->priority_xoff_tx[i]);
+ i40e_stat_update32(hw,
+ I40E_GLPRT_RXON2OFFCNT(hw->port, i),
+ pf->stat_offsets_loaded,
+ &osd->priority_xon_2_xoff[i],
+ &nsd->priority_xon_2_xoff[i]);
+ }
+
+ i40e_stat_update48(hw, I40E_GLPRT_PRC64H(hw->port),
+ I40E_GLPRT_PRC64L(hw->port),
+ pf->stat_offsets_loaded,
+ &osd->rx_size_64, &nsd->rx_size_64);
+ i40e_stat_update48(hw, I40E_GLPRT_PRC127H(hw->port),
+ I40E_GLPRT_PRC127L(hw->port),
+ pf->stat_offsets_loaded,
+ &osd->rx_size_127, &nsd->rx_size_127);
+ i40e_stat_update48(hw, I40E_GLPRT_PRC255H(hw->port),
+ I40E_GLPRT_PRC255L(hw->port),
+ pf->stat_offsets_loaded,
+ &osd->rx_size_255, &nsd->rx_size_255);
+ i40e_stat_update48(hw, I40E_GLPRT_PRC511H(hw->port),
+ I40E_GLPRT_PRC511L(hw->port),
+ pf->stat_offsets_loaded,
+ &osd->rx_size_511, &nsd->rx_size_511);
+ i40e_stat_update48(hw, I40E_GLPRT_PRC1023H(hw->port),
+ I40E_GLPRT_PRC1023L(hw->port),
+ pf->stat_offsets_loaded,
+ &osd->rx_size_1023, &nsd->rx_size_1023);
+ i40e_stat_update48(hw, I40E_GLPRT_PRC1522H(hw->port),
+ I40E_GLPRT_PRC1522L(hw->port),
+ pf->stat_offsets_loaded,
+ &osd->rx_size_1522, &nsd->rx_size_1522);
+ i40e_stat_update48(hw, I40E_GLPRT_PRC9522H(hw->port),
+ I40E_GLPRT_PRC9522L(hw->port),
+ pf->stat_offsets_loaded,
+ &osd->rx_size_big, &nsd->rx_size_big);
+
+ i40e_stat_update48(hw, I40E_GLPRT_PTC64H(hw->port),
+ I40E_GLPRT_PTC64L(hw->port),
+ pf->stat_offsets_loaded,
+ &osd->tx_size_64, &nsd->tx_size_64);
+ i40e_stat_update48(hw, I40E_GLPRT_PTC127H(hw->port),
+ I40E_GLPRT_PTC127L(hw->port),
+ pf->stat_offsets_loaded,
+ &osd->tx_size_127, &nsd->tx_size_127);
+ i40e_stat_update48(hw, I40E_GLPRT_PTC255H(hw->port),
+ I40E_GLPRT_PTC255L(hw->port),
+ pf->stat_offsets_loaded,
+ &osd->tx_size_255, &nsd->tx_size_255);
+ i40e_stat_update48(hw, I40E_GLPRT_PTC511H(hw->port),
+ I40E_GLPRT_PTC511L(hw->port),
+ pf->stat_offsets_loaded,
+ &osd->tx_size_511, &nsd->tx_size_511);
+ i40e_stat_update48(hw, I40E_GLPRT_PTC1023H(hw->port),
+ I40E_GLPRT_PTC1023L(hw->port),
+ pf->stat_offsets_loaded,
+ &osd->tx_size_1023, &nsd->tx_size_1023);
+ i40e_stat_update48(hw, I40E_GLPRT_PTC1522H(hw->port),
+ I40E_GLPRT_PTC1522L(hw->port),
+ pf->stat_offsets_loaded,
+ &osd->tx_size_1522, &nsd->tx_size_1522);
+ i40e_stat_update48(hw, I40E_GLPRT_PTC9522H(hw->port),
+ I40E_GLPRT_PTC9522L(hw->port),
+ pf->stat_offsets_loaded,
+ &osd->tx_size_big, &nsd->tx_size_big);
+
+ i40e_stat_update32(hw, I40E_GLPRT_RUC(hw->port),
+ pf->stat_offsets_loaded,
+ &osd->rx_undersize, &nsd->rx_undersize);
+ i40e_stat_update32(hw, I40E_GLPRT_RFC(hw->port),
+ pf->stat_offsets_loaded,
+ &osd->rx_fragments, &nsd->rx_fragments);
+ i40e_stat_update32(hw, I40E_GLPRT_ROC(hw->port),
+ pf->stat_offsets_loaded,
+ &osd->rx_oversize, &nsd->rx_oversize);
+ i40e_stat_update32(hw, I40E_GLPRT_RJC(hw->port),
+ pf->stat_offsets_loaded,
+ &osd->rx_jabber, &nsd->rx_jabber);
+ }
+
+ pf->stat_offsets_loaded = true;
+}
+
+/**
+ * i40e_find_filter - Search VSI filter list for specific mac/vlan filter
+ * @vsi: the VSI to be searched
+ * @macaddr: the MAC address
+ * @vlan: the vlan
+ * @is_vf: make sure its a vf filter, else doesn't matter
+ * @is_netdev: make sure its a netdev filter, else doesn't matter
+ *
+ * Returns ptr to the filter object or NULL
+ **/
+static struct i40e_mac_filter *i40e_find_filter(struct i40e_vsi *vsi,
+ u8 *macaddr, s16 vlan,
+ bool is_vf, bool is_netdev)
+{
+ struct i40e_mac_filter *f;
+
+ if (!vsi || !macaddr)
+ return NULL;
+
+ list_for_each_entry(f, &vsi->mac_filter_list, list) {
+ if ((ether_addr_equal(macaddr, f->macaddr)) &&
+ (vlan == f->vlan) &&
+ (!is_vf || f->is_vf) &&
+ (!is_netdev || f->is_netdev))
+ return f;
+ }
+ return NULL;
+}
+
+/**
+ * i40e_find_mac - Find a mac addr in the macvlan filters list
+ * @vsi: the VSI to be searched
+ * @macaddr: the MAC address we are searching for
+ * @is_vf: make sure its a vf filter, else doesn't matter
+ * @is_netdev: make sure its a netdev filter, else doesn't matter
+ *
+ * Returns the first filter with the provided MAC address or NULL if
+ * MAC address was not found
+ **/
+struct i40e_mac_filter *i40e_find_mac(struct i40e_vsi *vsi, u8 *macaddr,
+ bool is_vf, bool is_netdev)
+{
+ struct i40e_mac_filter *f;
+
+ if (!vsi || !macaddr)
+ return NULL;
+
+ list_for_each_entry(f, &vsi->mac_filter_list, list) {
+ if ((ether_addr_equal(macaddr, f->macaddr)) &&
+ (!is_vf || f->is_vf) &&
+ (!is_netdev || f->is_netdev))
+ return f;
+ }
+ return NULL;
+}
+
+/**
+ * i40e_is_vsi_in_vlan - Check if VSI is in vlan mode
+ * @vsi: the VSI to be searched
+ *
+ * Returns true if VSI is in vlan mode or false otherwise
+ **/
+bool i40e_is_vsi_in_vlan(struct i40e_vsi *vsi)
+{
+ struct i40e_mac_filter *f;
+
+ /* Only -1 for all the filters denotes not in vlan mode
+ * so we have to go through all the list in order to make sure
+ */
+ list_for_each_entry(f, &vsi->mac_filter_list, list) {
+ if (f->vlan >= 0)
+ return true;
+ }
+
+ return false;
+}
+
+/**
+ * i40e_put_mac_in_vlan - Make macvlan filters from macaddrs and vlans
+ * @vsi: the VSI to be searched
+ * @macaddr: the mac address to be filtered
+ * @is_vf: true if it is a vf
+ * @is_netdev: true if it is a netdev
+ *
+ * Goes through all the macvlan filters and adds a
+ * macvlan filter for each unique vlan that already exists
+ *
+ * Returns I40E_SUCCESS on success or -ENOMEM if it could not add a filter
+ **/
+i40e_status i40e_put_mac_in_vlan(struct i40e_vsi *vsi, u8 *macaddr,
+ bool is_vf, bool is_netdev)
+{
+ struct i40e_mac_filter *f, *add_f;
+
+ list_for_each_entry(f, &vsi->mac_filter_list, list) {
+ if (!i40e_find_filter(vsi, macaddr, f->vlan,
+ is_vf, is_netdev)) {
+ add_f = i40e_add_filter(vsi, macaddr, f->vlan,
+ is_vf, is_netdev);
+
+ if (!add_f) {
+ dev_info(&vsi->back->pdev->dev, "Could not add filter %d for %pM\n",
+ f->vlan, f->macaddr);
+ return -ENOMEM;
+ }
+ }
+ }
+ return I40E_SUCCESS;
+}
+
+/**
+ * i40e_add_filter - Add a mac/vlan filter to the VSI
+ * @vsi: the VSI to be searched
+ * @macaddr: the MAC address
+ * @vlan: the vlan
+ * @is_vf: make sure its a vf filter, else doesn't matter
+ * @is_netdev: make sure its a netdev filter, else doesn't matter
+ *
+ * Returns ptr to the filter object or NULL when no memory available.
+ **/
+struct i40e_mac_filter *i40e_add_filter(struct i40e_vsi *vsi,
+ u8 *macaddr, s16 vlan,
+ bool is_vf, bool is_netdev)
+{
+ struct i40e_mac_filter *f;
+
+ if (!vsi || !macaddr)
+ return NULL;
+
+ f = i40e_find_filter(vsi, macaddr, vlan, is_vf, is_netdev);
+ if (!f) {
+ f = kzalloc(sizeof(*f), GFP_ATOMIC);
+ if (!f)
+ goto add_filter_out;
+
+ memcpy(f->macaddr, macaddr, ETH_ALEN);
+ f->vlan = vlan;
+ f->changed = true;
+
+ INIT_LIST_HEAD(&f->list);
+ list_add(&f->list, &vsi->mac_filter_list);
+ }
+
+ /* increment counter and add a new flag if needed */
+ if (is_vf) {
+ if (!f->is_vf) {
+ f->is_vf = true;
+ f->counter++;
+ }
+ } else if (is_netdev) {
+ if (!f->is_netdev) {
+ f->is_netdev = true;
+ f->counter++;
+ }
+ } else {
+ f->counter++;
+ }
+
+ /* changed tells sync_filters_subtask to
+ * push the filter down to the firmware
+ */
+ if (f->changed) {
+ vsi->flags |= I40E_VSI_FLAG_FILTER_CHANGED;
+ vsi->back->flags |= I40E_FLAG_FILTER_SYNC;
+ }
+
+add_filter_out:
+ return f;
+}
+
+/**
+ * i40e_del_filter - Remove a mac/vlan filter from the VSI
+ * @vsi: the VSI to be searched
+ * @macaddr: the MAC address
+ * @vlan: the vlan
+ * @is_vf: make sure it's a vf filter, else doesn't matter
+ * @is_netdev: make sure it's a netdev filter, else doesn't matter
+ **/
+void i40e_del_filter(struct i40e_vsi *vsi,
+ u8 *macaddr, s16 vlan,
+ bool is_vf, bool is_netdev)
+{
+ struct i40e_mac_filter *f;
+
+ if (!vsi || !macaddr)
+ return;
+
+ f = i40e_find_filter(vsi, macaddr, vlan, is_vf, is_netdev);
+ if (!f || f->counter == 0)
+ return;
+
+ if (is_vf) {
+ if (f->is_vf) {
+ f->is_vf = false;
+ f->counter--;
+ }
+ } else if (is_netdev) {
+ if (f->is_netdev) {
+ f->is_netdev = false;
+ f->counter--;
+ }
+ } else {
+ /* make sure we don't remove a filter in use by vf or netdev */
+ int min_f = 0;
+ min_f += (f->is_vf ? 1 : 0);
+ min_f += (f->is_netdev ? 1 : 0);
+
+ if (f->counter > min_f)
+ f->counter--;
+ }
+
+ /* counter == 0 tells sync_filters_subtask to
+ * remove the filter from the firmware's list
+ */
+ if (f->counter == 0) {
+ f->changed = true;
+ vsi->flags |= I40E_VSI_FLAG_FILTER_CHANGED;
+ vsi->back->flags |= I40E_FLAG_FILTER_SYNC;
+ }
+}
+
+/**
+ * i40e_set_mac - NDO callback to set mac address
+ * @netdev: network interface device structure
+ * @p: pointer to an address structure
+ *
+ * Returns 0 on success, negative on failure
+ **/
+static int i40e_set_mac(struct net_device *netdev, void *p)
+{
+ struct i40e_netdev_priv *np = netdev_priv(netdev);
+ struct i40e_vsi *vsi = np->vsi;
+ struct sockaddr *addr = p;
+ struct i40e_mac_filter *f;
+
+ if (!is_valid_ether_addr(addr->sa_data))
+ return -EADDRNOTAVAIL;
+
+ netdev_info(netdev, "set mac address=%pM\n", addr->sa_data);
+
+ if (ether_addr_equal(netdev->dev_addr, addr->sa_data))
+ return 0;
+
+ if (vsi->type == I40E_VSI_MAIN) {
+ i40e_status ret;
+ ret = i40e_aq_mac_address_write(&vsi->back->hw,
+ I40E_AQC_WRITE_TYPE_LAA_ONLY,
+ addr->sa_data, NULL);
+ if (ret) {
+ netdev_info(netdev,
+ "Addr change for Main VSI failed: %d\n",
+ ret);
+ return -EADDRNOTAVAIL;
+ }
+
+ memcpy(vsi->back->hw.mac.addr, addr->sa_data, netdev->addr_len);
+ }
+
+ /* In order to be sure to not drop any packets, add the new address
+ * then delete the old one.
+ */
+ f = i40e_add_filter(vsi, addr->sa_data, I40E_VLAN_ANY, false, false);
+ if (!f)
+ return -ENOMEM;
+
+ i40e_sync_vsi_filters(vsi);
+ i40e_del_filter(vsi, netdev->dev_addr, I40E_VLAN_ANY, false, false);
+ i40e_sync_vsi_filters(vsi);
+
+ memcpy(netdev->dev_addr, addr->sa_data, netdev->addr_len);
+
+ return 0;
+}
+
+/**
+ * i40e_vsi_setup_queue_map - Setup a VSI queue map based on enabled_tc
+ * @vsi: the VSI being setup
+ * @ctxt: VSI context structure
+ * @enabled_tc: Enabled TCs bitmap
+ * @is_add: True if called before Add VSI
+ *
+ * Setup VSI queue mapping for enabled traffic classes.
+ **/
+static void i40e_vsi_setup_queue_map(struct i40e_vsi *vsi,
+ struct i40e_vsi_context *ctxt,
+ u8 enabled_tc,
+ bool is_add)
+{
+ struct i40e_pf *pf = vsi->back;
+ u16 sections = 0;
+ u8 netdev_tc = 0;
+ u16 numtc = 0;
+ u16 qcount;
+ u8 offset;
+ u16 qmap;
+ int i;
+
+ sections = I40E_AQ_VSI_PROP_QUEUE_MAP_VALID;
+ offset = 0;
+
+ if (enabled_tc && (vsi->back->flags & I40E_FLAG_DCB_ENABLED)) {
+ /* Find numtc from enabled TC bitmap */
+ for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++) {
+ if (enabled_tc & (1 << i)) /* TC is enabled */
+ numtc++;
+ }
+ if (!numtc) {
+ dev_warn(&pf->pdev->dev, "DCB is enabled but no TC enabled, forcing TC0\n");
+ numtc = 1;
+ }
+ } else {
+ /* At least TC0 is enabled in case of non-DCB case */
+ numtc = 1;
+ }
+
+ vsi->tc_config.numtc = numtc;
+ vsi->tc_config.enabled_tc = enabled_tc ? enabled_tc : 1;
+
+ /* Setup queue offset/count for all TCs for given VSI */
+ for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++) {
+ /* See if the given TC is enabled for the given VSI */
+ if (vsi->tc_config.enabled_tc & (1 << i)) { /* TC is enabled */
+ int pow, num_qps;
+
+ vsi->tc_config.tc_info[i].qoffset = offset;
+ switch (vsi->type) {
+ case I40E_VSI_MAIN:
+ if (i == 0)
+ qcount = pf->rss_size;
+ else
+ qcount = pf->num_tc_qps;
+ vsi->tc_config.tc_info[i].qcount = qcount;
+ break;
+ case I40E_VSI_FDIR:
+ case I40E_VSI_SRIOV:
+ case I40E_VSI_VMDQ2:
+ default:
+ qcount = vsi->alloc_queue_pairs;
+ vsi->tc_config.tc_info[i].qcount = qcount;
+ WARN_ON(i != 0);
+ break;
+ }
+
+ /* find the power-of-2 of the number of queue pairs */
+ num_qps = vsi->tc_config.tc_info[i].qcount;
+ pow = 0;
+ while (num_qps &&
+ ((1 << pow) < vsi->tc_config.tc_info[i].qcount)) {
+ pow++;
+ num_qps >>= 1;
+ }
+
+ vsi->tc_config.tc_info[i].netdev_tc = netdev_tc++;
+ qmap =
+ (offset << I40E_AQ_VSI_TC_QUE_OFFSET_SHIFT) |
+ (pow << I40E_AQ_VSI_TC_QUE_NUMBER_SHIFT);
+
+ offset += vsi->tc_config.tc_info[i].qcount;
+ } else {
+ /* TC is not enabled so set the offset to
+ * default queue and allocate one queue
+ * for the given TC.
+ */
+ vsi->tc_config.tc_info[i].qoffset = 0;
+ vsi->tc_config.tc_info[i].qcount = 1;
+ vsi->tc_config.tc_info[i].netdev_tc = 0;
+
+ qmap = 0;
+ }
+ ctxt->info.tc_mapping[i] = cpu_to_le16(qmap);
+ }
+
+ /* Set actual Tx/Rx queue pairs */
+ vsi->num_queue_pairs = offset;
+
+ /* Scheduler section valid can only be set for ADD VSI */
+ if (is_add) {
+ sections |= I40E_AQ_VSI_PROP_SCHED_VALID;
+
+ ctxt->info.up_enable_bits = enabled_tc;
+ }
+ if (vsi->type == I40E_VSI_SRIOV) {
+ ctxt->info.mapping_flags |=
+ cpu_to_le16(I40E_AQ_VSI_QUE_MAP_NONCONTIG);
+ for (i = 0; i < vsi->num_queue_pairs; i++)
+ ctxt->info.queue_mapping[i] =
+ cpu_to_le16(vsi->base_queue + i);
+ } else {
+ ctxt->info.mapping_flags |=
+ cpu_to_le16(I40E_AQ_VSI_QUE_MAP_CONTIG);
+ ctxt->info.queue_mapping[0] = cpu_to_le16(vsi->base_queue);
+ }
+ ctxt->info.valid_sections |= cpu_to_le16(sections);
+}
+
+/**
+ * i40e_set_rx_mode - NDO callback to set the netdev filters
+ * @netdev: network interface device structure
+ **/
+static void i40e_set_rx_mode(struct net_device *netdev)
+{
+ struct i40e_netdev_priv *np = netdev_priv(netdev);
+ struct i40e_mac_filter *f, *ftmp;
+ struct i40e_vsi *vsi = np->vsi;
+ struct netdev_hw_addr *uca;
+ struct netdev_hw_addr *mca;
+ struct netdev_hw_addr *ha;
+
+ /* add addr if not already in the filter list */
+ netdev_for_each_uc_addr(uca, netdev) {
+ if (!i40e_find_mac(vsi, uca->addr, false, true)) {
+ if (i40e_is_vsi_in_vlan(vsi))
+ i40e_put_mac_in_vlan(vsi, uca->addr,
+ false, true);
+ else
+ i40e_add_filter(vsi, uca->addr, I40E_VLAN_ANY,
+ false, true);
+ }
+ }
+
+ netdev_for_each_mc_addr(mca, netdev) {
+ if (!i40e_find_mac(vsi, mca->addr, false, true)) {
+ if (i40e_is_vsi_in_vlan(vsi))
+ i40e_put_mac_in_vlan(vsi, mca->addr,
+ false, true);
+ else
+ i40e_add_filter(vsi, mca->addr, I40E_VLAN_ANY,
+ false, true);
+ }
+ }
+
+ /* remove filter if not in netdev list */
+ list_for_each_entry_safe(f, ftmp, &vsi->mac_filter_list, list) {
+ bool found = false;
+
+ if (!f->is_netdev)
+ continue;
+
+ if (is_multicast_ether_addr(f->macaddr)) {
+ netdev_for_each_mc_addr(mca, netdev) {
+ if (ether_addr_equal(mca->addr, f->macaddr)) {
+ found = true;
+ break;
+ }
+ }
+ } else {
+ netdev_for_each_uc_addr(uca, netdev) {
+ if (ether_addr_equal(uca->addr, f->macaddr)) {
+ found = true;
+ break;
+ }
+ }
+
+ for_each_dev_addr(netdev, ha) {
+ if (ether_addr_equal(ha->addr, f->macaddr)) {
+ found = true;
+ break;
+ }
+ }
+ }
+ if (!found)
+ i40e_del_filter(
+ vsi, f->macaddr, I40E_VLAN_ANY, false, true);
+ }
+
+ /* check for other flag changes */
+ if (vsi->current_netdev_flags != vsi->netdev->flags) {
+ vsi->flags |= I40E_VSI_FLAG_FILTER_CHANGED;
+ vsi->back->flags |= I40E_FLAG_FILTER_SYNC;
+ }
+}
+
+/**
+ * i40e_sync_vsi_filters - Update the VSI filter list to the HW
+ * @vsi: ptr to the VSI
+ *
+ * Push any outstanding VSI filter changes through the AdminQ.
+ *
+ * Returns I40E_SUCCESS or error value
+ **/
+i40e_status i40e_sync_vsi_filters(struct i40e_vsi *vsi)
+{
+ struct i40e_mac_filter *f, *ftmp;
+ bool promisc_forced_on = false;
+ i40e_status ret = I40E_SUCCESS;
+ bool add_happened = false;
+ int filter_list_len = 0;
+ u32 changed_flags = 0;
+ struct i40e_pf *pf;
+ int num_add = 0;
+ int num_del = 0;
+ u16 cmd_flags;
+
+ /* empty array typed pointers, kcalloc later */
+ struct i40e_aqc_add_macvlan_element_data *add_list;
+ struct i40e_aqc_remove_macvlan_element_data *del_list;
+
+ if (!vsi)
+ return I40E_ERR_PARAM;
+ while (test_and_set_bit(__I40E_CONFIG_BUSY, &vsi->state))
+ usleep_range(1000, 2000);
+ pf = vsi->back;
+
+ if (vsi->netdev) {
+ changed_flags = vsi->current_netdev_flags ^ vsi->netdev->flags;
+ vsi->current_netdev_flags = vsi->netdev->flags;
+ }
+
+ if (vsi->flags & I40E_VSI_FLAG_FILTER_CHANGED) {
+ vsi->flags &= ~I40E_VSI_FLAG_FILTER_CHANGED;
+
+ filter_list_len = pf->hw.aq.asq_buf_size /
+ sizeof(struct i40e_aqc_remove_macvlan_element_data);
+ del_list = kcalloc(filter_list_len,
+ sizeof(struct i40e_aqc_remove_macvlan_element_data),
+ GFP_KERNEL);
+ if (!del_list)
+ return I40E_ERR_NO_MEMORY;
+
+ list_for_each_entry_safe(f, ftmp, &vsi->mac_filter_list, list) {
+ if (!f->changed)
+ continue;
+
+ if (f->counter != 0)
+ continue;
+ f->changed = false;
+ cmd_flags = 0;
+
+ /* add to delete list */
+ memcpy(del_list[num_del].mac_addr,
+ f->macaddr, ETH_ALEN);
+ del_list[num_del].vlan_tag =
+ cpu_to_le16((u16)(f->vlan ==
+ I40E_VLAN_ANY ? 0 : f->vlan));
+
+ /* vlan0 as wild card to allow packets from all vlans */
+ if (f->vlan == I40E_VLAN_ANY ||
+ (vsi->netdev && !(vsi->netdev->features &
+ NETIF_F_HW_VLAN_CTAG_FILTER)))
+ cmd_flags |= I40E_AQC_MACVLAN_DEL_IGNORE_VLAN;
+ cmd_flags |= I40E_AQC_MACVLAN_DEL_PERFECT_MATCH;
+ del_list[num_del].flags = cmd_flags;
+ num_del++;
+
+ /* unlink from filter list */
+ list_del(&f->list);
+ kfree(f);
+
+ /* flush a full buffer */
+ if (num_del == filter_list_len) {
+ ret = i40e_aq_remove_macvlan(&pf->hw,
+ vsi->seid, del_list, num_del,
+ NULL);
+ num_del = 0;
+ memset(del_list, 0, sizeof(*del_list));
+
+ if (ret != I40E_SUCCESS)
+ dev_info(&pf->pdev->dev,
+ "ignoring delete macvlan error, err %d, aq_err %d while flashing a full buffer\n",
+ ret,
+ pf->hw.aq.asq_last_status);
+ }
+ }
+ if (num_del) {
+ ret = i40e_aq_remove_macvlan(&pf->hw, vsi->seid,
+ del_list, num_del, NULL);
+ num_del = 0;
+
+ if (ret != I40E_SUCCESS)
+ dev_info(&pf->pdev->dev,
+ "ignoring delete macvlan error, err %d, aq_err %d\n",
+ ret, pf->hw.aq.asq_last_status);
+ }
+
+ kfree(del_list);
+ del_list = NULL;
+
+ /* do all the adds now */
+ filter_list_len = pf->hw.aq.asq_buf_size /
+ sizeof(struct i40e_aqc_add_macvlan_element_data),
+ add_list = kcalloc(filter_list_len,
+ sizeof(struct i40e_aqc_add_macvlan_element_data),
+ GFP_KERNEL);
+ if (!add_list)
+ return I40E_ERR_NO_MEMORY;
+
+ list_for_each_entry_safe(f, ftmp, &vsi->mac_filter_list, list) {
+ if (!f->changed)
+ continue;
+
+ if (f->counter == 0)
+ continue;
+ f->changed = false;
+ add_happened = true;
+ cmd_flags = 0;
+
+ /* add to add array */
+ memcpy(add_list[num_add].mac_addr,
+ f->macaddr, ETH_ALEN);
+ add_list[num_add].vlan_tag =
+ cpu_to_le16(
+ (u16)(f->vlan == I40E_VLAN_ANY ? 0 : f->vlan));
+ add_list[num_add].queue_number = 0;
+
+ cmd_flags |= I40E_AQC_MACVLAN_ADD_PERFECT_MATCH;
+
+ /* vlan0 as wild card to allow packets from all vlans */
+ if (f->vlan == I40E_VLAN_ANY || (vsi->netdev &&
+ !(vsi->netdev->features &
+ NETIF_F_HW_VLAN_CTAG_FILTER)))
+ cmd_flags |= I40E_AQC_MACVLAN_ADD_IGNORE_VLAN;
+ add_list[num_add].flags = cpu_to_le16(cmd_flags);
+ num_add++;
+
+ /* flush a full buffer */
+ if (num_add == filter_list_len) {
+ ret = i40e_aq_add_macvlan(&pf->hw,
+ vsi->seid,
+ add_list,
+ num_add,
+ NULL);
+ num_add = 0;
+
+ if (ret != I40E_SUCCESS)
+ break;
+ memset(add_list, 0, sizeof(*add_list));
+ }
+ }
+ if (num_add) {
+ ret = i40e_aq_add_macvlan(&pf->hw, vsi->seid,
+ add_list, num_add, NULL);
+ num_add = 0;
+ }
+ kfree(add_list);
+ add_list = NULL;
+
+ if (add_happened && (ret == I40E_SUCCESS)) {
+ /* do nothing */;
+ } else if (add_happened && (ret != I40E_SUCCESS)) {
+ dev_info(&pf->pdev->dev,
+ "add filter failed, err %d, aq_err %d\n",
+ ret, pf->hw.aq.asq_last_status);
+ if ((pf->hw.aq.asq_last_status == I40E_AQ_RC_ENOSPC) &&
+ !test_bit(__I40E_FILTER_OVERFLOW_PROMISC,
+ &vsi->state)) {
+ promisc_forced_on = true;
+ set_bit(__I40E_FILTER_OVERFLOW_PROMISC,
+ &vsi->state);
+ dev_info(&pf->pdev->dev, "promiscuous mode forced on\n");
+ }
+ }
+ }
+
+ /* check for changes in promiscuous modes */
+ if (changed_flags & IFF_ALLMULTI) {
+ bool cur_multipromisc;
+ cur_multipromisc = !!(vsi->current_netdev_flags & IFF_ALLMULTI);
+ ret = i40e_aq_set_vsi_multicast_promiscuous(&vsi->back->hw,
+ vsi->seid,
+ cur_multipromisc,
+ NULL);
+ if (ret != I40E_SUCCESS)
+ dev_info(&pf->pdev->dev,
+ "set multi promisc failed, err %d, aq_err %d\n",
+ ret, pf->hw.aq.asq_last_status);
+ }
+ if ((changed_flags & IFF_PROMISC) || promisc_forced_on) {
+ bool cur_promisc;
+ cur_promisc = (!!(vsi->current_netdev_flags & IFF_PROMISC) ||
+ test_bit(__I40E_FILTER_OVERFLOW_PROMISC,
+ &vsi->state));
+ ret = i40e_aq_set_vsi_unicast_promiscuous(&vsi->back->hw,
+ vsi->seid,
+ cur_promisc,
+ NULL);
+ if (ret != I40E_SUCCESS)
+ dev_info(&pf->pdev->dev,
+ "set uni promisc failed, err %d, aq_err %d\n",
+ ret, pf->hw.aq.asq_last_status);
+ }
+
+ clear_bit(__I40E_CONFIG_BUSY, &vsi->state);
+ return I40E_SUCCESS;
+}
+
+/**
+ * i40e_sync_filters_subtask - Sync the VSI filter list with HW
+ * @pf: board private structure
+ **/
+static void i40e_sync_filters_subtask(struct i40e_pf *pf)
+{
+ int v;
+
+ if (!pf || !(pf->flags & I40E_FLAG_FILTER_SYNC))
+ return;
+ pf->flags &= ~I40E_FLAG_FILTER_SYNC;
+
+ for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+ if (pf->vsi[v] &&
+ (pf->vsi[v]->flags & I40E_VSI_FLAG_FILTER_CHANGED))
+ i40e_sync_vsi_filters(pf->vsi[v]);
+ }
+}
+
+/**
+ * i40e_change_mtu - NDO callback to change the Maximum Transfer Unit
+ * @netdev: network interface device structure
+ * @new_mtu: new value for maximum frame size
+ *
+ * Returns 0 on success, negative on failure
+ **/
+static int i40e_change_mtu(struct net_device *netdev, int new_mtu)
+{
+ struct i40e_netdev_priv *np = netdev_priv(netdev);
+ int max_frame = new_mtu + ETH_HLEN + ETH_FCS_LEN;
+ struct i40e_vsi *vsi = np->vsi;
+
+ /* MTU < 68 is an error and causes problems on some kernels */
+ if ((new_mtu < 68) || (max_frame > I40E_MAX_RXBUFFER))
+ return -EINVAL;
+
+ netdev_info(netdev, "changing MTU from %d to %d\n",
+ netdev->mtu, new_mtu);
+ netdev->mtu = new_mtu;
+ if (netif_running(netdev))
+ i40e_vsi_reinit_locked(vsi);
+
+ return 0;
+}
+
+/**
+ * i40e_vlan_stripping_enable - Turn on vlan stripping for the VSI
+ * @vsi: the vsi being adjusted
+ **/
+void i40e_vlan_stripping_enable(struct i40e_vsi *vsi)
+{
+ struct i40e_vsi_context ctxt;
+ i40e_status ret;
+
+ if ((vsi->info.valid_sections &
+ cpu_to_le16(I40E_AQ_VSI_PROP_VLAN_VALID)) &&
+ ((vsi->info.port_vlan_flags & I40E_AQ_VSI_PVLAN_MODE_MASK) == 0))
+ return; /* already enabled */
+
+ vsi->info.valid_sections = cpu_to_le16(I40E_AQ_VSI_PROP_VLAN_VALID);
+ vsi->info.port_vlan_flags = I40E_AQ_VSI_PVLAN_MODE_ALL |
+ I40E_AQ_VSI_PVLAN_EMOD_STR_BOTH;
+
+ ctxt.seid = vsi->seid;
+ memcpy(&ctxt.info, &vsi->info, sizeof(vsi->info));
+ ret = i40e_aq_update_vsi_params(&vsi->back->hw, &ctxt, NULL);
+ if (ret != I40E_SUCCESS) {
+ dev_info(&vsi->back->pdev->dev,
+ "%s: update vsi failed, aq_err=%d\n",
+ __func__, vsi->back->hw.aq.asq_last_status);
+ }
+}
+
+/**
+ * i40e_vlan_stripping_disable - Turn off vlan stripping for the VSI
+ * @vsi: the vsi being adjusted
+ **/
+void i40e_vlan_stripping_disable(struct i40e_vsi *vsi)
+{
+ struct i40e_vsi_context ctxt;
+ i40e_status ret;
+
+ if ((vsi->info.valid_sections &
+ cpu_to_le16(I40E_AQ_VSI_PROP_VLAN_VALID)) &&
+ ((vsi->info.port_vlan_flags & I40E_AQ_VSI_PVLAN_EMOD_MASK) ==
+ I40E_AQ_VSI_PVLAN_EMOD_MASK))
+ return; /* already disabled */
+
+ vsi->info.valid_sections = cpu_to_le16(I40E_AQ_VSI_PROP_VLAN_VALID);
+ vsi->info.port_vlan_flags = I40E_AQ_VSI_PVLAN_MODE_ALL |
+ I40E_AQ_VSI_PVLAN_EMOD_NOTHING;
+
+ ctxt.seid = vsi->seid;
+ memcpy(&ctxt.info, &vsi->info, sizeof(vsi->info));
+ ret = i40e_aq_update_vsi_params(&vsi->back->hw, &ctxt, NULL);
+ if (ret != I40E_SUCCESS) {
+ dev_info(&vsi->back->pdev->dev,
+ "%s: update vsi failed, aq_err=%d\n",
+ __func__, vsi->back->hw.aq.asq_last_status);
+ }
+}
+
+/**
+ * i40e_vlan_rx_register - Setup or shutdown vlan offload
+ * @netdev: network interface to be adjusted
+ * @features: netdev features to test if VLAN offload is enabled or not
+ **/
+static void i40e_vlan_rx_register(struct net_device *netdev, u32 features)
+{
+ struct i40e_netdev_priv *np = netdev_priv(netdev);
+ struct i40e_vsi *vsi = np->vsi;
+
+ if (features & NETIF_F_HW_VLAN_CTAG_RX)
+ i40e_vlan_stripping_enable(vsi);
+ else
+ i40e_vlan_stripping_disable(vsi);
+}
+
+/**
+ * i40e_vsi_add_vlan - Add vsi membership for given vlan
+ * @vsi: the vsi being configured
+ * @vid: vlan id to be added (0 = untagged only , -1 = any)
+ **/
+int i40e_vsi_add_vlan(struct i40e_vsi *vsi, s16 vid)
+{
+ struct i40e_mac_filter *f, *add_f;
+ bool is_netdev, is_vf;
+ i40e_status ret;
+
+ is_vf = (vsi->type == I40E_VSI_SRIOV);
+ is_netdev = !!(vsi->netdev);
+
+ if (is_netdev) {
+ add_f = i40e_add_filter(vsi, vsi->netdev->dev_addr, vid,
+ is_vf, is_netdev);
+ if (!add_f) {
+ dev_info(&vsi->back->pdev->dev,
+ "Could not add vlan filter %d for %pM\n",
+ vid, vsi->netdev->dev_addr);
+ return -ENOMEM;
+ }
+ }
+
+ list_for_each_entry(f, &vsi->mac_filter_list, list) {
+ add_f = i40e_add_filter(vsi, f->macaddr, vid, is_vf, is_netdev);
+ if (!add_f) {
+ dev_info(&vsi->back->pdev->dev,
+ "Could not add vlan filter %d for %pM\n",
+ vid, f->macaddr);
+ return -ENOMEM;
+ }
+ }
+
+ ret = i40e_sync_vsi_filters(vsi);
+ if (ret != I40E_SUCCESS) {
+ dev_info(&vsi->back->pdev->dev,
+ "Could not sync filters for vid %d\n", vid);
+ return ret;
+ }
+
+ /* Now if we add a vlan tag, make sure to check if it is the first
+ * tag (i.e. a "tag" -1 does exist) and if so replace the -1 "tag"
+ * with 0, so we now accept untagged and specified tagged traffic
+ * (and not any taged and untagged)
+ */
+ if (vid > 0) {
+ if (is_netdev && i40e_find_filter(vsi, vsi->netdev->dev_addr,
+ I40E_VLAN_ANY,
+ is_vf, is_netdev)) {
+ i40e_del_filter(vsi, vsi->netdev->dev_addr,
+ I40E_VLAN_ANY, is_vf, is_netdev);
+ add_f = i40e_add_filter(vsi, vsi->netdev->dev_addr, 0,
+ is_vf, is_netdev);
+ if (!add_f) {
+ dev_info(&vsi->back->pdev->dev,
+ "Could not add filter 0 for %pM\n",
+ vsi->netdev->dev_addr);
+ return -ENOMEM;
+ }
+ }
+
+ list_for_each_entry(f, &vsi->mac_filter_list, list) {
+ if (i40e_find_filter(vsi, f->macaddr, I40E_VLAN_ANY,
+ is_vf, is_netdev)) {
+ i40e_del_filter(vsi, f->macaddr, I40E_VLAN_ANY,
+ is_vf, is_netdev);
+ add_f = i40e_add_filter(vsi, f->macaddr,
+ 0, is_vf, is_netdev);
+ if (!add_f) {
+ dev_info(&vsi->back->pdev->dev,
+ "Could not add filter 0 for %pM\n",
+ f->macaddr);
+ return -ENOMEM;
+ }
+ }
+ }
+ ret = i40e_sync_vsi_filters(vsi);
+ }
+
+ return ret;
+}
+
+/**
+ * i40e_vsi_kill_vlan - Remove vsi membership for given vlan
+ * @vsi: the vsi being configured
+ * @vid: vlan id to be removed (0 = untagged only , -1 = any)
+ **/
+int i40e_vsi_kill_vlan(struct i40e_vsi *vsi, s16 vid)
+{
+ struct net_device *netdev = vsi->netdev;
+ struct i40e_mac_filter *f, *add_f;
+ bool is_vf, is_netdev;
+ int filter_count = 0;
+ i40e_status ret;
+
+ is_vf = (vsi->type == I40E_VSI_SRIOV);
+ is_netdev = !!(netdev);
+
+ if (is_netdev)
+ i40e_del_filter(vsi, netdev->dev_addr, vid, is_vf, is_netdev);
+
+ list_for_each_entry(f, &vsi->mac_filter_list, list)
+ i40e_del_filter(vsi, f->macaddr, vid, is_vf, is_netdev);
+
+ ret = i40e_sync_vsi_filters(vsi);
+ if (ret != I40E_SUCCESS) {
+ dev_info(&vsi->back->pdev->dev, "Could not sync filters\n");
+ return ret;
+ }
+
+ /* go through all the filters for this VSI and if there is only
+ * vid == 0 it means there are no other filters, so vid 0 must
+ * be replaced with -1. This signifies that we should from now
+ * on accept any traffic (with any tag present, or untagged)
+ */
+ list_for_each_entry(f, &vsi->mac_filter_list, list) {
+ if (is_netdev) {
+ if (f->vlan &&
+ ether_addr_equal(netdev->dev_addr, f->macaddr))
+ filter_count++;
+ }
+
+ if (f->vlan)
+ filter_count++;
+ }
+
+ if (!filter_count && is_netdev) {
+ i40e_del_filter(vsi, netdev->dev_addr, 0, is_vf, is_netdev);
+ f = i40e_add_filter(vsi, netdev->dev_addr, I40E_VLAN_ANY,
+ is_vf, is_netdev);
+ if (!f) {
+ dev_info(&vsi->back->pdev->dev,
+ "Could not add filter %d for %pM\n",
+ I40E_VLAN_ANY, netdev->dev_addr);
+ return -ENOMEM;
+ }
+ }
+
+ if (!filter_count) {
+ list_for_each_entry(f, &vsi->mac_filter_list, list) {
+ i40e_del_filter(vsi, f->macaddr, 0, is_vf, is_netdev);
+ add_f = i40e_add_filter(vsi, f->macaddr, I40E_VLAN_ANY,
+ is_vf, is_netdev);
+ if (!add_f) {
+ dev_info(&vsi->back->pdev->dev,
+ "Could not add filter %d for %pM\n",
+ I40E_VLAN_ANY, f->macaddr);
+ return -ENOMEM;
+ }
+ }
+ }
+
+ return i40e_sync_vsi_filters(vsi);
+}
+
+/**
+ * i40e_vlan_rx_add_vid - Add a vlan id filter to HW offload
+ * @netdev: network interface to be adjusted
+ * @vid: vlan id to be added
+ **/
+static int i40e_vlan_rx_add_vid(struct net_device *netdev,
+ __always_unused __be16 proto, u16 vid)
+{
+ struct i40e_netdev_priv *np = netdev_priv(netdev);
+ struct i40e_vsi *vsi = np->vsi;
+ i40e_status ret;
+
+ if (vid > 4095)
+ return 0;
+
+ netdev_info(vsi->netdev, "adding %pM vid=%d\n",
+ netdev->dev_addr, vid);
+ /* If the network stack called us with vid = 0, we should
+ * indicate to i40e_vsi_add_vlan() that we want to receive
+ * any traffic (i.e. with any vlan tag, or untagged)
+ */
+ ret = i40e_vsi_add_vlan(vsi, vid ? vid : I40E_VLAN_ANY);
+
+ if (ret == I40E_SUCCESS) {
+ if (vid < VLAN_N_VID)
+ set_bit(vid, vsi->active_vlans);
+ }
+
+ return 0;
+}
+
+/**
+ * i40e_vlan_rx_kill_vid - Remove a vlan id filter from HW offload
+ * @netdev: network interface to be adjusted
+ * @vid: vlan id to be removed
+ **/
+static int i40e_vlan_rx_kill_vid(struct net_device *netdev,
+ __always_unused __be16 proto, u16 vid)
+{
+ struct i40e_netdev_priv *np = netdev_priv(netdev);
+ struct i40e_vsi *vsi = np->vsi;
+
+ netdev_info(vsi->netdev, "removing %pM vid=%d\n",
+ netdev->dev_addr, vid);
+ /* return code is ignored as there is nothing a user
+ * can do about failure to remove and a log message was
+ * already printed from another function
+ */
+ i40e_vsi_kill_vlan(vsi, vid);
+
+ clear_bit(vid, vsi->active_vlans);
+ return 0;
+}
+
+/**
+ * i40e_restore_vlan - Reinstate vlans when vsi/netdev comes back up
+ * @vsi: the vsi being brought back up
+ **/
+static void i40e_restore_vlan(struct i40e_vsi *vsi)
+{
+ u16 vid;
+
+ if (!vsi->netdev)
+ return;
+
+ i40e_vlan_rx_register(vsi->netdev, vsi->netdev->features);
+
+ for_each_set_bit(vid, vsi->active_vlans, VLAN_N_VID)
+ i40e_vlan_rx_add_vid(vsi->netdev, htons(ETH_P_8021Q),
+ vid);
+}
+
+/**
+ * i40e_vsi_add_pvid - Add pvid for the VSI
+ * @vsi: the vsi being adjusted
+ * @vid: the vlan id to set as a PVID
+ **/
+i40e_status i40e_vsi_add_pvid(struct i40e_vsi *vsi, u16 vid)
+{
+ struct i40e_vsi_context ctxt;
+ i40e_status ret;
+
+ vsi->info.valid_sections = cpu_to_le16(I40E_AQ_VSI_PROP_VLAN_VALID);
+ vsi->info.pvid = cpu_to_le16(vid);
+ vsi->info.port_vlan_flags |= I40E_AQ_VSI_PVLAN_INSERT_PVID;
+ vsi->info.port_vlan_flags |= I40E_AQ_VSI_PVLAN_MODE_UNTAGGED;
+
+ ctxt.seid = vsi->seid;
+ memcpy(&ctxt.info, &vsi->info, sizeof(vsi->info));
+ ret = i40e_aq_update_vsi_params(&vsi->back->hw, &ctxt, NULL);
+ if (ret != I40E_SUCCESS) {
+ dev_info(&vsi->back->pdev->dev,
+ "%s: update vsi failed, aq_err=%d\n",
+ __func__, vsi->back->hw.aq.asq_last_status);
+ }
+
+ return ret;
+}
+
+/**
+ * i40e_vsi_remove_pvid - Remove the pvid from the VSI
+ * @vsi: the vsi being adjusted
+ *
+ * Just use the vlan_rx_register() service to put it back to normal
+ **/
+i40e_status i40e_vsi_remove_pvid(struct i40e_vsi *vsi)
+{
+ vsi->info.pvid = 0;
+ i40e_vlan_rx_register(vsi->netdev, vsi->netdev->features);
+
+ return I40E_SUCCESS;
+}
+
+/**
+ * i40e_vsi_setup_tx_resources - Allocate VSI Tx queue resources
+ * @vsi: ptr to the VSI
+ *
+ * If this function returns with an error, then it's possible one or
+ * more of the rings is populated (while the rest are not). It is the
+ * callers duty to clean those orphaned rings.
