[dpdk-dev] [PATCH 01/50] net/bnxt: Basic infrastructure support for VF representors

[Somnath Kotur <somnath.kotur@broadcom.com>]

Defines data structures and code to init/uninit
VF representors during pci_probe and pci_remove
respectively.
Most of the dev_ops for the VF representor are just
stubs for now and will be will be filled out in next patch.

To create a representor using testpmd:
testpmd -c 0xff -wB:D.F,representor=1 -- -i
testpmd -c 0xff -w05:02.0,representor=[1] -- -i

To create a representor using ovs-dpdk:
1. Firt add the trusted VF port to a bridge
ovs-vsctl add-port ovsbr0 vf_rep1 -- set Interface vf_rep1 type=dpdk
options:dpdk-devargs=0000:06:02.0
2. Add the representor port to the bridge
ovs-vsctl add-port ovsbr0 vf_rep1 -- set Interface vf_rep1 type=dpdk
options:dpdk-devargs=0000:06:02.0,representor=1

Reviewed-by: Kalesh Anakkur Purayil <kalesh-anakkur.purayil@broadcom.com>
Signed-off-by: Somnath Kotur <somnath.kotur@broadcom.com>
Signed-off-by: Venkat Duvvuru <venkatkumar.duvvuru@broadcom.com>
---
 drivers/net/bnxt/Makefile      |   2 +
 drivers/net/bnxt/bnxt.h        |  64 ++++++++-
 drivers/net/bnxt/bnxt_ethdev.c | 225 +++++++++++++++++++++++++-------
 drivers/net/bnxt/bnxt_reps.c   | 287 +++++++++++++++++++++++++++++++++++++++++
 drivers/net/bnxt/bnxt_reps.h   |  35 +++++
 drivers/net/bnxt/meson.build   |   1 +
 6 files changed, 566 insertions(+), 48 deletions(-)
 create mode 100644 drivers/net/bnxt/bnxt_reps.c
 create mode 100644 drivers/net/bnxt/bnxt_reps.h

diff --git a/drivers/net/bnxt/Makefile b/drivers/net/bnxt/Makefile
index a375299..3656274 100644
--- a/drivers/net/bnxt/Makefile
+++ b/drivers/net/bnxt/Makefile
@@ -14,6 +14,7 @@ LIB = librte_pmd_bnxt.a
 EXPORT_MAP := rte_pmd_bnxt_version.map
 
 CFLAGS += -O3
+CFLAGS += -DALLOW_EXPERIMENTAL_API
 CFLAGS += $(WERROR_FLAGS)
 LDLIBS += -lrte_eal -lrte_mbuf -lrte_mempool -lrte_ring
 LDLIBS += -lrte_ethdev -lrte_net -lrte_kvargs
@@ -38,6 +39,7 @@ SRCS-$(CONFIG_RTE_LIBRTE_BNXT_PMD) += bnxt_txr.c
 SRCS-$(CONFIG_RTE_LIBRTE_BNXT_PMD) += bnxt_vnic.c
 SRCS-$(CONFIG_RTE_LIBRTE_BNXT_PMD) += bnxt_irq.c
 SRCS-$(CONFIG_RTE_LIBRTE_BNXT_PMD) += bnxt_util.c
+SRCS-$(CONFIG_RTE_LIBRTE_BNXT_PMD) += bnxt_reps.c
 SRCS-$(CONFIG_RTE_LIBRTE_BNXT_PMD) += rte_pmd_bnxt.c
 ifeq ($(CONFIG_RTE_ARCH_X86), y)
 SRCS-$(CONFIG_RTE_LIBRTE_BNXT_PMD) += bnxt_rxtx_vec_sse.c
diff --git a/drivers/net/bnxt/bnxt.h b/drivers/net/bnxt/bnxt.h
index d455f8d..9b7b87c 100644
--- a/drivers/net/bnxt/bnxt.h
+++ b/drivers/net/bnxt/bnxt.h
@@ -220,6 +220,7 @@ struct bnxt_child_vf_info {
 struct bnxt_pf_info {
 #define BNXT_FIRST_PF_FID	1
 #define BNXT_MAX_VFS(bp)	((bp)->pf->max_vfs)
+#define BNXT_MAX_VF_REPS	64
 #define BNXT_TOTAL_VFS(bp)	((bp)->pf->total_vfs)
 #define BNXT_FIRST_VF_FID	128
 #define BNXT_PF_RINGS_USED(bp)	bnxt_get_num_queues(bp)
@@ -492,6 +493,10 @@ struct bnxt_mark_info {
 	bool		valid;
 };
 
+struct bnxt_rep_info {
+	struct rte_eth_dev	*vfr_eth_dev;
+};
+
 /* address space location of register */
 #define BNXT_FW_STATUS_REG_TYPE_MASK	3
 /* register is located in PCIe config space */
@@ -515,6 +520,40 @@ struct bnxt_mark_info {
 #define BNXT_FW_STATUS_HEALTHY		0x8000
 #define BNXT_FW_STATUS_SHUTDOWN		0x100000
 
