* [Intel-wired-lan] [PATCH 11/13] ice: Enable firmware logging during device initialization.
From: Anirudh Venkataramanan @ 2018-07-23 23:13 UTC (permalink / raw)
To: intel-wired-lan
In-Reply-To: <20180723231348.14838-1-anirudh.venkataramanan@intel.com>
From: Hieu Tran <hieu.t.tran@intel.com>
To enable FW logging, the "cq_en" and "uart_en" enable bits of the
"fw_log" element in struct ice_hw need to set accordingly based on
some user-provided parameters during driver loading. To select which
FW log events to be emitted, the "cfg" elements of corresponding FW
modules in the "evnts" array member of "fw_log" need to be configured.
Signed-off-by: Hieu Tran <hieu.t.tran@intel.com>
Signed-off-by: Anirudh Venkataramanan <anirudh.venkataramanan@intel.com>
---
[Anirudh Venkataramanan <anirudh.venkataramanan@intel.com> cleaned up commit message]
---
drivers/net/ethernet/intel/ice/ice_adminq_cmd.h | 83 +++++++++++
drivers/net/ethernet/intel/ice/ice_common.c | 180 ++++++++++++++++++++++++
drivers/net/ethernet/intel/ice/ice_common.h | 1 +
drivers/net/ethernet/intel/ice/ice_main.c | 3 +
drivers/net/ethernet/intel/ice/ice_type.h | 19 +++
5 files changed, 286 insertions(+)
diff --git a/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h b/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
index ce3405f870d8..8fb54c3e7bf2 100644
--- a/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
@@ -1207,6 +1207,84 @@ struct ice_aqc_dis_txq {
struct ice_aqc_dis_txq_item qgrps[1];
};
+/* Configure Firmware Logging Command (indirect 0xFF09)
+ * Logging Information Read Response (indirect 0xFF10)
+ * Note: The 0xFF10 command has no input parameters.
+ */
+struct ice_aqc_fw_logging {
+ u8 log_ctrl;
+#define ICE_AQC_FW_LOG_AQ_EN BIT(0)
+#define ICE_AQC_FW_LOG_UART_EN BIT(1)
+ u8 rsvd0;
+ u8 log_ctrl_valid; /* Not used by 0xFF10 Response */
+#define ICE_AQC_FW_LOG_AQ_VALID BIT(0)
+#define ICE_AQC_FW_LOG_UART_VALID BIT(1)
+ u8 rsvd1[5];
+ __le32 addr_high;
+ __le32 addr_low;
+};
+
+enum ice_aqc_fw_logging_mod {
+ ICE_AQC_FW_LOG_ID_GENERAL = 0,
+ ICE_AQC_FW_LOG_ID_CTRL,
+ ICE_AQC_FW_LOG_ID_LINK,
+ ICE_AQC_FW_LOG_ID_LINK_TOPO,
+ ICE_AQC_FW_LOG_ID_DNL,
+ ICE_AQC_FW_LOG_ID_I2C,
+ ICE_AQC_FW_LOG_ID_SDP,
+ ICE_AQC_FW_LOG_ID_MDIO,
+ ICE_AQC_FW_LOG_ID_ADMINQ,
+ ICE_AQC_FW_LOG_ID_HDMA,
+ ICE_AQC_FW_LOG_ID_LLDP,
+ ICE_AQC_FW_LOG_ID_DCBX,
+ ICE_AQC_FW_LOG_ID_DCB,
+ ICE_AQC_FW_LOG_ID_NETPROXY,
+ ICE_AQC_FW_LOG_ID_NVM,
+ ICE_AQC_FW_LOG_ID_AUTH,
+ ICE_AQC_FW_LOG_ID_VPD,
+ ICE_AQC_FW_LOG_ID_IOSF,
+ ICE_AQC_FW_LOG_ID_PARSER,
+ ICE_AQC_FW_LOG_ID_SW,
+ ICE_AQC_FW_LOG_ID_SCHEDULER,
+ ICE_AQC_FW_LOG_ID_TXQ,
+ ICE_AQC_FW_LOG_ID_RSVD,
+ ICE_AQC_FW_LOG_ID_POST,
+ ICE_AQC_FW_LOG_ID_WATCHDOG,
+ ICE_AQC_FW_LOG_ID_TASK_DISPATCH,
+ ICE_AQC_FW_LOG_ID_MNG,
+ ICE_AQC_FW_LOG_ID_MAX,
+};
+
+/* This is the buffer for both of the logging commands.
+ * The entry array size depends on the datalen parameter in the descriptor.
+ * There will be a total of datalen / 2 entries.
+ */
+struct ice_aqc_fw_logging_data {
+ __le16 entry[1];
+#define ICE_AQC_FW_LOG_ID_S 0
+#define ICE_AQC_FW_LOG_ID_M (0xFFF << ICE_AQC_FW_LOG_ID_S)
+
+#define ICE_AQC_FW_LOG_CONF_SUCCESS 0 /* Used by response */
+#define ICE_AQC_FW_LOG_CONF_BAD_INDX BIT(12) /* Used by response */
+
+#define ICE_AQC_FW_LOG_EN_S 12
+#define ICE_AQC_FW_LOG_EN_M (0xF << ICE_AQC_FW_LOG_EN_S)
+#define ICE_AQC_FW_LOG_INFO_EN BIT(12) /* Used by command */
+#define ICE_AQC_FW_LOG_INIT_EN BIT(13) /* Used by command */
+#define ICE_AQC_FW_LOG_FLOW_EN BIT(14) /* Used by command */
+#define ICE_AQC_FW_LOG_ERR_EN BIT(15) /* Used by command */
+};
+
+/* Get/Clear FW Log (indirect 0xFF11) */
+struct ice_aqc_get_clear_fw_log {
+ u8 flags;
+#define ICE_AQC_FW_LOG_CLEAR BIT(0)
+#define ICE_AQC_FW_LOG_MORE_DATA_AVAIL BIT(1)
+ u8 rsvd1[7];
+ __le32 addr_high;
+ __le32 addr_low;
+};
+
/**
* struct ice_aq_desc - Admin Queue (AQ) descriptor
* @flags: ICE_AQ_FLAG_* flags
@@ -1257,6 +1335,8 @@ struct ice_aq_desc {
struct ice_aqc_dis_txqs dis_txqs;
struct ice_aqc_add_get_update_free_vsi vsi_cmd;
struct ice_aqc_add_update_free_vsi_resp add_update_free_vsi_res;
+ struct ice_aqc_fw_logging fw_logging;
+ struct ice_aqc_get_clear_fw_log get_clear_fw_log;
struct ice_aqc_alloc_free_res_cmd sw_res_ctrl;
struct ice_aqc_set_event_mask set_event_mask;
struct ice_aqc_get_link_status get_link_status;
@@ -1354,6 +1434,9 @@ enum ice_adminq_opc {
/* TX queue handling commands/events */
ice_aqc_opc_add_txqs = 0x0C30,
ice_aqc_opc_dis_txqs = 0x0C31,
+
+ /* debug commands */
+ ice_aqc_opc_fw_logging = 0xFF09,
};
#endif /* _ICE_ADMINQ_CMD_H_ */
diff --git a/drivers/net/ethernet/intel/ice/ice_common.c b/drivers/net/ethernet/intel/ice/ice_common.c
index 67c95fed320a..d6bfa90d9e19 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.c
+++ b/drivers/net/ethernet/intel/ice/ice_common.c
@@ -369,6 +369,177 @@ static void ice_cleanup_fltr_mgmt_struct(struct ice_hw *hw)
devm_kfree(ice_hw_to_dev(hw), sw);
}
+#define ICE_FW_LOG_DESC_SIZE(n) (sizeof(struct ice_aqc_fw_logging_data) + \
+ (((n) - 1) * sizeof(((struct ice_aqc_fw_logging_data *)0)->entry)))
+#define ICE_FW_LOG_DESC_SIZE_MAX \
+ ICE_FW_LOG_DESC_SIZE(ICE_AQC_FW_LOG_ID_MAX)
+
+/**
+ * ice_cfg_fw_log - configure FW logging
+ * @hw: pointer to the hw struct
+ * @enable: enable certain FW logging events if true, disable all if false
+ *
+ * This function enables/disables the FW logging via Rx CQ events and a UART
+ * port based on predetermined configurations. FW logging via the Rx CQ can be
+ * enabled/disabled for individual PF's. However, FW logging via the UART can
+ * only be enabled/disabled for all PFs on the same device.
+ *
+ * To enable overall FW logging, the "cq_en" and "uart_en" enable bits in
+ * hw->fw_log need to be set accordingly, e.g. based on user-provided input,
+ * before initializing the device.
+ *
+ * When re/configuring FW logging, callers need to update the "cfg" elements of
+ * the hw->fw_log.evnts array with the desired logging event configurations for
+ * modules of interest. When disabling FW logging completely, the callers can
+ * just pass false in the "enable" parameter. On completion, the function will
+ * update the "cur" element of the the hw->fw_log.evnts array with the resulting
+ * logging event configurations of the modules that are being re/configured. FW
+ * logging modules that are not part of a reconfiguration operation retain their
+ * previous states.
+ *
+ * Before resetting the device, it is recommended that the driver disables FW
+ * logging before shutting down the control queue. When disabling FW logging
+ * ("enable" = false), the latest configurations of FW logging events stored in
+ * hw->fw_log.evnts[] are not overridden to allow them to be reconfigured after
+ * a device reset.
+ *
+ * When enabling FW logging to emit log messages via the Rx CQ during the
+ * device's initialization phase, a mechanism alternative to interrupt handlers
+ * needs to be used to extract FW log messages from the Rx CQ periodically and
+ * to prevent the Rx CQ from being full and stalling other types of control
+ * messages from FW to SW. Interrupts are typically disabled during the device's
+ * initialization phase.
+ */
+
+static enum ice_status ice_cfg_fw_log(struct ice_hw *hw, bool enable)
+{
+ struct ice_aqc_fw_logging_data *data = NULL;
+ struct ice_aqc_fw_logging *cmd;
+ enum ice_status status = 0;
+ u16 i, chgs = 0, len = 0;
+ struct ice_aq_desc desc;
+ u8 actv_evnts = 0;
+ void *buf = NULL;
+
+ if (!hw->fw_log.cq_en && !hw->fw_log.uart_en)
+ return 0;
+
+ /* Disable FW logging only when the control queue is still responsive */
+ if (!enable &&
+ (!hw->fw_log.actv_evnts || !ice_check_sq_alive(hw, &hw->adminq)))
+ return 0;
+
+ ice_fill_dflt_direct_cmd_desc(&desc, ice_aqc_opc_fw_logging);
+ cmd = &desc.params.fw_logging;
+
+ /* Indicate which controls are valid */
+ if (hw->fw_log.cq_en)
+ cmd->log_ctrl_valid |= ICE_AQC_FW_LOG_AQ_VALID;
+
+ if (hw->fw_log.uart_en)
+ cmd->log_ctrl_valid |= ICE_AQC_FW_LOG_UART_VALID;
+
+ if (enable) {
+ /* Fill in an array of entries with FW logging modules and
+ * logging events being reconfigured.
+ */
+ for (i = 0; i < ICE_AQC_FW_LOG_ID_MAX; i++) {
+ u16 val;
+
+ /* Keep track of enabled event types */
+ actv_evnts |= hw->fw_log.evnts[i].cfg;
+
+ if (hw->fw_log.evnts[i].cfg == hw->fw_log.evnts[i].cur)
+ continue;
+
+ if (!data) {
+ data = devm_kzalloc(ice_hw_to_dev(hw),
+ ICE_FW_LOG_DESC_SIZE_MAX,
+ GFP_KERNEL);
+ if (!data)
+ return ICE_ERR_NO_MEMORY;
+ }
+
+ val = i << ICE_AQC_FW_LOG_ID_S;
+ val |= hw->fw_log.evnts[i].cfg << ICE_AQC_FW_LOG_EN_S;
+ data->entry[chgs++] = cpu_to_le16(val);
+ }
+
+ /* Only enable FW logging if at least one module is specified.
+ * If FW logging is currently enabled but all modules are not
+ * enabled to emit log messages, disable FW logging altogether.
+ */
+ if (actv_evnts) {
+ /* Leave if there is effectively no change */
+ if (!chgs)
+ goto out;
+
+ if (hw->fw_log.cq_en)
+ cmd->log_ctrl |= ICE_AQC_FW_LOG_AQ_EN;
+
+ if (hw->fw_log.uart_en)
+ cmd->log_ctrl |= ICE_AQC_FW_LOG_UART_EN;
+
+ buf = data;
+ len = ICE_FW_LOG_DESC_SIZE(chgs);
+ desc.flags |= cpu_to_le16(ICE_AQ_FLAG_RD);
+ }
+ }
+
+ status = ice_aq_send_cmd(hw, &desc, buf, len, NULL);
+ if (!status) {
+ /* Update the current configuration to reflect events enabled.
+ * hw->fw_log.cq_en and hw->fw_log.uart_en indicate if the FW
+ * logging mode is enabled for the device. They do not reflect
+ * actual modules being enabled to emit log messages. So, their
+ * values remain unchanged even when all modules are disabled.
+ */
+ u16 cnt = enable ? chgs : (u16)ICE_AQC_FW_LOG_ID_MAX;
+
+ hw->fw_log.actv_evnts = actv_evnts;
+ for (i = 0; i < cnt; i++) {
+ u16 v, m;
+
+ if (!enable) {
+ /* When disabling all FW logging events as part
+ * of device's de-initialization, the original
+ * configurations are retained, and can be used
+ * to reconfigure FW logging later if the device
+ * is re-initialized.
+ */
+ hw->fw_log.evnts[i].cur = 0;
+ continue;
+ }
+
+ v = le16_to_cpu(data->entry[i]);
+ m = (v & ICE_AQC_FW_LOG_ID_M) >> ICE_AQC_FW_LOG_ID_S;
+ hw->fw_log.evnts[m].cur = hw->fw_log.evnts[m].cfg;
+ }
+ }
+
+out:
+ if (data)
+ devm_kfree(ice_hw_to_dev(hw), data);
+
+ return status;
+}
+
+/**
+ * ice_output_fw_log
+ * @hw: pointer to the hw struct
+ * @desc: pointer to the AQ message descriptor
+ * @buf: pointer to the buffer accompanying the AQ message
+ *
+ * Formats a FW Log message and outputs it via the standard driver logs.
+ */
+void ice_output_fw_log(struct ice_hw *hw, struct ice_aq_desc *desc, void *buf)
+{
+ ice_debug(hw, ICE_DBG_AQ_MSG, "[ FW Log Msg Start ]\n");
+ ice_debug_array(hw, ICE_DBG_AQ_MSG, 16, 1, (u8 *)buf,
+ le16_to_cpu(desc->datalen));
+ ice_debug(hw, ICE_DBG_AQ_MSG, "[ FW Log Msg End ]\n");
+}
+
/**
* ice_init_hw - main hardware initialization routine
* @hw: pointer to the hardware structure
@@ -403,6 +574,11 @@ enum ice_status ice_init_hw(struct ice_hw *hw)
if (status)
goto err_unroll_cqinit;
+ /* Enable FW logging. Not fatal if this fails. */
+ status = ice_cfg_fw_log(hw, true);
+ if (status)
+ ice_debug(hw, ICE_DBG_INIT, "Failed to enable FW logging.\n");
+
status = ice_clear_pf_cfg(hw);
if (status)
goto err_unroll_cqinit;
@@ -509,6 +685,10 @@ enum ice_status ice_init_hw(struct ice_hw *hw)
void ice_deinit_hw(struct ice_hw *hw)
{
ice_sched_cleanup_all(hw);
+
+ /* Attempt to disable FW logging before shutting down control queues */
+ ice_cfg_fw_log(hw, false);
+
ice_shutdown_all_ctrlq(hw);
if (hw->port_info) {
diff --git a/drivers/net/ethernet/intel/ice/ice_common.h b/drivers/net/ethernet/intel/ice/ice_common.h
index 34c468138a8a..8e56bbbeda56 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.h
+++ b/drivers/net/ethernet/intel/ice/ice_common.h
@@ -95,4 +95,5 @@ enum ice_status
ice_ena_vsi_txq(struct ice_port_info *pi, u16 vsi_id, u8 tc, u8 num_qgrps,
struct ice_aqc_add_tx_qgrp *buf, u16 buf_size,
struct ice_sq_cd *cd);
+void ice_output_fw_log(struct ice_hw *hw, struct ice_aq_desc *desc, void *buf);
#endif /* _ICE_COMMON_H_ */
diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index 889631be02f9..e1ec525138ac 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -984,6 +984,9 @@ static int __ice_clean_ctrlq(struct ice_pf *pf, enum ice_ctl_q q_type)
dev_err(&pf->pdev->dev,
"Could not handle link event");
break;
+ case ice_aqc_opc_fw_logging:
+ ice_output_fw_log(hw, &event.desc, event.msg_buf);
+ break;
default:
dev_dbg(&pf->pdev->dev,
"%s Receive Queue unknown event 0x%04x ignored\n",
diff --git a/drivers/net/ethernet/intel/ice/ice_type.h b/drivers/net/ethernet/intel/ice/ice_type.h
index 75457afc912c..9f57306daf85 100644
--- a/drivers/net/ethernet/intel/ice/ice_type.h
+++ b/drivers/net/ethernet/intel/ice/ice_type.h
@@ -253,6 +253,24 @@ struct ice_switch_info {
struct ice_sw_recipe *recp_list;
};
+/* FW logging configuration */
+struct ice_fw_log_evnt {
+ u8 cfg : 4; /* New event enables to configure */
+ u8 cur : 4; /* Current/active event enables */
+};
+
+struct ice_fw_log_cfg {
+ u8 cq_en : 1; /* FW logging is enabled via the control queue */
+ u8 uart_en : 1; /* FW logging is enabled via UART for all PFs */
+ u8 actv_evnts; /* Cumulation of currently enabled log events */
+
+#define ICE_FW_LOG_EVNT_INFO (ICE_AQC_FW_LOG_INFO_EN >> ICE_AQC_FW_LOG_EN_S)
+#define ICE_FW_LOG_EVNT_INIT (ICE_AQC_FW_LOG_INIT_EN >> ICE_AQC_FW_LOG_EN_S)
+#define ICE_FW_LOG_EVNT_FLOW (ICE_AQC_FW_LOG_FLOW_EN >> ICE_AQC_FW_LOG_EN_S)
+#define ICE_FW_LOG_EVNT_ERR (ICE_AQC_FW_LOG_ERR_EN >> ICE_AQC_FW_LOG_EN_S)
+ struct ice_fw_log_evnt evnts[ICE_AQC_FW_LOG_ID_MAX];
+};
+
/* Port hardware description */
struct ice_hw {
u8 __iomem *hw_addr;
@@ -301,6 +319,7 @@ struct ice_hw {
u8 fw_patch; /* firmware patch version */
u32 fw_build; /* firmware build number */
+ struct ice_fw_log_cfg fw_log;
/* minimum allowed value for different speeds */
#define ICE_ITR_GRAN_MIN_200 1
#define ICE_ITR_GRAN_MIN_100 1
--
2.14.3
^ permalink raw reply related
* [Intel-wired-lan] [PATCH 12/13] ice: Make ice_vsi_kill_vlan return an error code
From: Anirudh Venkataramanan @ 2018-07-23 23:13 UTC (permalink / raw)
To: intel-wired-lan
In-Reply-To: <20180723231348.14838-1-anirudh.venkataramanan@intel.com>
From: Brett Creeley <brett.creeley@intel.com>
Prior to this patch ice_vsi_kill_vlan has a void return value and we
were ignoring if it fails. Make ice_vsi_kill_vlan return an integer
value so its return value can be checked and used to determine whether
or not to update VLAN counts, structures, pruning bits.