+ *
+ * Return 0 on success, negative on failure
+ **/
+static int i40e_vsi_setup_tx_resources(struct i40e_vsi *vsi)
+{
+ int i, err = 0;
+
+ for (i = 0; i < vsi->num_queue_pairs && !err; i++)
+ err = i40e_setup_tx_descriptors(&vsi->tx_rings[i]);
+
+ return err;
+}
+
+/**
+ * i40e_vsi_free_tx_resources - Free Tx resources for VSI queues
+ * @vsi: ptr to the VSI
+ *
+ * Free VSI's transmit software resources
+ **/
+static void i40e_vsi_free_tx_resources(struct i40e_vsi *vsi)
+{
+ int i;
+
+ for (i = 0; i < vsi->num_queue_pairs; i++)
+ if (vsi->tx_rings[i].desc)
+ i40e_free_tx_resources(&vsi->tx_rings[i]);
+}
+
+/**
+ * i40e_vsi_setup_rx_resources - Allocate VSI queues Rx resources
+ * @vsi: ptr to the VSI
+ *
+ * If this function returns with an error, then it's possible one or
+ * more of the rings is populated (while the rest are not). It is the
+ * callers duty to clean those orphaned rings.
+ *
+ * Return 0 on success, negative on failure
+ **/
+static int i40e_vsi_setup_rx_resources(struct i40e_vsi *vsi)
+{
+ int i, err = 0;
+
+ for (i = 0; i < vsi->num_queue_pairs && !err; i++)
+ err = i40e_setup_rx_descriptors(&vsi->rx_rings[i]);
+ return err;
+}
+
+/**
+ * i40e_vsi_free_rx_resources - Free Rx Resources for VSI queues
+ * @vsi: ptr to the VSI
+ *
+ * Free all receive software resources
+ **/
+static void i40e_vsi_free_rx_resources(struct i40e_vsi *vsi)
+{
+ int i;
+
+ for (i = 0; i < vsi->num_queue_pairs; i++)
+ if (vsi->rx_rings[i].desc)
+ i40e_free_rx_resources(&vsi->rx_rings[i]);
+}
+
+/**
+ * i40e_configure_tx_ring - Configure a transmit ring context and rest
+ * @ring: The Tx ring to configure
+ *
+ * Configure the Tx descriptor ring in the HMC context.
+ **/
+static s32 i40e_configure_tx_ring(struct i40e_ring *ring)
+{
+ struct i40e_vsi *vsi = ring->vsi;
+ u16 pf_q = vsi->base_queue + ring->queue_index;
+ struct i40e_hw *hw = &vsi->back->hw;
+ struct i40e_hmc_obj_txq tx_ctx;
+ i40e_status err = I40E_SUCCESS;
+ u32 qtx_ctl = 0;
+
+ /* some ATR related tx ring init */
+ if (vsi->back->flags & I40E_FLAG_FDIR_ATR_ENABLED) {
+ ring->atr_sample_rate = vsi->back->atr_sample_rate;
+ ring->atr_count = 0;
+ } else {
+ ring->atr_sample_rate = 0;
+ }
+
+ /* initialize XPS */
+ if (ring->q_vector && ring->netdev &&
+ !test_and_set_bit(__I40E_TX_XPS_INIT_DONE, &ring->state))
+ netif_set_xps_queue(ring->netdev,
+ &ring->q_vector->affinity_mask,
+ ring->queue_index);
+
+ /* clear the context structure first */
+ memset(&tx_ctx, 0, sizeof(tx_ctx));
+
+ tx_ctx.new_context = 1;
+ tx_ctx.base = (ring->dma / 128);
+ tx_ctx.qlen = ring->count;
+ tx_ctx.fd_ena = !!(vsi->back->flags & (I40E_FLAG_FDIR_ENABLED |
+ I40E_FLAG_FDIR_ATR_ENABLED));
+
+ /* As part of VSI creation/update, FW allocates certain
+ * Tx arbitration queue sets for each TC enabled for
+ * the VSI. The FW returns the handles to these queue
+ * sets as part of the response buffer to Add VSI,
+ * Update VSI, etc. AQ commands. It is expected that
+ * these queue set handles be associated with the Tx
+ * queues by the driver as part of the TX queue context
+ * initialization. This has to be done regardless of
+ * DCB as by default everything is mapped to TC0.
+ */
+ tx_ctx.rdylist = le16_to_cpu(vsi->info.qs_handle[ring->dcb_tc]);
+ tx_ctx.rdylist_act = 0;
+
+ /* clear the context in the HMC */
+ err = i40e_clear_lan_tx_queue_context(hw, pf_q);
+ if (err != I40E_SUCCESS) {
+ dev_info(&vsi->back->pdev->dev,
+ "Failed to clear LAN Tx queue context on Tx ring %d (pf_q %d), error: %d\n",
+ ring->queue_index, pf_q, err);
+ return err;
+ }
+
+ /* set the context in the HMC */
+ err = i40e_set_lan_tx_queue_context(hw, pf_q, &tx_ctx);
+ if (err != I40E_SUCCESS) {
+ dev_info(&vsi->back->pdev->dev,
+ "Failed to set LAN Tx queue context on Tx ring %d (pf_q %d, error: %d\n",
+ ring->queue_index, pf_q, err);
+ return err;
+ }
+
+ /* Now associate this queue with this PCI function */
+ qtx_ctl = I40E_QTX_CTL_PF_QUEUE;
+ qtx_ctl |= ((hw->hmc.hmc_fn_id << I40E_QTX_CTL_PF_INDX_SHIFT)
+ & I40E_QTX_CTL_PF_INDX_MASK);
+ wr32(hw, I40E_QTX_CTL(pf_q), qtx_ctl);
+ i40e_flush(hw);
+
+ clear_bit(__I40E_HANG_CHECK_ARMED, &ring->state);
+
+ /* cache tail off for easier writes later */
+ ring->tail = hw->hw_addr + I40E_QTX_TAIL(pf_q);
+
+ return err;
+}
+
+/**
+ * i40e_configure_rx_ring - Configure a receive ring context
+ * @ring: The Rx ring to configure
+ *
+ * Configure the Rx descriptor ring in the HMC context.
+ **/
+static s32 i40e_configure_rx_ring(struct i40e_ring *ring)
+{
+ struct i40e_vsi *vsi = ring->vsi;
+ u32 chain_len = vsi->back->hw.func_caps.rx_buf_chain_len;
+ u16 pf_q = vsi->base_queue + ring->queue_index;
+ struct i40e_hw *hw = &vsi->back->hw;
+ i40e_status err = I40E_SUCCESS;
+ struct i40e_hmc_obj_rxq rx_ctx;
+
+ ring->state = 0;
+
+ /* clear the context structure first */
+ memset(&rx_ctx, 0, sizeof(rx_ctx));
+
+ ring->rx_buf_len = vsi->rx_buf_len;
+ ring->rx_hdr_len = vsi->rx_hdr_len;
+
+ rx_ctx.dbuff = ring->rx_buf_len >> I40E_RXQ_CTX_DBUFF_SHIFT;
+ rx_ctx.hbuff = ring->rx_hdr_len >> I40E_RXQ_CTX_HBUFF_SHIFT;
+
+ rx_ctx.base = (ring->dma / 128);
+ rx_ctx.qlen = ring->count;
+
+ if (vsi->back->flags & I40E_FLAG_16BYTE_RX_DESC_ENABLED) {
+ set_ring_16byte_desc_enabled(ring);
+ rx_ctx.dsize = 0;
+ } else {
+ rx_ctx.dsize = 1;
+ }
+
+ rx_ctx.dtype = vsi->dtype;
+ if (vsi->dtype) {
+ set_ring_ps_enabled(ring);
+ rx_ctx.hsplit_0 = I40E_RX_SPLIT_L2 |
+ I40E_RX_SPLIT_IP |
+ I40E_RX_SPLIT_TCP_UDP |
+ I40E_RX_SPLIT_SCTP;
+ } else {
+ rx_ctx.hsplit_0 = 0;
+ }
+
+ rx_ctx.rxmax = min_t(u16, vsi->max_frame,
+ (chain_len * ring->rx_buf_len));
+ rx_ctx.tphrdesc_ena = 1;
+ rx_ctx.tphwdesc_ena = 1;
+ rx_ctx.tphdata_ena = 1;
+ rx_ctx.tphhead_ena = 1;
+ rx_ctx.lrxqthresh = 2;
+ rx_ctx.crcstrip = 1;
+ rx_ctx.l2tsel = 1;
+ rx_ctx.showiv = 1;
+
+ /* clear the context in the HMC */
+ err = i40e_clear_lan_rx_queue_context(hw, pf_q);
+ if (err != I40E_SUCCESS) {
+ dev_info(&vsi->back->pdev->dev,
+ "Failed to clear LAN Rx queue context on Rx ring %d (pf_q %d), error: %d\n",
+ ring->queue_index, pf_q, err);
+ return err;
+ }
+
+ /* set the context in the HMC */
+ err = i40e_set_lan_rx_queue_context(hw, pf_q, &rx_ctx);
+ if (err != I40E_SUCCESS) {
+ dev_info(&vsi->back->pdev->dev,
+ "Failed to set LAN Rx queue context on Rx ring %d (pf_q %d), error: %d\n",
+ ring->queue_index, pf_q, err);
+ return err;
+ }
+
+ /* cache tail for quicker writes, and clear the reg before use */
+ ring->tail = hw->hw_addr + I40E_QRX_TAIL(pf_q);
+ writel(0, ring->tail);
+
+ i40e_alloc_rx_buffers(ring, I40E_DESC_UNUSED(ring));
+
+ return err;
+}
+
+/**
+ * i40e_vsi_configure_tx - Configure the VSI for Tx
+ * @vsi: VSI structure describing this set of rings and resources
+ *
+ * Configure the Tx VSI for operation.
+ **/
+static s32 i40e_vsi_configure_tx(struct i40e_vsi *vsi)
+{
+ u32 err = I40E_SUCCESS;
+ u16 i;
+
+ for (i = 0; (i < vsi->num_queue_pairs) && (err == I40E_SUCCESS); i++)
+ err = i40e_configure_tx_ring(&vsi->tx_rings[i]);
+
+ return err;
+}
+
+/**
+ * i40e_vsi_configure_rx - Configure the VSI for Rx
+ * @vsi: the VSI being configured
+ *
+ * Configure the Rx VSI for operation.
+ **/
+static s32 i40e_vsi_configure_rx(struct i40e_vsi *vsi)
+{
+ u32 err = I40E_SUCCESS;
+ u16 i;
+
+ if (vsi->netdev && (vsi->netdev->mtu > ETH_DATA_LEN))
+ vsi->max_frame = vsi->netdev->mtu + ETH_HLEN
+ + ETH_FCS_LEN + VLAN_HLEN;
+ else
+ vsi->max_frame = I40E_RXBUFFER_2048;
+
+ /* figure out correct receive buffer length */
+ switch (vsi->back->flags & (I40E_FLAG_RX_1BUF_ENABLED |
+ I40E_FLAG_RX_PS_ENABLED)) {
+ case I40E_FLAG_RX_1BUF_ENABLED:
+ vsi->rx_hdr_len = 0;
+ vsi->rx_buf_len = vsi->max_frame;
+ vsi->dtype = I40E_RX_DTYPE_NO_SPLIT;
+ break;
+ case I40E_FLAG_RX_PS_ENABLED:
+ vsi->rx_hdr_len = I40E_RX_HDR_SIZE;
+ vsi->rx_buf_len = I40E_RXBUFFER_2048;
+ vsi->dtype = I40E_RX_DTYPE_HEADER_SPLIT;
+ break;
+ default:
+ vsi->rx_hdr_len = I40E_RX_HDR_SIZE;
+ vsi->rx_buf_len = I40E_RXBUFFER_2048;
+ vsi->dtype = I40E_RX_DTYPE_SPLIT_ALWAYS;
+ break;
+ }
+
+ /* round up for the chip's needs */
+ vsi->rx_hdr_len = ALIGN(vsi->rx_hdr_len,
+ (1 << I40E_RXQ_CTX_HBUFF_SHIFT));
+ vsi->rx_buf_len = ALIGN(vsi->rx_buf_len,
+ (1 << I40E_RXQ_CTX_DBUFF_SHIFT));
+
+ /* set up individual rings */
+ for (i = 0; i < vsi->num_queue_pairs && err == I40E_SUCCESS; i++)
+ err = i40e_configure_rx_ring(&vsi->rx_rings[i]);
+
+ return err;
+}
+
+/**
+ * i40e_vsi_config_dcb_rings - Update rings to reflect DCB TC
+ * @vsi: ptr to the VSI
+ **/
+static void i40e_vsi_config_dcb_rings(struct i40e_vsi *vsi)
+{
+ u16 qoffset, qcount;
+ int i, n;
+
+ if (!(vsi->back->flags & I40E_FLAG_DCB_ENABLED))
+ return;
+
+ for (n = 0; n < I40E_MAX_TRAFFIC_CLASS; n++) {
+ if (!(vsi->tc_config.enabled_tc & (1 << n)))
+ continue;
+
+ qoffset = vsi->tc_config.tc_info[n].qoffset;
+ qcount = vsi->tc_config.tc_info[n].qcount;
+ for (i = qoffset; i < (qoffset + qcount); i++) {
+ struct i40e_ring *rx_ring = &vsi->rx_rings[i];
+ struct i40e_ring *tx_ring = &vsi->tx_rings[i];
+ rx_ring->dcb_tc = n;
+ tx_ring->dcb_tc = n;
+ }
+ }
+}
+
+/**
+ * i40e_set_vsi_rx_mode - Call set_rx_mode on a VSI
+ * @vsi: ptr to the VSI
+ **/
+static void i40e_set_vsi_rx_mode(struct i40e_vsi *vsi)
+{
+ if (vsi->netdev)
+ i40e_set_rx_mode(vsi->netdev);
+}
+
+/**
+ * i40e_vsi_configure - Set up the VSI for action
+ * @vsi: the VSI being configured
+ **/
+static s32 i40e_vsi_configure(struct i40e_vsi *vsi)
+{
+ s32 err;
+
+ i40e_set_vsi_rx_mode(vsi);
+ i40e_restore_vlan(vsi);
+ i40e_vsi_config_dcb_rings(vsi);
+ err = i40e_vsi_configure_tx(vsi);
+ if (!err)
+ err = i40e_vsi_configure_rx(vsi);
+
+ return err;
+}
+
+/**
+ * i40e_vsi_configure_msix - MSIX mode Interrupt Config in the HW
+ * @vsi: the VSI being configured
+ **/
+static void i40e_vsi_configure_msix(struct i40e_vsi *vsi)
+{
+ struct i40e_pf *pf = vsi->back;
+ struct i40e_q_vector *q_vector;
+ struct i40e_hw *hw = &pf->hw;
+ u16 vector;
+ int i, q;
+ u32 val;
+ u32 qp;
+
+ /* The interrupt indexing is offset by 1 in the PFINT_ITRn
+ * and PFINT_LNKLSTn registers, e.g.:
+ * PFINT_ITRn[0..n-1] gets msix-1..msix-n (qpair interrupts)
+ */
+ qp = vsi->base_queue;
+ vector = vsi->base_vector;
+ q_vector = vsi->q_vectors;
+ for (i = 0; i < vsi->num_q_vectors; i++, q_vector++, vector++) {
+ q_vector->rx.itr = ITR_TO_REG(vsi->rx_itr_setting);
+ q_vector->rx.latency_range = I40E_LOW_LATENCY;
+ wr32(hw, I40E_PFINT_ITRN(I40E_RX_ITR, vector - 1),
+ q_vector->rx.itr);
+ q_vector->tx.itr = ITR_TO_REG(vsi->tx_itr_setting);
+ q_vector->tx.latency_range = I40E_LOW_LATENCY;
+ wr32(hw, I40E_PFINT_ITRN(I40E_TX_ITR, vector - 1),
+ q_vector->tx.itr);
+
+ /* Linked list for the queuepairs assigned to this vector */
+ wr32(hw, I40E_PFINT_LNKLSTN(vector - 1), qp);
+ for (q = 0; q < q_vector->num_ringpairs; q++) {
+ val = I40E_QINT_RQCTL_CAUSE_ENA_MASK |
+ (I40E_RX_ITR << I40E_QINT_RQCTL_ITR_INDX_SHIFT) |
+ (vector << I40E_QINT_RQCTL_MSIX_INDX_SHIFT) |
+ (qp << I40E_QINT_RQCTL_NEXTQ_INDX_SHIFT)|
+ (I40E_QUEUE_TYPE_TX
+ << I40E_QINT_RQCTL_NEXTQ_TYPE_SHIFT);
+
+ wr32(hw, I40E_QINT_RQCTL(qp), val);
+
+ val = I40E_QINT_TQCTL_CAUSE_ENA_MASK |
+ (I40E_TX_ITR << I40E_QINT_TQCTL_ITR_INDX_SHIFT) |
+ (vector << I40E_QINT_TQCTL_MSIX_INDX_SHIFT) |
+ ((qp+1) << I40E_QINT_TQCTL_NEXTQ_INDX_SHIFT)|
+ (I40E_QUEUE_TYPE_RX
+ << I40E_QINT_TQCTL_NEXTQ_TYPE_SHIFT);
+
+ /* Terminate the linked list */
+ if (q == (q_vector->num_ringpairs - 1))
+ val |= (I40E_QUEUE_END_OF_LIST
+ << I40E_QINT_TQCTL_NEXTQ_INDX_SHIFT);
+
+ wr32(hw, I40E_QINT_TQCTL(qp), val);
+ qp++;
+ }
+ }
+
+ i40e_flush(hw);
+}
+
+/**
+ * i40e_enable_misc_int_causes - enable the non-queue interrupts
+ * @hw: ptr to the hardware info
+ **/
+static void i40e_enable_misc_int_causes(struct i40e_hw *hw)
+{
+ u32 val;
+
+ /* clear things first */
+ wr32(hw, I40E_PFINT_ICR0_ENA, 0); /* disable all */
+ rd32(hw, I40E_PFINT_ICR0); /* read to clear */
+
+ val = I40E_PFINT_ICR0_ENA_ECC_ERR_MASK |
+ I40E_PFINT_ICR0_ENA_MAL_DETECT_MASK |
+ I40E_PFINT_ICR0_ENA_GRST_MASK |
+ I40E_PFINT_ICR0_ENA_PCI_EXCEPTION_MASK |
+ I40E_PFINT_ICR0_ENA_GPIO_MASK |
+ I40E_PFINT_ICR0_ENA_STORM_DETECT_MASK |
+ I40E_PFINT_ICR0_ENA_HMC_ERR_MASK |
+ I40E_PFINT_ICR0_ENA_VFLR_MASK |
+ I40E_PFINT_ICR0_ENA_ADMINQ_MASK;
+
+ wr32(hw, I40E_PFINT_ICR0_ENA, val);
+
+ /* SW_ITR_IDX = 0, but don't change INTENA */
+ wr32(hw, I40E_PFINT_DYN_CTL0, I40E_PFINT_DYN_CTLN_SW_ITR_INDX_MASK |
+ I40E_PFINT_DYN_CTLN_INTENA_MSK_MASK);
+
+ /* OTHER_ITR_IDX = 0 */
+ wr32(hw, I40E_PFINT_STAT_CTL0, 0);
+}
+
+/**
+ * i40e_configure_msi_and_legacy - Legacy mode interrupt config in the HW
+ * @vsi: the VSI being configured
+ **/
+static void i40e_configure_msi_and_legacy(struct i40e_vsi *vsi)
+{
+ struct i40e_q_vector *q_vector = vsi->q_vectors;
+ struct i40e_pf *pf = vsi->back;
+ struct i40e_hw *hw = &pf->hw;
+ u32 val;
+
+ /* set the ITR configuration */
+ q_vector->rx.itr = ITR_TO_REG(vsi->rx_itr_setting);
+ q_vector->rx.latency_range = I40E_LOW_LATENCY;
+ wr32(hw, I40E_PFINT_ITR0(I40E_RX_ITR), q_vector->rx.itr);
+ q_vector->tx.itr = ITR_TO_REG(vsi->tx_itr_setting);
+ q_vector->tx.latency_range = I40E_LOW_LATENCY;
+ wr32(hw, I40E_PFINT_ITR0(I40E_TX_ITR), q_vector->tx.itr);
+
+ i40e_enable_misc_int_causes(hw);
+
+ /* FIRSTQ_INDX = 0, FIRSTQ_TYPE = 0 (rx) */
+ wr32(hw, I40E_PFINT_LNKLST0, 0);
+
+ /* Associate the queue pair to the vector and enable the q int */
+ val = I40E_QINT_RQCTL_CAUSE_ENA_MASK |
+ (I40E_RX_ITR << I40E_QINT_RQCTL_ITR_INDX_SHIFT) |
+ (I40E_QUEUE_TYPE_TX << I40E_QINT_TQCTL_NEXTQ_TYPE_SHIFT);
+
+ wr32(hw, I40E_QINT_RQCTL(0), val);
+
+ val = I40E_QINT_TQCTL_CAUSE_ENA_MASK |
+ (I40E_TX_ITR << I40E_QINT_TQCTL_ITR_INDX_SHIFT) |
+ (I40E_QUEUE_END_OF_LIST << I40E_QINT_TQCTL_NEXTQ_INDX_SHIFT);
+
+ wr32(hw, I40E_QINT_TQCTL(0), val);
+ i40e_flush(hw);
+}
+
+/**
+ * i40e_irq_dynamic_enable_icr0 - Enable default interrupt generation for icr0
+ * @pf: board private structure
+ **/
+static void i40e_irq_dynamic_enable_icr0(struct i40e_pf *pf)
+{
+ struct i40e_hw *hw = &pf->hw;
+ u32 val;
+
+ val = I40E_PFINT_DYN_CTL0_INTENA_MASK |
+ I40E_PFINT_DYN_CTL0_CLEARPBA_MASK |
+ (I40E_ITR_NONE << I40E_PFINT_DYN_CTL0_ITR_INDX_SHIFT);
+
+ wr32(hw, I40E_PFINT_DYN_CTL0, val);
+ i40e_flush(hw);
+}
+
+/**
+ * i40e_irq_dynamic_enable - Enable default interrupt generation settings
+ * @vsi: pointer to a vsi
+ * @vector: enable a particular Hw Interrupt vector
+ **/
+void i40e_irq_dynamic_enable(struct i40e_vsi *vsi, int vector)
+{
+ struct i40e_pf *pf = vsi->back;
+ struct i40e_hw *hw = &pf->hw;
+ u32 val;
+
+ val = I40E_PFINT_DYN_CTLN_INTENA_MASK |
+ I40E_PFINT_DYN_CTLN_CLEARPBA_MASK |
+ (I40E_ITR_NONE << I40E_PFINT_DYN_CTLN_ITR_INDX_SHIFT);
+ wr32(hw, I40E_PFINT_DYN_CTLN(vector - 1), val);
+ i40e_flush(hw);
+}
+
+/**
+ * i40e_msix_clean_rings - MSIX mode Interrupt Handler
+ * @irq: interrupt number
+ * @data: pointer to a q_vector
+ **/
+static irqreturn_t i40e_msix_clean_rings(int irq, void *data)
+{
+ struct i40e_q_vector *q_vector = data;
+
+ if (!q_vector->tx.ring[0] && !q_vector->rx.ring[0])
+ return IRQ_HANDLED;
+
+ napi_schedule(&q_vector->napi);
+
+ return IRQ_HANDLED;
+}
+
+/**
+ * i40e_fdir_clean_rings - Interrupt Handler for FDIR rings
+ * @irq: interrupt number
+ * @data: pointer to a q_vector
+ **/
+static irqreturn_t i40e_fdir_clean_rings(int irq, void *data)
+{
+ struct i40e_q_vector *q_vector = data;
+
+ if (!q_vector->tx.ring[0] && !q_vector->rx.ring[0])
+ return IRQ_HANDLED;
+
+ pr_info("fdir ring cleaning needed\n");
+
+ return IRQ_HANDLED;
+}
+
+/**
+ * i40e_vsi_request_irq_msix - Initialize MSI-X interrupts
+ * @vsi: the VSI being configured
+ * @basename: name for the vector
+ *
+ * Allocates MSI-X vectors and requests interrupts from the kernel.
+ **/
+static int i40e_vsi_request_irq_msix(struct i40e_vsi *vsi, char *basename)
+{
+ int q_vectors = vsi->num_q_vectors;
+ struct i40e_pf *pf = vsi->back;
+ int base = vsi->base_vector;
+ int rx_int_idx = 0;
+ int tx_int_idx = 0;
+ int vector, err;
+
+ for (vector = 0; vector < q_vectors; vector++) {
+ struct i40e_q_vector *q_vector = &(vsi->q_vectors[vector]);
+
+ if (q_vector->tx.ring[0] && q_vector->rx.ring[0]) {
+ snprintf(q_vector->name, sizeof(q_vector->name) - 1,
+ "%s-%s-%d", basename, "TxRx", rx_int_idx++);
+ tx_int_idx++;
+ } else if (q_vector->rx.ring[0]) {
+ snprintf(q_vector->name, sizeof(q_vector->name) - 1,
+ "%s-%s-%d", basename, "rx", rx_int_idx++);
+ } else if (q_vector->tx.ring[0]) {
+ snprintf(q_vector->name, sizeof(q_vector->name) - 1,
+ "%s-%s-%d", basename, "tx", tx_int_idx++);
+ } else {
+ /* skip this unused q_vector */
+ continue;
+ }
+ err = request_irq(pf->msix_entries[base + vector].vector,
+ vsi->irq_handler,
+ 0,
+ q_vector->name,
+ q_vector);
+ if (err) {
+ dev_info(&pf->pdev->dev,
+ "%s: request_irq failed, error: %d\n",
+ __func__, err);
+ goto free_queue_irqs;
+ }
+ /* assign the mask for this irq */
+ irq_set_affinity_hint(pf->msix_entries[base + vector].vector,
+ &q_vector->affinity_mask);
+ }
+
+ return I40E_SUCCESS;
+
+free_queue_irqs:
+ while (vector) {
+ vector--;
+ irq_set_affinity_hint(pf->msix_entries[base + vector].vector,
+ NULL);
+ free_irq(pf->msix_entries[base + vector].vector,
+ &(vsi->q_vectors[vector]));
+ }
+ return err;
+}
+
+/**
+ * i40e_vsi_disable_irq - Mask off queue interrupt generation on the VSI
+ * @vsi: the VSI being un-configured
+ **/
+static void i40e_vsi_disable_irq(struct i40e_vsi *vsi)
+{
+ struct i40e_pf *pf = vsi->back;
+ struct i40e_hw *hw = &pf->hw;
+ int base = vsi->base_vector;
+ int i;
+
+ for (i = 0; i < vsi->num_queue_pairs; i++) {
+ wr32(hw, I40E_QINT_TQCTL(vsi->tx_rings[i].reg_idx), 0);
+ wr32(hw, I40E_QINT_RQCTL(vsi->rx_rings[i].reg_idx), 0);
+ }
+
+ if (pf->flags & I40E_FLAG_MSIX_ENABLED) {
+ for (i = vsi->base_vector;
+ i < (vsi->num_q_vectors + vsi->base_vector); i++)
+ wr32(hw, I40E_PFINT_DYN_CTLN(i - 1), 0);
+
+ i40e_flush(hw);
+ for (i = 0; i < vsi->num_q_vectors; i++)
+ synchronize_irq(pf->msix_entries[i + base].vector);
+ } else {
+ /* Legacy and MSI mode - this stops all interrupt handling */
+ wr32(hw, I40E_PFINT_ICR0_ENA, 0);
+ wr32(hw, I40E_PFINT_DYN_CTL0, 0);
+ i40e_flush(hw);
+ synchronize_irq(pf->pdev->irq);
+ }
+}
+
+/**
+ * i40e_vsi_enable_irq - Enable IRQ for the given VSI
+ * @vsi: the VSI being configured
+ **/
+static int i40e_vsi_enable_irq(struct i40e_vsi *vsi)
+{
+ struct i40e_pf *pf = vsi->back;
+ int i;
+
+ if (pf->flags & I40E_FLAG_MSIX_ENABLED) {
+ for (i = vsi->base_vector;
+ i < (vsi->num_q_vectors + vsi->base_vector); i++)
+ i40e_irq_dynamic_enable(vsi, i);
+ } else {
+ i40e_irq_dynamic_enable_icr0(pf);
+ }
+
+ return I40E_SUCCESS;
+}
+
+/**
+ * i40e_stop_misc_vector - Stop the vector that handles non-queue events
+ * @pf: board private structure
+ **/
+static void i40e_stop_misc_vector(struct i40e_pf *pf)
+{
+ /* Disable ICR 0 */
+ wr32(&pf->hw, I40E_PFINT_ICR0_ENA, 0);
+ i40e_flush(&pf->hw);
+}
+
+/**
+ * i40e_intr - MSI/Legacy and non-queue interrupt handler
+ * @irq: interrupt number
+ * @data: pointer to a q_vector
+ *
+ * This is the handler used for all MSI/Legacy interrupts, and deals
+ * with both queue and non-queue interrupts. This is also used in
+ * MSIX mode to handle the non-queue interrupts.
