[dpdk-dev] [PATCH 11/40] bnxt: add Rx queue create/destroy operations

Stephen Hurd stephen.hurd at broadcom.com
Fri May 6 21:25:51 CEST 2016


Initial create/destroy queue code.  Requires RX ring support to be
functional.

Signed-off-by: Stephen Hurd <stephen.hurd at broadcom.com>
Reviewed-by: Ajit Kumar Khaparde <ajit.khaparde at broadcom.com>
---
 drivers/net/bnxt/Makefile              |   1 +
 drivers/net/bnxt/bnxt.h                |   2 +
 drivers/net/bnxt/bnxt_ethdev.c         |   3 +
 drivers/net/bnxt/bnxt_rxq.c            | 286 +++++++++++++++++++++++++++++++++
 drivers/net/bnxt/bnxt_rxq.h            |  74 +++++++++
 drivers/net/bnxt/hsi_struct_def_dpdk.h | 121 ++++++++++++++
 6 files changed, 487 insertions(+)
 create mode 100644 drivers/net/bnxt/bnxt_rxq.c
 create mode 100644 drivers/net/bnxt/bnxt_rxq.h

diff --git a/drivers/net/bnxt/Makefile b/drivers/net/bnxt/Makefile
index 0759df0..7320fac 100644
--- a/drivers/net/bnxt/Makefile
+++ b/drivers/net/bnxt/Makefile
@@ -53,6 +53,7 @@ SRCS-$(CONFIG_RTE_LIBRTE_BNXT_PMD) += bnxt_ethdev.c
 SRCS-$(CONFIG_RTE_LIBRTE_BNXT_PMD) += bnxt_filter.c
 SRCS-$(CONFIG_RTE_LIBRTE_BNXT_PMD) += bnxt_hwrm.c
 SRCS-$(CONFIG_RTE_LIBRTE_BNXT_PMD) += bnxt_ring.c
+SRCS-$(CONFIG_RTE_LIBRTE_BNXT_PMD) += bnxt_rxq.c
 SRCS-$(CONFIG_RTE_LIBRTE_BNXT_PMD) += bnxt_txq.c
 SRCS-$(CONFIG_RTE_LIBRTE_BNXT_PMD) += bnxt_vnic.c
 
diff --git a/drivers/net/bnxt/bnxt.h b/drivers/net/bnxt/bnxt.h
index 54ddd24..38b590b 100644
--- a/drivers/net/bnxt/bnxt.h
+++ b/drivers/net/bnxt/bnxt.h
@@ -143,6 +143,8 @@ struct bnxt {
 	/* Default completion ring */
 	struct bnxt_cp_ring_info	def_cp_ring;
 
+	unsigned		nr_vnics;
+
 	struct bnxt_vnic_info	*vnic_info;
 	STAILQ_HEAD(, bnxt_vnic_info)	free_vnic_list;
 
diff --git a/drivers/net/bnxt/bnxt_ethdev.c b/drivers/net/bnxt/bnxt_ethdev.c
index d8dbc10..df39fae 100644
--- a/drivers/net/bnxt/bnxt_ethdev.c
+++ b/drivers/net/bnxt/bnxt_ethdev.c
@@ -41,6 +41,7 @@
 
 #include "bnxt.h"
 #include "bnxt_hwrm.h"
+#include "bnxt_rxq.h"
 #include "bnxt_txq.h"
 