+#define BNXT_ETH_RSS_SUPPORT (	\
+	ETH_RSS_IPV4 |		\
+	ETH_RSS_NONFRAG_IPV4_TCP |	\
+	ETH_RSS_NONFRAG_IPV4_UDP |	\
+	ETH_RSS_IPV6 |		\
+	ETH_RSS_NONFRAG_IPV6_TCP |	\
+	ETH_RSS_NONFRAG_IPV6_UDP)
+
+#define BNXT_DEV_TX_OFFLOAD_SUPPORT (DEV_TX_OFFLOAD_VLAN_INSERT | \
+				     DEV_TX_OFFLOAD_IPV4_CKSUM | \
+				     DEV_TX_OFFLOAD_TCP_CKSUM | \
+				     DEV_TX_OFFLOAD_UDP_CKSUM | \
+				     DEV_TX_OFFLOAD_TCP_TSO | \
+				     DEV_TX_OFFLOAD_OUTER_IPV4_CKSUM | \
+				     DEV_TX_OFFLOAD_VXLAN_TNL_TSO | \
+				     DEV_TX_OFFLOAD_GRE_TNL_TSO | \
+				     DEV_TX_OFFLOAD_IPIP_TNL_TSO | \
+				     DEV_TX_OFFLOAD_GENEVE_TNL_TSO | \
+				     DEV_TX_OFFLOAD_QINQ_INSERT | \
+				     DEV_TX_OFFLOAD_MULTI_SEGS)
+
+#define BNXT_DEV_RX_OFFLOAD_SUPPORT (DEV_RX_OFFLOAD_VLAN_FILTER | \
+				     DEV_RX_OFFLOAD_VLAN_STRIP | \
+				     DEV_RX_OFFLOAD_IPV4_CKSUM | \
+				     DEV_RX_OFFLOAD_UDP_CKSUM | \
+				     DEV_RX_OFFLOAD_TCP_CKSUM | \
+				     DEV_RX_OFFLOAD_OUTER_IPV4_CKSUM | \
+				     DEV_RX_OFFLOAD_JUMBO_FRAME | \
+				     DEV_RX_OFFLOAD_KEEP_CRC | \
+				     DEV_RX_OFFLOAD_VLAN_EXTEND | \
+				     DEV_RX_OFFLOAD_TCP_LRO | \
+				     DEV_RX_OFFLOAD_SCATTER | \
+				     DEV_RX_OFFLOAD_RSS_HASH)
+
 #define BNXT_HWRM_SHORT_REQ_LEN		sizeof(struct hwrm_short_input)
 
 struct bnxt_flow_stat_info {
@@ -682,6 +721,9 @@ struct bnxt {
 #define BNXT_MAX_RINGS(bp) \
 	(RTE_MIN((((bp)->max_cp_rings - BNXT_NUM_ASYNC_CPR(bp)) / 2U), \
 		 BNXT_MAX_TX_RINGS(bp)))
+
+#define BNXT_MAX_VF_REP_RINGS	8
+
 	uint16_t		max_nq_rings;
 	uint16_t		max_l2_ctx;
 	uint16_t		max_rx_em_flows;
@@ -711,7 +753,9 @@ struct bnxt {
 
 	uint16_t		fw_reset_min_msecs;
 	uint16_t		fw_reset_max_msecs;
-
+	uint16_t		switch_domain_id;
+	uint16_t		num_reps;
+	struct bnxt_rep_info	rep_info[BNXT_MAX_VF_REPS];
 	/* Struct to hold adapter error recovery related info */
 	struct bnxt_error_recovery_info *recovery_info;
 #define BNXT_MARK_TABLE_SZ	(sizeof(struct bnxt_mark_info)  * 64 * 1024)
@@ -732,6 +776,18 @@ struct bnxt {
 
 #define BNXT_FC_TIMER	1 /* Timer freq in Sec Flow Counters */
 
+/**
+ * Structure to store private data for each VF representor instance
+ */
+struct bnxt_vf_representor {
+	uint16_t switch_domain_id;
+	uint16_t vf_id;
+	/* Private data store of associated PF/Trusted VF */
+	struct bnxt	*parent_priv;
+	uint8_t		mac_addr[RTE_ETHER_ADDR_LEN];
+	uint8_t		dflt_mac_addr[RTE_ETHER_ADDR_LEN];
+};
+
 int bnxt_mtu_set_op(struct rte_eth_dev *eth_dev, uint16_t new_mtu);
 int bnxt_link_update(struct rte_eth_dev *eth_dev, int wait_to_complete,
 		     bool exp_link_status);
@@ -744,7 +800,13 @@ void bnxt_schedule_fw_health_check(struct bnxt *bp);
 
 bool is_bnxt_supported(struct rte_eth_dev *dev);
 bool bnxt_stratus_device(struct bnxt *bp);
+void bnxt_print_link_info(struct rte_eth_dev *eth_dev);
+uint16_t bnxt_rss_hash_tbl_size(const struct bnxt *bp);
+int bnxt_link_update_op(struct rte_eth_dev *eth_dev,
+			int wait_to_complete);
+
 extern const struct rte_flow_ops bnxt_flow_ops;
+
 #define bnxt_acquire_flow_lock(bp) \
 	pthread_mutex_lock(&(bp)->flow_lock)
 