Signed-off-by: Brett Creeley <brett.creeley@intel.com>
Signed-off-by: Anirudh Venkataramanan <anirudh.venkataramanan@intel.com>
---
[Anirudh Venkataramanan <anirudh.venkataramanan@intel.com> cleaned up commit message]
---
drivers/net/ethernet/intel/ice/ice_main.c | 94 ++++++++++++++++++++++++++++---
1 file changed, 85 insertions(+), 9 deletions(-)
diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index e1ec525138ac..cb1145515f98 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -349,6 +349,63 @@ static bool ice_vsi_fltr_changed(struct ice_vsi *vsi)
test_bit(ICE_VSI_FLAG_VLAN_FLTR_CHANGED, vsi->flags);
}
+/**
+ * ice_cfg_vlan_pruning - enable or disable VLAN pruning on the VSI
+ * @vsi: VSI to enable or disable VLAN pruning on
+ * @ena: set to true to enable VLAN pruning and false to disable it
+ *
+ * returns 0 if VSI is updated, negative otherwise
+ */
+static int ice_cfg_vlan_pruning(struct ice_vsi *vsi, bool ena)
+{
+ struct ice_vsi_ctx *ctxt;
+ struct device *dev;
+ int status;
+
+ if (!vsi)
+ return -EINVAL;
+
+ dev = &vsi->back->pdev->dev;
+ ctxt = devm_kzalloc(dev, sizeof(*ctxt), GFP_KERNEL);
+ if (!ctxt)
+ return -ENOMEM;
+
+ ctxt->info = vsi->info;
+
+ if (ena) {
+ ctxt->info.sec_flags |=
+ ICE_AQ_VSI_SEC_TX_VLAN_PRUNE_ENA <<
+ ICE_AQ_VSI_SEC_TX_PRUNE_ENA_S;
+ ctxt->info.sw_flags2 |= ICE_AQ_VSI_SW_FLAG_RX_VLAN_PRUNE_ENA;
+ } else {
+ ctxt->info.sec_flags &=
+ ~(ICE_AQ_VSI_SEC_TX_VLAN_PRUNE_ENA <<
+ ICE_AQ_VSI_SEC_TX_PRUNE_ENA_S);
+ ctxt->info.sw_flags2 &= ~ICE_AQ_VSI_SW_FLAG_RX_VLAN_PRUNE_ENA;
+ }
+
+ ctxt->info.valid_sections = cpu_to_le16(ICE_AQ_VSI_PROP_SECURITY_VALID |
+ ICE_AQ_VSI_PROP_SW_VALID);
+ ctxt->vsi_num = vsi->vsi_num;
+ status = ice_aq_update_vsi(&vsi->back->hw, ctxt, NULL);
+ if (status) {
+ netdev_err(vsi->netdev, "%sabling VLAN pruning on VSI %d failed, err = %d, aq_err = %d\n",
+ ena ? "Ena" : "Dis", vsi->vsi_num, status,
+ vsi->back->hw.adminq.sq_last_status);
+ goto err_out;
+ }
+
+ vsi->info.sec_flags = ctxt->info.sec_flags;
+ vsi->info.sw_flags2 = ctxt->info.sw_flags2;
+
+ devm_kfree(dev, ctxt);
+ return 0;
+
+err_out:
+ devm_kfree(dev, ctxt);
+ return -EIO;
+}
+
/**
* ice_vsi_sync_fltr - Update the VSI filter list to the HW
* @vsi: ptr to the VSI
@@ -3128,7 +3185,7 @@ static int ice_vlan_rx_add_vid(struct net_device *netdev,
{
struct ice_netdev_priv *np = netdev_priv(netdev);
struct ice_vsi *vsi = np->vsi;
- int ret = 0;
+ int ret;
if (vid >= VLAN_N_VID) {
netdev_err(netdev, "VLAN id requested %d is out of range %d\n",
@@ -3139,6 +3196,13 @@ static int ice_vlan_rx_add_vid(struct net_device *netdev,
if (vsi->info.pvid)
return -EINVAL;
+ /* Enable VLAN pruning when VLAN 0 is added */
+ if (unlikely(!vid)) {
+ ret = ice_cfg_vlan_pruning(vsi, true);
+ if (ret)
+ return ret;
+ }
+
/* Add all VLAN ids including 0 to the switch filter. VLAN id 0 is
* needed to continue allowing all untagged packets since VLAN prune
* list is applied to all packets by the switch
@@ -3155,16 +3219,19 @@ static int ice_vlan_rx_add_vid(struct net_device *netdev,
* ice_vsi_kill_vlan - Remove VSI membership for a given VLAN
* @vsi: the VSI being configured
* @vid: VLAN id to be removed
+ *
+ * Returns 0 on success and negative on failure
*/
-static void ice_vsi_kill_vlan(struct ice_vsi *vsi, u16 vid)
+static int ice_vsi_kill_vlan(struct ice_vsi *vsi, u16 vid)
{
struct ice_fltr_list_entry *list;
struct ice_pf *pf = vsi->back;
LIST_HEAD(tmp_add_list);
+ int status = 0;
list = devm_kzalloc(&pf->pdev->dev, sizeof(*list), GFP_KERNEL);
if (!list)
- return;
+ return -ENOMEM;
list->fltr_info.lkup_type = ICE_SW_LKUP_VLAN;
list->fltr_info.fwd_id.vsi_id = vsi->vsi_num;
@@ -3176,11 +3243,14 @@ static void ice_vsi_kill_vlan(struct ice_vsi *vsi, u16 vid)
INIT_LIST_HEAD(&list->list_entry);
list_add(&list->list_entry, &tmp_add_list);
- if (ice_remove_vlan(&pf->hw, &tmp_add_list))
+ if (ice_remove_vlan(&pf->hw, &tmp_add_list)) {
dev_err(&pf->pdev->dev, "Error removing VLAN %d on vsi %i\n",
vid, vsi->vsi_num);
+ status = -EIO;
+ }
ice_free_fltr_list(&pf->pdev->dev, &tmp_add_list);
+ return status;
}
/**
@@ -3196,19 +3266,25 @@ static int ice_vlan_rx_kill_vid(struct net_device *netdev,
{
struct ice_netdev_priv *np = netdev_priv(netdev);
struct ice_vsi *vsi = np->vsi;
+ int status;
if (vsi->info.pvid)
return -EINVAL;
- /* return code is ignored as there is nothing a user
- * can do about failure to remove and a log message was
- * already printed from the other function
+ /* Make sure ice_vsi_kill_vlan is successful before updating VLAN
+ * information
*/
- ice_vsi_kill_vlan(vsi, vid);
+ status = ice_vsi_kill_vlan(vsi, vid);
+ if (status)
+ return status;
clear_bit(vid, vsi->active_vlans);
- return 0;
+ /* Disable VLAN pruning when VLAN 0 is removed */
+ if (unlikely(!vid))
+ status = ice_cfg_vlan_pruning(vsi, false);
+
+ return status;
}
/**
--
2.14.3
^ permalink raw reply related
* [Intel-wired-lan] [PATCH 13/13] ice: Introduce SERVICE_DIS flag and service routine functions
From: Anirudh Venkataramanan @ 2018-07-23 23:13 UTC (permalink / raw)
To: intel-wired-lan
In-Reply-To: <20180723231348.14838-1-anirudh.venkataramanan@intel.com>
From: Akeem G Abodunrin <akeem.g.abodunrin@intel.com>
This patch introduces SERVICE_DIS flag to use for stopping service task.
This flag will be checked before scheduling new tasks. Also add new
functions ice_service_task_stop to stop service task.
Signed-off-by: Akeem G Abodunrin <akeem.g.abodunrin@intel.com>
Signed-off-by: Anirudh Venkataramanan <anirudh.venkataramanan@intel.com>
---
drivers/net/ethernet/intel/ice/ice.h | 1 +
drivers/net/ethernet/intel/ice/ice_main.c | 34 ++++++++++++++++++++++++-------
2 files changed, 28 insertions(+), 7 deletions(-)
diff --git a/drivers/net/ethernet/intel/ice/ice.h b/drivers/net/ethernet/intel/ice/ice.h
index 6f44a850c4b2..9cf233d085d8 100644
--- a/drivers/net/ethernet/intel/ice/ice.h
+++ b/drivers/net/ethernet/intel/ice/ice.h
@@ -138,6 +138,7 @@ enum ice_state {
__ICE_FLTR_OVERFLOW_PROMISC,
__ICE_CFG_BUSY,
__ICE_SERVICE_SCHED,
+ __ICE_SERVICE_DIS,
__ICE_STATE_NBITS /* must be last */
};
diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index cb1145515f98..3a77812e4be7 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -1107,7 +1107,7 @@ static void ice_clean_adminq_subtask(struct ice_pf *pf)
*/
static void ice_service_task_schedule(struct ice_pf *pf)
{
- if (!test_bit(__ICE_DOWN, pf->state) &&
+ if (!test_bit(__ICE_SERVICE_DIS, pf->state) &&
!test_and_set_bit(__ICE_SERVICE_SCHED, pf->state) &&
!test_bit(__ICE_NEEDS_RESTART, pf->state))
queue_work(ice_wq, &pf->serv_task);
@@ -1126,6 +1126,22 @@ static void ice_service_task_complete(struct ice_pf *pf)
clear_bit(__ICE_SERVICE_SCHED, pf->state);
}
+/**
+ * ice_service_task_stop - stop service task and cancel works
+ * @pf: board private structure
+ */
+static void ice_service_task_stop(struct ice_pf *pf)
+{
+ set_bit(__ICE_SERVICE_DIS, pf->state);
+
+ if (pf->serv_tmr.function)
+ del_timer_sync(&pf->serv_tmr);
+ if (pf->serv_task.func)
+ cancel_work_sync(&pf->serv_task);
+
+ clear_bit(__ICE_SERVICE_SCHED, pf->state);
+}
+
/**
* ice_service_timer - timer callback to schedule service task
* @t: pointer to timer_list
@@ -3391,10 +3407,7 @@ static void ice_determine_q_usage(struct ice_pf *pf)
*/
static void ice_deinit_pf(struct ice_pf *pf)
{
- if (pf->serv_tmr.function)
- del_timer_sync(&pf->serv_tmr);
- if (pf->serv_task.func)
- cancel_work_sync(&pf->serv_task);
+ ice_service_task_stop(pf);
mutex_destroy(&pf->sw_mutex);
mutex_destroy(&pf->avail_q_mutex);
}
@@ -3601,6 +3614,8 @@ static int ice_probe(struct pci_dev *pdev,
pf->pdev = pdev;
pci_set_drvdata(pdev, pf);
set_bit(__ICE_DOWN, pf->state);
+ /* Disable service task until DOWN bit is cleared */
+ set_bit(__ICE_SERVICE_DIS, pf->state);
hw = &pf->hw;
hw->hw_addr = pcim_iomap_table(pdev)[ICE_BAR0];
@@ -3658,6 +3673,9 @@ static int ice_probe(struct pci_dev *pdev,
goto err_init_interrupt_unroll;
}
+ /* Driver is mostly up */
+ clear_bit(__ICE_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
@@ -3697,8 +3715,7 @@ static int ice_probe(struct pci_dev *pdev,
goto err_alloc_sw_unroll;
}
- /* Driver is mostly up */
- clear_bit(__ICE_DOWN, pf->state);
+ clear_bit(__ICE_SERVICE_DIS, pf->state);
/* since everything is good, start the service timer */
mod_timer(&pf->serv_tmr, round_jiffies(jiffies + pf->serv_tmr_period));
@@ -3712,6 +3729,7 @@ static int ice_probe(struct pci_dev *pdev,
return 0;
err_alloc_sw_unroll:
+ set_bit(__ICE_SERVICE_DIS, pf->state);
set_bit(__ICE_DOWN, pf->state);
devm_kfree(&pf->pdev->dev, pf->first_sw);
err_msix_misc_unroll:
@@ -3739,6 +3757,7 @@ static void ice_remove(struct pci_dev *pdev)
return;
set_bit(__ICE_DOWN, pf->state);
+ ice_service_task_stop(pf);
ice_vsi_release_all(pf);
ice_free_irq_msix_misc(pf);
@@ -6003,6 +6022,7 @@ static void ice_tx_timeout(struct net_device *netdev)
netdev_err(netdev, "tx_timeout recovery unsuccessful, device is in unrecoverable state.\n");
set_bit(__ICE_DOWN, pf->state);
set_bit(__ICE_NEEDS_RESTART, vsi->state);
+ set_bit(__ICE_SERVICE_DIS, pf->state);
break;
}
--
2.14.3
^ permalink raw reply related
* Re: [Qemu-devel] [PATCH v1 0/5] riscv: Fix introspection problems
From: Alistair Francis @ 2018-07-23 23:13 UTC (permalink / raw)
To: Thomas Huth
Cc: Alistair Francis, qemu-devel@nongnu.org Developers, Michael Clark,
Markus Armbruster
In-Reply-To: <d10ee58e-2b2b-7064-ae13-c07ba2040a07@redhat.com>
On Thu, Jul 19, 2018 at 12:59 AM, Thomas Huth <thuth@redhat.com> wrote:
> On 18.07.2018 23:31, Alistair Francis wrote:
>> On Wed, Jul 18, 2018 at 12:22 AM, Thomas Huth <thuth@redhat.com> wrote:
>>> On 17.07.2018 22:27, Alistair Francis wrote:
>>>> This is based on Thomas's work fixing introspection problems [1] and
>>>> applied to the RISC-V port.
>>>>
>>>> 1: https://lists.gnu.org/archive/html/qemu-devel/2018-07/msg03261.html
>>>>
>>>> Alistair Francis (5):
>>>> sifive_e: Fix crash when introspecting the device
>>>> sifive_u: Fix crash when introspecting the device
>>>> virt: Fix crash when introspecting the device
>>>> riscv_hart: Fix crash when introspecting the device
>>>> spike: Fix crash when introspecting the device
>>>>
>>>> hw/riscv/riscv_hart.c | 7 +++----
>>>> hw/riscv/sifive_e.c | 12 ++++++------
>>>> hw/riscv/sifive_u.c | 15 +++++++--------
>>>> hw/riscv/spike.c | 10 ++++------
>>>> hw/riscv/virt.c | 5 ++---
>>>> 5 files changed, 22 insertions(+), 27 deletions(-)
>>>
>>> That's interesting, these issues did not appear in my tests (modified
>>> tests/device-introspect-test.c with hmp("info qtree")). Likely because
>>> device-introspect-test only checks with the "none" machine ==> one more
>>> reason to test here with all machines, too (I suggested such a patch a
>>> couple of months ago) ...
>>
>> What happened to the patch? More tests would be super handy.
>
> It was only a RFC since at that point in time, there were lots of other
> crashes, so it could not be merged yet:
>
> http://lists.gnu.org/archive/html/qemu-devel/2018-03/msg05033.html
>
> I think the other crashes are now fixed, too, so we could consider this
> again. I'll put it on my TODO list to respin it for 3.1 ...
It looks super useful! I'll use this to test the RISC-V stuff a bit
more as well.
Alistair
>
> Thomas
^ permalink raw reply
* Re: [PATCH] mm: thp: remove use_zero_page sysfs knob
From: David Rientjes @ 2018-07-23 23:14 UTC (permalink / raw)
To: Yang Shi
Cc: Kirill A. Shutemov, hughd, aaron.lu, akpm, linux-mm, linux-kernel
In-Reply-To: <91caed46-6437-a137-0dbc-dadd113f8d58@linux.alibaba.com>
On Mon, 23 Jul 2018, Yang Shi wrote:
> > > I agree to keep it for a while to let that security bug cool down,
> > > however, if
> > > there is no user anymore, it sounds pointless to still keep a dead knob.
> > >
> > It's not a dead knob. We use it, and for reasons other than
> > CVE-2017-1000405. To mitigate the cost of constantly compacting memory to
> > allocate it after it has been freed due to memry pressure, we can either
> > continue to disable it, allow it to be persistently available, or use a
> > new value for use_zero_page to specify it should be persistently
> > available.
>
> My understanding is the cost of memory compaction is *not* unique for huge
> zero page, right? It is expected when memory pressure is met, even though huge
> zero page is disabled.
>
It's caused by fragmentation, not necessarily memory pressure. We've
disabled it because compacting for tens of thousands of huge zero pages in
the background has a noticeable impact on cpu. Additionally, if the hzp
cannot be allocated at runtime it increases the rss of applications that
map it, making it unpredictable. Making it persistent, as I've been
suggesting, fixes these issues.
^ permalink raw reply
* Re: [PATCH] drivers/pci/probe: Move variable bridge inside ifdef
From: Bjorn Helgaas @ 2018-07-23 22:11 UTC (permalink / raw)
To: Anders Roxell; +Cc: bhelgaas, linux-pci, linux-kernel
In-Reply-To: <20180721214556.25651-1-anders.roxell@linaro.org>
On Sat, Jul 21, 2018 at 11:45:56PM +0200, Anders Roxell wrote:
> When CONFIG_PCI_QUIRKS isn't enabled we get the warning below:
> drivers/pci/probe.c: In function ‘pci_bus_read_dev_vendor_id’:
> drivers/pci/probe.c:2221:18: warning: unused variable ‘bridge’ [-Wunused-variable]
> struct pci_dev *bridge = bus->self;
> ^~~~~~
>
> Move the declaration of variable bridge to inside the ifdef
> CONFIG_PCI_QUIRKS.
>
> Fixes: ac5ea104a279 ("PCI: Workaround IDT switch ACS Source Validation erratum")
> Signed-off-by: Anders Roxell <anders.roxell@linaro.org>
I folded this into the original commit on pci/enumeration, thanks!
> ---
> drivers/pci/probe.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c
> index 1c581346c5b9..7a5323798312 100644
> --- a/drivers/pci/probe.c
> +++ b/drivers/pci/probe.c
> @@ -2218,9 +2218,9 @@ bool pci_bus_generic_read_dev_vendor_id(struct pci_bus *bus, int devfn, u32 *l,
> bool pci_bus_read_dev_vendor_id(struct pci_bus *bus, int devfn, u32 *l,
> int timeout)
> {
> +#ifdef CONFIG_PCI_QUIRKS
> struct pci_dev *bridge = bus->self;
>
> -#ifdef CONFIG_PCI_QUIRKS
> /*
> * Certain IDT switches have an issue where they improperly trigger
> * ACS Source Validation errors on completions for config reads.
> --
> 2.18.0
>
^ permalink raw reply
* ✓ Fi.CI.IGT: success for series starting with [1/2] drm/dp: add extended receiver capability field present bit
From: Patchwork @ 2018-07-23 23:14 UTC (permalink / raw)
To: matthew.s.atwood; +Cc: intel-gfx
In-Reply-To: <20180723212735.23893-1-matthew.s.atwood@intel.com>
== Series Details ==
Series: series starting with [1/2] drm/dp: add extended receiver capability field present bit
URL : https://patchwork.freedesktop.org/series/47087/
State : success
== Summary ==
= CI Bug Log - changes from CI_DRM_4530_full -> Patchwork_9753_full =
== Summary - WARNING ==
Minor unknown changes coming with Patchwork_9753_full need to be verified
manually.
If you think the reported changes have nothing to do with the changes
introduced in Patchwork_9753_full, please notify your bug team to allow them
to document this new failure mode, which will reduce false positives in CI.
== Possible new issues ==
Here are the unknown changes that may have been introduced in Patchwork_9753_full:
=== IGT changes ===
==== Warnings ====
igt@gem_exec_schedule@deep-bsd2:
shard-kbl: PASS -> SKIP +1
== Known issues ==
Here are the changes found in Patchwork_9753_full that come from known issues:
=== IGT changes ===
==== Issues hit ====
igt@kms_flip@plain-flip-fb-recreate:
shard-hsw: PASS -> FAIL (fdo#100368)
==== Possible fixes ====
igt@kms_flip@2x-dpms-vs-vblank-race-interruptible:
shard-hsw: FAIL (fdo#103060) -> PASS
igt@kms_flip@2x-flip-vs-wf_vblank:
shard-glk: FAIL (fdo#100368) -> PASS
igt@perf_pmu@rc6-runtime-pm-long:
shard-kbl: FAIL (fdo#105010) -> PASS
fdo#100368 https://bugs.freedesktop.org/show_bug.cgi?id=100368
fdo#103060 https://bugs.freedesktop.org/show_bug.cgi?id=103060
fdo#105010 https://bugs.freedesktop.org/show_bug.cgi?id=105010
== Participating hosts (5 -> 5) ==
No changes in participating hosts
== Build changes ==
* Linux: CI_DRM_4530 -> Patchwork_9753
CI_DRM_4530: d27cc4a37a5cc1ef14a3aafdcb6682e5f6a85d09 @ git://anongit.freedesktop.org/gfx-ci/linux
IGT_4571: 65fccc149b85968cdce4737266b056059c1510f3 @ git://anongit.freedesktop.org/xorg/app/intel-gpu-tools
Patchwork_9753: 23a1b794202633961ad336e9f57c1ba6cb184d93 @ git://anongit.freedesktop.org/gfx-ci/linux
piglit_4509: fdc5a4ca11124ab8413c7988896eec4c97336694 @ git://anongit.freedesktop.org/piglit
== Logs ==
For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_9753/shards.html
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
^ permalink raw reply
* [pull request][net-next V2 00/12] Mellanox, mlx5e updates 2018-07-18
From: Saeed Mahameed @ 2018-07-23 22:11 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Saeed Mahameed
Hi Dave,
This series includes updates for mlx5e net device driver, with a couple
of major features and some misc updates.