 #define DRV_MODULE_NAME		"bnxt"
@@ -177,6 +178,8 @@ static int bnxt_dev_configure_op(struct rte_eth_dev *eth_dev)
 static struct eth_dev_ops bnxt_dev_ops = {
 	.dev_infos_get = bnxt_dev_info_get_op,
 	.dev_configure = bnxt_dev_configure_op,
+	.rx_queue_setup = bnxt_rx_queue_setup_op,
+	.rx_queue_release = bnxt_rx_queue_release_op,
 	.tx_queue_setup = bnxt_tx_queue_setup_op,
 	.tx_queue_release = bnxt_tx_queue_release_op,
 };
diff --git a/drivers/net/bnxt/bnxt_rxq.c b/drivers/net/bnxt/bnxt_rxq.c
new file mode 100644
index 0000000..4ba5d75
--- /dev/null
+++ b/drivers/net/bnxt/bnxt_rxq.c
@@ -0,0 +1,286 @@
+/*-
+ *   BSD LICENSE
+ *
+ *   Copyright(c) Broadcom Limited.
+ *   All rights reserved.
+ *
+ *   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.
+ *     * Neither the name of Broadcom Corporation nor the names of its
+ *       contributors may be used to endorse or promote products derived
+ *       from this software without specific prior written permission.
+ *
+ *   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ *   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ *   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ *   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ *   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ *   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ *   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ *   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ *   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ *   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ *   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include <inttypes.h>
+
+#include <rte_malloc.h>
+
+#include "bnxt.h"
+#include "bnxt_filter.h"
+#include "bnxt_hwrm.h"
+#include "bnxt_ring.h"
+#include "bnxt_rxq.h"
+#include "bnxt_vnic.h"
+#include "hsi_struct_def_dpdk.h"
+
+/*
+ * RX Queues
+ */
+
+void bnxt_free_rxq_stats(struct bnxt_rx_queue *rxq)
+{
+	struct bnxt_cp_ring_info *cpr = rxq->cp_ring;
+
+	/* 'Unreserve' rte_memzone */
+	/* N/A */
+
+	if (cpr->hw_stats)
+		cpr->hw_stats = NULL;
+}
+
+int bnxt_mq_rx_configure(struct bnxt *bp)
+{
+	struct rte_eth_conf *dev_conf = &bp->eth_dev->data->dev_conf;
+	unsigned i, j, nb_q_per_grp, ring_idx;
+	int start_grp_id, end_grp_id, rc = 0;
+	struct bnxt_vnic_info *vnic;
+	struct bnxt_filter_info *filter;
+	struct bnxt_rx_queue *rxq;
+
+	bp->nr_vnics = 0;
+
+	/* Single queue mode */
+	if (bp->rx_cp_nr_rings < 2) {
+		vnic = bnxt_alloc_vnic(bp);
+		if (!vnic) {
+			RTE_LOG(ERR, PMD, "VNIC alloc failed\n");
+			rc = -ENOMEM;
+			goto err_out;
+		}
+		STAILQ_INSERT_TAIL(&bp->ff_pool[0], vnic, next);
+		bp->nr_vnics++;
+
+		rxq = bp->eth_dev->data->rx_queues[0];
+		rxq->vnic = vnic;
+
+		vnic->func_default = true;
+		vnic->ff_pool_idx = 0;
+		vnic->start_grp_id = 1;
+		vnic->end_grp_id = vnic->start_grp_id +
+				   bp->rx_cp_nr_rings - 1;
+		filter = bnxt_alloc_filter(bp);
+		if (!filter) {
+			RTE_LOG(ERR, PMD, "L2 filter alloc failed\n");
+			rc = -ENOMEM;
+			goto err_out;
+		}
+		STAILQ_INSERT_TAIL(&vnic->filter, filter, next);
+		goto out;
+	}
+
+	/* Multi-queue mode */
+	if (dev_conf->rxmode.mq_mode & ETH_MQ_RX_VMDQ_FLAG) {
+		/* VMDq ONLY, VMDq+RSS, VMDq+DCB, VMDq+DCB+RSS */
+		enum rte_eth_nb_pools pools;
+
+		switch (dev_conf->rxmode.mq_mode) {
+		case ETH_MQ_RX_VMDQ_RSS:
+		case ETH_MQ_RX_VMDQ_ONLY:
+			{
+				const struct rte_eth_vmdq_rx_conf *conf =
+				    &dev_conf->rx_adv_conf.vmdq_rx_conf;
+
+				/* ETH_8/64_POOLs */
+				pools = conf->nb_queue_pools;
+				break;
+			}
+		default:
+			RTE_LOG(ERR, PMD, "Unsupported mq_mod %d\n",