+ **/
+static irqreturn_t i40e_intr(int irq, void *data)
+{
+ struct i40e_pf *pf = (struct i40e_pf *)data;
+ struct i40e_hw *hw = &pf->hw;
+ u32 icr0, icr0_remaining;
+ u32 val, ena_mask;
+
+ icr0 = rd32(hw, I40E_PFINT_ICR0);
+
+ /* if sharing a legacy IRQ, we might get called w/o an intr pending */
+ if ((icr0 & I40E_PFINT_ICR0_INTEVENT_MASK) == 0)
+ return IRQ_NONE;
+
+ val = rd32(hw, I40E_PFINT_DYN_CTL0);
+ val = val | I40E_PFINT_DYN_CTL0_CLEARPBA_MASK;
+ wr32(hw, I40E_PFINT_DYN_CTL0, val);
+
+ ena_mask = rd32(hw, I40E_PFINT_ICR0_ENA);
+
+ /* only q0 is used in MSI/Legacy mode, and none are used in MSIX */
+ if (icr0 & I40E_PFINT_ICR0_QUEUE_0_MASK) {
+
+ /* temporarily disable queue cause for NAPI processing */
+ u32 qval = rd32(hw, I40E_QINT_RQCTL(0));
+ qval &= ~I40E_QINT_RQCTL_CAUSE_ENA_MASK;
+ wr32(hw, I40E_QINT_RQCTL(0), qval);
+
+ qval = rd32(hw, I40E_QINT_TQCTL(0));
+ qval &= ~I40E_QINT_TQCTL_CAUSE_ENA_MASK;
+ wr32(hw, I40E_QINT_TQCTL(0), qval);
+ i40e_flush(hw);
+
+ if (!test_bit(__I40E_DOWN, &pf->state))
+ napi_schedule(&pf->vsi[pf->lan_vsi]->q_vectors[0].napi);
+ }
+
+ if (icr0 & I40E_PFINT_ICR0_ADMINQ_MASK) {
+ ena_mask &= ~I40E_PFINT_ICR0_ENA_ADMINQ_MASK;
+ set_bit(__I40E_ADMINQ_EVENT_PENDING, &pf->state);
+ }
+
+ if (icr0 & I40E_PFINT_ICR0_MAL_DETECT_MASK) {
+ ena_mask &= ~I40E_PFINT_ICR0_ENA_MAL_DETECT_MASK;
+ set_bit(__I40E_MDD_EVENT_PENDING, &pf->state);
+ }
+
+ if (icr0 & I40E_PFINT_ICR0_VFLR_MASK) {
+ ena_mask &= ~I40E_PFINT_ICR0_ENA_VFLR_MASK;
+ set_bit(__I40E_VFLR_EVENT_PENDING, &pf->state);
+ }
+
+ if (icr0 & I40E_PFINT_ICR0_GRST_MASK) {
+ if (!test_bit(__I40E_RESET_RECOVERY_PENDING, &pf->state))
+ set_bit(__I40E_RESET_INTR_RECEIVED, &pf->state);
+ ena_mask &= ~I40E_PFINT_ICR0_ENA_GRST_MASK;
+ val = rd32(hw, I40E_GLGEN_RSTAT);
+ val = (val & I40E_GLGEN_RSTAT_RESET_TYPE_MASK)
+ >> I40E_GLGEN_RSTAT_RESET_TYPE_SHIFT;
+ if (val & I40E_RESET_CORER)
+ pf->corer_count++;
+ else if (val & I40E_RESET_GLOBR)
+ pf->globr_count++;
+ else if (val & I40E_RESET_EMPR)
+ pf->empr_count++;
+ }
+
+ /* If a critical error is pending we have no choice but to reset the
+ * device.
+ * Report and mask out any remaining unexpected interrupts.
+ */
+ icr0_remaining = icr0 & ena_mask;
+ if (icr0_remaining) {
+ dev_info(&pf->pdev->dev, "unhandled interrupt icr0=0x%08x\n",
+ icr0_remaining);
+ if ((icr0_remaining & I40E_PFINT_ICR0_HMC_ERR_MASK) ||
+ (icr0_remaining & I40E_PFINT_ICR0_PE_CRITERR_MASK) ||
+ (icr0_remaining & I40E_PFINT_ICR0_PCI_EXCEPTION_MASK) ||
+ (icr0_remaining & I40E_PFINT_ICR0_ECC_ERR_MASK) ||
+ (icr0_remaining & I40E_PFINT_ICR0_MAL_DETECT_MASK)) {
+ dev_info(&pf->pdev->dev, "device will be reset\n");
+ set_bit(__I40E_PF_RESET_REQUESTED, &pf->state);
+ i40e_service_event_schedule(pf);
+ }
+ ena_mask &= ~icr0_remaining;
+ }
+
+ /* re-enable interrupt causes */
+ wr32(hw, I40E_PFINT_ICR0_ENA, ena_mask);
+ i40e_flush(hw);
+ if (!test_bit(__I40E_DOWN, &pf->state)) {
+ i40e_service_event_schedule(pf);
+ i40e_irq_dynamic_enable_icr0(pf);
+ }
+
+ return IRQ_HANDLED;
+}
+
+/**
+ * i40e_map_vector_to_rxq - Assigns the Rx queue to the vector
+ * @vsi: the VSI being configured
+ * @v_idx: vector index
+ * @r_idx: rx queue index
+ **/
+static void map_vector_to_rxq(struct i40e_vsi *vsi, int v_idx, int r_idx)
+{
+ struct i40e_q_vector *q_vector = &(vsi->q_vectors[v_idx]);
+ struct i40e_ring *rx_ring = &(vsi->rx_rings[r_idx]);
+
+ rx_ring->q_vector = q_vector;
+ q_vector->rx.ring[q_vector->rx.count] = rx_ring;
+ q_vector->rx.count++;
+ q_vector->rx.latency_range = I40E_LOW_LATENCY;
+ q_vector->vsi = vsi;
+}
+
+/**
+ * i40e_map_vector_to_txq - Assigns the Tx queue to the vector
+ * @vsi: the VSI being configured
+ * @v_idx: vector index
+ * @t_idx: tx queue index
+ **/
+static void map_vector_to_txq(struct i40e_vsi *vsi, int v_idx, int t_idx)
+{
+ struct i40e_q_vector *q_vector = &(vsi->q_vectors[v_idx]);
+ struct i40e_ring *tx_ring = &(vsi->tx_rings[t_idx]);
+
+ tx_ring->q_vector = q_vector;
+ q_vector->tx.ring[q_vector->tx.count] = tx_ring;
+ q_vector->tx.count++;
+ q_vector->tx.latency_range = I40E_LOW_LATENCY;
+ q_vector->num_ringpairs++;
+ q_vector->vsi = vsi;
+}
+
+/**
+ * i40e_vsi_map_rings_to_vectors - Maps descriptor rings to vectors
+ * @vsi: the VSI being configured
+ *
+ * This function maps descriptor rings to the queue-specific vectors
+ * we were allotted through the MSI-X enabling code. Ideally, we'd have
+ * one vector per queue pair, but on a constrained vector budget, we
+ * group the queue pairs as "efficiently" as possible.
+ **/
+static void i40e_vsi_map_rings_to_vectors(struct i40e_vsi *vsi)
+{
+ int qp_remaining = vsi->num_queue_pairs;
+ int q_vectors = vsi->num_q_vectors;
+ int qp_per_vector;
+ int v_start = 0;
+ int qp_idx = 0;
+
+ /* If we don't have enough vectors for a 1-to-1 mapping, we'll have to
+ * group them so there are multiple queues per vector.
+ */
+ for (; v_start < q_vectors && qp_remaining; v_start++) {
+ qp_per_vector = DIV_ROUND_UP(qp_remaining, q_vectors - v_start);
+ for (; qp_per_vector;
+ qp_per_vector--, qp_idx++, qp_remaining--) {
+ map_vector_to_rxq(vsi, v_start, qp_idx);
+ map_vector_to_txq(vsi, v_start, qp_idx);
+ }
+ }
+}
+
+/**
+ * i40e_vsi_request_irq - Request IRQ from the OS
+ * @vsi: the VSI being configured
+ * @basename: name for the vector
+ **/
+static int i40e_vsi_request_irq(struct i40e_vsi *vsi, char *basename)
+{
+ struct i40e_pf *pf = vsi->back;
+ int err;
+
+ if (pf->flags & I40E_FLAG_MSIX_ENABLED)
+ err = i40e_vsi_request_irq_msix(vsi, basename);
+ else if (pf->flags & I40E_FLAG_MSI_ENABLED)
+ err = request_irq(pf->pdev->irq, i40e_intr, 0,
+ pf->misc_int_name, pf);
+ else
+ err = request_irq(pf->pdev->irq, i40e_intr, IRQF_SHARED,
+ pf->misc_int_name, pf);
+
+ if (err)
+ dev_info(&pf->pdev->dev, "request_irq failed, Error %d\n", err);
+
+ return err;
+}
+
+#ifdef CONFIG_NET_POLL_CONTROLLER
+/**
+ * i40e_netpoll - A Polling 'interrupt'handler
+ * @netdev: network interface device structure
+ *
+ * This is used by netconsole to send skbs without having to re-enable
+ * interrupts. It's not called while the normal interrupt routine is executing.
+ **/
+static void i40e_netpoll(struct net_device *netdev)
+{
+ struct i40e_netdev_priv *np = netdev_priv(netdev);
+ struct i40e_vsi *vsi = np->vsi;
+ struct i40e_pf *pf = vsi->back;
+ int i;
+
+ /* if interface is down do nothing */
+ if (test_bit(__I40E_DOWN, &vsi->state))
+ return;
+
+ pf->flags |= I40E_FLAG_IN_NETPOLL;
+ if (pf->flags & I40E_FLAG_MSIX_ENABLED) {
+ for (i = 0; i < vsi->num_q_vectors; i++)
+ i40e_msix_clean_rings(0, &vsi->q_vectors[i]);
+ } else {
+ i40e_intr(pf->pdev->irq, netdev);
+ }
+ pf->flags &= ~I40E_FLAG_IN_NETPOLL;
+}
+#endif
+
+/**
+ * i40e_vsi_control_tx - Start or stop a VSI's rings
+ * @vsi: the VSI being configured
+ * @enable: start or stop the rings
+ **/
+static int i40e_vsi_control_tx(struct i40e_vsi *vsi, bool enable)
+{
+ struct i40e_pf *pf = vsi->back;
+ struct i40e_hw *hw = &pf->hw;
+ int i, j, pf_q;
+ u32 tx_reg;
+
+ pf_q = vsi->base_queue;
+ for (i = 0; i < vsi->num_queue_pairs; i++, pf_q++) {
+ j = 1000;
+ do {
+ usleep_range(1000, 2000);
+ tx_reg = rd32(hw, I40E_QTX_ENA(pf_q));
+ } while (j-- && ((tx_reg >> I40E_QTX_ENA_QENA_REQ_SHIFT)
+ ^ (tx_reg >> I40E_QTX_ENA_QENA_STAT_SHIFT)) & 1);
+
+ if (enable) {
+ /* is STAT set ? */
+ if ((tx_reg & I40E_QTX_ENA_QENA_STAT_MASK)) {
+ dev_info(&pf->pdev->dev,
+ "Tx %d already enabled\n", i);
+ continue;
+ }
+ } else {
+ /* is !STAT set ? */
+ if (!(tx_reg & I40E_QTX_ENA_QENA_STAT_MASK)) {
+ dev_info(&pf->pdev->dev,
+ "Tx %d already disabled\n", i);
+ continue;
+ }
+ }
+
+ /* turn on/off the queue */
+ if (enable)
+ tx_reg |= I40E_QTX_ENA_QENA_REQ_MASK |
+ I40E_QTX_ENA_QENA_STAT_MASK;
+ else
+ tx_reg &= ~I40E_QTX_ENA_QENA_REQ_MASK;
+
+ wr32(hw, I40E_QTX_ENA(pf_q), tx_reg);
+
+ /* wait for the change to finish */
+ for (j = 0; j < 10; j++) {
+ tx_reg = rd32(hw, I40E_QTX_ENA(pf_q));
+ if (enable) {
+ if ((tx_reg & I40E_QTX_ENA_QENA_STAT_MASK))
+ break;
+ } else {
+ if (!(tx_reg & I40E_QTX_ENA_QENA_STAT_MASK))
+ break;
+ }
+
+ udelay(10);
+ }
+ if (j >= 10) {
+ dev_info(&pf->pdev->dev, "Tx ring %d %sable timeout\n",
+ pf_q, (enable ? "en" : "dis"));
+ }
+ }
+
+ return I40E_SUCCESS;
+}
+
+/**
+ * i40e_vsi_control_rx - Start or stop a VSI's rings
+ * @vsi: the VSI being configured
+ * @enable: start or stop the rings
+ **/
+static int i40e_vsi_control_rx(struct i40e_vsi *vsi, bool enable)
+{
+ struct i40e_pf *pf = vsi->back;
+ struct i40e_hw *hw = &pf->hw;
+ int i, j, pf_q;
+ u32 rx_reg;
+
+ pf_q = vsi->base_queue;
+ for (i = 0; i < vsi->num_queue_pairs; i++, pf_q++) {
+ j = 1000;
+ do {
+ usleep_range(1000, 2000);
+ rx_reg = rd32(hw, I40E_QRX_ENA(pf_q));
+ } while (j-- && ((rx_reg >> I40E_QRX_ENA_QENA_REQ_SHIFT)
+ ^ (rx_reg >> I40E_QRX_ENA_QENA_STAT_SHIFT)) & 1);
+
+ if (enable) {
+ /* is STAT set ? */
+ if ((rx_reg & I40E_QRX_ENA_QENA_STAT_MASK))
+ continue;
+ } else {
+ /* is !STAT set ? */
+ if (!(rx_reg & I40E_QRX_ENA_QENA_STAT_MASK))
+ continue;
+ }
+
+ /* turn on/off the queue */
+ if (enable)
+ rx_reg |= I40E_QRX_ENA_QENA_REQ_MASK |
+ I40E_QRX_ENA_QENA_STAT_MASK;
+ else
+ rx_reg &= ~(I40E_QRX_ENA_QENA_REQ_MASK |
+ I40E_QRX_ENA_QENA_STAT_MASK);
+ wr32(hw, I40E_QRX_ENA(pf_q), rx_reg);
+
+ /* wait for the change to finish */
+ for (j = 0; j < 10; j++) {
+ rx_reg = rd32(hw, I40E_QRX_ENA(pf_q));
+
+ if (enable) {
+ if ((rx_reg & I40E_QRX_ENA_QENA_STAT_MASK))
+ break;
+ } else {
+ if (!(rx_reg & I40E_QRX_ENA_QENA_STAT_MASK))
+ break;
+ }
+
+ udelay(10);
+ }
+ if (j >= 10) {
+ dev_info(&pf->pdev->dev, "Rx ring %d %sable timeout\n",
+ pf_q, (enable ? "en" : "dis"));
+ return I40E_ERR_TIMEOUT;
+ }
+ }
+
+ return I40E_SUCCESS;
+}
+
+/**
+ * i40e_vsi_control_rings - Start or stop a VSI's rings
+ * @vsi: the VSI being configured
+ * @enable: start or stop the rings
+ **/
+static int i40e_vsi_control_rings(struct i40e_vsi *vsi, bool request)
+{
+ int ret;
+
+ /* do rx first for enable and last for disable */
+ if (request) {
+ ret = i40e_vsi_control_rx(vsi, request);
+ if (ret)
+ return ret;
+ ret = i40e_vsi_control_tx(vsi, request);
+ } else {
+ ret = i40e_vsi_control_tx(vsi, request);
+ if (ret)
+ return ret;
+ ret = i40e_vsi_control_rx(vsi, request);
+ }
+
+ return ret;
+}
+
+/**
+ * i40e_vsi_free_irq - Free the irq association with the OS
+ * @vsi: the VSI being configured
+ **/
+static void i40e_vsi_free_irq(struct i40e_vsi *vsi)
+{
+ struct i40e_pf *pf = vsi->back;
+ struct i40e_hw *hw = &pf->hw;
+ int base = vsi->base_vector;
+ u32 val, qp;
+ int i;
+
+ if (pf->flags & I40E_FLAG_MSIX_ENABLED) {
+ if (!vsi->q_vectors)
+ return;
+
+ for (i = 0; i < vsi->num_q_vectors; i++) {
+ u16 vector = i + base;
+
+ /* free only the irqs that were actually requested */
+ if (vsi->q_vectors[i].num_ringpairs == 0)
+ continue;
+
+ /* clear the affinity_mask in the IRQ descriptor */
+ irq_set_affinity_hint(pf->msix_entries[vector].vector,
+ NULL);
+ free_irq(pf->msix_entries[vector].vector,
+ &vsi->q_vectors[i]);
+
+ /* Tear down the interrupt queue link list
+ *
+ * We know that they come in pairs and always
+ * the Rx first, then the Tx. To clear the
+ * link list, stick the EOL value into the
+ * next_q field of the registers.
+ */
+ val = rd32(hw, I40E_PFINT_LNKLSTN(vector - 1));
+ qp = (val & I40E_PFINT_LNKLSTN_FIRSTQ_INDX_MASK)
+ >> I40E_PFINT_LNKLSTN_FIRSTQ_INDX_SHIFT;
+ val |= I40E_QUEUE_END_OF_LIST
+ << I40E_PFINT_LNKLSTN_FIRSTQ_INDX_SHIFT;
+ wr32(hw, I40E_PFINT_LNKLSTN(vector - 1), val);
+
+ while (qp != I40E_QUEUE_END_OF_LIST) {
+ u32 next;
+
+ val = rd32(hw, I40E_QINT_RQCTL(qp));
+
+ val &= ~(I40E_QINT_RQCTL_MSIX_INDX_MASK |
+ I40E_QINT_RQCTL_MSIX0_INDX_MASK |
+ I40E_QINT_RQCTL_CAUSE_ENA_MASK |
+ I40E_QINT_RQCTL_INTEVENT_MASK);
+
+ val |= (I40E_QINT_RQCTL_ITR_INDX_MASK |
+ I40E_QINT_RQCTL_NEXTQ_INDX_MASK);
+
+ wr32(hw, I40E_QINT_RQCTL(qp), val);
+
+ val = rd32(hw, I40E_QINT_TQCTL(qp));
+
+ next = (val & I40E_QINT_TQCTL_NEXTQ_INDX_MASK)
+ >> I40E_QINT_TQCTL_NEXTQ_INDX_SHIFT;
+
+ val &= ~(I40E_QINT_TQCTL_MSIX_INDX_MASK |
+ I40E_QINT_TQCTL_MSIX0_INDX_MASK |
+ I40E_QINT_TQCTL_CAUSE_ENA_MASK |
+ I40E_QINT_TQCTL_INTEVENT_MASK);
+
+ val |= (I40E_QINT_TQCTL_ITR_INDX_MASK |
+ I40E_QINT_TQCTL_NEXTQ_INDX_MASK);
+
+ wr32(hw, I40E_QINT_TQCTL(qp), val);
+ qp = next;
+ }
+ }
+ } else {
+ free_irq(pf->pdev->irq, pf);
+
+ val = rd32(hw, I40E_PFINT_LNKLST0);
+ qp = (val & I40E_PFINT_LNKLSTN_FIRSTQ_INDX_MASK)
+ >> I40E_PFINT_LNKLSTN_FIRSTQ_INDX_SHIFT;
+ val |= I40E_QUEUE_END_OF_LIST
+ << I40E_PFINT_LNKLST0_FIRSTQ_INDX_SHIFT;
+ wr32(hw, I40E_PFINT_LNKLST0, val);
+
+ val = rd32(hw, I40E_QINT_RQCTL(qp));
+ val &= ~(I40E_QINT_RQCTL_MSIX_INDX_MASK |
+ I40E_QINT_RQCTL_MSIX0_INDX_MASK |
+ I40E_QINT_RQCTL_CAUSE_ENA_MASK |
+ I40E_QINT_RQCTL_INTEVENT_MASK);
+
+ val |= (I40E_QINT_RQCTL_ITR_INDX_MASK |
+ I40E_QINT_RQCTL_NEXTQ_INDX_MASK);
+
+ wr32(hw, I40E_QINT_RQCTL(qp), val);
+
+ val = rd32(hw, I40E_QINT_TQCTL(qp));
+
+ val &= ~(I40E_QINT_TQCTL_MSIX_INDX_MASK |
+ I40E_QINT_TQCTL_MSIX0_INDX_MASK |
+ I40E_QINT_TQCTL_CAUSE_ENA_MASK |
+ I40E_QINT_TQCTL_INTEVENT_MASK);
+
+ val |= (I40E_QINT_TQCTL_ITR_INDX_MASK |
+ I40E_QINT_TQCTL_NEXTQ_INDX_MASK);
+
+ wr32(hw, I40E_QINT_TQCTL(qp), val);
+ }
+}
+
+/**
+ * i40e_vsi_free_q_vectors - Free memory allocated for interrupt vectors
+ * @vsi: the VSI being un-configured
+ *
+ * This frees the memory allocated to the q_vectors and
+ * deletes references to the NAPI struct.
+ **/
+static void i40e_vsi_free_q_vectors(struct i40e_vsi *vsi)
+{
+ int v_idx;
+
+ for (v_idx = 0; v_idx < vsi->num_q_vectors; v_idx++) {
+ struct i40e_q_vector *q_vector = &vsi->q_vectors[v_idx];
+ int r_idx;
+
+ if (!q_vector)
+ continue;
+
+ /* disassociate q_vector from rings */
+ for (r_idx = 0; r_idx < q_vector->tx.count; r_idx++)
+ q_vector->tx.ring[r_idx]->q_vector = NULL;
+ for (r_idx = 0; r_idx < q_vector->rx.count; r_idx++)
+ q_vector->rx.ring[r_idx]->q_vector = NULL;
+
+ /* only VSI w/ an associated netdev is set up w/ NAPI */
+ if (vsi->netdev)
+ netif_napi_del(&q_vector->napi);
+ }
+ kfree(vsi->q_vectors);
+}
+
+/**
+ * i40e_reset_interrupt_capability - Disable interrupt setup in OS
+ * @pf: board private structure
+ **/
+static void i40e_reset_interrupt_capability(struct i40e_pf *pf)
+{
+ /* If we're in Legacy mode, the interrupt was cleaned in vsi_close */
+ if (pf->flags & I40E_FLAG_MSIX_ENABLED) {
+ pci_disable_msix(pf->pdev);
+ kfree(pf->msix_entries);
+ pf->msix_entries = NULL;
+ } else if (pf->flags & I40E_FLAG_MSI_ENABLED) {
+ pci_disable_msi(pf->pdev);
+ }
+ pf->flags &= ~(I40E_FLAG_MSIX_ENABLED | I40E_FLAG_MSI_ENABLED);
+}
+
+/**
+ * i40e_clear_interrupt_scheme - Clear the current interrupt scheme settings
+ * @pf: board private structure
+ *
+ * We go through and clear interrupt specific resources and reset the structure
+ * to pre-load conditions
+ **/
+static void i40e_clear_interrupt_scheme(struct i40e_pf *pf)
+{
+ int i;
+
+ i40e_put_lump(pf->irq_pile, 0, I40E_PILE_VALID_BIT-1);
+ for (i = 0; i < pf->hw.func_caps.num_vsis; i++)
+ if (pf->vsi[i])
+ i40e_vsi_free_q_vectors(pf->vsi[i]);
+ i40e_reset_interrupt_capability(pf);
+}
+
+/**
+ * i40e_napi_enable_all - Enable NAPI for all q_vectors in the VSI
+ * @vsi: the VSI being configured
+ **/
+static void i40e_napi_enable_all(struct i40e_vsi *vsi)
+{
+ int q_idx;
+
+ if (!vsi->netdev)
+ return;
+
+ for (q_idx = 0; q_idx < vsi->num_q_vectors; q_idx++)
+ napi_enable(&vsi->q_vectors[q_idx].napi);
+}
+
+/**
+ * i40e_napi_disable_all - Disable NAPI for all q_vectors in the VSI
+ * @vsi: the VSI being configured
+ **/
+static void i40e_napi_disable_all(struct i40e_vsi *vsi)
+{
+ int q_idx;
+
+ if (!vsi->netdev)
+ return;
+
+ for (q_idx = 0; q_idx < vsi->num_q_vectors; q_idx++)
+ napi_disable(&vsi->q_vectors[q_idx].napi);
+}
+
+/**
+ * i40e_quiesce_vsi - Pause a given VSI
+ * @vsi: the VSI being paused
+ **/
+static void i40e_quiesce_vsi(struct i40e_vsi *vsi)
+{
+ if (test_bit(__I40E_DOWN, &vsi->state))
+ return;
+
+ set_bit(__I40E_NEEDS_RESTART, &vsi->state);
+ if (vsi->netdev && netif_running(vsi->netdev)) {
+ vsi->netdev->netdev_ops->ndo_stop(vsi->netdev);
+ } else {
+ set_bit(__I40E_DOWN, &vsi->state);
+ i40e_down(vsi);
+ }
+}
+
+/**
+ * i40e_unquiesce_vsi - Resume a given VSI
+ * @vsi: the VSI being resumed
+ **/
+static void i40e_unquiesce_vsi(struct i40e_vsi *vsi)
+{
+ if (!test_bit(__I40E_NEEDS_RESTART, &vsi->state))
+ return;
+
+ clear_bit(__I40E_NEEDS_RESTART, &vsi->state);
+ if (vsi->netdev && netif_running(vsi->netdev))
+ vsi->netdev->netdev_ops->ndo_open(vsi->netdev);
+ else
+ i40e_up(vsi); /* this clears the DOWN bit */
+}
+
+/**
+ * i40e_pf_quiesce_all_vsi - Pause all VSIs on a PF
+ * @pf: the PF
+ **/
+static void i40e_pf_quiesce_all_vsi(struct i40e_pf *pf)
+{
+ int v;
+
+ for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+ if (pf->vsi[v])
+ i40e_quiesce_vsi(pf->vsi[v]);
+ }
+}
+
+/**
+ * i40e_pf_unquiesce_all_vsi - Resume all VSIs on a PF
+ * @pf: the PF
+ **/
+static void i40e_pf_unquiesce_all_vsi(struct i40e_pf *pf)
+{
+ int v;
+
+ for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+ if (pf->vsi[v])
+ i40e_unquiesce_vsi(pf->vsi[v]);
+ }
+}
+
+/**
+ * i40e_dcb_get_num_tc - Get the number of TCs from DCBx config
+ * @dcbcfg: the corresponding DCBx configuration structure
+ *
+ * Return the number of TCs from given DCBx configuration
+ **/
+static u8 i40e_dcb_get_num_tc(struct i40e_dcbx_config *dcbcfg)
+{
+ int num_tc = 0, i;
+
+ /* Scan the ETS Config Priority Table to find
+ * traffic class enabled for a given priority
+ * and use the traffic class index to get the
+ * number of traffic classes enabled
+ */
+ for (i = 0; i < I40E_MAX_USER_PRIORITY; i++) {
+ if (dcbcfg->etscfg.prioritytable[i] > num_tc)
+ num_tc = dcbcfg->etscfg.prioritytable[i];
+ }
+
+ /* Traffic class index starts from zero so
+ * increment to return the actual count
+ */
+ num_tc++;
+
+ return num_tc;
+}
+
+/**
+ * i40e_dcb_get_enabled_tc - Get enabled traffic classes
+ * @dcbcfg: the corresponding DCBx configuration structure
+ *
+ * Query the current DCB configuration and return the number of
+ * traffic classes enabled from the given DCBX config
+ **/
+static u8 i40e_dcb_get_enabled_tc(struct i40e_dcbx_config *dcbcfg)
+{
+ u8 num_tc = i40e_dcb_get_num_tc(dcbcfg);
+ u8 enabled_tc = 1;
+ u8 i;
+
+ for (i = 0; i < num_tc; i++)
+ enabled_tc |= 1 << i;
+
+ return enabled_tc;
+}
+
+/**
+ * i40e_pf_get_num_tc - Get enabled traffic classes for PF
+ * @pf: PF being queried
+ *
+ * Return number of traffic classes enabled for the given PF
+ **/
+static u8 i40e_pf_get_num_tc(struct i40e_pf *pf)
+{
+ struct i40e_hw *hw = &pf->hw;
+ u8 i, enabled_tc;
+ u8 num_tc = 0;
+ struct i40e_dcbx_config *dcbcfg = &hw->local_dcbx_config;
+
+ /* If DCB is not enabled then always in single TC */
+ if (!(pf->flags & I40E_FLAG_DCB_ENABLED))
+ return 1;
+
+ /* MFP mode return count of enabled TCs for this PF */
+ if (pf->flags & I40E_FLAG_MFP_ENABLED) {
+ enabled_tc = pf->hw.func_caps.enabled_tcmap;
+ for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++) {
+ if (enabled_tc & (1 << i))
+ num_tc++;
+ }
+ return num_tc;
+ }
+
+ /* SFP mode will be enabled for all TCs on port */
+ return i40e_dcb_get_num_tc(dcbcfg);
+}
+
+/**
+ * i40e_pf_get_default_tc - Get bitmap for first enabled TC
+ * @pf: PF being queried
+ *
+ * Return a bitmap for first enabled traffic class for this PF.
+ **/
+static u8 i40e_pf_get_default_tc(struct i40e_pf *pf)
+{
+ u8 enabled_tc = pf->hw.func_caps.enabled_tcmap;
+ u8 i = 0;
+
+ if (!enabled_tc)
+ return 0x1; /* TC0 */
+
+ /* Find the first enabled TC */
+ for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++) {
+ if (enabled_tc & (1 << i))
+ break;
+ }
+
+ return 1 << i;
+}
+
+/**
+ * i40e_pf_get_pf_tc_map - Get bitmap for enabled traffic classes
+ * @pf: PF being queried
+ *
+ * Return a bitmap for enabled traffic classes for this PF.
+ **/
+static u8 i40e_pf_get_tc_map(struct i40e_pf *pf)
+{
+ /* If DCB is not enabled for this PF then just return default TC */
+ if (!(pf->flags & I40E_FLAG_DCB_ENABLED))
+ return i40e_pf_get_default_tc(pf);
+
+ /* MFP mode will have enabled TCs set by FW */
+ if (pf->flags & I40E_FLAG_MFP_ENABLED)
+ return pf->hw.func_caps.enabled_tcmap;
+
+ /* SFP mode we want PF to be enabled for all TCs */
+ return i40e_dcb_get_enabled_tc(&pf->hw.local_dcbx_config);
+}
+
+/**
+ * i40e_vsi_get_bw_info - Query VSI BW Information
+ * @vsi: the VSI being queried
+ *
+ * Returns 0 on success, negative value on failure
+ **/
+static s32 i40e_vsi_get_bw_info(struct i40e_vsi *vsi)
+{
+ struct i40e_aqc_query_vsi_ets_sla_config_resp bw_ets_config = {0};
+ struct i40e_aqc_query_vsi_bw_config_resp bw_config = {0};
+ int ret = I40E_ERR_NOT_IMPLEMENTED;
+ struct i40e_pf *pf = vsi->back;
+ struct i40e_hw *hw = &pf->hw;
+ u32 tc_bw_max;
+ int i;
+
+ /* Get the VSI level BW configuration */
+ ret = i40e_aq_query_vsi_bw_config(hw, vsi->seid, &bw_config, NULL);
+ if (ret) {
+ dev_info(&pf->pdev->dev,
+ "couldn't get pf vsi bw config, err %d, aq_err %d\n",
+ ret, pf->hw.aq.asq_last_status);
+ return ret;
+ }
+
+ /* Get the VSI level BW configuration per TC */
+ ret = i40e_aq_query_vsi_ets_sla_config(hw, vsi->seid,
+ &bw_ets_config,
+ NULL);
+ if (ret) {
+ dev_info(&pf->pdev->dev,
+ "couldn't get pf vsi ets bw config, err %d, aq_err %d\n",
+ ret, pf->hw.aq.asq_last_status);
+ return ret;
+ }
+
+ if (bw_config.tc_valid_bits != bw_ets_config.tc_valid_bits) {
+ dev_info(&pf->pdev->dev,
+ "Enabled TCs mismatch from querying VSI BW info 0x%08x 0x%08x\n",
+ bw_config.tc_valid_bits,
+ bw_ets_config.tc_valid_bits);
+ /* Still continuing */
+ }
+
+ vsi->bw_limit = le16_to_cpu(bw_config.port_bw_limit);
+ vsi->bw_max_quanta = bw_config.max_bw;
+ tc_bw_max = le16_to_cpu(bw_ets_config.tc_bw_max[0]) |
+ (le16_to_cpu(bw_ets_config.tc_bw_max[1]) << 16);
+ for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++) {
+ vsi->bw_ets_share_credits[i] = bw_ets_config.share_credits[i];
+ vsi->bw_ets_limit_credits[i] =
+ le16_to_cpu(bw_ets_config.credits[i]);
+ /* 3 bits out of 4 for each TC */
+ vsi->bw_ets_max_quanta[i] = (u8)((tc_bw_max >> (i*4)) & 0x7);
+ }
+ return ret;
+}
+
+/**
+ * i40e_vsi_configure_bw_alloc - Configure VSI BW allocation per TC
+ * @vsi: the VSI being configured
+ * @enabled_tc: TC bitmap
+ * @bw_credits: BW shared credits per TC
+ *
+ * Returns 0 on success, negative value on failure
+ **/
+static s32 i40e_vsi_configure_bw_alloc(struct i40e_vsi *vsi,
+ u8 enabled_tc,
+ u8 *bw_share)
+{
+ struct i40e_aqc_configure_vsi_tc_bw_data bw_data;
+ int i, ret = 0;
+
+ bw_data.tc_valid_bits = enabled_tc;
+ for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++)
+ bw_data.tc_bw_credits[i] = bw_share[i];
+
+ ret = i40e_aq_config_vsi_tc_bw(&vsi->back->hw, vsi->seid,
+ &bw_data, NULL);
+ if (ret != I40E_SUCCESS) {
+ dev_info(&vsi->back->pdev->dev,
+ "%s: AQ command Config VSI BW allocation per TC failed = %d\n",
+ __func__, vsi->back->hw.aq.asq_last_status);
+ return ret;
+ }
+
+ for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++)
+ vsi->info.qs_handle[i] = bw_data.qs_handles[i];
+
+ return ret;
+}
+
+/**
+ * i40e_vsi_config_netdev_tc - Setup the netdev TC configuration
+ * @vsi: the VSI being configured
+ * @enabled_tc: TC map to be enabled
+ *
+ **/
+static void i40e_vsi_config_netdev_tc(struct i40e_vsi *vsi, u8 enabled_tc)
+{
+ struct net_device *netdev = vsi->netdev;
+ struct i40e_pf *pf = vsi->back;
+ struct i40e_hw *hw = &pf->hw;
+ u8 netdev_tc = 0;
+ int i;
+ struct i40e_dcbx_config *dcbcfg = &hw->local_dcbx_config;
+
+ if (!netdev)
+ return;
+
+ if (!enabled_tc) {
+ netdev_reset_tc(netdev);
+ return;
+ }
+
+ /* Set up actual enabled TCs on the VSI */
+ if (netdev_set_num_tc(netdev, vsi->tc_config.numtc))
+ return;
+
+ /* set per TC queues for the VSI */
+ for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++) {
+ /* Only set TC queues for enabled tcs
+ *
+ * e.g. For a VSI that has TC0 and TC3 enabled the
+ * enabled_tc bitmap would be 0x00001001; the driver
+ * will set the numtc for netdev as 2 that will be
+ * referenced by the netdev layer as TC 0 and 1.
+ */
+ if (vsi->tc_config.enabled_tc & (1 << i))
+ netdev_set_tc_queue(netdev,
+ vsi->tc_config.tc_info[i].netdev_tc,
+ vsi->tc_config.tc_info[i].qcount,
+ vsi->tc_config.tc_info[i].qoffset);
+ }
+
+ /* Assign UP2TC map for the VSI */
+ for (i = 0; i < I40E_MAX_USER_PRIORITY; i++) {
+ /* Get the actual TC# for the UP */
+ u8 ets_tc = dcbcfg->etscfg.prioritytable[i];
+ /* Get the mapped netdev TC# for the UP */
+ netdev_tc = vsi->tc_config.tc_info[ets_tc].netdev_tc;
+ netdev_set_prio_tc_map(netdev, i, netdev_tc);
+ }
+}
+
+/**
+ * i40e_vsi_update_queue_map - Update our copy of VSi info with new queue map
+ * @vsi: the VSI being configured
+ * @ctxt: the ctxt buffer returned from AQ VSI update param command
+ **/
+static void i40e_vsi_update_queue_map(struct i40e_vsi *vsi,
+ struct i40e_vsi_context *ctxt)
+{
+ /* copy just the sections touched not the entire info
+ * since not all sections are valid as returned by
+ * update vsi params
+ */
+ vsi->info.mapping_flags = ctxt->info.mapping_flags;
+ memcpy(&vsi->info.queue_mapping,
+ &ctxt->info.queue_mapping, sizeof(vsi->info.queue_mapping));
+ memcpy(&vsi->info.tc_mapping, ctxt->info.tc_mapping,
+ sizeof(vsi->info.tc_mapping));
+}
+
+/**
+ * i40e_vsi_config_tc - Configure VSI Tx Scheduler for given TC map
+ * @vsi: VSI to be configured
+ * @enabled_tc: TC bitmap
+ *
+ * This configures a particular VSI for TCs that are mapped to the
+ * given TC bitmap. It uses default bandwidth share for TCs across
+ * VSIs to configure TC for a particular VSI.