Please notice the mlx5-next merge patch at the beginning:
"Merge branch 'mlx5-next' of git://git.kernel.org/pub/scm/linux/kernel/git/mellanox/linux"
For more information please see tag log below.
Please pull and let me know if there's any problem.
v1->v2:
- Dropped "Support PCIe buffer congestion handling via Devlink" patches until the
comments are addressed.
Thanks,
Saeed.
---
The following changes since commit 7854ac44fe86548f8a6c6001938a1a2593b255e4:
Merge branch 'mlx5-next' of git://git.kernel.org/pub/scm/linux/kernel/git/mellanox/linux (2018-07-23 14:58:46 -0700)
are available in the Git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux.git tags/mlx5e-updates-2018-07-18-v2
for you to fetch changes up to 3f44899ef2ce0c9da49feb0d6f08098a08cb96ae:
net/mlx5e: Use PARTIAL_GSO for UDP segmentation (2018-07-23 15:01:11 -0700)
----------------------------------------------------------------
mlx5e-updates-2018-07-18
This series includes update for mlx5e net device driver.
1) From Feras Daoud, Added the support for firmware log tracing,
first by introducing the firmware API needed for the task and then
For each PF do the following:
1- Allocate memory for the tracer strings database and read it from the FW to the SW.
2- Allocate and dma map tracer buffers.
Traces that will be written into the buffer will be parsed as a group
of one or more traces, referred to as trace message. The trace message
represents a C-like printf string.
Once a new trace is available FW will generate an event indicates new trace/s are
available and the driver will parse them and dump them using tracepoints
event tracing
Enable mlx5 fw tracing by:
echo 1 > /sys/kernel/debug/tracing/events/mlx5/mlx5_fw/enable
Read traces by:
cat /sys/kernel/debug/tracing/trace
2) From Roi Dayan, Remove redundant WARN when we cannot find neigh entry
3) From Jianbo Liu, TC double vlan support
- Support offloading tc double vlan headers match
- Support offloading double vlan push/pop tc actions
4) From Boris, re-visit UDP GSO, remove the splitting of UDP_GSO_L4 packets
in the driver, and exposes UDP_GSO_L4 as a PARTIAL_GSO feature.
----------------------------------------------------------------
Boris Pismenny (1):
net/mlx5e: Use PARTIAL_GSO for UDP segmentation
Feras Daoud (5):
net/mlx5: FW tracer, implement tracer logic
net/mlx5: FW tracer, create trace buffer and copy strings database
net/mlx5: FW tracer, events handling
net/mlx5: FW tracer, parse traces and kernel tracing support
net/mlx5: FW tracer, Enable tracing
Jianbo Liu (3):
net/mlx5e: Support offloading tc double vlan headers match
net/mlx5e: Refactor tc vlan push/pop actions offloading
net/mlx5e: Support offloading double vlan push/pop tc actions
Roi Dayan (1):
net/mlx5e: Remove redundant WARN when we cannot find neigh entry
Saeed Mahameed (2):
net/mlx5: FW tracer, register log buffer memory key
net/mlx5: FW tracer, Add debug prints
drivers/net/ethernet/mellanox/mlx5/core/Makefile | 6 +-
.../ethernet/mellanox/mlx5/core/diag/fw_tracer.c | 947 +++++++++++++++++++++
.../ethernet/mellanox/mlx5/core/diag/fw_tracer.h | 175 ++++
.../mellanox/mlx5/core/diag/fw_tracer_tracepoint.h | 78 ++
.../mellanox/mlx5/core/en_accel/en_accel.h | 27 +-
.../ethernet/mellanox/mlx5/core/en_accel/rxtx.c | 109 ---
.../ethernet/mellanox/mlx5/core/en_accel/rxtx.h | 14 -
drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 9 +-
drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 134 ++-
drivers/net/ethernet/mellanox/mlx5/core/eq.c | 11 +
drivers/net/ethernet/mellanox/mlx5/core/eswitch.h | 21 +-
.../ethernet/mellanox/mlx5/core/eswitch_offloads.c | 23 +-
drivers/net/ethernet/mellanox/mlx5/core/main.c | 18 +-
include/linux/mlx5/device.h | 7 +
include/linux/mlx5/driver.h | 3 +
15 files changed, 1399 insertions(+), 183 deletions(-)
create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h
create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer_tracepoint.h
delete mode 100644 drivers/net/ethernet/mellanox/mlx5/core/en_accel/rxtx.c
delete mode 100644 drivers/net/ethernet/mellanox/mlx5/core/en_accel/rxtx.h
^ permalink raw reply
* [net-next V2 01/12] net/mlx5: FW tracer, implement tracer logic
From: Saeed Mahameed @ 2018-07-23 22:11 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Feras Daoud, Saeed Mahameed
In-Reply-To: <20180723221129.21625-1-saeedm@mellanox.com>
From: Feras Daoud <ferasda@mellanox.com>
Implement FW tracer logic and registers access, initialization and
cleanup flows.
Initializing the tracer will be part of load one flow, as multiple
PFs will try to acquire ownership but only one will succeed and will
be the tracer owner.
Signed-off-by: Feras Daoud <ferasda@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../mellanox/mlx5/core/diag/fw_tracer.c | 196 ++++++++++++++++++
.../mellanox/mlx5/core/diag/fw_tracer.h | 66 ++++++
include/linux/mlx5/driver.h | 3 +
3 files changed, 265 insertions(+)
create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
new file mode 100644
index 000000000000..3ecbf06b4d71
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
@@ -0,0 +1,196 @@
+/*
+ * Copyright (c) 2018, Mellanox Technologies. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses. You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ * Redistribution and use in source and binary forms, with or
+ * without modification, are permitted provided that the following
+ * conditions are met:
+ *
+ * - Redistributions of source code must retain the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer.
+ *
+ * - Redistributions in binary form must reproduce the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer in the documentation and/or other materials
+ * provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#include "fw_tracer.h"
+
+static int mlx5_query_mtrc_caps(struct mlx5_fw_tracer *tracer)
+{
+ u32 *string_db_base_address_out = tracer->str_db.base_address_out;
+ u32 *string_db_size_out = tracer->str_db.size_out;
+ struct mlx5_core_dev *dev = tracer->dev;
+ u32 out[MLX5_ST_SZ_DW(mtrc_cap)] = {0};
+ u32 in[MLX5_ST_SZ_DW(mtrc_cap)] = {0};
+ void *mtrc_cap_sp;
+ int err, i;
+
+ err = mlx5_core_access_reg(dev, in, sizeof(in), out, sizeof(out),
+ MLX5_REG_MTRC_CAP, 0, 0);
+ if (err) {
+ mlx5_core_warn(dev, "FWTracer: Error reading tracer caps %d\n",
+ err);
+ return err;
+ }
+
+ if (!MLX5_GET(mtrc_cap, out, trace_to_memory)) {
+ mlx5_core_dbg(dev, "FWTracer: Device does not support logging traces to memory\n");
+ return -ENOTSUPP;
+ }
+
+ tracer->trc_ver = MLX5_GET(mtrc_cap, out, trc_ver);
+ tracer->str_db.first_string_trace =
+ MLX5_GET(mtrc_cap, out, first_string_trace);
+ tracer->str_db.num_string_trace =
+ MLX5_GET(mtrc_cap, out, num_string_trace);
+ tracer->str_db.num_string_db = MLX5_GET(mtrc_cap, out, num_string_db);
+ tracer->owner = !!MLX5_GET(mtrc_cap, out, trace_owner);
+
+ for (i = 0; i < tracer->str_db.num_string_db; i++) {
+ mtrc_cap_sp = MLX5_ADDR_OF(mtrc_cap, out, string_db_param[i]);
+ string_db_base_address_out[i] = MLX5_GET(mtrc_string_db_param,
+ mtrc_cap_sp,
+ string_db_base_address);
+ string_db_size_out[i] = MLX5_GET(mtrc_string_db_param,
+ mtrc_cap_sp, string_db_size);
+ }
+
+ return err;
+}
+
+static int mlx5_set_mtrc_caps_trace_owner(struct mlx5_fw_tracer *tracer,
+ u32 *out, u32 out_size,
+ u8 trace_owner)
+{
+ struct mlx5_core_dev *dev = tracer->dev;
+ u32 in[MLX5_ST_SZ_DW(mtrc_cap)] = {0};
+
+ MLX5_SET(mtrc_cap, in, trace_owner, trace_owner);
+
+ return mlx5_core_access_reg(dev, in, sizeof(in), out, out_size,
+ MLX5_REG_MTRC_CAP, 0, 1);
+}
+
+static int mlx5_fw_tracer_ownership_acquire(struct mlx5_fw_tracer *tracer)
+{
+ struct mlx5_core_dev *dev = tracer->dev;
+ u32 out[MLX5_ST_SZ_DW(mtrc_cap)] = {0};
+ int err;
+
+ err = mlx5_set_mtrc_caps_trace_owner(tracer, out, sizeof(out),
+ MLX5_FW_TRACER_ACQUIRE_OWNERSHIP);
+ if (err) {
+ mlx5_core_warn(dev, "FWTracer: Acquire tracer ownership failed %d\n",
+ err);
+ return err;
+ }
+
+ tracer->owner = !!MLX5_GET(mtrc_cap, out, trace_owner);
+
+ if (!tracer->owner)
+ return -EBUSY;
+
+ return 0;
+}
+
+static void mlx5_fw_tracer_ownership_release(struct mlx5_fw_tracer *tracer)
+{
+ u32 out[MLX5_ST_SZ_DW(mtrc_cap)] = {0};
+
+ mlx5_set_mtrc_caps_trace_owner(tracer, out, sizeof(out),
+ MLX5_FW_TRACER_RELEASE_OWNERSHIP);
+ tracer->owner = false;
+}
+
+static void mlx5_fw_tracer_ownership_change(struct work_struct *work)
+{
+ struct mlx5_fw_tracer *tracer = container_of(work, struct mlx5_fw_tracer,
+ ownership_change_work);
+ struct mlx5_core_dev *dev = tracer->dev;
+ int err;
+
+ if (tracer->owner) {
+ mlx5_fw_tracer_ownership_release(tracer);
+ return;
+ }
+
+ err = mlx5_fw_tracer_ownership_acquire(tracer);
+ if (err) {
+ mlx5_core_dbg(dev, "FWTracer: Ownership was not granted %d\n", err);
+ return;
+ }
+}
+
+struct mlx5_fw_tracer *mlx5_fw_tracer_create(struct mlx5_core_dev *dev)
+{
+ struct mlx5_fw_tracer *tracer = NULL;
+ int err;
+
+ if (!MLX5_CAP_MCAM_REG(dev, tracer_registers)) {
+ mlx5_core_dbg(dev, "FWTracer: Tracer capability not present\n");
+ return NULL;
+ }
+
+ tracer = kzalloc(sizeof(*tracer), GFP_KERNEL);
+ if (!tracer)
+ return ERR_PTR(-ENOMEM);
+
+ tracer->work_queue = create_singlethread_workqueue("mlx5_fw_tracer");
+ if (!tracer->work_queue) {
+ err = -ENOMEM;
+ goto free_tracer;
+ }
+
+ tracer->dev = dev;
+
+ INIT_WORK(&tracer->ownership_change_work, mlx5_fw_tracer_ownership_change);
+
+ err = mlx5_query_mtrc_caps(tracer);
+ if (err) {
+ mlx5_core_dbg(dev, "FWTracer: Failed to query capabilities %d\n", err);
+ goto destroy_workqueue;
+ }
+
+ mlx5_fw_tracer_ownership_change(&tracer->ownership_change_work);
+
+ return tracer;
+
+destroy_workqueue:
+ tracer->dev = NULL;
+ destroy_workqueue(tracer->work_queue);
+free_tracer:
+ kfree(tracer);
+ return ERR_PTR(err);
+}
+
+void mlx5_fw_tracer_destroy(struct mlx5_fw_tracer *tracer)
+{
+ if (!tracer)
+ return;
+
+ cancel_work_sync(&tracer->ownership_change_work);
+
+ if (tracer->owner)
+ mlx5_fw_tracer_ownership_release(tracer);
+
+ flush_workqueue(tracer->work_queue);
+ destroy_workqueue(tracer->work_queue);
+ kfree(tracer);
+}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h
new file mode 100644
index 000000000000..721c41a5e827
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h
@@ -0,0 +1,66 @@
+/*
+ * Copyright (c) 2018, Mellanox Technologies. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses. You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ * Redistribution and use in source and binary forms, with or
+ * without modification, are permitted provided that the following
+ * conditions are met:
+ *
+ * - Redistributions of source code must retain the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer.
+ *
+ * - Redistributions in binary form must reproduce the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer in the documentation and/or other materials
+ * provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#ifndef __LIB_TRACER_H__
+#define __LIB_TRACER_H__
+
+#include <linux/mlx5/driver.h>
+#include "mlx5_core.h"
+
+#define STRINGS_DB_SECTIONS_NUM 8
+
+struct mlx5_fw_tracer {
+ struct mlx5_core_dev *dev;
+ bool owner;
+ u8 trc_ver;
+ struct workqueue_struct *work_queue;
+ struct work_struct ownership_change_work;
+
+ /* Strings DB */
+ struct {
+ u8 first_string_trace;
+ u8 num_string_trace;
+ u32 num_string_db;
+ u32 base_address_out[STRINGS_DB_SECTIONS_NUM];
+ u32 size_out[STRINGS_DB_SECTIONS_NUM];
+ } str_db;
+};
+
+enum mlx5_fw_tracer_ownership_state {
+ MLX5_FW_TRACER_RELEASE_OWNERSHIP,
+ MLX5_FW_TRACER_ACQUIRE_OWNERSHIP,
+};
+
+struct mlx5_fw_tracer *mlx5_fw_tracer_create(struct mlx5_core_dev *dev);
+void mlx5_fw_tracer_destroy(struct mlx5_fw_tracer *tracer);
+
+#endif
diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h
index 957199c20a0f..86cb0ebf92fa 100644
--- a/include/linux/mlx5/driver.h
+++ b/include/linux/mlx5/driver.h
@@ -816,6 +816,8 @@ struct mlx5_clock {
struct mlx5_pps pps_info;
};
+struct mlx5_fw_tracer;
+
struct mlx5_core_dev {
struct pci_dev *pdev;
/* sync pci state */
@@ -860,6 +862,7 @@ struct mlx5_core_dev {
struct mlx5_clock clock;
struct mlx5_ib_clock_info *clock_info;
struct page *clock_info_page;
+ struct mlx5_fw_tracer *tracer;
};
struct mlx5_db {
--
2.17.0
^ permalink raw reply related
* [net-next V2 03/12] net/mlx5: FW tracer, register log buffer memory key
From: Saeed Mahameed @ 2018-07-23 22:11 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Saeed Mahameed
In-Reply-To: <20180723221129.21625-1-saeedm@mellanox.com>
Create a memory key and protection domain for the tracer log buffer.
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../mellanox/mlx5/core/diag/fw_tracer.c | 64 ++++++++++++++++++-
1 file changed, 61 insertions(+), 3 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
index 35107b8f76df..d6cc27b0ff34 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
@@ -169,6 +169,48 @@ static void mlx5_fw_tracer_destroy_log_buf(struct mlx5_fw_tracer *tracer)
free_pages((unsigned long)tracer->buff.log_buf, get_order(tracer->buff.size));
}
+static int mlx5_fw_tracer_create_mkey(struct mlx5_fw_tracer *tracer)
+{
+ struct mlx5_core_dev *dev = tracer->dev;
+ int err, inlen, i;
+ __be64 *mtt;
+ void *mkc;
+ u32 *in;
+
+ inlen = MLX5_ST_SZ_BYTES(create_mkey_in) +
+ sizeof(*mtt) * round_up(TRACER_BUFFER_PAGE_NUM, 2);
+
+ in = kvzalloc(inlen, GFP_KERNEL);
+ if (!in)
+ return -ENOMEM;
+
+ MLX5_SET(create_mkey_in, in, translations_octword_actual_size,
+ DIV_ROUND_UP(TRACER_BUFFER_PAGE_NUM, 2));
+ mtt = (u64 *)MLX5_ADDR_OF(create_mkey_in, in, klm_pas_mtt);
+ for (i = 0 ; i < TRACER_BUFFER_PAGE_NUM ; i++)
+ mtt[i] = cpu_to_be64(tracer->buff.dma + i * PAGE_SIZE);
+
+ mkc = MLX5_ADDR_OF(create_mkey_in, in, memory_key_mkey_entry);
+ MLX5_SET(mkc, mkc, access_mode_1_0, MLX5_MKC_ACCESS_MODE_MTT);
+ MLX5_SET(mkc, mkc, lr, 1);
+ MLX5_SET(mkc, mkc, lw, 1);
+ MLX5_SET(mkc, mkc, pd, tracer->buff.pdn);
+ MLX5_SET(mkc, mkc, bsf_octword_size, 0);
+ MLX5_SET(mkc, mkc, qpn, 0xffffff);
+ MLX5_SET(mkc, mkc, log_page_size, PAGE_SHIFT);
+ MLX5_SET(mkc, mkc, translations_octword_size,
+ DIV_ROUND_UP(TRACER_BUFFER_PAGE_NUM, 2));
+ MLX5_SET64(mkc, mkc, start_addr, tracer->buff.dma);
+ MLX5_SET64(mkc, mkc, len, tracer->buff.size);
+ err = mlx5_core_create_mkey(dev, &tracer->buff.mkey, in, inlen);
+ if (err)
+ mlx5_core_warn(dev, "FWTracer: Failed to create mkey, %d\n", err);
+
+ kvfree(in);
+
+ return err;
+}
+
static void mlx5_fw_tracer_free_strings_db(struct mlx5_fw_tracer *tracer)
{
u32 num_string_db = tracer->str_db.num_string_db;
@@ -363,13 +405,26 @@ int mlx5_fw_tracer_init(struct mlx5_fw_tracer *tracer)
if (!tracer->str_db.loaded)
queue_work(tracer->work_queue, &tracer->read_fw_strings_work);
- err = mlx5_fw_tracer_ownership_acquire(tracer);
+ err = mlx5_core_alloc_pd(dev, &tracer->buff.pdn);
if (err) {
- mlx5_core_dbg(dev, "FWTracer: Ownership was not granted %d\n", err);
- return 0; /* return 0 since ownership can be acquired on a later FW event */
+ mlx5_core_warn(dev, "FWTracer: Failed to allocate PD %d\n", err);
+ return err;
}
+ err = mlx5_fw_tracer_create_mkey(tracer);
+ if (err) {
+ mlx5_core_warn(dev, "FWTracer: Failed to create mkey %d\n", err);
+ goto err_dealloc_pd;
+ }
+
+ err = mlx5_fw_tracer_ownership_acquire(tracer);
+ if (err) /* Don't fail since ownership can be acquired on a later FW event */
+ mlx5_core_dbg(dev, "FWTracer: Ownership was not granted %d\n", err);
+
return 0;
+err_dealloc_pd:
+ mlx5_core_dealloc_pd(dev, tracer->buff.pdn);
+ return err;
}
void mlx5_fw_tracer_cleanup(struct mlx5_fw_tracer *tracer)
@@ -381,6 +436,9 @@ void mlx5_fw_tracer_cleanup(struct mlx5_fw_tracer *tracer)
if (tracer->owner)
mlx5_fw_tracer_ownership_release(tracer);
+
+ mlx5_core_destroy_mkey(tracer->dev, &tracer->buff.mkey);
+ mlx5_core_dealloc_pd(tracer->dev, tracer->buff.pdn);
}
void mlx5_fw_tracer_destroy(struct mlx5_fw_tracer *tracer)
--
2.17.0
^ permalink raw reply related
* [net-next V2 02/12] net/mlx5: FW tracer, create trace buffer and copy strings database
From: Saeed Mahameed @ 2018-07-23 22:11 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Feras Daoud, Saeed Mahameed
In-Reply-To: <20180723221129.21625-1-saeedm@mellanox.com>
From: Feras Daoud <ferasda@mellanox.com>
For each PF do the following:
1- Allocate memory for the tracer strings database and read the
strings from the FW to the SW. These strings will be used later for
parsing traces.
