/* disable the VSIs and their queues that are not already DOWN */
ice_pf_dis_all_vsi(pf);
+ if (hw->port_info)
+ ice_sched_clear_port(hw->port_info);
+
ice_shutdown_all_ctrlq(hw);
set_bit(__ICE_PREPARED_FOR_RESET, pf->state);
/* When a CORER/GLOBR/EMPR is about to happen, the hardware triggers an
* OICR interrupt. The OICR handler (ice_misc_intr) determines what type
* of reset is pending and sets bits in pf->state indicating the reset
- * type and __ICE_RESET_OICR_RECV. So, if the latter bit is set
+ * type and __ICE_RESET_OICR_RECV. So, if the latter bit is set
* prepare for pending reset if not already (for PF software-initiated
* global resets the software should already be prepared for it as
* indicated by __ICE_PREPARED_FOR_RESET; for global resets initiated
* @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
+ * non-queue interrupts, e.g. AdminQ and errors. This is not used
* when in MSI or Legacy interrupt mode.
*/
static int ice_req_irq_msix_misc(struct ice_pf *pf)
{
struct ice_hw *hw = &pf->hw;
int oicr_idx, err = 0;
- u8 itr_gran;
u32 val;
if (!pf->int_name[0])
PFINT_MBX_CTL_CAUSE_ENA_M);
wr32(hw, PFINT_MBX_CTL, val);
- itr_gran = hw->itr_gran;
-
wr32(hw, GLINT_ITR(ICE_RX_ITR, pf->hw_oicr_idx),
- ITR_TO_REG(ICE_ITR_8K, itr_gran));
+ ITR_REG_ALIGN(ICE_ITR_8K) >> ICE_ITR_GRAN_S);
ice_flush(hw);
ice_irq_dynamic_ena(hw, NULL, NULL);
csumo_features = NETIF_F_RXCSUM |
NETIF_F_IP_CSUM |
+ NETIF_F_SCTP_CRC |
NETIF_F_IPV6_CSUM;
vlano_features = NETIF_F_HW_VLAN_CTAG_FILTER |
pf->num_lan_tx = min_t(int, q_left_tx, num_online_cpus());
- /* only 1 rx queue unless RSS is enabled */
+ /* only 1 Rx queue unless RSS is enabled */
if (!test_bit(ICE_FLAG_RSS_ENA, pf->flags))
pf->num_lan_rx = 1;
else
return 0;
}
+/**
+ * ice_verify_itr_gran - verify driver's assumption of ITR granularity
+ * @pf: pointer to the PF structure
+ *
+ * There is no error returned here because the driver will be able to handle a
+ * different ITR granularity, but interrupt moderation will not be accurate if
+ * the driver's assumptions are not verified. This assumption is made so we can
+ * use constants in the hot path instead of accessing structure members.
+ */
+static void ice_verify_itr_gran(struct ice_pf *pf)
+{
+ if (pf->hw.itr_gran != (ICE_ITR_GRAN_S << 1))
+ dev_warn(&pf->pdev->dev,
+ "%d ITR granularity assumption is invalid, actual ITR granularity is %d. Interrupt moderation will be inaccurate!\n",
+ (ICE_ITR_GRAN_S << 1), pf->hw.itr_gran);
+}
+
/**
* ice_verify_cacheline_size - verify driver's assumption of 64 Byte cache lines
* @pf: pointer to the PF structure
ice_determine_q_usage(pf);
- pf->num_alloc_vsi = min_t(u16, ICE_MAX_VSI_ALLOC,
- hw->func_caps.guaranteed_num_vsi);
+ pf->num_alloc_vsi = hw->func_caps.guar_num_vsi;
if (!pf->num_alloc_vsi) {
err = -EIO;
goto err_init_pf_unroll;
mod_timer(&pf->serv_tmr, round_jiffies(jiffies + pf->serv_tmr_period));
ice_verify_cacheline_size(pf);
+ ice_verify_itr_gran(pf);
return 0;
*/
static int ice_fdb_add(struct ndmsg *ndm, struct nlattr __always_unused *tb[],
struct net_device *dev, const unsigned char *addr,
- u16 vid, u16 flags)
