/* SPDX-License-Identifier: BSD-3-Clause
- * Copyright(c) 2019-2020 Broadcom
+ * Copyright(c) 2019-2021 Broadcom
* All rights reserved.
*/
#include <rte_flow.h>
#include <rte_flow_driver.h>
#include <rte_tailq.h>
+#include <rte_spinlock.h>
+#include "bnxt.h"
#include "bnxt_ulp.h"
#include "bnxt_tf_common.h"
-#include "bnxt.h"
#include "tf_core.h"
#include "tf_ext_flow_handle.h"
-#include "ulp_template_db.h"
+#include "ulp_template_db_enum.h"
#include "ulp_template_struct.h"
#include "ulp_mark_mgr.h"
+#include "ulp_fc_mgr.h"
#include "ulp_flow_db.h"
+#include "ulp_mapper.h"
+#include "ulp_port_db.h"
+#include "ulp_tun.h"
+#include "ulp_ha_mgr.h"
+#include "bnxt_tf_pmd_shim.h"
/* Linked list of all TF sessions. */
STAILQ_HEAD(, bnxt_ulp_session_state) bnxt_ulp_session_list =
/* Mutex to synchronize bnxt_ulp_session_list operations. */
static pthread_mutex_t bnxt_ulp_global_mutex = PTHREAD_MUTEX_INITIALIZER;
+/* Spin lock to protect context global list */
+rte_spinlock_t bnxt_ulp_ctxt_lock;
+TAILQ_HEAD(cntx_list_entry_list, ulp_context_list_entry);
+static struct cntx_list_entry_list ulp_cntx_list =
+ TAILQ_HEAD_INITIALIZER(ulp_cntx_list);
+
+/* Static function declarations */
+static int32_t bnxt_ulp_cntxt_list_init(void);
+static int32_t bnxt_ulp_cntxt_list_add(struct bnxt_ulp_context *ulp_ctx);
+static void bnxt_ulp_cntxt_list_del(struct bnxt_ulp_context *ulp_ctx);
+
/*
* Allow the deletion of context only for the bnxt device that
- * created the session
- * TBD - The implementation of the function should change to
- * using the reference count once tf_session_attach functionality
- * is fixed.
+ * created the session.
*/
bool
-ulp_ctx_deinit_allowed(void *ptr)
+ulp_ctx_deinit_allowed(struct bnxt_ulp_context *ulp_ctx)
{
- struct bnxt *bp = (struct bnxt *)ptr;
+ if (!ulp_ctx || !ulp_ctx->cfg_data)
+ return false;
- if (!bp)
+ if (!ulp_ctx->cfg_data->ref_cnt) {
+ BNXT_TF_DBG(DEBUG, "ulp ctx shall initiate deinit\n");
+ return true;
+ }
+
+ return false;
+}
+
+static int32_t
+bnxt_ulp_devid_get(struct bnxt *bp,
+ enum bnxt_ulp_device_id *ulp_dev_id)
+{
+ if (BNXT_CHIP_P5(bp)) {
+ *ulp_dev_id = BNXT_ULP_DEVICE_ID_THOR;
return 0;
+ }
+
+ if (BNXT_STINGRAY(bp))
+ *ulp_dev_id = BNXT_ULP_DEVICE_ID_STINGRAY;
+ else
+ /* Assuming Whitney */
+ *ulp_dev_id = BNXT_ULP_DEVICE_ID_WH_PLUS;
+
+ return 0;
+}
+
+struct bnxt_ulp_app_capabilities_info *
+bnxt_ulp_app_cap_list_get(uint32_t *num_entries)
+{
+ if (!num_entries)
+ return NULL;
+ *num_entries = BNXT_ULP_APP_CAP_TBL_MAX_SZ;
+ return ulp_app_cap_info_list;
+}
+
+static struct bnxt_ulp_resource_resv_info *
+bnxt_ulp_app_resource_resv_list_get(uint32_t *num_entries)
+{
+ if (num_entries == NULL)
+ return NULL;
+ *num_entries = BNXT_ULP_APP_RESOURCE_RESV_LIST_MAX_SZ;
+ return ulp_app_resource_resv_list;
+}
+
+struct bnxt_ulp_resource_resv_info *
+bnxt_ulp_resource_resv_list_get(uint32_t *num_entries)
+{
+ if (!num_entries)
+ return NULL;
+ *num_entries = BNXT_ULP_RESOURCE_RESV_LIST_MAX_SZ;
+ return ulp_resource_resv_list;
+}
+
+struct bnxt_ulp_glb_resource_info *
+bnxt_ulp_app_glb_resource_info_list_get(uint32_t *num_entries)
+{
+ if (!num_entries)
+ return NULL;
+ *num_entries = BNXT_ULP_APP_GLB_RESOURCE_TBL_MAX_SZ;
+ return ulp_app_glb_resource_tbl;
+}
+
+static int32_t
+bnxt_ulp_named_resources_calc(struct bnxt_ulp_context *ulp_ctx,
+ struct bnxt_ulp_glb_resource_info *info,
+ uint32_t num,
+ struct tf_session_resources *res)
+{
+ uint32_t dev_id = BNXT_ULP_DEVICE_ID_LAST, res_type, i;
+ enum tf_dir dir;
+ uint8_t app_id;
+ int32_t rc = 0;
+
+ if (ulp_ctx == NULL || info == NULL || res == NULL || num == 0) {
+ BNXT_TF_DBG(ERR, "Invalid parms to named resources calc.\n");
+ return -EINVAL;
+ }
+
+ rc = bnxt_ulp_cntxt_app_id_get(ulp_ctx, &app_id);
+ if (rc) {
+ BNXT_TF_DBG(ERR, "Unable to get the app id from ulp.\n");
+ return -EINVAL;
+ }
+
+ rc = bnxt_ulp_cntxt_dev_id_get(ulp_ctx, &dev_id);
+ if (rc) {
+ BNXT_TF_DBG(ERR, "Unable to get the dev id from ulp.\n");
+ return -EINVAL;
+ }
+
+ for (i = 0; i < num; i++) {
+ if (dev_id != info[i].device_id || app_id != info[i].app_id)
+ continue;
+ dir = info[i].direction;
+ res_type = info[i].resource_type;
+
+ switch (info[i].resource_func) {
+ case BNXT_ULP_RESOURCE_FUNC_IDENTIFIER:
+ res->ident_cnt[dir].cnt[res_type]++;
+ break;
+ case BNXT_ULP_RESOURCE_FUNC_INDEX_TABLE:
+ res->tbl_cnt[dir].cnt[res_type]++;
+ break;
+ case BNXT_ULP_RESOURCE_FUNC_TCAM_TABLE:
+ res->tcam_cnt[dir].cnt[res_type]++;
+ break;
+ case BNXT_ULP_RESOURCE_FUNC_EM_TABLE:
+ res->em_cnt[dir].cnt[res_type]++;
+ break;
+ default:
+ BNXT_TF_DBG(ERR, "Unknown resource func (0x%x)\n,",
+ info[i].resource_func);
+ continue;
+ }
+ }
+
+ return 0;
+}
+
+static int32_t
+bnxt_ulp_unnamed_resources_calc(struct bnxt_ulp_context *ulp_ctx,
+ struct bnxt_ulp_resource_resv_info *info,
+ uint32_t num,
+ struct tf_session_resources *res)
+{
+ uint32_t dev_id, res_type, i;
+ enum tf_dir dir;
+ uint8_t app_id;
+ int32_t rc = 0;
+
+ if (ulp_ctx == NULL || res == NULL || info == NULL || num == 0) {
+ BNXT_TF_DBG(ERR, "Invalid arguments to get resources.\n");
+ return -EINVAL;
+ }
+
+ rc = bnxt_ulp_cntxt_app_id_get(ulp_ctx, &app_id);
+ if (rc) {
+ BNXT_TF_DBG(ERR, "Unable to get the app id from ulp.\n");
+ return -EINVAL;
+ }
- if (&bp->tfp == bp->ulp_ctx.g_tfp)
- return 1;
+ rc = bnxt_ulp_cntxt_dev_id_get(ulp_ctx, &dev_id);
+ if (rc) {
+ BNXT_TF_DBG(ERR, "Unable to get the dev id from ulp.\n");
+ return -EINVAL;
+ }
+
+ for (i = 0; i < num; i++) {
+ if (app_id != info[i].app_id || dev_id != info[i].device_id)
+ continue;
+ dir = info[i].direction;
+ res_type = info[i].resource_type;
+
+ switch (info[i].resource_func) {
+ case BNXT_ULP_RESOURCE_FUNC_IDENTIFIER:
+ res->ident_cnt[dir].cnt[res_type] = info[i].count;
+ break;
+ case BNXT_ULP_RESOURCE_FUNC_INDEX_TABLE:
+ res->tbl_cnt[dir].cnt[res_type] = info[i].count;
+ break;
+ case BNXT_ULP_RESOURCE_FUNC_TCAM_TABLE:
+ res->tcam_cnt[dir].cnt[res_type] = info[i].count;
+ break;
+ case BNXT_ULP_RESOURCE_FUNC_EM_TABLE:
+ res->em_cnt[dir].cnt[res_type] = info[i].count;
+ break;
+ default:
+ break;
+ }
+ }
+ return 0;
+}
+
+static int32_t
+bnxt_ulp_tf_resources_get(struct bnxt_ulp_context *ulp_ctx,
+ struct tf_session_resources *res)
+{
+ struct bnxt_ulp_resource_resv_info *unnamed = NULL;
+ uint32_t unum;
+ int32_t rc = 0;
+
+ if (ulp_ctx == NULL || res == NULL) {
+ BNXT_TF_DBG(ERR, "Invalid arguments to get resources.\n");
+ return -EINVAL;
+ }
+
+ unnamed = bnxt_ulp_resource_resv_list_get(&unum);
+ if (unnamed == NULL) {
+ BNXT_TF_DBG(ERR, "Unable to get resource resv list.\n");
+ return -EINVAL;
+ }
+
+ rc = bnxt_ulp_unnamed_resources_calc(ulp_ctx, unnamed, unum, res);
+ if (rc)
+ BNXT_TF_DBG(ERR, "Unable to calc resources for session.\n");
+
+ return rc;
+}
+
+static int32_t
+bnxt_ulp_tf_shared_session_resources_get(struct bnxt_ulp_context *ulp_ctx,
+ struct tf_session_resources *res)
+{
+ struct bnxt_ulp_resource_resv_info *unnamed;
+ struct bnxt_ulp_glb_resource_info *named;
+ uint32_t unum, nnum;
+ int32_t rc;
+
+ if (ulp_ctx == NULL || res == NULL) {
+ BNXT_TF_DBG(ERR, "Invalid arguments to get resources.\n");
+ return -EINVAL;
+ }
+
+ /* Make sure the resources are zero before accumulating. */
+ memset(res, 0, sizeof(struct tf_session_resources));
+
+ /*
+ * Shared resources are comprised of both named and unnamed resources.
+ * First get the unnamed counts, and then add the named to the result.
+ */
+ /* Get the baseline counts */
+ unnamed = bnxt_ulp_app_resource_resv_list_get(&unum);
+ if (unnamed == NULL) {
+ BNXT_TF_DBG(ERR, "Unable to get shared resource resv list.\n");
+ return -EINVAL;
+ }
+ rc = bnxt_ulp_unnamed_resources_calc(ulp_ctx, unnamed, unum, res);
+ if (rc) {
+ BNXT_TF_DBG(ERR, "Unable to calc resources for shared session.\n");
+ return -EINVAL;
+ }
+
+ /* Get the named list and add the totals */
+ named = bnxt_ulp_app_glb_resource_info_list_get(&nnum);
+ if (named == NULL) {
+ BNXT_TF_DBG(ERR, "Unable to get app global resource list\n");
+ return -EINVAL;
+ }
+ rc = bnxt_ulp_named_resources_calc(ulp_ctx, named, nnum, res);
+ if (rc)
+ BNXT_TF_DBG(ERR, "Unable to calc named resources\n");
+
+ return rc;
+}
+
+int32_t
+bnxt_ulp_cntxt_app_caps_init(struct bnxt_ulp_context *ulp_ctx,
+ uint8_t app_id, uint32_t dev_id)
+{
+ struct bnxt_ulp_app_capabilities_info *info;
+ uint32_t num = 0;
+ uint16_t i;
+ bool found = false;
+
+ if (ULP_APP_DEV_UNSUPPORTED_ENABLED(ulp_ctx->cfg_data->ulp_flags)) {
+ BNXT_TF_DBG(ERR, "APP ID %d, Device ID: 0x%x not supported.\n",
+ app_id, dev_id);
+ return -EINVAL;
+ }
+
+ info = bnxt_ulp_app_cap_list_get(&num);
+ if (!info || !num) {
+ BNXT_TF_DBG(ERR, "Failed to get app capabilities.\n");
+ return -EINVAL;
+ }
+
+ for (i = 0; i < num; i++) {
+ if (info[i].app_id != app_id || info[i].device_id != dev_id)
+ continue;
+ found = true;
+ if (info[i].flags & BNXT_ULP_APP_CAP_SHARED_EN)
+ ulp_ctx->cfg_data->ulp_flags |=
+ BNXT_ULP_SHARED_SESSION_ENABLED;
+ if (info[i].flags & BNXT_ULP_APP_CAP_HOT_UPGRADE_EN)
+ ulp_ctx->cfg_data->ulp_flags |=
+ BNXT_ULP_HIGH_AVAIL_ENABLED;
+ if (info[i].flags & BNXT_ULP_APP_CAP_UNICAST_ONLY)
+ ulp_ctx->cfg_data->ulp_flags |=
+ BNXT_ULP_APP_UNICAST_ONLY;
+ }
+ if (!found) {
+ BNXT_TF_DBG(ERR, "APP ID %d, Device ID: 0x%x not supported.\n",
+ app_id, dev_id);
+ ulp_ctx->cfg_data->ulp_flags |= BNXT_ULP_APP_DEV_UNSUPPORTED;
+ return -EINVAL;
+ }
return 0;
}
+static void
+ulp_ctx_shared_session_close(struct bnxt *bp,
+ struct bnxt_ulp_session_state *session)
+{
+ struct tf *tfp;
+ int32_t rc;
+
+ if (!bnxt_ulp_cntxt_shared_session_enabled(bp->ulp_ctx))
+ return;
+
+ tfp = bnxt_ulp_cntxt_shared_tfp_get(bp->ulp_ctx);
+ if (!tfp) {
+ /*
+ * Log it under debug since this is likely a case of the
+ * shared session not being created. For example, a failed
+ * initialization.
+ */
+ BNXT_TF_DBG(DEBUG, "Failed to get shared tfp on close.\n");
+ return;
+ }
+ rc = tf_close_session(tfp);
+ if (rc)
+ BNXT_TF_DBG(ERR, "Failed to close the shared session rc=%d.\n",
+ rc);
+ (void)bnxt_ulp_cntxt_shared_tfp_set(bp->ulp_ctx, NULL);
+
+ session->g_shared_tfp.session = NULL;
+}
+
+static int32_t
+ulp_ctx_shared_session_open(struct bnxt *bp,
+ struct bnxt_ulp_session_state *session)
+{
+ struct rte_eth_dev *ethdev = bp->eth_dev;
+ struct tf_session_resources *resources;
+ struct tf_open_session_parms parms;
+ size_t copy_nbytes;
+ uint32_t ulp_dev_id = BNXT_ULP_DEVICE_ID_LAST;
+ int32_t rc = 0;
+
+ /* only perform this if shared session is enabled. */
+ if (!bnxt_ulp_cntxt_shared_session_enabled(bp->ulp_ctx))
+ return 0;
+
+ memset(&parms, 0, sizeof(parms));
+
+ rc = rte_eth_dev_get_name_by_port(ethdev->data->port_id,
+ parms.ctrl_chan_name);
+ if (rc) {
+ BNXT_TF_DBG(ERR, "Invalid port %d, rc = %d\n",
+ ethdev->data->port_id, rc);
+ return rc;
+ }
+ resources = &parms.resources;
+
+ /*
+ * Need to account for size of ctrl_chan_name and 1 extra for Null
+ * terminator
+ */
+ copy_nbytes = sizeof(parms.ctrl_chan_name) -
+ strlen(parms.ctrl_chan_name) - 1;
+
+ /*
+ * Build the ctrl_chan_name with shared token.
+ * When HA is enabled, the WC TCAM needs extra management by the core,
+ * so add the wc_tcam string to the control channel.
+ */
+ if (bnxt_ulp_cntxt_ha_enabled(bp->ulp_ctx))
+ strncat(parms.ctrl_chan_name, "-tf_shared-wc_tcam",
+ copy_nbytes);
+ else
+ strncat(parms.ctrl_chan_name, "-tf_shared", copy_nbytes);
+
+ rc = bnxt_ulp_tf_shared_session_resources_get(bp->ulp_ctx, resources);
+ if (rc)
+ return rc;
+
+ rc = bnxt_ulp_cntxt_dev_id_get(bp->ulp_ctx, &ulp_dev_id);
+ if (rc) {
+ BNXT_TF_DBG(ERR, "Unable to get device id from ulp.\n");
+ return rc;
+ }
+
+ switch (ulp_dev_id) {
+ case BNXT_ULP_DEVICE_ID_WH_PLUS:
+ parms.device_type = TF_DEVICE_TYPE_WH;
+ break;
+ case BNXT_ULP_DEVICE_ID_STINGRAY:
+ parms.device_type = TF_DEVICE_TYPE_SR;
+ break;
+ case BNXT_ULP_DEVICE_ID_THOR:
+ parms.device_type = TF_DEVICE_TYPE_THOR;
+ break;
+ default:
+ BNXT_TF_DBG(ERR, "Unable to determine dev for opening session.\n");
+ return rc;
+ }
+
+ parms.shadow_copy = true;
+ parms.bp = bp;
+
+ /*
+ * Open the session here, but the collect the resources during the
+ * mapper initialization.
+ */
+ rc = tf_open_session(&bp->tfp_shared, &parms);
+ if (rc)
+ return rc;
+
+ if (parms.shared_session_creator)
+ BNXT_TF_DBG(DEBUG, "Shared session creator.\n");
+ else
+ BNXT_TF_DBG(DEBUG, "Shared session attached.\n");
+
+ /* Save the shared session in global data */
+ if (!session->g_shared_tfp.session)
+ session->g_shared_tfp.session = bp->tfp_shared.session;
+
+ rc = bnxt_ulp_cntxt_shared_tfp_set(bp->ulp_ctx, &bp->tfp_shared);
+ if (rc)
+ BNXT_TF_DBG(ERR, "Failed to add shared tfp to ulp (%d)\n", rc);
+
+ return rc;
+}
+
+static int32_t
+ulp_ctx_shared_session_attach(struct bnxt *bp,
+ struct bnxt_ulp_session_state *session)
+{
+ int32_t rc = 0;
+
+ /* Simply return success if shared session not enabled */
+ if (bnxt_ulp_cntxt_shared_session_enabled(bp->ulp_ctx)) {
+ bp->tfp_shared.session = session->g_shared_tfp.session;
+ rc = ulp_ctx_shared_session_open(bp, session);
+ }
+
+ return rc;
+}
+
+static void
+ulp_ctx_shared_session_detach(struct bnxt *bp)
+{
+ if (bnxt_ulp_cntxt_shared_session_enabled(bp->ulp_ctx)) {
+ if (bp->tfp_shared.session) {
+ tf_close_session(&bp->tfp_shared);
+ bp->tfp_shared.session = NULL;
+ }
+ }
+}
+
/*
* Initialize an ULP session.
* An ULP session will contain all the resources needed to support rte flow
struct rte_eth_dev *ethdev = bp->eth_dev;
int32_t rc = 0;
struct tf_open_session_parms params;
+ struct tf_session_resources *resources;
+ uint32_t ulp_dev_id = BNXT_ULP_DEVICE_ID_LAST;
memset(¶ms, 0, sizeof(params));
return rc;
}
+ params.shadow_copy = true;
+
+ rc = bnxt_ulp_cntxt_dev_id_get(bp->ulp_ctx, &ulp_dev_id);
+ if (rc) {
+ BNXT_TF_DBG(ERR, "Unable to get device id from ulp.\n");
+ return rc;
+ }
+
+ switch (ulp_dev_id) {
+ case BNXT_ULP_DEVICE_ID_WH_PLUS:
+ params.device_type = TF_DEVICE_TYPE_WH;
+ break;
+ case BNXT_ULP_DEVICE_ID_STINGRAY:
+ params.device_type = TF_DEVICE_TYPE_SR;
+ break;
+ case BNXT_ULP_DEVICE_ID_THOR:
+ params.device_type = TF_DEVICE_TYPE_THOR;
+ break;
+ default:
+ BNXT_TF_DBG(ERR, "Unable to determine device for opening session.\n");
+ return rc;
+ }
+
+ resources = ¶ms.resources;
+ rc = bnxt_ulp_tf_resources_get(bp->ulp_ctx, resources);
+ if (rc)
+ return rc;
+
+ params.bp = bp;
rc = tf_open_session(&bp->tfp, ¶ms);
if (rc) {
BNXT_TF_DBG(ERR, "Failed to open TF session - %s, rc = %d\n",
params.ctrl_chan_name, rc);
return -EINVAL;
}
- session->session_opened = 1;
- session->g_tfp = &bp->tfp;
+ if (!session->session_opened) {
+ session->session_opened = 1;
+ session->g_tfp = rte_zmalloc("bnxt_ulp_session_tfp",
+ sizeof(struct tf), 0);
+ session->g_tfp->session = bp->tfp.session;
+ }
return rc;
}
if (session->session_opened)
tf_close_session(&bp->tfp);
session->session_opened = 0;
+ rte_free(session->g_tfp);
session->g_tfp = NULL;
- bp->ulp_ctx.g_tfp = NULL;
}
static void
uint32_t dev_id;
int rc;
- rc = bnxt_ulp_cntxt_dev_id_get(&bp->ulp_ctx, &dev_id);
+ rc = bnxt_ulp_cntxt_dev_id_get(bp->ulp_ctx, &dev_id);
if (rc)
/* TBD: For now, just use default. */
dparms = 0;
else
dparms = bnxt_ulp_device_params_get(dev_id);
+ /*
+ * Set the flush timer for EEM entries. The value is in 100ms intervals,
+ * so 100 is 10s.
+ */
+ params->hw_flow_cache_flush_timer = 100;
+
if (!dparms) {
params->rx_max_key_sz_in_bits = BNXT_ULP_DFLT_RX_MAX_KEY;
params->rx_max_action_entry_sz_in_bits =
BNXT_ULP_DFLT_RX_MAX_ACTN_ENTRY;
params->rx_mem_size_in_mb = BNXT_ULP_DFLT_RX_MEM;
params->rx_num_flows_in_k = BNXT_ULP_RX_NUM_FLOWS;
- params->rx_tbl_if_id = BNXT_ULP_RX_TBL_IF_ID;
params->tx_max_key_sz_in_bits = BNXT_ULP_DFLT_TX_MAX_KEY;
params->tx_max_action_entry_sz_in_bits =
BNXT_ULP_DFLT_TX_MAX_ACTN_ENTRY;
params->tx_mem_size_in_mb = BNXT_ULP_DFLT_TX_MEM;
params->tx_num_flows_in_k = BNXT_ULP_TX_NUM_FLOWS;
- params->tx_tbl_if_id = BNXT_ULP_TX_TBL_IF_ID;
} else {
params->rx_max_key_sz_in_bits = BNXT_ULP_DFLT_RX_MAX_KEY;
params->rx_max_action_entry_sz_in_bits =
BNXT_ULP_DFLT_RX_MAX_ACTN_ENTRY;
params->rx_mem_size_in_mb = BNXT_ULP_DFLT_RX_MEM;
- params->rx_num_flows_in_k = dparms->num_flows / (1024);
- params->rx_tbl_if_id = BNXT_ULP_RX_TBL_IF_ID;
+ params->rx_num_flows_in_k =
+ dparms->ext_flow_db_num_entries / 1024;
params->tx_max_key_sz_in_bits = BNXT_ULP_DFLT_TX_MAX_KEY;
params->tx_max_action_entry_sz_in_bits =
BNXT_ULP_DFLT_TX_MAX_ACTN_ENTRY;
params->tx_mem_size_in_mb = BNXT_ULP_DFLT_TX_MEM;
- params->tx_num_flows_in_k = dparms->num_flows / (1024);
- params->tx_tbl_if_id = BNXT_ULP_TX_TBL_IF_ID;
+ params->tx_num_flows_in_k =
+ dparms->ext_flow_db_num_entries / 1024;
}
+ BNXT_TF_DBG(INFO, "Table Scope initialized with %uK flows.\n",
+ params->rx_num_flows_in_k);
}
/* Initialize Extended Exact Match host memory. */
ulp_eem_tbl_scope_init(struct bnxt *bp)
{
struct tf_alloc_tbl_scope_parms params = {0};
+ struct bnxt_ulp_device_params *dparms;
+ enum bnxt_ulp_flow_mem_type mtype;
+ uint32_t dev_id;
int rc;
- bnxt_init_tbl_scope_parms(bp, ¶ms);
+ /* Get the dev specific number of flows that needed to be supported. */
+ if (bnxt_ulp_cntxt_dev_id_get(bp->ulp_ctx, &dev_id)) {
+ BNXT_TF_DBG(ERR, "Invalid device id\n");
+ return -EINVAL;
+ }
+
+ dparms = bnxt_ulp_device_params_get(dev_id);
+ if (!dparms) {
+ BNXT_TF_DBG(ERR, "could not fetch the device params\n");
+ return -ENODEV;
+ }
+
+ if (bnxt_ulp_cntxt_mem_type_get(bp->ulp_ctx, &mtype))
+ return -EINVAL;
+ if (mtype != BNXT_ULP_FLOW_MEM_TYPE_EXT) {
+ BNXT_TF_DBG(INFO, "Table Scope alloc is not required\n");
+ return 0;
+ }
+ bnxt_init_tbl_scope_parms(bp, ¶ms);
rc = tf_alloc_tbl_scope(&bp->tfp, ¶ms);
if (rc) {
BNXT_TF_DBG(ERR, "Unable to allocate eem table scope rc = %d\n",
rc);
return rc;
}
-
- rc = bnxt_ulp_cntxt_tbl_scope_id_set(&bp->ulp_ctx, params.tbl_scope_id);
+ rc = bnxt_ulp_cntxt_tbl_scope_id_set(bp->ulp_ctx, params.tbl_scope_id);
if (rc) {
BNXT_TF_DBG(ERR, "Unable to set table scope id\n");
return rc;
struct tf_free_tbl_scope_parms params = {0};
struct tf *tfp;
int32_t rc = 0;
+ struct bnxt_ulp_device_params *dparms;
+ enum bnxt_ulp_flow_mem_type mtype;
+ uint32_t dev_id;
if (!ulp_ctx || !ulp_ctx->cfg_data)
return -EINVAL;
- /* Free the resources for the last device */
- if (!ulp_ctx_deinit_allowed(bp))
- return rc;
-
- tfp = bnxt_ulp_cntxt_tfp_get(ulp_ctx);
+ tfp = bnxt_ulp_cntxt_tfp_get(ulp_ctx, BNXT_ULP_SHARED_SESSION_NO);
if (!tfp) {
BNXT_TF_DBG(ERR, "Failed to get the truflow pointer\n");
return -EINVAL;
}
+ /* Get the dev specific number of flows that needed to be supported. */
+ if (bnxt_ulp_cntxt_dev_id_get(bp->ulp_ctx, &dev_id)) {
+ BNXT_TF_DBG(ERR, "Invalid device id\n");
+ return -EINVAL;
+ }
+
+ dparms = bnxt_ulp_device_params_get(dev_id);
+ if (!dparms) {
+ BNXT_TF_DBG(ERR, "could not fetch the device params\n");
+ return -ENODEV;
+ }
+
+ if (bnxt_ulp_cntxt_mem_type_get(ulp_ctx, &mtype))
+ return -EINVAL;
+ if (mtype != BNXT_ULP_FLOW_MEM_TYPE_EXT) {
+ BNXT_TF_DBG(INFO, "Table Scope free is not required\n");
+ return 0;
+ }
+
rc = bnxt_ulp_cntxt_tbl_scope_id_get(ulp_ctx, ¶ms.tbl_scope_id);
if (rc) {
BNXT_TF_DBG(ERR, "Failed to get the table scope id\n");
ulp_ctx_deinit(struct bnxt *bp,
struct bnxt_ulp_session_state *session)
{
- if (!session || !bp) {
- BNXT_TF_DBG(ERR, "Invalid Arguments\n");
- return -EINVAL;
- }
-
/* close the tf session */
ulp_ctx_session_close(bp, session);
+ /* The shared session must be closed last. */
+ ulp_ctx_shared_session_close(bp, session);
+
/* Free the contents */
if (session->cfg_data) {
rte_free(session->cfg_data);
- bp->ulp_ctx.cfg_data = NULL;
+ bp->ulp_ctx->cfg_data = NULL;
session->cfg_data = NULL;
}
return 0;
{
struct bnxt_ulp_data *ulp_data;
int32_t rc = 0;
+ enum bnxt_ulp_device_id devid;
- if (!session || !bp) {
- BNXT_TF_DBG(ERR, "Invalid Arguments\n");
- return -EINVAL;
+ /* Initialize the context entries list */
+ bnxt_ulp_cntxt_list_init();
+
+ /* Add the context to the context entries list */
+ rc = bnxt_ulp_cntxt_list_add(bp->ulp_ctx);
+ if (rc) {
+ BNXT_TF_DBG(ERR, "Failed to add the context list entry\n");
+ return -ENOMEM;
}
/* Allocate memory to hold ulp context data. */
}
/* Increment the ulp context data reference count usage. */
- bp->ulp_ctx.cfg_data = ulp_data;
+ bp->ulp_ctx->cfg_data = ulp_data;
session->cfg_data = ulp_data;
ulp_data->ref_cnt++;
+ ulp_data->ulp_flags |= BNXT_ULP_VF_REP_ENABLED;
- /* Open the ulp session. */
- rc = ulp_ctx_session_open(bp, session);
+ rc = bnxt_ulp_devid_get(bp, &devid);
if (rc) {
- (void)ulp_ctx_deinit(bp, session);
- return rc;
+ BNXT_TF_DBG(ERR, "Unable to determine device for ULP init.\n");
+ goto error_deinit;
}
- bnxt_ulp_cntxt_tfp_set(&bp->ulp_ctx, session->g_tfp);
- return rc;
-}
-static int32_t
-ulp_ctx_attach(struct bnxt_ulp_context *ulp_ctx,
- struct bnxt_ulp_session_state *session)
-{
- if (!ulp_ctx || !session) {
- BNXT_TF_DBG(ERR, "Invalid Arguments\n");
- return -EINVAL;
+ rc = bnxt_ulp_cntxt_dev_id_set(bp->ulp_ctx, devid);
+ if (rc) {
+ BNXT_TF_DBG(ERR, "Unable to set device for ULP init.\n");
+ goto error_deinit;
}
- /* Increment the ulp context data reference count usage. */
- ulp_ctx->cfg_data = session->cfg_data;
- ulp_ctx->cfg_data->ref_cnt++;
-
- /* TBD call TF_session_attach. */
- ulp_ctx->g_tfp = session->g_tfp;
- return 0;
-}
+ rc = bnxt_ulp_cntxt_app_id_set(bp->ulp_ctx, bp->app_id);
+ if (rc) {
+ BNXT_TF_DBG(ERR, "Unable to set app_id for ULP init.\n");
+ goto error_deinit;
+ }
+ rc = bnxt_ulp_cntxt_app_caps_init(bp->ulp_ctx, bp->app_id, devid);
+ if (rc) {
+ BNXT_TF_DBG(ERR, "Unable to set caps for app(%x)/dev(%x)\n",
+ bp->app_id, devid);
+ goto error_deinit;
+ }
+
+ /*
+ * Shared session must be created before first regular session but after
+ * the ulp_ctx is valid.
+ */
+ rc = ulp_ctx_shared_session_open(bp, session);
+ if (rc) {
+ BNXT_TF_DBG(ERR, "Unable to open shared session (%d)\n", rc);
+ goto error_deinit;
+ }
+
+ /* Open the ulp session. */
+ rc = ulp_ctx_session_open(bp, session);
+ if (rc)
+ goto error_deinit;
+
+ ulp_tun_tbl_init(ulp_data->tun_tbl);
+
+ bnxt_ulp_cntxt_tfp_set(bp->ulp_ctx, &bp->tfp);
+ return rc;
+
+error_deinit:
+ session->session_opened = 1;
+ (void)ulp_ctx_deinit(bp, session);
+ return rc;
+}
+
+/* The function to initialize ulp dparms with devargs */
static int32_t
-ulp_ctx_detach(struct bnxt *bp,
- struct bnxt_ulp_session_state *session)
+ulp_dparms_init(struct bnxt *bp, struct bnxt_ulp_context *ulp_ctx)
{
- struct bnxt_ulp_context *ulp_ctx;
+ struct bnxt_ulp_device_params *dparms;
+ uint32_t dev_id = BNXT_ULP_DEVICE_ID_LAST;
- if (!bp || !session) {
- BNXT_TF_DBG(ERR, "Invalid Arguments\n");
+ if (!bp->max_num_kflows) {
+ /* Defaults to Internal */
+ bnxt_ulp_cntxt_mem_type_set(ulp_ctx,
+ BNXT_ULP_FLOW_MEM_TYPE_INT);
+ return 0;
+ }
+
+ /* The max_num_kflows were set, so move to external */
+ if (bnxt_ulp_cntxt_mem_type_set(ulp_ctx, BNXT_ULP_FLOW_MEM_TYPE_EXT))
+ return -EINVAL;
+
+ if (bnxt_ulp_cntxt_dev_id_get(ulp_ctx, &dev_id)) {
+ BNXT_TF_DBG(DEBUG, "Failed to get device id\n");
return -EINVAL;
}
- ulp_ctx = &bp->ulp_ctx;
- if (!ulp_ctx->cfg_data)
- return 0;
+ dparms = bnxt_ulp_device_params_get(dev_id);
+ if (!dparms) {
+ BNXT_TF_DBG(DEBUG, "Failed to get device parms\n");
+ return -EINVAL;
+ }
+
+ /* num_flows = max_num_kflows * 1024 */
+ dparms->ext_flow_db_num_entries = bp->max_num_kflows * 1024;
+ /* GFID = 2 * num_flows */
+ dparms->mark_db_gfid_entries = dparms->ext_flow_db_num_entries * 2;
+ BNXT_TF_DBG(DEBUG, "Set the number of flows = %"PRIu64"\n",
+ dparms->ext_flow_db_num_entries);
+
+ return 0;
+}
+
+/* The function to initialize bp flags with truflow features */
+static int32_t
+ulp_dparms_dev_port_intf_update(struct bnxt *bp,
+ struct bnxt_ulp_context *ulp_ctx)
+{
+ enum bnxt_ulp_flow_mem_type mtype;
+
+ if (bnxt_ulp_cntxt_mem_type_get(ulp_ctx, &mtype))
+ return -EINVAL;
+ /* Update the bp flag with gfid flag */
+ if (mtype == BNXT_ULP_FLOW_MEM_TYPE_EXT)
+ bp->flags |= BNXT_FLAG_GFID_ENABLE;
+
+ return 0;
+}
- /* TBD call TF_session_detach */
+static int32_t
+ulp_ctx_attach(struct bnxt *bp,
+ struct bnxt_ulp_session_state *session)
+{
+ int32_t rc = 0;
+ uint32_t flags, dev_id = BNXT_ULP_DEVICE_ID_LAST;
+ uint8_t app_id;
/* Increment the ulp context data reference count usage. */
- if (ulp_ctx->cfg_data->ref_cnt >= 1) {
- ulp_ctx->cfg_data->ref_cnt--;
- if (ulp_ctx_deinit_allowed(bp))
- ulp_ctx_deinit(bp, session);
- ulp_ctx->cfg_data = NULL;
- ulp_ctx->g_tfp = NULL;
- return 0;
+ bp->ulp_ctx->cfg_data = session->cfg_data;
+ bp->ulp_ctx->cfg_data->ref_cnt++;
+
+ /* update the session details in bnxt tfp */
+ bp->tfp.session = session->g_tfp->session;
+
+ /* Add the context to the context entries list */
+ rc = bnxt_ulp_cntxt_list_add(bp->ulp_ctx);
+ if (rc) {
+ BNXT_TF_DBG(ERR, "Failed to add the context list entry\n");
+ return -EINVAL;
+ }
+
+ /*
+ * The supported flag will be set during the init. Use it now to
+ * know if we should go through the attach.
+ */
+ rc = bnxt_ulp_cntxt_app_id_get(bp->ulp_ctx, &app_id);
+ if (rc) {
+ BNXT_TF_DBG(ERR, "Unable to get the app id from ulp.\n");
+ return -EINVAL;
+ }
+
+ rc = bnxt_ulp_cntxt_dev_id_get(bp->ulp_ctx, &dev_id);
+ if (rc) {
+ BNXT_TF_DBG(ERR, "Unable do get the dev_id.\n");
+ return -EINVAL;
+ }
+
+ flags = bp->ulp_ctx->cfg_data->ulp_flags;
+ if (ULP_APP_DEV_UNSUPPORTED_ENABLED(flags)) {
+ BNXT_TF_DBG(ERR, "APP ID %d, Device ID: 0x%x not supported.\n",
+ app_id, dev_id);
+ return -EINVAL;
+ }
+
+ /* Create a TF Client */
+ rc = ulp_ctx_session_open(bp, session);
+ if (rc) {
+ PMD_DRV_LOG(ERR, "Failed to open ctxt session, rc:%d\n", rc);
+ bp->tfp.session = NULL;
+ return rc;
+ }
+
+ bnxt_ulp_cntxt_tfp_set(bp->ulp_ctx, &bp->tfp);
+ return rc;
+}
+
+static void
+ulp_ctx_detach(struct bnxt *bp)
+{
+ if (bp->tfp.session) {
+ tf_close_session(&bp->tfp);
+ bp->tfp.session = NULL;
}
- BNXT_TF_DBG(ERR, "context deatach on invalid data\n");
- return 0;
}
/*
struct rte_pci_device *pci_dev;
struct rte_pci_addr *pci_addr;
struct bnxt_ulp_session_state *session;
+ int rc = 0;
if (!bp)
return NULL;
/* Add it to the queue */
session->pci_info.domain = pci_addr->domain;
session->pci_info.bus = pci_addr->bus;
- pthread_mutex_init(&session->bnxt_ulp_mutex, NULL);
+ rc = pthread_mutex_init(&session->bnxt_ulp_mutex, NULL);
+ if (rc) {
+ BNXT_TF_DBG(ERR, "mutex create failed\n");
+ pthread_mutex_unlock(&bnxt_ulp_global_mutex);
+ return NULL;
+ }
STAILQ_INSERT_TAIL(&bnxt_ulp_session_list,
session, next);
}
}
}
+/*
+ * Internal api to enable NAT feature.
+ * Set set_flag to 1 to set the value or zero to reset the value.
+ * returns 0 on success.
+ */
+static int32_t
+bnxt_ulp_global_cfg_update(struct bnxt *bp,
+ enum tf_dir dir,
+ enum tf_global_config_type type,
+ uint32_t offset,
+ uint32_t value,
+ uint32_t set_flag)
+{
+ uint32_t global_cfg = 0;
+ int rc;
+ struct tf_global_cfg_parms parms = { 0 };
+
+ /* Initialize the params */
+ parms.dir = dir,
+ parms.type = type,
+ parms.offset = offset,
+ parms.config = (uint8_t *)&global_cfg,
+ parms.config_sz_in_bytes = sizeof(global_cfg);
+
+ rc = tf_get_global_cfg(&bp->tfp, &parms);
+ if (rc) {
+ BNXT_TF_DBG(ERR, "Failed to get global cfg 0x%x rc:%d\n",
+ type, rc);
+ return rc;
+ }
+
+ if (set_flag)
+ global_cfg |= value;
+ else
+ global_cfg &= ~value;
+
+ /* SET the register RE_CFA_REG_ACT_TECT */
+ rc = tf_set_global_cfg(&bp->tfp, &parms);
+ if (rc) {
+ BNXT_TF_DBG(ERR, "Failed to set global cfg 0x%x rc:%d\n",
+ type, rc);
+ return rc;
+ }
+ return rc;
+}
+
+/* Internal function to delete all the flows belonging to the given port */
+static void
+bnxt_ulp_flush_port_flows(struct bnxt *bp)
+{
+ uint16_t func_id;
+
+ /* it is assumed that port is either TVF or PF */
+ if (ulp_port_db_port_func_id_get(bp->ulp_ctx,
+ bp->eth_dev->data->port_id,
+ &func_id)) {
+ BNXT_TF_DBG(ERR, "Invalid argument\n");
+ return;
+ }
+ (void)ulp_flow_db_function_flow_flush(bp->ulp_ctx, func_id);
+}
+
+/* Internal function to delete the VFR default flows */
+static void
+bnxt_ulp_destroy_vfr_default_rules(struct bnxt *bp, bool global)
+{
+ struct bnxt_ulp_vfr_rule_info *info;
+ uint16_t port_id;
+ struct rte_eth_dev *vfr_eth_dev;
+ struct bnxt_representor *vfr_bp;
+
+ if (!BNXT_TRUFLOW_EN(bp) || BNXT_ETH_DEV_IS_REPRESENTOR(bp->eth_dev))
+ return;
+
+ if (!bp->ulp_ctx || !bp->ulp_ctx->cfg_data)
+ return;
+
+ /* Delete default rules for all ports */
+ for (port_id = 0; port_id < RTE_MAX_ETHPORTS; port_id++) {
+ info = &bp->ulp_ctx->cfg_data->vfr_rule_info[port_id];
+ if (!info->valid)
+ continue;
+
+ if (!global && info->parent_port_id !=
+ bp->eth_dev->data->port_id)
+ continue;
+
+ /* Destroy the flows */
+ ulp_default_flow_destroy(bp->eth_dev, info->vfr_flow_id);
+ /* Clean up the tx action pointer */
+ vfr_eth_dev = &rte_eth_devices[port_id];
+ if (vfr_eth_dev) {
+ vfr_bp = vfr_eth_dev->data->dev_private;
+ vfr_bp->vfr_tx_cfa_action = 0;
+ }
+ memset(info, 0, sizeof(struct bnxt_ulp_vfr_rule_info));
+ }
+}
+
+/*
+ * When a port is deinit'ed by dpdk. This function is called
+ * and this function clears the ULP context and rest of the
+ * infrastructure associated with it.
+ */
+static void
+bnxt_ulp_deinit(struct bnxt *bp,
+ struct bnxt_ulp_session_state *session)
+{
+ bool ha_enabled;
+
+ if (!bp->ulp_ctx || !bp->ulp_ctx->cfg_data)
+ return;
+
+ ha_enabled = bnxt_ulp_cntxt_ha_enabled(bp->ulp_ctx);
+ if (ha_enabled && session->session_opened) {
+ int32_t rc = ulp_ha_mgr_close(bp->ulp_ctx);
+ if (rc)
+ BNXT_TF_DBG(ERR, "Failed to close HA (%d)\n", rc);
+ }
+
+ /* clean up default flows */
+ bnxt_ulp_destroy_df_rules(bp, true);
+
+ /* clean up default VFR flows */
+ bnxt_ulp_destroy_vfr_default_rules(bp, true);
+
+ /* clean up regular flows */
+ ulp_flow_db_flush_flows(bp->ulp_ctx, BNXT_ULP_FDB_TYPE_REGULAR);
+
+ /* cleanup the eem table scope */
+ ulp_eem_tbl_scope_deinit(bp, bp->ulp_ctx);
+
+ /* cleanup the flow database */
+ ulp_flow_db_deinit(bp->ulp_ctx);
+
+ /* Delete the Mark database */
+ ulp_mark_db_deinit(bp->ulp_ctx);
+
+ /* cleanup the ulp mapper */
+ ulp_mapper_deinit(bp->ulp_ctx);
+
+ /* Delete the Flow Counter Manager */
+ ulp_fc_mgr_deinit(bp->ulp_ctx);
+
+ /* Delete the Port database */
+ ulp_port_db_deinit(bp->ulp_ctx);
+
+ /* Disable NAT feature */
+ (void)bnxt_ulp_global_cfg_update(bp, TF_DIR_RX, TF_TUNNEL_ENCAP,
+ TF_TUNNEL_ENCAP_NAT,
+ BNXT_ULP_NAT_OUTER_MOST_FLAGS, 0);
+
+ (void)bnxt_ulp_global_cfg_update(bp, TF_DIR_TX, TF_TUNNEL_ENCAP,
+ TF_TUNNEL_ENCAP_NAT,
+ BNXT_ULP_NAT_OUTER_MOST_FLAGS, 0);
+
+ /* free the flow db lock */
+ pthread_mutex_destroy(&bp->ulp_ctx->cfg_data->flow_db_lock);
+
+ if (ha_enabled)
+ ulp_ha_mgr_deinit(bp->ulp_ctx);
+
+ /* Delete the ulp context and tf session and free the ulp context */
+ ulp_ctx_deinit(bp, session);
+ BNXT_TF_DBG(DEBUG, "ulp ctx has been deinitialized\n");
+}
+
/*
* When a port is initialized by dpdk. This functions is called
* and this function initializes the ULP context and rest of the
* infrastructure associated with it.
*/
+static int32_t
+bnxt_ulp_init(struct bnxt *bp,
+ struct bnxt_ulp_session_state *session)
+{
+ int rc;
+
+ /* Allocate and Initialize the ulp context. */
+ rc = ulp_ctx_init(bp, session);
+ if (rc) {
+ BNXT_TF_DBG(ERR, "Failed to create the ulp context\n");
+ goto jump_to_error;
+ }
+
+ rc = pthread_mutex_init(&bp->ulp_ctx->cfg_data->flow_db_lock, NULL);
+ if (rc) {
+ BNXT_TF_DBG(ERR, "Unable to initialize flow db lock\n");
+ goto jump_to_error;
+ }
+
+ /* Initialize ulp dparms with values devargs passed */
+ rc = ulp_dparms_init(bp, bp->ulp_ctx);
+ if (rc) {
+ BNXT_TF_DBG(ERR, "Failed to initialize the dparms\n");
+ goto jump_to_error;
+ }
+
+ /* create the port database */
+ rc = ulp_port_db_init(bp->ulp_ctx, bp->port_cnt);
+ if (rc) {
+ BNXT_TF_DBG(ERR, "Failed to create the port database\n");
+ goto jump_to_error;
+ }
+
+ /* Create the Mark database. */
+ rc = ulp_mark_db_init(bp->ulp_ctx);
+ if (rc) {
+ BNXT_TF_DBG(ERR, "Failed to create the mark database\n");
+ goto jump_to_error;
+ }
+
+ /* Create the flow database. */
+ rc = ulp_flow_db_init(bp->ulp_ctx);
+ if (rc) {
+ BNXT_TF_DBG(ERR, "Failed to create the flow database\n");
+ goto jump_to_error;
+ }
+
+ /* Create the eem table scope. */
+ rc = ulp_eem_tbl_scope_init(bp);
+ if (rc) {
+ BNXT_TF_DBG(ERR, "Failed to create the eem scope table\n");
+ goto jump_to_error;
+ }
+
+ rc = ulp_mapper_init(bp->ulp_ctx);
+ if (rc) {
+ BNXT_TF_DBG(ERR, "Failed to initialize ulp mapper\n");
+ goto jump_to_error;
+ }
+
+ rc = ulp_fc_mgr_init(bp->ulp_ctx);
+ if (rc) {
+ BNXT_TF_DBG(ERR, "Failed to initialize ulp flow counter mgr\n");
+ goto jump_to_error;
+ }
+
+ /*
+ * Enable NAT feature. Set the global configuration register
+ * Tunnel encap to enable NAT with the reuse of existing inner
+ * L2 header smac and dmac
+ */
+ rc = bnxt_ulp_global_cfg_update(bp, TF_DIR_RX, TF_TUNNEL_ENCAP,
+ TF_TUNNEL_ENCAP_NAT,
+ BNXT_ULP_NAT_OUTER_MOST_FLAGS, 1);
+ if (rc) {
+ BNXT_TF_DBG(ERR, "Failed to set rx global configuration\n");
+ goto jump_to_error;
+ }
+
+ rc = bnxt_ulp_global_cfg_update(bp, TF_DIR_TX, TF_TUNNEL_ENCAP,
+ TF_TUNNEL_ENCAP_NAT,
+ BNXT_ULP_NAT_OUTER_MOST_FLAGS, 1);
+ if (rc) {
+ BNXT_TF_DBG(ERR, "Failed to set tx global configuration\n");
+ goto jump_to_error;
+ }
+
+ if (bnxt_ulp_cntxt_ha_enabled(bp->ulp_ctx)) {
+ rc = ulp_ha_mgr_init(bp->ulp_ctx);
+ if (rc) {
+ BNXT_TF_DBG(ERR, "Failed to initialize HA %d\n", rc);
+ goto jump_to_error;
+ }
+ rc = ulp_ha_mgr_open(bp->ulp_ctx);
+ if (rc) {
+ BNXT_TF_DBG(ERR, "Failed to Process HA Open %d\n", rc);
+ goto jump_to_error;
+ }
+ }
+ BNXT_TF_DBG(DEBUG, "ulp ctx has been initialized\n");
+ return rc;
+
+jump_to_error:
+ bnxt_ulp_deinit(bp, session);
+ return rc;
+}
+
+/*
+ * When a port is initialized by dpdk. This functions sets up
+ * the port specific details.
+ */
int32_t
-bnxt_ulp_init(struct bnxt *bp)
+bnxt_ulp_port_init(struct bnxt *bp)
{
struct bnxt_ulp_session_state *session;
- bool init;
- int rc;
+ bool initialized;
+ enum bnxt_ulp_device_id devid = BNXT_ULP_DEVICE_ID_LAST;
+ uint32_t ulp_flags;
+ int32_t rc = 0;
+
+ if (!BNXT_PF(bp) && !BNXT_VF_IS_TRUSTED(bp)) {
+ BNXT_TF_DBG(ERR,
+ "Skip ulp init for port: %d, not a TVF or PF\n",
+ bp->eth_dev->data->port_id);
+ return rc;
+ }
+
+ if (!BNXT_TRUFLOW_EN(bp)) {
+ BNXT_TF_DBG(DEBUG,
+ "Skip ulp init for port: %d, truflow is not enabled\n",
+ bp->eth_dev->data->port_id);
+ return rc;
+ }
+
+ if (bp->ulp_ctx) {
+ BNXT_TF_DBG(DEBUG, "ulp ctx already allocated\n");
+ return rc;
+ }
+
+ bp->ulp_ctx = rte_zmalloc("bnxt_ulp_ctx",
+ sizeof(struct bnxt_ulp_context), 0);
+ if (!bp->ulp_ctx) {
+ BNXT_TF_DBG(ERR, "Failed to allocate ulp ctx\n");
+ return -ENOMEM;
+ }
/*
* Multiple uplink ports can be associated with a single vswitch.
* Make sure only the port that is started first will initialize
* the TF session.
*/
- session = ulp_session_init(bp, &init);
+ session = ulp_session_init(bp, &initialized);
if (!session) {
BNXT_TF_DBG(ERR, "Failed to initialize the tf session\n");
- return -EINVAL;
+ rc = -EIO;
+ goto jump_to_error;
}
- /*
- * If ULP is already initialized for a specific domain then simply
- * assign the ulp context to this rte_eth_dev.
- */
- if (init) {
- rc = ulp_ctx_attach(&bp->ulp_ctx, session);
+ if (initialized) {
+ /*
+ * If ULP is already initialized for a specific domain then
+ * simply assign the ulp context to this rte_eth_dev.
+ */
+ rc = ulp_ctx_attach(bp, session);
+ if (rc) {
+ BNXT_TF_DBG(ERR, "Failed to attach the ulp context\n");
+ goto jump_to_error;
+ }
+
+ /*
+ * Attach to the shared session, must be called after the
+ * ulp_ctx_attach in order to ensure that ulp data is available
+ * for attaching.
+ */
+ rc = ulp_ctx_shared_session_attach(bp, session);
if (rc) {
BNXT_TF_DBG(ERR,
- "Failed to attach the ulp context\n");
+ "Failed attach to shared session (%d)", rc);
+ goto jump_to_error;
+ }
+ } else {
+ rc = bnxt_ulp_init(bp, session);
+ if (rc) {
+ BNXT_TF_DBG(ERR, "Failed to initialize the ulp init\n");
+ goto jump_to_error;
}
- return rc;
}
- /* Allocate and Initialize the ulp context. */
- rc = ulp_ctx_init(bp, session);
+ /* Update bnxt driver flags */
+ rc = ulp_dparms_dev_port_intf_update(bp, bp->ulp_ctx);
if (rc) {
- BNXT_TF_DBG(ERR, "Failed to create the ulp context\n");
+ BNXT_TF_DBG(ERR, "Failed to update driver flags\n");
goto jump_to_error;
}
- /* Create the Mark database. */
- rc = ulp_mark_db_init(&bp->ulp_ctx);
+ /* update the port database for the given interface */
+ rc = ulp_port_db_dev_port_intf_update(bp->ulp_ctx, bp->eth_dev);
if (rc) {
- BNXT_TF_DBG(ERR, "Failed to create the mark database\n");
+ BNXT_TF_DBG(ERR, "Failed to update port database\n");
goto jump_to_error;
}
-
- /* Create the flow database. */
- rc = ulp_flow_db_init(&bp->ulp_ctx);
+ /* create the default rules */
+ rc = bnxt_ulp_create_df_rules(bp);
if (rc) {
- BNXT_TF_DBG(ERR, "Failed to create the flow database\n");
+ BNXT_TF_DBG(ERR, "Failed to create default flow\n");
goto jump_to_error;
}
- /* Create the eem table scope. */
- rc = ulp_eem_tbl_scope_init(bp);
+ rc = bnxt_ulp_devid_get(bp, &devid);
if (rc) {
- BNXT_TF_DBG(ERR, "Failed to create the eem scope table\n");
+ BNXT_TF_DBG(ERR, "Unable to determine device for ULP port init.\n");
+ goto jump_to_error;
+ }
+
+ if (devid != BNXT_ULP_DEVICE_ID_THOR && BNXT_ACCUM_STATS_EN(bp))
+ bp->ulp_ctx->cfg_data->accum_stats = true;
+
+ BNXT_TF_DBG(DEBUG, "BNXT Port:%d ULP port init, accum_stats:%d\n",
+ bp->eth_dev->data->port_id,
+ bp->ulp_ctx->cfg_data->accum_stats);
+
+ /* set the unicast mode */
+ if (bnxt_ulp_cntxt_ptr2_ulp_flags_get(bp->ulp_ctx, &ulp_flags)) {
+ BNXT_TF_DBG(ERR, "Error in getting ULP context flags\n");
goto jump_to_error;
}
+ if (ulp_flags & BNXT_ULP_APP_UNICAST_ONLY) {
+ if (bnxt_pmd_set_unicast_rxmask(bp->eth_dev)) {
+ BNXT_TF_DBG(ERR, "Error in setting unicast rxmode\n");
+ goto jump_to_error;
+ }
+ }
return rc;
jump_to_error:
- bnxt_ulp_deinit(bp);
- return -ENOMEM;
+ bnxt_ulp_port_deinit(bp);
+ return rc;
}
-/* Below are the access functions to access internal data of ulp context. */
-
/*
- * When a port is deinit'ed by dpdk. This function is called
- * and this function clears the ULP context and rest of the
- * infrastructure associated with it.
+ * When a port is de-initialized by dpdk. This functions clears up
+ * the port specific details.
*/
void
-bnxt_ulp_deinit(struct bnxt *bp)
+bnxt_ulp_port_deinit(struct bnxt *bp)
{
- struct bnxt_ulp_session_state *session;
- struct rte_pci_device *pci_dev;
- struct rte_pci_addr *pci_addr;
+ struct bnxt_ulp_session_state *session;
+ struct rte_pci_device *pci_dev;
+ struct rte_pci_addr *pci_addr;
+
+ if (!BNXT_PF(bp) && !BNXT_VF_IS_TRUSTED(bp)) {
+ BNXT_TF_DBG(ERR,
+ "Skip ULP deinit port:%d, not a TVF or PF\n",
+ bp->eth_dev->data->port_id);
+ return;
+ }
+
+ if (!BNXT_TRUFLOW_EN(bp)) {
+ BNXT_TF_DBG(DEBUG,
+ "Skip ULP deinit for port:%d, truflow is not enabled\n",
+ bp->eth_dev->data->port_id);
+ return;
+ }
+
+ if (!bp->ulp_ctx) {
+ BNXT_TF_DBG(DEBUG, "ulp ctx already de-allocated\n");
+ return;
+ }
- /* Get the session first */
+ BNXT_TF_DBG(DEBUG, "BNXT Port:%d ULP port deinit\n",
+ bp->eth_dev->data->port_id);
+
+ /* Free the ulp context in the context entry list */
+ bnxt_ulp_cntxt_list_del(bp->ulp_ctx);
+
+ /* Get the session details */
pci_dev = RTE_DEV_TO_PCI(bp->eth_dev->device);
pci_addr = &pci_dev->addr;
pthread_mutex_lock(&bnxt_ulp_global_mutex);
pthread_mutex_unlock(&bnxt_ulp_global_mutex);
/* session not found then just exit */
- if (!session)
+ if (!session) {
+ /* Free the ulp context */
+ rte_free(bp->ulp_ctx);
+ bp->ulp_ctx = NULL;
return;
+ }
- /* clean up regular flows */
- ulp_flow_db_flush_flows(&bp->ulp_ctx, BNXT_ULP_REGULAR_FLOW_TABLE);
-
- /* cleanup the eem table scope */
- ulp_eem_tbl_scope_deinit(bp, &bp->ulp_ctx);
+ /* Check the reference count to deinit or deattach*/
+ if (bp->ulp_ctx->cfg_data && bp->ulp_ctx->cfg_data->ref_cnt) {
+ bp->ulp_ctx->cfg_data->ref_cnt--;
+ if (bp->ulp_ctx->cfg_data->ref_cnt) {
+ /* free the port details */
+ /* Free the default flow rule associated to this port */
+ bnxt_ulp_destroy_df_rules(bp, false);
+ bnxt_ulp_destroy_vfr_default_rules(bp, false);
- /* cleanup the flow database */
- ulp_flow_db_deinit(&bp->ulp_ctx);
+ /* free flows associated with this port */
+ bnxt_ulp_flush_port_flows(bp);
- /* Delete the Mark database */
- ulp_mark_db_deinit(&bp->ulp_ctx);
+ /* close the session associated with this port */
+ ulp_ctx_detach(bp);
- /* Delete the ulp context and tf session */
- ulp_ctx_detach(bp, session);
+ /* always detach/close shared after the session. */
+ ulp_ctx_shared_session_detach(bp);
+ } else {
+ /* Perform ulp ctx deinit */
+ bnxt_ulp_deinit(bp, session);
+ }
+ }
- /* Finally delete the bnxt session*/
+ /* clean up the session */
ulp_session_deinit(session);
+
+ /* Free the ulp context */
+ rte_free(bp->ulp_ctx);
+ bp->ulp_ctx = NULL;
}
+/* Below are the access functions to access internal data of ulp context. */
/* Function to set the Mark DB into the context */
int32_t
bnxt_ulp_cntxt_ptr2_mark_db_set(struct bnxt_ulp_context *ulp_ctx,
return ulp_ctx->cfg_data->mark_tbl;
}
+bool
+bnxt_ulp_cntxt_shared_session_enabled(struct bnxt_ulp_context *ulp_ctx)
+{
+ return ULP_SHARED_SESSION_IS_ENABLED(ulp_ctx->cfg_data->ulp_flags);
+}
+
+int32_t
+bnxt_ulp_cntxt_app_id_set(struct bnxt_ulp_context *ulp_ctx, uint8_t app_id)
+{
+ if (!ulp_ctx)
+ return -EINVAL;
+ ulp_ctx->cfg_data->app_id = app_id;
+ return 0;
+}
+
+int32_t
+bnxt_ulp_cntxt_app_id_get(struct bnxt_ulp_context *ulp_ctx, uint8_t *app_id)
+{
+ /* Default APP id is zero */
+ if (!ulp_ctx || !app_id)
+ return -EINVAL;
+ *app_id = ulp_ctx->cfg_data->app_id;
+ return 0;
+}
+
/* Function to set the device id of the hardware. */
int32_t
bnxt_ulp_cntxt_dev_id_set(struct bnxt_ulp_context *ulp_ctx,
*dev_id = ulp_ctx->cfg_data->dev_id;
return 0;
}
+ *dev_id = BNXT_ULP_DEVICE_ID_LAST;
+ BNXT_TF_DBG(ERR, "Failed to read dev_id from ulp ctxt\n");
+ return -EINVAL;
+}
+
+int32_t
+bnxt_ulp_cntxt_mem_type_set(struct bnxt_ulp_context *ulp_ctx,
+ enum bnxt_ulp_flow_mem_type mem_type)
+{
+ if (ulp_ctx && ulp_ctx->cfg_data) {
+ ulp_ctx->cfg_data->mem_type = mem_type;
+ return 0;
+ }
+ BNXT_TF_DBG(ERR, "Failed to write mem_type in ulp ctxt\n");
+ return -EINVAL;
+}
+int32_t
+bnxt_ulp_cntxt_mem_type_get(struct bnxt_ulp_context *ulp_ctx,
+ enum bnxt_ulp_flow_mem_type *mem_type)
+{
+ if (ulp_ctx && ulp_ctx->cfg_data) {
+ *mem_type = ulp_ctx->cfg_data->mem_type;
+ return 0;
+ }
+ *mem_type = BNXT_ULP_FLOW_MEM_TYPE_LAST;
+ BNXT_TF_DBG(ERR, "Failed to read mem_type in ulp ctxt\n");
return -EINVAL;
}
return -EINVAL;
}
+/* Function to set the shared tfp session details from the ulp context. */
+int32_t
+bnxt_ulp_cntxt_shared_tfp_set(struct bnxt_ulp_context *ulp, struct tf *tfp)
+{
+ if (!ulp) {
+ BNXT_TF_DBG(ERR, "Invalid arguments\n");
+ return -EINVAL;
+ }
+
+ if (tfp == NULL) {
+ if (ulp->cfg_data->num_shared_clients > 0)
+ ulp->cfg_data->num_shared_clients--;
+ } else {
+ ulp->cfg_data->num_shared_clients++;
+ }
+
+ ulp->g_shared_tfp = tfp;
+ return 0;
+}
+
+/* Function to get the shared tfp session details from the ulp context. */
+struct tf *
+bnxt_ulp_cntxt_shared_tfp_get(struct bnxt_ulp_context *ulp)
+{
+ if (!ulp) {
+ BNXT_TF_DBG(ERR, "Invalid arguments\n");
+ return NULL;
+ }
+ return ulp->g_shared_tfp;
+}
+
+/* Function to get the number of shared clients attached */
+uint8_t
+bnxt_ulp_cntxt_num_shared_clients_get(struct bnxt_ulp_context *ulp)
+{
+ if (ulp == NULL || ulp->cfg_data == NULL) {
+ BNXT_TF_DBG(ERR, "Invalid arguments\n");
+ return 0;
+ }
+ return ulp->cfg_data->num_shared_clients;
+}
+
/* Function to set the tfp session details from the ulp context. */
int32_t
bnxt_ulp_cntxt_tfp_set(struct bnxt_ulp_context *ulp, struct tf *tfp)
return -EINVAL;
}
- /* TBD The tfp should be removed once tf_attach is implemented. */
ulp->g_tfp = tfp;
return 0;
}
/* Function to get the tfp session details from the ulp context. */
struct tf *
-bnxt_ulp_cntxt_tfp_get(struct bnxt_ulp_context *ulp)
+bnxt_ulp_cntxt_tfp_get(struct bnxt_ulp_context *ulp,
+ enum bnxt_ulp_shared_session shared)
{
if (!ulp) {
BNXT_TF_DBG(ERR, "Invalid arguments\n");
return NULL;
}
- /* TBD The tfp should be removed once tf_attach is implemented. */
- return ulp->g_tfp;
+ if (shared)
+ return ulp->g_shared_tfp;
+ else
+ return ulp->g_tfp;
}
/*
bnxt_ulp_cntxt_ptr2_flow_db_set(struct bnxt_ulp_context *ulp_ctx,
struct bnxt_ulp_flow_db *flow_db)
{
- if (!ulp_ctx || !ulp_ctx->cfg_data) {
- BNXT_TF_DBG(ERR, "Invalid ulp context data\n");
+ if (!ulp_ctx || !ulp_ctx->cfg_data)
return -EINVAL;
- }
ulp_ctx->cfg_data->flow_db = flow_db;
return 0;
struct bnxt_ulp_flow_db *
bnxt_ulp_cntxt_ptr2_flow_db_get(struct bnxt_ulp_context *ulp_ctx)
{
- if (!ulp_ctx || !ulp_ctx->cfg_data) {
- BNXT_TF_DBG(ERR, "Invalid ulp context data\n");
+ if (!ulp_ctx || !ulp_ctx->cfg_data)
return NULL;
- }
return ulp_ctx->cfg_data->flow_db;
}
+/* Function to get the tunnel cache table info from the ulp context. */
+struct bnxt_tun_cache_entry *
+bnxt_ulp_cntxt_ptr2_tun_tbl_get(struct bnxt_ulp_context *ulp_ctx)
+{
+ if (!ulp_ctx || !ulp_ctx->cfg_data)
+ return NULL;
+
+ return ulp_ctx->cfg_data->tun_tbl;
+}
+
/* Function to get the ulp context from eth device. */
struct bnxt_ulp_context *
bnxt_ulp_eth_dev_ptr2_cntxt_get(struct rte_eth_dev *dev)
{
- struct bnxt *bp;
+ struct bnxt *bp = (struct bnxt *)dev->data->dev_private;
+
+ if (BNXT_ETH_DEV_IS_REPRESENTOR(dev)) {
+ struct bnxt_representor *vfr = dev->data->dev_private;
+
+ bp = vfr->parent_dev->data->dev_private;
+ }
- bp = (struct bnxt *)dev->data->dev_private;
if (!bp) {
BNXT_TF_DBG(ERR, "Bnxt private data is not initialized\n");
return NULL;
}
- return &bp->ulp_ctx;
+ return bp->ulp_ctx;
+}
+
+int32_t
+bnxt_ulp_cntxt_ptr2_mapper_data_set(struct bnxt_ulp_context *ulp_ctx,
+ void *mapper_data)
+{
+ if (!ulp_ctx || !ulp_ctx->cfg_data) {
+ BNXT_TF_DBG(ERR, "Invalid ulp context data\n");
+ return -EINVAL;
+ }
+
+ ulp_ctx->cfg_data->mapper_data = mapper_data;
+ return 0;
+}
+
+void *
+bnxt_ulp_cntxt_ptr2_mapper_data_get(struct bnxt_ulp_context *ulp_ctx)
+{
+ if (!ulp_ctx || !ulp_ctx->cfg_data) {
+ BNXT_TF_DBG(ERR, "Invalid ulp context data\n");
+ return NULL;
+ }
+
+ return ulp_ctx->cfg_data->mapper_data;
+}
+
+/* Function to set the port database to the ulp context. */
+int32_t
+bnxt_ulp_cntxt_ptr2_port_db_set(struct bnxt_ulp_context *ulp_ctx,
+ struct bnxt_ulp_port_db *port_db)
+{
+ if (!ulp_ctx || !ulp_ctx->cfg_data)
+ return -EINVAL;
+
+ ulp_ctx->cfg_data->port_db = port_db;
+ return 0;
+}
+
+/* Function to get the port database from the ulp context. */
+struct bnxt_ulp_port_db *
+bnxt_ulp_cntxt_ptr2_port_db_get(struct bnxt_ulp_context *ulp_ctx)
+{
+ if (!ulp_ctx || !ulp_ctx->cfg_data)
+ return NULL;
+
+ return ulp_ctx->cfg_data->port_db;
+}
+
+/* Function to set the flow counter info into the context */
+int32_t
+bnxt_ulp_cntxt_ptr2_fc_info_set(struct bnxt_ulp_context *ulp_ctx,
+ struct bnxt_ulp_fc_info *ulp_fc_info)
+{
+ if (!ulp_ctx || !ulp_ctx->cfg_data) {
+ BNXT_TF_DBG(ERR, "Invalid ulp context data\n");
+ return -EINVAL;
+ }
+
+ ulp_ctx->cfg_data->fc_info = ulp_fc_info;
+
+ return 0;
+}
+
+/* Function to retrieve the flow counter info from the context. */
+struct bnxt_ulp_fc_info *
+bnxt_ulp_cntxt_ptr2_fc_info_get(struct bnxt_ulp_context *ulp_ctx)
+{
+ if (!ulp_ctx || !ulp_ctx->cfg_data)
+ return NULL;
+
+ return ulp_ctx->cfg_data->fc_info;
+}
+
+/* Function to get the ulp flags from the ulp context. */
+int32_t
+bnxt_ulp_cntxt_ptr2_ulp_flags_get(struct bnxt_ulp_context *ulp_ctx,
+ uint32_t *flags)
+{
+ if (!ulp_ctx || !ulp_ctx->cfg_data)
+ return -1;
+
+ *flags = ulp_ctx->cfg_data->ulp_flags;
+ return 0;
+}
+
+/* Function to get the ulp vfr info from the ulp context. */
+struct bnxt_ulp_vfr_rule_info*
+bnxt_ulp_cntxt_ptr2_ulp_vfr_info_get(struct bnxt_ulp_context *ulp_ctx,
+ uint32_t port_id)
+{
+ if (!ulp_ctx || !ulp_ctx->cfg_data || port_id >= RTE_MAX_ETHPORTS)
+ return NULL;
+
+ return &ulp_ctx->cfg_data->vfr_rule_info[port_id];
+}
+
+/* Function to acquire the flow database lock from the ulp context. */
+int32_t
+bnxt_ulp_cntxt_acquire_fdb_lock(struct bnxt_ulp_context *ulp_ctx)
+{
+ if (!ulp_ctx || !ulp_ctx->cfg_data)
+ return -1;
+
+ if (pthread_mutex_lock(&ulp_ctx->cfg_data->flow_db_lock)) {
+ BNXT_TF_DBG(ERR, "unable to acquire fdb lock\n");
+ return -1;
+ }
+ return 0;
+}
+
+/* Function to release the flow database lock from the ulp context. */
+void
+bnxt_ulp_cntxt_release_fdb_lock(struct bnxt_ulp_context *ulp_ctx)
+{
+ if (!ulp_ctx || !ulp_ctx->cfg_data)
+ return;
+
+ pthread_mutex_unlock(&ulp_ctx->cfg_data->flow_db_lock);
+}
+
+/* Function to set the ha info into the context */
+int32_t
+bnxt_ulp_cntxt_ptr2_ha_info_set(struct bnxt_ulp_context *ulp_ctx,
+ struct bnxt_ulp_ha_mgr_info *ulp_ha_info)
+{
+ if (ulp_ctx == NULL || ulp_ctx->cfg_data == NULL) {
+ BNXT_TF_DBG(ERR, "Invalid ulp context data\n");
+ return -EINVAL;
+ }
+ ulp_ctx->cfg_data->ha_info = ulp_ha_info;
+ return 0;
+}
+
+/* Function to retrieve the ha info from the context. */
+struct bnxt_ulp_ha_mgr_info *
+bnxt_ulp_cntxt_ptr2_ha_info_get(struct bnxt_ulp_context *ulp_ctx)
+{
+ if (ulp_ctx == NULL || ulp_ctx->cfg_data == NULL)
+ return NULL;
+ return ulp_ctx->cfg_data->ha_info;
+}
+
+bool
+bnxt_ulp_cntxt_ha_enabled(struct bnxt_ulp_context *ulp_ctx)
+{
+ if (ulp_ctx == NULL || ulp_ctx->cfg_data == NULL)
+ return false;
+ return !!ULP_HIGH_AVAIL_IS_ENABLED(ulp_ctx->cfg_data->ulp_flags);
+}
+
+static int32_t
+bnxt_ulp_cntxt_list_init(void)
+{
+ /* Create the cntxt spin lock */
+ rte_spinlock_init(&bnxt_ulp_ctxt_lock);
+
+ return 0;
+}
+
+static int32_t
+bnxt_ulp_cntxt_list_add(struct bnxt_ulp_context *ulp_ctx)
+{
+ struct ulp_context_list_entry *entry;
+
+ entry = rte_zmalloc(NULL, sizeof(struct ulp_context_list_entry), 0);
+ if (entry == NULL) {
+ BNXT_TF_DBG(ERR, "unable to allocate memory\n");
+ return -ENOMEM;
+ }
+
+ rte_spinlock_lock(&bnxt_ulp_ctxt_lock);
+ entry->ulp_ctx = ulp_ctx;
+ TAILQ_INSERT_TAIL(&ulp_cntx_list, entry, next);
+ rte_spinlock_unlock(&bnxt_ulp_ctxt_lock);
+ return 0;
+}
+
+static void
+bnxt_ulp_cntxt_list_del(struct bnxt_ulp_context *ulp_ctx)
+{
+ struct ulp_context_list_entry *entry, *temp;
+
+ rte_spinlock_lock(&bnxt_ulp_ctxt_lock);
+ TAILQ_FOREACH_SAFE(entry, &ulp_cntx_list, next, temp) {
+ if (entry->ulp_ctx == ulp_ctx) {
+ TAILQ_REMOVE(&ulp_cntx_list, entry, next);
+ rte_free(entry);
+ break;
+ }
+ }
+ rte_spinlock_unlock(&bnxt_ulp_ctxt_lock);
+}
+
+struct bnxt_ulp_context *
+bnxt_ulp_cntxt_entry_acquire(void)
+{
+ struct ulp_context_list_entry *entry;
+
+ /* take a lock and get the first ulp context available */
+ if (rte_spinlock_trylock(&bnxt_ulp_ctxt_lock)) {
+ TAILQ_FOREACH(entry, &ulp_cntx_list, next)
+ if (entry->ulp_ctx)
+ return entry->ulp_ctx;
+ }
+ return NULL;
+}
+
+void
+bnxt_ulp_cntxt_entry_release(void)
+{
+ rte_spinlock_unlock(&bnxt_ulp_ctxt_lock);
}