+ *
+ * NOTE:
+ * It is expected that the VSI queues have been quisced before calling
+ * this function.
+ **/
+static s32 i40e_vsi_config_tc(struct i40e_vsi *vsi, u8 enabled_tc)
+{
+ u8 bw_share[I40E_MAX_TRAFFIC_CLASS] = {0};
+ struct i40e_vsi_context ctxt;
+ int ret = I40E_SUCCESS;
+ int i;
+
+ /* Check if enabled_tc is same as existing or new TCs */
+ if (vsi->tc_config.enabled_tc == enabled_tc)
+ return ret;
+
+ /* Enable ETS TCs with equal BW Share for now across all VSIs */
+ for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++) {
+ if (enabled_tc & (1 << i))
+ bw_share[i] = 1;
+ }
+
+ ret = i40e_vsi_configure_bw_alloc(vsi, enabled_tc, bw_share);
+ if (ret) {
+ dev_info(&vsi->back->pdev->dev,
+ "Failed configuring TC map %d for VSI %d\n",
+ enabled_tc, vsi->seid);
+ goto out;
+ }
+
+ /* Update Queue Pairs Mapping for currently enabled UPs */
+ ctxt.seid = vsi->seid;
+ ctxt.pf_num = vsi->back->hw.pf_id;
+ ctxt.vf_num = 0;
+ ctxt.uplink_seid = vsi->uplink_seid;
+ memcpy(&ctxt.info, &vsi->info, sizeof(vsi->info));
+ i40e_vsi_setup_queue_map(vsi, &ctxt, enabled_tc, false);
+
+ /* Update the VSI after updating the VSI queue-mapping information */
+ ret = i40e_aq_update_vsi_params(&vsi->back->hw, &ctxt, NULL);
+ if (ret != I40E_SUCCESS) {
+ dev_info(&vsi->back->pdev->dev,
+ "update vsi failed, aq_err=%d\n",
+ vsi->back->hw.aq.asq_last_status);
+ goto out;
+ }
+ /* update the local VSI info with updated queue map */
+ i40e_vsi_update_queue_map(vsi, &ctxt);
+ vsi->info.valid_sections = 0;
+
+ /* Update current VSI BW information */
+ ret = i40e_vsi_get_bw_info(vsi);
+ if (ret != I40E_SUCCESS) {
+ dev_info(&vsi->back->pdev->dev,
+ "Failed updating vsi bw info, aq_err=%d\n",
+ vsi->back->hw.aq.asq_last_status);
+ goto out;
+ }
+
+ /* Update the netdev TC setup */
+ i40e_vsi_config_netdev_tc(vsi, enabled_tc);
+out:
+ return ret;
+}
+
+/**
+ * i40e_up_complete - Finish the last steps of bringing up a connection
+ * @vsi: the VSI being configured
+ **/
+static int i40e_up_complete(struct i40e_vsi *vsi)
+{
+ struct i40e_pf *pf = vsi->back;
+ int err;
+
+ if (pf->flags & I40E_FLAG_MSIX_ENABLED)
+ i40e_vsi_configure_msix(vsi);
+ else
+ i40e_configure_msi_and_legacy(vsi);
+
+ /* start rings */
+ err = i40e_vsi_control_rings(vsi, true);
+ if (err)
+ return err;
+
+ clear_bit(__I40E_DOWN, &vsi->state);
+ i40e_napi_enable_all(vsi);
+ i40e_vsi_enable_irq(vsi);
+
+ if ((pf->hw.phy.link_info.link_info & I40E_AQ_LINK_UP) &&
+ (vsi->netdev)) {
+ netif_tx_start_all_queues(vsi->netdev);
+ netif_carrier_on(vsi->netdev);
+ }
+ i40e_service_event_schedule(pf);
+
+ return 0;
+}
+
+/**
+ * i40e_vsi_reinit_locked - Reset the VSI
+ * @vsi: the VSI being configured
+ *
+ * Rebuild the ring structs after some configuration
+ * has changed, e.g. MTU size.
+ **/
+static void i40e_vsi_reinit_locked(struct i40e_vsi *vsi)
+{
+ struct i40e_pf *pf = vsi->back;
+
+ WARN_ON(in_interrupt());
+ while (test_and_set_bit(__I40E_CONFIG_BUSY, &pf->state))
+ usleep_range(1000, 2000);
+ i40e_down(vsi);
+
+ /* Give a VF some time to respond to the reset. The
+ * two second wait is based upon the watchdog cycle in
+ * the VF driver.
+ */
+ if (vsi->type == I40E_VSI_SRIOV)
+ msleep(2000);
+ i40e_up(vsi);
+ clear_bit(__I40E_CONFIG_BUSY, &pf->state);
+}
+
+/**
+ * i40e_up - Bring the connection back up after being down
+ * @vsi: the VSI being configured
+ **/
+int i40e_up(struct i40e_vsi *vsi)
+{
+ int err;
+
+ err = i40e_vsi_configure(vsi);
+ if (err == I40E_SUCCESS)
+ err = i40e_up_complete(vsi);
+
+ return err;
+}
+
+/**
+ * i40e_down - Shutdown the connection processing
+ * @vsi: the VSI being stopped
+ **/
+void i40e_down(struct i40e_vsi *vsi)
+{
+ int i;
+
+ /* It is assumed that the caller of this function
+ * sets the vsi->state __I40E_DOWN bit.
+ */
+ if (vsi->netdev) {
+ netif_carrier_off(vsi->netdev);
+ netif_tx_disable(vsi->netdev);
+ }
+ i40e_vsi_disable_irq(vsi);
+ i40e_vsi_control_rings(vsi, false);
+ i40e_napi_disable_all(vsi);
+
+ for (i = 0; i < vsi->num_queue_pairs; i++) {
+ i40e_clean_tx_ring(&vsi->tx_rings[i]);
+ i40e_clean_rx_ring(&vsi->rx_rings[i]);
+ }
+}
+
+/**
+ * i40e_setup_tc - configure multiple traffic classes
+ * @netdev: net device to configure
+ * @tc: number of traffic classes to enable
+ **/
+static int i40e_setup_tc(struct net_device *netdev, u8 tc)
+{
+ struct i40e_netdev_priv *np = netdev_priv(netdev);
+ struct i40e_vsi *vsi = np->vsi;
+ struct i40e_pf *pf = vsi->back;
+ u8 enabled_tc = 0;
+ int ret = -EINVAL;
+ int i;
+
+ /* Check if DCB enabled to continue */
+ if (!(pf->flags & I40E_FLAG_DCB_ENABLED)) {
+ netdev_info(netdev, "DCB is not enabled for adapter\n");
+ goto exit;
+ }
+
+ /* Check if MFP enabled */
+ if (pf->flags & I40E_FLAG_MFP_ENABLED) {
+ netdev_info(netdev, "Configuring TC not supported in MFP mode\n");
+ goto exit;
+ }
+
+ /* Check whether tc count is within enabled limit */
+ if (tc > i40e_pf_get_num_tc(pf)) {
+ netdev_info(netdev, "TC count greater than enabled on link for adapter\n");
+ goto exit;
+ }
+
+ /* Generate TC map for number of tc requested */
+ for (i = 0; i < tc; i++)
+ enabled_tc |= (1 << i);
+
+ /* Requesting same TC configuration as already enabled */
+ if (enabled_tc == vsi->tc_config.enabled_tc)
+ return 0;
+
+ /* Quiesce VSI queues */
+ i40e_quiesce_vsi(vsi);
+
+ /* Configure VSI for enabled TCs */
+ ret = i40e_vsi_config_tc(vsi, enabled_tc);
+ if (ret != I40E_SUCCESS) {
+ netdev_info(netdev, "Failed configuring TC for VSI seid=%d\n",
+ vsi->seid);
+ goto exit;
+ }
+
+ /* Unquiesce VSI */
+ i40e_unquiesce_vsi(vsi);
+
+exit:
+ return ret;
+}
+
+/**
+ * i40e_open - Called when a network interface is made active
+ * @netdev: network interface device structure
+ *
+ * The open entry point is called when a network interface is made
+ * active by the system (IFF_UP). At this point all resources needed
+ * for transmit and receive operations are allocated, the interrupt
+ * handler is registered with the OS, the netdev watchdog subtask is
+ * enabled, and the stack is notified that the interface is ready.
+ *
+ * Returns 0 on success, negative value on failure
+ **/
+static int i40e_open(struct net_device *netdev)
+{
+ struct i40e_netdev_priv *np = netdev_priv(netdev);
+ struct i40e_vsi *vsi = np->vsi;
+ struct i40e_pf *pf = vsi->back;
+ char int_name[IFNAMSIZ];
+ int err;
+
+ /* disallow open during test */
+ if (test_bit(__I40E_TESTING, &pf->state))
+ return -EBUSY;
+
+ netif_carrier_off(netdev);
+
+ /* allocate descriptors */
+ err = i40e_vsi_setup_tx_resources(vsi);
+ if (err)
+ goto err_setup_tx;
+ err = i40e_vsi_setup_rx_resources(vsi);
+ if (err)
+ goto err_setup_rx;
+
+ err = i40e_vsi_configure(vsi);
+ if (err)
+ goto err_setup_rx;
+
+ snprintf(int_name, sizeof(int_name) - 1, "%s-%s",
+ dev_driver_string(&pf->pdev->dev), netdev->name);
+ err = i40e_vsi_request_irq(vsi, int_name);
+ if (err)
+ goto err_setup_rx;
+
+ err = i40e_up_complete(vsi);
+ if (err)
+ goto err_up_complete;
+
+ if ((vsi->type == I40E_VSI_MAIN) || (vsi->type == I40E_VSI_VMDQ2)) {
+ err = i40e_aq_set_vsi_broadcast(&pf->hw, vsi->seid, true, NULL);
+ if (err)
+ netdev_info(netdev,
+ "couldn't set broadcast err %d aq_err %d\n",
+ err, pf->hw.aq.asq_last_status);
+ }
+
+ return I40E_SUCCESS;
+
+err_up_complete:
+ i40e_down(vsi);
+ i40e_vsi_free_irq(vsi);
+err_setup_rx:
+ i40e_vsi_free_rx_resources(vsi);
+err_setup_tx:
+ i40e_vsi_free_tx_resources(vsi);
+ if (vsi == pf->vsi[pf->lan_vsi])
+ i40e_do_reset(pf, (1 << __I40E_PF_RESET_REQUESTED));
+
+ return err;
+}
+
+/**
+ * i40e_close - Disables a network interface
+ * @netdev: network interface device structure
+ *
+ * The close entry point is called when an interface is de-activated
+ * by the OS. The hardware is still under the driver's control, but
+ * this netdev interface is disabled.
+ *
+ * Returns 0, this is not allowed to fail
+ **/
+static int i40e_close(struct net_device *netdev)
+{
+ struct i40e_netdev_priv *np = netdev_priv(netdev);
+ struct i40e_vsi *vsi = np->vsi;
+
+ if (test_and_set_bit(__I40E_DOWN, &vsi->state))
+ return I40E_SUCCESS;
+
+ i40e_down(vsi);
+ i40e_vsi_free_irq(vsi);
+
+ i40e_vsi_free_tx_resources(vsi);
+ i40e_vsi_free_rx_resources(vsi);
+
+ return I40E_SUCCESS;
+}
+
+/**
+ * i40e_do_reset - Start a PF or Core Reset sequence
+ * @pf: board private structure
+ * @reset_flags: which reset is requested
+ *
+ * The essential difference in resets is that the PF Reset
+ * doesn't clear the packet buffers, doesn't reset the PE
+ * firmware, and doesn't bother the other PFs on the chip.
+ **/
+void i40e_do_reset(struct i40e_pf *pf, u32 reset_flags)
+{
+ u32 val;
+
+ WARN_ON(in_interrupt());
+
+ /* do the biggest reset indicated */
+ if (reset_flags & (1 << __I40E_GLOBAL_RESET_REQUESTED)) {
+
+ /* Request a Global Reset
+ *
+ * This will start the chip's countdown to the actual full
+ * chip reset event, and a warning interrupt to be sent
+ * to all PFs, including the requestor. Our handler
+ * for the warning interrupt will deal with the shutdown
+ * and recovery of the switch setup.
+ */
+ dev_info(&pf->pdev->dev, "GlobalR requested\n");
+ val = rd32(&pf->hw, I40E_GLGEN_RTRIG);
+ val |= I40E_GLGEN_RTRIG_GLOBR_MASK;
+ wr32(&pf->hw, I40E_GLGEN_RTRIG, val);
+
+ } else if (reset_flags & (1 << __I40E_CORE_RESET_REQUESTED)) {
+
+ /* Request a Core Reset
+ *
+ * Same as Global Reset, except does *not* include the MAC/PHY
+ */
+ dev_info(&pf->pdev->dev, "CoreR requested\n");
+ val = rd32(&pf->hw, I40E_GLGEN_RTRIG);
+ val |= I40E_GLGEN_RTRIG_CORER_MASK;
+ wr32(&pf->hw, I40E_GLGEN_RTRIG, val);
+ i40e_flush(&pf->hw);
+
+ } else if (reset_flags & (1 << __I40E_PF_RESET_REQUESTED)) {
+
+ /* Request a PF Reset
+ *
+ * Resets only the PF-specific registers
+ *
+ * This goes directly to the tear-down and rebuild of
+ * the switch, since we need to do all the recovery as
+ * for the Core Reset.
+ */
+ dev_info(&pf->pdev->dev, "PFR requested\n");
+ i40e_handle_reset_warning(pf);
+
+ } else if (reset_flags & (1 << __I40E_REINIT_REQUESTED)) {
+ int v;
+
+ /* Find the VSI(s) that requested a re-init */
+ dev_info(&pf->pdev->dev,
+ "VSI reinit requested\n");
+ for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+ struct i40e_vsi *vsi = pf->vsi[v];
+ if (vsi != NULL &&
+ test_bit(__I40E_REINIT_REQUESTED, &vsi->state)) {
+ i40e_vsi_reinit_locked(pf->vsi[v]);
+ clear_bit(__I40E_REINIT_REQUESTED, &vsi->state);
+ }
+ }
+
+ /* no further action needed, so return now */
+ return;
+ } else {
+ dev_info(&pf->pdev->dev,
+ "bad reset request 0x%08x\n", reset_flags);
+ return;
+ }
+}
+
+/**
+ * i40e_handle_lan_overflow_event - Handler for LAN queue overflow event
+ * @pf: board private structure
+ * @e: event info posted on ARQ
+ *
+ * Handler for LAN Queue Overflow Event generated by the firmware for PF
+ * and VF queues
+ **/
+static void i40e_handle_lan_overflow_event(struct i40e_pf *pf,
+ struct i40e_arq_event_info *e)
+{
+ struct i40e_aqc_lan_overflow *data =
+ (struct i40e_aqc_lan_overflow *)&e->desc.params.raw;
+ u32 queue = le32_to_cpu(data->prtdcb_rupto);
+ u32 qtx_ctl = le32_to_cpu(data->otx_ctl);
+ struct i40e_hw *hw = &pf->hw;
+ struct i40e_vf *vf;
+ u16 vf_id;
+
+ dev_info(&pf->pdev->dev, "%s: Rx Queue Number = %d QTX_CTL=0x%08x\n",
+ __func__, queue, qtx_ctl);
+
+ /* Queue belongs to VF, find the VF and issue VF reset */
+ if (((qtx_ctl & I40E_QTX_CTL_PFVF_Q_MASK)
+ >> I40E_QTX_CTL_PFVF_Q_SHIFT) == I40E_QTX_CTL_VF_QUEUE) {
+ vf_id = (u16)((qtx_ctl & I40E_QTX_CTL_VFVM_INDX_MASK)
+ >> I40E_QTX_CTL_VFVM_INDX_SHIFT);
+ vf_id -= hw->func_caps.vf_base_id;
+ vf = &pf->vf[vf_id];
+ i40e_vc_notify_vf_reset(vf);
+ /* Allow VF to process pending reset notification */
+ msleep(20);
+ i40e_reset_vf(vf, false);
+ }
+}
+
+/**
+ * i40e_service_event_complete - Finish up the service event
+ * @pf: board private structure
+ **/
+static void i40e_service_event_complete(struct i40e_pf *pf)
+{
+ BUG_ON(!test_bit(__I40E_SERVICE_SCHED, &pf->state));
+
+ /* flush memory to make sure state is correct before next watchog */
+ smp_mb__before_clear_bit();
+ clear_bit(__I40E_SERVICE_SCHED, &pf->state);
+}
+
+/**
+ * i40e_fdir_reinit_subtask - Worker thread to reinit FDIR filter table
+ * @pf: board private structure
+ **/
+static void i40e_fdir_reinit_subtask(struct i40e_pf *pf)
+{
+ if (!(pf->flags & I40E_FLAG_FDIR_REQUIRES_REINIT))
+ return;
+
+ pf->flags &= ~I40E_FLAG_FDIR_REQUIRES_REINIT;
+
+ /* if interface is down do nothing */
+ if (test_bit(__I40E_DOWN, &pf->state))
+ return;
+}
+
+/**
+ * i40e_vsi_link_event - notify VSI of a link event
+ * @vsi: vsi to be notified
+ * @link_up: link up or down
+ **/
+static void i40e_vsi_link_event(struct i40e_vsi *vsi, bool link_up)
+{
+ if (!vsi)
+ return;
+
+ switch (vsi->type) {
+ case I40E_VSI_MAIN:
+ if (!vsi->netdev || !vsi->netdev_registered)
+ break;
+
+ if (link_up) {
+ netif_carrier_on(vsi->netdev);
+ netif_tx_wake_all_queues(vsi->netdev);
+ } else {
+ netif_carrier_off(vsi->netdev);
+ netif_tx_stop_all_queues(vsi->netdev);
+ }
+ break;
+
+ case I40E_VSI_SRIOV:
+ break;
+
+ case I40E_VSI_VMDQ2:
+ case I40E_VSI_CTRL:
+ case I40E_VSI_MIRROR:
+ default:
+ /* there is no notification for other VSIs */
+ break;
+ }
+}
+
+/**
+ * i40e_veb_link_event - notify elements on the veb of a link event
+ * @veb: veb to be notified
+ * @link_up: link up or down
+ **/
+static void i40e_veb_link_event(struct i40e_veb *veb, bool link_up)
+{
+ struct i40e_pf *pf;
+ int i;
+
+ if (!veb || !veb->pf)
+ return;
+ pf = veb->pf;
+
+ /* depth first... */
+ for (i = 0; i < I40E_MAX_VEB; i++)
+ if (pf->veb[i] && (pf->veb[i]->uplink_seid == veb->seid))
+ i40e_veb_link_event(pf->veb[i], link_up);
+
+ /* ... now the local VSIs */
+ for (i = 0; i < pf->hw.func_caps.num_vsis; i++)
+ if (pf->vsi[i] && (pf->vsi[i]->uplink_seid == veb->seid))
+ i40e_vsi_link_event(pf->vsi[i], link_up);
+}
+
+/**
+ * i40e_link_event - Update netif_carrier status
+ * @pf: board private structure
+ **/
+static void i40e_link_event(struct i40e_pf *pf)
+{
+ bool new_link, old_link;
+
+ new_link = (pf->hw.phy.link_info.link_info & I40E_AQ_LINK_UP);
+ old_link = (pf->hw.phy.link_info_old.link_info & I40E_AQ_LINK_UP);
+
+ if (new_link == old_link)
+ return;
+
+ netdev_info(pf->vsi[pf->lan_vsi]->netdev,
+ "NIC Link is %s\n", (new_link ? "Up" : "Down"));
+
+ /* Notify the base of the switch tree connected to
+ * the link. Floating VEBs are not notified.
+ */
+ if (pf->lan_veb != I40E_NO_VEB && pf->veb[pf->lan_veb])
+ i40e_veb_link_event(pf->veb[pf->lan_veb], new_link);
+ else
+ i40e_vsi_link_event(pf->vsi[pf->lan_vsi], new_link);
+
+ if (pf->vf)
+ i40e_vc_notify_link_state(pf);
+}
+
+/**
+ * i40e_check_hang_subtask - Check for hung queues and dropped interrupts
+ * @pf: board private structure
+ *
+ * Set the per-queue flags to request a check for stuck queues in the irq
+ * clean functions, then force interrupts to be sure the irq clean is called.
+ **/
+static void i40e_check_hang_subtask(struct i40e_pf *pf)
+{
+ int i, v;
+
+ /* If we're down or resetting, just bail */
+ if (test_bit(__I40E_CONFIG_BUSY, &pf->state))
+ return;
+
+ /* for each VSI/netdev
+ * for each Tx queue
+ * set the check flag
+ * for each q_vector
+ * force an interrupt
+ */
+ for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+ struct i40e_vsi *vsi = pf->vsi[v];
+ int armed = 0;
+
+ if (!pf->vsi[v] ||
+ test_bit(__I40E_DOWN, &vsi->state) ||
+ (vsi->netdev && !netif_carrier_ok(vsi->netdev)))
+ continue;
+
+ for (i = 0; i < vsi->num_queue_pairs; i++) {
+ set_check_for_tx_hang(&vsi->tx_rings[i]);
+ if (test_bit(__I40E_HANG_CHECK_ARMED,
+ &vsi->tx_rings[i].state))
+ armed++;
+ }
+
+ if (armed) {
+ if (!(pf->flags & I40E_FLAG_MSIX_ENABLED)) {
+ wr32(&vsi->back->hw, I40E_PFINT_DYN_CTL0,
+ (I40E_PFINT_DYN_CTL0_INTENA_MASK |
+ I40E_PFINT_DYN_CTL0_SWINT_TRIG_MASK));
+ } else {
+ u16 vec = vsi->base_vector - 1;
+ u32 val = (I40E_PFINT_DYN_CTLN_INTENA_MASK |
+ I40E_PFINT_DYN_CTLN_SWINT_TRIG_MASK);
+ for (i = 0; i < vsi->num_q_vectors; i++, vec++)
+ wr32(&vsi->back->hw,
+ I40E_PFINT_DYN_CTLN(vec), val);
+ }
+ i40e_flush(&vsi->back->hw);
+ }
+ }
+}
+
+/**
+ * i40e_watchdog_subtask - Check and bring link up
+ * @pf: board private structure
+ **/
+static void i40e_watchdog_subtask(struct i40e_pf *pf)
+{
+ int i;
+
+ /* if interface is down do nothing */
+ if (test_bit(__I40E_DOWN, &pf->state) ||
+ test_bit(__I40E_CONFIG_BUSY, &pf->state))
+ return;
+
+ /* Update the stats for active netdevs so the network stack
+ * can look at updated numbers whenever it cares to
+ */
+ for (i = 0; i < pf->hw.func_caps.num_vsis; i++)
+ if (pf->vsi[i] && pf->vsi[i]->netdev)
+ i40e_update_stats(pf->vsi[i]);
+
+ /* Update the stats for the active switching components */
+ for (i = 0; i < I40E_MAX_VEB; i++)
+ if (pf->veb[i])
+ i40e_update_veb_stats(pf->veb[i]);
+}
+
+/**
+ * i40e_reset_subtask - Set up for resetting the device and driver
+ * @pf: board private structure
+ **/
+static void i40e_reset_subtask(struct i40e_pf *pf)
+{
+ u32 reset_flags = 0;
+
+ if (test_bit(__I40E_REINIT_REQUESTED, &pf->state)) {
+ reset_flags |= (1 << __I40E_REINIT_REQUESTED);
+ clear_bit(__I40E_REINIT_REQUESTED, &pf->state);
+ }
+ if (test_bit(__I40E_PF_RESET_REQUESTED, &pf->state)) {
+ reset_flags |= (1 << __I40E_PF_RESET_REQUESTED);
+ clear_bit(__I40E_PF_RESET_REQUESTED, &pf->state);
+ }
+ if (test_bit(__I40E_CORE_RESET_REQUESTED, &pf->state)) {
+ reset_flags |= (1 << __I40E_CORE_RESET_REQUESTED);
+ clear_bit(__I40E_CORE_RESET_REQUESTED, &pf->state);
+ }
+ if (test_bit(__I40E_GLOBAL_RESET_REQUESTED, &pf->state)) {
+ reset_flags |= (1 << __I40E_GLOBAL_RESET_REQUESTED);
+ clear_bit(__I40E_GLOBAL_RESET_REQUESTED, &pf->state);
+ }
+
+ /* If there's a recovery already waiting, it takes
+ * precedence before starting a new reset sequence.
+ */
+ if (test_bit(__I40E_RESET_INTR_RECEIVED, &pf->state)) {
+ i40e_handle_reset_warning(pf);
+ return;
+ }
+
+ /* If we're already down or resetting, just bail */
+ if (reset_flags &&
+ !test_bit(__I40E_DOWN, &pf->state) &&
+ !test_bit(__I40E_CONFIG_BUSY, &pf->state))
+ i40e_do_reset(pf, reset_flags);
+}
+
+/**
+ * i40e_handle_link_event - Handle link event
+ * @pf: board private structure
+ * @e: event info posted on ARQ
+ **/
+static void i40e_handle_link_event(struct i40e_pf *pf,
+ struct i40e_arq_event_info *e)
+{
+ struct i40e_hw *hw = &pf->hw;
+ struct i40e_aqc_get_link_status *status =
+ (struct i40e_aqc_get_link_status *)&e->desc.params.raw;
+ struct i40e_link_status *hw_link_info = &hw->phy.link_info;
+
+ /* save off old link status information */
+ memcpy(&pf->hw.phy.link_info_old, hw_link_info,
+ sizeof(pf->hw.phy.link_info_old));
+
+ /* update link status */
+ hw_link_info->phy_type = (enum i40e_aq_phy_type)status->phy_type;
+ hw_link_info->link_speed = (enum i40e_aq_link_speed)status->link_speed;
+ hw_link_info->link_info = status->link_info;
+ hw_link_info->an_info = status->an_info;
+ hw_link_info->ext_info = status->ext_info;
+ hw_link_info->lse_enable =
+ le16_to_cpu(status->command_flags) &
+ I40E_AQ_LSE_ENABLE;
+
+ /* process the event */
+ i40e_link_event(pf);
+
+ /* Do a new status request to re-enable LSE reporting
+ * and load new status information into the hw struct,
+ * then see if the status changed while processing the
+ * initial event.
+ */
+ i40e_aq_get_link_info(&pf->hw, true, NULL, NULL);
+ i40e_link_event(pf);
+}
+
+/**
+ * i40e_clean_adminq_subtask - Clean the AdminQ rings
+ * @pf: board private structure
+ **/
+static void i40e_clean_adminq_subtask(struct i40e_pf *pf)
+{
+ struct i40e_arq_event_info event;
+ struct i40e_hw *hw = &pf->hw;
+ u16 pending, i = 0;
+ i40e_status ret;
+ u16 opcode;
+ u32 val;
+
+ if (!test_bit(__I40E_ADMINQ_EVENT_PENDING, &pf->state))
+ return;
+
+ event.msg_size = I40E_MAX_AQ_BUF_SIZE;
+ event.msg_buf = kzalloc(event.msg_size, GFP_KERNEL);
+ if (!event.msg_buf)
+ return;
+
+ do {
+ ret = i40e_clean_arq_element(hw, &event, &pending);
+ if (ret == I40E_ERR_ADMIN_QUEUE_NO_WORK) {
+ dev_info(&pf->pdev->dev, "No ARQ event found\n");
+ break;
+ } else if (ret) {
+ dev_info(&pf->pdev->dev, "ARQ event error %d\n", ret);
+ break;
+ }
+
+ opcode = le16_to_cpu(event.desc.opcode);
+ switch (opcode) {
+
+ case i40e_aqc_opc_get_link_status:
+ i40e_handle_link_event(pf, &event);
+ break;
+ case i40e_aqc_opc_send_msg_to_pf:
+ ret = i40e_vc_process_vf_msg(pf,
+ le16_to_cpu(event.desc.retval),
+ le32_to_cpu(event.desc.cookie_high),
+ le32_to_cpu(event.desc.cookie_low),
+ event.msg_buf,
+ event.msg_size);
+ break;
+ case i40e_aqc_opc_lldp_update_mib:
+ dev_info(&pf->pdev->dev, "ARQ: Update LLDP MIB event received\n");
+ break;
+ case i40e_aqc_opc_event_lan_overflow:
+ dev_info(&pf->pdev->dev, "ARQ LAN queue overflow event received\n");
+ i40e_handle_lan_overflow_event(pf, &event);
+ break;
+ default:
+ dev_info(&pf->pdev->dev,
+ "ARQ Error: Unknown event %d received\n",
+ event.desc.opcode);
+ break;
+ }
+ if (pending != 0)
+ dev_info(&pf->pdev->dev, "ARQ: Pending events %d\n",
+ pending);
+ } while (pending && (i++ < pf->adminq_work_limit));
+
+ clear_bit(__I40E_ADMINQ_EVENT_PENDING, &pf->state);
+ /* re-enable Admin queue interrupt cause */
+ val = rd32(hw, I40E_PFINT_ICR0_ENA);
+ val |= I40E_PFINT_ICR0_ENA_ADMINQ_MASK;
+ wr32(hw, I40E_PFINT_ICR0_ENA, val);
+ i40e_flush(hw);
+
+ kfree(event.msg_buf);
+}
+
+/**
+ * i40e_reconstitute_veb - rebuild the VEB and anything connected to it
+ * @veb: pointer to the VEB instance
+ *
+ * This is a recursive function that first builds the attached VSIs then
+ * recurses in to build the next layer of VEB. We track the connections
+ * through our own index numbers because the seid's from the HW could
+ * change across the reset.
+ **/
+static s32 i40e_reconstitute_veb(struct i40e_veb *veb)
+{
+ struct i40e_vsi *ctl_vsi = NULL;
+ struct i40e_pf *pf = veb->pf;
+ int v, veb_idx;
+ int ret;
+
+ /* build VSI that owns this VEB, temporarily attached to base VEB */
+ for (v = 0; v < pf->hw.func_caps.num_vsis && !ctl_vsi; v++) {
+ if (pf->vsi[v] &&
+ pf->vsi[v]->veb_idx == veb->idx &&
+ pf->vsi[v]->flags & I40E_VSI_FLAG_VEB_OWNER) {
+ ctl_vsi = pf->vsi[v];
+ break;
+ }
+ }
+ if (!ctl_vsi) {
+ dev_info(&pf->pdev->dev,
+ "missing owner VSI for veb_idx %d\n", veb->idx);
+ ret = I40E_ERR_NO_AVAILABLE_VSI;
+ goto end_reconstitute;
+ }
+ if (ctl_vsi != pf->vsi[pf->lan_vsi])
+ ctl_vsi->uplink_seid = pf->vsi[pf->lan_vsi]->uplink_seid;
+ ret = i40e_add_vsi(ctl_vsi);
+ if (ret != I40E_SUCCESS) {
+ dev_info(&pf->pdev->dev,
+ "rebuild of owner VSI failed: %d\n", ret);
+ goto end_reconstitute;
+ }
+ i40e_sys_add_vsi(ctl_vsi);
+ i40e_vsi_reset_stats(ctl_vsi);
+
+ /* create the VEB in the switch and move the VSI onto the VEB */
+ ret = i40e_add_veb(veb, ctl_vsi);
+ if (ret != I40E_SUCCESS)
+ goto end_reconstitute;
+
+ /* create the remaining VSIs attached to this VEB */
+ for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+ if (!pf->vsi[v] || pf->vsi[v] == ctl_vsi)
+ continue;
+
+ if (pf->vsi[v]->veb_idx == veb->idx) {
+ struct i40e_vsi *vsi = pf->vsi[v];
+ vsi->uplink_seid = veb->seid;
+ ret = i40e_add_vsi(vsi);
+ if (ret != I40E_SUCCESS) {
+ dev_info(&pf->pdev->dev,
+ "rebuild of vsi_idx %d failed: %d\n",
+ v, ret);
+ goto end_reconstitute;
+ }
+ i40e_sys_add_vsi(vsi);
+ i40e_vsi_reset_stats(vsi);
+ }
+ }
+
+ /* create any VEBs attached to this VEB - RECURSION */
+ for (veb_idx = 0; veb_idx < I40E_MAX_VEB; veb_idx++) {
+ if (pf->veb[veb_idx] && pf->veb[veb_idx]->veb_idx == veb->idx) {
+ pf->veb[veb_idx]->uplink_seid = veb->seid;
+ ret = i40e_reconstitute_veb(pf->veb[veb_idx]);
+ if (ret != I40E_SUCCESS)
+ break;
+ }
+ }
+
+end_reconstitute:
+ return ret;
+}
+
+/**
+ * i40e_get_capabilities - get info about the HW
+ * @pf: the PF struct
+ **/
+static i40e_status i40e_get_capabilities(struct i40e_pf *pf)
+{
+ struct i40e_aqc_list_capabilities_element_resp *cap_buf;
+ u16 data_size;
+ int buf_len;
+ int err;
+
+ buf_len = 40 * sizeof(struct i40e_aqc_list_capabilities_element_resp);
+ do {
+ cap_buf = kzalloc(buf_len, GFP_KERNEL);
+ if (!cap_buf)
+ return -ENOMEM;
+
+ /* this loads the data into the hw struct for us */
+ err = i40e_aq_discover_capabilities(&pf->hw, cap_buf, buf_len,
+ &data_size,
+ i40e_aqc_opc_list_func_capabilities,
+ NULL);
+ /* data loaded, buffer no longer needed */
+ kfree(cap_buf);
+
+ if (pf->hw.aq.asq_last_status == I40E_AQ_RC_ENOMEM) {
+ /* retry with a larger buffer */
+ buf_len = data_size;
+ } else if (pf->hw.aq.asq_last_status != I40E_AQ_RC_OK) {
+ dev_info(&pf->pdev->dev,
+ "capability discovery failed: aq=%d\n",
+ pf->hw.aq.asq_last_status);
+ return err;
+ }
+ } while (err != I40E_SUCCESS);
+
+ if (pf->hw.debug_mask & I40E_DEBUG_USER)
+ dev_info(&pf->pdev->dev,
+ "pf=%d, num_vfs=%d, msix_pf=%d, msix_vf=%d, fd_g=%d, fd_b=%d, pf_max_q=%d num_vsi=%d\n",
+ pf->hw.pf_id, pf->hw.func_caps.num_vfs,
+ pf->hw.func_caps.num_msix_vectors,
+ pf->hw.func_caps.num_msix_vectors_vf,
+ pf->hw.func_caps.fd_filters_guaranteed,
+ pf->hw.func_caps.fd_filters_best_effort,
+ pf->hw.func_caps.num_tx_qp,
+ pf->hw.func_caps.num_vsis);
+
+ return I40E_SUCCESS;
+}
+
+/**
+ * i40e_fdir_setup - initialize the Flow Director resources
+ * @pf: board private structure
+ **/
+static void i40e_fdir_setup(struct i40e_pf *pf)
+{
+ struct i40e_vsi *vsi;
+ bool new_vsi = false;
+ int err, i;
+
+ if (!(pf->flags & (I40E_FLAG_FDIR_ENABLED|I40E_FLAG_FDIR_ATR_ENABLED)))
+ return;
+
+ pf->atr_sample_rate = I40E_DEFAULT_ATR_SAMPLE_RATE;
+
+ /* find existing or make new FDIR VSI */
+ vsi = NULL;
+ for (i = 0; i < pf->hw.func_caps.num_vsis; i++)
+ if (pf->vsi[i] && pf->vsi[i]->type == I40E_VSI_FDIR)
+ vsi = pf->vsi[i];
+ if (!vsi) {
+ vsi = i40e_vsi_setup(pf, I40E_VSI_FDIR, pf->mac_seid, 0);
+ if (!vsi) {
+ dev_info(&pf->pdev->dev, "Couldn't create FDir VSI\n");
+ pf->flags &= ~I40E_FLAG_FDIR_ENABLED;
+ return;
+ }
+ new_vsi = true;
+ }
+ WARN_ON(vsi->base_queue != I40E_FDIR_RING);
+ i40e_vsi_setup_irqhandler(vsi, i40e_fdir_clean_rings);
+
+ err = i40e_vsi_setup_tx_resources(vsi);
+ if (!err)
+ err = i40e_vsi_setup_rx_resources(vsi);
+ if (!err)
+ err = i40e_vsi_configure(vsi);
+ if (!err && new_vsi) {
+ char int_name[IFNAMSIZ + 9];
+ snprintf(int_name, sizeof(int_name) - 1, "%s-fdir",
+ dev_driver_string(&pf->pdev->dev));
+ err = i40e_vsi_request_irq(vsi, int_name);
+ }
+ if (!err)
+ err = i40e_up_complete(vsi);
+
+ clear_bit(__I40E_NEEDS_RESTART, &vsi->state);
+}
+
+/**
+ * i40e_fdir_teardown - release the Flow Director resources
+ * @pf: board private structure
+ **/
+static void i40e_fdir_teardown(struct i40e_pf *pf)
+{
+ int i;
+
+ for (i = 0; i < pf->hw.func_caps.num_vsis; i++) {
+ if (pf->vsi[i] && pf->vsi[i]->type == I40E_VSI_FDIR) {
+ i40e_vsi_release(pf->vsi[i]);
+ break;
+ }
+ }
+}
+
+/**
+ * i40e_handle_reset_warning - prep for the core to reset
+ * @pf: board private structure
+ *
+ * Close up the VFs and other things in prep for a Core Reset,
+ * then get ready to rebuild the world.