2- Allocate and dma map tracer buffers.
Traces that will be written into the buffer will be parsed as a group
of one or more traces, referred to as trace message. The trace message
represents a C-like printf string.
First trace of a message holds the pointer to the correct string in
strings database. The following traces holds the variables of the
message.
Signed-off-by: Feras Daoud <ferasda@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../mellanox/mlx5/core/diag/fw_tracer.c | 209 +++++++++++++++++-
.../mellanox/mlx5/core/diag/fw_tracer.h | 18 ++
2 files changed, 224 insertions(+), 3 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
index 3ecbf06b4d71..35107b8f76df 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
@@ -119,6 +119,163 @@ static void mlx5_fw_tracer_ownership_release(struct mlx5_fw_tracer *tracer)
tracer->owner = false;
}
+static int mlx5_fw_tracer_create_log_buf(struct mlx5_fw_tracer *tracer)
+{
+ struct mlx5_core_dev *dev = tracer->dev;
+ struct device *ddev = &dev->pdev->dev;
+ dma_addr_t dma;
+ void *buff;
+ gfp_t gfp;
+ int err;
+
+ tracer->buff.size = TRACE_BUFFER_SIZE_BYTE;
+
+ gfp = GFP_KERNEL | __GFP_ZERO;
+ buff = (void *)__get_free_pages(gfp,
+ get_order(tracer->buff.size));
+ if (!buff) {
+ err = -ENOMEM;
+ mlx5_core_warn(dev, "FWTracer: Failed to allocate pages, %d\n", err);
+ return err;
+ }
+ tracer->buff.log_buf = buff;
+
+ dma = dma_map_single(ddev, buff, tracer->buff.size, DMA_FROM_DEVICE);
+ if (dma_mapping_error(ddev, dma)) {
+ mlx5_core_warn(dev, "FWTracer: Unable to map DMA: %d\n",
+ dma_mapping_error(ddev, dma));
+ err = -ENOMEM;
+ goto free_pages;
+ }
+ tracer->buff.dma = dma;
+
+ return 0;
+
+free_pages:
+ free_pages((unsigned long)tracer->buff.log_buf, get_order(tracer->buff.size));
+
+ return err;
+}
+
+static void mlx5_fw_tracer_destroy_log_buf(struct mlx5_fw_tracer *tracer)
+{
+ struct mlx5_core_dev *dev = tracer->dev;
+ struct device *ddev = &dev->pdev->dev;
+
+ if (!tracer->buff.log_buf)
+ return;
+
+ dma_unmap_single(ddev, tracer->buff.dma, tracer->buff.size, DMA_FROM_DEVICE);
+ free_pages((unsigned long)tracer->buff.log_buf, get_order(tracer->buff.size));
+}
+
+static void mlx5_fw_tracer_free_strings_db(struct mlx5_fw_tracer *tracer)
+{
+ u32 num_string_db = tracer->str_db.num_string_db;
+ int i;
+
+ for (i = 0; i < num_string_db; i++) {
+ kfree(tracer->str_db.buffer[i]);
+ tracer->str_db.buffer[i] = NULL;
+ }
+}
+
+static int mlx5_fw_tracer_allocate_strings_db(struct mlx5_fw_tracer *tracer)
+{
+ u32 *string_db_size_out = tracer->str_db.size_out;
+ u32 num_string_db = tracer->str_db.num_string_db;
+ int i;
+
+ for (i = 0; i < num_string_db; i++) {
+ tracer->str_db.buffer[i] = kzalloc(string_db_size_out[i], GFP_KERNEL);
+ if (!tracer->str_db.buffer[i])
+ goto free_strings_db;
+ }
+
+ return 0;
+
+free_strings_db:
+ mlx5_fw_tracer_free_strings_db(tracer);
+ return -ENOMEM;
+}
+
+static void mlx5_tracer_read_strings_db(struct work_struct *work)
+{
+ struct mlx5_fw_tracer *tracer = container_of(work, struct mlx5_fw_tracer,
+ read_fw_strings_work);
+ u32 num_of_reads, num_string_db = tracer->str_db.num_string_db;
+ struct mlx5_core_dev *dev = tracer->dev;
+ u32 in[MLX5_ST_SZ_DW(mtrc_cap)] = {0};
+ u32 leftovers, offset;
+ int err = 0, i, j;
+ u32 *out, outlen;
+ void *out_value;
+
+ outlen = MLX5_ST_SZ_BYTES(mtrc_stdb) + STRINGS_DB_READ_SIZE_BYTES;
+ out = kzalloc(outlen, GFP_KERNEL);
+ if (!out) {
+ err = -ENOMEM;
+ goto out;
+ }
+
+ for (i = 0; i < num_string_db; i++) {
+ offset = 0;
+ MLX5_SET(mtrc_stdb, in, string_db_index, i);
+ num_of_reads = tracer->str_db.size_out[i] /
+ STRINGS_DB_READ_SIZE_BYTES;
+ leftovers = (tracer->str_db.size_out[i] %
+ STRINGS_DB_READ_SIZE_BYTES) /
+ STRINGS_DB_LEFTOVER_SIZE_BYTES;
+
+ MLX5_SET(mtrc_stdb, in, read_size, STRINGS_DB_READ_SIZE_BYTES);
+ for (j = 0; j < num_of_reads; j++) {
+ MLX5_SET(mtrc_stdb, in, start_offset, offset);
+
+ err = mlx5_core_access_reg(dev, in, sizeof(in), out,
+ outlen, MLX5_REG_MTRC_STDB,
+ 0, 1);
+ if (err) {
+ mlx5_core_dbg(dev, "FWTracer: Failed to read strings DB %d\n",
+ err);
+ goto out_free;
+ }
+
+ out_value = MLX5_ADDR_OF(mtrc_stdb, out, string_db_data);
+ memcpy(tracer->str_db.buffer[i] + offset, out_value,
+ STRINGS_DB_READ_SIZE_BYTES);
+ offset += STRINGS_DB_READ_SIZE_BYTES;
+ }
+
+ /* Strings database is aligned to 64, need to read leftovers*/
+ MLX5_SET(mtrc_stdb, in, read_size,
+ STRINGS_DB_LEFTOVER_SIZE_BYTES);
+ for (j = 0; j < leftovers; j++) {
+ MLX5_SET(mtrc_stdb, in, start_offset, offset);
+
+ err = mlx5_core_access_reg(dev, in, sizeof(in), out,
+ outlen, MLX5_REG_MTRC_STDB,
+ 0, 1);
+ if (err) {
+ mlx5_core_dbg(dev, "FWTracer: Failed to read strings DB %d\n",
+ err);
+ goto out_free;
+ }
+
+ out_value = MLX5_ADDR_OF(mtrc_stdb, out, string_db_data);
+ memcpy(tracer->str_db.buffer[i] + offset, out_value,
+ STRINGS_DB_LEFTOVER_SIZE_BYTES);
+ offset += STRINGS_DB_LEFTOVER_SIZE_BYTES;
+ }
+ }
+
+ tracer->str_db.loaded = true;
+
+out_free:
+ kfree(out);
+out:
+ return;
+}
+
static void mlx5_fw_tracer_ownership_change(struct work_struct *work)
{
struct mlx5_fw_tracer *tracer = container_of(work, struct mlx5_fw_tracer,
@@ -161,6 +318,7 @@ struct mlx5_fw_tracer *mlx5_fw_tracer_create(struct mlx5_core_dev *dev)
tracer->dev = dev;
INIT_WORK(&tracer->ownership_change_work, mlx5_fw_tracer_ownership_change);
+ INIT_WORK(&tracer->read_fw_strings_work, mlx5_tracer_read_strings_db);
err = mlx5_query_mtrc_caps(tracer);
if (err) {
@@ -168,10 +326,22 @@ struct mlx5_fw_tracer *mlx5_fw_tracer_create(struct mlx5_core_dev *dev)
goto destroy_workqueue;
}
- mlx5_fw_tracer_ownership_change(&tracer->ownership_change_work);
+ err = mlx5_fw_tracer_create_log_buf(tracer);
+ if (err) {
+ mlx5_core_warn(dev, "FWTracer: Create log buffer failed %d\n", err);
+ goto destroy_workqueue;
+ }
+
+ err = mlx5_fw_tracer_allocate_strings_db(tracer);
+ if (err) {
+ mlx5_core_warn(dev, "FWTracer: Allocate strings database failed %d\n", err);
+ goto free_log_buf;
+ }
return tracer;
+free_log_buf:
+ mlx5_fw_tracer_destroy_log_buf(tracer);
destroy_workqueue:
tracer->dev = NULL;
destroy_workqueue(tracer->work_queue);
@@ -180,17 +350,50 @@ struct mlx5_fw_tracer *mlx5_fw_tracer_create(struct mlx5_core_dev *dev)
return ERR_PTR(err);
}
-void mlx5_fw_tracer_destroy(struct mlx5_fw_tracer *tracer)
+int mlx5_fw_tracer_init(struct mlx5_fw_tracer *tracer)
{
- if (!tracer)
+ struct mlx5_core_dev *dev;
+ int err;
+
+ if (IS_ERR_OR_NULL(tracer))
+ return 0;
+
+ dev = tracer->dev;
+
+ if (!tracer->str_db.loaded)
+ queue_work(tracer->work_queue, &tracer->read_fw_strings_work);
+
+ err = mlx5_fw_tracer_ownership_acquire(tracer);
+ if (err) {
+ mlx5_core_dbg(dev, "FWTracer: Ownership was not granted %d\n", err);
+ return 0; /* return 0 since ownership can be acquired on a later FW event */
+ }
+
+ return 0;
+}
+
+void mlx5_fw_tracer_cleanup(struct mlx5_fw_tracer *tracer)
+{
+ if (IS_ERR_OR_NULL(tracer))
return;
cancel_work_sync(&tracer->ownership_change_work);
if (tracer->owner)
mlx5_fw_tracer_ownership_release(tracer);
+}
+void mlx5_fw_tracer_destroy(struct mlx5_fw_tracer *tracer)
+{
+ if (IS_ERR_OR_NULL(tracer))
+ return;
+
+ cancel_work_sync(&tracer->read_fw_strings_work);
+ mlx5_fw_tracer_free_strings_db(tracer);
+ mlx5_fw_tracer_destroy_log_buf(tracer);
flush_workqueue(tracer->work_queue);
destroy_workqueue(tracer->work_queue);
kfree(tracer);
}
+
+EXPORT_TRACEPOINT_SYMBOL(mlx5_fw);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h
index 721c41a5e827..66cb7e7ada28 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h
@@ -37,6 +37,11 @@
#include "mlx5_core.h"
#define STRINGS_DB_SECTIONS_NUM 8
+#define STRINGS_DB_READ_SIZE_BYTES 256
+#define STRINGS_DB_LEFTOVER_SIZE_BYTES 64
+#define TRACER_BUFFER_PAGE_NUM 64
+#define TRACER_BUFFER_CHUNK 4096
+#define TRACE_BUFFER_SIZE_BYTE (TRACER_BUFFER_PAGE_NUM * TRACER_BUFFER_CHUNK)
struct mlx5_fw_tracer {
struct mlx5_core_dev *dev;
@@ -44,6 +49,7 @@ struct mlx5_fw_tracer {
u8 trc_ver;
struct workqueue_struct *work_queue;
struct work_struct ownership_change_work;
+ struct work_struct read_fw_strings_work;
/* Strings DB */
struct {
@@ -52,7 +58,19 @@ struct mlx5_fw_tracer {
u32 num_string_db;
u32 base_address_out[STRINGS_DB_SECTIONS_NUM];
u32 size_out[STRINGS_DB_SECTIONS_NUM];
+ void *buffer[STRINGS_DB_SECTIONS_NUM];
+ bool loaded;
} str_db;
+
+ /* Log Buffer */
+ struct {
+ u32 pdn;
+ void *log_buf;
+ dma_addr_t dma;
+ u32 size;
+ struct mlx5_core_mkey mkey;
+
+ } buff;
};
enum mlx5_fw_tracer_ownership_state {
--
2.17.0
^ permalink raw reply related
* [net-next V2 05/12] net/mlx5: FW tracer, parse traces and kernel tracing support
From: Saeed Mahameed @ 2018-07-23 22:11 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Feras Daoud, Erez Shitrit, Saeed Mahameed
In-Reply-To: <20180723221129.21625-1-saeedm@mellanox.com>
From: Feras Daoud <ferasda@mellanox.com>
For each message the driver should do the following:
1- Find the message string in the strings database
2- Count the param number of each message
3- Wait for the param events and accumulate them
4- Calculate the event timestamp using the local event timestamp
and the first timestamp event following it.
5- Print message to trace log
Enable the tracing by:
echo 1 > /sys/kernel/debug/tracing/events/mlx5/mlx5_fw/enable
Read traces by:
cat /sys/kernel/debug/tracing/trace
Signed-off-by: Feras Daoud <ferasda@mellanox.com>
Signed-off-by: Erez Shitrit <erezsh@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../mellanox/mlx5/core/diag/fw_tracer.c | 235 +++++++++++++++++-
.../mellanox/mlx5/core/diag/fw_tracer.h | 22 ++
.../mlx5/core/diag/fw_tracer_tracepoint.h | 78 ++++++
3 files changed, 333 insertions(+), 2 deletions(-)
create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer_tracepoint.h
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
index bd887d1d3396..309842de272c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
@@ -29,8 +29,9 @@
* CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*/
-
+#define CREATE_TRACE_POINTS
#include "fw_tracer.h"
+#include "fw_tracer_tracepoint.h"
static int mlx5_query_mtrc_caps(struct mlx5_fw_tracer *tracer)
{
@@ -332,6 +333,109 @@ static void mlx5_fw_tracer_arm(struct mlx5_core_dev *dev)
mlx5_core_warn(dev, "FWTracer: Failed to arm tracer event %d\n", err);
}
+static const char *VAL_PARM = "%llx";
+static const char *REPLACE_64_VAL_PARM = "%x%x";
+static const char *PARAM_CHAR = "%";
+
+static int mlx5_tracer_message_hash(u32 message_id)
+{
+ return jhash_1word(message_id, 0) & (MESSAGE_HASH_SIZE - 1);
+}
+
+static struct tracer_string_format *mlx5_tracer_message_insert(struct mlx5_fw_tracer *tracer,
+ struct tracer_event *tracer_event)
+{
+ struct hlist_head *head =
+ &tracer->hash[mlx5_tracer_message_hash(tracer_event->string_event.tmsn)];
+ struct tracer_string_format *cur_string;
+
+ cur_string = kzalloc(sizeof(*cur_string), GFP_KERNEL);
+ if (!cur_string)
+ return NULL;
+
+ hlist_add_head(&cur_string->hlist, head);
+
+ return cur_string;
+}
+
+static struct tracer_string_format *mlx5_tracer_get_string(struct mlx5_fw_tracer *tracer,
+ struct tracer_event *tracer_event)
+{
+ struct tracer_string_format *cur_string;
+ u32 str_ptr, offset;
+ int i;
+
+ str_ptr = tracer_event->string_event.string_param;
+
+ for (i = 0; i < tracer->str_db.num_string_db; i++) {
+ if (str_ptr > tracer->str_db.base_address_out[i] &&
+ str_ptr < tracer->str_db.base_address_out[i] +
+ tracer->str_db.size_out[i]) {
+ offset = str_ptr - tracer->str_db.base_address_out[i];
+ /* add it to the hash */
+ cur_string = mlx5_tracer_message_insert(tracer, tracer_event);
+ if (!cur_string)
+ return NULL;
+ cur_string->string = (char *)(tracer->str_db.buffer[i] +
+ offset);
+ return cur_string;
+ }
+ }
+
+ return NULL;
+}
+
+static void mlx5_tracer_clean_message(struct tracer_string_format *str_frmt)
+{
+ hlist_del(&str_frmt->hlist);
+ kfree(str_frmt);
+}
+
+static int mlx5_tracer_get_num_of_params(char *str)
+{
+ char *substr, *pstr = str;
+ int num_of_params = 0;
+
+ /* replace %llx with %x%x */
+ substr = strstr(pstr, VAL_PARM);
+ while (substr) {
+ memcpy(substr, REPLACE_64_VAL_PARM, 4);
+ pstr = substr;
+ substr = strstr(pstr, VAL_PARM);
+ }
+
+ /* count all the % characters */
+ substr = strstr(str, PARAM_CHAR);
+ while (substr) {
+ num_of_params += 1;
+ str = substr + 1;
+ substr = strstr(str, PARAM_CHAR);
+ }
+
+ return num_of_params;
+}
+
+static struct tracer_string_format *mlx5_tracer_message_find(struct hlist_head *head,
+ u8 event_id, u32 tmsn)
+{
+ struct tracer_string_format *message;
+
+ hlist_for_each_entry(message, head, hlist)
+ if (message->event_id == event_id && message->tmsn == tmsn)
+ return message;
+
+ return NULL;
+}
+
+static struct tracer_string_format *mlx5_tracer_message_get(struct mlx5_fw_tracer *tracer,
+ struct tracer_event *tracer_event)
+{
+ struct hlist_head *head =
+ &tracer->hash[mlx5_tracer_message_hash(tracer_event->string_event.tmsn)];
+
+ return mlx5_tracer_message_find(head, tracer_event->event_id, tracer_event->string_event.tmsn);
+}
+
static void poll_trace(struct mlx5_fw_tracer *tracer,
struct tracer_event *tracer_event, u64 *trace)
{
@@ -396,6 +500,128 @@ static u64 get_block_timestamp(struct mlx5_fw_tracer *tracer, u64 *ts_event)
return tracer_event.timestamp_event.timestamp;
}
+static void mlx5_fw_tracer_clean_print_hash(struct mlx5_fw_tracer *tracer)
+{
+ struct tracer_string_format *str_frmt;
+ struct hlist_node *n;
+ int i;
+
+ for (i = 0; i < MESSAGE_HASH_SIZE; i++) {
+ hlist_for_each_entry_safe(str_frmt, n, &tracer->hash[i], hlist)
+ mlx5_tracer_clean_message(str_frmt);
+ }
+}
+
+static void mlx5_fw_tracer_clean_ready_list(struct mlx5_fw_tracer *tracer)
+{
+ struct tracer_string_format *str_frmt, *tmp_str;
+
+ list_for_each_entry_safe(str_frmt, tmp_str, &tracer->ready_strings_list,
+ list)
+ list_del(&str_frmt->list);
+}
+
+static void mlx5_tracer_print_trace(struct tracer_string_format *str_frmt,
+ struct mlx5_core_dev *dev,
+ u64 trace_timestamp)
+{
+ char tmp[512];
+