diff --git a/drivers/net/bnxt/bnxt_ethdev.c b/drivers/net/bnxt/bnxt_ethdev.c
index 7022f6d..4911745 100644
--- a/drivers/net/bnxt/bnxt_ethdev.c
+++ b/drivers/net/bnxt/bnxt_ethdev.c
@@ -18,6 +18,7 @@
 #include "bnxt_filter.h"
 #include "bnxt_hwrm.h"
 #include "bnxt_irq.h"
+#include "bnxt_reps.h"
 #include "bnxt_ring.h"
 #include "bnxt_rxq.h"
 #include "bnxt_rxr.h"
@@ -93,40 +94,6 @@ static const struct rte_pci_id bnxt_pci_id_map[] = {
 	{ .vendor_id = 0, /* sentinel */ },
 };
 
-#define BNXT_ETH_RSS_SUPPORT (	\
-	ETH_RSS_IPV4 |		\
-	ETH_RSS_NONFRAG_IPV4_TCP |	\
-	ETH_RSS_NONFRAG_IPV4_UDP |	\
-	ETH_RSS_IPV6 |		\
-	ETH_RSS_NONFRAG_IPV6_TCP |	\
-	ETH_RSS_NONFRAG_IPV6_UDP)
-
-#define BNXT_DEV_TX_OFFLOAD_SUPPORT (DEV_TX_OFFLOAD_VLAN_INSERT | \
-				     DEV_TX_OFFLOAD_IPV4_CKSUM | \
-				     DEV_TX_OFFLOAD_TCP_CKSUM | \
-				     DEV_TX_OFFLOAD_UDP_CKSUM | \
-				     DEV_TX_OFFLOAD_TCP_TSO | \
-				     DEV_TX_OFFLOAD_OUTER_IPV4_CKSUM | \
-				     DEV_TX_OFFLOAD_VXLAN_TNL_TSO | \
-				     DEV_TX_OFFLOAD_GRE_TNL_TSO | \
-				     DEV_TX_OFFLOAD_IPIP_TNL_TSO | \
-				     DEV_TX_OFFLOAD_GENEVE_TNL_TSO | \
-				     DEV_TX_OFFLOAD_QINQ_INSERT | \
-				     DEV_TX_OFFLOAD_MULTI_SEGS)
-
-#define BNXT_DEV_RX_OFFLOAD_SUPPORT (DEV_RX_OFFLOAD_VLAN_FILTER | \
-				     DEV_RX_OFFLOAD_VLAN_STRIP | \
-				     DEV_RX_OFFLOAD_IPV4_CKSUM | \
-				     DEV_RX_OFFLOAD_UDP_CKSUM | \
-				     DEV_RX_OFFLOAD_TCP_CKSUM | \
-				     DEV_RX_OFFLOAD_OUTER_IPV4_CKSUM | \
-				     DEV_RX_OFFLOAD_JUMBO_FRAME | \
-				     DEV_RX_OFFLOAD_KEEP_CRC | \
-				     DEV_RX_OFFLOAD_VLAN_EXTEND | \
-				     DEV_RX_OFFLOAD_TCP_LRO | \
-				     DEV_RX_OFFLOAD_SCATTER | \
-				     DEV_RX_OFFLOAD_RSS_HASH)
-
 #define BNXT_DEVARG_TRUFLOW	"host-based-truflow"
 #define BNXT_DEVARG_FLOW_XSTAT	"flow-xstat"
 #define BNXT_DEVARG_MAX_NUM_KFLOWS  "max-num-kflows"
@@ -163,7 +130,6 @@ static int bnxt_devarg_max_num_kflow_invalid(uint16_t max_num_kflows)
 }
 
 static int bnxt_vlan_offload_set_op(struct rte_eth_dev *dev, int mask);
-static void bnxt_print_link_info(struct rte_eth_dev *eth_dev);
 static int bnxt_dev_uninit(struct rte_eth_dev *eth_dev);
 static int bnxt_init_resources(struct bnxt *bp, bool reconfig_dev);
 static int bnxt_uninit_resources(struct bnxt *bp, bool reconfig_dev);
@@ -198,7 +164,7 @@ static uint16_t bnxt_rss_ctxts(const struct bnxt *bp)
 				    BNXT_RSS_ENTRIES_PER_CTX_THOR;
 }
 
-static uint16_t  bnxt_rss_hash_tbl_size(const struct bnxt *bp)
+uint16_t bnxt_rss_hash_tbl_size(const struct bnxt *bp)
 {
 	if (!BNXT_CHIP_THOR(bp))
 		return HW_HASH_INDEX_SIZE;
@@ -1047,7 +1013,7 @@ static int bnxt_dev_configure_op(struct rte_eth_dev *eth_dev)
 	return -ENOSPC;
 }
 
-static void bnxt_print_link_info(struct rte_eth_dev *eth_dev)
+void bnxt_print_link_info(struct rte_eth_dev *eth_dev)
 {
 	struct rte_eth_link *link = &eth_dev->data->dev_link;
 