+ **/
+static void i40e_handle_reset_warning(struct i40e_pf *pf)
+{
+ struct i40e_driver_version dv;
+ struct i40e_hw *hw = &pf->hw;
+ i40e_status ret;
+ u32 v;
+
+ clear_bit(__I40E_RESET_INTR_RECEIVED, &pf->state);
+ if (test_and_set_bit(__I40E_RESET_RECOVERY_PENDING, &pf->state))
+ return;
+
+ dev_info(&pf->pdev->dev, "Tearing down internal switch for reset\n");
+
+ i40e_vc_notify_reset(pf);
+
+ /* quiesce the VSIs and their queues that are not already DOWN */
+ i40e_pf_quiesce_all_vsi(pf);
+
+ /* remove everything from sysfs */
+ for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+ i40e_sys_del_vsi(pf->vsi[v]);
+ if (pf->vsi[v])
+ pf->vsi[v]->seid = 0;
+ }
+ for (v = 0; v < I40E_MAX_VEB; v++)
+ i40e_sys_del_veb(pf->veb[v]);
+
+ i40e_shutdown_adminq(&pf->hw);
+
+ /* Now we wait for GRST to settle out.
+ * We don't have to delete the VEBs or VSIs from the hw switch
+ * because the reset will make them disappear.
+ */
+ ret = i40e_pf_reset(hw);
+ if (ret != I40E_SUCCESS)
+ dev_info(&pf->pdev->dev, "PF reset failed, %d\n", ret);
+ pf->pfr_count++;
+
+ if (test_bit(__I40E_DOWN, &pf->state))
+ goto end_core_reset;
+ dev_info(&pf->pdev->dev, "Rebuilding internal switch\n");
+
+ /* rebuild the basics for the AdminQ, HMC, and initial HW switch */
+ ret = i40e_init_adminq(&pf->hw);
+ if (ret != I40E_SUCCESS) {
+ dev_info(&pf->pdev->dev, "Rebuild AdminQ failed, %d\n", ret);
+ goto end_core_reset;
+ }
+
+ ret = i40e_get_capabilities(pf);
+ if (ret) {
+ dev_info(&pf->pdev->dev, "i40e_get_capabilities failed, %d\n",
+ ret);
+ goto end_core_reset;
+ }
+
+ /* call shutdown HMC */
+ ret = i40e_shutdown_lan_hmc(hw);
+ if (ret) {
+ dev_info(&pf->pdev->dev, "shutdown_lan_hmc failed: %d\n", ret);
+ goto end_core_reset;
+ }
+
+ ret = i40e_init_lan_hmc(hw, hw->func_caps.num_tx_qp,
+ hw->func_caps.num_rx_qp,
+ pf->fcoe_hmc_cntx_num, pf->fcoe_hmc_filt_num);
+ if (ret) {
+ dev_info(&pf->pdev->dev, "init_lan_hmc failed: %d\n", ret);
+ goto end_core_reset;
+ }
+ ret = i40e_configure_lan_hmc(hw, I40E_HMC_MODEL_DIRECT_ONLY);
+ if (ret) {
+ dev_info(&pf->pdev->dev, "configure_lan_hmc failed: %d\n", ret);
+ goto end_core_reset;
+ }
+
+ /* do basic switch setup */
+ ret = i40e_setup_pf_switch(pf);
+ if (ret != I40E_SUCCESS)
+ goto end_core_reset;
+
+ /* Rebuild the VSIs and VEBs that existed before reset.
+ * They are still in our local switch element arrays, so only
+ * need to rebuild the switch model in the HW.
+ *
+ * If there were VEBs but the reconstitution failed, we'll try
+ * try to recover minimal use by getting the basic PF VSI working.
+ */
+ if (pf->vsi[pf->lan_vsi]->uplink_seid != pf->mac_seid) {
+ dev_info(&pf->pdev->dev, "attempting to rebuild switch\n");
+ /* find the one VEB connected to the MAC, and find orphans */
+ for (v = 0; v < I40E_MAX_VEB; v++) {
+ if (!pf->veb[v])
+ continue;
+
+ if (pf->veb[v]->uplink_seid == pf->mac_seid ||
+ pf->veb[v]->uplink_seid == 0) {
+ ret = i40e_reconstitute_veb(pf->veb[v]);
+
+ if (ret == I40E_SUCCESS)
+ continue;
+
+ /* If Main VEB failed, we're in deep doodoo,
+ * so give up rebuilding the switch and set up
+ * for minimal rebuild of PF VSI.
+ * If orphan failed, we'll report the error
+ * but try to keep going.
+ */
+ if (pf->veb[v]->uplink_seid == pf->mac_seid) {
+ dev_info(&pf->pdev->dev,
+ "rebuild of switch failed: %d, will try to set up simple PF connection\n",
+ ret);
+ pf->vsi[pf->lan_vsi]->uplink_seid
+ = pf->mac_seid;
+ break;
+ } else if (pf->veb[v]->uplink_seid == 0) {
+ dev_info(&pf->pdev->dev,
+ "rebuild of orphan VEB failed: %d\n",
+ ret);
+ }
+ }
+ }
+ }
+
+ if (pf->vsi[pf->lan_vsi]->uplink_seid == pf->mac_seid) {
+ dev_info(&pf->pdev->dev, "attempting to rebuild PF VSI\n");
+ /* no VEB, so rebuild only the Main VSI */
+ ret = i40e_add_vsi(pf->vsi[pf->lan_vsi]);
+ if (ret != I40E_SUCCESS) {
+ dev_info(&pf->pdev->dev,
+ "rebuild of Main VSI failed: %d\n", ret);
+ goto end_core_reset;
+ }
+ i40e_sys_add_vsi(pf->vsi[pf->lan_vsi]);
+ }
+
+ /* reinit the misc interrupt */
+ if (pf->flags & I40E_FLAG_MSIX_ENABLED)
+ ret = i40e_setup_misc_vector(pf);
+
+ /* restart the VSIs that were rebuilt and running before the reset */
+ i40e_pf_unquiesce_all_vsi(pf);
+
+ /* tell the firmware that we're starting */
+ dv.major_version = DRV_VERSION_MAJOR;
+ dv.minor_version = DRV_VERSION_MINOR;
+ dv.build_version = DRV_VERSION_BUILD;
+ dv.subbuild_version = 0;
+ i40e_aq_send_driver_version(&pf->hw, &dv, NULL);
+
+ dev_info(&pf->pdev->dev, "PF reset done\n");
+
+end_core_reset:
+ clear_bit(__I40E_RESET_RECOVERY_PENDING, &pf->state);
+}
+
+/**
+ * i40e_handle_mdd_event
+ * @pf: pointer to the pf structure
+ *
+ * Called from the MDD irq handler to identify possibly malicious vfs
+ **/
+static i40e_status i40e_handle_mdd_event(struct i40e_pf *pf)
+{
+ struct i40e_hw *hw = &pf->hw;
+ bool mdd_detected = false;
+ struct i40e_vf *vf;
+ u32 reg;
+ int i;
+
+ if (!test_bit(__I40E_MDD_EVENT_PENDING, &pf->state))
+ return I40E_SUCCESS;
+
+ /* find what triggered the MDD event */
+ reg = rd32(hw, I40E_GL_MDET_TX);
+ if (reg & I40E_GL_MDET_TX_VALID_MASK) {
+ u8 func = (reg & I40E_GL_MDET_TX_FUNCTION_MASK)
+ >> I40E_GL_MDET_TX_FUNCTION_SHIFT;
+ u8 event = (reg & I40E_GL_MDET_TX_EVENT_SHIFT)
+ >> I40E_GL_MDET_TX_EVENT_SHIFT;
+ u8 queue = (reg & I40E_GL_MDET_TX_QUEUE_MASK)
+ >> I40E_GL_MDET_TX_QUEUE_SHIFT;
+ dev_info(&pf->pdev->dev,
+ "Malicious Driver Detection TX event 0x%02x on q %d of function 0x%02x\n",
+ event, queue, func);
+ wr32(hw, I40E_GL_MDET_TX, 0xffffffff);
+ mdd_detected = true;
+ }
+ reg = rd32(hw, I40E_GL_MDET_RX);
+ if (reg & I40E_GL_MDET_RX_VALID_MASK) {
+ u8 func = (reg & I40E_GL_MDET_RX_FUNCTION_MASK)
+ >> I40E_GL_MDET_RX_FUNCTION_SHIFT;
+ u8 event = (reg & I40E_GL_MDET_RX_EVENT_SHIFT)
+ >> I40E_GL_MDET_RX_EVENT_SHIFT;
+ u8 queue = (reg & I40E_GL_MDET_RX_QUEUE_MASK)
+ >> I40E_GL_MDET_RX_QUEUE_SHIFT;
+ dev_info(&pf->pdev->dev,
+ "Malicious Driver Detection RX event 0x%02x on q %d of function 0x%02x\n",
+ event, queue, func);
+ wr32(hw, I40E_GL_MDET_RX, 0xffffffff);
+ mdd_detected = true;
+ }
+
+ /* see if one of the VFs needs its hand slapped */
+ for (i = 0; i < pf->num_alloc_vfs && mdd_detected; i++) {
+ vf = &(pf->vf[i]);
+ reg = rd32(hw, I40E_VP_MDET_TX(i));
+ if (reg & I40E_VP_MDET_TX_VALID_MASK) {
+ wr32(hw, I40E_VP_MDET_TX(i), 0xFFFF);
+ vf->num_mdd_events++;
+ dev_info(&pf->pdev->dev, "MDD TX event on VF %d\n", i);
+ }
+
+ reg = rd32(hw, I40E_VP_MDET_RX(i));
+ if (reg & I40E_VP_MDET_RX_VALID_MASK) {
+ wr32(hw, I40E_VP_MDET_RX(i), 0xFFFF);
+ vf->num_mdd_events++;
+ dev_info(&pf->pdev->dev, "MDD RX event on VF %d\n", i);
+ }
+
+ if (vf->num_mdd_events > I40E_DEFAULT_NUM_MDD_EVENTS_ALLOWED) {
+ dev_info(&pf->pdev->dev,
+ "Too many MDD events on VF %d, disabled\n", i);
+ dev_info(&pf->pdev->dev,
+ "Use PF Control I/F to re-enable the VF\n");
+ set_bit(I40E_VF_STAT_DISABLED, &vf->vf_states);
+ }
+ }
+
+ /* re-enable mdd interrupt cause */
+ clear_bit(__I40E_MDD_EVENT_PENDING, &pf->state);
+ reg = rd32(hw, I40E_PFINT_ICR0_ENA);
+ reg |= I40E_PFINT_ICR0_ENA_MAL_DETECT_MASK;
+ wr32(hw, I40E_PFINT_ICR0_ENA, reg);
+ i40e_flush(hw);
+
+ return I40E_SUCCESS;
+}
+
+/**
+ * i40e_service_task - Run the driver's async subtasks
+ * @work: pointer to work_struct containing our data
+ **/
+static void i40e_service_task(struct work_struct *work)
+{
+ struct i40e_pf *pf = container_of(work,
+ struct i40e_pf,
+ service_task);
+ unsigned long start_time = jiffies;
+
+ i40e_reset_subtask(pf);
+ i40e_handle_mdd_event(pf);
+ i40e_vc_process_vflr_event(pf);
+ i40e_watchdog_subtask(pf);
+ i40e_fdir_reinit_subtask(pf);
+ i40e_check_hang_subtask(pf);
+ i40e_sync_filters_subtask(pf);
+ i40e_clean_adminq_subtask(pf);
+
+ i40e_service_event_complete(pf);
+
+ /* If the tasks have taken longer than one timer cycle or there
+ * is more work to be done, reschedule the service task now
+ * rather than wait for the timer to tick again.
+ */
+ if (time_after(jiffies, (start_time + pf->service_timer_period)) ||
+ test_bit(__I40E_ADMINQ_EVENT_PENDING, &pf->state) ||
+ test_bit(__I40E_MDD_EVENT_PENDING, &pf->state) ||
+ test_bit(__I40E_VFLR_EVENT_PENDING, &pf->state))
+ i40e_service_event_schedule(pf);
+}
+
+/**
+ * i40e_service_timer - timer callback
+ * @data: pointer to PF struct
+ **/
+static void i40e_service_timer(unsigned long data)
+{
+ struct i40e_pf *pf = (struct i40e_pf *)data;
+
+ mod_timer(&pf->service_timer, jiffies + pf->service_timer_period);
+ i40e_service_event_schedule(pf);
+}
+
+/**
+ * i40e_set_num_rings_in_vsi - Determine number of rings in the VSI
+ * @vsi: the VSI being configured
+ **/
+static s32 i40e_set_num_rings_in_vsi(struct i40e_vsi *vsi)
+{
+ struct i40e_pf *pf = vsi->back;
+
+ switch (vsi->type) {
+ case I40E_VSI_MAIN:
+ vsi->alloc_queue_pairs = pf->num_lan_qps;
+ vsi->num_desc = ALIGN(I40E_DEFAULT_NUM_DESCRIPTORS,
+ I40E_REQ_DESCRIPTOR_MULTIPLE);
+ if (pf->flags & I40E_FLAG_MSIX_ENABLED)
+ vsi->num_q_vectors = pf->num_lan_msix;
+ else
+ vsi->num_q_vectors = 1;
+
+ break;
+
+ case I40E_VSI_FDIR:
+ vsi->alloc_queue_pairs = 1;
+ vsi->num_desc = ALIGN(I40E_FDIR_RING_COUNT,
+ I40E_REQ_DESCRIPTOR_MULTIPLE);
+ vsi->num_q_vectors = 1;
+ break;
+
+ case I40E_VSI_VMDQ2:
+ vsi->alloc_queue_pairs = pf->num_vmdq_qps;
+ vsi->num_desc = ALIGN(I40E_DEFAULT_NUM_DESCRIPTORS,
+ I40E_REQ_DESCRIPTOR_MULTIPLE);
+ vsi->num_q_vectors = pf->num_vmdq_msix;
+ break;
+
+ case I40E_VSI_SRIOV:
+ vsi->alloc_queue_pairs = pf->num_vf_qps;
+ vsi->num_desc = ALIGN(I40E_DEFAULT_NUM_DESCRIPTORS,
+ I40E_REQ_DESCRIPTOR_MULTIPLE);
+ break;
+
+ default:
+ return I40E_ERR_PARAM;
+ }
+
+ return I40E_SUCCESS;
+}
+
+/**
+ * i40e_vsi_mem_alloc - Allocates the next available struct vsi in the PF
+ * @pf: board private structure
+ * @type: type of VSI
+ *
+ * On error: returns error code (negative)
+ * On success: returns vsi index in PF (positive)
+ **/
+static s32 i40e_vsi_mem_alloc(struct i40e_pf *pf, enum i40e_vsi_type type)
+{
+ int ret = I40E_ERR_NO_AVAILABLE_VSI;
+ struct i40e_vsi *vsi;
+ int vsi_idx;
+ int i;
+
+ /* Need to protect the allocation of the VSIs at the PF level */
+ mutex_lock(&pf->switch_mutex);
+
+ /* VSI list may be fragmented if VSI creation/destruction has
+ * been happening. We can afford to do a quick scan to look
+ * for any free VSIs in the list.
+ *
+ * find next empty vsi slot, looping back around if necessary
+ */
+ i = pf->next_vsi;
+ while (i < pf->hw.func_caps.num_vsis && pf->vsi[i])
+ i++;
+ if (i >= pf->hw.func_caps.num_vsis) {
+ i = 0;
+ while (i < pf->next_vsi && pf->vsi[i])
+ i++;
+ }
+
+ if (i < pf->hw.func_caps.num_vsis && !pf->vsi[i]) {
+ vsi_idx = i; /* Found one! */
+ } else {
+ ret = I40E_ERR_NO_AVAILABLE_VSI;
+ goto err_alloc_vsi; /* out of VSI slots! */
+ }
+ pf->next_vsi = ++i;
+
+ vsi = kzalloc(sizeof(*vsi), GFP_KERNEL);
+ if (!vsi) {
+ ret = -ENOMEM;
+ goto err_alloc_vsi;
+ }
+ vsi->type = type;
+ vsi->back = pf;
+ set_bit(__I40E_DOWN, &vsi->state);
+ vsi->flags = 0;
+ vsi->idx = vsi_idx;
+ vsi->rx_itr_setting = pf->rx_itr_default;
+ vsi->tx_itr_setting = pf->tx_itr_default;
+ vsi->netdev_registered = false;
+ vsi->work_limit = I40E_DEFAULT_IRQ_WORK;
+ INIT_LIST_HEAD(&vsi->mac_filter_list);
+
+ i40e_set_num_rings_in_vsi(vsi);
+
+ /* Setup default MSIX irq handler for VSI */
+ i40e_vsi_setup_irqhandler(vsi, i40e_msix_clean_rings);
+
+ pf->vsi[vsi_idx] = vsi;
+ ret = vsi_idx;
+err_alloc_vsi:
+ mutex_unlock(&pf->switch_mutex);
+ return ret;
+}
+
+/**
+ * i40e_vsi_clear - Deallocate the VSI provided
+ * @vsi: the VSI being un-configured
+ **/
+static s32 i40e_vsi_clear(struct i40e_vsi *vsi)
+{
+ struct i40e_pf *pf;
+
+ if (!vsi)
+ return I40E_SUCCESS;
+
+ if (!vsi->back)
+ goto free_vsi;
+ pf = vsi->back;
+
+ mutex_lock(&pf->switch_mutex);
+ if (!pf->vsi[vsi->idx]) {
+ dev_err(&pf->pdev->dev, "pf->vsi[%d] is NULL, just free vsi[%d](%p,type %d)\n",
+ vsi->idx, vsi->idx, vsi, vsi->type);
+ goto unlock_vsi;
+ }
+
+ if (pf->vsi[vsi->idx] != vsi) {
+ dev_err(&pf->pdev->dev,
+ "pf->vsi[%d](%p, type %d) != vsi[%d](%p,type %d): no free!\n",
+ pf->vsi[vsi->idx]->idx,
+ pf->vsi[vsi->idx],
+ pf->vsi[vsi->idx]->type,
+ vsi->idx, vsi, vsi->type);
+ goto unlock_vsi;
+ }
+
+ /* updates the pf for this cleared vsi */
+ i40e_put_lump(pf->qp_pile, vsi->base_queue, vsi->idx);
+ i40e_put_lump(pf->irq_pile, vsi->base_vector, vsi->idx);
+
+ pf->vsi[vsi->idx] = NULL;
+ if (vsi->idx < pf->next_vsi)
+ pf->next_vsi = vsi->idx;
+
+unlock_vsi:
+ mutex_unlock(&pf->switch_mutex);
+free_vsi:
+ kfree(vsi);
+
+ return I40E_SUCCESS;
+}
+
+/**
+ * i40e_alloc_rings - Allocates the Rx and Tx rings for the provided VSI
+ * @vsi: the VSI being configured
+ **/
+static s32 i40e_alloc_rings(struct i40e_vsi *vsi)
+{
+ struct i40e_pf *pf = vsi->back;
+ int ret = I40E_SUCCESS;
+ int i;
+
+ vsi->rx_rings = kcalloc(vsi->alloc_queue_pairs,
+ sizeof(struct i40e_ring), GFP_KERNEL);
+ if (!vsi->rx_rings) {
+ ret = -ENOMEM;
+ goto err_alloc_rings;
+ }
+
+ vsi->tx_rings = kcalloc(vsi->alloc_queue_pairs,
+ sizeof(struct i40e_ring), GFP_KERNEL);
+ if (!vsi->tx_rings) {
+ ret = -ENOMEM;
+ kfree(vsi->rx_rings);
+ goto err_alloc_rings;
+ }
+
+ /* Set basic values in the rings to be used later during open() */
+ for (i = 0; i < vsi->alloc_queue_pairs; i++) {
+ struct i40e_ring *rx_ring = &vsi->rx_rings[i];
+ struct i40e_ring *tx_ring = &vsi->tx_rings[i];
+
+ tx_ring->queue_index = i;
+ tx_ring->reg_idx = vsi->base_queue + i;
+ tx_ring->ring_active = false;
+ tx_ring->vsi = vsi;
+ tx_ring->netdev = vsi->netdev;
+ tx_ring->dev = &pf->pdev->dev;
+ tx_ring->count = vsi->num_desc;
+ tx_ring->size = 0;
+ tx_ring->dcb_tc = 0;
+
+ rx_ring->queue_index = i;
+ rx_ring->reg_idx = vsi->base_queue + i;
+ rx_ring->ring_active = false;
+ rx_ring->vsi = vsi;
+ rx_ring->netdev = vsi->netdev;
+ rx_ring->dev = &pf->pdev->dev;
+ rx_ring->count = vsi->num_desc;
+ rx_ring->size = 0;
+ rx_ring->dcb_tc = 0;
+ if (pf->flags & I40E_FLAG_16BYTE_RX_DESC_ENABLED)
+ set_ring_16byte_desc_enabled(rx_ring);
+ else
+ clear_ring_16byte_desc_enabled(rx_ring);
+ }
+
+err_alloc_rings:
+ return ret;
+}
+
+/**
+ * i40e_vsi_clear_rings - Deallocates the Rx and Tx rings for the provided VSI
+ * @vsi: the VSI being cleaned
+ **/
+static s32 i40e_vsi_clear_rings(struct i40e_vsi *vsi)
+{
+ if (vsi) {
+ kfree(vsi->rx_rings);
+ kfree(vsi->tx_rings);
+ }
+
+ return I40E_SUCCESS;
+}
+
+/**
+ * i40e_reserve_msix_vectors - Reserve MSI-X vectors in the kernel
+ * @pf: board private structure
+ * @vectors: the number of MSI-X vectors to request
+ *
+ * Returns the number of vectors reserved, or error
+ **/
+static int i40e_reserve_msix_vectors(struct i40e_pf *pf, int vectors)
+{
+ int err = 0;
+
+ pf->num_msix_entries = 0;
+ while (vectors >= I40E_MIN_MSIX) {
+ err = pci_enable_msix(pf->pdev, pf->msix_entries, vectors);
+ if (err == 0) {
+ /* good to go */
+ pf->num_msix_entries = vectors;
+ break;
+ } else if (err < 0) {
+ /* total failure */
+ dev_info(&pf->pdev->dev,
+ "MSI-X vector reservation failed: %d\n", err);
+ vectors = 0;
+ break;
+ } else {
+ /* err > 0 is the hint for retry */
+ dev_info(&pf->pdev->dev,
+ "MSI-X vectors wanted %d, retrying with %d\n",
+ vectors, err);
+ vectors = err;
+ }
+ }
+
+ if (vectors > 0 && vectors < I40E_MIN_MSIX) {
+ dev_info(&pf->pdev->dev,
+ "Couldn't get enough vectors, only %d available\n",
+ vectors);
+ vectors = 0;
+ }
+
+ return vectors;
+}
+
+/**
+ * i40e_init_msix - Setup the MSIX capability
+ * @pf: board private structure
+ *
+ * Work with the OS to set up the MSIX vectors needed.
+ *
+ * Returns 0 on success, negative on failure
+ **/
+static i40e_status i40e_init_msix(struct i40e_pf *pf)
+{
+ i40e_status err = I40E_SUCCESS;
+ struct i40e_hw *hw = &pf->hw;
+ int v_budget, i;
+ int vec;
+
+ if (!(pf->flags & I40E_FLAG_MSIX_ENABLED))
+ return I40E_NOT_SUPPORTED;
+
+ /* The number of vectors we'll request will be comprised of:
+ * - Add 1 for "other" cause for Admin Queue events, etc.
+ * - The number of LAN queue pairs
+ * already adjusted for the NUMA node
+ * assumes symmetric Tx/Rx pairing
+ * - The number of VMDq pairs
+ * Once we count this up, try the request.
+ *
+ * If we can't get what we want, we'll simplify to nearly nothing
+ * and try again. If that still fails, we punt.
+ */
+ pf->num_lan_msix = pf->num_lan_qps;
+ pf->num_vmdq_msix = pf->num_vmdq_qps;
+ v_budget = 1 + pf->num_lan_msix;
+ v_budget += (pf->num_vmdq_vsis * pf->num_vmdq_msix);
+ if (pf->flags & I40E_FLAG_FDIR_ENABLED)
+ v_budget++;
+
+ /* Scale down if necessary, and the rings will share vectors */
+ v_budget = min_t(int, v_budget, hw->func_caps.num_msix_vectors);
+
+ pf->msix_entries = kcalloc(v_budget, sizeof(struct msix_entry),
+ GFP_KERNEL);
+ if (!pf->msix_entries)
+ return I40E_ERR_NO_MEMORY;
+
+ for (i = 0; i < v_budget; i++)
+ pf->msix_entries[i].entry = i;
+ vec = i40e_reserve_msix_vectors(pf, v_budget);
+ if (vec < I40E_MIN_MSIX) {
+ pf->flags &= ~I40E_FLAG_MSIX_ENABLED;
+ kfree(pf->msix_entries);
+ pf->msix_entries = NULL;
+ return I40E_NOT_SUPPORTED;
+
+ } else if (vec == I40E_MIN_MSIX) {
+ /* Adjust for minimal MSIX use */
+ dev_info(&pf->pdev->dev, "Features disabled, not enough MSIX vectors\n");
+ pf->flags &= ~I40E_FLAG_VMDQ_ENABLED;
+ pf->num_vmdq_vsis = 0;
+ pf->num_vmdq_qps = 0;
+ pf->num_vmdq_msix = 0;
+ pf->num_lan_qps = 1;
+ pf->num_lan_msix = 1;
+
+ } else if (vec != v_budget) {
+ /* Scale vector usage down */
+ pf->num_vmdq_msix = 1; /* force VMDqs to only one vector */
+ vec--; /* reserve the misc vector */
+
+ /* partition out the remaining vectors */
+ switch (vec) {
+ case 2:
+ pf->num_vmdq_vsis = 1;
+ pf->num_lan_msix = 1;
+ break;
+ case 3:
+ pf->num_vmdq_vsis = 1;
+ pf->num_lan_msix = 2;
+ break;
+ default:
+ pf->num_lan_msix = min_t(int, (vec / 2),
+ pf->num_lan_qps);
+ pf->num_vmdq_vsis = min_t(int, (vec - pf->num_lan_msix),
+ I40E_DEFAULT_NUM_VMDQ_VSI);
+ break;
+ }
+ }
+
+ return err;
+}
+
+/**
+ * i40e_alloc_q_vectors - Allocate memory for interrupt vectors
+ * @vsi: the VSI being configured
+ *
+ * We allocate one q_vector per queue interrupt. If allocation fails we
+ * return -ENOMEM.
+ **/
+static int i40e_alloc_q_vectors(struct i40e_vsi *vsi)
+{
+ struct i40e_pf *pf = vsi->back;
+ int v_idx, num_q_vectors;
+
+ /* if not MSIX, give the one vector only to the LAN VSI */
+ if (pf->flags & I40E_FLAG_MSIX_ENABLED)
+ num_q_vectors = vsi->num_q_vectors;
+ else if (vsi == pf->vsi[pf->lan_vsi])
+ num_q_vectors = 1;
+ else
+ return -EINVAL;
+
+ vsi->q_vectors = kcalloc(num_q_vectors,
+ sizeof(struct i40e_q_vector),
+ GFP_KERNEL);
+ if (!vsi->q_vectors)
+ return -ENOMEM;
+
+ for (v_idx = 0; v_idx < num_q_vectors; v_idx++) {
+ vsi->q_vectors[v_idx].vsi = vsi;
+ vsi->q_vectors[v_idx].v_idx = v_idx;
+ cpumask_set_cpu(v_idx, &vsi->q_vectors[v_idx].affinity_mask);
+ if (vsi->netdev)
+ netif_napi_add(vsi->netdev, &vsi->q_vectors[v_idx].napi,
+ i40e_napi_poll, vsi->work_limit);
+ }
+
+ return I40E_SUCCESS;
+}
+
+/**
+ * i40e_init_interrupt_scheme - Determine proper interrupt scheme
+ * @pf: board private structure to initialize
+ **/
+static void i40e_init_interrupt_scheme(struct i40e_pf *pf)
+{
+ int err = I40E_SUCCESS;
+
+ if (pf->flags & I40E_FLAG_MSIX_ENABLED) {
+ err = i40e_init_msix(pf);
+ if (err) {
+ pf->flags &= ~(I40E_FLAG_RSS_ENABLED |
+ I40E_FLAG_MQ_ENABLED |
+ I40E_FLAG_DCB_ENABLED |
+ I40E_FLAG_SRIOV_ENABLED |
+ I40E_FLAG_FDIR_ENABLED |
+ I40E_FLAG_FDIR_ATR_ENABLED |
+ I40E_FLAG_VMDQ_ENABLED);
+
+ /* rework the queue expectations without MSIX */
+ i40e_determine_queue_usage(pf);
+ }
+ }
+
+ if (!(pf->flags & I40E_FLAG_MSIX_ENABLED) &&
+ (pf->flags & I40E_FLAG_MSI_ENABLED)) {
+ err = pci_enable_msi(pf->pdev);
+ if (err) {
+ dev_info(&pf->pdev->dev,
+ "MSI init failed (%d), trying legacy.\n", err);
+ pf->flags &= ~I40E_FLAG_MSI_ENABLED;
+ }
+ }
+
+ /* track first vector for misc interrupts */
+ err = i40e_get_lump(pf, pf->irq_pile, 1, I40E_PILE_VALID_BIT-1);
+}
+
+/**
+ * i40e_setup_misc_vector - Setup the misc vector to handle non queue events
+ * @pf: board private structure
+ *
+ * This sets up the handler for MSIX 0, which is used to manage the
+ * non-queue interrupts, e.g. AdminQ and errors. This is not used
+ * when in MSI or Legacy interrupt mode.
+ **/
+static int i40e_setup_misc_vector(struct i40e_pf *pf)
+{
+ struct i40e_hw *hw = &pf->hw;
+ int err = I40E_SUCCESS;
+
+ /* Only request the irq if this is the first time through, and
+ * not when we're rebuilding after a Reset
+ */
+ if (!test_bit(__I40E_RESET_RECOVERY_PENDING, &pf->state)) {
+ err = request_irq(pf->msix_entries[0].vector,
+ i40e_intr, 0, pf->misc_int_name, pf);
+ if (err) {
+ dev_info(&pf->pdev->dev,
+ "request_irq for msix_misc failed: %d\n", err);
+ return I40E_ERR_CONFIG;
+ }
+ }
+
+ i40e_enable_misc_int_causes(hw);
+
+ /* associate no queues to the misc vector */
+ wr32(hw, I40E_PFINT_LNKLST0, I40E_QUEUE_END_OF_LIST);
+ wr32(hw, I40E_PFINT_ITR0(I40E_RX_ITR), I40E_ITR_8K);
+
+ i40e_flush(hw);
+
+ i40e_irq_dynamic_enable_icr0(pf);
+
+ return err;
+}
+
+/**
+ * i40e_config_rss - Prepare for RSS if used
+ * @pf: board private structure
+ **/
+static s32 i40e_config_rss(struct i40e_pf *pf)
+{
+ struct i40e_hw *hw = &pf->hw;
+ u32 lut = 0;
+ int i, j;
+ u64 hena;
+ /* Set of random keys generated using kernel random number generator */
+ static const u32 seed[I40E_PFQF_HKEY_MAX_INDEX + 1] = {0x41b01687,
+ 0x183cfd8c, 0xce880440, 0x580cbc3c, 0x35897377,
+ 0x328b25e1, 0x4fa98922, 0xb7d90c14, 0xd5bad70d,
+ 0xcd15a2c1, 0xe8580225, 0x4a1e9d11, 0xfe5731be};
+
+ /* Fill out hash function seed */
+ for (i = 0; i <= I40E_PFQF_HKEY_MAX_INDEX; i++)
+ wr32(hw, I40E_PFQF_HKEY(i), seed[i]);
+
+ /* By default we enable TCP/UDP with IPv4/IPv6 ptypes */
+ hena = (u64)rd32(hw, I40E_PFQF_HENA(0)) |
+ ((u64)rd32(hw, I40E_PFQF_HENA(1)) << 32);
+ hena |= ((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV4_UDP) |
+ ((u64)1 << I40E_FILTER_PCTYPE_NONF_UNICAST_IPV4_UDP) |
+ ((u64)1 << I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV4_UDP) |
+ ((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV4_TCP) |
+ ((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV6_TCP) |
+ ((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV6_UDP) |
+ ((u64)1 << I40E_FILTER_PCTYPE_NONF_UNICAST_IPV6_UDP) |
+ ((u64)1 << I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV6_UDP) |
+ ((u64)1 << I40E_FILTER_PCTYPE_FRAG_IPV4)|
+ ((u64)1 << I40E_FILTER_PCTYPE_FRAG_IPV6);
+ wr32(hw, I40E_PFQF_HENA(0), (u32)hena);
+ wr32(hw, I40E_PFQF_HENA(1), (u32)(hena >> 32));
+
+ /* Populate the LUT with max no. of queues in round robin fashion */
+ for (i = 0, j = 0; i < pf->hw.func_caps.rss_table_size; i++, j++) {
+
+ /* The assumption is that lan qp count will be the highest
+ * qp count for any PF VSI that needs RSS.
+ * If multiple VSIs need RSS support, all the qp counts
+ * for those VSIs should be a power of 2 for RSS to work.
+ * If LAN VSI is the only consumer for RSS then this requirement
+ * is not necessary.
+ */
+ if (j == pf->rss_size)
+ j = 0;
+ /* lut = 4-byte sliding window of 4 lut entries */
+ lut = (lut << 8) | (j &
+ ((0x1 << pf->hw.func_caps.rss_table_entry_width) - 1));
+ /* On i = 3, we have 4 entries in lut; write to the register */
+ if ((i & 3) == 3)
+ wr32(hw, I40E_PFQF_HLUT(i >> 2), lut);
+ }
+ i40e_flush(hw);
+
+ return I40E_SUCCESS;
+}
+
+/**
+ * i40e_sw_init - Initialize general software structures (struct i40e_pf)
+ * @pf: board private structure to initialize
+ *
+ * i40e_sw_init initializes the Adapter private data structure.
+ * Fields are initialized based on PCI device information and
+ * OS network device settings (MTU size).