+ u16 vid, u16 flags,
+ struct netlink_ext_ack *extack)
{
int err;
return err;
}
- err = ice_vsi_cfg_txqs(vsi);
+ err = ice_vsi_cfg_lan_txqs(vsi);
if (!err)
err = ice_vsi_cfg_rxqs(vsi);
if (!vsi->netdev)
return;
- for (q_idx = 0; q_idx < vsi->num_q_vectors; q_idx++)
- napi_enable(&vsi->q_vectors[q_idx]->napi);
+ for (q_idx = 0; q_idx < vsi->num_q_vectors; q_idx++) {
+ struct ice_q_vector *q_vector = vsi->q_vectors[q_idx];
+
+ if (q_vector->rx.ring || q_vector->tx.ring)
+ napi_enable(&q_vector->napi);
+ }
}
/**
if (!vsi->netdev)
return;
- for (q_idx = 0; q_idx < vsi->num_q_vectors; q_idx++)
- napi_disable(&vsi->q_vectors[q_idx]->napi);
+ for (q_idx = 0; q_idx < vsi->num_q_vectors; q_idx++) {
+ struct ice_q_vector *q_vector = vsi->q_vectors[q_idx];
+
+ if (q_vector->rx.ring || q_vector->tx.ring)
+ napi_disable(&q_vector->napi);
+ }
+}
+
+/**
+ * ice_force_phys_link_state - Force the physical link state
+ * @vsi: VSI to force the physical link state to up/down
+ * @link_up: true/false indicates to set the physical link to up/down
+ *
+ * Force the physical link state by getting the current PHY capabilities from
+ * hardware and setting the PHY config based on the determined capabilities. If
+ * link changes a link event will be triggered because both the Enable Automatic
+ * Link Update and LESM Enable bits are set when setting the PHY capabilities.
+ *
+ * Returns 0 on success, negative on failure
+ */
+static int ice_force_phys_link_state(struct ice_vsi *vsi, bool link_up)
+{
+ struct ice_aqc_get_phy_caps_data *pcaps;
+ struct ice_aqc_set_phy_cfg_data *cfg;
+ struct ice_port_info *pi;
+ struct device *dev;
+ int retcode;
+
+ if (!vsi || !vsi->port_info || !vsi->back)
+ return -EINVAL;
+ if (vsi->type != ICE_VSI_PF)
+ return 0;
+
+ dev = &vsi->back->pdev->dev;
+
+ pi = vsi->port_info;
+
+ pcaps = devm_kzalloc(dev, sizeof(*pcaps), GFP_KERNEL);
+ if (!pcaps)
+ return -ENOMEM;
+
+ retcode = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_SW_CFG, pcaps,
+ NULL);
+ if (retcode) {
+ dev_err(dev,
+ "Failed to get phy capabilities, VSI %d error %d\n",
+ vsi->vsi_num, retcode);
+ retcode = -EIO;
+ goto out;
+ }
+
+ /* No change in link */
+ if (link_up == !!(pcaps->caps & ICE_AQC_PHY_EN_LINK) &&
+ link_up == !!(pi->phy.link_info.link_info & ICE_AQ_LINK_UP))
+ goto out;
+
+ cfg = devm_kzalloc(dev, sizeof(*cfg), GFP_KERNEL);
+ if (!cfg) {
+ retcode = -ENOMEM;
+ goto out;
+ }
+
+ cfg->phy_type_low = pcaps->phy_type_low;
+ cfg->phy_type_high = pcaps->phy_type_high;
+ cfg->caps = pcaps->caps | ICE_AQ_PHY_ENA_AUTO_LINK_UPDT;
+ cfg->low_power_ctrl = pcaps->low_power_ctrl;
+ cfg->eee_cap = pcaps->eee_cap;
+ cfg->eeer_value = pcaps->eeer_value;
+ cfg->link_fec_opt = pcaps->link_fec_options;
+ if (link_up)
+ cfg->caps |= ICE_AQ_PHY_ENA_LINK;
+ else
+ cfg->caps &= ~ICE_AQ_PHY_ENA_LINK;
+
+ retcode = ice_aq_set_phy_cfg(&vsi->back->hw, pi->lport, cfg, NULL);
+ if (retcode) {
+ dev_err(dev, "Failed to set phy config, VSI %d error %d\n",
+ vsi->vsi_num, retcode);
+ retcode = -EIO;
+ }
+
+ devm_kfree(dev, cfg);
+out:
+ devm_kfree(dev, pcaps);
+ return retcode;
}
/**
*/
int ice_down(struct ice_vsi *vsi)
{
- int i, tx_err, rx_err;
+ int i, tx_err, rx_err, link_err = 0;
/* Caller of this function is expected to set the
* vsi->state __ICE_DOWN bit
}
ice_vsi_dis_irq(vsi);
- tx_err = ice_vsi_stop_tx_rings(vsi, ICE_NO_RESET, 0);
+
+ tx_err = ice_vsi_stop_lan_tx_rings(vsi, ICE_NO_RESET, 0);
if (tx_err)
netdev_err(vsi->netdev,
"Failed stop Tx rings, VSI %d error %d\n",
ice_napi_disable_all(vsi);
+ if (test_bit(ICE_FLAG_LINK_DOWN_ON_CLOSE_ENA, vsi->back->flags)) {
+ link_err = ice_force_phys_link_state(vsi, false);
+ if (link_err)
+ netdev_err(vsi->netdev,
+ "Failed to set physical link down, VSI %d error %d\n",
+ vsi->vsi_num, link_err);
+ }
+
ice_for_each_txq(vsi, i)
ice_clean_tx_ring(vsi->tx_rings[i]);
ice_for_each_rxq(vsi, i)
ice_clean_rx_ring(vsi->rx_rings[i]);
- if (tx_err || rx_err) {
+ if (tx_err || rx_err || link_err) {
netdev_err(vsi->netdev,
"Failed to close VSI 0x%04X on switch 0x%04X\n",
vsi->vsi_num, vsi->vsw->sw_id);
/**
* ice_dis_vsi - pause a VSI
* @vsi: the VSI being paused
+ * @locked: is the rtnl_lock already held
*/
-static void ice_dis_vsi(struct ice_vsi *vsi)
+static void ice_dis_vsi(struct ice_vsi *vsi, bool locked)
{
if (test_bit(__ICE_DOWN, vsi->state))
return;
if (vsi->type == ICE_VSI_PF && vsi->netdev) {
if (netif_running(vsi->netdev)) {
- rtnl_lock();
- vsi->netdev->netdev_ops->ndo_stop(vsi->netdev);
- rtnl_unlock();
+ if (!locked) {
+ rtnl_lock();
+ vsi->netdev->netdev_ops->ndo_stop(vsi->netdev);
+ rtnl_unlock();
+ } else {
+ vsi->netdev->netdev_ops->ndo_stop(vsi->netdev);
+ }
} else {
ice_vsi_close(vsi);
}
ice_for_each_vsi(pf, v)
if (pf->vsi[v])
- ice_dis_vsi(pf->vsi[v]);
+ ice_dis_vsi(pf->vsi[v], false);
}
/**
* @dev: the netdev being configured
* @nlh: RTNL message
* @flags: bridge setlink flags
+ * @extack: netlink extended ack
*
* Sets the bridge mode (VEB/VEPA) of the switch to which the netdev (VSI) is
* hooked up to. Iterates through the PF VSI list and sets the loopback mode (if
*/
static int
ice_bridge_setlink(struct net_device *dev, struct nlmsghdr *nlh,
- u16 __always_unused flags)
+ u16 __always_unused flags,
+ struct netlink_ext_ack __always_unused *extack)
{
struct ice_netdev_priv *np = netdev_priv(dev);
struct ice_pf *pf = np->vsi->back;
*/
status = ice_update_sw_rule_bridge_mode(hw);
if (status) {
- netdev_err(dev, "update SW_RULE for bridge mode failed, = %d err %d aq_err %d\n",
+ netdev_err(dev, "switch rule update failed, mode = %d err %d aq_err %d\n",
mode, status, hw->adminq.sq_last_status);
/* revert hw->evb_veb */
hw->evb_veb = (pf_sw->bridge_mode == BRIDGE_MODE_VEB);
struct ice_ring *tx_ring = NULL;
struct ice_vsi *vsi = np->vsi;
struct ice_pf *pf = vsi->back;
- u32 head, val = 0, i;
int hung_queue = -1;
+ u32 i;
pf->tx_timeout_count++;
- /* find the stopped queue the same way the stack does */
+ /* find the stopped queue the same way dev_watchdog() does */
for (i = 0; i < netdev->num_tx_queues; i++) {
- struct netdev_queue *q;
unsigned long trans_start;
+ struct netdev_queue *q;
q = netdev_get_tx_queue(netdev, i);
trans_start = q->trans_start;
if (netif_xmit_stopped(q) &&
time_after(jiffies,
- (trans_start + netdev->watchdog_timeo))) {
+ trans_start + netdev->watchdog_timeo)) {
hung_queue = i;
break;
}
}
- if (i == netdev->num_tx_queues) {
+ if (i == netdev->num_tx_queues)
netdev_info(netdev, "tx_timeout: no netdev hung queue found\n");
- } else {
+ else
/* now that we have an index, find the tx_ring struct */
- for (i = 0; i < vsi->num_txq; i++) {
- if (vsi->tx_rings[i] && vsi->tx_rings[i]->desc) {
- if (hung_queue ==
- vsi->tx_rings[i]->q_index) {
+ for (i = 0; i < vsi->num_txq; i++)
+ if (vsi->tx_rings[i] && vsi->tx_rings[i]->desc)
+ if (hung_queue == vsi->tx_rings[i]->q_index) {
tx_ring = vsi->tx_rings[i];
break;
}
- }
- }
- }
/* Reset recovery level if enough time has elapsed after last timeout.
* Also ensure no new reset action happens before next timeout period.
return;
if (tx_ring) {
- head = tx_ring->next_to_clean;
+ struct ice_hw *hw = &pf->hw;
+ u32 head, val = 0;
+
+ head = (rd32(hw, QTX_COMM_HEAD(vsi->txq_map[hung_queue])) &
+ QTX_COMM_HEAD_HEAD_M) >> QTX_COMM_HEAD_HEAD_S;
/* Read interrupt register */
if (test_bit(ICE_FLAG_MSIX_ENA, pf->flags))
- val = rd32(&pf->hw,
+ val = rd32(hw,
GLINT_DYN_CTL(tx_ring->q_vector->v_idx +
- tx_ring->vsi->hw_base_vector));
+ tx_ring->vsi->hw_base_vector));
- netdev_info(netdev, "tx_timeout: VSI_num: %d, Q %d, NTC: 0x%x, HWB: 0x%x, NTU: 0x%x, TAIL: 0x%x, INT: 0x%x\n",
+ netdev_info(netdev, "tx_timeout: VSI_num: %d, Q %d, NTC: 0x%x, HW_HEAD: 0x%x, NTU: 0x%x, INT: 0x%x\n",
vsi->vsi_num, hung_queue, tx_ring->next_to_clean,
- head, tx_ring->next_to_use,
- readl(tx_ring->tail), val);
+ head, tx_ring->next_to_use, val);
}
pf->tx_timeout_last_recovery = jiffies;
* @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
+ * 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 is enabled,
* and the stack is notified that the interface is ready.
netif_carrier_off(netdev);
- err = ice_vsi_open(vsi);
+ err = ice_force_phys_link_state(vsi, true);
+ if (err) {
+ netdev_err(netdev,
+ "Failed to set physical link up, error %d\n", err);
+ return err;
+ }
+ err = ice_vsi_open(vsi);
if (err)
netdev_err(netdev, "Failed to open VSI 0x%04X on switch 0x%04X\n",
vsi->vsi_num, vsi->vsw->sw_id);
* @netdev: network interface device structure
*
* The stop entry point is called when an interface is de-activated by the OS,
- * and the netdevice enters the DOWN state. The hardware is still under the
+ * and the netdevice enters the DOWN state. The hardware is still under the
* driver's control, but the netdev interface is disabled.
*
* Returns success only - not allowed to fail
size_t len;
/* No point in doing any of this if neither checksum nor GSO are
- * being requested for this frame. We can rule out both by just
+ * being requested for this frame. We can rule out both by just
* checking for CHECKSUM_PARTIAL
*/
if (skb->ip_summed != CHECKSUM_PARTIAL)
return features;
/* We cannot support GSO if the MSS is going to be less than
- * 64 bytes. If it is then we need to drop support for GSO.
+ * 64 bytes. If it is then we need to drop support for GSO.
*/
if (skb_is_gso(skb) && (skb_shinfo(skb)->gso_size < 64))
features &= ~NETIF_F_GSO_MASK;