@@ -1273,6 +1239,12 @@ static int bnxt_dev_set_link_down_op(struct rte_eth_dev *eth_dev)
 	return 0;
 }
 
+static void bnxt_free_switch_domain(struct bnxt *bp)
+{
+	if (bp->switch_domain_id)
+		rte_eth_switch_domain_free(bp->switch_domain_id);
+}
+
 /* Unload the driver, release resources */
 static void bnxt_dev_stop_op(struct rte_eth_dev *eth_dev)
 {
@@ -1341,6 +1313,8 @@ static void bnxt_dev_close_op(struct rte_eth_dev *eth_dev)
 	if (eth_dev->data->dev_started)
 		bnxt_dev_stop_op(eth_dev);
 
+	bnxt_free_switch_domain(bp);
+
 	bnxt_uninit_resources(bp, false);
 
 	bnxt_free_leds_info(bp);
@@ -1522,8 +1496,8 @@ int bnxt_link_update(struct rte_eth_dev *eth_dev, int wait_to_complete,
 	return rc;
 }
 
-static int bnxt_link_update_op(struct rte_eth_dev *eth_dev,
-			       int wait_to_complete)
+int bnxt_link_update_op(struct rte_eth_dev *eth_dev,
+			int wait_to_complete)
 {
 	return bnxt_link_update(eth_dev, wait_to_complete, ETH_LINK_UP);
 }
@@ -5477,8 +5451,26 @@ bnxt_parse_dev_args(struct bnxt *bp, struct rte_devargs *devargs)
 	rte_kvargs_free(kvlist);
 }
 
+static int bnxt_alloc_switch_domain(struct bnxt *bp)
+{
+	int rc = 0;
+
+	if (BNXT_PF(bp) || BNXT_VF_IS_TRUSTED(bp)) {
+		rc = rte_eth_switch_domain_alloc(&bp->switch_domain_id);
+		if (rc)
+			PMD_DRV_LOG(ERR,
+				    "Failed to alloc switch domain: %d\n", rc);
+		else
+			PMD_DRV_LOG(INFO,
+				    "Switch domain allocated %d\n",
+				    bp->switch_domain_id);
+	}
+
+	return rc;
+}
+
 static int
-bnxt_dev_init(struct rte_eth_dev *eth_dev)
+bnxt_dev_init(struct rte_eth_dev *eth_dev, void *params __rte_unused)
 {
 	struct rte_pci_device *pci_dev = RTE_ETH_DEV_TO_PCI(eth_dev);
 	static int version_printed;
@@ -5557,6 +5549,8 @@ bnxt_dev_init(struct rte_eth_dev *eth_dev)
 	if (rc)
 		goto error_free;
 
+	bnxt_alloc_switch_domain(bp);
+
 	/* Pass the information to the rte_eth_dev_close() that it should also
 	 * release the private port resources.
 	 */
@@ -5689,25 +5683,162 @@ bnxt_dev_uninit(struct rte_eth_dev *eth_dev)
 	return 0;
 }
 