+ snprintf(tmp, sizeof(tmp), str_frmt->string,
+ str_frmt->params[0],
+ str_frmt->params[1],
+ str_frmt->params[2],
+ str_frmt->params[3],
+ str_frmt->params[4],
+ str_frmt->params[5],
+ str_frmt->params[6]);
+
+ trace_mlx5_fw(dev->tracer, trace_timestamp, str_frmt->lost,
+ str_frmt->event_id, tmp);
+
+ /* remove it from hash */
+ mlx5_tracer_clean_message(str_frmt);
+}
+
+static int mlx5_tracer_handle_string_trace(struct mlx5_fw_tracer *tracer,
+ struct tracer_event *tracer_event)
+{
+ struct tracer_string_format *cur_string;
+
+ if (tracer_event->string_event.tdsn == 0) {
+ cur_string = mlx5_tracer_get_string(tracer, tracer_event);
+ if (!cur_string)
+ return -1;
+
+ cur_string->num_of_params = mlx5_tracer_get_num_of_params(cur_string->string);
+ cur_string->last_param_num = 0;
+ cur_string->event_id = tracer_event->event_id;
+ cur_string->tmsn = tracer_event->string_event.tmsn;
+ cur_string->timestamp = tracer_event->string_event.timestamp;
+ cur_string->lost = tracer_event->lost_event;
+ if (cur_string->num_of_params == 0) /* trace with no params */
+ list_add_tail(&cur_string->list, &tracer->ready_strings_list);
+ } else {
+ cur_string = mlx5_tracer_message_get(tracer, tracer_event);
+ if (!cur_string) {
+ pr_debug("%s Got string event for unknown string tdsm: %d\n",
+ __func__, tracer_event->string_event.tmsn);
+ return -1;
+ }
+ cur_string->last_param_num += 1;
+ if (cur_string->last_param_num > TRACER_MAX_PARAMS) {
+ pr_debug("%s Number of params exceeds the max (%d)\n",
+ __func__, TRACER_MAX_PARAMS);
+ list_add_tail(&cur_string->list, &tracer->ready_strings_list);
+ return 0;
+ }
+ /* keep the new parameter */
+ cur_string->params[cur_string->last_param_num - 1] =
+ tracer_event->string_event.string_param;
+ if (cur_string->last_param_num == cur_string->num_of_params)
+ list_add_tail(&cur_string->list, &tracer->ready_strings_list);
+ }
+
+ return 0;
+}
+
+static void mlx5_tracer_handle_timestamp_trace(struct mlx5_fw_tracer *tracer,
+ struct tracer_event *tracer_event)
+{
+ struct tracer_timestamp_event timestamp_event =
+ tracer_event->timestamp_event;
+ struct tracer_string_format *str_frmt, *tmp_str;
+ struct mlx5_core_dev *dev = tracer->dev;
+ u64 trace_timestamp;
+
+ list_for_each_entry_safe(str_frmt, tmp_str, &tracer->ready_strings_list, list) {
+ list_del(&str_frmt->list);
+ if (str_frmt->timestamp < (timestamp_event.timestamp & MASK_6_0))
+ trace_timestamp = (timestamp_event.timestamp & MASK_52_7) |
+ (str_frmt->timestamp & MASK_6_0);
+ else
+ trace_timestamp = ((timestamp_event.timestamp & MASK_52_7) - 1) |
+ (str_frmt->timestamp & MASK_6_0);
+
+ mlx5_tracer_print_trace(str_frmt, dev, trace_timestamp);
+ }
+}
+
+static int mlx5_tracer_handle_trace(struct mlx5_fw_tracer *tracer,
+ struct tracer_event *tracer_event)
+{
+ if (tracer_event->type == TRACER_EVENT_TYPE_STRING) {
+ mlx5_tracer_handle_string_trace(tracer, tracer_event);
+ } else if (tracer_event->type == TRACER_EVENT_TYPE_TIMESTAMP) {
+ if (!tracer_event->timestamp_event.unreliable)
+ mlx5_tracer_handle_timestamp_trace(tracer, tracer_event);
+ } else {
+ pr_debug("%s Got unrecognised type %d for parsing, exiting..\n",
+ __func__, tracer_event->type);
+ }
+ return 0;
+}
+
static void mlx5_fw_tracer_handle_traces(struct work_struct *work)
{
struct mlx5_fw_tracer *tracer =
@@ -452,8 +678,10 @@ static void mlx5_fw_tracer_handle_traces(struct work_struct *work)
}
/* Parse events */
- for (i = 0; i < TRACES_PER_BLOCK ; i++)
+ for (i = 0; i < TRACES_PER_BLOCK ; i++) {
poll_trace(tracer, &tracer_event, &tmp_trace_block[i]);
+ mlx5_tracer_handle_trace(tracer, &tracer_event);
+ }
tracer->buff.consumer_index =
(tracer->buff.consumer_index + 1) & (block_count - 1);
@@ -578,6 +806,7 @@ struct mlx5_fw_tracer *mlx5_fw_tracer_create(struct mlx5_core_dev *dev)
tracer->dev = dev;
+ INIT_LIST_HEAD(&tracer->ready_strings_list);
INIT_WORK(&tracer->ownership_change_work, mlx5_fw_tracer_ownership_change);
INIT_WORK(&tracer->read_fw_strings_work, mlx5_tracer_read_strings_db);
INIT_WORK(&tracer->handle_traces_work, mlx5_fw_tracer_handle_traces);
@@ -675,6 +904,8 @@ void mlx5_fw_tracer_destroy(struct mlx5_fw_tracer *tracer)
return;
cancel_work_sync(&tracer->read_fw_strings_work);
+ mlx5_fw_tracer_clean_ready_list(tracer);
+ mlx5_fw_tracer_clean_print_hash(tracer);
mlx5_fw_tracer_free_strings_db(tracer);
mlx5_fw_tracer_destroy_log_buf(tracer);
flush_workqueue(tracer->work_queue);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h
index 3915e91486b2..8d310e7d6743 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h
@@ -46,6 +46,13 @@
#define TRACER_BLOCK_SIZE_BYTE 256
#define TRACES_PER_BLOCK 32
+#define TRACER_MAX_PARAMS 7
+#define MESSAGE_HASH_BITS 6
+#define MESSAGE_HASH_SIZE BIT(MESSAGE_HASH_BITS)
+
+#define MASK_52_7 (0x1FFFFFFFFFFF80)
+#define MASK_6_0 (0x7F)
+
struct mlx5_fw_tracer {
struct mlx5_core_dev *dev;
bool owner;
@@ -77,6 +84,21 @@ struct mlx5_fw_tracer {
u64 last_timestamp;
struct work_struct handle_traces_work;
+ struct hlist_head hash[MESSAGE_HASH_SIZE];
+ struct list_head ready_strings_list;
+};
+
+struct tracer_string_format {
+ char *string;
+ int params[TRACER_MAX_PARAMS];
+ int num_of_params;
+ int last_param_num;
+ u8 event_id;
+ u32 tmsn;
+ struct hlist_node hlist;
+ struct list_head list;
+ u32 timestamp;
+ bool lost;
};
enum mlx5_fw_tracer_ownership_state {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer_tracepoint.h b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer_tracepoint.h
new file mode 100644
index 000000000000..83f90e9aff45
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer_tracepoint.h
@@ -0,0 +1,78 @@
+/*
+ * Copyright (c) 2018, Mellanox Technologies. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses. You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ * Redistribution and use in source and binary forms, with or
+ * without modification, are permitted provided that the following
+ * conditions are met:
+ *
+ * - Redistributions of source code must retain the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer.
+ *
+ * - Redistributions in binary form must reproduce the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer in the documentation and/or other materials
+ * provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#if !defined(__LIB_TRACER_TRACEPOINT_H__) || defined(TRACE_HEADER_MULTI_READ)
+#define __LIB_TRACER_TRACEPOINT_H__
+
+#include <linux/tracepoint.h>
+#include "fw_tracer.h"
+
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM mlx5
+
+/* Tracepoint for FWTracer messages: */
+TRACE_EVENT(mlx5_fw,
+ TP_PROTO(const struct mlx5_fw_tracer *tracer, u64 trace_timestamp,
+ bool lost, u8 event_id, const char *msg),
+
+ TP_ARGS(tracer, trace_timestamp, lost, event_id, msg),
+
+ TP_STRUCT__entry(
+ __string(dev_name, dev_name(&tracer->dev->pdev->dev))
+ __field(u64, trace_timestamp)
+ __field(bool, lost)
+ __field(u8, event_id)
+ __string(msg, msg)
+ ),
+
+ TP_fast_assign(
+ __assign_str(dev_name, dev_name(&tracer->dev->pdev->dev));
+ __entry->trace_timestamp = trace_timestamp;
+ __entry->lost = lost;
+ __entry->event_id = event_id;
+ __assign_str(msg, msg);
+ ),
+
+ TP_printk("%s [0x%llx] %d [0x%x] %s",
+ __get_str(dev_name),
+ __entry->trace_timestamp,
+ __entry->lost, __entry->event_id,
+ __get_str(msg))
+);
+
+#endif
+
+#undef TRACE_INCLUDE_PATH
+#undef TRACE_INCLUDE_FILE
+#define TRACE_INCLUDE_PATH ./diag
+#define TRACE_INCLUDE_FILE fw_tracer_tracepoint
+#include <trace/define_trace.h>
--
2.17.0
^ permalink raw reply related
* [net-next V2 04/12] net/mlx5: FW tracer, events handling
From: Saeed Mahameed @ 2018-07-23 22:11 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Feras Daoud, Saeed Mahameed
In-Reply-To: <20180723221129.21625-1-saeedm@mellanox.com>
From: Feras Daoud <ferasda@mellanox.com>
The tracer has one event, event 0x26, with two subtypes:
- Subtype 0: Ownership change
- Subtype 1: Traces available
An ownership change occurs in the following cases:
1- Owner releases his ownership, in this case, an event will be
sent to inform others to reattempt acquire ownership.
2- Ownership was taken by a higher priority tool, in this case
the owner should understand that it lost ownership, and go through
tear down flow.
The second subtype indicates that there are traces in the trace buffer,
in this case, the driver polls the tracer buffer for new traces, parse
them and prepares the messages for printing.
The HW starts tracing from the first address in the tracer buffer.
Driver receives an event notifying that new trace block exists.
HW posts a timestamp event at the last 8B of every 256B block.
Comparing the timestamp to the last handled timestamp would indicate
that this is a new trace block. Once the new timestamp is detected,
the entire block is considered valid.
Block validation and parsing, should be done after copying the current
block to a different location, in order to avoid block overwritten
during processing.
Signed-off-by: Feras Daoud <ferasda@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../mellanox/mlx5/core/diag/fw_tracer.c | 268 +++++++++++++++++-
.../mellanox/mlx5/core/diag/fw_tracer.h | 71 ++++-
drivers/net/ethernet/mellanox/mlx5/core/eq.c | 11 +
include/linux/mlx5/device.h | 7 +
4 files changed, 347 insertions(+), 10 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
index d6cc27b0ff34..bd887d1d3396 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
@@ -318,25 +318,244 @@ static void mlx5_tracer_read_strings_db(struct work_struct *work)
return;
}
-static void mlx5_fw_tracer_ownership_change(struct work_struct *work)
+static void mlx5_fw_tracer_arm(struct mlx5_core_dev *dev)
{
- struct mlx5_fw_tracer *tracer = container_of(work, struct mlx5_fw_tracer,
- ownership_change_work);
- struct mlx5_core_dev *dev = tracer->dev;
+ u32 out[MLX5_ST_SZ_DW(mtrc_ctrl)] = {0};
+ u32 in[MLX5_ST_SZ_DW(mtrc_ctrl)] = {0};
int err;
- if (tracer->owner) {
- mlx5_fw_tracer_ownership_release(tracer);
+ MLX5_SET(mtrc_ctrl, in, arm_event, 1);
+
+ err = mlx5_core_access_reg(dev, in, sizeof(in), out, sizeof(out),
+ MLX5_REG_MTRC_CTRL, 0, 1);
+ if (err)
+ mlx5_core_warn(dev, "FWTracer: Failed to arm tracer event %d\n", err);
+}
+
+static void poll_trace(struct mlx5_fw_tracer *tracer,
+ struct tracer_event *tracer_event, u64 *trace)
+{
+ u32 timestamp_low, timestamp_mid, timestamp_high, urts;
+
+ tracer_event->event_id = MLX5_GET(tracer_event, trace, event_id);
+ tracer_event->lost_event = MLX5_GET(tracer_event, trace, lost);
+
+ switch (tracer_event->event_id) {
+ case TRACER_EVENT_TYPE_TIMESTAMP:
+ tracer_event->type = TRACER_EVENT_TYPE_TIMESTAMP;
+ urts = MLX5_GET(tracer_timestamp_event, trace, urts);
+ if (tracer->trc_ver == 0)
+ tracer_event->timestamp_event.unreliable = !!(urts >> 2);
+ else
+ tracer_event->timestamp_event.unreliable = !!(urts & 1);
+
+ timestamp_low = MLX5_GET(tracer_timestamp_event,
+ trace, timestamp7_0);
+ timestamp_mid = MLX5_GET(tracer_timestamp_event,
+ trace, timestamp39_8);
+ timestamp_high = MLX5_GET(tracer_timestamp_event,
+ trace, timestamp52_40);
+
+ tracer_event->timestamp_event.timestamp =
+ ((u64)timestamp_high << 40) |
+ ((u64)timestamp_mid << 8) |
+ (u64)timestamp_low;
+ break;
+ default:
+ if (tracer_event->event_id >= tracer->str_db.first_string_trace ||
+ tracer_event->event_id <= tracer->str_db.first_string_trace +
+ tracer->str_db.num_string_trace) {
+ tracer_event->type = TRACER_EVENT_TYPE_STRING;
+ tracer_event->string_event.timestamp =
+ MLX5_GET(tracer_string_event, trace, timestamp);
+ tracer_event->string_event.string_param =
+ MLX5_GET(tracer_string_event, trace, string_param);
+ tracer_event->string_event.tmsn =
+ MLX5_GET(tracer_string_event, trace, tmsn);
+ tracer_event->string_event.tdsn =
+ MLX5_GET(tracer_string_event, trace, tdsn);
+ } else {
+ tracer_event->type = TRACER_EVENT_TYPE_UNRECOGNIZED;
+ }
+ break;
+ }
+}
+
+static u64 get_block_timestamp(struct mlx5_fw_tracer *tracer, u64 *ts_event)
+{
+ struct tracer_event tracer_event;
+ u8 event_id;
+
+ event_id = MLX5_GET(tracer_event, ts_event, event_id);
+
+ if (event_id == TRACER_EVENT_TYPE_TIMESTAMP)
+ poll_trace(tracer, &tracer_event, ts_event);
+ else
+ tracer_event.timestamp_event.timestamp = 0;
+
+ return tracer_event.timestamp_event.timestamp;
+}
+
+static void mlx5_fw_tracer_handle_traces(struct work_struct *work)
+{
+ struct mlx5_fw_tracer *tracer =
+ container_of(work, struct mlx5_fw_tracer, handle_traces_work);
+ u64 block_timestamp, last_block_timestamp, tmp_trace_block[TRACES_PER_BLOCK];
+ u32 block_count, start_offset, prev_start_offset, prev_consumer_index;
+ u32 trace_event_size = MLX5_ST_SZ_BYTES(tracer_event);
+ struct tracer_event tracer_event;
+ struct mlx5_core_dev *dev;
+ int i;
+
+ if (!tracer->owner)
return;
+
+ dev = tracer->dev;
+ block_count = tracer->buff.size / TRACER_BLOCK_SIZE_BYTE;
+ start_offset = tracer->buff.consumer_index * TRACER_BLOCK_SIZE_BYTE;
+
+ /* Copy the block to local buffer to avoid HW override while being processed*/
+ memcpy(tmp_trace_block, tracer->buff.log_buf + start_offset,
+ TRACER_BLOCK_SIZE_BYTE);
+
+ block_timestamp =
+ get_block_timestamp(tracer, &tmp_trace_block[TRACES_PER_BLOCK - 1]);
+
+ while (block_timestamp > tracer->last_timestamp) {
+ /* Check block override if its not the first block */
+ if (!tracer->last_timestamp) {
+ u64 *ts_event;
+ /* To avoid block override be the HW in case of buffer
+ * wraparound, the time stamp of the previous block
+ * should be compared to the last timestamp handled
+ * by the driver.
+ */
+ prev_consumer_index =
+ (tracer->buff.consumer_index - 1) & (block_count - 1);
+ prev_start_offset = prev_consumer_index * TRACER_BLOCK_SIZE_BYTE;
+
+ ts_event = tracer->buff.log_buf + prev_start_offset +
+ (TRACES_PER_BLOCK - 1) * trace_event_size;
+ last_block_timestamp = get_block_timestamp(tracer, ts_event);
+ /* If previous timestamp different from last stored
+ * timestamp then there is a good chance that the
+ * current buffer is overwritten and therefore should
+ * not be parsed.
+ */
+ if (tracer->last_timestamp != last_block_timestamp) {
+ mlx5_core_warn(dev, "FWTracer: Events were lost\n");
+ tracer->last_timestamp = block_timestamp;
+ tracer->buff.consumer_index =
+ (tracer->buff.consumer_index + 1) & (block_count - 1);
+ break;
+ }
+ }
+
+ /* Parse events */
+ for (i = 0; i < TRACES_PER_BLOCK ; i++)
+ poll_trace(tracer, &tracer_event, &tmp_trace_block[i]);
+
+ tracer->buff.consumer_index =
+ (tracer->buff.consumer_index + 1) & (block_count - 1);
+
+ tracer->last_timestamp = block_timestamp;
+ start_offset = tracer->buff.consumer_index * TRACER_BLOCK_SIZE_BYTE;
+ memcpy(tmp_trace_block, tracer->buff.log_buf + start_offset,
+ TRACER_BLOCK_SIZE_BYTE);
+ block_timestamp = get_block_timestamp(tracer,
+ &tmp_trace_block[TRACES_PER_BLOCK - 1]);
}
+ mlx5_fw_tracer_arm(dev);
+}
+
+static int mlx5_fw_tracer_set_mtrc_conf(struct mlx5_fw_tracer *tracer)
+{
+ struct mlx5_core_dev *dev = tracer->dev;
+ u32 out[MLX5_ST_SZ_DW(mtrc_conf)] = {0};
+ u32 in[MLX5_ST_SZ_DW(mtrc_conf)] = {0};
+ int err;
+
+ MLX5_SET(mtrc_conf, in, trace_mode, TRACE_TO_MEMORY);
+ MLX5_SET(mtrc_conf, in, log_trace_buffer_size,
+ ilog2(TRACER_BUFFER_PAGE_NUM));
+ MLX5_SET(mtrc_conf, in, trace_mkey, tracer->buff.mkey.key);
+
+ err = mlx5_core_access_reg(dev, in, sizeof(in), out, sizeof(out),
+ MLX5_REG_MTRC_CONF, 0, 1);
+ if (err)
+ mlx5_core_warn(dev, "FWTracer: Failed to set tracer configurations %d\n", err);
+
+ return err;
+}
+
+static int mlx5_fw_tracer_set_mtrc_ctrl(struct mlx5_fw_tracer *tracer, u8 status, u8 arm)
+{
+ struct mlx5_core_dev *dev = tracer->dev;
+ u32 out[MLX5_ST_SZ_DW(mtrc_ctrl)] = {0};
+ u32 in[MLX5_ST_SZ_DW(mtrc_ctrl)] = {0};
+ int err;
+
+ MLX5_SET(mtrc_ctrl, in, modify_field_select, TRACE_STATUS);
+ MLX5_SET(mtrc_ctrl, in, trace_status, status);
+ MLX5_SET(mtrc_ctrl, in, arm_event, arm);
+
+ err = mlx5_core_access_reg(dev, in, sizeof(in), out, sizeof(out),
+ MLX5_REG_MTRC_CTRL, 0, 1);
+
+ if (!err && status)
+ tracer->last_timestamp = 0;
+
+ return err;
+}
+
+static int mlx5_fw_tracer_start(struct mlx5_fw_tracer *tracer)
+{
+ struct mlx5_core_dev *dev = tracer->dev;
+ int err;
+
err = mlx5_fw_tracer_ownership_acquire(tracer);
if (err) {
mlx5_core_dbg(dev, "FWTracer: Ownership was not granted %d\n", err);
+ /* Don't fail since ownership can be acquired on a later FW event */
+ return 0;
+ }
+
+ err = mlx5_fw_tracer_set_mtrc_conf(tracer);
+ if (err) {
+ mlx5_core_warn(dev, "FWTracer: Failed to set tracer configuration %d\n", err);
+ goto release_ownership;
+ }
+
+ /* enable tracer & trace events */
+ err = mlx5_fw_tracer_set_mtrc_ctrl(tracer, 1, 1);
+ if (err) {
+ mlx5_core_warn(dev, "FWTracer: Failed to enable tracer %d\n", err);
+ goto release_ownership;
+ }
+
+ return 0;
+
+release_ownership:
+ mlx5_fw_tracer_ownership_release(tracer);
+ return err;
+}
+
+static void mlx5_fw_tracer_ownership_change(struct work_struct *work)
+{
+ struct mlx5_fw_tracer *tracer =
+ container_of(work, struct mlx5_fw_tracer, ownership_change_work);
+
+ if (tracer->owner) {
+ tracer->owner = false;
+ tracer->buff.consumer_index = 0;
return;
}
+
+ mlx5_fw_tracer_start(tracer);
}
+/* Create software resources (Buffers, etc ..) */
struct mlx5_fw_tracer *mlx5_fw_tracer_create(struct mlx5_core_dev *dev)
{
struct mlx5_fw_tracer *tracer = NULL;
@@ -361,6 +580,8 @@ struct mlx5_fw_tracer *mlx5_fw_tracer_create(struct mlx5_core_dev *dev)
INIT_WORK(&tracer->ownership_change_work, mlx5_fw_tracer_ownership_change);
INIT_WORK(&tracer->read_fw_strings_work, mlx5_tracer_read_strings_db);
+ INIT_WORK(&tracer->handle_traces_work, mlx5_fw_tracer_handle_traces);
+
err = mlx5_query_mtrc_caps(tracer);
if (err) {
@@ -392,6 +613,9 @@ struct mlx5_fw_tracer *mlx5_fw_tracer_create(struct mlx5_core_dev *dev)
return ERR_PTR(err);
}
+/* Create HW resources + start tracer
+ * must be called before Async EQ is created
+ */
int mlx5_fw_tracer_init(struct mlx5_fw_tracer *tracer)
{
struct mlx5_core_dev *dev;
@@ -417,22 +641,25 @@ int mlx5_fw_tracer_init(struct mlx5_fw_tracer *tracer)
goto err_dealloc_pd;
}
- err = mlx5_fw_tracer_ownership_acquire(tracer);
- if (err) /* Don't fail since ownership can be acquired on a later FW event */
- mlx5_core_dbg(dev, "FWTracer: Ownership was not granted %d\n", err);
+ mlx5_fw_tracer_start(tracer);
return 0;
+
err_dealloc_pd:
mlx5_core_dealloc_pd(dev, tracer->buff.pdn);
return err;
}
+/* Stop tracer + Cleanup HW resources
+ * must be called after Async EQ is destroyed
+ */
void mlx5_fw_tracer_cleanup(struct mlx5_fw_tracer *tracer)
{
if (IS_ERR_OR_NULL(tracer))
return;
cancel_work_sync(&tracer->ownership_change_work);
+ cancel_work_sync(&tracer->handle_traces_work);
if (tracer->owner)
mlx5_fw_tracer_ownership_release(tracer);
@@ -441,6 +668,7 @@ void mlx5_fw_tracer_cleanup(struct mlx5_fw_tracer *tracer)
mlx5_core_dealloc_pd(tracer->dev, tracer->buff.pdn);
}
+/* Free software resources (Buffers, etc ..) */
void mlx5_fw_tracer_destroy(struct mlx5_fw_tracer *tracer)
{
if (IS_ERR_OR_NULL(tracer))
@@ -454,4 +682,26 @@ void mlx5_fw_tracer_destroy(struct mlx5_fw_tracer *tracer)
kfree(tracer);
}
+void mlx5_fw_tracer_event(struct mlx5_core_dev *dev, struct mlx5_eqe *eqe)
+{
+ struct mlx5_fw_tracer *tracer = dev->tracer;
+
+ if (!tracer)
+ return;
+
+ switch (eqe->sub_type) {
+ case MLX5_TRACER_SUBTYPE_OWNERSHIP_CHANGE:
+ if (test_bit(MLX5_INTERFACE_STATE_UP, &dev->intf_state))
+ queue_work(tracer->work_queue, &tracer->ownership_change_work);
+ break;
+ case MLX5_TRACER_SUBTYPE_TRACES_AVAILABLE:
+ if (likely(tracer->str_db.loaded))
+ queue_work(tracer->work_queue, &tracer->handle_traces_work);
+ break;
+ default:
+ mlx5_core_dbg(dev, "FWTracer: Event with unrecognized subtype: sub_type %d\n",
+ eqe->sub_type);
+ }
+}
+
EXPORT_TRACEPOINT_SYMBOL(mlx5_fw);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h
index 66cb7e7ada28..3915e91486b2 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h
@@ -43,6 +43,9 @@
#define TRACER_BUFFER_CHUNK 4096
#define TRACE_BUFFER_SIZE_BYTE (TRACER_BUFFER_PAGE_NUM * TRACER_BUFFER_CHUNK)
+#define TRACER_BLOCK_SIZE_BYTE 256
+#define TRACES_PER_BLOCK 32
+
struct mlx5_fw_tracer {
struct mlx5_core_dev *dev;
bool owner;
@@ -69,8 +72,11 @@ struct mlx5_fw_tracer {
dma_addr_t dma;
u32 size;
struct mlx5_core_mkey mkey;
-
+ u32 consumer_index;
} buff;
+
+ u64 last_timestamp;
+ struct work_struct handle_traces_work;
};
enum mlx5_fw_tracer_ownership_state {
@@ -78,7 +84,70 @@ enum mlx5_fw_tracer_ownership_state {
MLX5_FW_TRACER_ACQUIRE_OWNERSHIP,
};
+enum tracer_ctrl_fields_select {
+ TRACE_STATUS = 1 << 0,
+};
+
+enum tracer_event_type {
+ TRACER_EVENT_TYPE_STRING,
+ TRACER_EVENT_TYPE_TIMESTAMP = 0xFF,
+ TRACER_EVENT_TYPE_UNRECOGNIZED,
+};
+
+enum tracing_mode {
+ TRACE_TO_MEMORY = 1 << 0,
+};
+
+struct tracer_timestamp_event {
+ u64 timestamp;
+ u8 unreliable;
+};
+
+struct tracer_string_event {
+ u32 timestamp;
+ u32 tmsn;
+ u32 tdsn;
+ u32 string_param;
+};
+
+struct tracer_event {
+ bool lost_event;
+ u32 type;
+ u8 event_id;
+ union {
+ struct tracer_string_event string_event;
+ struct tracer_timestamp_event timestamp_event;
+ };
+};
+
+struct mlx5_ifc_tracer_event_bits {
+ u8 lost[0x1];
+ u8 timestamp[0x7];
+ u8 event_id[0x8];
+ u8 event_data[0x30];
+};
+
+struct mlx5_ifc_tracer_string_event_bits {
+ u8 lost[0x1];
+ u8 timestamp[0x7];
+ u8 event_id[0x8];
+ u8 tmsn[0xd];
+ u8 tdsn[0x3];
+ u8 string_param[0x20];
+};
+
+struct mlx5_ifc_tracer_timestamp_event_bits {
+ u8 timestamp7_0[0x8];
+ u8 event_id[0x8];
+ u8 urts[0x3];
+ u8 timestamp52_40[0xd];
+ u8 timestamp39_8[0x20];
+};
+
struct mlx5_fw_tracer *mlx5_fw_tracer_create(struct mlx5_core_dev *dev);
+int mlx5_fw_tracer_init(struct mlx5_fw_tracer *tracer);
+void mlx5_fw_tracer_cleanup(struct mlx5_fw_tracer *tracer);
void mlx5_fw_tracer_destroy(struct mlx5_fw_tracer *tracer);
+void mlx5_fw_tracer_event(struct mlx5_core_dev *dev, struct mlx5_eqe *eqe) { return; }
#endif
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eq.c b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
index 406c23862f5f..7669b4380779 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
@@ -40,6 +40,7 @@
#include "mlx5_core.h"
#include "fpga/core.h"
#include "eswitch.h"
+#include "diag/fw_tracer.h"
enum {
MLX5_EQE_SIZE = sizeof(struct mlx5_eqe),
@@ -168,6 +169,8 @@ static const char *eqe_type_str(u8 type)
return "MLX5_EVENT_TYPE_FPGA_QP_ERROR";
case MLX5_EVENT_TYPE_GENERAL_EVENT:
return "MLX5_EVENT_TYPE_GENERAL_EVENT";
+ case MLX5_EVENT_TYPE_DEVICE_TRACER:
+ return "MLX5_EVENT_TYPE_DEVICE_TRACER";
default:
return "Unrecognized event";
}
@@ -576,6 +579,11 @@ static irqreturn_t mlx5_eq_int(int irq, void *eq_ptr)
case MLX5_EVENT_TYPE_GENERAL_EVENT:
general_event_handler(dev, eqe);
break;
+
+ case MLX5_EVENT_TYPE_DEVICE_TRACER:
+ mlx5_fw_tracer_event(dev, eqe);
+ break;
+
default:
mlx5_core_warn(dev, "Unhandled event 0x%x on EQ 0x%x\n",
eqe->type, eq->eqn);
@@ -853,6 +861,9 @@ int mlx5_start_eqs(struct mlx5_core_dev *dev)
if (MLX5_CAP_GEN(dev, temp_warn_event))
async_event_mask |= (1ull << MLX5_EVENT_TYPE_TEMP_WARN_EVENT);
+ if (MLX5_CAP_MCAM_REG(dev, tracer_registers))
+ async_event_mask |= (1ull << MLX5_EVENT_TYPE_DEVICE_TRACER);
+
err = mlx5_create_map_eq(dev, &table->cmd_eq, MLX5_EQ_VEC_CMD,
MLX5_NUM_CMD_EQE, 1ull << MLX5_EVENT_TYPE_CMD,
"mlx5_cmd_eq", MLX5_EQ_TYPE_ASYNC);
diff --git a/include/linux/mlx5/device.h b/include/linux/mlx5/device.h
index 0566c6a94805..d489494b0a84 100644
--- a/include/linux/mlx5/device.h
+++ b/include/linux/mlx5/device.h
@@ -332,6 +332,13 @@ enum mlx5_event {
MLX5_EVENT_TYPE_FPGA_ERROR = 0x20,
MLX5_EVENT_TYPE_FPGA_QP_ERROR = 0x21,
+
+ MLX5_EVENT_TYPE_DEVICE_TRACER = 0x26,
+};
+
+enum {
+ MLX5_TRACER_SUBTYPE_OWNERSHIP_CHANGE = 0x0,
+ MLX5_TRACER_SUBTYPE_TRACES_AVAILABLE = 0x1,
};
enum {
--
2.17.0
^ permalink raw reply related
* [net-next V2 07/12] net/mlx5: FW tracer, Add debug prints
From: Saeed Mahameed @ 2018-07-23 22:11 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Saeed Mahameed
In-Reply-To: <20180723221129.21625-1-saeedm@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../ethernet/mellanox/mlx5/core/diag/fw_tracer.c | 13 +++++++++++--
1 file changed, 11 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
index 309842de272c..d4ec93bde4de 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
@@ -629,14 +629,14 @@ static void mlx5_fw_tracer_handle_traces(struct work_struct *work)
u64 block_timestamp, last_block_timestamp, tmp_trace_block[TRACES_PER_BLOCK];
u32 block_count, start_offset, prev_start_offset, prev_consumer_index;
u32 trace_event_size = MLX5_ST_SZ_BYTES(tracer_event);
+ struct mlx5_core_dev *dev = tracer->dev;
struct tracer_event tracer_event;
- struct mlx5_core_dev *dev;
int i;
+ mlx5_core_dbg(dev, "FWTracer: Handle Trace event, owner=(%d)\n", tracer->owner);
if (!tracer->owner)
return;
- dev = tracer->dev;
block_count = tracer->buff.size / TRACER_BLOCK_SIZE_BYTE;
start_offset = tracer->buff.consumer_index * TRACER_BLOCK_SIZE_BYTE;
@@ -762,6 +762,7 @@ static int mlx5_fw_tracer_start(struct mlx5_fw_tracer *tracer)
goto release_ownership;
}
+ mlx5_core_dbg(dev, "FWTracer: Ownership granted and active\n");
return 0;
release_ownership:
@@ -774,6 +775,7 @@ static void mlx5_fw_tracer_ownership_change(struct work_struct *work)
struct mlx5_fw_tracer *tracer =
container_of(work, struct mlx5_fw_tracer, ownership_change_work);
+ mlx5_core_dbg(tracer->dev, "FWTracer: ownership changed, current=(%d)\n", tracer->owner);
if (tracer->owner) {
tracer->owner = false;
tracer->buff.consumer_index = 0;
@@ -830,6 +832,8 @@ struct mlx5_fw_tracer *mlx5_fw_tracer_create(struct mlx5_core_dev *dev)
goto free_log_buf;
}
+ mlx5_core_dbg(dev, "FWTracer: Tracer created\n");
+
return tracer;
free_log_buf:
@@ -887,6 +891,9 @@ void mlx5_fw_tracer_cleanup(struct mlx5_fw_tracer *tracer)
if (IS_ERR_OR_NULL(tracer))
return;
+ mlx5_core_dbg(tracer->dev, "FWTracer: Cleanup, is owner ? (%d)\n",
+ tracer->owner);
+
cancel_work_sync(&tracer->ownership_change_work);
cancel_work_sync(&tracer->handle_traces_work);
@@ -903,6 +910,8 @@ void mlx5_fw_tracer_destroy(struct mlx5_fw_tracer *tracer)
if (IS_ERR_OR_NULL(tracer))
return;
+ mlx5_core_dbg(tracer->dev, "FWTracer: Destroy\n");
+
cancel_work_sync(&tracer->read_fw_strings_work);
mlx5_fw_tracer_clean_ready_list(tracer);
mlx5_fw_tracer_clean_print_hash(tracer);
--
2.17.0
^ permalink raw reply related
* [net-next V2 06/12] net/mlx5: FW tracer, Enable tracing
From: Saeed Mahameed @ 2018-07-23 22:11 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Feras Daoud, Saeed Mahameed
In-Reply-To: <20180723221129.21625-1-saeedm@mellanox.com>
From: Feras Daoud <ferasda@mellanox.com>
Add the tracer file to the makefile and add the init
function to the load one flow.
Signed-off-by: Feras Daoud <ferasda@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../net/ethernet/mellanox/mlx5/core/Makefile | 2 +-
.../mellanox/mlx5/core/diag/fw_tracer.h | 2 +-
drivers/net/ethernet/mellanox/mlx5/core/main.c | 18 ++++++++++++++++--
3 files changed, 18 insertions(+), 4 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
index d923f2f58608..55d5a5c2e9d8 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile
+++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
@@ -6,7 +6,7 @@ mlx5_core-y := main.o cmd.o debugfs.o fw.o eq.o uar.o pagealloc.o \
health.o mcg.o cq.o srq.o alloc.o qp.o port.o mr.o pd.o \
mad.o transobj.o vport.o sriov.o fs_cmd.o fs_core.o \
fs_counters.o rl.o lag.o dev.o wq.o lib/gid.o lib/clock.o \
- diag/fs_tracepoint.o
+ diag/fs_tracepoint.o diag/fw_tracer.o
mlx5_core-$(CONFIG_MLX5_ACCEL) += accel/ipsec.o accel/tls.o
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h
index 8d310e7d6743..0347f2dd5cee 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h
@@ -170,6 +170,6 @@ struct mlx5_fw_tracer *mlx5_fw_tracer_create(struct mlx5_core_dev *dev);
int mlx5_fw_tracer_init(struct mlx5_fw_tracer *tracer);
void mlx5_fw_tracer_cleanup(struct mlx5_fw_tracer *tracer);
void mlx5_fw_tracer_destroy(struct mlx5_fw_tracer *tracer);
-void mlx5_fw_tracer_event(struct mlx5_core_dev *dev, struct mlx5_eqe *eqe) { return; }
+void mlx5_fw_tracer_event(struct mlx5_core_dev *dev, struct mlx5_eqe *eqe);
#endif
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c
index f9b950e1bd85..6ddbb70e95de 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c
@@ -62,6 +62,7 @@
#include "accel/ipsec.h"
#include "accel/tls.h"
#include "lib/clock.h"
+#include "diag/fw_tracer.h"
MODULE_AUTHOR("Eli Cohen <eli@mellanox.com>");
MODULE_DESCRIPTION("Mellanox 5th generation network adapters (ConnectX series) core driver");
@@ -990,6 +991,8 @@ static int mlx5_init_once(struct mlx5_core_dev *dev, struct mlx5_priv *priv)
goto err_sriov_cleanup;
}
+ dev->tracer = mlx5_fw_tracer_create(dev);
+
return 0;
err_sriov_cleanup:
@@ -1015,6 +1018,7 @@ static int mlx5_init_once(struct mlx5_core_dev *dev, struct mlx5_priv *priv)
static void mlx5_cleanup_once(struct mlx5_core_dev *dev)
{
+ mlx5_fw_tracer_destroy(dev->tracer);
mlx5_fpga_cleanup(dev);
mlx5_sriov_cleanup(dev);
mlx5_eswitch_cleanup(dev->priv.eswitch);
@@ -1167,10 +1171,16 @@ static int mlx5_load_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv,
goto err_put_uars;
}
+ err = mlx5_fw_tracer_init(dev->tracer);
+ if (err) {
+ dev_err(&pdev->dev, "Failed to init FW tracer\n");
+ goto err_fw_tracer;
+ }
+
err = alloc_comp_eqs(dev);
if (err) {
dev_err(&pdev->dev, "Failed to alloc completion EQs\n");
- goto err_stop_eqs;
+ goto err_comp_eqs;
}
err = mlx5_irq_set_affinity_hints(dev);
@@ -1252,7 +1262,10 @@ static int mlx5_load_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv,
err_affinity_hints:
free_comp_eqs(dev);
-err_stop_eqs:
+err_comp_eqs:
+ mlx5_fw_tracer_cleanup(dev->tracer);
+
+err_fw_tracer:
mlx5_stop_eqs(dev);
err_put_uars:
@@ -1320,6 +1333,7 @@ static int mlx5_unload_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv,
mlx5_fpga_device_stop(dev);
mlx5_irq_clear_affinity_hints(dev);
free_comp_eqs(dev);
+ mlx5_fw_tracer_cleanup(dev->tracer);
mlx5_stop_eqs(dev);
mlx5_put_uars_page(dev, priv->uar);
mlx5_free_irq_vectors(dev);
--
2.17.0
^ permalink raw reply related
* [net-next V2 08/12] net/mlx5e: Remove redundant WARN when we cannot find neigh entry
From: Saeed Mahameed @ 2018-07-23 22:11 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Roi Dayan, Saeed Mahameed
In-Reply-To: <20180723221129.21625-1-saeedm@mellanox.com>
From: Roi Dayan <roid@mellanox.com>
It is possible for neigh entry not to exist if it was cleaned already.
When we bring down an interface the neigh gets deleted but it could be
that our listener for neigh event to clear the encap valid bit didn't
start yet and the neigh update last used work is started first.
In this scenario the encap entry has valid bit set but the neigh entry
doesn't exist.
Signed-off-by: Roi Dayan <roid@mellanox.com>
Reviewed-by: Paul Blakey <paulb@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 4 +---
1 file changed, 1 insertion(+), 3 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 0edf4751a8ba..335a08bc381d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -1032,10 +1032,8 @@ void mlx5e_tc_update_neigh_used_value(struct mlx5e_neigh_hash_entry *nhe)
* dst ip pair
*/
n = neigh_lookup(tbl, &m_neigh->dst_ip, m_neigh->dev);
- if (!n) {
- WARN(1, "The neighbour already freed\n");
+ if (!n)
return;
- }
neigh_event_send(n, NULL);
neigh_release(n);
--
2.17.0
^ permalink raw reply related
* [net-next V2 09/12] net/mlx5e: Support offloading tc double vlan headers match
From: Saeed Mahameed @ 2018-07-23 22:11 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Jianbo Liu, Saeed Mahameed
In-Reply-To: <20180723221129.21625-1-saeedm@mellanox.com>
From: Jianbo Liu <jianbol@mellanox.com>
We can match on both outer and inner vlan tags, add support for
offloading that.
Signed-off-by: Jianbo Liu <jianbol@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../net/ethernet/mellanox/mlx5/core/en_tc.c | 55 ++++++++++++++++++-
1 file changed, 52 insertions(+), 3 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 335a08bc381d..dcb8c4993811 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -1235,6 +1235,10 @@ static int __parse_cls_flower(struct mlx5e_priv *priv,
outer_headers);
void *headers_v = MLX5_ADDR_OF(fte_match_param, spec->match_value,
outer_headers);
+ void *misc_c = MLX5_ADDR_OF(fte_match_param, spec->match_criteria,
+ misc_parameters);
+ void *misc_v = MLX5_ADDR_OF(fte_match_param, spec->match_value,
+ misc_parameters);
u16 addr_type = 0;
u8 ip_proto = 0;
@@ -1245,6 +1249,7 @@ static int __parse_cls_flower(struct mlx5e_priv *priv,
BIT(FLOW_DISSECTOR_KEY_BASIC) |
BIT(FLOW_DISSECTOR_KEY_ETH_ADDRS) |
BIT(FLOW_DISSECTOR_KEY_VLAN) |
+ BIT(FLOW_DISSECTOR_KEY_CVLAN) |
BIT(FLOW_DISSECTOR_KEY_IPV4_ADDRS) |
BIT(FLOW_DISSECTOR_KEY_IPV6_ADDRS) |
BIT(FLOW_DISSECTOR_KEY_PORTS) |
@@ -1325,9 +1330,18 @@ static int __parse_cls_flower(struct mlx5e_priv *priv,
skb_flow_dissector_target(f->dissector,
FLOW_DISSECTOR_KEY_VLAN,
f->mask);
- if (mask->vlan_id || mask->vlan_priority) {
- MLX5_SET(fte_match_set_lyr_2_4, headers_c, cvlan_tag, 1);
- MLX5_SET(fte_match_set_lyr_2_4, headers_v, cvlan_tag, 1);
+ if (mask->vlan_id || mask->vlan_priority || mask->vlan_tpid) {
+ if (key->vlan_tpid == htons(ETH_P_8021AD)) {
+ MLX5_SET(fte_match_set_lyr_2_4, headers_c,
+ svlan_tag, 1);
+ MLX5_SET(fte_match_set_lyr_2_4, headers_v,
+ svlan_tag, 1);
+ } else {
+ MLX5_SET(fte_match_set_lyr_2_4, headers_c,
+ cvlan_tag, 1);
+ MLX5_SET(fte_match_set_lyr_2_4, headers_v,
+ cvlan_tag, 1);
+ }
MLX5_SET(fte_match_set_lyr_2_4, headers_c, first_vid, mask->vlan_id);
MLX5_SET(fte_match_set_lyr_2_4, headers_v, first_vid, key->vlan_id);
@@ -1339,6 +1353,41 @@ static int __parse_cls_flower(struct mlx5e_priv *priv,
}
}
+ if (dissector_uses_key(f->dissector, FLOW_DISSECTOR_KEY_CVLAN)) {
+ struct flow_dissector_key_vlan *key =
+ skb_flow_dissector_target(f->dissector,
+ FLOW_DISSECTOR_KEY_CVLAN,
+ f->key);
+ struct flow_dissector_key_vlan *mask =
+ skb_flow_dissector_target(f->dissector,
+ FLOW_DISSECTOR_KEY_CVLAN,
+ f->mask);
+ if (mask->vlan_id || mask->vlan_priority || mask->vlan_tpid) {
+ if (key->vlan_tpid == htons(ETH_P_8021AD)) {
+ MLX5_SET(fte_match_set_misc, misc_c,
+ outer_second_svlan_tag, 1);
+ MLX5_SET(fte_match_set_misc, misc_v,
+ outer_second_svlan_tag, 1);
+ } else {
+ MLX5_SET(fte_match_set_misc, misc_c,
+ outer_second_cvlan_tag, 1);
+ MLX5_SET(fte_match_set_misc, misc_v,
+ outer_second_cvlan_tag, 1);
+ }
+
+ MLX5_SET(fte_match_set_misc, misc_c, outer_second_vid,
+ mask->vlan_id);
+ MLX5_SET(fte_match_set_misc, misc_v, outer_second_vid,
+ key->vlan_id);
+ MLX5_SET(fte_match_set_misc, misc_c, outer_second_prio,
+ mask->vlan_priority);
+ MLX5_SET(fte_match_set_misc, misc_v, outer_second_prio,
+ key->vlan_priority);
+
+ *match_level = MLX5_MATCH_L2;
+ }
+ }
+
if (dissector_uses_key(f->dissector, FLOW_DISSECTOR_KEY_BASIC)) {
struct flow_dissector_key_basic *key =
skb_flow_dissector_target(f->dissector,
--
2.17.0
^ permalink raw reply related
* [net-next V2 10/12] net/mlx5e: Refactor tc vlan push/pop actions offloading
From: Saeed Mahameed @ 2018-07-23 22:11 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Jianbo Liu, Saeed Mahameed
In-Reply-To: <20180723221129.21625-1-saeedm@mellanox.com>
From: Jianbo Liu <jianbol@mellanox.com>
Extract actions offloading code to a new function, and also extend data
structures for double vlan actions.
Signed-off-by: Jianbo Liu <jianbol@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../net/ethernet/mellanox/mlx5/core/en_tc.c | 51 ++++++++++++-------
.../net/ethernet/mellanox/mlx5/core/eswitch.h | 6 +--
.../mellanox/mlx5/core/eswitch_offloads.c | 12 ++---
3 files changed, 41 insertions(+), 28 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index dcb8c4993811..35b3e135ae1d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -2578,6 +2578,32 @@ static int mlx5e_attach_encap(struct mlx5e_priv *priv,
return err;
}
+static int parse_tc_vlan_action(struct mlx5e_priv *priv,
+ const struct tc_action *a,
+ struct mlx5_esw_flow_attr *attr,
+ u32 *action)
+{
+ if (tcf_vlan_action(a) == TCA_VLAN_ACT_POP) {
+ *action |= MLX5_FLOW_CONTEXT_ACTION_VLAN_POP;
+ } else if (tcf_vlan_action(a) == TCA_VLAN_ACT_PUSH) {
+ *action |= MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH;
+ attr->vlan_vid[0] = tcf_vlan_push_vid(a);
+ if (mlx5_eswitch_vlan_actions_supported(priv->mdev)) {
+ attr->vlan_prio[0] = tcf_vlan_push_prio(a);
+ attr->vlan_proto[0] = tcf_vlan_push_proto(a);
+ if (!attr->vlan_proto[0])
+ attr->vlan_proto[0] = htons(ETH_P_8021Q);
+ } else if (tcf_vlan_push_proto(a) != htons(ETH_P_8021Q) ||
+ tcf_vlan_push_prio(a)) {
+ return -EOPNOTSUPP;
+ }
+ } else { /* action is TCA_VLAN_ACT_MODIFY */
+ return -EOPNOTSUPP;
+ }
+
+ return 0;
+}
+
static int parse_tc_fdb_actions(struct mlx5e_priv *priv, struct tcf_exts *exts,
struct mlx5e_tc_flow_parse_attr *parse_attr,
struct mlx5e_tc_flow *flow)
@@ -2589,6 +2615,7 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv, struct tcf_exts *exts,
LIST_HEAD(actions);
bool encap = false;
u32 action = 0;
+ int err;
if (!tcf_exts_has_actions(exts))
return -EINVAL;
@@ -2605,8 +2632,6 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv, struct tcf_exts *exts,
}
if (is_tcf_pedit(a)) {
- int err;
-
err = parse_tc_pedit_action(priv, a, MLX5_FLOW_NAMESPACE_FDB,
parse_attr);
if (err)
@@ -2673,23 +2698,11 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv, struct tcf_exts *exts,
}
if (is_tcf_vlan(a)) {
- if (tcf_vlan_action(a) == TCA_VLAN_ACT_POP) {
- action |= MLX5_FLOW_CONTEXT_ACTION_VLAN_POP;
- } else if (tcf_vlan_action(a) == TCA_VLAN_ACT_PUSH) {
- action |= MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH;
- attr->vlan_vid = tcf_vlan_push_vid(a);
- if (mlx5_eswitch_vlan_actions_supported(priv->mdev)) {
- attr->vlan_prio = tcf_vlan_push_prio(a);
- attr->vlan_proto = tcf_vlan_push_proto(a);
- if (!attr->vlan_proto)
- attr->vlan_proto = htons(ETH_P_8021Q);
- } else if (tcf_vlan_push_proto(a) != htons(ETH_P_8021Q) ||
- tcf_vlan_push_prio(a)) {
- return -EOPNOTSUPP;
- }
- } else { /* action is TCA_VLAN_ACT_MODIFY */
- return -EOPNOTSUPP;
- }
+ err = parse_tc_vlan_action(priv, a, attr, &action);
+
+ if (err)
+ return err;
+
attr->mirror_count = attr->out_count;
continue;
}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
index b174da2884c5..befa0011efee 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
@@ -256,9 +256,9 @@ struct mlx5_esw_flow_attr {
int out_count;
int action;
- __be16 vlan_proto;
- u16 vlan_vid;
- u8 vlan_prio;
+ __be16 vlan_proto[1];
+ u16 vlan_vid[1];
+ u8 vlan_prio[1];
bool vlan_handled;
u32 encap_id;
u32 mod_hdr_id;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index f32e69170b30..552954d7184e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -70,9 +70,9 @@ mlx5_eswitch_add_offloaded_rule(struct mlx5_eswitch *esw,
flow_act.action &= ~(MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH |
MLX5_FLOW_CONTEXT_ACTION_VLAN_POP);
else if (flow_act.action & MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH) {
- flow_act.vlan[0].ethtype = ntohs(attr->vlan_proto);
- flow_act.vlan[0].vid = attr->vlan_vid;
- flow_act.vlan[0].prio = attr->vlan_prio;
+ flow_act.vlan[0].ethtype = ntohs(attr->vlan_proto[0]);
+ flow_act.vlan[0].vid = attr->vlan_vid[0];
+ flow_act.vlan[0].prio = attr->vlan_prio[0];
}
if (flow_act.action & MLX5_FLOW_CONTEXT_ACTION_FWD_DEST) {
@@ -266,7 +266,7 @@ static int esw_add_vlan_action_check(struct mlx5_esw_flow_attr *attr,
/* protects against (1) setting rules with different vlans to push and
* (2) setting rules w.o vlans (attr->vlan = 0) && w. vlans to push (!= 0)
*/
- if (push && in_rep->vlan_refcount && (in_rep->vlan != attr->vlan_vid))
+ if (push && in_rep->vlan_refcount && (in_rep->vlan != attr->vlan_vid[0]))
goto out_notsupp;
return 0;
@@ -324,11 +324,11 @@ int mlx5_eswitch_add_vlan_action(struct mlx5_eswitch *esw,
if (vport->vlan_refcount)
goto skip_set_push;
- err = __mlx5_eswitch_set_vport_vlan(esw, vport->vport, attr->vlan_vid, 0,
+ err = __mlx5_eswitch_set_vport_vlan(esw, vport->vport, attr->vlan_vid[0], 0,
SET_VLAN_INSERT | SET_VLAN_STRIP);
if (err)
goto out;
- vport->vlan = attr->vlan_vid;
+ vport->vlan = attr->vlan_vid[0];
skip_set_push:
vport->vlan_refcount++;
}
--
2.17.0
^ permalink raw reply related
* [net-next V2 11/12] net/mlx5e: Support offloading double vlan push/pop tc actions
From: Saeed Mahameed @ 2018-07-23 22:11 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Jianbo Liu, Saeed Mahameed
In-Reply-To: <20180723221129.21625-1-saeedm@mellanox.com>
From: Jianbo Liu <jianbol@mellanox.com>
As we can configure two push/pop actions in one flow table entry,
add support to offload those double vlan actions in a rule to HW.
Signed-off-by: Jianbo Liu <jianbol@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../net/ethernet/mellanox/mlx5/core/en_tc.c | 46 ++++++++++++++-----
.../net/ethernet/mellanox/mlx5/core/eswitch.h | 21 ++++++---
.../mellanox/mlx5/core/eswitch_offloads.c | 11 +++--
3 files changed, 58 insertions(+), 20 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 35b3e135ae1d..e9888d6c1f7c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -2583,24 +2583,48 @@ static int parse_tc_vlan_action(struct mlx5e_priv *priv,
struct mlx5_esw_flow_attr *attr,
u32 *action)
{
+ u8 vlan_idx = attr->total_vlan;
+
+ if (vlan_idx >= MLX5_FS_VLAN_DEPTH)
+ return -EOPNOTSUPP;
+
if (tcf_vlan_action(a) == TCA_VLAN_ACT_POP) {
- *action |= MLX5_FLOW_CONTEXT_ACTION_VLAN_POP;
+ if (vlan_idx) {
+ if (!mlx5_eswitch_vlan_actions_supported(priv->mdev,
+ MLX5_FS_VLAN_DEPTH))
+ return -EOPNOTSUPP;
+
+ *action |= MLX5_FLOW_CONTEXT_ACTION_VLAN_POP_2;
+ } else {
+ *action |= MLX5_FLOW_CONTEXT_ACTION_VLAN_POP;
+ }
} else if (tcf_vlan_action(a) == TCA_VLAN_ACT_PUSH) {
- *action |= MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH;
- attr->vlan_vid[0] = tcf_vlan_push_vid(a);
- if (mlx5_eswitch_vlan_actions_supported(priv->mdev)) {
- attr->vlan_prio[0] = tcf_vlan_push_prio(a);
- attr->vlan_proto[0] = tcf_vlan_push_proto(a);
- if (!attr->vlan_proto[0])
- attr->vlan_proto[0] = htons(ETH_P_8021Q);
- } else if (tcf_vlan_push_proto(a) != htons(ETH_P_8021Q) ||
- tcf_vlan_push_prio(a)) {
- return -EOPNOTSUPP;
+ attr->vlan_vid[vlan_idx] = tcf_vlan_push_vid(a);
+ attr->vlan_prio[vlan_idx] = tcf_vlan_push_prio(a);
+ attr->vlan_proto[vlan_idx] = tcf_vlan_push_proto(a);
+ if (!attr->vlan_proto[vlan_idx])
+ attr->vlan_proto[vlan_idx] = htons(ETH_P_8021Q);
+
+ if (vlan_idx) {
+ if (!mlx5_eswitch_vlan_actions_supported(priv->mdev,
+ MLX5_FS_VLAN_DEPTH))
+ return -EOPNOTSUPP;
+
+ *action |= MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH_2;
+ } else {
+ if (!mlx5_eswitch_vlan_actions_supported(priv->mdev, 1) &&
+ (tcf_vlan_push_proto(a) != htons(ETH_P_8021Q) ||
+ tcf_vlan_push_prio(a)))
+ return -EOPNOTSUPP;
+
+ *action |= MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH;
}
} else { /* action is TCA_VLAN_ACT_MODIFY */
return -EOPNOTSUPP;
}
+ attr->total_vlan = vlan_idx + 1;
+
return 0;
}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
index befa0011efee..c17bfcab517c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
@@ -38,6 +38,7 @@
#include <net/devlink.h>
#include <linux/mlx5/device.h>
#include <linux/mlx5/eswitch.h>
+#include <linux/mlx5/fs.h>
#include "lib/mpfs.h"
#ifdef CONFIG_MLX5_ESWITCH
@@ -256,9 +257,10 @@ struct mlx5_esw_flow_attr {
int out_count;
int action;
- __be16 vlan_proto[1];
- u16 vlan_vid[1];
- u8 vlan_prio[1];
+ __be16 vlan_proto[MLX5_FS_VLAN_DEPTH];
+ u16 vlan_vid[MLX5_FS_VLAN_DEPTH];
+ u8 vlan_prio[MLX5_FS_VLAN_DEPTH];
+ u8 total_vlan;
bool vlan_handled;
u32 encap_id;
u32 mod_hdr_id;
@@ -282,10 +284,17 @@ int mlx5_eswitch_del_vlan_action(struct mlx5_eswitch *esw,
int __mlx5_eswitch_set_vport_vlan(struct mlx5_eswitch *esw,
int vport, u16 vlan, u8 qos, u8 set_flags);
-static inline bool mlx5_eswitch_vlan_actions_supported(struct mlx5_core_dev *dev)
+static inline bool mlx5_eswitch_vlan_actions_supported(struct mlx5_core_dev *dev,
+ u8 vlan_depth)
{
- return MLX5_CAP_ESW_FLOWTABLE_FDB(dev, pop_vlan) &&
- MLX5_CAP_ESW_FLOWTABLE_FDB(dev, push_vlan);
+ bool ret = MLX5_CAP_ESW_FLOWTABLE_FDB(dev, pop_vlan) &&
+ MLX5_CAP_ESW_FLOWTABLE_FDB(dev, push_vlan);
+
+ if (vlan_depth == 1)
+ return ret;
+
+ return ret && MLX5_CAP_ESW_FLOWTABLE_FDB(dev, pop_vlan_2) &&
+ MLX5_CAP_ESW_FLOWTABLE_FDB(dev, push_vlan_2);
}
#define MLX5_DEBUG_ESWITCH_MASK BIT(3)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index 552954d7184e..f72b5c9dcfe9 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -66,13 +66,18 @@ mlx5_eswitch_add_offloaded_rule(struct mlx5_eswitch *esw,
flow_act.action = attr->action;
/* if per flow vlan pop/push is emulated, don't set that into the firmware */
- if (!mlx5_eswitch_vlan_actions_supported(esw->dev))
+ if (!mlx5_eswitch_vlan_actions_supported(esw->dev, 1))
flow_act.action &= ~(MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH |
MLX5_FLOW_CONTEXT_ACTION_VLAN_POP);
else if (flow_act.action & MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH) {
flow_act.vlan[0].ethtype = ntohs(attr->vlan_proto[0]);
flow_act.vlan[0].vid = attr->vlan_vid[0];
flow_act.vlan[0].prio = attr->vlan_prio[0];
+ if (flow_act.action & MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH_2) {
+ flow_act.vlan[1].ethtype = ntohs(attr->vlan_proto[1]);
+ flow_act.vlan[1].vid = attr->vlan_vid[1];
+ flow_act.vlan[1].prio = attr->vlan_prio[1];
+ }
}
if (flow_act.action & MLX5_FLOW_CONTEXT_ACTION_FWD_DEST) {
@@ -284,7 +289,7 @@ int mlx5_eswitch_add_vlan_action(struct mlx5_eswitch *esw,
int err = 0;
/* nop if we're on the vlan push/pop non emulation mode */
- if (mlx5_eswitch_vlan_actions_supported(esw->dev))
+ if (mlx5_eswitch_vlan_actions_supported(esw->dev, 1))
return 0;
push = !!(attr->action & MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH);
@@ -347,7 +352,7 @@ int mlx5_eswitch_del_vlan_action(struct mlx5_eswitch *esw,
int err = 0;
/* nop if we're on the vlan push/pop non emulation mode */
- if (mlx5_eswitch_vlan_actions_supported(esw->dev))
+ if (mlx5_eswitch_vlan_actions_supported(esw->dev, 1))
return 0;
if (!attr->vlan_handled)
--
2.17.0
^ permalink raw reply related
* [net-next V2 12/12] net/mlx5e: Use PARTIAL_GSO for UDP segmentation
From: Saeed Mahameed @ 2018-07-23 22:11 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Boris Pismenny, Saeed Mahameed
In-Reply-To: <20180723221129.21625-1-saeedm@mellanox.com>
From: Boris Pismenny <borisp@mellanox.com>
This patch removes the splitting of UDP_GSO_L4 packets in the driver,
and exposes UDP_GSO_L4 as a PARTIAL_GSO feature. Thus, the network stack
is not responsible for splitting the packet into two.
Signed-off-by: Boris Pismenny <borisp@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../net/ethernet/mellanox/mlx5/core/Makefile | 4 +-
.../mellanox/mlx5/core/en_accel/en_accel.h | 27 +++--
.../mellanox/mlx5/core/en_accel/rxtx.c | 109 ------------------
.../mellanox/mlx5/core/en_accel/rxtx.h | 14 ---
.../net/ethernet/mellanox/mlx5/core/en_main.c | 9 +-
5 files changed, 23 insertions(+), 140 deletions(-)
delete mode 100644 drivers/net/ethernet/mellanox/mlx5/core/en_accel/rxtx.c
delete mode 100644 drivers/net/ethernet/mellanox/mlx5/core/en_accel/rxtx.h
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
index 55d5a5c2e9d8..fa7fcca5dc78 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile
+++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
@@ -14,8 +14,8 @@ mlx5_core-$(CONFIG_MLX5_FPGA) += fpga/cmd.o fpga/core.o fpga/conn.o fpga/sdk.o \
fpga/ipsec.o fpga/tls.o
mlx5_core-$(CONFIG_MLX5_CORE_EN) += en_main.o en_common.o en_fs.o en_ethtool.o \
- en_tx.o en_rx.o en_dim.o en_txrx.o en_accel/rxtx.o en_stats.o \
- vxlan.o en_arfs.o en_fs_ethtool.o en_selftest.o en/port.o
+ en_tx.o en_rx.o en_dim.o en_txrx.o en_stats.o vxlan.o \
+ en_arfs.o en_fs_ethtool.o en_selftest.o en/port.o
mlx5_core-$(CONFIG_MLX5_MPFS) += lib/mpfs.o
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/en_accel.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/en_accel.h
index 39a5d13ba459..1dd225380a66 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/en_accel.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/en_accel.h
@@ -38,14 +38,22 @@
#include <linux/netdevice.h>
#include "en_accel/ipsec_rxtx.h"
#include "en_accel/tls_rxtx.h"
-#include "en_accel/rxtx.h"
#include "en.h"
-static inline struct sk_buff *mlx5e_accel_handle_tx(struct sk_buff *skb,
- struct mlx5e_txqsq *sq,
- struct net_device *dev,
- struct mlx5e_tx_wqe **wqe,
- u16 *pi)
+static inline void
+mlx5e_udp_gso_handle_tx_skb(struct sk_buff *skb)
+{
+ int payload_len = skb_shinfo(skb)->gso_size + sizeof(struct udphdr);
+
+ udp_hdr(skb)->len = htons(payload_len);
+}
+
+static inline struct sk_buff *
+mlx5e_accel_handle_tx(struct sk_buff *skb,
+ struct mlx5e_txqsq *sq,
+ struct net_device *dev,
+ struct mlx5e_tx_wqe **wqe,
+ u16 *pi)
{
#ifdef CONFIG_MLX5_EN_TLS
if (test_bit(MLX5E_SQ_STATE_TLS, &sq->state)) {
@@ -63,11 +71,8 @@ static inline struct sk_buff *mlx5e_accel_handle_tx(struct sk_buff *skb,
}
#endif
- if (skb_shinfo(skb)->gso_type & SKB_GSO_UDP_L4) {
- skb = mlx5e_udp_gso_handle_tx_skb(dev, sq, skb, wqe, pi);
- if (unlikely(!skb))
- return NULL;
- }
+ if (skb_is_gso(skb) && skb_shinfo(skb)->gso_type & SKB_GSO_UDP_L4)
+ mlx5e_udp_gso_handle_tx_skb(skb);
return skb;
}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/rxtx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/rxtx.c
deleted file mode 100644
index 7b7ec3998e84..000000000000
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/rxtx.c
+++ /dev/null
@@ -1,109 +0,0 @@
-#include "en_accel/rxtx.h"
-
-static void mlx5e_udp_gso_prepare_last_skb(struct sk_buff *skb,
- struct sk_buff *nskb,
- int remaining)
-{
- int bytes_needed = remaining, remaining_headlen, remaining_page_offset;
- int headlen = skb_transport_offset(skb) + sizeof(struct udphdr);
- int payload_len = remaining + sizeof(struct udphdr);
- int k = 0, i, j;
-
- skb_copy_bits(skb, 0, nskb->data, headlen);
- nskb->dev = skb->dev;
- skb_reset_mac_header(nskb);
- skb_set_network_header(nskb, skb_network_offset(skb));
- skb_set_transport_header(nskb, skb_transport_offset(skb));
- skb_set_tail_pointer(nskb, headlen);
-
- /* How many frags do we need? */
- for (i = skb_shinfo(skb)->nr_frags - 1; i >= 0; i--) {
- bytes_needed -= skb_frag_size(&skb_shinfo(skb)->frags[i]);
- k++;
- if (bytes_needed <= 0)
- break;
- }
-
- /* Fill the first frag and split it if necessary */
- j = skb_shinfo(skb)->nr_frags - k;
- remaining_page_offset = -bytes_needed;
- skb_fill_page_desc(nskb, 0,
- skb_shinfo(skb)->frags[j].page.p,
- skb_shinfo(skb)->frags[j].page_offset + remaining_page_offset,
- skb_shinfo(skb)->frags[j].size - remaining_page_offset);
-
- skb_frag_ref(skb, j);
-
- /* Fill the rest of the frags */
- for (i = 1; i < k; i++) {
- j = skb_shinfo(skb)->nr_frags - k + i;
-
- skb_fill_page_desc(nskb, i,
- skb_shinfo(skb)->frags[j].page.p,
- skb_shinfo(skb)->frags[j].page_offset,
- skb_shinfo(skb)->frags[j].size);
- skb_frag_ref(skb, j);
- }
- skb_shinfo(nskb)->nr_frags = k;
-
- remaining_headlen = remaining - skb->data_len;
-
- /* headlen contains remaining data? */
- if (remaining_headlen > 0)
- skb_copy_bits(skb, skb->len - remaining, nskb->data + headlen,
- remaining_headlen);
- nskb->len = remaining + headlen;
- nskb->data_len = payload_len - sizeof(struct udphdr) +
- max_t(int, 0, remaining_headlen);
- nskb->protocol = skb->protocol;
- if (nskb->protocol == htons(ETH_P_IP)) {
- ip_hdr(nskb)->id = htons(ntohs(ip_hdr(nskb)->id) +
- skb_shinfo(skb)->gso_segs);
- ip_hdr(nskb)->tot_len =
- htons(payload_len + sizeof(struct iphdr));
- } else {
- ipv6_hdr(nskb)->payload_len = htons(payload_len);
- }
- udp_hdr(nskb)->len = htons(payload_len);
- skb_shinfo(nskb)->gso_size = 0;
- nskb->ip_summed = skb->ip_summed;
- nskb->csum_start = skb->csum_start;
- nskb->csum_offset = skb->csum_offset;
- nskb->queue_mapping = skb->queue_mapping;
-}
-
-/* might send skbs and update wqe and pi */
-struct sk_buff *mlx5e_udp_gso_handle_tx_skb(struct net_device *netdev,
- struct mlx5e_txqsq *sq,
- struct sk_buff *skb,
- struct mlx5e_tx_wqe **wqe,
- u16 *pi)
-{
- int payload_len = skb_shinfo(skb)->gso_size + sizeof(struct udphdr);
- int headlen = skb_transport_offset(skb) + sizeof(struct udphdr);
- int remaining = (skb->len - headlen) % skb_shinfo(skb)->gso_size;
- struct sk_buff *nskb;
-
- if (skb->protocol == htons(ETH_P_IP))
- ip_hdr(skb)->tot_len = htons(payload_len + sizeof(struct iphdr));
- else
- ipv6_hdr(skb)->payload_len = htons(payload_len);
- udp_hdr(skb)->len = htons(payload_len);
- if (!remaining)
- return skb;
-
- sq->stats->udp_seg_rem++;
- nskb = alloc_skb(max_t(int, headlen, headlen + remaining - skb->data_len), GFP_ATOMIC);
- if (unlikely(!nskb)) {
- sq->stats->dropped++;
- return NULL;
- }
-
- mlx5e_udp_gso_prepare_last_skb(skb, nskb, remaining);
-
- skb_shinfo(skb)->gso_segs--;
- pskb_trim(skb, skb->len - remaining);
- mlx5e_sq_xmit(sq, skb, *wqe, *pi);
- mlx5e_sq_fetch_wqe(sq, wqe, pi);
- return nskb;
-}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/rxtx.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/rxtx.h
deleted file mode 100644
index ed42699a78b3..000000000000
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/rxtx.h
+++ /dev/null
@@ -1,14 +0,0 @@
-
-#ifndef __MLX5E_EN_ACCEL_RX_TX_H__
-#define __MLX5E_EN_ACCEL_RX_TX_H__
-
-#include <linux/skbuff.h>
-#include "en.h"
-
-struct sk_buff *mlx5e_udp_gso_handle_tx_skb(struct net_device *netdev,
- struct mlx5e_txqsq *sq,
- struct sk_buff *skb,
- struct mlx5e_tx_wqe **wqe,
- u16 *pi);
-
-#endif
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 712b9766485f..dccde18f6170 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -4538,7 +4538,6 @@ static void mlx5e_build_nic_netdev(struct net_device *netdev)
netdev->hw_features |= NETIF_F_HW_VLAN_STAG_TX;
if (mlx5e_vxlan_allowed(mdev) || MLX5_CAP_ETH(mdev, tunnel_stateless_gre)) {
- netdev->hw_features |= NETIF_F_GSO_PARTIAL;
netdev->hw_enc_features |= NETIF_F_IP_CSUM;
netdev->hw_enc_features |= NETIF_F_IPV6_CSUM;
netdev->hw_enc_features |= NETIF_F_TSO;
@@ -4563,6 +4562,11 @@ static void mlx5e_build_nic_netdev(struct net_device *netdev)
NETIF_F_GSO_GRE_CSUM;
}
+ netdev->hw_features |= NETIF_F_GSO_PARTIAL;
+ netdev->gso_partial_features |= NETIF_F_GSO_UDP_L4;
+ netdev->hw_features |= NETIF_F_GSO_UDP_L4;
+ netdev->features |= NETIF_F_GSO_UDP_L4;
+
mlx5_query_port_fcs(mdev, &fcs_supported, &fcs_enabled);
if (fcs_supported)
@@ -4595,9 +4599,6 @@ static void mlx5e_build_nic_netdev(struct net_device *netdev)
netdev->features |= NETIF_F_HIGHDMA;
netdev->features |= NETIF_F_HW_VLAN_STAG_FILTER;
- netdev->features |= NETIF_F_GSO_UDP_L4;
- netdev->hw_features |= NETIF_F_GSO_UDP_L4;
-
netdev->priv_flags |= IFF_UNICAST_FLT;
mlx5e_set_netdev_dev_addr(netdev);
--
2.17.0
^ permalink raw reply related
* Re: Linux 4.18-rc6
From: Linus Torvalds @ 2018-07-23 23:16 UTC (permalink / raw)
To: Guenter Roeck; +Cc: David Miller, Martin Schwidefsky, Linux Kernel Mailing List
In-Reply-To: <20180723212330.GA7441@roeck-us.net>
On Mon, Jul 23, 2018 at 2:23 PM Guenter Roeck <linux@roeck-us.net> wrote:
>
> My patch is also at
>
> https://patchwork.ozlabs.org/patch/937283/
Ah, ok, so that just adds the forward-declaration of 'struct page' in
the right global namespace.
Anyway, I'll just re-order the includes as I suggested, which I think
is the right fix and makes the forward-declaration unnecessary
(although not _wrong_)
Linus
^ permalink raw reply
* Re: [PATCH v2] perf/core: fix a possible deadlock scenario
From: Cong Wang @ 2018-07-23 23:16 UTC (permalink / raw)
To: Peter Zijlstra
Cc: LKML, Ingo Molnar, Linus Torvalds, Arnaldo Carvalho de Melo,
Alexander Shishkin, Jiri Olsa, Namhyung Kim, stable
In-Reply-To: <20180720115217.GQ2494@hirez.programming.kicks-ass.net>
On Fri, Jul 20, 2018 at 4:52 AM Peter Zijlstra <peterz@infradead.org> wrote:
>
> On Thu, Jul 19, 2018 at 12:12:53PM -0700, Cong Wang wrote:
> > hrtimer_cancel() busy-waits for the hrtimer callback to stop,
> > pretty much like del_timer_sync(). This creates a possible deadlock
> > scenario where we hold a spinlock before calling hrtimer_cancel()
> > while in trying to acquire the same spinlock in the callback.
>
> Has this actually been observed?
Without lockdep annotation, it is not easy to observe.
>
> > cpu_clock_event_init():
> > perf_swevent_init_hrtimer():
> > hwc->hrtimer.function = perf_swevent_hrtimer;
> >
> > perf_swevent_hrtimer():
> > __perf_event_overflow():
> > __perf_event_account_interrupt():
> > perf_adjust_period():
> > pmu->stop():
> > cpu_clock_event_stop():
> > perf_swevent_cancel():
> > hrtimer_cancel()
>
> Please explain how a hrtimer event ever gets to perf_adjust_period().
> Last I checked perf_swevent_init_hrtimer() results in attr.freq=0.
Good point.
I thought attr.freq is specified by user-space, but seems
perf_swevent_init_hrtimer() clears it purposely and will not change
after initialization, interesting...
>
> > Getting stuck in an hrtimer is a disaster:
>
> You'll get NMI watchdog splats. Getting stuck in NMI context is far more
> 'interesting :-)
Yes, I did see a stack trace in perf_swevent_hrtimer() which led
me here. But I have to admit among those hundreds of soft lockup's,
I only saw one showing swevent hrtimer backtrace.
Previously I thought this is because of NMI handler race, but Jiri
pointed out the race doesn't exist.
>
> > +#define PERF_EF_NO_WAIT 0x08 /* do not wait when stopping, for
> > + * example, waiting for a timer
> > + */
>
> That's a broken comment style.
It is picked by checkpatch.pl, not me, I chose a different one and got
a complain. :)
Thanks!
^ permalink raw reply
* Re: [PATCH v2] systemd: Allow custom coredump config options
From: Alistair Francis @ 2018-07-23 23:16 UTC (permalink / raw)
To: Andre McCurdy; +Cc: OE Core mailing list
In-Reply-To: <CAJ86T=XuEUrsb8gZAXmh3m+sXFpU2T=bXqNRhjLYUNCH9DNtDg@mail.gmail.com>
On Wed, Jul 18, 2018 at 4:48 PM, Andre McCurdy <armccurdy@gmail.com> wrote:
> On Wed, Jul 18, 2018 at 3:53 PM, Alistair Francis
> <alistair.francis@wdc.com> wrote:
>> If the user has enabled coredump let's allow them to customise the
>> config options.
>>
>> Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
>> ---
>> meta/recipes-core/systemd/systemd_239.bb | 12 ++++++++++++
>> 1 file changed, 12 insertions(+)
>>
>> diff --git a/meta/recipes-core/systemd/systemd_239.bb b/meta/recipes-core/systemd/systemd_239.bb
>> index 7822548993..a625d060a9 100644
>> --- a/meta/recipes-core/systemd/systemd_239.bb
>> +++ b/meta/recipes-core/systemd/systemd_239.bb
>> @@ -279,6 +279,18 @@ do_install() {
>> chown polkitd:root ${D}${datadir}/polkit-1/rules.d
>> fi
>> fi
>> +
>> + # If coredump was enabled, enable it in the config.
>> + # This just sets the default, but can be used to customise the values.
>> + if ${@bb.utils.contains('PACKAGECONFIG', 'coredump', 'true', 'false', d)}; then
>> + sed 's|#Storage.*|Storage=external|g' -i ${D}${sysconfdir}/systemd/coredump.conf
>> + sed 's|#Compress.*|Compress=yes|g' -i ${D}${sysconfdir}/systemd/coredump.conf
>> + sed 's|#ProcessSizeMax.*|ProcessSizeMax=2G|g' -i ${D}${sysconfdir}/systemd/coredump.conf
>> + sed 's|#ExternalSizeMax.*|ExternalSizeMax=2G|g' -i ${D}${sysconfdir}/systemd/coredump.conf
>> + sed 's|#JournalSizeMax.*|JournalSizeMax=767M|g' -i ${D}${sysconfdir}/systemd/coredump.conf
>> + sed 's|#MaxUse.*|MaxUse=|g' -i ${D}${sysconfdir}/systemd/coredump.conf
>> + sed 's|#KeepFree.*|KeepFree=|g' -i ${D}${sysconfdir}/systemd/coredump.conf
>
> How does this allow the user to customise the config options?
They can edit the sed command line and have different options.
>
> By forcing the options to the _current_ default values you are
> creating a maintenance problem - ie you'll still be forcing these
> values even if/when the upstream defaults are changed.
That is a good point, is there a usual way to allow edits to config
files like this?
Alistair
>
>> + fi
>> }
>>
>>
>> --
>> 2.17.1
>>
>> --
>> _______________________________________________
>> Openembedded-core mailing list
>> Openembedded-core@lists.openembedded.org
>> http://lists.openembedded.org/mailman/listinfo/openembedded-core
> --
> _______________________________________________
> Openembedded-core mailing list
> Openembedded-core@lists.openembedded.org
> http://lists.openembedded.org/mailman/listinfo/openembedded-core
^ permalink raw reply
* Re: Linux 4.18-rc6
From: Linus Torvalds @ 2018-07-23 23:17 UTC (permalink / raw)
To: Guenter Roeck; +Cc: David Miller, Martin Schwidefsky, Linux Kernel Mailing List
In-Reply-To: <20180723212330.GA7441@roeck-us.net>
On Mon, Jul 23, 2018 at 2:23 PM Guenter Roeck <linux@roeck-us.net> wrote:
>
> >
> > Martin - can we just remove the
> >
> > select HAVE_GCC_PLUGINS
> >
> > from the s390 Kconfig file (or perhaps add "if BROKEN" or something to
> > disable it).
> >
> > Because if it's not getting fixed, it shouldn't be exposed.
> >
> The problem only affects 4.18 - the code has been rearranged in -next.
> Only, in my builders, I can't disable a flag for individual releases,
> so I just disabled it completely for s390.
Well, I'm not going to release a 4.18 with a known problem, so in 4.18
this *will* be disabled if it's not fixed.
The fact that it might be fixed in linux-next is entirely immaterial
to the release of 4.18.
Linus
^ permalink raw reply
* Re: [PATCH v5] PCI: Check for PCIe downtraining conditions
From: Jakub Kicinski @ 2018-07-23 22:14 UTC (permalink / raw)
To: Tal Gilboa
Cc: Alexandru Gagniuc, linux-pci@vger.kernel.org, bhelgaas@google.com,
keith.busch@intel.com, alex_gagniuc@dellteam.com,
austin_bolen@dell.com, shyam_iyer@dell.com,
jeffrey.t.kirsher@intel.com, ariel.elior@cavium.com,
michael.chan@broadcom.com, ganeshgr@chelsio.com, Tariq Toukan,
airlied@gmail.com, alexander.deucher@amd.com,
mike.marciniszyn@intel.com, linux-kernel@vger.kernel.org
In-Reply-To: <b9a6b3b8-56cf-8bb5-2568-071ccd8a4f28@mellanox.com>
On Tue, 24 Jul 2018 00:52:22 +0300, Tal Gilboa wrote:
> On 7/24/2018 12:01 AM, Jakub Kicinski wrote:
> > On Mon, 23 Jul 2018 15:03:38 -0500, Alexandru Gagniuc wrote:
> >> PCIe downtraining happens when both the device and PCIe port are
> >> capable of a larger bus width or higher speed than negotiated.
> >> Downtraining might be indicative of other problems in the system, and
> >> identifying this from userspace is neither intuitive, nor
> >> straightforward.
> >>
> >> The easiest way to detect this is with pcie_print_link_status(),
> >> since the bottleneck is usually the link that is downtrained. It's not
> >> a perfect solution, but it works extremely well in most cases.
> >>
> >> Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
> >> ---
> >>
> >> For the sake of review, I've created a __pcie_print_link_status() which
> >> takes a 'verbose' argument. If we agree want to go this route, and update
> >> the users of pcie_print_link_status(), I can split this up in two patches.
> >> I prefer just printing this information in the core functions, and letting
> >> drivers not have to worry about this. Though there seems to be strong for
> >> not going that route, so here it goes:
> >
> > FWIW the networking drivers print PCIe BW because sometimes the network
> > bandwidth is simply over-provisioned on multi port cards, e.g. 80Gbps
> > card on a x8 link.
> >
> > Sorry to bike shed, but currently the networking cards print the info
> > during probe. Would it make sense to move your message closer to probe
> > time? Rather than when device is added. If driver structure is
> > available, we could also consider adding a boolean to struct pci_driver
> > to indicate if driver wants the verbose message? This way we avoid
> > duplicated prints.
> >
> > I have no objection to current patch, it LGTM. Just a thought.
>
> I don't see the reason for having two functions. What's the problem with
> adding the verbose argument to the original function?
IMHO it's reasonable to keep the default parameter to what 90% of users
want by a means on a wrapper. The non-verbose output is provided by
the core already for all devices.
What do you think of my proposal above Tal? That would make the extra
wrapper unnecessary since the verbose parameter would be part of the
driver structure, and it would avoid the duplicated output.
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.