+ **/
+static int i40e_sw_init(struct i40e_pf *pf)
+{
+ int err = 0;
+ int size;
+
+ pf->msg_enable = netif_msg_init(I40E_DEFAULT_MSG_ENABLE,
+ (NETIF_MSG_DRV|NETIF_MSG_PROBE|NETIF_MSG_LINK));
+ if (debug != -1 && debug != I40E_DEFAULT_MSG_ENABLE) {
+ if (I40E_DEBUG_USER & debug)
+ pf->hw.debug_mask = debug;
+ pf->msg_enable = netif_msg_init((debug & ~I40E_DEBUG_USER),
+ I40E_DEFAULT_MSG_ENABLE);
+ }
+
+ /* Set default capability flags */
+ pf->flags = I40E_FLAG_RX_CSUM_ENABLED |
+ I40E_FLAG_MSI_ENABLED |
+ I40E_FLAG_MSIX_ENABLED |
+ I40E_FLAG_RX_PS_ENABLED |
+ I40E_FLAG_MQ_ENABLED |
+ I40E_FLAG_RX_1BUF_ENABLED;
+
+ pf->rss_size_max = 0x1 << pf->hw.func_caps.rss_table_entry_width;
+ if (pf->hw.func_caps.rss) {
+ pf->flags |= I40E_FLAG_RSS_ENABLED;
+ pf->rss_size = min_t(int, pf->rss_size_max,
+ nr_cpus_node(numa_node_id()));
+ } else {
+ pf->rss_size = 1;
+ }
+
+ if (pf->hw.func_caps.dcb)
+ pf->num_tc_qps = I40E_DEFAULT_QUEUES_PER_TC;
+ else
+ pf->num_tc_qps = 0;
+
+ if (pf->hw.func_caps.fd) {
+ /* FW/NVM is not yet fixed in this regard */
+ if ((pf->hw.func_caps.fd_filters_guaranteed > 0) ||
+ (pf->hw.func_caps.fd_filters_best_effort > 0)) {
+ pf->flags |= I40E_FLAG_FDIR_ATR_ENABLED;
+ dev_info(&pf->pdev->dev,
+ "Flow Director ATR mode Enabled\n");
+ pf->flags |= I40E_FLAG_FDIR_ENABLED;
+ dev_info(&pf->pdev->dev,
+ "Flow Director Side Band mode Enabled\n");
+ pf->fdir_pf_filter_count =
+ pf->hw.func_caps.fd_filters_guaranteed;
+ }
+ } else {
+ pf->fdir_pf_filter_count = 0;
+ }
+
+ if (pf->hw.func_caps.vmdq) {
+ pf->flags |= I40E_FLAG_VMDQ_ENABLED;
+ pf->num_vmdq_vsis = I40E_DEFAULT_NUM_VMDQ_VSI;
+ pf->num_vmdq_qps = I40E_DEFAULT_QUEUES_PER_VMDQ;
+ }
+
+ /* MFP mode enabled */
+ if (pf->hw.func_caps.npar_enable || pf->hw.func_caps.mfp_mode_1) {
+ pf->flags |= I40E_FLAG_MFP_ENABLED;
+ dev_info(&pf->pdev->dev, "MFP mode Enabled\n");
+ }
+
+#ifdef CONFIG_PCI_IOV
+ if (pf->hw.func_caps.num_vfs) {
+ pf->num_vf_qps = I40E_DEFAULT_QUEUES_PER_VF;
+ pf->flags |= I40E_FLAG_SRIOV_ENABLED;
+ pf->num_req_vfs = min_t(int,
+ pf->hw.func_caps.num_vfs,
+ I40E_MAX_VF_COUNT);
+ }
+#endif /* CONFIG_PCI_IOV */
+ pf->eeprom_version = 0xDEAD;
+ pf->lan_veb = I40E_NO_VEB;
+ pf->lan_vsi = I40E_NO_VSI;
+
+ /* set up queue assignment tracking */
+ size = sizeof(struct i40e_lump_tracking)
+ + (sizeof(u16) * pf->hw.func_caps.num_tx_qp);
+ pf->qp_pile = kzalloc(size, GFP_KERNEL);
+ if (!pf->qp_pile) {
+ err = -ENOMEM;
+ goto sw_init_done;
+ }
+ pf->qp_pile->num_entries = pf->hw.func_caps.num_tx_qp;
+ pf->qp_pile->search_hint = 0;
+
+ /* set up vector assignment tracking */
+ size = sizeof(struct i40e_lump_tracking)
+ + (sizeof(u16) * pf->hw.func_caps.num_msix_vectors);
+ pf->irq_pile = kzalloc(size, GFP_KERNEL);
+ if (!pf->irq_pile) {
+ kfree(pf->qp_pile);
+ err = -ENOMEM;
+ goto sw_init_done;
+ }
+ pf->irq_pile->num_entries = pf->hw.func_caps.num_msix_vectors;
+ pf->irq_pile->search_hint = 0;
+
+ mutex_init(&pf->switch_mutex);
+
+sw_init_done:
+ return err;
+}
+
+/**
+ * i40e_set_features - set the netdev feature flags
+ * @netdev: ptr to the netdev being adjusted
+ * @features: the feature set that the stack is suggesting
+ **/
+static int i40e_set_features(struct net_device *netdev,
+ netdev_features_t features)
+{
+ struct i40e_netdev_priv *np = netdev_priv(netdev);
+ struct i40e_vsi *vsi = np->vsi;
+
+ if (features & NETIF_F_HW_VLAN_CTAG_RX)
+ i40e_vlan_stripping_enable(vsi);
+ else
+ i40e_vlan_stripping_disable(vsi);
+
+ return 0;
+}
+
+static const struct net_device_ops i40e_netdev_ops = {
+ .ndo_open = i40e_open,
+ .ndo_stop = i40e_close,
+ .ndo_start_xmit = i40e_lan_xmit_frame,
+ .ndo_get_stats64 = i40e_get_netdev_stats_struct,
+ .ndo_set_rx_mode = i40e_set_rx_mode,
+ .ndo_validate_addr = eth_validate_addr,
+ .ndo_set_mac_address = i40e_set_mac,
+ .ndo_change_mtu = i40e_change_mtu,
+ .ndo_tx_timeout = i40e_tx_timeout,
+ .ndo_vlan_rx_add_vid = i40e_vlan_rx_add_vid,
+ .ndo_vlan_rx_kill_vid = i40e_vlan_rx_kill_vid,
+#ifdef CONFIG_NET_POLL_CONTROLLER
+ .ndo_poll_controller = i40e_netpoll,
+#endif
+ .ndo_setup_tc = i40e_setup_tc,
+ .ndo_set_features = i40e_set_features,
+ .ndo_set_vf_mac = i40e_ndo_set_vf_mac,
+ .ndo_set_vf_vlan = i40e_ndo_set_vf_port_vlan,
+ .ndo_set_vf_tx_rate = i40e_ndo_set_vf_bw,
+ .ndo_get_vf_config = i40e_ndo_get_vf_config,
+};
+
+/**
+ * i40e_config_netdev - Setup the netdev flags
+ * @vsi: the VSI being configured
+ *
+ * Returns 0 on success, negative value on failure
+ **/
+static s32 i40e_config_netdev(struct i40e_vsi *vsi)
+{
+ struct i40e_pf *pf = vsi->back;
+ struct i40e_hw *hw = &pf->hw;
+ struct i40e_netdev_priv *np;
+ struct net_device *netdev;
+ u8 mac_addr[ETH_ALEN];
+ int etherdev_size;
+
+ etherdev_size = (sizeof(struct i40e_netdev_priv));
+ netdev = alloc_etherdev_mq(etherdev_size, vsi->alloc_queue_pairs);
+ if (!netdev)
+ return -ENOMEM;
+
+ vsi->netdev = netdev;
+ np = netdev_priv(netdev);
+ np->vsi = vsi;
+
+ netdev->hw_enc_features = NETIF_F_IP_CSUM |
+ NETIF_F_GSO_UDP_TUNNEL |
+ NETIF_F_TSO |
+ NETIF_F_SG;
+
+ netdev->features = NETIF_F_SG |
+ NETIF_F_IP_CSUM |
+ NETIF_F_SCTP_CSUM |
+ NETIF_F_HIGHDMA |
+ NETIF_F_GSO_UDP_TUNNEL |
+ NETIF_F_HW_VLAN_CTAG_TX |
+ NETIF_F_HW_VLAN_CTAG_RX |
+ NETIF_F_HW_VLAN_CTAG_FILTER |
+ NETIF_F_IPV6_CSUM |
+ NETIF_F_TSO |
+ NETIF_F_TSO6 |
+ NETIF_F_RXCSUM |
+ NETIF_F_RXHASH |
+ 0;
+
+ /* copy netdev features into list of user selectable features */
+ netdev->hw_features |= netdev->features;
+
+ if (vsi->type == I40E_VSI_MAIN) {
+ SET_NETDEV_DEV(netdev, &pf->pdev->dev);
+ memcpy(mac_addr, hw->mac.perm_addr, ETH_ALEN);
+ strlcpy(netdev->name, "eth%d", IFNAMSIZ-1);
+ } else {
+ random_ether_addr(mac_addr);
+ snprintf(netdev->name, IFNAMSIZ, "%sv%%d",
+ pf->vsi[pf->lan_vsi]->netdev->name);
+ i40e_add_filter(vsi, mac_addr, I40E_VLAN_ANY, false, false);
+ }
+
+ memcpy(netdev->dev_addr, mac_addr, netdev->addr_len);
+ memcpy(netdev->perm_addr, mac_addr, netdev->addr_len);
+ /* vlan gets same features (except vlan offload)
+ * after any tweaks for specific VSI types
+ */
+ netdev->vlan_features = netdev->features & ~(NETIF_F_HW_VLAN_CTAG_TX |
+ NETIF_F_HW_VLAN_CTAG_RX |
+ NETIF_F_HW_VLAN_CTAG_FILTER);
+ netdev->priv_flags |= IFF_UNICAST_FLT;
+ netdev->priv_flags |= IFF_SUPP_NOFCS;
+ /* Setup netdev TC information */
+ i40e_vsi_config_netdev_tc(vsi, vsi->tc_config.enabled_tc);
+
+ netdev->netdev_ops = &i40e_netdev_ops;
+ netdev->watchdog_timeo = 5 * HZ;
+ i40e_set_ethtool_ops(netdev);
+
+ return I40E_SUCCESS;
+}
+
+/**
+ * i40e_vsi_delete - Delete a VSI from the switch
+ * @vsi: the VSI being removed
+ *
+ * Returns 0 on success, negative value on failure
+ **/
+static s32 i40e_vsi_delete(struct i40e_vsi *vsi)
+{
+ int ret = I40E_ERR_PARAM;
+
+ /* remove default VSI is not allowed */
+ if (vsi == vsi->back->vsi[vsi->back->lan_vsi])
+ return ret;
+
+ /* there is no HW VSI for FDIR */
+ if (vsi->type == I40E_VSI_FDIR)
+ return I40E_SUCCESS;
+
+ ret = i40e_aq_delete_element(&vsi->back->hw, vsi->seid, NULL);
+ return ret;
+}
+
+/**
+ * i40e_add_vsi - Add a VSI to the switch
+ * @vsi: the VSI being configured
+ *
+ * This initializes a VSI context depending on the VSI type to be added and
+ * passes it down to the add_vsi aq command.
+ **/
+static s32 i40e_add_vsi(struct i40e_vsi *vsi)
+{
+ int ret = I40E_ERR_NOT_IMPLEMENTED;
+ struct i40e_mac_filter *f, *ftmp;
+ struct i40e_pf *pf = vsi->back;
+ struct i40e_hw *hw = &pf->hw;
+ struct i40e_vsi_context ctxt;
+ u8 enabled_tc = 0x1; /* TC0 enabled */
+ int f_count = 0;
+
+ memset(&ctxt, 0, sizeof(ctxt));
+ switch (vsi->type) {
+ case I40E_VSI_MAIN:
+ /* The PF's main VSI is already setup as part of the
+ * device initialization, so we'll not bother with
+ * the add_vsi call, but we will retrieve the current
+ * VSI context.
+ */
+ ctxt.seid = pf->main_vsi_seid;
+ ctxt.pf_num = pf->hw.pf_id;
+ ctxt.vf_num = 0;
+ ret = i40e_aq_get_vsi_params(&pf->hw, &ctxt, NULL);
+ ctxt.flags = I40E_AQ_VSI_TYPE_PF;
+ if (ret) {
+ dev_info(&pf->pdev->dev,
+ "couldn't get pf vsi config, err %d, aq_err %d\n",
+ ret, pf->hw.aq.asq_last_status);
+ return ret;
+ }
+ memcpy(&vsi->info, &ctxt.info, sizeof(ctxt.info));
+ vsi->info.valid_sections = 0;
+
+ vsi->seid = ctxt.seid;
+ vsi->id = ctxt.vsi_number;
+
+ enabled_tc = i40e_pf_get_tc_map(pf);
+
+ /* MFP mode setup queue map and update VSI */
+ if (pf->flags & I40E_FLAG_MFP_ENABLED) {
+ memset(&ctxt, 0, sizeof(ctxt));
+ ctxt.seid = pf->main_vsi_seid;
+ ctxt.pf_num = pf->hw.pf_id;
+ ctxt.vf_num = 0;
+ i40e_vsi_setup_queue_map(vsi, &ctxt, enabled_tc, false);
+ ret = i40e_aq_update_vsi_params(hw, &ctxt, NULL);
+ if (ret != I40E_SUCCESS) {
+ dev_info(&pf->pdev->dev,
+ "update vsi failed, aq_err=%d\n",
+ pf->hw.aq.asq_last_status);
+ goto err;
+ }
+ /* update the local VSI info queue map */
+ i40e_vsi_update_queue_map(vsi, &ctxt);
+ vsi->info.valid_sections = 0;
+ } else {
+ /* Default/Main VSI is only enabled for TC0
+ * reconfigure it to enable all TCs that are
+ * available on the port in SFP mode.
+ */
+ ret = i40e_vsi_config_tc(vsi, enabled_tc);
+ if (ret) {
+ dev_info(&pf->pdev->dev,
+ "failed to configure TCs for main VSI tc_map 0x%08x, err %d, aq_err %d\n",
+ enabled_tc, ret,
+ pf->hw.aq.asq_last_status);
+ }
+ }
+ break;
+
+ case I40E_VSI_FDIR:
+ /* no queue mapping or actual HW VSI needed */
+ vsi->info.valid_sections = 0;
+ vsi->seid = 0;
+ vsi->id = 0;
+ i40e_vsi_setup_queue_map(vsi, &ctxt, enabled_tc, true);
+ return I40E_SUCCESS;
+ break;
+
+ case I40E_VSI_VMDQ2:
+ ctxt.pf_num = hw->pf_id;
+ ctxt.vf_num = 0;
+ ctxt.uplink_seid = vsi->uplink_seid;
+ ctxt.connection_type = 0x1; /* regular data port */
+ ctxt.flags = I40E_AQ_VSI_TYPE_VMDQ2;
+
+ ctxt.info.valid_sections |= cpu_to_le16(I40E_AQ_VSI_PROP_SWITCH_VALID);
+
+ /* This VSI is connected to VEB so the switch_id
+ * should be set to zero by default.
+ */
+ ctxt.info.switch_id = 0;
+ ctxt.info.switch_id |= cpu_to_le16(I40E_AQ_VSI_SW_ID_FLAG_LOCAL_LB);
+ ctxt.info.switch_id |= cpu_to_le16(I40E_AQ_VSI_SW_ID_FLAG_ALLOW_LB);
+
+ /* Setup the VSI tx/rx queue map for TC0 only for now */
+ i40e_vsi_setup_queue_map(vsi, &ctxt, enabled_tc, true);
+ break;
+
+ case I40E_VSI_SRIOV:
+ ctxt.pf_num = hw->pf_id;
+ ctxt.vf_num = vsi->vf_id + hw->func_caps.vf_base_id;
+ ctxt.uplink_seid = vsi->uplink_seid;
+ ctxt.connection_type = 0x1; /* regular data port */
+ ctxt.flags = I40E_AQ_VSI_TYPE_VF;
+
+ ctxt.info.valid_sections |= cpu_to_le16(I40E_AQ_VSI_PROP_SWITCH_VALID);
+
+ /* This VSI is connected to VEB so the switch_id
+ * should be set to zero by default.
+ */
+ ctxt.info.switch_id = cpu_to_le16(I40E_AQ_VSI_SW_ID_FLAG_ALLOW_LB);
+
+ ctxt.info.valid_sections |= cpu_to_le16(I40E_AQ_VSI_PROP_VLAN_VALID);
+ ctxt.info.port_vlan_flags |= I40E_AQ_VSI_PVLAN_MODE_ALL;
+ /* Setup the VSI tx/rx queue map for TC0 only for now */
+ i40e_vsi_setup_queue_map(vsi, &ctxt, enabled_tc, true);
+ break;
+
+ default:
+ return ret;
+ }
+
+ if (vsi->type != I40E_VSI_MAIN) {
+ ret = i40e_aq_add_vsi(hw, &ctxt, NULL);
+ if (ret) {
+ dev_info(&vsi->back->pdev->dev,
+ "add vsi failed, aq_err=%d\n",
+ vsi->back->hw.aq.asq_last_status);
+ goto err;
+ }
+ memcpy(&vsi->info, &ctxt.info, sizeof(ctxt.info));
+ vsi->info.valid_sections = 0;
+ vsi->seid = ctxt.seid;
+ vsi->id = ctxt.vsi_number;
+ }
+
+ /* If macvlan filters already exist, force them to get loaded */
+ list_for_each_entry_safe(f, ftmp, &vsi->mac_filter_list, list) {
+ f->changed = true;
+ f_count++;
+ }
+ if (f_count) {
+ vsi->flags |= I40E_VSI_FLAG_FILTER_CHANGED;
+ pf->flags |= I40E_FLAG_FILTER_SYNC;
+ }
+
+ /* Update VSI BW information */
+ ret = i40e_vsi_get_bw_info(vsi);
+ if (ret) {
+ dev_info(&pf->pdev->dev,
+ "couldn't get vsi bw info, err %d, aq_err %d\n",
+ ret, pf->hw.aq.asq_last_status);
+ /* VSI is already added so not tearing that up */
+ ret = I40E_SUCCESS;
+ }
+
+err:
+ return ret;
+}
+
+/**
+ * i40e_vsi_release - Delete a VSI and free its resources
+ * @vsi: the VSI being removed
+ *
+ * Returns I40E_SUCCESS on success or < 0 on error
+ **/
+s32 i40e_vsi_release(struct i40e_vsi *vsi)
+{
+ struct i40e_mac_filter *f, *ftmp;
+ struct i40e_veb *veb = NULL;
+ struct i40e_pf *pf;
+ u16 uplink_seid;
+ int i, n;
+
+ if (!vsi)
+ return I40E_SUCCESS;
+
+ pf = vsi->back;
+
+ /* release of a VEB-owner or last VSI is not allowed */
+ if (vsi->flags & I40E_VSI_FLAG_VEB_OWNER) {
+ dev_info(&pf->pdev->dev, "VSI %d has existing VEB %d\n",
+ vsi->seid, vsi->uplink_seid);
+ return I40E_ERR_DEVICE_NOT_SUPPORTED;
+ }
+ if (vsi == pf->vsi[pf->lan_vsi] &&
+ !test_bit(__I40E_DOWN, &pf->state)) {
+ dev_info(&pf->pdev->dev, "Can't remove PF VSI\n");
+ return I40E_ERR_DEVICE_NOT_SUPPORTED;
+ }
+
+ uplink_seid = vsi->uplink_seid;
+ i40e_sys_del_vsi(vsi);
+ if (vsi->type != I40E_VSI_SRIOV) {
+ if (vsi->netdev_registered) {
+ vsi->netdev_registered = false;
+ if (vsi->netdev) {
+ /* results in a call to i40e_close() */
+ unregister_netdev(vsi->netdev);
+ free_netdev(vsi->netdev);
+ }
+ } else {
+ if (!test_and_set_bit(__I40E_DOWN, &vsi->state))
+ i40e_down(vsi);
+ i40e_vsi_free_irq(vsi);
+ i40e_vsi_free_tx_resources(vsi);
+ i40e_vsi_free_rx_resources(vsi);
+ }
+ i40e_vsi_disable_irq(vsi);
+ }
+
+ list_for_each_entry_safe(f, ftmp, &vsi->mac_filter_list, list)
+ i40e_del_filter(vsi, f->macaddr, f->vlan,
+ f->is_vf, f->is_netdev);
+ i40e_sync_vsi_filters(vsi);
+
+ i40e_vsi_delete(vsi);
+ i40e_vsi_free_q_vectors(vsi);
+ i40e_vsi_clear_rings(vsi);
+ i40e_vsi_clear(vsi);
+
+ /* If this was the last thing on the VEB, except for the
+ * controlling VSI, remove the VEB, which puts the controlling
+ * VSI onto the next level down in the switch.
+ *
+ * Well, okay, there's one more exception here: don't remove
+ * the orphan VEBs yet. We'll wait for an explicit remove request
+ * from up the network stack.
+ */
+ for (n = 0, i = 0; i < pf->hw.func_caps.num_vsis; i++) {
+ if (pf->vsi[i] &&
+ pf->vsi[i]->uplink_seid == uplink_seid &&
+ (pf->vsi[i]->flags & I40E_VSI_FLAG_VEB_OWNER) == 0) {
+ n++; /* count the VSIs */
+ }
+ }
+ for (i = 0; i < I40E_MAX_VEB; i++) {
+ if (!pf->veb[i])
+ continue;
+ if (pf->veb[i]->uplink_seid == uplink_seid)
+ n++; /* count the VEBs */
+ if (pf->veb[i]->seid == uplink_seid)
+ veb = pf->veb[i];
+ }
+ if (n == 0 && veb && veb->uplink_seid != 0)
+ i40e_veb_release(veb);
+
+ return I40E_SUCCESS;
+}
+
+/**
+ * i40e_vsi_setup_vectors - Set up the q_vectors for the given VSI
+ * @vsi: ptr to the VSI
+ *
+ * This should only be called after i40e_vsi_mem_alloc() which allocates the
+ * corresponding SW VSI structure and initializes num_queue_pairs for the
+ * newly allocated VSI.
+ *
+ * Returns 0 on success or negative on failure
+ **/
+static s32 i40e_vsi_setup_vectors(struct i40e_vsi *vsi)
+{
+ int ret = I40E_ERR_NOT_IMPLEMENTED;
+ struct i40e_pf *pf = vsi->back;
+
+ if (vsi->q_vectors) {
+ dev_info(&pf->pdev->dev, "VSI %d has existing q_vectors\n",
+ vsi->seid);
+ return I40E_ERR_CONFIG;
+ }
+
+ if (vsi->base_vector) {
+ dev_info(&pf->pdev->dev,
+ "VSI %d has non-zero base vector %d\n",
+ vsi->seid, vsi->base_vector);
+ return I40E_ERR_CONFIG;
+ }
+
+ ret = i40e_alloc_q_vectors(vsi);
+ if (ret != I40E_SUCCESS) {
+ dev_info(&pf->pdev->dev,
+ "failed to allocate %d q_vector for VSI %d, ret=%d\n",
+ vsi->num_q_vectors, vsi->seid, ret);
+ vsi->num_q_vectors = 0;
+ goto vector_setup_out;
+ }
+
+ vsi->base_vector = i40e_get_lump(pf, pf->irq_pile,
+ vsi->num_q_vectors, vsi->idx);
+ if (vsi->base_vector < 0) {
+ dev_info(&pf->pdev->dev,
+ "failed to get q tracking for VSI %d, err=%d\n",
+ vsi->seid, vsi->base_vector);
+ i40e_vsi_free_q_vectors(vsi);
+ ret = I40E_ERR_CONFIG;
+ goto vector_setup_out;
+ }
+
+vector_setup_out:
+ return ret;
+}
+
+/**
+ * i40e_vsi_setup - Set up a VSI by a given type
+ * @pf: board private structure
+ * @type: VSI type
+ * @uplink_seid: the switch element to link to
+ * @param1: usage depends upon VSI type. For VF types, indicates VF id
+ *
+ * This allocates the sw VSI structure and its queue resources, then add a VSI
+ * to the identified VEB.
+ *
+ * Returns pointer to the successfully allocated and configure VSI sw struct on
+ * success, otherwise returns NULL on failure.
+ **/
+struct i40e_vsi *i40e_vsi_setup(struct i40e_pf *pf, u8 type,
+ u16 uplink_seid, u32 param1)
+{
+ struct i40e_vsi *vsi = NULL;
+ struct i40e_veb *veb = NULL;
+ int ret, i;
+ int v_idx;
+
+ /* The requested uplink_seid must be either
+ * - the PF's port seid
+ * no VEB is needed because this is the PF
+ * or this is a Flow Director special case VSI
+ * - seid of an existing VEB
+ * - seid of a VSI that owns an existing VEB
+ * - seid of a VSI that doesn't own a VEB
+ * a new VEB is created and the VSI becomes the owner
+ * - seid of the PF VSI, which is what creates the first VEB
+ * this is a special case of the previous
+ *
+ * Find which uplink_seid we were given and create a new VEB if needed
+ */
+ for (i = 0; i < I40E_MAX_VEB; i++) {
+ if (pf->veb[i] && pf->veb[i]->seid == uplink_seid) {
+ veb = pf->veb[i];
+ break;
+ }
+ }
+
+ if (!veb && uplink_seid != pf->mac_seid) {
+
+ for (i = 0; i < pf->hw.func_caps.num_vsis; i++) {
+ if (pf->vsi[i] && pf->vsi[i]->seid == uplink_seid) {
+ vsi = pf->vsi[i];
+ break;
+ }
+ }
+ if (!vsi) {
+ dev_info(&pf->pdev->dev, "no such uplink_seid %d\n",
+ uplink_seid);
+ return NULL;
+ }
+
+ if (vsi->uplink_seid == pf->mac_seid)
+ veb = i40e_veb_setup(pf, 0, pf->mac_seid, vsi->seid,
+ vsi->tc_config.enabled_tc);
+ else if ((vsi->flags & I40E_VSI_FLAG_VEB_OWNER) == 0)
+ veb = i40e_veb_setup(pf, 0, vsi->uplink_seid, vsi->seid,
+ vsi->tc_config.enabled_tc);
+
+ for (i = 0; i < I40E_MAX_VEB && !veb; i++) {
+ if (pf->veb[i] && pf->veb[i]->seid == vsi->uplink_seid)
+ veb = pf->veb[i];
+ }
+ if (!veb) {
+ dev_info(&pf->pdev->dev, "couldn't add VEB\n");
+ return NULL;
+ }
+
+ vsi->flags |= I40E_VSI_FLAG_VEB_OWNER;
+ uplink_seid = veb->seid;
+ }
+
+ /* get vsi sw struct */
+ v_idx = i40e_vsi_mem_alloc(pf, type);
+ if (v_idx < 0)
+ goto err_alloc;
+ vsi = pf->vsi[v_idx];
+ vsi->type = type;
+ vsi->veb_idx = (veb ? veb->idx : I40E_NO_VEB);
+
+ if (type == I40E_VSI_MAIN)
+ pf->lan_vsi = v_idx;
+ else if (type == I40E_VSI_SRIOV)
+ vsi->vf_id = param1;
+ /* assign it some queues */
+ ret = i40e_get_lump(pf, pf->qp_pile, vsi->alloc_queue_pairs, vsi->idx);
+ if (ret < 0) {
+ dev_info(&pf->pdev->dev, "VSI %d get_lump failed %d\n",
+ vsi->seid, ret);
+ goto err_vsi;
+ }
+ vsi->base_queue = ret;
+
+ /* get a VSI from the hardware */
+ vsi->uplink_seid = uplink_seid;
+ ret = i40e_add_vsi(vsi);
+ if (ret != I40E_SUCCESS)
+ goto err_vsi;
+
+ switch (vsi->type) {
+ /* setup the netdev if needed */
+ case I40E_VSI_MAIN:
+ case I40E_VSI_VMDQ2:
+ ret = i40e_config_netdev(vsi);
+ if (ret)
+ goto err_netdev;
+ ret = register_netdev(vsi->netdev);
+ if (ret)
+ goto err_netdev;
+ vsi->netdev_registered = true;
+ netif_carrier_off(vsi->netdev);
+ /* fall through */
+
+ case I40E_VSI_FDIR:
+ /* set up vectors and rings if needed */
+ ret = i40e_vsi_setup_vectors(vsi);
+ if (ret)
+ goto err_msix;
+
+ ret = i40e_alloc_rings(vsi);
+ if (ret)
+ goto err_rings;
+
+ /* map all of the rings to the q_vectors */
+ i40e_vsi_map_rings_to_vectors(vsi);
+
+ i40e_vsi_reset_stats(vsi);
+ break;
+
+ default:
+ /* no netdev or rings for the other VSI types */
+ break;
+ }
+
+ i40e_sys_add_vsi(vsi);
+ return vsi;
+
+err_rings:
+ i40e_vsi_free_q_vectors(vsi);
+err_msix:
+ if (vsi->netdev_registered) {
+ vsi->netdev_registered = false;
+ unregister_netdev(vsi->netdev);
+ free_netdev(vsi->netdev);
+ vsi->netdev = NULL;
+ }
+err_netdev:
+ i40e_aq_delete_element(&pf->hw, vsi->seid, NULL);
+err_vsi:
+ i40e_vsi_clear(vsi);
+err_alloc:
+ return NULL;
+}
+
+/**
+ * i40e_veb_get_bw_info - Query VEB BW information
+ * @veb: the veb to query
+ *
+ * Query the Tx scheduler BW configuration data for given VEB
+ **/
+static s32 i40e_veb_get_bw_info(struct i40e_veb *veb)
+{
+ struct i40e_aqc_query_switching_comp_ets_config_resp ets_data;
+ struct i40e_aqc_query_switching_comp_bw_config_resp bw_data;
+ struct i40e_pf *pf = veb->pf;
+ struct i40e_hw *hw = &pf->hw;
+ int ret = I40E_SUCCESS;
+ u32 tc_bw_max;
+ int i;
+
+ ret = i40e_aq_query_switch_comp_bw_config(hw, veb->seid,
+ &bw_data, NULL);
+ if (ret != I40E_SUCCESS) {
+ dev_info(&pf->pdev->dev,
+ "query veb bw config failed, aq_err=%d\n",
+ hw->aq.asq_last_status);
+ goto out;
+ }
+
+ ret = i40e_aq_query_switch_comp_ets_config(hw, veb->seid,
+ &ets_data, NULL);
+ if (ret != I40E_SUCCESS) {
+ dev_info(&pf->pdev->dev,
+ "query veb bw ets config failed, aq_err=%d\n",
+ hw->aq.asq_last_status);
+ goto out;
+ }
+
+ veb->bw_limit = le16_to_cpu(ets_data.port_bw_limit);
+ veb->bw_max_quanta = ets_data.tc_bw_max;
+ veb->is_abs_credits = bw_data.absolute_credits_enable;
+ tc_bw_max = le16_to_cpu(bw_data.tc_bw_max[0]) |
+ (le16_to_cpu(bw_data.tc_bw_max[1]) << 16);
+ for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++) {
+ veb->bw_tc_share_credits[i] = bw_data.tc_bw_share_credits[i];
+ veb->bw_tc_limit_credits[i] =
+ le16_to_cpu(bw_data.tc_bw_limits[i]);
+ veb->bw_tc_max_quanta[i] = ((tc_bw_max >> (i*4)) & 0x7);
+ }
+
+out:
+ return ret;
+}
+
+/**
+ * i40e_veb_mem_alloc - Allocates the next available struct veb in the PF
+ * @pf: board private structure
+ *
+ * On error: returns error code (negative)
+ * On success: returns vsi index in PF (positive)
+ **/
+static s32 i40e_veb_mem_alloc(struct i40e_pf *pf)
+{
+ int ret = I40E_ERR_NO_AVAILABLE_VSI;
+ struct i40e_veb *veb;
+ int i;
+
+ /* Need to protect the allocation of switch elements at the PF level */
+ mutex_lock(&pf->switch_mutex);
+
+ /* VEB list may be fragmented if VEB creation/destruction has
+ * been happening. We can afford to do a quick scan to look
+ * for any free slots in the list.
+ *
+ * find next empty veb slot, looping back around if necessary
+ */
+ i = 0;
+ while ((i < I40E_MAX_VEB) && (pf->veb[i] != NULL))
+ i++;
+ if (i >= I40E_MAX_VEB) {
+ ret = I40E_ERR_NO_MEMORY;
+ goto err_alloc_veb; /* out of VEB slots! */
+ }
+
+ veb = kzalloc(sizeof(*veb), GFP_KERNEL);
+ if (!veb) {
+ ret = -ENOMEM;
+ goto err_alloc_veb;
+ }
+ veb->pf = pf;
+ veb->idx = i;
+ veb->enabled_tc = 1;
+
+ pf->veb[i] = veb;
+ ret = i;
+err_alloc_veb:
+ mutex_unlock(&pf->switch_mutex);
+ return ret;
+}
+
+/**
+ * i40e_switch_branch_release - Delete a branch of the switch tree
+ * @branch: where to start deleting
+ *
+ * This uses recursion to find the tips of the branch to be
+ * removed, deleting until we get back to and can delete this VEB.
+ **/
+static void i40e_switch_branch_release(struct i40e_veb *branch)
+{
+ struct i40e_pf *pf = branch->pf;
+ u16 branch_seid = branch->seid;
+ u16 veb_idx = branch->idx;
+ int i;
+
+ /* release any VEBs on this VEB - RECURSION */
+ for (i = 0; i < I40E_MAX_VEB; i++) {
+ if (!pf->veb[i])
+ continue;
+ if (pf->veb[i]->uplink_seid == branch->seid)
+ i40e_switch_branch_release(pf->veb[i]);
+ }
+
+ /* Release the VSIs on this VEB, but not the owner VSI.
+ *
+ * NOTE: Removing the last VSI on a VEB has the SIDE EFFECT of removing
+ * the VEB itself, so don't use (*branch) after this loop.
+ */
+ for (i = 0; i < pf->hw.func_caps.num_vsis; i++) {
+ if (!pf->vsi[i])
+ continue;
+ if (pf->vsi[i]->uplink_seid == branch_seid &&
+ (pf->vsi[i]->flags & I40E_VSI_FLAG_VEB_OWNER) == 0) {
+ i40e_vsi_release(pf->vsi[i]);
+ }
+ }
+
+ /* There's one corner case where the VEB might not have been
+ * removed, so double check it here and remove it if needed.
+ * This case happens if the veb was created from the debugfs
+ * commands and no VSIs were added to it.
+ */
+ if (pf->veb[veb_idx])
+ i40e_veb_release(pf->veb[veb_idx]);
+}
+
+/**
+ * i40e_veb_clear - remove veb struct
+ * @veb: the veb to remove
+ **/
+static void i40e_veb_clear(struct i40e_veb *veb)
+{
+ if (!veb)
+ return;
+
+ if (veb->pf) {
+ struct i40e_pf *pf = veb->pf;
+
+ mutex_lock(&pf->switch_mutex);
+ if (pf->veb[veb->idx] == veb)
+ pf->veb[veb->idx] = NULL;
+ mutex_unlock(&pf->switch_mutex);
+ }
+
+ kfree(veb);
+}
+
+/**
+ * i40e_veb_release - Delete a VEB and free its resources
+ * @veb: the VEB being removed
+ **/
+s32 i40e_veb_release(struct i40e_veb *veb)
+{
+ struct i40e_vsi *vsi = NULL;
+ int ret = I40E_ERR_PARAM;
+ struct i40e_pf *pf;
+ int i, n = 0;
+
+ if (!veb)
+ return I40E_SUCCESS;
+ pf = veb->pf;
+
+ /* find the remaining VSI and check for extras */
+ for (i = 0; i < pf->hw.func_caps.num_vsis; i++) {
+ if (pf->vsi[i] && pf->vsi[i]->uplink_seid == veb->seid) {
+ n++;
+ vsi = pf->vsi[i];
+ }
+ }
+ if (n != 1) {
+ dev_info(&pf->pdev->dev,
+ "can't remove VEB %d with %d VSIs left\n",
+ veb->seid, n);
+ return I40E_ERR_NOT_READY;
+ }
+
+ /* move the remaining VSI to uplink veb */
+ i40e_sys_del_vsi(vsi);
+ vsi->flags &= ~I40E_VSI_FLAG_VEB_OWNER;
+ if (veb->uplink_seid) {
+ vsi->uplink_seid = veb->uplink_seid;
+ if (veb->uplink_seid == pf->mac_seid)
+ vsi->veb_idx = I40E_NO_VEB;
+ else
+ vsi->veb_idx = veb->veb_idx;
+ } else {
+ /* floating VEB */
+ vsi->uplink_seid = pf->vsi[pf->lan_vsi]->uplink_seid;
+ vsi->veb_idx = pf->vsi[pf->lan_vsi]->veb_idx;
+ }
+ i40e_sys_add_vsi(vsi);
+
+ i40e_sys_del_veb(veb);
+ ret = i40e_aq_delete_element(&pf->hw, veb->seid, NULL);
+ i40e_veb_clear(veb);
+
+ return ret;
+}
+
+/**
+ * i40e_add_veb - create the VEB in the switch
+ * @veb: the VEB to be instantiated
+ * @vsi: the controlling VSI
+ **/
+static s32 i40e_add_veb(struct i40e_veb *veb, struct i40e_vsi *vsi)
+{
+ bool is_default = (vsi->idx == vsi->back->lan_vsi);
+ int ret;
+
+ /* get a VEB from the hardware */
+ ret = i40e_aq_add_veb(&veb->pf->hw, veb->uplink_seid, vsi->seid,
+ veb->enabled_tc, is_default, &veb->seid, NULL);
+ if (ret != I40E_SUCCESS) {
+ dev_info(&veb->pf->pdev->dev,
+ "couldn't add VEB, err %d, aq_err %d\n",
+ ret, veb->pf->hw.aq.asq_last_status);
+ return ret;
+ }
+
+ /* get statistics counter */
+ ret = i40e_aq_get_veb_parameters(&veb->pf->hw, veb->seid, NULL, NULL,
+ &veb->stats_idx, NULL, NULL, NULL);
+ if (ret != I40E_SUCCESS) {
+ dev_info(&veb->pf->pdev->dev,
+ "couldn't get VEB statistics idx, err %d, aq_err %d\n",
+ ret, veb->pf->hw.aq.asq_last_status);
+ return ret;
+ }
+ ret = i40e_veb_get_bw_info(veb);
+ if (ret != I40E_SUCCESS) {
+ dev_info(&veb->pf->pdev->dev,
+ "couldn't get VEB bw info, err %d, aq_err %d\n",
+ ret, veb->pf->hw.aq.asq_last_status);
+ i40e_aq_delete_element(&veb->pf->hw, veb->seid, NULL);
+ return ret;
+ }
+
+ i40e_sys_add_veb(veb);
+
+ /* move the VSI to this VEB in our model */
+ i40e_sys_del_vsi(vsi);
+ vsi->uplink_seid = veb->seid;
+ vsi->veb_idx = veb->idx;
+ vsi->flags |= I40E_VSI_FLAG_VEB_OWNER;
+ i40e_sys_add_vsi(vsi);
+
+ return I40E_SUCCESS;
+}
+
+/**
+ * i40e_veb_setup - Set up a VEB
+ * @pf: board private structure
+ * @flags: VEB setup flags
+ * @uplink_seid: the switch element to link to
+ * @vsi_seid: the initial VSI seid
+ * @enabled_tc: Enabled TC bit-map
+ *
+ * This allocates the sw VEB structure and links it into the switch
+ * It is possible and legal for this to be a duplicate of an already
+ * existing VEB. It is also possible for both uplink and vsi seids
+ * to be zero, in order to create a floating VEB.
+ *
+ * Returns pointer to the successfully allocated VEB sw struct on
+ * success, otherwise returns NULL on failure.