+static int bnxt_pci_remove_dev_with_reps(struct rte_eth_dev *eth_dev)
+{
+	struct bnxt *bp = eth_dev->data->dev_private;
+	struct rte_eth_dev *vf_rep_eth_dev;
+	int ret = 0, i;
+
+	if (!bp)
+		return -EINVAL;
+
+	for (i = 0; i < bp->num_reps; i++) {
+		vf_rep_eth_dev = bp->rep_info[i].vfr_eth_dev;
+		if (!vf_rep_eth_dev)
+			continue;
+		rte_eth_dev_destroy(vf_rep_eth_dev, bnxt_vf_representor_uninit);
+	}
+	ret = rte_eth_dev_destroy(eth_dev, bnxt_dev_uninit);
+
+	return ret;
+}
+
 static int bnxt_pci_probe(struct rte_pci_driver *pci_drv __rte_unused,
 	struct rte_pci_device *pci_dev)
 {
-	return rte_eth_dev_pci_generic_probe(pci_dev, sizeof(struct bnxt),
-		bnxt_dev_init);
+	char name[RTE_ETH_NAME_MAX_LEN];
+	struct rte_eth_devargs eth_da = { .nb_representor_ports = 0 };
+	struct rte_eth_dev *backing_eth_dev, *vf_rep_eth_dev;
+	uint16_t num_rep;
+	int i, ret = 0;
+	struct bnxt *backing_bp;
+
+	if (pci_dev->device.devargs) {
+		ret = rte_eth_devargs_parse(pci_dev->device.devargs->args,
+					    &eth_da);
+		if (ret)
+			return ret;
+	}
+
+	num_rep = eth_da.nb_representor_ports;
+	PMD_DRV_LOG(DEBUG, "nb_representor_ports = %d\n",
+		    num_rep);
+
+	/* We could come here after first level of probe is already invoked
+	 * as part of an application bringup(OVS-DPDK vswitchd), so first check
+	 * for already allocated eth_dev for the backing device (PF/Trusted VF)
+	 */
+	backing_eth_dev = rte_eth_dev_allocated(pci_dev->device.name);
+	if (backing_eth_dev == NULL) {
+		ret = rte_eth_dev_create(&pci_dev->device, pci_dev->device.name,
+					 sizeof(struct bnxt),
+					 eth_dev_pci_specific_init, pci_dev,
+					 bnxt_dev_init, NULL);
+
+		if (ret || !num_rep)
+			return ret;
+	}
+
+	if (num_rep > BNXT_MAX_VF_REPS) {
+		PMD_DRV_LOG(ERR, "nb_representor_ports = %d > %d MAX VF REPS\n",
+			    eth_da.nb_representor_ports, BNXT_MAX_VF_REPS);
+		ret = -EINVAL;
+		return ret;
+	}
+
+	/* probe representor ports now */
+	if (!backing_eth_dev)
+		backing_eth_dev = rte_eth_dev_allocated(pci_dev->device.name);
+	if (backing_eth_dev == NULL) {
+		ret = -ENODEV;
+		return ret;
+	}
+	backing_bp = backing_eth_dev->data->dev_private;
+
+	if (!(BNXT_PF(backing_bp) || BNXT_VF_IS_TRUSTED(backing_bp))) {
+		PMD_DRV_LOG(ERR,
+			    "Not a PF or trusted VF. No Representor support\n");
+		/* Returning an error is not an option.
+		 * Applications are not handling this correctly
+		 */
+		return ret;
+	}
+
+	for (i = 0; i < eth_da.nb_representor_ports; i++) {
+		struct bnxt_vf_representor representor = {
+			.vf_id = eth_da.representor_ports[i],
+			.switch_domain_id = backing_bp->switch_domain_id,
+			.parent_priv = backing_bp
+		};
+
+		if (representor.vf_id >= BNXT_MAX_VF_REPS) {
+			PMD_DRV_LOG(ERR, "VF-Rep id %d >= %d MAX VF ID\n",
+				    representor.vf_id, BNXT_MAX_VF_REPS);
+			continue;
+		}
+
+		/* representor port net_bdf_port */
+		snprintf(name, sizeof(name), "net_%s_representor_%d",
+			 pci_dev->device.name, eth_da.representor_ports[i]);
+
+		ret = rte_eth_dev_create(&pci_dev->device, name,
+					 sizeof(struct bnxt_vf_representor),
+					 NULL, NULL,
+					 bnxt_vf_representor_init,
+					 &representor);
+
+		if (!ret) {
+			vf_rep_eth_dev = rte_eth_dev_allocated(name);
+			if (!vf_rep_eth_dev) {
+				PMD_DRV_LOG(ERR, "Failed to find the eth_dev"
+					    " for VF-Rep: %s.", name);
+				bnxt_pci_remove_dev_with_reps(backing_eth_dev);
+				ret = -ENODEV;
+				return ret;
+			}
+			backing_bp->rep_info[representor.vf_id].vfr_eth_dev =
+				vf_rep_eth_dev;
+			backing_bp->num_reps++;
+		} else {
+			PMD_DRV_LOG(ERR, "failed to create bnxt vf "
+				    "representor %s.", name);
+			bnxt_pci_remove_dev_with_reps(backing_eth_dev);
+		}
+	}
+
+	return ret;
 }
 
 static int bnxt_pci_remove(struct rte_pci_device *pci_dev)
 {
-	if (rte_eal_process_type() == RTE_PROC_PRIMARY)
-		return rte_eth_dev_pci_generic_remove(pci_dev,
-				bnxt_dev_uninit);
-	else
+	struct rte_eth_dev *eth_dev;
+
+	eth_dev = rte_eth_dev_allocated(pci_dev->device.name);
+	if (!eth_dev)
+		return ENODEV; /* Invoked typically only by OVS-DPDK, by the
+				* time it comes here the eth_dev is already
+				* deleted by rte_eth_dev_close(), so returning
+				* +ve value will atleast help in proper cleanup
+				*/
+
+	if (rte_eal_process_type() == RTE_PROC_PRIMARY) {
+		if (eth_dev->data->dev_flags & RTE_ETH_DEV_REPRESENTOR)
+			return rte_eth_dev_destroy(eth_dev,
+						   bnxt_vf_representor_uninit);
+		else
+			return rte_eth_dev_destroy(eth_dev,
+						   bnxt_dev_uninit);
+	} else {
 		return rte_eth_dev_pci_generic_remove(pci_dev, NULL);
+	}
 }
 
 static struct rte_pci_driver bnxt_rte_pmd = {
 	.id_table = bnxt_pci_id_map,
-	.drv_flags = RTE_PCI_DRV_NEED_MAPPING | RTE_PCI_DRV_INTR_LSC,
+	.drv_flags = RTE_PCI_DRV_NEED_MAPPING | RTE_PCI_DRV_INTR_LSC |
+			RTE_PCI_DRV_PROBE_AGAIN, /* Needed in case of VF-REPs
+						  * and OVS-DPDK
+						  */
 	.probe = bnxt_pci_probe,
 	.remove = bnxt_pci_remove,
 };
diff --git a/drivers/net/bnxt/bnxt_reps.c b/drivers/net/bnxt/bnxt_reps.c
new file mode 100644
index 0000000..7033d62
--- /dev/null
+++ b/drivers/net/bnxt/bnxt_reps.c
@@ -0,0 +1,287 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2014-2018 Broadcom
+ * All rights reserved.
+ */
+
+#include "bnxt.h"
+#include "bnxt_ring.h"
+#include "bnxt_reps.h"
+#include "hsi_struct_def_dpdk.h"
+
+static const struct eth_dev_ops bnxt_vf_rep_dev_ops = {
+	.dev_infos_get = bnxt_vf_rep_dev_info_get_op,
+	.dev_configure = bnxt_vf_rep_dev_configure_op,
+	.dev_start = bnxt_vf_rep_dev_start_op,
+	.rx_queue_setup = bnxt_vf_rep_rx_queue_setup_op,
+	.tx_queue_setup = bnxt_vf_rep_tx_queue_setup_op,
+	.link_update = bnxt_vf_rep_link_update_op,
+	.dev_close = bnxt_vf_rep_dev_close_op,
+	.dev_stop = bnxt_vf_rep_dev_stop_op
+};
+
+static uint16_t
+bnxt_vf_rep_rx_burst(__rte_unused void *rx_queue,
+		     __rte_unused struct rte_mbuf **rx_pkts,
+		     __rte_unused uint16_t nb_pkts)
+{
+	return 0;
+}
+
+static uint16_t
+bnxt_vf_rep_tx_burst(__rte_unused void *tx_queue,
+		     __rte_unused struct rte_mbuf **tx_pkts,
+		     __rte_unused uint16_t nb_pkts)
+{
+	return 0;
+}
+
+int bnxt_vf_representor_init(struct rte_eth_dev *eth_dev, void *params)
+{
+	struct bnxt_vf_representor *vf_rep_bp = eth_dev->data->dev_private;
+	struct bnxt_vf_representor *rep_params =
+				 (struct bnxt_vf_representor *)params;
+	struct rte_eth_link *link;
+	struct bnxt *parent_bp;
+
+	vf_rep_bp->vf_id = rep_params->vf_id;
+	vf_rep_bp->switch_domain_id = rep_params->switch_domain_id;
+	vf_rep_bp->parent_priv = rep_params->parent_priv;
+
+	eth_dev->data->dev_flags |= RTE_ETH_DEV_REPRESENTOR;
+	eth_dev->data->representor_id = rep_params->vf_id;
+
+	rte_eth_random_addr(vf_rep_bp->dflt_mac_addr);
+	memcpy(vf_rep_bp->mac_addr, vf_rep_bp->dflt_mac_addr,
+	       sizeof(vf_rep_bp->mac_addr));
+	eth_dev->data->mac_addrs =
+		(struct rte_ether_addr *)&vf_rep_bp->mac_addr;
+	eth_dev->dev_ops = &bnxt_vf_rep_dev_ops;
+
+	/* No data-path, but need stub Rx/Tx functions to avoid crash
+	 * when testing with ovs-dpdk
+	 */
+	eth_dev->rx_pkt_burst = bnxt_vf_rep_rx_burst;
+	eth_dev->tx_pkt_burst = bnxt_vf_rep_tx_burst;
+	/* Link state. Inherited from PF or trusted VF */
+	parent_bp = vf_rep_bp->parent_priv;
+	link = &parent_bp->eth_dev->data->dev_link;
+
+	eth_dev->data->dev_link.link_speed = link->link_speed;
+	eth_dev->data->dev_link.link_duplex = link->link_duplex;
+	eth_dev->data->dev_link.link_status = link->link_status;
+	eth_dev->data->dev_link.link_autoneg = link->link_autoneg;
+
+	PMD_DRV_LOG(INFO, "calling bnxt_print_link_info\n");
+	bnxt_print_link_info(eth_dev);
+
+	/* Pass the information to the rte_eth_dev_close() that it should also
+	 * release the private port resources.
+	 */
+	eth_dev->data->dev_flags |= RTE_ETH_DEV_CLOSE_REMOVE;
+	PMD_DRV_LOG(INFO,
+		    "Switch domain id %d: Representor Device %d init done\n",
+		    vf_rep_bp->switch_domain_id, vf_rep_bp->vf_id);
+
+	return 0;
+}
+
+int bnxt_vf_representor_uninit(struct rte_eth_dev *eth_dev)
+{
+	struct bnxt *parent_bp;
+	struct bnxt_vf_representor *rep =
+		(struct bnxt_vf_representor *)eth_dev->data->dev_private;
+
+	uint16_t vf_id;
+
+	eth_dev->data->mac_addrs = NULL;
+
+	parent_bp = rep->parent_priv;
+	if (parent_bp) {
+		parent_bp->num_reps--;
+		vf_id = rep->vf_id;
+		if (parent_bp->rep_info) {
+			memset(&parent_bp->rep_info[vf_id], 0,
+			       sizeof(parent_bp->rep_info[vf_id]));
+			/* mark that this representor has been freed */
+		}
+	}
+	eth_dev->dev_ops = NULL;
+	return 0;
+}
+
+int bnxt_vf_rep_link_update_op(struct rte_eth_dev *eth_dev, int wait_to_compl)
+{
+	struct bnxt *parent_bp;
+	struct bnxt_vf_representor *rep =
+		(struct bnxt_vf_representor *)eth_dev->data->dev_private;
+	struct rte_eth_link *link;
+	int rc;
+
+	parent_bp = rep->parent_priv;
+	rc = bnxt_link_update_op(parent_bp->eth_dev, wait_to_compl);
+
+	/* Link state. Inherited from PF or trusted VF */
+	link = &parent_bp->eth_dev->data->dev_link;
+
+	eth_dev->data->dev_link.link_speed = link->link_speed;
+	eth_dev->data->dev_link.link_duplex = link->link_duplex;
+	eth_dev->data->dev_link.link_status = link->link_status;
+	eth_dev->data->dev_link.link_autoneg = link->link_autoneg;
+	bnxt_print_link_info(eth_dev);
+
+	return rc;
+}
+
+int bnxt_vf_rep_dev_start_op(struct rte_eth_dev *eth_dev)
+{
+	bnxt_vf_rep_link_update_op(eth_dev, 1);
+
+	return 0;
+}
+
+void bnxt_vf_rep_dev_stop_op(struct rte_eth_dev *eth_dev)
+{
+	eth_dev = eth_dev;
+}
+
+void bnxt_vf_rep_dev_close_op(struct rte_eth_dev *eth_dev)
+{
+	bnxt_vf_representor_uninit(eth_dev);
+}
+
+int bnxt_vf_rep_dev_info_get_op(struct rte_eth_dev *eth_dev,
+				struct rte_eth_dev_info *dev_info)
+{
+	struct bnxt_vf_representor *rep_bp = eth_dev->data->dev_private;
+	struct bnxt *parent_bp;
+	uint16_t max_vnics, i, j, vpool, vrxq;
+	unsigned int max_rx_rings;
+	int rc = 0;
+
+	/* MAC Specifics */
+	parent_bp = rep_bp->parent_priv;
+	if (!parent_bp) {
+		PMD_DRV_LOG(ERR, "Rep parent NULL!\n");
+		return rc;
+	}
+	PMD_DRV_LOG(DEBUG, "Representor dev_info_get_op\n");
+	dev_info->max_mac_addrs = parent_bp->max_l2_ctx;
+	dev_info->max_hash_mac_addrs = 0;
+
+	max_rx_rings = BNXT_MAX_VF_REP_RINGS;
+	/* For the sake of symmetry, max_rx_queues = max_tx_queues */
+	dev_info->max_rx_queues = max_rx_rings;
+	dev_info->max_tx_queues = max_rx_rings;
+	dev_info->reta_size = bnxt_rss_hash_tbl_size(parent_bp);
+	dev_info->hash_key_size = 40;
+	max_vnics = parent_bp->max_vnics;
+
+	/* MTU specifics */
+	dev_info->min_mtu = RTE_ETHER_MIN_MTU;
+	dev_info->max_mtu = BNXT_MAX_MTU;
+
+	/* Fast path specifics */
+	dev_info->min_rx_bufsize = 1;
+	dev_info->max_rx_pktlen = BNXT_MAX_PKT_LEN;
+
+	dev_info->rx_offload_capa = BNXT_DEV_RX_OFFLOAD_SUPPORT;
+	if (parent_bp->flags & BNXT_FLAG_PTP_SUPPORTED)
+		dev_info->rx_offload_capa |= DEV_RX_OFFLOAD_TIMESTAMP;
+	dev_info->tx_offload_capa = BNXT_DEV_TX_OFFLOAD_SUPPORT;
+	dev_info->flow_type_rss_offloads = BNXT_ETH_RSS_SUPPORT;
+
+	/* *INDENT-OFF* */
+	dev_info->default_rxconf = (struct rte_eth_rxconf) {
+		.rx_thresh = {
+			.pthresh = 8,
+			.hthresh = 8,
+			.wthresh = 0,
+		},
+		.rx_free_thresh = 32,
+		/* If no descriptors available, pkts are dropped by default */
+		.rx_drop_en = 1,
+	};
+
+	dev_info->default_txconf = (struct rte_eth_txconf) {
+		.tx_thresh = {
+			.pthresh = 32,
+			.hthresh = 0,
+			.wthresh = 0,
+		},
+		.tx_free_thresh = 32,
+		.tx_rs_thresh = 32,
+	};
+	eth_dev->data->dev_conf.intr_conf.lsc = 1;
+
+	eth_dev->data->dev_conf.intr_conf.rxq = 1;
+	dev_info->rx_desc_lim.nb_min = BNXT_MIN_RING_DESC;
+	dev_info->rx_desc_lim.nb_max = BNXT_MAX_RX_RING_DESC;
+	dev_info->tx_desc_lim.nb_min = BNXT_MIN_RING_DESC;
+	dev_info->tx_desc_lim.nb_max = BNXT_MAX_TX_RING_DESC;
+
+	/* *INDENT-ON* */
+
+	/*
+	 * TODO: default_rxconf, default_txconf, rx_desc_lim, and tx_desc_lim
+	 *       need further investigation.
+	 */
+
+	/* VMDq resources */
+	vpool = 64; /* ETH_64_POOLS */
+	vrxq = 128; /* ETH_VMDQ_DCB_NUM_QUEUES */
+	for (i = 0; i < 4; vpool >>= 1, i++) {
+		if (max_vnics > vpool) {
+			for (j = 0; j < 5; vrxq >>= 1, j++) {
+				if (dev_info->max_rx_queues > vrxq) {
+					if (vpool > vrxq)
+						vpool = vrxq;
+					goto found;
+				}
+			}
+			/* Not enough resources to support VMDq */
+			break;
+		}
+	}
+	/* Not enough resources to support VMDq */
+	vpool = 0;
+	vrxq = 0;
+found:
+	dev_info->max_vmdq_pools = vpool;
+	dev_info->vmdq_queue_num = vrxq;
+
+	dev_info->vmdq_pool_base = 0;
+	dev_info->vmdq_queue_base = 0;
+
+	return 0;
+}
+
+int bnxt_vf_rep_dev_configure_op(__rte_unused struct rte_eth_dev *eth_dev)
+{
+	PMD_DRV_LOG(DEBUG, "Representor dev_configure_op\n");
+	return 0;
+}
+
+int bnxt_vf_rep_rx_queue_setup_op(struct rte_eth_dev *eth_dev,
+				  __rte_unused uint16_t queue_idx,
+				  __rte_unused uint16_t nb_desc,
+				  __rte_unused unsigned int socket_id,
+				  __rte_unused const struct rte_eth_rxconf *
+				  rx_conf,
+				  __rte_unused struct rte_mempool *mp)
+{
+	eth_dev = eth_dev;
+
+	return 0;
+}
+
+int bnxt_vf_rep_tx_queue_setup_op(struct rte_eth_dev *eth_dev,
+				  __rte_unused uint16_t queue_idx,
+				  __rte_unused uint16_t nb_desc,
+				  __rte_unused unsigned int socket_id,
+				  __rte_unused const struct rte_eth_txconf *
+				  tx_conf)
+{
+	eth_dev = eth_dev;
+
+	return 0;
+}
diff --git a/drivers/net/bnxt/bnxt_reps.h b/drivers/net/bnxt/bnxt_reps.h
new file mode 100644
index 0000000..f4c033a
--- /dev/null
+++ b/drivers/net/bnxt/bnxt_reps.h
@@ -0,0 +1,35 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2014-2018 Broadcom
+ * All rights reserved.
+ */
+
+#ifndef _BNXT_REPS_H_
+#define _BNXT_REPS_H_
+
+#include <rte_malloc.h>
+#include <rte_ethdev.h>
+
+int bnxt_vf_representor_init(struct rte_eth_dev *eth_dev, void *params);
+int bnxt_vf_representor_uninit(struct rte_eth_dev *eth_dev);
+int bnxt_vf_rep_dev_info_get_op(struct rte_eth_dev *eth_dev,
+				struct rte_eth_dev_info *dev_info);
+int bnxt_vf_rep_dev_configure_op(struct rte_eth_dev *eth_dev);
+
+int bnxt_vf_rep_link_update_op(struct rte_eth_dev *eth_dev, int wait_to_compl);
+int bnxt_vf_rep_dev_start_op(struct rte_eth_dev *eth_dev);
+int bnxt_vf_rep_rx_queue_setup_op(struct rte_eth_dev *eth_dev,
+				  __rte_unused uint16_t queue_idx,
+				  __rte_unused uint16_t nb_desc,
+				  __rte_unused unsigned int socket_id,
+				  __rte_unused const struct rte_eth_rxconf *
+				  rx_conf,
+				  __rte_unused struct rte_mempool *mp);
+int bnxt_vf_rep_tx_queue_setup_op(struct rte_eth_dev *eth_dev,
+				  __rte_unused uint16_t queue_idx,
+				  __rte_unused uint16_t nb_desc,
+				  __rte_unused unsigned int socket_id,
+				  __rte_unused const struct rte_eth_txconf *
+				  tx_conf);
+void bnxt_vf_rep_dev_stop_op(struct rte_eth_dev *eth_dev);
+void bnxt_vf_rep_dev_close_op(struct rte_eth_dev *eth_dev);
+#endif /* _BNXT_REPS_H_ */
diff --git a/drivers/net/bnxt/meson.build b/drivers/net/bnxt/meson.build
index 4306c60..5c7859c 100644
--- a/drivers/net/bnxt/meson.build
+++ b/drivers/net/bnxt/meson.build
@@ -21,6 +21,7 @@ sources = files('bnxt_cpr.c',
 	'bnxt_txr.c',
 	'bnxt_util.c',
 	'bnxt_vnic.c',
+	'bnxt_reps.c',
 
 	'tf_core/tf_core.c',
 	'tf_core/bitalloc.c',
-- 
2.7.4