+				dev_conf->rxmode.mq_mode);
+			rc = -EINVAL;
+			goto err_out;
+		}
+		/* For each pool, allocate MACVLAN CFA rule & VNIC */
+		if (!pools) {
+			RTE_LOG(ERR, PMD,
+				"VMDq pool not set, defaulted to 64\n");
+			pools = ETH_64_POOLS;
+		}
+		nb_q_per_grp = bp->rx_cp_nr_rings / pools;
+		start_grp_id = 1;
+		end_grp_id = start_grp_id + nb_q_per_grp - 1;
+
+		ring_idx = 0;
+		for (i = 0; i < pools; i++) {
+			vnic = bnxt_alloc_vnic(bp);
+			if (!vnic) {
+				RTE_LOG(ERR, PMD,
+					"VNIC alloc failed\n");
+				rc = -ENOMEM;
+				goto err_out;
+			}
+			STAILQ_INSERT_TAIL(&bp->ff_pool[i], vnic, next);
+			bp->nr_vnics++;
+
+			for (j = 0; j < nb_q_per_grp; j++, ring_idx++) {
+				rxq = bp->eth_dev->data->rx_queues[ring_idx];
+				rxq->vnic = vnic;
+			}
+			if (i == 0)
+				vnic->func_default = true;
+			vnic->ff_pool_idx = i;
+			vnic->start_grp_id = start_grp_id;
+			vnic->end_grp_id = end_grp_id;
+
+			filter = bnxt_alloc_filter(bp);
+			if (!filter) {
+				RTE_LOG(ERR, PMD,
+					"L2 filter alloc failed\n");
+				rc = -ENOMEM;
+				goto err_out;
+			}
+			/* TODO: Configure & associate CFA rule for
+			   each VNIC for each VMDq with MACVLAN, MACVLAN+TC */
+			STAILQ_INSERT_TAIL(&vnic->filter, filter, next);
+
+			start_grp_id = end_grp_id + 1;
+			end_grp_id += nb_q_per_grp;
+		}
+		goto out;
+	}
+
+	/* Non-VMDq mode - RSS, DCB, RSS+DCB */
+	/* Init default VNIC for RSS or DCB only */
+	vnic = bnxt_alloc_vnic(bp);
+	if (!vnic) {
+		RTE_LOG(ERR, PMD, "VNIC alloc failed\n");
+		rc = -ENOMEM;
+		goto err_out;
+	}
+	/* Partition the rx queues for the single pool */
+	for (i = 0; i < bp->rx_cp_nr_rings; i++) {
+		rxq = bp->eth_dev->data->rx_queues[i];
+		rxq->vnic = vnic;
+	}
+	STAILQ_INSERT_TAIL(&bp->ff_pool[0], vnic, next);
+	bp->nr_vnics++;
+
+	vnic->func_default = true;
+	vnic->ff_pool_idx = 0;
+	vnic->start_grp_id = 1;
+	vnic->end_grp_id = vnic->start_grp_id +
+			   bp->rx_cp_nr_rings - 1;
+	filter = bnxt_alloc_filter(bp);
+	if (!filter) {
+		RTE_LOG(ERR, PMD, "L2 filter alloc failed\n");
+		rc = -ENOMEM;
+		goto err_out;
+	}
+	STAILQ_INSERT_TAIL(&vnic->filter, filter, next);
+
+	if (dev_conf->rxmode.mq_mode & ETH_MQ_RX_RSS_FLAG)
+		vnic->hash_type =
+			HWRM_VNIC_RSS_CFG_INPUT_HASH_TYPE_IPV4 |
+			HWRM_VNIC_RSS_CFG_INPUT_HASH_TYPE_IPV6;
+
+out:
+	return rc;
+
+err_out:
+	/* Free allocated vnic/filters */
+
+	return rc;
+}
+
+static void bnxt_rx_queue_release_mbufs(struct bnxt_rx_queue *rxq __rte_unused)
+{
+	/* TODO: Requires interaction with TX ring */
+}
+
+void bnxt_free_rx_mbufs(struct bnxt *bp)
+{
+	struct bnxt_rx_queue *rxq;
+	int i;
+
+	for (i = 0; i < (int)bp->rx_nr_rings; i++) {
+		rxq = bp->rx_queues[i];
+		bnxt_rx_queue_release_mbufs(rxq);
+	}
+}
+
+void bnxt_rx_queue_release_op(void *rx_queue)
+{
+	struct bnxt_rx_queue *rxq = (struct bnxt_rx_queue *)rx_queue;
+
+	if (rxq) {
+		bnxt_rx_queue_release_mbufs(rxq);
+
+		/* TODO: Free ring and stats here */
+
+		rte_free(rxq);
+	}
+}
+
+int bnxt_rx_queue_setup_op(struct rte_eth_dev *eth_dev,
+			       uint16_t queue_idx,
+			       uint16_t nb_desc,
+			       unsigned int socket_id,
+			       const struct rte_eth_rxconf *rx_conf,
+			       struct rte_mempool *mp)
+{
+	struct bnxt *bp = (struct bnxt *)eth_dev->data->dev_private;
+	struct bnxt_rx_queue *rxq;
+
+	if (!nb_desc || nb_desc > MAX_RX_DESC_CNT) {
+		RTE_LOG(ERR, PMD, "nb_desc %d is invalid", nb_desc);
+		return -EINVAL;
+	}
+
+	if (eth_dev->data->rx_queues) {
+		rxq = eth_dev->data->rx_queues[queue_idx];
+		if (rxq)
+			bnxt_rx_queue_release_op(rxq);
+	}
+	rxq = rte_zmalloc_socket("bnxt_rx_queue", sizeof(struct bnxt_rx_queue),
+				 RTE_CACHE_LINE_SIZE, socket_id);
+	if (!rxq) {
+		RTE_LOG(ERR, PMD, "bnxt_rx_queue allocation failed!");
+		return -ENOMEM;
+	}
+	rxq->bp = bp;
+	rxq->mb_pool = mp;
+	rxq->nb_rx_desc = nb_desc;
+	rxq->rx_free_thresh = rx_conf->rx_free_thresh;
+
+	/* TODO: Initialize ring structure */
+
+	rxq->queue_id = queue_idx;
+	rxq->port_id = eth_dev->data->port_id;
+	rxq->crc_len = (uint8_t)((eth_dev->data->dev_conf.rxmode.hw_strip_crc) ?
+				0 : ETHER_CRC_LEN);
+
+	eth_dev->data->rx_queues[queue_idx] = rxq;
+	/* TODO: Allocate RX ring hardware descriptors */
+
+	return 0;
+}
diff --git a/drivers/net/bnxt/bnxt_rxq.h b/drivers/net/bnxt/bnxt_rxq.h
new file mode 100644
index 0000000..9554329
--- /dev/null
+++ b/drivers/net/bnxt/bnxt_rxq.h
@@ -0,0 +1,74 @@
+/*-
+ *   BSD LICENSE
+ *
+ *   Copyright(c) Broadcom Limited.
+ *   All rights reserved.
+ *
+ *   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.
+ *     * Neither the name of Broadcom Corporation nor the names of its
+ *       contributors may be used to endorse or promote products derived
+ *       from this software without specific prior written permission.
+ *
+ *   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ *   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ *   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ *   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ *   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ *   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ *   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ *   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ *   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ *   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ *   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#ifndef _BNXT_RQX_H_
+#define _BNXT_RQX_H_
+
+struct bnxt;
+struct bnxt_rx_ring_info;
+struct bnxt_cp_ring_info;
+struct bnxt_rx_queue {
+	struct rte_mempool	*mb_pool; /* mbuf pool for RX ring */
+	struct rte_mbuf		*pkt_first_seg; /* 1st seg of pkt */
+	struct rte_mbuf		*pkt_last_seg; /* Last seg of pkt */
+	uint64_t		mbuf_initializer; /* val to init mbuf */
+	uint16_t		nb_rx_desc; /* num of RX desc */
+	uint16_t		rx_tail; /* cur val of RDT register */
+	uint16_t		nb_rx_hold; /* num held free RX desc */
+	uint16_t		rx_free_thresh; /* max free RX desc to hold */
+	uint16_t		queue_id; /* RX queue index */
+	uint16_t		reg_idx; /* RX queue register index */
+	uint8_t			port_id; /* Device port identifier */
+	uint8_t			crc_len; /* 0 if CRC stripped, 4 otherwise */
+
+	struct bnxt		*bp;
+	struct bnxt_vnic_info	*vnic;
+
+	uint32_t			rx_buf_size;
+	uint32_t			rx_buf_use_size;  /* useable size */
+	struct bnxt_rx_ring_info	*rx_ring;
+	struct bnxt_cp_ring_info	*cp_ring;
+};
+
+void bnxt_free_rxq_stats(struct bnxt_rx_queue *rxq);
+int bnxt_mq_rx_configure(struct bnxt *bp);
+void bnxt_rx_queue_release_op(void *rx_queue);
+int bnxt_rx_queue_setup_op(struct rte_eth_dev *eth_dev,
+			       uint16_t queue_idx,
+			       uint16_t nb_desc,
+			       unsigned int socket_id,
+			       const struct rte_eth_rxconf *rx_conf,
+			       struct rte_mempool *mp);
+void bnxt_free_rx_mbufs(struct bnxt *bp);
+
+#endif
diff --git a/drivers/net/bnxt/hsi_struct_def_dpdk.h b/drivers/net/bnxt/hsi_struct_def_dpdk.h
index bfa8a7c..c5ff9ff 100644
--- a/drivers/net/bnxt/hsi_struct_def_dpdk.h
+++ b/drivers/net/bnxt/hsi_struct_def_dpdk.h
@@ -1946,6 +1946,127 @@ struct hwrm_queue_qportcfg_input {
 	uint16_t unused_0;
 } __attribute__((packed));
 
+/* hwrm_vnic_rss_cfg */
+/* Description: This function is used to enable RSS configuration. */
+
+/* Input (48 bytes) */
+struct hwrm_vnic_rss_cfg_input {
+	/*
+	 * This value indicates what type of request this is. The format for the
+	 * rest of the command is determined by this field.
+	 */
+	uint16_t req_type;
+
+	/*
+	 * This value indicates the what completion ring the request will be
+	 * optionally completed on. If the value is -1, then no CR completion
+	 * will be generated. Any other value must be a valid CR ring_id value
+	 * for this function.
+	 */
+	uint16_t cmpl_ring;
+
+	/* This value indicates the command sequence number. */
+	uint16_t seq_id;
+
+	/*
+	 * Target ID of this command. 0x0 - 0xFFF8 - Used for function ids
+	 * 0xFFF8 - 0xFFFE - Reserved for internal processors 0xFFFF - HWRM
+	 */
+	uint16_t target_id;
+
+	/*
+	 * This is the host address where the response will be written when the
+	 * request is complete. This area must be 16B aligned and must be
+	 * cleared to zero before the request is made.
+	 */
+	uint64_t resp_addr;
+
+	/*
+	 * When this bit is '1', the RSS hash shall be computed over source and
+	 * destination IPv4 addresses of IPv4 packets.
+	 */
+	#define HWRM_VNIC_RSS_CFG_INPUT_HASH_TYPE_IPV4		UINT32_C(0x1)
+	/*
+	 * When this bit is '1', the RSS hash shall be computed over
+	 * source/destination IPv4 addresses and source/destination ports of
+	 * TCP/IPv4 packets.
+	 */
+	#define HWRM_VNIC_RSS_CFG_INPUT_HASH_TYPE_TCP_IPV4	UINT32_C(0x2)
+	/*
+	 * When this bit is '1', the RSS hash shall be computed over
+	 * source/destination IPv4 addresses and source/destination ports of
+	 * UDP/IPv4 packets.
+	 */
+	#define HWRM_VNIC_RSS_CFG_INPUT_HASH_TYPE_UDP_IPV4	UINT32_C(0x4)
+	/*
+	 * When this bit is '1', the RSS hash shall be computed over source and
+	 * destination IPv4 addresses of IPv6 packets.
+	 */
+	#define HWRM_VNIC_RSS_CFG_INPUT_HASH_TYPE_IPV6		UINT32_C(0x8)
+	/*
+	 * When this bit is '1', the RSS hash shall be computed over
+	 * source/destination IPv6 addresses and source/destination ports of
+	 * TCP/IPv6 packets.
+	 */
+	#define HWRM_VNIC_RSS_CFG_INPUT_HASH_TYPE_TCP_IPV6	UINT32_C(0x10)
+	/*
+	 * When this bit is '1', the RSS hash shall be computed over
+	 * source/destination IPv6 addresses and source/destination ports of
+	 * UDP/IPv6 packets.
+	 */
+	#define HWRM_VNIC_RSS_CFG_INPUT_HASH_TYPE_UDP_IPV6	UINT32_C(0x20)
+	uint32_t hash_type;
+
+	uint32_t unused_0;
+
+	/* This is the address for rss ring group table */
+	uint64_t ring_grp_tbl_addr;
+
+	/* This is the address for rss hash key table */
+	uint64_t hash_key_tbl_addr;
+
+	/* Index to the rss indirection table. */
+	uint16_t rss_ctx_idx;
+
+	uint16_t unused_1[3];
+} __attribute__((packed));
+
+/* Output (16 bytes) */
+struct hwrm_vnic_rss_cfg_output {
+	/*
+	 * Pass/Fail or error type Note: receiver to verify the in parameters,
+	 * and fail the call with an error when appropriate
+	 */
+	uint16_t error_code;
+
+	/* This field returns the type of original request. */
+	uint16_t req_type;
+
+	/* This field provides original sequence number of the command. */
+	uint16_t seq_id;
+
+	/*
+	 * This field is the length of the response in bytes. The last byte of
+	 * the response is a valid flag that will read as '1' when the command
+	 * has been completely written to memory.
+	 */
+	uint16_t resp_len;
+
+	uint32_t unused_0;
+	uint8_t unused_1;
+	uint8_t unused_2;
+	uint8_t unused_3;
+
+	/*
+	 * This field is used in Output records to indicate that the output is
+	 * completely written to RAM. This field should be read as '1' to
+	 * indicate that the output has been completely written. When writing a
+	 * command completion or response to an internal processor, the order of
+	 * writes has to be such that this field is written last.
+	 */
+	uint8_t valid;
+} __attribute__((packed));
+
 /* Output (32 bytes) */
 struct hwrm_queue_qportcfg_output {
 	/*
-- 
1.9.1



More information about the dev mailing list