+ **/
+struct i40e_veb *i40e_veb_setup(struct i40e_pf *pf, u16 flags,
+ u16 uplink_seid, u16 vsi_seid,
+ u8 enabled_tc)
+{
+ struct i40e_veb *veb, *uplink_veb = NULL;
+ int vsi_idx, veb_idx;
+ int ret;
+
+ /* if one seid is 0, the other must be 0 to create a floating relay */
+ if ((uplink_seid == 0 || vsi_seid == 0) &&
+ (uplink_seid + vsi_seid != 0)) {
+ dev_info(&pf->pdev->dev,
+ "one, not both seid's are 0: uplink=%d vsi=%d\n",
+ uplink_seid, vsi_seid);
+ return NULL;
+ }
+
+ /* make sure there is such a vsi and uplink */
+ for (vsi_idx = 0; vsi_idx < pf->hw.func_caps.num_vsis; vsi_idx++)
+ if (pf->vsi[vsi_idx] && pf->vsi[vsi_idx]->seid == vsi_seid)
+ break;
+ if (vsi_idx >= pf->hw.func_caps.num_vsis && vsi_seid != 0) {
+ dev_info(&pf->pdev->dev, "vsi seid %d not found\n",
+ vsi_seid);
+ return NULL;
+ }
+
+ if (uplink_seid && uplink_seid != pf->mac_seid) {
+ for (veb_idx = 0; veb_idx < I40E_MAX_VEB; veb_idx++) {
+ if (pf->veb[veb_idx] &&
+ pf->veb[veb_idx]->seid == uplink_seid) {
+ uplink_veb = pf->veb[veb_idx];
+ break;
+ }
+ }
+ if (!uplink_veb) {
+ dev_info(&pf->pdev->dev,
+ "uplink seid %d not found\n", uplink_seid);
+ return NULL;
+ }
+ }
+
+ /* get veb sw struct */
+ veb_idx = i40e_veb_mem_alloc(pf);
+ if (veb_idx < 0)
+ goto err_alloc;
+ veb = pf->veb[veb_idx];
+ veb->flags = flags;
+ veb->uplink_seid = uplink_seid;
+ veb->veb_idx = (uplink_veb ? uplink_veb->idx : I40E_NO_VEB);
+ veb->enabled_tc = (enabled_tc ? enabled_tc : 0x1);
+
+ /* create the VEB in the switch */
+ ret = i40e_add_veb(veb, pf->vsi[vsi_idx]);
+ if (ret != I40E_SUCCESS)
+ goto err_veb;
+
+ return veb;
+
+err_veb:
+ i40e_veb_clear(veb);
+err_alloc:
+ return NULL;
+}
+
+/**
+ * i40e_setup_pf_switch_element - set pf vars based on switch type
+ * @pf: board private structure
+ * @ele: element we are building info from
+ * @num_reported: total number of elements
+ * @printconfig: should we print the contents
+ *
+ * helper function to assist in extracting a few useful SEID values.
+ **/
+static void i40e_setup_pf_switch_element(struct i40e_pf *pf,
+ struct i40e_aqc_switch_config_element_resp *ele,
+ u16 num_reported, bool printconfig)
+{
+ u16 downlink_seid = le16_to_cpu(ele->downlink_seid);
+ u16 uplink_seid = le16_to_cpu(ele->uplink_seid);
+ u8 element_type = ele->element_type;
+ u16 seid = le16_to_cpu(ele->seid);
+
+ if (printconfig)
+ dev_info(&pf->pdev->dev,
+ "type=%d seid=%d uplink=%d downlink=%d\n",
+ element_type, seid, uplink_seid, downlink_seid);
+
+ switch (element_type) {
+ case I40E_SWITCH_ELEMENT_TYPE_MAC:
+ pf->mac_seid = seid;
+ break;
+ case I40E_SWITCH_ELEMENT_TYPE_VEB:
+ /* Main VEB? */
+ if (uplink_seid != pf->mac_seid)
+ break;
+ if (pf->lan_veb == I40E_NO_VEB) {
+ int v;
+
+ /* find existing or else empty VEB */
+ for (v = 0; v < I40E_MAX_VEB; v++) {
+ if (pf->veb[v] && (pf->veb[v]->seid == seid)) {
+ pf->lan_veb = v;
+ break;
+ }
+ }
+ if (pf->lan_veb == I40E_NO_VEB) {
+ v = i40e_veb_mem_alloc(pf);
+ if (v < 0)
+ break;
+ pf->lan_veb = v;
+ }
+ }
+
+ pf->veb[pf->lan_veb]->seid = seid;
+ pf->veb[pf->lan_veb]->uplink_seid = pf->mac_seid;
+ pf->veb[pf->lan_veb]->pf = pf;
+ pf->veb[pf->lan_veb]->veb_idx = I40E_NO_VEB;
+ break;
+ case I40E_SWITCH_ELEMENT_TYPE_VSI:
+ if (num_reported != 1)
+ break;
+ /* This is immediately after a reset so we can assume this is
+ * the PF's VSI
+ */
+ pf->mac_seid = uplink_seid;
+ pf->pf_seid = downlink_seid;
+ pf->main_vsi_seid = seid;
+ if (printconfig)
+ dev_info(&pf->pdev->dev,
+ "pf_seid=%d main_vsi_seid=%d\n",
+ pf->pf_seid, pf->main_vsi_seid);
+ break;
+ case I40E_SWITCH_ELEMENT_TYPE_PF:
+ case I40E_SWITCH_ELEMENT_TYPE_VF:
+ case I40E_SWITCH_ELEMENT_TYPE_EMP:
+ case I40E_SWITCH_ELEMENT_TYPE_BMC:
+ case I40E_SWITCH_ELEMENT_TYPE_PE:
+ case I40E_SWITCH_ELEMENT_TYPE_PA:
+ /* ignore these for now */
+ break;
+ default:
+ dev_info(&pf->pdev->dev, "unknown element type=%d seid=%d\n",
+ element_type, seid);
+ break;
+ }
+}
+
+/**
+ * i40e_fetch_switch_configuration - Get switch config from firmware
+ * @pf: board private structure
+ * @printconfig: should we print the contents
+ *
+ * Get the current switch configuration from the device and
+ * extract a few useful SEID values.
+ **/
+s32 i40e_fetch_switch_configuration(struct i40e_pf *pf, bool printconfig)
+{
+ struct i40e_aqc_get_switch_config_resp *sw_config;
+ int ret = I40E_SUCCESS;
+ u16 next_seid = 0;
+ u8 *aq_buf;
+ int i;
+
+ if (!pf)
+ return I40E_ERR_BAD_PTR;
+
+ aq_buf = kzalloc(I40E_AQ_LARGE_BUF, GFP_KERNEL);
+ if (!aq_buf)
+ return -ENOMEM;
+
+ sw_config = (struct i40e_aqc_get_switch_config_resp *)aq_buf;
+ do {
+ u16 num_reported, num_total;
+
+ ret = i40e_aq_get_switch_config(&pf->hw, sw_config,
+ I40E_AQ_LARGE_BUF,
+ &next_seid, NULL);
+ if (ret) {
+ dev_info(&pf->pdev->dev,
+ "get switch config failed %d aq_err=%x\n",
+ ret, pf->hw.aq.asq_last_status);
+ kfree(aq_buf);
+ return ret;
+ }
+
+ num_reported = le16_to_cpu(sw_config->header.num_reported);
+ num_total = le16_to_cpu(sw_config->header.num_total);
+
+ if (printconfig)
+ dev_info(&pf->pdev->dev,
+ "header: %d reported %d total\n",
+ num_reported, num_total);
+
+ if (num_reported) {
+ int sz = sizeof(*sw_config) * num_reported;
+
+ kfree(pf->sw_config);
+ pf->sw_config = kzalloc(sz, GFP_KERNEL);
+ if (pf->sw_config)
+ memcpy(pf->sw_config, sw_config, sz);
+ }
+
+ for (i = 0; i < num_reported; i++) {
+ struct i40e_aqc_switch_config_element_resp *ele =
+ &sw_config->element[i];
+
+ i40e_setup_pf_switch_element(pf, ele, num_reported,
+ printconfig);
+ }
+ } while (next_seid != 0);
+
+ kfree(aq_buf);
+ return ret;
+}
+
+/**
+ * i40e_setup_pf_switch - Setup the HW switch on startup or after reset
+ * @pf: board private structure
+ *
+ * Returns 0 on success, negative value on failure
+ **/
+static s32 i40e_setup_pf_switch(struct i40e_pf *pf)
+{
+ int ret;
+
+ /* find out what's out there already */
+ ret = i40e_fetch_switch_configuration(pf, false);
+ if (ret) {
+ dev_info(&pf->pdev->dev,
+ "couldn't fetch switch config, err %d, aq_err %d\n",
+ ret, pf->hw.aq.asq_last_status);
+ return ret;
+ }
+ i40e_pf_reset_stats(pf);
+
+ /* fdir VSI must happen first to be sure it gets queue 0, but only
+ * if there is enough room for the fdir VSI
+ */
+ if (pf->num_lan_qps > 1)
+ i40e_fdir_setup(pf);
+
+ /* first time setup */
+ if (pf->lan_vsi == I40E_NO_VSI) {
+ struct i40e_vsi *vsi = NULL;
+ u16 uplink_seid;
+
+ /* Set up the PF VSI associated with the PF's main VSI
+ * that is already in the HW switch
+ */
+ if (pf->lan_veb != I40E_NO_VEB && pf->veb[pf->lan_veb])
+ uplink_seid = pf->veb[pf->lan_veb]->seid;
+ else
+ uplink_seid = pf->mac_seid;
+
+ vsi = i40e_vsi_setup(pf, I40E_VSI_MAIN, uplink_seid, 0);
+ if (!vsi) {
+ dev_info(&pf->pdev->dev, "setup of MAIN VSI failed\n");
+ i40e_fdir_teardown(pf);
+ return I40E_ERR_NOT_READY;
+ }
+ /* accommodate kcompat by copying the main VSI queue count
+ * into the pf, since this newer code pushes the pf queue
+ * info down a level into a VSI
+ */
+ pf->num_rx_queues = vsi->alloc_queue_pairs;
+ pf->num_tx_queues = vsi->alloc_queue_pairs;
+ } else {
+ /* force a reset of TC and queue layout configurations */
+ u8 enabled_tc = pf->vsi[pf->lan_vsi]->tc_config.enabled_tc;
+ pf->vsi[pf->lan_vsi]->tc_config.enabled_tc = 0;
+ pf->vsi[pf->lan_vsi]->seid = pf->main_vsi_seid;
+ i40e_vsi_config_tc(pf->vsi[pf->lan_vsi], enabled_tc);
+ }
+ i40e_vlan_stripping_disable(pf->vsi[pf->lan_vsi]);
+
+ /* Setup static PF queue filter control settings */
+ ret = i40e_setup_pf_filter_control(pf);
+ if (ret) {
+ dev_info(&pf->pdev->dev, "setup_pf_filter_control failed: %d\n",
+ ret);
+ /* Failure here should not stop continuing other steps */
+ }
+
+ /* enable RSS in the HW, even for only one queue, as the stack can use
+ * the hash
+ */
+ if ((pf->flags & I40E_FLAG_RSS_ENABLED))
+ i40e_config_rss(pf);
+
+ /* fill in link information and enable LSE reporting */
+ i40e_aq_get_link_info(&pf->hw, true, NULL, NULL);
+ i40e_link_event(pf);
+
+ /* Initialize user-specifics link properties */
+ pf->fc_autoneg_status = ((pf->hw.phy.link_info.an_info &
+ I40E_AQ_AN_COMPLETED) ? true : false);
+ pf->hw.fc.requested_mode = I40E_FC_DEFAULT;
+ if (pf->hw.phy.link_info.an_info &
+ (I40E_AQ_LINK_PAUSE_TX | I40E_AQ_LINK_PAUSE_RX))
+ pf->hw.fc.current_mode = I40E_FC_FULL;
+ else if (pf->hw.phy.link_info.an_info & I40E_AQ_LINK_PAUSE_TX)
+ pf->hw.fc.current_mode = I40E_FC_TX_PAUSE;
+ else if (pf->hw.phy.link_info.an_info & I40E_AQ_LINK_PAUSE_RX)
+ pf->hw.fc.current_mode = I40E_FC_RX_PAUSE;
+ else
+ pf->hw.fc.current_mode = I40E_FC_DEFAULT;
+
+ return ret;
+}
+
+/**
+ * i40e_determine_queue_usage - Work out queue distribution
+ * @pf: board private structure
+ **/
+static i40e_status i40e_determine_queue_usage(struct i40e_pf *pf)
+{
+ int accum_tc_size;
+ int queues_left;
+ int num_tc0;
+
+ pf->num_lan_qps = 0;
+ pf->num_tc_qps = rounddown_pow_of_two(pf->num_tc_qps);
+ accum_tc_size = (I40E_MAX_TRAFFIC_CLASS - 1) * pf->num_tc_qps;
+
+ /* a quicky macro for a repeated set of lines */
+#define SET_RSS_SIZE do { \
+num_tc0 = min_t(int, queues_left, pf->rss_size_max); \
+num_tc0 = min_t(int, num_tc0, nr_cpus_node(numa_node_id())); \
+num_tc0 = rounddown_pow_of_two(num_tc0); \
+pf->rss_size = num_tc0; \
+} while (0)
+
+ /* Find the max queues to be put into basic use. We'll always be
+ * using TC0, whether or not DCB is running, and TC0 will get the
+ * big RSS set.
+ */
+ queues_left = pf->hw.func_caps.num_tx_qp;
+
+ if (!((pf->flags & I40E_FLAG_MSIX_ENABLED) &&
+ (pf->flags & I40E_FLAG_MQ_ENABLED)) ||
+ !(pf->flags & (I40E_FLAG_RSS_ENABLED |
+ I40E_FLAG_FDIR_ENABLED | I40E_FLAG_DCB_ENABLED)) ||
+ (queues_left == 1)) {
+
+ /* one qp for PF, no queues for anything else */
+ queues_left = 0;
+ pf->rss_size = pf->num_lan_qps = 1;
+
+ /* make sure all the fancies are disabled */
+ pf->flags &= ~(I40E_FLAG_RSS_ENABLED |
+ I40E_FLAG_MQ_ENABLED |
+ I40E_FLAG_FDIR_ENABLED |
+ I40E_FLAG_FDIR_ATR_ENABLED |
+ I40E_FLAG_DCB_ENABLED |
+ I40E_FLAG_SRIOV_ENABLED |
+ I40E_FLAG_VMDQ_ENABLED);
+
+ } else if (pf->flags & I40E_FLAG_RSS_ENABLED &&
+ !(pf->flags & I40E_FLAG_FDIR_ENABLED) &&
+ !(pf->flags & I40E_FLAG_DCB_ENABLED)) {
+
+ SET_RSS_SIZE;
+
+ queues_left -= pf->rss_size;
+ pf->num_lan_qps = pf->rss_size;
+
+ } else if (pf->flags & I40E_FLAG_RSS_ENABLED &&
+ !(pf->flags & I40E_FLAG_FDIR_ENABLED) &&
+ (pf->flags & I40E_FLAG_DCB_ENABLED)) {
+
+ /* save num_tc_qps queues for TCs 1 thru 7 and the rest
+ * are set up for RSS in TC0
+ */
+ queues_left -= accum_tc_size;
+
+ SET_RSS_SIZE;
+
+ queues_left -= pf->rss_size;
+ if (queues_left < 0) {
+ dev_info(&pf->pdev->dev, "not enough queues for DCB\n");
+ return I40E_ERR_CONFIG;
+ }
+
+ pf->num_lan_qps = pf->rss_size + accum_tc_size;
+
+ } else if (pf->flags & I40E_FLAG_RSS_ENABLED &&
+ (pf->flags & I40E_FLAG_FDIR_ENABLED) &&
+ !(pf->flags & I40E_FLAG_DCB_ENABLED)) {
+
+ queues_left -= 1; /* save 1 queue for FD */
+
+ SET_RSS_SIZE;
+
+ queues_left -= pf->rss_size;
+ if (queues_left < 0) {
+ dev_info(&pf->pdev->dev, "not enough queues for Flow Director\n");
+ return I40E_ERR_CONFIG;
+ }
+
+ pf->num_lan_qps = pf->rss_size;
+
+ } else if (pf->flags & I40E_FLAG_RSS_ENABLED &&
+ (pf->flags & I40E_FLAG_FDIR_ENABLED) &&
+ (pf->flags & I40E_FLAG_DCB_ENABLED)) {
+
+ /* save 1 queue for TCs 1 thru 7,
+ * 1 queue for flow director,
+ * and the rest are set up for RSS in TC0
+ */
+ queues_left -= 1;
+ queues_left -= accum_tc_size;
+
+ SET_RSS_SIZE;
+ queues_left -= pf->rss_size;
+ if (queues_left < 0) {
+ dev_info(&pf->pdev->dev, "not enough queues for DCB and Flow Director\n");
+ return I40E_ERR_CONFIG;
+ }
+
+ pf->num_lan_qps = pf->rss_size + accum_tc_size;
+
+ } else {
+ dev_info(&pf->pdev->dev,
+ "Invalid configuration, flags=0x%08llx\n", pf->flags);
+ return I40E_ERR_CONFIG;
+ }
+
+ if ((pf->flags & I40E_FLAG_SRIOV_ENABLED) &&
+ pf->num_vf_qps && pf->num_req_vfs && queues_left) {
+ pf->num_req_vfs = min_t(int, pf->num_req_vfs, (queues_left /
+ pf->num_vf_qps));
+ queues_left -= (pf->num_req_vfs * pf->num_vf_qps);
+ }
+
+ if ((pf->flags & I40E_FLAG_VMDQ_ENABLED) &&
+ pf->num_vmdq_vsis && pf->num_vmdq_qps && queues_left) {
+ pf->num_vmdq_vsis = min_t(int, pf->num_vmdq_vsis,
+ (queues_left / pf->num_vmdq_qps));
+ queues_left -= (pf->num_vmdq_vsis * pf->num_vmdq_qps);
+ }
+
+ return I40E_SUCCESS;
+}
+
+/**
+ * i40e_setup_pf_filter_control - Setup PF static filter control
+ * @pf: PF to be setup
+ *
+ * i40e_setup_pf_filter_control sets up a pf's initial filter control
+ * settings. If PE/FCoE are enabled then it will also set the per PF
+ * based filter sizes required for them. It also enables Flow director,
+ * ethertype and macvlan type filter settings for the pf.
+ *
+ * Returns 0 on success, negative on failure
+ **/
+static int i40e_setup_pf_filter_control(struct i40e_pf *pf)
+{
+ struct i40e_filter_control_settings *settings = &pf->filter_settings;
+
+ settings->hash_lut_size = I40E_HASH_LUT_SIZE_128;
+
+ /* Flow Director is enabled */
+ if (pf->flags & (I40E_FLAG_FDIR_ENABLED | I40E_FLAG_FDIR_ATR_ENABLED))
+ settings->enable_fdir = true;
+
+ /* Ethtype and MACVLAN filters enabled for PF */
+ settings->enable_ethtype = true;
+ settings->enable_macvlan = true;
+
+ return i40e_set_filter_control(&pf->hw, settings);
+}
+
+/**
+ * i40e_probe - Device initialization routine
+ * @pdev: PCI device information struct
+ * @ent: entry in i40e_pci_tbl
+ *
+ * i40e_probe initializes a pf identified by a pci_dev structure.
+ * The OS initialization, configuring of the pf private structure,
+ * and a hardware reset occur.
+ *
+ * Returns 0 on success, negative on failure
+ **/
+static int i40e_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
+{
+ struct i40e_driver_version dv;
+ struct i40e_pf *pf;
+ struct i40e_hw *hw;
+ int err = 0;
+ u32 len;
+
+ err = pci_enable_device_mem(pdev);
+ if (err)
+ return err;
+
+ /* set up for high or low dma */
+ if (!dma_set_mask(&pdev->dev, DMA_BIT_MASK(64))) {
+ /* coherent mask for the same size will always succeed if
+ * dma_set_mask does
+ */
+ dma_set_coherent_mask(&pdev->dev, DMA_BIT_MASK(64));
+ } else if (!dma_set_mask(&pdev->dev, DMA_BIT_MASK(32))) {
+ dma_set_coherent_mask(&pdev->dev, DMA_BIT_MASK(32));
+ } else {
+ dev_err(&pdev->dev, "DMA configuration failed: %d\n", err);
+ err = -EIO;
+ goto err_dma;
+ }
+
+ /* set up pci connections */
+ err = pci_request_selected_regions(pdev, pci_select_bars(pdev,
+ IORESOURCE_MEM), i40e_driver_name);
+ if (err) {
+ dev_info(&pdev->dev,
+ "pci_request_selected_regions failed %d\n", err);
+ goto err_pci_reg;
+ }
+
+ pci_enable_pcie_error_reporting(pdev);
+ pci_set_master(pdev);
+
+ /* Now that we have a PCI connection, we need to do the
+ * low level device setup. This is primarily setting up
+ * the Admin Queue structures and then querying for the
+ * device's current profile information.
+ */
+ pf = kzalloc(sizeof(*pf), GFP_KERNEL);
+ if (!pf) {
+ err = -ENOMEM;
+ goto err_pf_alloc;
+ }
+ pf->next_vsi = 0;
+ pf->pdev = pdev;
+ set_bit(__I40E_DOWN, &pf->state);
+
+ hw = &pf->hw;
+ hw->back = pf;
+ hw->hw_addr = ioremap(pci_resource_start(pdev, 0),
+ pci_resource_len(pdev, 0));
+ if (!hw->hw_addr) {
+ err = -EIO;
+ dev_info(&pdev->dev, "ioremap(0x%04x, 0x%04x) failed: 0x%x\n",
+ (unsigned int)pci_resource_start(pdev, 0),
+ (unsigned int)pci_resource_len(pdev, 0), err);
+ goto err_ioremap;
+ }
+ hw->vendor_id = pdev->vendor;
+ hw->device_id = pdev->device;
+ pci_read_config_byte(pdev, PCI_REVISION_ID, &hw->revision_id);
+ hw->subsystem_vendor_id = pdev->subsystem_vendor;
+ hw->subsystem_device_id = pdev->subsystem_device;
+ hw->bus.device = PCI_SLOT(pdev->devfn);
+ hw->bus.func = PCI_FUNC(pdev->devfn);
+
+ /* Reset here to make sure all is clean and to define PF 'n' */
+ err = i40e_pf_reset(hw);
+ if (err) {
+ dev_info(&pdev->dev, "Initial pf_reset failed: %d\n", err);
+ goto err_pf_reset;
+ }
+ pf->pfr_count++;
+
+ hw->aq.num_arq_entries = I40E_AQ_LEN;
+ hw->aq.num_asq_entries = I40E_AQ_LEN;
+ hw->aq.arq_buf_size = I40E_MAX_AQ_BUF_SIZE;
+ hw->aq.asq_buf_size = I40E_MAX_AQ_BUF_SIZE;
+ pf->adminq_work_limit = I40E_AQ_WORK_LIMIT;
+ snprintf(pf->misc_int_name, sizeof(pf->misc_int_name) - 1,
+ "%s-pf%d:misc",
+ dev_driver_string(&pf->pdev->dev), pf->hw.pf_id);
+
+ err = i40e_init_shared_code(hw);
+ if (err) {
+ dev_info(&pdev->dev, "init_shared_code failed: %d\n", err);
+ goto err_pf_reset;
+ }
+
+ err = i40e_init_adminq(hw);
+ dev_info(&pdev->dev, "%s\n", i40e_fw_version_str(hw));
+ if (err) {
+ dev_info(&pdev->dev,
+ "init_adminq failed: %d expecting API %02x.%02x\n",
+ err,
+ I40E_FW_API_VERSION_MAJOR, I40E_FW_API_VERSION_MINOR);
+ goto err_pf_reset;
+ }
+
+ err = i40e_get_capabilities(pf);
+ if (err)
+ goto err_adminq_setup;
+
+ err = i40e_sw_init(pf);
+ if (err) {
+ dev_info(&pdev->dev, "sw_init failed: %d\n", err);
+ goto err_sw_init;
+ }
+
+ err = i40e_init_lan_hmc(hw, hw->func_caps.num_tx_qp,
+ hw->func_caps.num_rx_qp,
+ pf->fcoe_hmc_cntx_num, pf->fcoe_hmc_filt_num);
+ if (err) {
+ dev_info(&pdev->dev, "init_lan_hmc failed: %d\n", err);
+ goto err_init_lan_hmc;
+ }
+
+ err = i40e_configure_lan_hmc(hw, I40E_HMC_MODEL_DIRECT_ONLY);
+ if (err) {
+ dev_info(&pdev->dev, "configure_lan_hmc failed: %d\n", err);
+ goto err_configure_lan_hmc;
+ }
+
+ i40e_get_mac_addr(hw, hw->mac.addr);
+ if (i40e_validate_mac_addr(hw->mac.addr)) {
+ dev_info(&pdev->dev, "invalid MAC address %pM\n", hw->mac.addr);
+ err = -EIO;
+ goto err_mac_addr;
+ }
+ dev_info(&pdev->dev, "MAC address: %pM\n", hw->mac.addr);
+ memcpy(hw->mac.perm_addr, hw->mac.addr, ETH_ALEN);
+
+ pci_set_drvdata(pdev, pf);
+ pci_save_state(pdev);
+
+ /* set up periodic task facility */
+ setup_timer(&pf->service_timer, i40e_service_timer, (unsigned long)pf);
+ pf->service_timer_period = HZ;
+
+ INIT_WORK(&pf->service_task, i40e_service_task);
+ clear_bit(__I40E_SERVICE_SCHED, &pf->state);
+ pf->flags |= I40E_FLAG_NEED_LINK_UPDATE;
+ pf->link_check_timeout = jiffies;
+
+ /* set up the main switch operations */
+ i40e_determine_queue_usage(pf);
+ i40e_init_interrupt_scheme(pf);
+
+ /* Set up the *vsi struct based on the number of VSIs in the HW,
+ * and set up our local tracking of the MAIN PF vsi.
+ */
+ len = sizeof(struct i40e_vsi *) * pf->hw.func_caps.num_vsis;
+ pf->vsi = kzalloc(len, GFP_KERNEL);
+ if (!pf->vsi)
+ goto err_switch_setup;
+
+ err = i40e_setup_pf_switch(pf);
+ if (err) {
+ dev_info(&pdev->dev, "setup_pf_switch failed: %d\n", err);
+ goto err_vsis;
+ }
+
+ /* The main driver is (mostly) up and happy. We need to set this state
+ * before setting up the misc vector or we get a race and the vector
+ * ends up disabled forever.
+ */
+ clear_bit(__I40E_DOWN, &pf->state);
+
+ /* In case of MSIX we are going to setup the misc vector right here
+ * to handle admin queue events etc. In case of legacy and MSI
+ * the misc functionality and queue processing is combined in
+ * the same vector and that gets setup at open.
+ */
+ if (pf->flags & I40E_FLAG_MSIX_ENABLED) {
+ err = i40e_setup_misc_vector(pf);
+ if (err) {
+ dev_info(&pdev->dev,
+ "setup of misc vector failed: %d\n", err);
+ goto err_vsis;
+ }
+ }
+
+ /* prep for VF support */
+ if ((pf->flags & I40E_FLAG_SRIOV_ENABLED) &&
+ (pf->flags & I40E_FLAG_MSIX_ENABLED)) {
+ u32 val;
+
+ /* disable link interrupts for VFs */
+ val = rd32(hw, I40E_PFGEN_PORTMDIO_NUM);
+ val &= ~I40E_PFGEN_PORTMDIO_NUM_VFLINK_STAT_ENA_MASK;
+ wr32(hw, I40E_PFGEN_PORTMDIO_NUM, val);
+ i40e_flush(hw);
+ }
+
+ i40e_sys_init(pf);
+ i40e_dbg_pf_init(pf);
+
+ /* tell the firmware that we're starting */
+ dv.major_version = DRV_VERSION_MAJOR;
+ dv.minor_version = DRV_VERSION_MINOR;
+ dv.build_version = DRV_VERSION_BUILD;
+ dv.subbuild_version = 0;
+ i40e_aq_send_driver_version(&pf->hw, &dv, NULL);
+
+ /* since everything's happy, start the service_task timer */
+ mod_timer(&pf->service_timer, jiffies + pf->service_timer_period);
+
+ return 0;
+
+ /* Unwind what we've done if something failed in the setup */
+err_vsis:
+ set_bit(__I40E_DOWN, &pf->state);
+err_switch_setup:
+ i40e_clear_interrupt_scheme(pf);
+ kfree(pf->vsi);
+ del_timer_sync(&pf->service_timer);
+err_mac_addr:
+err_configure_lan_hmc:
+ (void)i40e_shutdown_lan_hmc(hw);
+err_init_lan_hmc:
+ kfree(pf->qp_pile);
+ kfree(pf->irq_pile);
+err_sw_init:
+err_adminq_setup:
+ (void)i40e_shutdown_adminq(hw);
+err_pf_reset:
+ iounmap(hw->hw_addr);
+err_ioremap:
+ kfree(pf);
+err_pf_alloc:
+ pci_disable_pcie_error_reporting(pdev);
+ pci_release_selected_regions(pdev,
+ pci_select_bars(pdev, IORESOURCE_MEM));
+err_pci_reg:
+err_dma:
+ pci_disable_device(pdev);
+ return err;
+}
+
+/**
+ * i40e_remove - Device removal routine
+ * @pdev: PCI device information struct
+ *
+ * i40e_remove is called by the PCI subsystem to alert the driver
+ * that is should release a PCI device. This could be caused by a
+ * Hot-Plug event, or because the driver is going to be removed from
+ * memory.
+ **/
+static void i40e_remove(struct pci_dev *pdev)
+{
+ struct i40e_pf *pf = pci_get_drvdata(pdev);
+ i40e_status ret_code;
+ u32 reg;
+ int i;
+
+ i40e_dbg_pf_exit(pf);
+
+
+ if (pf->flags & I40E_FLAG_SRIOV_ENABLED) {
+ ret_code = i40e_free_vfs(pf);
+ if (ret_code != I40E_SUCCESS)
+ dev_warn(&pdev->dev, "Failed to free vfs: %d\n",
+ ret_code);
+ pf->flags &= ~I40E_FLAG_SRIOV_ENABLED;
+ }
+
+ /* no more scheduling of any task */
+ set_bit(__I40E_DOWN, &pf->state);
+ del_timer_sync(&pf->service_timer);
+ cancel_work_sync(&pf->service_task);
+
+ i40e_fdir_teardown(pf);
+
+ /* If there is a switch structure or any orphans, remove them.
+ * This will leave only the PF's VSI remaining.
+ */
+ for (i = 0; i < I40E_MAX_VEB; i++) {
+ if (!pf->veb[i])
+ continue;
+
+ if (pf->veb[i]->uplink_seid == pf->mac_seid ||
+ pf->veb[i]->uplink_seid == 0)
+ i40e_switch_branch_release(pf->veb[i]);
+ }
+
+ /* Now we can shutdown the PF's VSI, just before we kill
+ * adminq and hmc.
+ */
+ i40e_vsi_release(pf->vsi[pf->lan_vsi]);
+
+ i40e_stop_misc_vector(pf);
+ if (pf->flags & I40E_FLAG_MSIX_ENABLED) {
+ synchronize_irq(pf->msix_entries[0].vector);
+ free_irq(pf->msix_entries[0].vector, pf);
+ }
+
+ /* shutdown and destroy the HMC */
+ ret_code = i40e_shutdown_lan_hmc(&pf->hw);
+ if (ret_code != I40E_SUCCESS)
+ dev_warn(&pdev->dev,
+ "Failed to destroy the HMC resources: %d\n", ret_code);
+
+ /* shutdown the adminq */
+ i40e_aq_queue_shutdown(&pf->hw, true);
+ ret_code = i40e_shutdown_adminq(&pf->hw);
+ if (ret_code != I40E_SUCCESS)
+ dev_warn(&pdev->dev,
+ "Failed to destroy the Admin Queue resources: %d\n",
+ ret_code);
+
+ /* Clear all dynamic memory lists of rings, q_vectors, and VSIs */
+ i40e_clear_interrupt_scheme(pf);
+ for (i = 0; i < pf->hw.func_caps.num_vsis; i++) {
+ if (pf->vsi[i]) {
+ i40e_sys_del_vsi(pf->vsi[i]);
+ i40e_vsi_clear_rings(pf->vsi[i]);
+ i40e_vsi_clear(pf->vsi[i]);
+ pf->vsi[i] = NULL;
+ }
+ }
+
+ i40e_sys_exit(pf);
+
+ for (i = 0; i < I40E_MAX_VEB; i++) {
+ kfree(pf->veb[i]);
+ pf->veb[i] = NULL;
+ }
+
+ kfree(pf->qp_pile);
+ kfree(pf->irq_pile);
+ kfree(pf->sw_config);
+ kfree(pf->vsi);
+
+ /* force a PF reset to clean anything leftover */
+ reg = rd32(&pf->hw, I40E_PFGEN_CTRL);
+ wr32(&pf->hw, I40E_PFGEN_CTRL, (reg | I40E_PFGEN_CTRL_PFSWR_MASK));
+ i40e_flush(&pf->hw);
+
+ iounmap(pf->hw.hw_addr);
+ kfree(pf);
+ pci_release_selected_regions(pdev,
+ pci_select_bars(pdev, IORESOURCE_MEM));
+
+ pci_disable_pcie_error_reporting(pdev);
+ pci_disable_device(pdev);
+}
+
+/**
+ * i40e_pci_error_detected - warning that something funky happened in PCI land
+ * @pdev: PCI device information struct
+ *
+ * Called to warn that something happened and the error handling steps
+ * are in progress. Allows the driver to quiesce things, be ready for
+ * remediation.
+ **/
+static pci_ers_result_t i40e_pci_error_detected(struct pci_dev *pdev,
+ enum pci_channel_state error)
+{
+ struct i40e_pf *pf = pci_get_drvdata(pdev);
+
+ dev_info(&pdev->dev, "%s: error %d\n", __func__, error);
+
+ /* shutdown all operations */
+ i40e_pf_quiesce_all_vsi(pf);
+
+ /* Request a slot reset */
+ return PCI_ERS_RESULT_NEED_RESET;
+}
+
+/**
+ * i40e_pci_error_slot_reset - a PCI slot reset just happened
+ * @pdev: PCI device information struct
+ *
+ * Called to find if the driver can work with the device now that
+ * the pci slot has been reset. If a basic connection seems good
+ * (registers are readable and have sane content) then return a
+ * happy little PCI_ERS_RESULT_xxx.
+ **/
+static pci_ers_result_t i40e_pci_error_slot_reset(struct pci_dev *pdev)
+{
+ struct i40e_pf *pf = pci_get_drvdata(pdev);
+ pci_ers_result_t result;
+ int err;
+ u32 reg;
+
+ dev_info(&pdev->dev, "%s\n", __func__);
+ if (pci_enable_device_mem(pdev)) {
+ dev_info(&pdev->dev,
+ "Cannot re-enable PCI device after reset.\n");
+ result = PCI_ERS_RESULT_DISCONNECT;
+ } else {
+ pci_set_master(pdev);
+ pci_restore_state(pdev);
+ pci_save_state(pdev);
+ pci_wake_from_d3(pdev, false);
+
+ reg = rd32(&pf->hw, I40E_GLGEN_RTRIG);
+ if (reg == 0)
+ result = PCI_ERS_RESULT_RECOVERED;
+ else
+ result = PCI_ERS_RESULT_DISCONNECT;
+ }
+
+ err = pci_cleanup_aer_uncorrect_error_status(pdev);
+ if (err) {
+ dev_info(&pdev->dev,
+ "pci_cleanup_aer_uncorrect_error_status failed 0x%0x\n",
+ err);
+ /* non-fatal, continue */
+ }
+
+ return result;
+}
+
+/**
+ * i40e_pci_error_resume - restart operations after PCI error recovery
+ * @pdev: PCI device information struct
+ *
+ * Called to allow the driver to bring things back up after PCI error
+ * and/or reset recovery has finished.
+ **/
+static void i40e_pci_error_resume(struct pci_dev *pdev)
+{
+ struct i40e_pf *pf = pci_get_drvdata(pdev);
+
+ dev_info(&pdev->dev, "%s\n", __func__);
+ i40e_handle_reset_warning(pf);
+}
+
+static const struct pci_error_handlers i40e_err_handler = {
+ .error_detected = i40e_pci_error_detected,
+ .slot_reset = i40e_pci_error_slot_reset,
+ .resume = i40e_pci_error_resume,
+};
+
+static struct pci_driver i40e_driver = {
+ .name = i40e_driver_name,
+ .id_table = i40e_pci_tbl,
+ .probe = i40e_probe,
+ .remove = i40e_remove,
+ .err_handler = &i40e_err_handler,
+ .sriov_configure = i40e_pci_sriov_configure,
+};
+
+/**
+ * i40e_init_module - Driver registration routine
+ *
+ * i40e_init_module is the first routine called when the driver is
+ * loaded. All it does is register with the PCI subsystem.
+ **/
+static int __init i40e_init_module(void)
+{
+ int ret;
+ pr_info("%s: %s - version %s\n", i40e_driver_name,
+ i40e_driver_string, i40e_driver_version_str);
+ pr_info("%s: %s\n", i40e_driver_name, i40e_copyright);
+
+ i40e_dbg_init();
+ ret = pci_register_driver(&i40e_driver);
+ return ret;
+}
+module_init(i40e_init_module);
+
+/**
+ * i40e_exit_module - Driver exit cleanup routine
+ *
+ * i40e_exit_module is called just before the driver is removed
+ * from memory.
+ **/
+static void __exit i40e_exit_module(void)
+{
+ pci_unregister_driver(&i40e_driver);
+ i40e_dbg_exit();
+}
+module_exit(i40e_exit_module);
--
1.8.3.1
------------------------------------------------------------------------------
Learn the latest--Visual Studio 2012, SharePoint 2013, SQL 2012, more!
Discover the easy way to master current and previous Microsoft technologies
and advance your career. Get an incredible 1,500+ hours of step-by-step
tutorial videos with LearnDevNow. Subscribe today and save!
http://pubads.g.doubleclick.net/gampad/clk?id=58041391&iu=/4140/ostg.clktrk
_______________________________________________
E1000-devel mailing list
E1000-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/e1000-devel
To learn more about Intel® Ethernet, visit http://communities.intel.com/community/wired
^ permalink raw reply related
* Re: [PATCH RFC] solos-pci: Fix BUG() with shared skb
From: David Miller @ 2013-09-04 21:51 UTC (permalink / raw)
To: dwmw2; +Cc: netdev, simon, linux-atm-general, nathan
In-Reply-To: <1378327275.2627.28.camel@shinybook.infradead.org>
From: David Woodhouse <dwmw2@infradead.org>
Date: Wed, 04 Sep 2013 21:41:15 +0100
> On Wed, 2013-09-04 at 14:30 -0400, David Miller wrote:
>> skb_realloc_headroom() should do everything you need.
>
> Great, thanks! Something like this then... ?
>
> Do I really need the truesize check? And if so, is there a better way to
> handle the ATM accounting? It just *happens* to be the case that the the
> br2684_pop() and pppoatm_pop() functions don't mind being bypassed in
> this fashion, and we should probably get away with tweaking the core ATM
> accounting directly like this. Doesn't make me happy though...
...
> + nskb = skb_realloc_headroom(skb, sizeof(*header));
> + if (!nskb) {
> + solos_pop(vcc, skb);
> + return -ENOMEM;
> + }
> + if (skb->truesize != nskb->truesize)
> + atm_force_charge(vcc, nskb->truesize - skb->truesize);
My understanding is that truesize will not be changed by calls to
skb_realloc_headroom(), because if it did then every ethernet driver
would be screwed as the skb->truesize adjustment would corrupt socket
memory accounting.
The only thing you need to be mindful is that after the
skb_realloc_headroom() call all skb data pointers change, and
therefore things like packet pointers are now stale and need to be
recalculated.
^ permalink raw reply
* [PATCH] can: pcan_usb_core: fix memory leak on failure paths in peak_usb_start()
From: Alexey Khoroshilov @ 2013-09-04 21:46 UTC (permalink / raw)
To: Marc Kleine-Budde
Cc: Alexey Khoroshilov, Wolfgang Grandegger, linux-can, netdev,
linux-kernel, ldv-project
Tx and rx urbs are not deallocated if something goes wrong in peak_usb_start().
The patch fixes error handling to deallocate all the resources.
Found by Linux Driver Verification project (linuxtesting.org).
Signed-off-by: Alexey Khoroshilov <khoroshilov@ispras.ru>
---
drivers/net/can/usb/peak_usb/pcan_usb_core.c | 15 +++++++++++----
1 file changed, 11 insertions(+), 4 deletions(-)
diff --git a/drivers/net/can/usb/peak_usb/pcan_usb_core.c b/drivers/net/can/usb/peak_usb/pcan_usb_core.c
index a0f647f..0b7a4c3 100644
--- a/drivers/net/can/usb/peak_usb/pcan_usb_core.c
+++ b/drivers/net/can/usb/peak_usb/pcan_usb_core.c
@@ -463,7 +463,7 @@ static int peak_usb_start(struct peak_usb_device *dev)
if (i < PCAN_USB_MAX_TX_URBS) {
if (i == 0) {
netdev_err(netdev, "couldn't setup any tx URB\n");
- return err;
+ goto err_tx;
}
netdev_warn(netdev, "tx performance may be slow\n");
@@ -472,7 +472,7 @@ static int peak_usb_start(struct peak_usb_device *dev)
if (dev->adapter->dev_start) {
err = dev->adapter->dev_start(dev);
if (err)
- goto failed;
+ goto err_adapter;
}
dev->state |= PCAN_USB_STATE_STARTED;
@@ -481,19 +481,26 @@ static int peak_usb_start(struct peak_usb_device *dev)
if (dev->adapter->dev_set_bus) {
err = dev->adapter->dev_set_bus(dev, 1);
if (err)
- goto failed;
+ goto err_adapter;
}
dev->can.state = CAN_STATE_ERROR_ACTIVE;
return 0;
-failed:
+err_adapter:
if (err == -ENODEV)
netif_device_detach(dev->netdev);
netdev_warn(netdev, "couldn't submit control: %d\n", err);
+ for (i = 0; i < PCAN_USB_MAX_TX_URBS; i++) {
+ usb_free_urb(dev->tx_contexts[i].urb);
+ dev->tx_contexts[i].urb = NULL;
+ }
+err_tx:
+ usb_kill_anchored_urbs(&dev->rx_submitted);
+
return err;
}
--
1.8.1.2
^ permalink raw reply related
* [PATCH net] vxlan: Fix kernel panic on device delete.
From: Pravin B Shelar @ 2013-09-04 21:12 UTC (permalink / raw)
To: netdev; +Cc: Pravin B Shelar
On vxlan device create if socket create fails vxlan device is not
added to hash table. Therefore we need to check if device
is in hashtable before we delete it from hlist.
Following patch avoid the crash. net-next already has this fix.
---8<---
BUG: unable to handle kernel NULL pointer dereference at (null)
IP: [<ffffffffa05f9ca7>] vxlan_dellink+0x77/0xf0 [vxlan]
PGD 42b2d9067 PUD 42e04c067 PMD 0
Oops: 0002 [#1] SMP
Modules linked in: vxlan(-)
Hardware name: Dell Inc. PowerEdge R620/0KCKR5, BIOS 1.4.8 10/25/2012
task: ffff88042ecf8760 ti: ffff88042f106000 task.ti: ffff88042f106000
RIP: 0010:[<ffffffffa05f9ca7>] [<ffffffffa05f9ca7>]
vxlan_dellink+0x77/0xf0 [vxlan]
RSP: 0018:ffff88042f107e28 EFLAGS: 00010246
RAX: 0000000000000000 RBX: ffff88082af08000 RCX: ffff88083fd80000
RDX: 0000000000000000 RSI: ffff88042f107e58 RDI: ffff88042e12f810
RBP: ffff88042f107e48 R08: ffffffff8166eca0 R09: 0000000000000000
R10: 0000000000000001 R11: 0000000000000000 R12: ffff88082af087c0
R13: ffff88042e12f000 R14: ffff88042f107e58 R15: ffff88042f107e58
FS: 00007f4ed2de7700(0000) GS:ffff88043fc80000(0000)
knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000000000000000 CR3: 000000042e076000 CR4: 00000000000407e0
Stack:
ffff88082af08000 ffffffff81654848 ffffffffa05fb4e0 ffffffff81654780
ffff88042f107e98 ffffffff813b9c7a ffff88042f107e58 ffff88042f107e58
ffff88042f107e88 ffffffffa05fb4e0 ffffffffa05fb780 ffff88042f107f18
Call Trace:
[<ffffffff813b9c7a>] __rtnl_link_unregister+0xca/0xd0
[<ffffffff813bb0e9>] rtnl_link_unregister+0x19/0x30
[<ffffffffa05faa4c>] vxlan_cleanup_module+0x10/0x2f [vxlan]
[<ffffffff81099fef>] SyS_delete_module+0x1cf/0x2c0
[<ffffffff8146c069>] ? do_page_fault+0x9/0x10
[<ffffffff8146f012>] system_call_fastpath+0x16/0x1b
Code: 4d 85 ed 0f 84 95 00 00 00 4c 8d a7 c0 07 00 00 49 8d bd 10 08 00
00 e8 28 e8 e6 e0 48 8b 83 c0 07 00 00 49 8b 54 24 08 48 85 c0 <48> 89
02 74 04 48 89 50 08 49 b8 00 02 20 00 00 00 ad de 4d 89
RIP [<ffffffffa05f9ca7>] vxlan_dellink+0x77/0xf0 [vxlan]
RSP <ffff88042f107e28>
CR2: 0000000000000000
Signed-off-by: Pravin B Shelar <pshelar@nicira.com>
---
drivers/net/vxlan.c | 3 ++-
1 files changed, 2 insertions(+), 1 deletions(-)
diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index 767f7af..2c15f6c 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -1794,7 +1794,8 @@ static void vxlan_dellink(struct net_device *dev, struct list_head *head)
struct vxlan_dev *vxlan = netdev_priv(dev);
spin_lock(&vn->sock_lock);
- hlist_del_rcu(&vxlan->hlist);
+ if (!hlist_unhashed(&vxlan->hlist))
+ hlist_del_rcu(&vxlan->hlist);
spin_unlock(&vn->sock_lock);
list_del(&vxlan->next);
--
1.7.1
^ permalink raw reply related
* Re: [PATCH] net/core: Order-3 frag allocator causes SWIOTLB bouncing under Xen
From: Konrad Rzeszutek Wilk @ 2013-09-04 21:11 UTC (permalink / raw)
To: Eric Dumazet
Cc: Zoltan Kiss, David S. Miller, Eric Dumazet, Neil Horman, Li Zefan,
Eliezer Tamir, netdev, linux-kernel, malcolm.crossley,
david.vrabel, xen-devel
In-Reply-To: <1378328440.7360.124.camel@edumazet-glaptop>
On Wed, Sep 04, 2013 at 02:00:40PM -0700, Eric Dumazet wrote:
> On Wed, 2013-09-04 at 21:47 +0100, Zoltan Kiss wrote:
> > THIS PATCH IS NOT INTENDED TO BE UPSTREAMED, IT HAS ONLY INFORMING PURPOSES!
> >
> > I've noticed a performance regression with upstream kernels when used as Dom0
> > under Xen. The classic kernel can utilize the whole bandwidth of a 10G NIC
> > (ca. 9.3 Gbps), but upstream can reach only ca. 7 Gbps. I found that it
> > happens because SWIOTLB has to do double buffering. The per task frag
> > allocator introduced in 5640f7 creates 32 kb frags, which are not contiguous
> > in mfn space.
> > This patch provides a workaround by going back to the old way. The possible
> > ideas came up to solve this:
> >
> > * make sure Dom0 memory is contiguous: it sounds trivial, but doesn't work with
> > driver domains, and there are lots of situations where this is not possible.
> > * use PVH Dom0: so we will have IOMMU. In the future sometime.
> > * use IOMMU with PV Dom0: this seems to happen earlier.
> >
> > Signed-off-by: Zoltan Kiss <zoltan.kiss@citrix.com>
> > ---
> > net/core/sock.c | 2 +-
> > 1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/net/core/sock.c b/net/core/sock.c
> > index 2c097c5..854a0ea 100644
> > --- a/net/core/sock.c
> > +++ b/net/core/sock.c
> > @@ -1812,7 +1812,7 @@ struct sk_buff *sock_alloc_send_skb(struct sock *sk, unsigned long size,
> > EXPORT_SYMBOL(sock_alloc_send_skb);
> >
> > /* On 32bit arches, an skb frag is limited to 2^15 */
> > -#define SKB_FRAG_PAGE_ORDER get_order(32768)
> > +#define SKB_FRAG_PAGE_ORDER get_order(4096)
> >
>
> Well, this hack is not new...
>
> We have dev->gso_max_size and dev->gso_max_segs
>
> We also have in net-next sk_pacing_rate and dynamic TSO sizing.
>
> Maybe you could add proper infrastructure to deal with Xen limitations.
I think Ian posted at some point an sysctl patch for that (more for
debugging that anything else). And it kind
of stalled: http://lists.xen.org/archives/html/xen-devel/2012-10/msg01832.html
Is that what you mean by proper infrastructure ?
Oh wait, did you mean via dev and not the whole system wide sysctl?
>
>
>
^ permalink raw reply
* Re: linux-next: Tree for Sep 4 (netfilter: xt_TPROXY)
From: Florian Westphal @ 2013-09-04 21:09 UTC (permalink / raw)
To: Randy Dunlap
Cc: Stephen Rothwell, linux-next, linux-kernel, netfilter-devel,
netdev@vger.kernel.org
In-Reply-To: <52277212.5020006@infradead.org>
Randy Dunlap <rdunlap@infradead.org> wrote:
> On 09/04/13 01:13, Stephen Rothwell wrote:
> > Hi all,
> >
> > Please do not add any code for v3.13 to your linux-next included branches
> > until after v3.12-rc1 is released.
> >
> > Changes since 20130902:
> >
>
> on x86_64:
>
> when CONFIG_IPV6=m
> and CONFIG_NETFILTER_XT_TARGET_TPROXY=y:
>
> net/built-in.o: In function `tproxy_tg6_v1':
> xt_TPROXY.c:(.text+0x5dc05): undefined reference to `udp6_lib_lookup'
> xt_TPROXY.c:(.text+0x5e32f): undefined reference to `udp6_lib_lookup'
> xt_TPROXY.c:(.text+0x5e432): undefined reference to `udp6_lib_lookup'
> net/built-in.o: In function `tproxy_tg_init':
> xt_TPROXY.c:(.init.text+0x1540): undefined reference to `nf_defrag_ipv6_enable'
Thanks for reporting.
I can reproduce same error with 3.10.6 stable tree, so its not
a recent problem.
As always, the tempting solution is to just forbid TPROXY=y with IPV6=m but
it would be better to get rid of the ipv6 link time deps..
^ permalink raw reply
* Re: [PATCH] net/core: Order-3 frag allocator causes SWIOTLB bouncing under Xen
From: Eric Dumazet @ 2013-09-04 21:00 UTC (permalink / raw)
To: Zoltan Kiss
Cc: Konrad Rzeszutek Wilk, David S. Miller, Eric Dumazet, Neil Horman,
Li Zefan, Eliezer Tamir, netdev, linux-kernel, malcolm.crossley,
david.vrabel, xen-devel
In-Reply-To: <1378327638-23956-1-git-send-email-zoltan.kiss@citrix.com>
On Wed, 2013-09-04 at 21:47 +0100, Zoltan Kiss wrote:
> THIS PATCH IS NOT INTENDED TO BE UPSTREAMED, IT HAS ONLY INFORMING PURPOSES!
>
> I've noticed a performance regression with upstream kernels when used as Dom0
> under Xen. The classic kernel can utilize the whole bandwidth of a 10G NIC
> (ca. 9.3 Gbps), but upstream can reach only ca. 7 Gbps. I found that it
> happens because SWIOTLB has to do double buffering. The per task frag
> allocator introduced in 5640f7 creates 32 kb frags, which are not contiguous
> in mfn space.
> This patch provides a workaround by going back to the old way. The possible
> ideas came up to solve this:
>
> * make sure Dom0 memory is contiguous: it sounds trivial, but doesn't work with
> driver domains, and there are lots of situations where this is not possible.
> * use PVH Dom0: so we will have IOMMU. In the future sometime.
> * use IOMMU with PV Dom0: this seems to happen earlier.
>
> Signed-off-by: Zoltan Kiss <zoltan.kiss@citrix.com>
> ---
> net/core/sock.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/net/core/sock.c b/net/core/sock.c
> index 2c097c5..854a0ea 100644
> --- a/net/core/sock.c
> +++ b/net/core/sock.c
> @@ -1812,7 +1812,7 @@ struct sk_buff *sock_alloc_send_skb(struct sock *sk, unsigned long size,
> EXPORT_SYMBOL(sock_alloc_send_skb);
>
> /* On 32bit arches, an skb frag is limited to 2^15 */
> -#define SKB_FRAG_PAGE_ORDER get_order(32768)
> +#define SKB_FRAG_PAGE_ORDER get_order(4096)
>
Well, this hack is not new...
We have dev->gso_max_size and dev->gso_max_segs
We also have in net-next sk_pacing_rate and dynamic TSO sizing.
Maybe you could add proper infrastructure to deal with Xen limitations.
^ permalink raw reply
* [PATCH] net/core: Order-3 frag allocator causes SWIOTLB bouncing under Xen
From: Zoltan Kiss @ 2013-09-04 20:47 UTC (permalink / raw)
To: Konrad Rzeszutek Wilk, David S. Miller, Eric Dumazet, Neil Horman,
Li Zefan, Eliezer Tamir, netdev, linux-kernel, malcolm.crossley,
david.vrabel, xen-devel
Cc: Zoltan Kiss
THIS PATCH IS NOT INTENDED TO BE UPSTREAMED, IT HAS ONLY INFORMING PURPOSES!
I've noticed a performance regression with upstream kernels when used as Dom0
under Xen. The classic kernel can utilize the whole bandwidth of a 10G NIC
(ca. 9.3 Gbps), but upstream can reach only ca. 7 Gbps. I found that it
happens because SWIOTLB has to do double buffering. The per task frag
allocator introduced in 5640f7 creates 32 kb frags, which are not contiguous
in mfn space.
This patch provides a workaround by going back to the old way. The possible
ideas came up to solve this:
* make sure Dom0 memory is contiguous: it sounds trivial, but doesn't work with
driver domains, and there are lots of situations where this is not possible.
* use PVH Dom0: so we will have IOMMU. In the future sometime.
* use IOMMU with PV Dom0: this seems to happen earlier.
Signed-off-by: Zoltan Kiss <zoltan.kiss@citrix.com>
---
net/core/sock.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/net/core/sock.c b/net/core/sock.c
index 2c097c5..854a0ea 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -1812,7 +1812,7 @@ struct sk_buff *sock_alloc_send_skb(struct sock *sk, unsigned long size,
EXPORT_SYMBOL(sock_alloc_send_skb);
/* On 32bit arches, an skb frag is limited to 2^15 */
-#define SKB_FRAG_PAGE_ORDER get_order(32768)
+#define SKB_FRAG_PAGE_ORDER get_order(4096)
bool sk_page_frag_refill(struct sock *sk, struct page_frag *pfrag)
{
^ permalink raw reply related
* Re: [PATCH RFC] solos-pci: Fix BUG() with shared skb
From: David Woodhouse @ 2013-09-04 20:41 UTC (permalink / raw)
To: David Miller; +Cc: netdev, simon, linux-atm-general, nathan
In-Reply-To: <20130904.143056.1558570610941220565.davem@davemloft.net>
[-- Attachment #1: Type: text/plain, Size: 1591 bytes --]
On Wed, 2013-09-04 at 14:30 -0400, David Miller wrote:
> skb_realloc_headroom() should do everything you need.
Great, thanks! Something like this then... ?
Do I really need the truesize check? And if so, is there a better way to
handle the ATM accounting? It just *happens* to be the case that the the
br2684_pop() and pppoatm_pop() functions don't mind being bypassed in
this fashion, and we should probably get away with tweaking the core ATM
accounting directly like this. Doesn't make me happy though...
diff --git a/drivers/atm/solos-pci.c b/drivers/atm/solos-pci.c
index 32784d1..4492c0f 100644
--- a/drivers/atm/solos-pci.c
+++ b/drivers/atm/solos-pci.c
@@ -1145,19 +1145,19 @@ static int psend(struct atm_vcc *vcc, struct sk_buff *skb)
return 0;
}
- if (!skb_clone_writable(skb, sizeof(*header))) {
- int expand_by = 0;
- int ret;
-
- if (skb_headroom(skb) < sizeof(*header))
- expand_by = sizeof(*header) - skb_headroom(skb);
-
- ret = pskb_expand_head(skb, expand_by, 0, GFP_ATOMIC);
- if (ret) {
- dev_warn(&card->dev->dev, "pskb_expand_head failed.\n");
- solos_pop(vcc, skb);
- return ret;
- }
+ if (skb_headroom(skb) < sizeof(*header)) {
+ struct sk_buff *nskb;
+
+ nskb = skb_realloc_headroom(skb, sizeof(*header));
+ if (!nskb) {
+ solos_pop(vcc, skb);
+ return -ENOMEM;
+ }
+ if (skb->truesize != nskb->truesize)
+ atm_force_charge(vcc, nskb->truesize - skb->truesize);
+
+ dev_kfree_skb_any(skb);
+ skb = nskb;
}
header = (void *)skb_push(skb, sizeof(*header));
--
dwmw2
[-- Attachment #2: smime.p7s --]
[-- Type: application/x-pkcs7-signature, Size: 5745 bytes --]
^ permalink raw reply related
* Re: [PATCH 2/2] sh_eth: add device tree support
From: Sergei Shtylyov @ 2013-09-04 19:42 UTC (permalink / raw)
To: Kumar Gala
Cc: netdev, rob.herring, pawel.moll, mark.rutland, swarren,
ian.campbell, grant.likely, devicetree, nobuhiro.iwamatsu.yj,
linux-sh, rob, linux-doc
In-Reply-To: <E4C90BAA-A89F-4C7C-B186-12C2DF7E6F31@codeaurora.org>
Hello.
On 09/04/2013 02:08 AM, Kumar Gala wrote:
>> Add support of the device tree probing for the Renesas SH-Mobile SoCs.
>> This work is loosely based on an original patch by Nobuhiro Iwamatsu
>> <nobuhiro.iwamatsu.yj@renesas.com>.
>> Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
>> ---
>> This patch is against Dave's 'net-next.git' repo.
>> Documentation/devicetree/bindings/net/sh_eth.txt | 40 +++++++++++++
>> drivers/net/ethernet/renesas/sh_eth.c | 66 ++++++++++++++++++++++-
>> 2 files changed, 105 insertions(+), 1 deletion(-)
>> Index: net-next/Documentation/devicetree/bindings/net/sh_eth.txt
>> ===================================================================
>> --- /dev/null
>> +++ net-next/Documentation/devicetree/bindings/net/sh_eth.txt
>> @@ -0,0 +1,40 @@
>> +* Renesas Electronics SH EtherMAC
>> +
>> +This file provides information on what the device node for the SH EtherMAC
>> +interface contains.
>> +
>> +Required properties:
>> +- compatible: "renesas,gether-r8a7740" if the device is a part of R8A7740 SoC.
>> + "renesas,ether-r8a7779" if the device is a part of R8A7778/9 SoCs.
>> + "renesas,ether-r8a7790" if the device is a part of R8A7790/1 SoCs.
> Curious, 7779 (vs 7778) but 7790 (vs 7791).
Support for R8A7779 appeared in Linux before support for R8A7778 (probably
the chips appeared on the market in the same order). The same can be said
about R8A7790 -- it appeared first; R8A7791 isn't yet supported by Linux (it
should be in 3.13). R8A7779/90 are high end chips, R8A7778/91 are middle class
chips. Anyway, by request of Magnus Damm, I'll have to use the individual
"compatible" properties.
>> +- reg: offset and length of the register set for the device; if the device has
>> + TSU registers, you need to specify two register sets here.
>> +- interrupt-parent: the phandle for the interrupt controller that services
>> + interrupts for this device.
>> +- interrupts: interrupt mapping for the interrupt source.
>> +- phy-mode: string, operation mode of the PHY interface (a string that
>> + of_get_phy_mode() can understand).
>> +- phy-handle: phandle of the PHY device.
>> +
> Should add something about requiring a phy subnode
OK.
>> +Optional properties:
>> +- local-mac-address: 6 bytes, MAC address.
> What's the assumption if local-mac-address is not specified?
The driver will try to read the chip's MAC address registers hoping to
find an address left there programmed by U-Boot. If no valid address is found
(usual case on R8A7778/9 development boards which have an extra network chip),
the random one will be used.
>> +- renesas,no-ether-link: specify when a board does not provide a proper LINK
>> + signal.
> How is this different from the fixed link concept we have?
Here the chip-specific LINK signal is meant which is supposed to go from a
PHY to the SoC. Sometimes it's wired from the PHY's link/active LED output or
just absent on the board, so we aren't able to use it for the link detection.
In this case, phylib's notification is used.
>> +- renesas,ether-link-active-low: specify when the LINK signal is active-low.
>> +
>> +Example (Armadillo800EVA board):
>> +
>> + ethernet@e9a00000 {
>> + compatible = "renesas,gether-r8a7740";
>> + reg = <0xe9a00000 0x800>, <0xe9a01800 0x800>;
>> + interrupt-parent = <&gic>;
>> + interrupts = <0 142 0x4>;
>> + phy-mode = "mii";
>> + phy-handle = <&phy0>;
>> + #address-cells = <1>;
>> + #size-cells = <0>;
>> +
>> + phy0: ethernet-phy@0 {
>> + reg = <0>;
>> + };
>> + };
> - k
WBR, Sergei
^ permalink raw reply
* Re: [PATCH] ipv6: fix null pointer dereference in __ip6addrlbl_add
From: David Miller @ 2013-09-04 18:58 UTC (permalink / raw)
To: hannes; +Cc: michele, netdev, yoshfuji, sasha.levin
In-Reply-To: <20130904185145.GC13136@order.stressinduktion.org>
From: Hannes Frederic Sowa <hannes@stressinduktion.org>
Date: Wed, 4 Sep 2013 20:51:45 +0200
> Sorry I forgot to mention that this patch is also applicable to solve
> this issue in the longterm 3.10 kernel. Maybe this should go to stable?
I did queue it up for -stable already.
^ permalink raw reply
* Re: [PATCH net] net: ipv6: tcp: fix potential use after free in tcp_v6_do_rcv
From: David Miller @ 2013-09-04 18:57 UTC (permalink / raw)
To: dborkman; +Cc: netdev, eric.dumazet
In-Reply-To: <1378229352-9779-1-git-send-email-dborkman@redhat.com>
From: Daniel Borkmann <dborkman@redhat.com>
Date: Tue, 3 Sep 2013 19:29:12 +0200
> In tcp_v6_do_rcv() code, when processing pkt options, we soley work
> on our skb clone opt_skb that we've created earlier before entering
> tcp_rcv_established() on our way. However, only in condition ...
>
> if (np->rxopt.bits.rxtclass)
> np->rcv_tclass = ipv6_get_dsfield(ipv6_hdr(skb));
>
> ... we work on skb itself. As we extract every other information out
> of opt_skb in ipv6_pktoptions path, this seems wrong, since skb can
> already be released by tcp_rcv_established() earlier on. When we try
> to access it in ipv6_hdr(), we will dereference freed skb.
>
> Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
> Cc: Eric Dumazet <eric.dumazet@gmail.com>
Applied.
^ permalink raw reply
* Re: [PATCH] sh_eth: fix napi_{en|dis}able() calls racing against interrupts
From: David Miller @ 2013-09-04 18:56 UTC (permalink / raw)
To: sergei.shtylyov; +Cc: netdev, nobuhiro.iwamatsu.yj, linux-sh
In-Reply-To: <201309040241.27834.sergei.shtylyov@cogentembedded.com>
From: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Date: Wed, 4 Sep 2013 02:41:27 +0400
> While implementing NAPI for the driver, I overlooked the race conditions where
> interrupt handler might have called napi_schedule_prep() before napi_enable()
> was called or after napi_disable() was called. If RX interrupt happens, this
> would cause the endless interrupts and messages like:
>
> sh-eth eth0: ignoring interrupt, status 0x00040000, mask 0x01ff009f.
>
> The interrupt wouldn't even be masked by the kernel eventually since the handler
> would return IRQ_HANDLED all the time.
>
> As a fix, move napi_enable() call before request_irq() call and napi_disable()
> call after free_irq() call.
>
> Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Applied and queued up for -stable, thanks!
^ permalink raw reply
* Re: [PATCH net-next v2 0/8] IPv6 MLD updates
From: David Miller @ 2013-09-04 18:54 UTC (permalink / raw)
To: dborkman; +Cc: netdev
In-Reply-To: <1378246784-21067-1-git-send-email-dborkman@redhat.com>
From: Daniel Borkmann <dborkman@redhat.com>
Date: Wed, 4 Sep 2013 00:19:36 +0200
> This set contains the non-RFC version of the previous submission entitled
> "[RFC PATCH net-next 0/7] IPv6 MLD updates". Most importantly it contains
> a fix for MLDv1/v2 switchback timeout where hosts currently are switching
> back from v1 compat mode to normal v2 too early (i.e. switchback time was
> <= 30secs instead of >= 260secs on default), and the set also contains a
> patch that allows for v2-only mode as per RFC recommendation. The rest is
> related to cleanups that make the code more readable resp. maintainable.
Series applied, thanks in particular for the 'net' merge conflict resolution
as that helps me a lot.
^ permalink raw reply
* Re: [PATCH] ipv6: fix null pointer dereference in __ip6addrlbl_add
From: Hannes Frederic Sowa @ 2013-09-04 18:51 UTC (permalink / raw)
To: David Miller; +Cc: michele, netdev, yoshfuji, sasha.levin
In-Reply-To: <20130904.141301.41040920905693536.davem@davemloft.net>
On Wed, Sep 04, 2013 at 02:13:01PM -0400, David Miller wrote:
> From: Hannes Frederic Sowa <hannes@stressinduktion.org>
> Date: Tue, 3 Sep 2013 02:13:31 +0200
>
> > Hi!
> >
> > On Mon, Sep 02, 2013 at 10:31:28PM +0100, Michele Baldessari wrote:
> >> with the latest linux master git tree from Linus
> >> (248d296d6d9df384996c2ed95676b367d876d48c - 2 Sep) I can reproduceably oops
> >> the kernel with the following commands:
> >> ip addrlabel flush
> >> ip addrlabel add prefix ::1/128 label 0
> >> ip addrlabel add prefix ::/0 label 1
> >
> > Thanks for the report! This patch should fix this issue:
> >
> > [PATCH] ipv6: fix null pointer dereference in __ip6addrlbl_add
> >
> > Commit b67bfe0d42cac56c512dd5da4b1b347a23f4b70a ("hlist: drop
> > the node parameter from iterators") changed the behavior of
> > hlist_for_each_entry_safe to leave the p argument NULL.
> >
> > Fix this up by tracking the last argument.
> >
> > Reported-by: Michele Baldessari <michele@acksyn.org>
> > Cc: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>
> > Cc: Sasha Levin <sasha.levin@oracle.com>
> > Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
>
> Applied.
Sorry I forgot to mention that this patch is also applicable to solve
this issue in the longterm 3.10 kernel. Maybe this should go to stable?
Thanks,
Hannes
^ permalink raw reply
* Re: [PATCH] xen-netback: count number required slots for an skb more carefully
From: David Miller @ 2013-09-04 18:48 UTC (permalink / raw)
To: david.vrabel
Cc: xen-devel, konrad.wilk, boris.ostrovsky, ian.campbell, netdev
In-Reply-To: <1378229390-12919-1-git-send-email-david.vrabel@citrix.com>
Can you guys just post the patch you want applied once this thread
moves closer to a consensus? Thanks.
^ permalink raw reply
* Re: [PATCH net-next] tcp: better comments for RTO initiallization
From: David Miller @ 2013-09-04 18:42 UTC (permalink / raw)
To: ycheng; +Cc: ncardwell, edumazet, netdev
In-Reply-To: <1378242875-5172-1-git-send-email-ycheng@google.com>
From: Yuchung Cheng <ycheng@google.com>
Date: Tue, 3 Sep 2013 14:14:35 -0700
> Commit 1b7fdd2ab585("tcp: do not use cached RTT for RTT estimation")
> removes important comments on how RTO is initialized and updated.
> Hopefully this patch puts those information back.
>
> Signed-off-by: Yuchung Cheng <ycheng@google.com>
Applied.
^ permalink raw reply
* Re: [PATCH net-next] vxlan: Optimize vxlan rcv
From: David Miller @ 2013-09-04 18:42 UTC (permalink / raw)
To: pshelar; +Cc: netdev, stephen
In-Reply-To: <1378226684-7751-1-git-send-email-pshelar@nicira.com>
From: Pravin B Shelar <pshelar@nicira.com>
Date: Tue, 3 Sep 2013 09:44:44 -0700
> vxlan-udp-recv function lookup vxlan_sock struct on every packet
> recv by using udp-port number. we can use sk->sk_user_data to
> store vxlan_sock and avoid lookup.
> I have open coded rcu-api to store and read vxlan_sock from
> sk_user_data to avoid sparse warning as sk_user_data is not
> __rcu pointer.
>
> Signed-off-by: Pravin B Shelar <pshelar@nicira.com>
Applied.
^ permalink raw reply
* Re: [PATCH v2 2/2] atm: he: print MAC via %pM
From: David Miller @ 2013-09-04 18:42 UTC (permalink / raw)
To: andriy.shevchenko; +Cc: chas, linux-atm-general, netdev
In-Reply-To: <1378210424-14781-2-git-send-email-andriy.shevchenko@linux.intel.com>
From: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Date: Tue, 3 Sep 2013 15:13:44 +0300
> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Applied.
^ permalink raw reply
* Re: [PATCH v2 1/2] atm: nicstar: re-use native mac_pton() helper
From: David Miller @ 2013-09-04 18:42 UTC (permalink / raw)
To: andriy.shevchenko; +Cc: chas, linux-atm-general, netdev, andy.shevchenko
In-Reply-To: <1378210424-14781-1-git-send-email-andriy.shevchenko@linux.intel.com>
From: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Date: Tue, 3 Sep 2013 15:13:43 +0300
> From: Andy Shevchenko <andy.shevchenko@gmail.com>
>
> There is a nice helper to parse MAC. Let's use it and remove custom
> implementation.
>
> Signed-off-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Applied.
^ permalink raw reply
* Re: [PATCH v2] ipv6: Don't depend on per socket memory for neighbour discovery messages
From: David Miller @ 2013-09-04 18:41 UTC (permalink / raw)
To: tgraf; +Cc: netdev, eric.dumazet, hannes, swarren, festevam
In-Reply-To: <e4600dac657e1160da7a5e7758dcb973b616a10e.1378207925.git.tgraf@suug.ch>
From: Thomas Graf <tgraf@suug.ch>
Date: Tue, 3 Sep 2013 13:37:01 +0200
> Allocating skbs when sending out neighbour discovery messages
> currently uses sock_alloc_send_skb() based on a per net namespace
> socket and thus share a socket wmem buffer space.
>
> If a netdevice is temporarily unable to transmit due to carrier
> loss or for other reasons, the queued up ndisc messages will cosnume
> all of the wmem space and will thus prevent from any more skbs to
> be allocated even for netdevices that are able to transmit packets.
>
> The number of neighbour discovery messages sent is very limited,
> use of alloc_skb() bypasses the socket wmem buffer size enforcement
> while the manual call to skb_set_owner_w() maintains the socket
> reference needed for the IPv6 output path.
>
> This patch has orginally been posted by Eric Dumazet in a modified
> form.
>
> Signed-off-by: Thomas Graf <tgraf@suug.ch>
Applied, and queued up for -stable along with Cong Wang's ndisc_send_rs()
locking change.
Thanks.
^ permalink raw reply
* Re: [PATCH v5] ipv6:introduce function to find route for redirect
From: Hannes Frederic Sowa @ 2013-09-04 18:36 UTC (permalink / raw)
To: Duan Jiong; +Cc: davem, netdev
In-Reply-To: <52271D15.1000806@cn.fujitsu.com>
On Wed, Sep 04, 2013 at 07:44:21PM +0800, Duan Jiong wrote:
> From: Duan Jiong <duanj.fnst@cn.fujitsu.com>
>
> RFC 4861 says that the IP source address of the Redirect is the
> same as the current first-hop router for the specified ICMP
> Destination Address, so the gateway should be taken into
> consideration when we find the route for redirect.
>
> There was once a check in commit
> a6279458c534d01ccc39498aba61c93083ee0372 ("NDISC: Search over
> all possible rules on receipt of redirect.") and the check
> went away in commit b94f1c0904da9b8bf031667afc48080ba7c3e8c9
> ("ipv6: Use icmpv6_notify() to propagate redirect, instead of
> rt6_redirect()").
>
> The bug is only "exploitable" on layer-2 because the source
> address of the redirect is checked to be a valid link-local
> address but it makes spoofing a lot easier in the same L2
> domain nonetheless.
>
> Thanks very much for Hannes's help.
>
> Signed-off-by: Duan Jiong <duanj.fnst@cn.fujitsu.com>
Tested and looks good!
Thanks!
Acked-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
^ permalink raw reply
* Re: [PATCH RFC] solos-pci: Fix BUG() with shared skb
From: David Miller @ 2013-09-04 18:30 UTC (permalink / raw)
To: dwmw2; +Cc: netdev, simon, linux-atm-general, nathan
In-Reply-To: <1378223125.4210.11.camel@i7.infradead.org>
From: David Woodhouse <dwmw2@infradead.org>
Date: Tue, 03 Sep 2013 16:45:25 +0100
> Simon reported this BUG():
>
> kernel BUG at net/core/skbuff.c:1065!
> Call Trace:
> [<f9b7c12c>] ? pppoatm_send+0x3f/0x1a0 [pppoatm]
> [<f8751797>] psend+0xa9/0x14a [solos_pci]
> [<f9b7c248>] pppoatm_send+0x15b/0x1a0 [pppoatm]
> [<f8a2f77d>] ppp_push+0x76/0x533 [ppp_generic]
>
> (Rest of backtrace at http://s85.org/mn0aOxMN ― the skb appears to be
> IPv6, forwarded from another interface over PPPoATM.)
>
> I wasn't expecting to see shared skbs in the ATM driver's ->send()
> function. Is this the right fix?
skb_realloc_headroom() should do everything you need.
This is what ethernet drivers do to prepend custom headers
when skb_headroom() is not large enough.
For example, see drivers/net/ethernet/sun/niu.c:niu_start_xmit().
There, the driver is attempting to prepend a TX descriptor to the
SKB.
If the SKB is shared, skb_realloc_headroom() will do the clone
for you if necessary.
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox