DPDK-dev Archive on lore.kernel.org
 help / color / mirror / Atom feed
* Re: [PATCH 02/13] rte_device: make driver pointer const
From: Jan Blunck @ 2016-12-20 11:14 UTC (permalink / raw)
  To: Stephen Hemminger; +Cc: dev, Stephen Hemminger
In-Reply-To: <20161219215944.17226-3-sthemmin@microsoft.com>

On Mon, Dec 19, 2016 at 10:59 PM, Stephen Hemminger
<stephen@networkplumber.org> wrote:
> The info in rte_device about driver is immutable and
> shouldn't change.
> ---
>  lib/librte_eal/common/include/rte_dev.h | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/lib/librte_eal/common/include/rte_dev.h b/lib/librte_eal/common/include/rte_dev.h
> index 8840380d..e5471a22 100644
> --- a/lib/librte_eal/common/include/rte_dev.h
> +++ b/lib/librte_eal/common/include/rte_dev.h
> @@ -122,7 +122,7 @@ struct rte_driver;
>   */
>  struct rte_device {
>         TAILQ_ENTRY(rte_device) next; /**< Next device */
> -       struct rte_driver *driver;    /**< Associated driver */
> +       const struct rte_driver *driver;/**< Associated driver */
>         int numa_node;                /**< NUMA node connection */
>         struct rte_devargs *devargs;  /**< Device user arguments */
>  };
> --
> 2.11.0
>

Acked-by: Jan Blunck <jblunck@infradead.org>

^ permalink raw reply

* Re: [PATCH v3 0/6] libeventdev API and northbound implementation
From: Bruce Richardson @ 2016-12-20 11:13 UTC (permalink / raw)
  To: Jerin Jacob
  Cc: dev, thomas.monjalon, hemant.agrawal, gage.eads, harry.van.haaren
In-Reply-To: <1482070895-32491-1-git-send-email-jerin.jacob@caviumnetworks.com>

On Sun, Dec 18, 2016 at 07:51:29PM +0530, Jerin Jacob wrote:
> As previously discussed in RFC v1 [1], RFC v2 [2], with changes
> described in [3] (also pasted below), here is the first non-draft series
> for this new API.
> 
> [1] http://dpdk.org/ml/archives/dev/2016-August/045181.html
> [2] http://dpdk.org/ml/archives/dev/2016-October/048592.html
> [3] http://dpdk.org/ml/archives/dev/2016-October/048196.html
> 
> v2..v3:
> 
> 1) Changed struct rte_event layout more aligment balanced(Harry, Jerin)
> 2) Changed event_ptr type to void* from uintptr_t(Bruce)
> 3) Changed ev[] as const in rte_event_enqueue_burst  to disallow
> drivers from modifying the events passed in(Bruce)
> 4) Removed queue memory allocation from common code as some drivers may not need
> it(Bruce)
> 5) Removed "struct rte_event_queue_link" and replaced with queues and priorities
> in the link and link_get API to avoid one redirection to use the API(Bruce)
> 
> v1..v2:
> 1) Remove unnecessary header files from rte_eventdev.h(Thomas)
> 2) Removed PMD driver name(EVENTDEV_NAME_SKELETON_PMD) from rte_eventdev.h(Thomas)
> 3) Removed different #define for different priority schemes. Changed to
> one event device RTE_EVENT_DEV_PRIORITY_* priority (Bruce)
> 4) add const to rte_event_dev_configure(), rte_event_queue_setup(),
> rte_event_port_setup(), rte_event_port_link()(Bruce)
> 5) Fixed missing dev argument in dev->schedule() function(Bruce)
> 6) Changed \see to @see in doxgen comments(Thomas)
> 7) Added additional text in specification to clarify the queue depth(Thomas)
> 8) Changed wait to timeout across the specification(Thomas)
> 9) Added longer explanation for RTE_EVENT_OP_NEW and RTE_EVENT_OP_FORWARD(Thomas)
> 10) Fixed issue with RTE_EVENT_OP_RELEASE doxgen formatting(Thomas)
> 11) Changed to RTE_EVENT_DEV_CFG_FLAG_ from RTE_EVENT_DEV_CFG_(Thomas)
> 12) Changed to EVENT_QUEUE_CFG_FLAG_ from EVENT_QUEUE_CFG_(Thomas)
> 13) s/RTE_EVENT_TYPE_CORE/RTE_EVENT_TYPE_CPU/(Thomas, Gage)
> 14) Removed non burst API and kept only the burst API in the API specification
> (Thomas, Bruce, Harry, Jerin)
> -- Driver interface has non burst API, selection of the non burst API is based
> on num_objects == 1
> 15) sizeeof(struct rte_event) was not 16 in v1. Fixed it in v2
> -- reduced the width of event_type to 4bit to save space for future change
> -- introduced impl_opaque for implementation specific opaque data(Harry),
> Something useful for HW driver too, in the context of removal the need for sepeare
> release API.
> -- squashed other element size and provided enough space to impl_opaque(Jerin)
> -- added RTE_BUILD_BUG_ON(sizeof(struct rte_event) != 16); check
> 16) add union of uint64_t in the second element in struct rte_event to
> make sure the structure has 16byte address all arch(Thomas)
> 17) Fixed invalid check of nb_atomic_order_sequences in implementation(Gage)
> 18) s/EDEV_LOG_ERR/RTE_EDEV_LOG_ERR(Thomas)
> 19) s/rte_eventdev_pmd_/rte_event_pmd_/(Bruce)
> 20) added fine details of distributed vs centralized scheduling information
> in the specification and introduced RTE_EVENT_DEV_CAP_FLAG_DISTRIBUTED_SCHED
> flag(Gage)
> 21)s/RTE_EVENT_QUEUE_CFG_FLAG_SINGLE_CONSUMER/RTE_EVENT_QUEUE_CFG_FLAG_SINGLE_LINK (Jerin)
> to remove the confusion to between another producer and consumer in sw eventdev driver
> 22) Northbound api implementation  patch spited to more logical patches(Thomas)
> 
> Changes since RFC v2:
> 
> - Updated the documentation to define the need for this library[Jerin]
> - Added RTE_EVENT_QUEUE_CFG_*_ONLY configuration parameters in
>   struct rte_event_queue_conf to enable optimized sw implementation [Bruce]
> - Introduced RTE_EVENT_OP* ops [Bruce]
> - Added nb_event_queue_flows,nb_event_port_dequeue_depth, nb_event_port_enqueue_depth
>   in rte_event_dev_configure() like ethdev and crypto library[Jerin]
> - Removed rte_event_release() and replaced with RTE_EVENT_OP_RELEASE ops to
>   reduce fast path APIs and it is redundant too[Jerin]
> - In the view of better application portability, Removed pin_event
>   from rte_event_enqueue as it is just hint and Intel/NXP can not support it[Jerin]
> - Added rte_event_port_links_get()[Jerin]
> - Added rte_event_dev_dump[Harry]
> 
> Notes:
> 
> - This patch set is check-patch clean with an exception that
> 03/06 has one WARNING:MACRO_WITH_FLOW_CONTROL
> - Looking forward to getting additional maintainers for libeventdev
> 
> TODO:
> 1) Create user guide
> 
> Jerin Jacob (6):
>   eventdev: introduce event driven programming model
>   eventdev: define southbound driver interface
>   eventdev: implement the northbound APIs
>   eventdev: implement PMD registration functions
>   event/skeleton: add skeleton eventdev driver
>   app/test: unit test case for eventdev APIs
> 
Hi Jerin,

other than the couple of comments I've made in replies to the individual
patches, this looks pretty good to me. Only additional comment I have is
that some of the macro names are a little long, and maybe we can shorten
them  For example, you've added "_FLAG_" into the config flag macros,
and I'm not sure that is necessary. Similarly, I think we can drop
"_DEV_" from the PRIORITY names to shorten them.

Irrespective of these naming suggestions, once the other couple of
comments are taken care of, I think this set is suitable for merging to
the next-event tree.

Series Acked-by: Bruce Richardson <bruce.richardson@intel.com>

Regards,
/Bruce

^ permalink raw reply

* [PATCH v3 9/9] ethdev: Decouple struct rte_eth_dev from struct rte_pci_device
From: Jan Blunck @ 2016-12-20 11:11 UTC (permalink / raw)
  To: dev; +Cc: shreyansh.jain, david.marchand, stephen
In-Reply-To: <1482232315-21626-1-git-send-email-jblunck@infradead.org>

Signed-off-by: Jan Blunck <jblunck@infradead.org>
---
 app/test/virtual_pmd.c                  |  4 ++--
 drivers/net/bonding/rte_eth_bond_args.c | 12 ++++++++++--
 drivers/net/cxgbe/cxgbe_main.c          |  4 ++--
 drivers/net/fm10k/fm10k_ethdev.c        |  6 +++---
 drivers/net/virtio/virtio_user_ethdev.c |  1 -
 lib/librte_eal/common/include/rte_pci.h |  6 ++++++
 lib/librte_ether/rte_ethdev.c           |  6 +++---
 lib/librte_ether/rte_ethdev.h           |  4 ++--
 8 files changed, 28 insertions(+), 15 deletions(-)

diff --git a/app/test/virtual_pmd.c b/app/test/virtual_pmd.c
index 65b44c6..bd30b14 100644
--- a/app/test/virtual_pmd.c
+++ b/app/test/virtual_pmd.c
@@ -625,8 +625,8 @@ virtual_ethdev_create(const char *name, struct ether_addr *mac_addr,
 	dev_private->dev_ops = virtual_ethdev_default_dev_ops;
 	eth_dev->dev_ops = &dev_private->dev_ops;
 
-	eth_dev->pci_dev = pci_dev;
-	eth_dev->pci_dev->device.driver = &eth_drv->pci_drv.driver;
+	pci_dev->device.driver = &eth_drv->pci_drv.driver;
+	eth_dev->device = &pci_dev->device;
 
 	eth_dev->rx_pkt_burst = virtual_ethdev_rx_burst_success;
 	eth_dev->tx_pkt_burst = virtual_ethdev_tx_burst_success;
diff --git a/drivers/net/bonding/rte_eth_bond_args.c b/drivers/net/bonding/rte_eth_bond_args.c
index 02ecde6..29436ea 100644
--- a/drivers/net/bonding/rte_eth_bond_args.c
+++ b/drivers/net/bonding/rte_eth_bond_args.c
@@ -54,15 +54,23 @@ const char *pmd_bond_init_valid_arguments[] = {
 static inline int
 find_port_id_by_pci_addr(const struct rte_pci_addr *pci_addr)
 {
+	struct rte_pci_device *pci_dev;
 	struct rte_pci_addr *eth_pci_addr;
 	unsigned i;
 
 	for (i = 0; i < rte_eth_dev_count(); i++) {
 
-		if (rte_eth_devices[i].pci_dev == NULL)
+		/* Currently populated by rte_eth_copy_pci_info().
+		 *
+		 * TODO: Once the PCI bus has arrived we should have a better
+		 * way to test for being a PCI device or not.
+		 */
+		if (rte_eth_devices[i].data->kdrv == RTE_KDRV_UNKNOWN ||
+		    rte_eth_devices[i].data->kdrv == RTE_KDRV_NONE)
 			continue;
 
-		eth_pci_addr = &(rte_eth_devices[i].pci_dev->addr);
+		pci_dev = ETH_DEV_PCI_DEV(&rte_eth_devices[i]);
+		eth_pci_addr = &pci_dev->addr;
 
 		if (pci_addr->bus == eth_pci_addr->bus &&
 			pci_addr->devid == eth_pci_addr->devid &&
diff --git a/drivers/net/cxgbe/cxgbe_main.c b/drivers/net/cxgbe/cxgbe_main.c
index 922155b..ea13c7e 100644
--- a/drivers/net/cxgbe/cxgbe_main.c
+++ b/drivers/net/cxgbe/cxgbe_main.c
@@ -1163,14 +1163,14 @@ int cxgbe_probe(struct adapter *adapter)
 		pi->eth_dev->data = data;
 
 allocate_mac:
-		pi->eth_dev->pci_dev = adapter->pdev;
+		pi->eth_dev->device = &adapter->pdev->device;
 		pi->eth_dev->data->dev_private = pi;
 		pi->eth_dev->driver = adapter->eth_dev->driver;
 		pi->eth_dev->dev_ops = adapter->eth_dev->dev_ops;
 		pi->eth_dev->tx_pkt_burst = adapter->eth_dev->tx_pkt_burst;
 		pi->eth_dev->rx_pkt_burst = adapter->eth_dev->rx_pkt_burst;
 
-		rte_eth_copy_pci_info(pi->eth_dev, pi->eth_dev->pci_dev);
+		rte_eth_copy_pci_info(pi->eth_dev, adapter->pdev);
 
 		TAILQ_INIT(&pi->eth_dev->link_intr_cbs);
 
diff --git a/drivers/net/fm10k/fm10k_ethdev.c b/drivers/net/fm10k/fm10k_ethdev.c
index 5d0bce5..5718b39 100644
--- a/drivers/net/fm10k/fm10k_ethdev.c
+++ b/drivers/net/fm10k/fm10k_ethdev.c
@@ -678,7 +678,7 @@ fm10k_dev_tx_init(struct rte_eth_dev *dev)
 		/* Enable use of FTAG bit in TX descriptor, PFVTCTL
 		 * register is read-only for VF.
 		 */
-		if (fm10k_check_ftag(dev->pci_dev->device.devargs)) {
+		if (fm10k_check_ftag(dev->device->devargs)) {
 			if (hw->mac.type == fm10k_mac_pf) {
 				FM10K_WRITE_REG(hw, FM10K_PFVTCTL(i),
 						FM10K_PFVTCTL_FTAG_DESC_ENABLE);
@@ -2740,7 +2740,7 @@ fm10k_set_tx_function(struct rte_eth_dev *dev)
 	int use_sse = 1;
 	uint16_t tx_ftag_en = 0;
 
-	if (fm10k_check_ftag(dev->pci_dev->device.devargs))
+	if (fm10k_check_ftag(dev->device->devargs))
 		tx_ftag_en = 1;
 
 	for (i = 0; i < dev->data->nb_tx_queues; i++) {
@@ -2771,7 +2771,7 @@ fm10k_set_rx_function(struct rte_eth_dev *dev)
 	uint16_t i, rx_using_sse;
 	uint16_t rx_ftag_en = 0;
 
-	if (fm10k_check_ftag(dev->pci_dev->device.devargs))
+	if (fm10k_check_ftag(dev->device->devargs))
 		rx_ftag_en = 1;
 
 	/* In order to allow Vector Rx there are a few configuration
diff --git a/drivers/net/virtio/virtio_user_ethdev.c b/drivers/net/virtio/virtio_user_ethdev.c
index 406beea..8cb983c 100644
--- a/drivers/net/virtio/virtio_user_ethdev.c
+++ b/drivers/net/virtio/virtio_user_ethdev.c
@@ -310,7 +310,6 @@ virtio_user_eth_dev_alloc(const char *name)
 	data->numa_node = SOCKET_ID_ANY;
 	data->kdrv = RTE_KDRV_NONE;
 	data->dev_flags = RTE_ETH_DEV_DETACHABLE;
-	eth_dev->pci_dev = NULL;
 	eth_dev->driver = NULL;
 	return eth_dev;
 }
diff --git a/lib/librte_eal/common/include/rte_pci.h b/lib/librte_eal/common/include/rte_pci.h
index 9ce8847..19deaf6 100644
--- a/lib/librte_eal/common/include/rte_pci.h
+++ b/lib/librte_eal/common/include/rte_pci.h
@@ -160,6 +160,12 @@ struct rte_pci_device {
 	enum rte_kernel_driver kdrv;            /**< Kernel driver passthrough */
 };
 
+/**
+ * @internal
+ * Helper macro for drivers that need to convert to struct rte_pci_device.
+ */
+#define DEV_PCI_DEV(ptr) container_of(ptr, struct rte_pci_device, device)
+
 /** Any PCI device identifier (vendor, device, ...) */
 #define PCI_ANY_ID (0xffff)
 #define RTE_CLASS_ANY_ID (0xffffff)
diff --git a/lib/librte_ether/rte_ethdev.c b/lib/librte_ether/rte_ethdev.c
index 4ecea50..cc01672 100644
--- a/lib/librte_ether/rte_ethdev.c
+++ b/lib/librte_ether/rte_ethdev.c
@@ -257,7 +257,7 @@ rte_eth_dev_pci_probe(struct rte_pci_driver *pci_drv,
 		if (eth_dev->data->dev_private == NULL)
 			rte_panic("Cannot allocate memzone for private port data\n");
 	}
-	eth_dev->pci_dev = pci_dev;
+	eth_dev->device = &pci_dev->device;
 	eth_dev->intr_handle = &pci_dev->intr_handle;
 	eth_dev->driver = eth_drv;
 	eth_dev->data->rx_mbuf_alloc_failed = 0;
@@ -318,7 +318,7 @@ rte_eth_dev_pci_remove(struct rte_pci_device *pci_dev)
 	if (rte_eal_process_type() == RTE_PROC_PRIMARY)
 		rte_free(eth_dev->data->dev_private);
 
-	eth_dev->pci_dev = NULL;
+	eth_dev->device = NULL;
 	eth_dev->driver = NULL;
 	eth_dev->data = NULL;
 
@@ -2578,7 +2578,7 @@ rte_eth_dma_zone_reserve(const struct rte_eth_dev *dev, const char *ring_name,
 	const struct rte_memzone *mz;
 
 	snprintf(z_name, sizeof(z_name), "%s_%s_%d_%d",
-		 dev->driver->pci_drv.driver.name, ring_name,
+		 dev->data->drv_name, ring_name,
 		 dev->data->port_id, queue_id);
 
 	mz = rte_memzone_lookup(z_name);
diff --git a/lib/librte_ether/rte_ethdev.h b/lib/librte_ether/rte_ethdev.h
index f1f656a..58ef798 100644
--- a/lib/librte_ether/rte_ethdev.h
+++ b/lib/librte_ether/rte_ethdev.h
@@ -1628,7 +1628,7 @@ struct rte_eth_dev {
 	struct rte_eth_dev_data *data;  /**< Pointer to device data */
 	const struct eth_driver *driver;/**< Driver for this device */
 	const struct eth_dev_ops *dev_ops; /**< Functions exported by PMD */
-	struct rte_pci_device *pci_dev; /**< PCI info. supplied by probing */
+	struct rte_device *device; /**< Backing device for this device */
 	struct rte_intr_handle *intr_handle; /**< Device interrupt handle */
 	/** User application callbacks for NIC interrupts */
 	struct rte_eth_dev_cb_list link_intr_cbs;
@@ -1649,7 +1649,7 @@ struct rte_eth_dev {
  * @internal
  * Helper for drivers that need to convert from rte_eth_dev to rte_pci_device.
  */
-#define ETH_DEV_PCI_DEV(ptr) ((ptr)->pci_dev)
+#define ETH_DEV_PCI_DEV(ptr) DEV_PCI_DEV((ptr)->device)
 
 struct rte_eth_dev_sriov {
 	uint8_t active;               /**< SRIOV is active with 16, 32 or 64 pools */
-- 
2.7.4

^ permalink raw reply related

* [PATCH v3 8/9] ethdev: Decouple interrupt handling from PCI device
From: Jan Blunck @ 2016-12-20 11:11 UTC (permalink / raw)
  To: dev; +Cc: shreyansh.jain, david.marchand, stephen
In-Reply-To: <1482232315-21626-1-git-send-email-jblunck@infradead.org>

The struct rte_intr_handle is an abstraction layer for different types of
interrupt mechanisms. It is embedded in the low-level device (e.g. PCI).
On allocation of a struct rte_eth_dev a reference to the intr_handle
should be stored for devices supporting interrupts.

Signed-off-by: Jan Blunck <jblunck@infradead.org>
---
 lib/librte_ether/rte_ethdev.c | 18 ++++++++++++++++--
 lib/librte_ether/rte_ethdev.h |  1 +
 2 files changed, 17 insertions(+), 2 deletions(-)

diff --git a/lib/librte_ether/rte_ethdev.c b/lib/librte_ether/rte_ethdev.c
index 4288577..4ecea50 100644
--- a/lib/librte_ether/rte_ethdev.c
+++ b/lib/librte_ether/rte_ethdev.c
@@ -258,6 +258,7 @@ rte_eth_dev_pci_probe(struct rte_pci_driver *pci_drv,
 			rte_panic("Cannot allocate memzone for private port data\n");
 	}
 	eth_dev->pci_dev = pci_dev;
+	eth_dev->intr_handle = &pci_dev->intr_handle;
 	eth_dev->driver = eth_drv;
 	eth_dev->data->rx_mbuf_alloc_failed = 0;
 
@@ -2543,7 +2544,13 @@ rte_eth_dev_rx_intr_ctl(uint8_t port_id, int epfd, int op, void *data)
 	RTE_ETH_VALID_PORTID_OR_ERR_RET(port_id, -ENODEV);
 
 	dev = &rte_eth_devices[port_id];
-	intr_handle = &dev->pci_dev->intr_handle;
+
+	if (!dev->intr_handle) {
+		RTE_PMD_DEBUG_TRACE("RX Intr handle unset\n");
+		return -ENOTSUP;
+	}
+
+	intr_handle = dev->intr_handle;
 	if (!intr_handle->intr_vec) {
 		RTE_PMD_DEBUG_TRACE("RX Intr vector unset\n");
 		return -EPERM;
@@ -2603,7 +2610,12 @@ rte_eth_dev_rx_intr_ctl_q(uint8_t port_id, uint16_t queue_id,
 		return -EINVAL;
 	}
 
-	intr_handle = &dev->pci_dev->intr_handle;
+	if (!dev->intr_handle) {
+		RTE_PMD_DEBUG_TRACE("RX Intr handle unset\n");
+		return -ENOTSUP;
+	}
+
+	intr_handle = dev->intr_handle;
 	if (!intr_handle->intr_vec) {
 		RTE_PMD_DEBUG_TRACE("RX Intr vector unset\n");
 		return -EPERM;
@@ -3205,6 +3217,8 @@ rte_eth_copy_pci_info(struct rte_eth_dev *eth_dev, struct rte_pci_device *pci_de
 		return;
 	}
 
+	eth_dev->intr_handle = &pci_dev->intr_handle;
+
 	eth_dev->data->dev_flags = 0;
 	if (pci_dev->driver->drv_flags & RTE_PCI_DRV_INTR_LSC)
 		eth_dev->data->dev_flags |= RTE_ETH_DEV_INTR_LSC;
diff --git a/lib/librte_ether/rte_ethdev.h b/lib/librte_ether/rte_ethdev.h
index 3adbb2b..f1f656a 100644
--- a/lib/librte_ether/rte_ethdev.h
+++ b/lib/librte_ether/rte_ethdev.h
@@ -1629,6 +1629,7 @@ struct rte_eth_dev {
 	const struct eth_driver *driver;/**< Driver for this device */
 	const struct eth_dev_ops *dev_ops; /**< Functions exported by PMD */
 	struct rte_pci_device *pci_dev; /**< PCI info. supplied by probing */
+	struct rte_intr_handle *intr_handle; /**< Device interrupt handle */
 	/** User application callbacks for NIC interrupts */
 	struct rte_eth_dev_cb_list link_intr_cbs;
 	/**
-- 
2.7.4

^ permalink raw reply related

* [PATCH v3 7/9] ethdev: Move filling of rte_eth_dev_info->pci_dev to dev_infos_get()
From: Jan Blunck @ 2016-12-20 11:11 UTC (permalink / raw)
  To: dev; +Cc: shreyansh.jain, david.marchand, stephen
In-Reply-To: <1482232315-21626-1-git-send-email-jblunck@infradead.org>

Only the device itself can decide its PCI or not.

Signed-off-by: Jan Blunck <jblunck@infradead.org>
---
 drivers/net/bnx2x/bnx2x_ethdev.c        | 1 +
 drivers/net/bnxt/bnxt_ethdev.c          | 2 ++
 drivers/net/cxgbe/cxgbe_ethdev.c        | 2 ++
 drivers/net/e1000/em_ethdev.c           | 1 +
 drivers/net/e1000/igb_ethdev.c          | 2 ++
 drivers/net/ena/ena_ethdev.c            | 2 ++
 drivers/net/enic/enic_ethdev.c          | 1 +
 drivers/net/fm10k/fm10k_ethdev.c        | 1 +
 drivers/net/i40e/i40e_ethdev.c          | 1 +
 drivers/net/i40e/i40e_ethdev_vf.c       | 1 +
 drivers/net/ixgbe/ixgbe_ethdev.c        | 2 ++
 drivers/net/mlx4/mlx4.c                 | 2 ++
 drivers/net/mlx5/mlx5_ethdev.c          | 2 ++
 drivers/net/nfp/nfp_net.c               | 1 +
 drivers/net/qede/qede_ethdev.c          | 1 +
 drivers/net/szedata2/rte_eth_szedata2.c | 1 +
 drivers/net/thunderx/nicvf_ethdev.c     | 2 ++
 drivers/net/virtio/virtio_ethdev.c      | 1 +
 drivers/net/vmxnet3/vmxnet3_ethdev.c    | 2 ++
 lib/librte_ether/rte_ethdev.c           | 1 -
 20 files changed, 28 insertions(+), 1 deletion(-)

diff --git a/drivers/net/bnx2x/bnx2x_ethdev.c b/drivers/net/bnx2x/bnx2x_ethdev.c
index c331ccd..551dbad 100644
--- a/drivers/net/bnx2x/bnx2x_ethdev.c
+++ b/drivers/net/bnx2x/bnx2x_ethdev.c
@@ -431,6 +431,7 @@ static void
 bnx2x_dev_infos_get(struct rte_eth_dev *dev, __rte_unused struct rte_eth_dev_info *dev_info)
 {
 	struct bnx2x_softc *sc = dev->data->dev_private;
+	dev_info->pci_dev = ETH_DEV_PCI_DEV(dev);
 	dev_info->max_rx_queues  = sc->max_rx_queues;
 	dev_info->max_tx_queues  = sc->max_tx_queues;
 	dev_info->min_rx_bufsize = BNX2X_MIN_RX_BUF_SIZE;
diff --git a/drivers/net/bnxt/bnxt_ethdev.c b/drivers/net/bnxt/bnxt_ethdev.c
index cd50f11..bf39fbe 100644
--- a/drivers/net/bnxt/bnxt_ethdev.c
+++ b/drivers/net/bnxt/bnxt_ethdev.c
@@ -303,6 +303,8 @@ static void bnxt_dev_info_get_op(struct rte_eth_dev *eth_dev,
 	struct bnxt *bp = (struct bnxt *)eth_dev->data->dev_private;
 	uint16_t max_vnics, i, j, vpool, vrxq;
 
+	dev_info->pci_dev = ETH_DEV_PCI_DEV(eth_dev);
+
 	/* MAC Specifics */
 	dev_info->max_mac_addrs = MAX_NUM_MAC_ADDR;
 	dev_info->max_hash_mac_addrs = 0;
diff --git a/drivers/net/cxgbe/cxgbe_ethdev.c b/drivers/net/cxgbe/cxgbe_ethdev.c
index 8bfdda8..8938b08 100644
--- a/drivers/net/cxgbe/cxgbe_ethdev.c
+++ b/drivers/net/cxgbe/cxgbe_ethdev.c
@@ -147,6 +147,8 @@ static void cxgbe_dev_info_get(struct rte_eth_dev *eth_dev,
 		.nb_align = 1,
 	};
 
+	device_info->pci_dev = ETH_DEV_PCI_DEV(eth_dev);
+
 	device_info->min_rx_bufsize = CXGBE_MIN_RX_BUFSIZE;
 	device_info->max_rx_pktlen = CXGBE_MAX_RX_PKTLEN;
 	device_info->max_rx_queues = max_queues;
diff --git a/drivers/net/e1000/em_ethdev.c b/drivers/net/e1000/em_ethdev.c
index 7f2f521..3d34e5b 100644
--- a/drivers/net/e1000/em_ethdev.c
+++ b/drivers/net/e1000/em_ethdev.c
@@ -1048,6 +1048,7 @@ eth_em_infos_get(struct rte_eth_dev *dev, struct rte_eth_dev_info *dev_info)
 {
 	struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
 
+	dev_info->pci_dev = ETH_DEV_PCI_DEV(dev);
 	dev_info->min_rx_bufsize = 256; /* See BSIZE field of RCTL register. */
 	dev_info->max_rx_pktlen = em_get_max_pktlen(hw);
 	dev_info->max_mac_addrs = hw->mac.rar_entry_count;
diff --git a/drivers/net/e1000/igb_ethdev.c b/drivers/net/e1000/igb_ethdev.c
index b25c66e..7d77561 100644
--- a/drivers/net/e1000/igb_ethdev.c
+++ b/drivers/net/e1000/igb_ethdev.c
@@ -1983,6 +1983,7 @@ eth_igb_infos_get(struct rte_eth_dev *dev, struct rte_eth_dev_info *dev_info)
 {
 	struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
 
+	dev_info->pci_dev = ETH_DEV_PCI_DEV(dev);
 	dev_info->min_rx_bufsize = 256; /* See BSIZE field of RCTL register. */
 	dev_info->max_rx_pktlen  = 0x3FFF; /* See RLPML register. */
 	dev_info->max_mac_addrs = hw->mac.rar_entry_count;
@@ -2111,6 +2112,7 @@ eth_igbvf_infos_get(struct rte_eth_dev *dev, struct rte_eth_dev_info *dev_info)
 {
 	struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
 
+	dev_info->pci_dev = ETH_DEV_PCI_DEV(dev);
 	dev_info->min_rx_bufsize = 256; /* See BSIZE field of RCTL register. */
 	dev_info->max_rx_pktlen  = 0x3FFF; /* See RLPML register. */
 	dev_info->max_mac_addrs = hw->mac.rar_entry_count;
diff --git a/drivers/net/ena/ena_ethdev.c b/drivers/net/ena/ena_ethdev.c
index c17d969..051275e 100644
--- a/drivers/net/ena/ena_ethdev.c
+++ b/drivers/net/ena/ena_ethdev.c
@@ -1436,6 +1436,8 @@ static void ena_infos_get(struct rte_eth_dev *dev,
 	ena_dev = &adapter->ena_dev;
 	ena_assert_msg(ena_dev != NULL, "Uninitialized device");
 
+	dev_info->pci_dev = ETH_DEV_PCI_DEV(dev);
+
 	dev_info->speed_capa =
 			ETH_LINK_SPEED_1G   |
 			ETH_LINK_SPEED_2_5G |
diff --git a/drivers/net/enic/enic_ethdev.c b/drivers/net/enic/enic_ethdev.c
index 553a88e..bb5dfe6 100644
--- a/drivers/net/enic/enic_ethdev.c
+++ b/drivers/net/enic/enic_ethdev.c
@@ -459,6 +459,7 @@ static void enicpmd_dev_info_get(struct rte_eth_dev *eth_dev,
 	struct enic *enic = pmd_priv(eth_dev);
 
 	ENICPMD_FUNC_TRACE();
+	device_info->pci_dev = ETH_DEV_PCI_DEV(eth_dev);
 	/* Scattered Rx uses two receive queues per rx queue exposed to dpdk */
 	device_info->max_rx_queues = enic->conf_rq_count / 2;
 	device_info->max_tx_queues = enic->conf_wq_count;
diff --git a/drivers/net/fm10k/fm10k_ethdev.c b/drivers/net/fm10k/fm10k_ethdev.c
index e1250f6..5d0bce5 100644
--- a/drivers/net/fm10k/fm10k_ethdev.c
+++ b/drivers/net/fm10k/fm10k_ethdev.c
@@ -1392,6 +1392,7 @@ fm10k_dev_infos_get(struct rte_eth_dev *dev,
 
 	PMD_INIT_FUNC_TRACE();
 
+	dev_info->pci_dev            = ETH_DEV_PCI_DEV(dev);
 	dev_info->min_rx_bufsize     = FM10K_MIN_RX_BUF_SIZE;
 	dev_info->max_rx_pktlen      = FM10K_MAX_PKT_SIZE;
 	dev_info->max_rx_queues      = hw->mac.max_queues;
diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40e_ethdev.c
index 8a63a8c..29c0277 100644
--- a/drivers/net/i40e/i40e_ethdev.c
+++ b/drivers/net/i40e/i40e_ethdev.c
@@ -2583,6 +2583,7 @@ i40e_dev_info_get(struct rte_eth_dev *dev, struct rte_eth_dev_info *dev_info)
 	struct i40e_hw *hw = I40E_DEV_PRIVATE_TO_HW(dev->data->dev_private);
 	struct i40e_vsi *vsi = pf->main_vsi;
 
+	dev_info->pci_dev = ETH_DEV_PCI_DEV(dev);
 	dev_info->max_rx_queues = vsi->nb_qps;
 	dev_info->max_tx_queues = vsi->nb_qps;
 	dev_info->min_rx_bufsize = I40E_BUF_SIZE_MIN;
diff --git a/drivers/net/i40e/i40e_ethdev_vf.c b/drivers/net/i40e/i40e_ethdev_vf.c
index 781e658..caef72c 100644
--- a/drivers/net/i40e/i40e_ethdev_vf.c
+++ b/drivers/net/i40e/i40e_ethdev_vf.c
@@ -2217,6 +2217,7 @@ i40evf_dev_info_get(struct rte_eth_dev *dev, struct rte_eth_dev_info *dev_info)
 	struct i40e_vf *vf = I40EVF_DEV_PRIVATE_TO_VF(dev->data->dev_private);
 
 	memset(dev_info, 0, sizeof(*dev_info));
+	dev_info->pci_dev = ETH_DEV_PCI_DEV(dev);
 	dev_info->max_rx_queues = vf->vsi_res->num_queue_pairs;
 	dev_info->max_tx_queues = vf->vsi_res->num_queue_pairs;
 	dev_info->min_rx_bufsize = I40E_BUF_SIZE_MIN;
diff --git a/drivers/net/ixgbe/ixgbe_ethdev.c b/drivers/net/ixgbe/ixgbe_ethdev.c
index f17da46..c88b7bf 100644
--- a/drivers/net/ixgbe/ixgbe_ethdev.c
+++ b/drivers/net/ixgbe/ixgbe_ethdev.c
@@ -3043,6 +3043,7 @@ ixgbe_dev_info_get(struct rte_eth_dev *dev, struct rte_eth_dev_info *dev_info)
 	struct rte_eth_conf *dev_conf = &dev->data->dev_conf;
 	struct rte_pci_device *pci_dev = ETH_DEV_PCI_DEV(dev);
 
+	dev_info->pci_dev = ETH_DEV_PCI_DEV(dev);
 	dev_info->max_rx_queues = (uint16_t)hw->mac.max_rx_queues;
 	dev_info->max_tx_queues = (uint16_t)hw->mac.max_tx_queues;
 	if (RTE_ETH_DEV_SRIOV(dev).active == 0) {
@@ -3175,6 +3176,7 @@ ixgbevf_dev_info_get(struct rte_eth_dev *dev,
 	struct ixgbe_hw *hw = IXGBE_DEV_PRIVATE_TO_HW(dev->data->dev_private);
 	struct rte_pci_device *pci_dev = ETH_DEV_PCI_DEV(dev);
 
+	dev_info->pci_dev = ETH_DEV_PCI_DEV(dev);
 	dev_info->max_rx_queues = (uint16_t)hw->mac.max_rx_queues;
 	dev_info->max_tx_queues = (uint16_t)hw->mac.max_tx_queues;
 	dev_info->min_rx_bufsize = 1024; /* cf BSIZEPACKET in SRRCTL reg */
diff --git a/drivers/net/mlx4/mlx4.c b/drivers/net/mlx4/mlx4.c
index da61a85..5455fea 100644
--- a/drivers/net/mlx4/mlx4.c
+++ b/drivers/net/mlx4/mlx4.c
@@ -4421,6 +4421,8 @@ mlx4_dev_infos_get(struct rte_eth_dev *dev, struct rte_eth_dev_info *info)
 	unsigned int max;
 	char ifname[IF_NAMESIZE];
 
+	info->pci_dev = ETH_DEV_PCI_DEV(dev);
+
 	if (priv == NULL)
 		return;
 	priv_lock(priv);
diff --git a/drivers/net/mlx5/mlx5_ethdev.c b/drivers/net/mlx5/mlx5_ethdev.c
index c0f73e9..489a4f6 100644
--- a/drivers/net/mlx5/mlx5_ethdev.c
+++ b/drivers/net/mlx5/mlx5_ethdev.c
@@ -562,6 +562,8 @@ mlx5_dev_infos_get(struct rte_eth_dev *dev, struct rte_eth_dev_info *info)
 	unsigned int max;
 	char ifname[IF_NAMESIZE];
 
+	info->pci_dev = ETH_DEV_PCI_DEV(dev);
+
 	priv_lock(priv);
 	/* FIXME: we should ask the device for these values. */
 	info->min_rx_bufsize = 32;
diff --git a/drivers/net/nfp/nfp_net.c b/drivers/net/nfp/nfp_net.c
index 7299fad..c49d38a 100644
--- a/drivers/net/nfp/nfp_net.c
+++ b/drivers/net/nfp/nfp_net.c
@@ -1008,6 +1008,7 @@ nfp_net_infos_get(struct rte_eth_dev *dev, struct rte_eth_dev_info *dev_info)
 
 	hw = NFP_NET_DEV_PRIVATE_TO_HW(dev->data->dev_private);
 
+	dev_info->pci_dev = ETH_DEV_PCI_DEV(dev);
 	dev_info->driver_name = dev->driver->pci_drv.driver.name;
 	dev_info->max_rx_queues = (uint16_t)hw->max_rx_queues;
 	dev_info->max_tx_queues = (uint16_t)hw->max_tx_queues;
diff --git a/drivers/net/qede/qede_ethdev.c b/drivers/net/qede/qede_ethdev.c
index 959ff0f..2c4de92 100644
--- a/drivers/net/qede/qede_ethdev.c
+++ b/drivers/net/qede/qede_ethdev.c
@@ -652,6 +652,7 @@ qede_dev_info_get(struct rte_eth_dev *eth_dev,
 
 	PMD_INIT_FUNC_TRACE(edev);
 
+	dev_info->pci_dev = ETH_DEV_PCI_DEV(eth_dev);
 	dev_info->min_rx_bufsize = (uint32_t)(ETHER_MIN_MTU +
 					      QEDE_ETH_OVERHEAD);
 	dev_info->max_rx_pktlen = (uint32_t)ETH_TX_MAX_NON_LSO_PKT_LEN;
diff --git a/drivers/net/szedata2/rte_eth_szedata2.c b/drivers/net/szedata2/rte_eth_szedata2.c
index 6d80a5b..6356257 100644
--- a/drivers/net/szedata2/rte_eth_szedata2.c
+++ b/drivers/net/szedata2/rte_eth_szedata2.c
@@ -1030,6 +1030,7 @@ eth_dev_info(struct rte_eth_dev *dev,
 		struct rte_eth_dev_info *dev_info)
 {
 	struct pmd_internals *internals = dev->data->dev_private;
+	dev_info->pci_dev = ETH_DEV_PCI_DEV(dev);
 	dev_info->if_index = 0;
 	dev_info->max_mac_addrs = 1;
 	dev_info->max_rx_pktlen = (uint32_t)-1;
diff --git a/drivers/net/thunderx/nicvf_ethdev.c b/drivers/net/thunderx/nicvf_ethdev.c
index 099b252..daac853 100644
--- a/drivers/net/thunderx/nicvf_ethdev.c
+++ b/drivers/net/thunderx/nicvf_ethdev.c
@@ -1339,6 +1339,8 @@ nicvf_dev_info_get(struct rte_eth_dev *dev, struct rte_eth_dev_info *dev_info)
 
 	PMD_INIT_FUNC_TRACE();
 
+	dev_info->pci_dev = pci_dev;
+
 	dev_info->min_rx_bufsize = ETHER_MIN_MTU;
 	dev_info->max_rx_pktlen = NIC_HW_MAX_FRS;
 	dev_info->max_rx_queues =
diff --git a/drivers/net/virtio/virtio_ethdev.c b/drivers/net/virtio/virtio_ethdev.c
index 023101d..2eb6a06 100644
--- a/drivers/net/virtio/virtio_ethdev.c
+++ b/drivers/net/virtio/virtio_ethdev.c
@@ -1625,6 +1625,7 @@ virtio_dev_info_get(struct rte_eth_dev *dev, struct rte_eth_dev_info *dev_info)
 	uint64_t tso_mask;
 	struct virtio_hw *hw = dev->data->dev_private;
 
+	dev_info->pci_dev = hw->dev;
 	dev_info->max_rx_queues =
 		RTE_MIN(hw->max_queue_pairs, VIRTIO_MAX_RX_QUEUES);
 	dev_info->max_tx_queues =
diff --git a/drivers/net/vmxnet3/vmxnet3_ethdev.c b/drivers/net/vmxnet3/vmxnet3_ethdev.c
index bcb3751..f7c0382 100644
--- a/drivers/net/vmxnet3/vmxnet3_ethdev.c
+++ b/drivers/net/vmxnet3/vmxnet3_ethdev.c
@@ -709,6 +709,8 @@ static void
 vmxnet3_dev_info_get(__rte_unused struct rte_eth_dev *dev,
 		     struct rte_eth_dev_info *dev_info)
 {
+	dev_info->pci_dev = ETH_DEV_PCI_DEV(dev);
+
 	dev_info->max_rx_queues = VMXNET3_MAX_RX_QUEUES;
 	dev_info->max_tx_queues = VMXNET3_MAX_TX_QUEUES;
 	dev_info->min_rx_bufsize = 1518 + RTE_PKTMBUF_HEADROOM;
diff --git a/lib/librte_ether/rte_ethdev.c b/lib/librte_ether/rte_ethdev.c
index fde8112..4288577 100644
--- a/lib/librte_ether/rte_ethdev.c
+++ b/lib/librte_ether/rte_ethdev.c
@@ -1556,7 +1556,6 @@ rte_eth_dev_info_get(uint8_t port_id, struct rte_eth_dev_info *dev_info)
 
 	RTE_FUNC_PTR_OR_RET(*dev->dev_ops->dev_infos_get);
 	(*dev->dev_ops->dev_infos_get)(dev, dev_info);
-	dev_info->pci_dev = dev->pci_dev;
 	dev_info->driver_name = dev->data->drv_name;
 	dev_info->nb_rx_queues = dev->data->nb_rx_queues;
 	dev_info->nb_tx_queues = dev->data->nb_tx_queues;
-- 
2.7.4

^ permalink raw reply related

* [PATCH v3 6/9] virtio: Don't depend on struct rte_eth_dev's pci_dev
From: Jan Blunck @ 2016-12-20 11:11 UTC (permalink / raw)
  To: dev; +Cc: shreyansh.jain, david.marchand, stephen
In-Reply-To: <1482232315-21626-1-git-send-email-jblunck@infradead.org>

We don't need to depend on rte_eth_dev->pci_dev to differentiate between
the virtio_user and the virtio_pci case. Instead we can use the private
virtio_hw struct to get that information.

Signed-off-by: Jan Blunck <jblunck@infradead.org>
---
 drivers/net/virtio/virtio_ethdev.c | 21 ++++++++++-----------
 1 file changed, 10 insertions(+), 11 deletions(-)

diff --git a/drivers/net/virtio/virtio_ethdev.c b/drivers/net/virtio/virtio_ethdev.c
index da9668e..023101d 100644
--- a/drivers/net/virtio/virtio_ethdev.c
+++ b/drivers/net/virtio/virtio_ethdev.c
@@ -483,11 +483,11 @@ virtio_init_queue(struct rte_eth_dev *dev, uint16_t vtpci_queue_idx)
 		hw->cvq = cvq;
 	}
 
-	/* For virtio_user case (that is when dev->pci_dev is NULL), we use
+	/* For virtio_user case (that is when hw->dev is NULL), we use
 	 * virtual address. And we need properly set _offset_, please see
 	 * VIRTIO_MBUF_DATA_DMA_ADDR in virtqueue.h for more information.
 	 */
-	if (dev->pci_dev)
+	if (hw->dev)
 		vq->offset = offsetof(struct rte_mbuf, buf_physaddr);
 	else {
 		vq->vq_ring_mem = (uintptr_t)mz->addr;
@@ -1190,7 +1190,7 @@ virtio_init_device(struct rte_eth_dev *eth_dev, uint64_t req_features)
 	struct virtio_hw *hw = eth_dev->data->dev_private;
 	struct virtio_net_config *config;
 	struct virtio_net_config local_config;
-	struct rte_pci_device *pci_dev = eth_dev->pci_dev;
+	struct rte_pci_device *pci_dev = hw->dev;
 	int ret;
 
 	/* Reset the device although not necessary at startup */
@@ -1294,7 +1294,6 @@ int
 eth_virtio_dev_init(struct rte_eth_dev *eth_dev)
 {
 	struct virtio_hw *hw = eth_dev->data->dev_private;
-	struct rte_pci_device *pci_dev;
 	uint32_t dev_flags = RTE_ETH_DEV_DETACHABLE;
 	int ret;
 
@@ -1317,10 +1316,11 @@ eth_virtio_dev_init(struct rte_eth_dev *eth_dev)
 		return -ENOMEM;
 	}
 
-	pci_dev = eth_dev->pci_dev;
-
-	if (pci_dev) {
-		ret = vtpci_init(pci_dev, hw, &dev_flags);
+	/* For virtio_user case the hw->virtio_user_dev is populated by
+	 * virtio_user_eth_dev_alloc() before eth_virtio_dev_init() is called.
+	 */
+	if (!hw->virtio_user_dev) {
+		ret = vtpci_init(ETH_DEV_PCI_DEV(eth_dev), hw, &dev_flags);
 		if (ret)
 			return ret;
 	}
@@ -1343,7 +1343,6 @@ eth_virtio_dev_init(struct rte_eth_dev *eth_dev)
 static int
 eth_virtio_dev_uninit(struct rte_eth_dev *eth_dev)
 {
-	struct rte_pci_device *pci_dev;
 	struct virtio_hw *hw = eth_dev->data->dev_private;
 
 	PMD_INIT_FUNC_TRACE();
@@ -1353,7 +1352,6 @@ eth_virtio_dev_uninit(struct rte_eth_dev *eth_dev)
 
 	virtio_dev_stop(eth_dev);
 	virtio_dev_close(eth_dev);
-	pci_dev = eth_dev->pci_dev;
 
 	eth_dev->dev_ops = NULL;
 	eth_dev->tx_pkt_burst = NULL;
@@ -1367,7 +1365,8 @@ eth_virtio_dev_uninit(struct rte_eth_dev *eth_dev)
 		rte_intr_callback_unregister(vtpci_intr_handle(hw),
 						virtio_interrupt_handler,
 						eth_dev);
-	rte_eal_pci_unmap_device(pci_dev);
+	if (hw->dev)
+		rte_eal_pci_unmap_device(hw->dev);
 
 	PMD_INIT_LOG(DEBUG, "dev_uninit completed");
 
-- 
2.7.4

^ permalink raw reply related

* [PATCH v3 5/9] virtio: Add vtpci_intr_handle() helper to get rte_intr_handle
From: Jan Blunck @ 2016-12-20 11:11 UTC (permalink / raw)
  To: dev; +Cc: shreyansh.jain, david.marchand, stephen
In-Reply-To: <1482232315-21626-1-git-send-email-jblunck@infradead.org>

This adds a helper to get the rte_intr_handle from the virtio_hw. This is
safe to do since the usage of the helper is guarded by RTE_ETH_DEV_INTR_LSC
which is only set if we found a PCI device during initialization.

Signed-off-by: Jan Blunck <jblunck@infradead.org>
---
 drivers/net/virtio/virtio_ethdev.c | 12 +++++++-----
 drivers/net/virtio/virtio_pci.h    |  6 ++++++
 2 files changed, 13 insertions(+), 5 deletions(-)

diff --git a/drivers/net/virtio/virtio_ethdev.c b/drivers/net/virtio/virtio_ethdev.c
index 741688e..da9668e 100644
--- a/drivers/net/virtio/virtio_ethdev.c
+++ b/drivers/net/virtio/virtio_ethdev.c
@@ -1162,7 +1162,7 @@ virtio_interrupt_handler(__rte_unused struct rte_intr_handle *handle,
 	isr = vtpci_isr(hw);
 	PMD_DRV_LOG(INFO, "interrupt status = %#x", isr);
 
-	if (rte_intr_enable(&dev->pci_dev->intr_handle) < 0)
+	if (rte_intr_enable(vtpci_intr_handle(hw)) < 0)
 		PMD_DRV_LOG(ERR, "interrupt enable failed");
 
 	if (isr & VIRTIO_PCI_ISR_CONFIG) {
@@ -1334,7 +1334,7 @@ eth_virtio_dev_init(struct rte_eth_dev *eth_dev)
 
 	/* Setup interrupt callback  */
 	if (eth_dev->data->dev_flags & RTE_ETH_DEV_INTR_LSC)
-		rte_intr_callback_register(&pci_dev->intr_handle,
+		rte_intr_callback_register(vtpci_intr_handle(hw),
 			virtio_interrupt_handler, eth_dev);
 
 	return 0;
@@ -1344,6 +1344,7 @@ static int
 eth_virtio_dev_uninit(struct rte_eth_dev *eth_dev)
 {
 	struct rte_pci_device *pci_dev;
+	struct virtio_hw *hw = eth_dev->data->dev_private;
 
 	PMD_INIT_FUNC_TRACE();
 
@@ -1363,7 +1364,7 @@ eth_virtio_dev_uninit(struct rte_eth_dev *eth_dev)
 
 	/* reset interrupt callback  */
 	if (eth_dev->data->dev_flags & RTE_ETH_DEV_INTR_LSC)
-		rte_intr_callback_unregister(&pci_dev->intr_handle,
+		rte_intr_callback_unregister(vtpci_intr_handle(hw),
 						virtio_interrupt_handler,
 						eth_dev);
 	rte_eal_pci_unmap_device(pci_dev);
@@ -1481,7 +1482,7 @@ virtio_dev_start(struct rte_eth_dev *dev)
 			return -ENOTSUP;
 		}
 
-		if (rte_intr_enable(&dev->pci_dev->intr_handle) < 0) {
+		if (rte_intr_enable(vtpci_intr_handle(hw)) < 0) {
 			PMD_DRV_LOG(ERR, "interrupt enable failed");
 			return -EIO;
 		}
@@ -1573,12 +1574,13 @@ static void virtio_dev_free_mbufs(struct rte_eth_dev *dev)
 static void
 virtio_dev_stop(struct rte_eth_dev *dev)
 {
+	struct virtio_hw *hw = dev->data->dev_private;
 	struct rte_eth_link link;
 
 	PMD_INIT_LOG(DEBUG, "stop");
 
 	if (dev->data->dev_conf.intr_conf.lsc)
-		rte_intr_disable(&dev->pci_dev->intr_handle);
+		rte_intr_disable(vtpci_intr_handle(hw));
 
 	memset(&link, 0, sizeof(link));
 	virtio_dev_atomic_write_link_status(dev, &link);
diff --git a/drivers/net/virtio/virtio_pci.h b/drivers/net/virtio/virtio_pci.h
index de271bf..5373e39 100644
--- a/drivers/net/virtio/virtio_pci.h
+++ b/drivers/net/virtio/virtio_pci.h
@@ -317,4 +317,10 @@ uint8_t vtpci_isr(struct virtio_hw *);
 
 uint16_t vtpci_irq_config(struct virtio_hw *, uint16_t);
 
+static inline struct rte_intr_handle *
+vtpci_intr_handle(struct virtio_hw *hw)
+{
+	return hw->dev ? &hw->dev->intr_handle : NULL;
+}
+
 #endif /* _VIRTIO_PCI_H_ */
-- 
2.7.4

^ permalink raw reply related

* [PATCH v3 4/9] virtio: Don't fill dev_info->driver_name
From: Jan Blunck @ 2016-12-20 11:11 UTC (permalink / raw)
  To: dev; +Cc: shreyansh.jain, david.marchand, stephen
In-Reply-To: <1482232315-21626-1-git-send-email-jblunck@infradead.org>

This is overwritten in rte_eth_dev_info_get().

Signed-off-by: Jan Blunck <jblunck@infradead.org>
Reviewed-by: David Marchand <david.marchand@6wind.com>
---
 drivers/net/virtio/virtio_ethdev.c | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/drivers/net/virtio/virtio_ethdev.c b/drivers/net/virtio/virtio_ethdev.c
index 079fd6c..741688e 100644
--- a/drivers/net/virtio/virtio_ethdev.c
+++ b/drivers/net/virtio/virtio_ethdev.c
@@ -1624,10 +1624,6 @@ virtio_dev_info_get(struct rte_eth_dev *dev, struct rte_eth_dev_info *dev_info)
 	uint64_t tso_mask;
 	struct virtio_hw *hw = dev->data->dev_private;
 
-	if (dev->pci_dev)
-		dev_info->driver_name = dev->driver->pci_drv.driver.name;
-	else
-		dev_info->driver_name = "virtio_user PMD";
 	dev_info->max_rx_queues =
 		RTE_MIN(hw->max_queue_pairs, VIRTIO_MAX_RX_QUEUES);
 	dev_info->max_tx_queues =
-- 
2.7.4

^ permalink raw reply related

* [PATCH v3 3/9] drivers: Use ETH_DEV_PCI_DEV() helper
From: Jan Blunck @ 2016-12-20 11:11 UTC (permalink / raw)
  To: dev; +Cc: shreyansh.jain, david.marchand, stephen
In-Reply-To: <1482232315-21626-1-git-send-email-jblunck@infradead.org>

The drivers should not directly access the rte_eth_dev->pci_dev but use
a macro instead. This is a preparation for replacing the pci_dev with
a struct rte_device member in the future.

Signed-off-by: Jan Blunck <jblunck@infradead.org>
---
 drivers/net/bnx2x/bnx2x_ethdev.c        | 12 +++---
 drivers/net/bnxt/bnxt_ethdev.c          | 19 ++++++----
 drivers/net/bnxt/bnxt_ring.c            | 11 +++---
 drivers/net/cxgbe/cxgbe_ethdev.c        |  2 +-
 drivers/net/e1000/em_ethdev.c           | 20 ++++++----
 drivers/net/e1000/igb_ethdev.c          | 50 ++++++++++++++-----------
 drivers/net/e1000/igb_pf.c              |  3 +-
 drivers/net/ena/ena_ethdev.c            |  2 +-
 drivers/net/enic/enic_ethdev.c          |  2 +-
 drivers/net/fm10k/fm10k_ethdev.c        | 49 +++++++++++++------------
 drivers/net/i40e/i40e_ethdev.c          | 44 +++++++++++-----------
 drivers/net/i40e/i40e_ethdev.h          |  4 ++
 drivers/net/i40e/i40e_ethdev_vf.c       | 38 +++++++++----------
 drivers/net/ixgbe/ixgbe_ethdev.c        | 65 ++++++++++++++++++++-------------
 drivers/net/ixgbe/ixgbe_pf.c            |  2 +-
 drivers/net/nfp/nfp_net.c               | 16 ++++----
 drivers/net/qede/qede_ethdev.c          | 17 +++++----
 drivers/net/szedata2/rte_eth_szedata2.c | 27 +++++++-------
 drivers/net/szedata2/rte_eth_szedata2.h | 34 ++++++++---------
 drivers/net/thunderx/nicvf_ethdev.c     |  5 ++-
 drivers/net/vmxnet3/vmxnet3_ethdev.c    |  4 +-
 21 files changed, 234 insertions(+), 192 deletions(-)

diff --git a/drivers/net/bnx2x/bnx2x_ethdev.c b/drivers/net/bnx2x/bnx2x_ethdev.c
index 0eae433..c331ccd 100644
--- a/drivers/net/bnx2x/bnx2x_ethdev.c
+++ b/drivers/net/bnx2x/bnx2x_ethdev.c
@@ -124,7 +124,7 @@ bnx2x_interrupt_handler(__rte_unused struct rte_intr_handle *handle, void *param
 	struct rte_eth_dev *dev = (struct rte_eth_dev *)param;
 
 	bnx2x_interrupt_action(dev);
-	rte_intr_enable(&(dev->pci_dev->intr_handle));
+	rte_intr_enable(&(ETH_DEV_PCI_DEV(dev)->intr_handle));
 }
 
 /*
@@ -187,10 +187,10 @@ bnx2x_dev_start(struct rte_eth_dev *dev)
 	}
 
 	if (IS_PF(sc)) {
-		rte_intr_callback_register(&(dev->pci_dev->intr_handle),
+		rte_intr_callback_register(&sc->pci_dev->intr_handle,
 				bnx2x_interrupt_handler, (void *)dev);
 
-		if(rte_intr_enable(&(dev->pci_dev->intr_handle)))
+		if (rte_intr_enable(&sc->pci_dev->intr_handle))
 			PMD_DRV_LOG(ERR, "rte_intr_enable failed");
 	}
 
@@ -215,8 +215,8 @@ bnx2x_dev_stop(struct rte_eth_dev *dev)
 	PMD_INIT_FUNC_TRACE();
 
 	if (IS_PF(sc)) {
-		rte_intr_disable(&(dev->pci_dev->intr_handle));
-		rte_intr_callback_unregister(&(dev->pci_dev->intr_handle),
+		rte_intr_disable(&sc->pci_dev->intr_handle);
+		rte_intr_callback_unregister(&sc->pci_dev->intr_handle,
 				bnx2x_interrupt_handler, (void *)dev);
 	}
 
@@ -516,7 +516,7 @@ bnx2x_common_dev_init(struct rte_eth_dev *eth_dev, int is_vf)
 	PMD_INIT_FUNC_TRACE();
 
 	eth_dev->dev_ops = is_vf ? &bnx2xvf_eth_dev_ops : &bnx2x_eth_dev_ops;
-	pci_dev = eth_dev->pci_dev;
+	pci_dev = ETH_DEV_PCI_DEV(eth_dev);
 
 	rte_eth_copy_pci_info(eth_dev, pci_dev);
 
diff --git a/drivers/net/bnxt/bnxt_ethdev.c b/drivers/net/bnxt/bnxt_ethdev.c
index 035fe07..cd50f11 100644
--- a/drivers/net/bnxt/bnxt_ethdev.c
+++ b/drivers/net/bnxt/bnxt_ethdev.c
@@ -743,6 +743,7 @@ static int bnxt_reta_query_op(struct rte_eth_dev *eth_dev,
 {
 	struct bnxt *bp = (struct bnxt *)eth_dev->data->dev_private;
 	struct bnxt_vnic_info *vnic = &bp->vnic_info[0];
+	struct rte_pci_device *pci_dev = ETH_DEV_PCI_DEV(eth_dev);
 
 	/* Retrieve from the default VNIC */
 	if (!vnic)
@@ -759,7 +760,7 @@ static int bnxt_reta_query_op(struct rte_eth_dev *eth_dev,
 	/* EW - need to revisit here copying from u64 to u16 */
 	memcpy(reta_conf, vnic->rss_table, reta_size);
 
-	if (rte_intr_allow_others(&eth_dev->pci_dev->intr_handle)) {
+	if (rte_intr_allow_others(&pci_dev->intr_handle)) {
 		if (eth_dev->data->dev_conf.intr_conf.lsc != 0)
 			bnxt_dev_lsc_intr_setup(eth_dev);
 	}
@@ -1011,9 +1012,10 @@ static int bnxt_init_board(struct rte_eth_dev *eth_dev)
 {
 	int rc;
 	struct bnxt *bp = eth_dev->data->dev_private;
+	struct rte_pci_device *pci_dev = ETH_DEV_PCI_DEV(eth_dev);
 
 	/* enable device (incl. PCI PM wakeup), and bus-mastering */
-	if (!eth_dev->pci_dev->mem_resource[0].addr) {
+	if (!pci_dev->mem_resource[0].addr) {
 		RTE_LOG(ERR, PMD,
 			"Cannot find PCI device base address, aborting\n");
 		rc = -ENODEV;
@@ -1021,9 +1023,9 @@ static int bnxt_init_board(struct rte_eth_dev *eth_dev)
 	}
 
 	bp->eth_dev = eth_dev;
-	bp->pdev = eth_dev->pci_dev;
+	bp->pdev = pci_dev;
 
-	bp->bar0 = (void *)eth_dev->pci_dev->mem_resource[0].addr;
+	bp->bar0 = (void *)pci_dev->mem_resource[0].addr;
 	if (!bp->bar0) {
 		RTE_LOG(ERR, PMD, "Cannot map device registers, aborting\n");
 		rc = -ENOMEM;
@@ -1043,6 +1045,7 @@ static int bnxt_init_board(struct rte_eth_dev *eth_dev)
 static int
 bnxt_dev_init(struct rte_eth_dev *eth_dev)
 {
+	struct rte_pci_device *pci_dev = ETH_DEV_PCI_DEV(eth_dev);
 	static int version_printed;
 	struct bnxt *bp;
 	int rc;
@@ -1050,10 +1053,10 @@ bnxt_dev_init(struct rte_eth_dev *eth_dev)
 	if (version_printed++ == 0)
 		RTE_LOG(INFO, PMD, "%s", bnxt_version);
 
-	rte_eth_copy_pci_info(eth_dev, eth_dev->pci_dev);
+	rte_eth_copy_pci_info(eth_dev, pci_dev);
 	bp = eth_dev->data->dev_private;
 
-	if (bnxt_vf_pciid(eth_dev->pci_dev->id.device_id))
+	if (bnxt_vf_pciid(pci_dev->id.device_id))
 		bp->flags |= BNXT_FLAG_VF;
 
 	rc = bnxt_init_board(eth_dev);
@@ -1121,8 +1124,8 @@ bnxt_dev_init(struct rte_eth_dev *eth_dev)
 
 	RTE_LOG(INFO, PMD,
 		DRV_MODULE_NAME " found at mem %" PRIx64 ", node addr %pM\n",
-		eth_dev->pci_dev->mem_resource[0].phys_addr,
-		eth_dev->pci_dev->mem_resource[0].addr);
+		pci_dev->mem_resource[0].phys_addr,
+		pci_dev->mem_resource[0].addr);
 
 	bp->dev_stopped = 0;
 
diff --git a/drivers/net/bnxt/bnxt_ring.c b/drivers/net/bnxt/bnxt_ring.c
index 3f81ffc..6793d75 100644
--- a/drivers/net/bnxt/bnxt_ring.c
+++ b/drivers/net/bnxt/bnxt_ring.c
@@ -209,6 +209,7 @@ int bnxt_alloc_rings(struct bnxt *bp, uint16_t qidx,
  */
 int bnxt_alloc_hwrm_rings(struct bnxt *bp)
 {
+	struct rte_pci_device *pci_dev = ETH_DEV_PCI_DEV(bp->eth_dev);
 	unsigned int i;
 	int rc = 0;
 
@@ -223,7 +224,7 @@ int bnxt_alloc_hwrm_rings(struct bnxt *bp)
 		if (rc)
 			goto err_out;
 		cpr->cp_doorbell =
-		    (char *)bp->eth_dev->pci_dev->mem_resource[2].addr;
+		    (char *)pci_dev->mem_resource[2].addr;
 		B_CP_DIS_DB(cpr, cpr->cp_raw_cons);
 		bp->grp_info[0].cp_fw_ring_id = cp_ring->fw_ring_id;
 	}
@@ -243,7 +244,7 @@ int bnxt_alloc_hwrm_rings(struct bnxt *bp)
 		if (rc)
 			goto err_out;
 		cpr->cp_doorbell =
-		    (char *)bp->eth_dev->pci_dev->mem_resource[2].addr +
+		    (char *)pci_dev->mem_resource[2].addr +
 		    idx * 0x80;
 		bp->grp_info[idx].cp_fw_ring_id = cp_ring->fw_ring_id;
 		B_CP_DIS_DB(cpr, cpr->cp_raw_cons);
@@ -256,7 +257,7 @@ int bnxt_alloc_hwrm_rings(struct bnxt *bp)
 			goto err_out;
 		rxr->rx_prod = 0;
 		rxr->rx_doorbell =
-		    (char *)bp->eth_dev->pci_dev->mem_resource[2].addr +
+		    (char *)pci_dev->mem_resource[2].addr +
 		    idx * 0x80;
 		bp->grp_info[idx].rx_fw_ring_id = ring->fw_ring_id;
 		B_RX_DB(rxr->rx_doorbell, rxr->rx_prod);
@@ -284,7 +285,7 @@ int bnxt_alloc_hwrm_rings(struct bnxt *bp)
 			goto err_out;
 
 		cpr->cp_doorbell =
-		    (char *)bp->eth_dev->pci_dev->mem_resource[2].addr +
+		    (char *)pci_dev->mem_resource[2].addr +
 		    idx * 0x80;
 		bp->grp_info[idx].cp_fw_ring_id = cp_ring->fw_ring_id;
 		B_CP_DIS_DB(cpr, cpr->cp_raw_cons);
@@ -297,7 +298,7 @@ int bnxt_alloc_hwrm_rings(struct bnxt *bp)
 			goto err_out;
 
 		txr->tx_doorbell =
-		    (char *)bp->eth_dev->pci_dev->mem_resource[2].addr +
+		    (char *)pci_dev->mem_resource[2].addr +
 		    idx * 0x80;
 	}
 
diff --git a/drivers/net/cxgbe/cxgbe_ethdev.c b/drivers/net/cxgbe/cxgbe_ethdev.c
index b7f28eb..8bfdda8 100644
--- a/drivers/net/cxgbe/cxgbe_ethdev.c
+++ b/drivers/net/cxgbe/cxgbe_ethdev.c
@@ -1005,7 +1005,7 @@ static int eth_cxgbe_dev_init(struct rte_eth_dev *eth_dev)
 	if (rte_eal_process_type() != RTE_PROC_PRIMARY)
 		return 0;
 
-	pci_dev = eth_dev->pci_dev;
+	pci_dev = ETH_DEV_PCI_DEV(eth_dev);
 
 	snprintf(name, sizeof(name), "cxgbeadapter%d", eth_dev->data->port_id);
 	adapter = rte_zmalloc(name, sizeof(*adapter), 0);
diff --git a/drivers/net/e1000/em_ethdev.c b/drivers/net/e1000/em_ethdev.c
index aee3d34..7f2f521 100644
--- a/drivers/net/e1000/em_ethdev.c
+++ b/drivers/net/e1000/em_ethdev.c
@@ -295,7 +295,7 @@ eth_em_dev_init(struct rte_eth_dev *eth_dev)
 	struct e1000_vfta * shadow_vfta =
 		E1000_DEV_PRIVATE_TO_VFTA(eth_dev->data->dev_private);
 
-	pci_dev = eth_dev->pci_dev;
+	pci_dev = ETH_DEV_PCI_DEV(eth_dev);
 
 	eth_dev->dev_ops = &eth_em_ops;
 	eth_dev->rx_pkt_burst = (eth_rx_burst_t)&eth_em_recv_pkts;
@@ -369,7 +369,7 @@ eth_em_dev_uninit(struct rte_eth_dev *eth_dev)
 	if (rte_eal_process_type() != RTE_PROC_PRIMARY)
 		return -EPERM;
 
-	pci_dev = eth_dev->pci_dev;
+	pci_dev = ETH_DEV_PCI_DEV(eth_dev);
 
 	if (adapter->stopped == 0)
 		eth_em_close(eth_dev);
@@ -556,7 +556,8 @@ eth_em_start(struct rte_eth_dev *dev)
 		E1000_DEV_PRIVATE(dev->data->dev_private);
 	struct e1000_hw *hw =
 		E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
-	struct rte_intr_handle *intr_handle = &dev->pci_dev->intr_handle;
+	struct rte_pci_device *pci_dev = ETH_DEV_PCI_DEV(dev);
+	struct rte_intr_handle *intr_handle = &pci_dev->intr_handle;
 	int ret, mask;
 	uint32_t intr_vector = 0;
 	uint32_t *speeds;
@@ -738,7 +739,8 @@ eth_em_stop(struct rte_eth_dev *dev)
 {
 	struct rte_eth_link link;
 	struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
-	struct rte_intr_handle *intr_handle = &dev->pci_dev->intr_handle;
+	struct rte_pci_device *pci_dev = ETH_DEV_PCI_DEV(dev);
+	struct rte_intr_handle *intr_handle = &pci_dev->intr_handle;
 
 	em_rxq_intr_disable(hw);
 	em_lsc_intr_disable(hw);
@@ -999,9 +1001,10 @@ static int
 eth_em_rx_queue_intr_enable(struct rte_eth_dev *dev, __rte_unused uint16_t queue_id)
 {
 	struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
+	struct rte_pci_device *pci_dev = ETH_DEV_PCI_DEV(dev);
 
 	em_rxq_intr_enable(hw);
-	rte_intr_enable(&dev->pci_dev->intr_handle);
+	rte_intr_enable(&pci_dev->intr_handle);
 
 	return 0;
 }
@@ -1542,6 +1545,7 @@ eth_em_interrupt_action(struct rte_eth_dev *dev)
 		E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
 	struct e1000_interrupt *intr =
 		E1000_DEV_PRIVATE_TO_INTR(dev->data->dev_private);
+	struct rte_pci_device *pci_dev = ETH_DEV_PCI_DEV(dev);
 	uint32_t tctl, rctl;
 	struct rte_eth_link link;
 	int ret;
@@ -1550,7 +1554,7 @@ eth_em_interrupt_action(struct rte_eth_dev *dev)
 		return -1;
 
 	intr->flags &= ~E1000_FLAG_NEED_LINK_UPDATE;
-	rte_intr_enable(&(dev->pci_dev->intr_handle));
+	rte_intr_enable(&pci_dev->intr_handle);
 
 	/* set get_link_status to check register later */
 	hw->mac.get_link_status = 1;
@@ -1571,8 +1575,8 @@ eth_em_interrupt_action(struct rte_eth_dev *dev)
 		PMD_INIT_LOG(INFO, " Port %d: Link Down", dev->data->port_id);
 	}
 	PMD_INIT_LOG(DEBUG, "PCI Address: %04d:%02d:%02d:%d",
-		     dev->pci_dev->addr.domain, dev->pci_dev->addr.bus,
-		     dev->pci_dev->addr.devid, dev->pci_dev->addr.function);
+		     pci_dev->addr.domain, pci_dev->addr.bus,
+		     pci_dev->addr.devid, pci_dev->addr.function);
 
 	tctl = E1000_READ_REG(hw, E1000_TCTL);
 	rctl = E1000_READ_REG(hw, E1000_RCTL);
diff --git a/drivers/net/e1000/igb_ethdev.c b/drivers/net/e1000/igb_ethdev.c
index 2fddf0c..b25c66e 100644
--- a/drivers/net/e1000/igb_ethdev.c
+++ b/drivers/net/e1000/igb_ethdev.c
@@ -672,11 +672,11 @@ igb_identify_hardware(struct rte_eth_dev *dev)
 {
 	struct e1000_hw *hw =
 		E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
-
-	hw->vendor_id = dev->pci_dev->id.vendor_id;
-	hw->device_id = dev->pci_dev->id.device_id;
-	hw->subsystem_vendor_id = dev->pci_dev->id.subsystem_vendor_id;
-	hw->subsystem_device_id = dev->pci_dev->id.subsystem_device_id;
+	struct rte_pci_device *pci_dev = ETH_DEV_PCI_DEV(dev);
+	hw->vendor_id = pci_dev->id.vendor_id;
+	hw->device_id = pci_dev->id.device_id;
+	hw->subsystem_vendor_id = pci_dev->id.subsystem_vendor_id;
+	hw->subsystem_device_id = pci_dev->id.subsystem_device_id;
 
 	e1000_set_mac_type(hw);
 
@@ -755,7 +755,7 @@ eth_igb_dev_init(struct rte_eth_dev *eth_dev)
 
 	uint32_t ctrl_ext;
 
-	pci_dev = eth_dev->pci_dev;
+	pci_dev = ETH_DEV_PCI_DEV(eth_dev);
 
 	eth_dev->dev_ops = &eth_igb_ops;
 	eth_dev->rx_pkt_burst = &eth_igb_recv_pkts;
@@ -918,7 +918,7 @@ eth_igb_dev_uninit(struct rte_eth_dev *eth_dev)
 		return -EPERM;
 
 	hw = E1000_DEV_PRIVATE_TO_HW(eth_dev->data->dev_private);
-	pci_dev = eth_dev->pci_dev;
+	pci_dev = ETH_DEV_PCI_DEV(eth_dev);
 
 	if (adapter->stopped == 0)
 		eth_igb_close(eth_dev);
@@ -973,7 +973,7 @@ eth_igbvf_dev_init(struct rte_eth_dev *eth_dev)
 		return 0;
 	}
 
-	pci_dev = eth_dev->pci_dev;
+	pci_dev = ETH_DEV_PCI_DEV(eth_dev);
 
 	rte_eth_copy_pci_info(eth_dev, pci_dev);
 
@@ -1050,7 +1050,7 @@ eth_igbvf_dev_uninit(struct rte_eth_dev *eth_dev)
 {
 	struct e1000_adapter *adapter =
 		E1000_DEV_PRIVATE(eth_dev->data->dev_private);
-	struct rte_pci_device *pci_dev = eth_dev->pci_dev;
+	struct rte_pci_device *pci_dev = ETH_DEV_PCI_DEV(eth_dev);
 
 	PMD_INIT_FUNC_TRACE();
 
@@ -1217,7 +1217,8 @@ eth_igb_start(struct rte_eth_dev *dev)
 		E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
 	struct e1000_adapter *adapter =
 		E1000_DEV_PRIVATE(dev->data->dev_private);
-	struct rte_intr_handle *intr_handle = &dev->pci_dev->intr_handle;
+	struct rte_intr_handle *intr_handle =
+		&ETH_DEV_PCI_DEV(dev)->intr_handle;
 	int ret, mask;
 	uint32_t intr_vector = 0;
 	uint32_t ctrl_ext;
@@ -1429,7 +1430,8 @@ eth_igb_stop(struct rte_eth_dev *dev)
 	struct e1000_flex_filter *p_flex;
 	struct e1000_5tuple_filter *p_5tuple, *p_5tuple_next;
 	struct e1000_2tuple_filter *p_2tuple, *p_2tuple_next;
-	struct rte_intr_handle *intr_handle = &dev->pci_dev->intr_handle;
+	struct rte_intr_handle *intr_handle =
+		&ETH_DEV_PCI_DEV(dev)->intr_handle;
 
 	igb_intr_disable(hw);
 
@@ -1549,7 +1551,7 @@ eth_igb_close(struct rte_eth_dev *dev)
 
 	igb_dev_free_queues(dev);
 
-	pci_dev = dev->pci_dev;
+	pci_dev = ETH_DEV_PCI_DEV(dev);
 	if (pci_dev->intr_handle.intr_vec) {
 		rte_free(pci_dev->intr_handle.intr_vec);
 		pci_dev->intr_handle.intr_vec = NULL;
@@ -2639,6 +2641,7 @@ eth_igb_interrupt_action(struct rte_eth_dev *dev)
 		E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
 	struct e1000_interrupt *intr =
 		E1000_DEV_PRIVATE_TO_INTR(dev->data->dev_private);
+	struct rte_pci_device *pci_dev = ETH_DEV_PCI_DEV(dev);
 	uint32_t tctl, rctl;
 	struct rte_eth_link link;
 	int ret;
@@ -2649,7 +2652,7 @@ eth_igb_interrupt_action(struct rte_eth_dev *dev)
 	}
 
 	igb_intr_enable(dev);
-	rte_intr_enable(&(dev->pci_dev->intr_handle));
+	rte_intr_enable(&pci_dev->intr_handle);
 
 	if (intr->flags & E1000_FLAG_NEED_LINK_UPDATE) {
 		intr->flags &= ~E1000_FLAG_NEED_LINK_UPDATE;
@@ -2677,10 +2680,10 @@ eth_igb_interrupt_action(struct rte_eth_dev *dev)
 		}
 
 		PMD_INIT_LOG(DEBUG, "PCI Address: %04d:%02d:%02d:%d",
-			     dev->pci_dev->addr.domain,
-			     dev->pci_dev->addr.bus,
-			     dev->pci_dev->addr.devid,
-			     dev->pci_dev->addr.function);
+			     pci_dev->addr.domain,
+			     pci_dev->addr.bus,
+			     pci_dev->addr.devid,
+			     pci_dev->addr.function);
 		tctl = E1000_READ_REG(hw, E1000_TCTL);
 		rctl = E1000_READ_REG(hw, E1000_RCTL);
 		if (link.link_status) {
@@ -2770,7 +2773,7 @@ eth_igbvf_interrupt_action(struct rte_eth_dev *dev)
 	}
 
 	igbvf_intr_enable(dev);
-	rte_intr_enable(&dev->pci_dev->intr_handle);
+	rte_intr_enable(&ETH_DEV_PCI_DEV(dev)->intr_handle);
 
 	return 0;
 }
@@ -3056,7 +3059,8 @@ igbvf_dev_start(struct rte_eth_dev *dev)
 	struct e1000_adapter *adapter =
 		E1000_DEV_PRIVATE(dev->data->dev_private);
 	int ret;
-	struct rte_intr_handle *intr_handle = &dev->pci_dev->intr_handle;
+	struct rte_intr_handle *intr_handle =
+		&ETH_DEV_PCI_DEV(dev)->intr_handle;
 	uint32_t intr_vector = 0;
 
 	PMD_INIT_FUNC_TRACE();
@@ -3110,7 +3114,8 @@ igbvf_dev_start(struct rte_eth_dev *dev)
 static void
 igbvf_dev_stop(struct rte_eth_dev *dev)
 {
-	struct rte_intr_handle *intr_handle = &dev->pci_dev->intr_handle;
+	struct rte_intr_handle *intr_handle =
+		&ETH_DEV_PCI_DEV(dev)->intr_handle;
 
 	PMD_INIT_FUNC_TRACE();
 
@@ -5102,7 +5107,7 @@ eth_igb_rx_queue_intr_enable(struct rte_eth_dev *dev, uint16_t queue_id)
 	E1000_WRITE_REG(hw, E1000_EIMS, regval | mask);
 	E1000_WRITE_FLUSH(hw);
 
-	rte_intr_enable(&dev->pci_dev->intr_handle);
+	rte_intr_enable(&ETH_DEV_PCI_DEV(dev)->intr_handle);
 
 	return 0;
 }
@@ -5167,7 +5172,8 @@ eth_igb_configure_msix_intr(struct rte_eth_dev *dev)
 	uint32_t base = E1000_MISC_VEC_ID;
 	uint32_t misc_shift = 0;
 
-	struct rte_intr_handle *intr_handle = &dev->pci_dev->intr_handle;
+	struct rte_intr_handle *intr_handle =
+		&ETH_DEV_PCI_DEV(dev)->intr_handle;
 
 	/* won't configure msix register if no mapping is done
 	 * between intr vector and event fd
diff --git a/drivers/net/e1000/igb_pf.c b/drivers/net/e1000/igb_pf.c
index 5845bc2..6a72ee1 100644
--- a/drivers/net/e1000/igb_pf.c
+++ b/drivers/net/e1000/igb_pf.c
@@ -57,7 +57,8 @@
 static inline uint16_t
 dev_num_vf(struct rte_eth_dev *eth_dev)
 {
-	return eth_dev->pci_dev->max_vfs;
+	struct rte_pci_device *pci_dev = ETH_DEV_PCI_DEV(eth_dev);
+	return pci_dev->max_vfs;
 }
 
 static inline
diff --git a/drivers/net/ena/ena_ethdev.c b/drivers/net/ena/ena_ethdev.c
index ab9a178..c17d969 100644
--- a/drivers/net/ena/ena_ethdev.c
+++ b/drivers/net/ena/ena_ethdev.c
@@ -1278,7 +1278,7 @@ static int eth_ena_dev_init(struct rte_eth_dev *eth_dev)
 	if (rte_eal_process_type() != RTE_PROC_PRIMARY)
 		return 0;
 
-	pci_dev = eth_dev->pci_dev;
+	pci_dev = ETH_DEV_PCI_DEV(eth_dev);
 	adapter->pdev = pci_dev;
 
 	PMD_INIT_LOG(INFO, "Initializing %x:%x:%x.%d\n",
diff --git a/drivers/net/enic/enic_ethdev.c b/drivers/net/enic/enic_ethdev.c
index 2b154ec..553a88e 100644
--- a/drivers/net/enic/enic_ethdev.c
+++ b/drivers/net/enic/enic_ethdev.c
@@ -621,7 +621,7 @@ static int eth_enicpmd_dev_init(struct rte_eth_dev *eth_dev)
 	eth_dev->rx_pkt_burst = &enic_recv_pkts;
 	eth_dev->tx_pkt_burst = &enic_xmit_pkts;
 
-	pdev = eth_dev->pci_dev;
+	pdev = ETH_DEV_PCI_DEV(eth_dev);
 	rte_eth_copy_pci_info(eth_dev, pdev);
 	enic->pdev = pdev;
 	addr = &pdev->addr;
diff --git a/drivers/net/fm10k/fm10k_ethdev.c b/drivers/net/fm10k/fm10k_ethdev.c
index 923690c..e1250f6 100644
--- a/drivers/net/fm10k/fm10k_ethdev.c
+++ b/drivers/net/fm10k/fm10k_ethdev.c
@@ -59,7 +59,8 @@
 #define BIT_MASK_PER_UINT32 ((1 << CHARS_PER_UINT32) - 1)
 
 /* default 1:1 map from queue ID to interrupt vector ID */
-#define Q2V(dev, queue_id) (dev->pci_dev->intr_handle.intr_vec[queue_id])
+#define D2IH(dev) (&ETH_DEV_PCI_DEV(dev)->intr_handle)
+#define Q2V(dev, queue_id) (D2IH(dev)->intr_vec[queue_id])
 
 /* First 64 Logical ports for PF/VMDQ, second 64 for Flow director */
 #define MAX_LPORT_NUM    128
@@ -711,7 +712,7 @@ fm10k_dev_rx_init(struct rte_eth_dev *dev)
 {
 	struct fm10k_hw *hw = FM10K_DEV_PRIVATE_TO_HW(dev->data->dev_private);
 	struct fm10k_macvlan_filter_info *macvlan;
-	struct rte_intr_handle *intr_handle = &dev->pci_dev->intr_handle;
+	struct rte_intr_handle *intr_handle = D2IH(dev);
 	int i, ret;
 	struct fm10k_rx_queue *rxq;
 	uint64_t base_addr;
@@ -1171,7 +1172,7 @@ static void
 fm10k_dev_stop(struct rte_eth_dev *dev)
 {
 	struct fm10k_hw *hw = FM10K_DEV_PRIVATE_TO_HW(dev->data->dev_private);
-	struct rte_intr_handle *intr_handle = &dev->pci_dev->intr_handle;
+	struct rte_intr_handle *intr_handle = D2IH(dev);
 	int i;
 
 	PMD_INIT_FUNC_TRACE();
@@ -1387,6 +1388,7 @@ fm10k_dev_infos_get(struct rte_eth_dev *dev,
 	struct rte_eth_dev_info *dev_info)
 {
 	struct fm10k_hw *hw = FM10K_DEV_PRIVATE_TO_HW(dev->data->dev_private);
+	struct rte_pci_device *pci_dev = ETH_DEV_PCI_DEV(dev);
 
 	PMD_INIT_FUNC_TRACE();
 
@@ -1396,7 +1398,7 @@ fm10k_dev_infos_get(struct rte_eth_dev *dev,
 	dev_info->max_tx_queues      = hw->mac.max_queues;
 	dev_info->max_mac_addrs      = FM10K_MAX_MACADDR_NUM;
 	dev_info->max_hash_mac_addrs = 0;
-	dev_info->max_vfs            = dev->pci_dev->max_vfs;
+	dev_info->max_vfs            = pci_dev->max_vfs;
 	dev_info->vmdq_pool_base     = 0;
 	dev_info->vmdq_queue_base    = 0;
 	dev_info->max_vmdq_pools     = ETH_32_POOLS;
@@ -2341,7 +2343,7 @@ fm10k_dev_rx_queue_intr_enable(struct rte_eth_dev *dev, uint16_t queue_id)
 	else
 		FM10K_WRITE_REG(hw, FM10K_VFITR(Q2V(dev, queue_id)),
 			FM10K_ITR_AUTOMASK | FM10K_ITR_MASK_CLEAR);
-	rte_intr_enable(&dev->pci_dev->intr_handle);
+	rte_intr_enable(D2IH(dev));
 	return 0;
 }
 
@@ -2364,7 +2366,7 @@ static int
 fm10k_dev_rxq_interrupt_setup(struct rte_eth_dev *dev)
 {
 	struct fm10k_hw *hw = FM10K_DEV_PRIVATE_TO_HW(dev->data->dev_private);
-	struct rte_intr_handle *intr_handle = &dev->pci_dev->intr_handle;
+	struct rte_intr_handle *intr_handle = D2IH(dev);
 	uint32_t intr_vector, vec;
 	uint16_t queue_id;
 	int result = 0;
@@ -2380,7 +2382,7 @@ fm10k_dev_rxq_interrupt_setup(struct rte_eth_dev *dev)
 	intr_vector = dev->data->nb_rx_queues;
 
 	/* disable interrupt first */
-	rte_intr_disable(&dev->pci_dev->intr_handle);
+	rte_intr_disable(intr_handle);
 	if (hw->mac.type == fm10k_mac_pf)
 		fm10k_dev_disable_intr_pf(dev);
 	else
@@ -2415,7 +2417,7 @@ fm10k_dev_rxq_interrupt_setup(struct rte_eth_dev *dev)
 		fm10k_dev_enable_intr_pf(dev);
 	else
 		fm10k_dev_enable_intr_vf(dev);
-	rte_intr_enable(&dev->pci_dev->intr_handle);
+	rte_intr_enable(intr_handle);
 	hw->mac.ops.update_int_moderator(hw);
 	return result;
 }
@@ -2581,7 +2583,7 @@ fm10k_dev_interrupt_handler_pf(
 	FM10K_WRITE_REG(hw, FM10K_ITR(0), FM10K_ITR_AUTOMASK |
 					FM10K_ITR_MASK_CLEAR);
 	/* Re-enable interrupt from host side */
-	rte_intr_enable(&(dev->pci_dev->intr_handle));
+	rte_intr_enable(D2IH(dev));
 }
 
 /**
@@ -2615,7 +2617,7 @@ fm10k_dev_interrupt_handler_vf(
 	FM10K_WRITE_REG(hw, FM10K_VFITR(0), FM10K_ITR_AUTOMASK |
 					FM10K_ITR_MASK_CLEAR);
 	/* Re-enable interrupt from host side */
-	rte_intr_enable(&(dev->pci_dev->intr_handle));
+	rte_intr_enable(D2IH(dev));
 }
 
 /* Mailbox message handler in VF */
@@ -2827,6 +2829,7 @@ static int
 eth_fm10k_dev_init(struct rte_eth_dev *dev)
 {
 	struct fm10k_hw *hw = FM10K_DEV_PRIVATE_TO_HW(dev->data->dev_private);
+	struct rte_pci_device *pci_dev = ETH_DEV_PCI_DEV(dev);
 	int diag, i;
 	struct fm10k_macvlan_filter_info *macvlan;
 
@@ -2840,18 +2843,18 @@ eth_fm10k_dev_init(struct rte_eth_dev *dev)
 	if (rte_eal_process_type() != RTE_PROC_PRIMARY)
 		return 0;
 
-	rte_eth_copy_pci_info(dev, dev->pci_dev);
+	rte_eth_copy_pci_info(dev, pci_dev);
 
 	macvlan = FM10K_DEV_PRIVATE_TO_MACVLAN(dev->data->dev_private);
 	memset(macvlan, 0, sizeof(*macvlan));
 	/* Vendor and Device ID need to be set before init of shared code */
 	memset(hw, 0, sizeof(*hw));
-	hw->device_id = dev->pci_dev->id.device_id;
-	hw->vendor_id = dev->pci_dev->id.vendor_id;
-	hw->subsystem_device_id = dev->pci_dev->id.subsystem_device_id;
-	hw->subsystem_vendor_id = dev->pci_dev->id.subsystem_vendor_id;
+	hw->device_id = pci_dev->id.device_id;
+	hw->vendor_id = pci_dev->id.vendor_id;
+	hw->subsystem_device_id = pci_dev->id.subsystem_device_id;
+	hw->subsystem_vendor_id = pci_dev->id.subsystem_vendor_id;
 	hw->revision_id = 0;
-	hw->hw_addr = (void *)dev->pci_dev->mem_resource[0].addr;
+	hw->hw_addr = (void *)pci_dev->mem_resource[0].addr;
 	if (hw->hw_addr == NULL) {
 		PMD_INIT_LOG(ERR, "Bad mem resource."
 			" Try to blacklist unused devices.");
@@ -2921,20 +2924,20 @@ eth_fm10k_dev_init(struct rte_eth_dev *dev)
 	/*PF/VF has different interrupt handling mechanism */
 	if (hw->mac.type == fm10k_mac_pf) {
 		/* register callback func to eal lib */
-		rte_intr_callback_register(&(dev->pci_dev->intr_handle),
+		rte_intr_callback_register(D2IH(dev),
 			fm10k_dev_interrupt_handler_pf, (void *)dev);
 
 		/* enable MISC interrupt */
 		fm10k_dev_enable_intr_pf(dev);
 	} else { /* VF */
-		rte_intr_callback_register(&(dev->pci_dev->intr_handle),
+		rte_intr_callback_register(D2IH(dev),
 			fm10k_dev_interrupt_handler_vf, (void *)dev);
 
 		fm10k_dev_enable_intr_vf(dev);
 	}
 
 	/* Enable intr after callback registered */
-	rte_intr_enable(&(dev->pci_dev->intr_handle));
+	rte_intr_enable(D2IH(dev));
 
 	hw->mac.ops.update_int_moderator(hw);
 
@@ -3004,7 +3007,7 @@ static int
 eth_fm10k_dev_uninit(struct rte_eth_dev *dev)
 {
 	struct fm10k_hw *hw = FM10K_DEV_PRIVATE_TO_HW(dev->data->dev_private);
-
+	struct rte_intr_handle *intr_handle = D2IH(dev);
 	PMD_INIT_FUNC_TRACE();
 
 	/* only uninitialize in the primary process */
@@ -3019,7 +3022,7 @@ eth_fm10k_dev_uninit(struct rte_eth_dev *dev)
 	dev->tx_pkt_burst = NULL;
 
 	/* disable uio/vfio intr */
-	rte_intr_disable(&(dev->pci_dev->intr_handle));
+	rte_intr_disable(intr_handle);
 
 	/*PF/VF has different interrupt handling mechanism */
 	if (hw->mac.type == fm10k_mac_pf) {
@@ -3027,13 +3030,13 @@ eth_fm10k_dev_uninit(struct rte_eth_dev *dev)
 		fm10k_dev_disable_intr_pf(dev);
 
 		/* unregister callback func to eal lib */
-		rte_intr_callback_unregister(&(dev->pci_dev->intr_handle),
+		rte_intr_callback_unregister(intr_handle,
 			fm10k_dev_interrupt_handler_pf, (void *)dev);
 	} else {
 		/* disable interrupt */
 		fm10k_dev_disable_intr_vf(dev);
 
-		rte_intr_callback_unregister(&(dev->pci_dev->intr_handle),
+		rte_intr_callback_unregister(intr_handle,
 			fm10k_dev_interrupt_handler_vf, (void *)dev);
 	}
 
diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40e_ethdev.c
index 67778ba..8a63a8c 100644
--- a/drivers/net/i40e/i40e_ethdev.c
+++ b/drivers/net/i40e/i40e_ethdev.c
@@ -907,7 +907,7 @@ is_floating_veb_supported(struct rte_devargs *devargs)
 static void
 config_floating_veb(struct rte_eth_dev *dev)
 {
-	struct rte_pci_device *pci_dev = dev->pci_dev;
+	struct rte_pci_device *pci_dev = ETH_DEV_PCI_DEV(dev);
 	struct i40e_pf *pf = I40E_DEV_PRIVATE_TO_PF(dev->data->dev_private);
 	struct i40e_hw *hw = I40E_DEV_PRIVATE_TO_HW(dev->data->dev_private);
 
@@ -952,7 +952,7 @@ eth_i40e_dev_init(struct rte_eth_dev *dev)
 		i40e_set_tx_function(dev);
 		return 0;
 	}
-	pci_dev = dev->pci_dev;
+	pci_dev = ETH_DEV_PCI_DEV(dev);
 
 	rte_eth_copy_pci_info(dev, pci_dev);
 
@@ -1215,7 +1215,7 @@ eth_i40e_dev_uninit(struct rte_eth_dev *dev)
 		return 0;
 
 	hw = I40E_DEV_PRIVATE_TO_HW(dev->data->dev_private);
-	pci_dev = dev->pci_dev;
+	pci_dev = ETH_DEV_PCI_DEV(dev);
 
 	if (hw->adapter_stopped == 0)
 		i40e_dev_close(dev);
@@ -1335,7 +1335,7 @@ void
 i40e_vsi_queues_unbind_intr(struct i40e_vsi *vsi)
 {
 	struct rte_eth_dev *dev = vsi->adapter->eth_dev;
-	struct rte_intr_handle *intr_handle = &dev->pci_dev->intr_handle;
+	struct rte_intr_handle *intr_handle = ETH_DEV_TO_INTR_HANDLE(dev);
 	struct i40e_hw *hw = I40E_VSI_TO_HW(vsi);
 	uint16_t msix_vect = vsi->msix_intr;
 	uint16_t i;
@@ -1448,7 +1448,7 @@ void
 i40e_vsi_queues_bind_intr(struct i40e_vsi *vsi)
 {
 	struct rte_eth_dev *dev = vsi->adapter->eth_dev;
-	struct rte_intr_handle *intr_handle = &dev->pci_dev->intr_handle;
+	struct rte_intr_handle *intr_handle = ETH_DEV_TO_INTR_HANDLE(dev);
 	struct i40e_hw *hw = I40E_VSI_TO_HW(vsi);
 	uint16_t msix_vect = vsi->msix_intr;
 	uint16_t nb_msix = RTE_MIN(vsi->nb_msix, intr_handle->nb_efd);
@@ -1519,7 +1519,7 @@ static void
 i40e_vsi_enable_queues_intr(struct i40e_vsi *vsi)
 {
 	struct rte_eth_dev *dev = vsi->adapter->eth_dev;
-	struct rte_intr_handle *intr_handle = &dev->pci_dev->intr_handle;
+	struct rte_intr_handle *intr_handle = ETH_DEV_TO_INTR_HANDLE(dev);
 	struct i40e_hw *hw = I40E_VSI_TO_HW(vsi);
 	uint16_t interval = i40e_calc_itr_interval(\
 		RTE_LIBRTE_I40E_ITR_INTERVAL);
@@ -1550,7 +1550,7 @@ static void
 i40e_vsi_disable_queues_intr(struct i40e_vsi *vsi)
 {
 	struct rte_eth_dev *dev = vsi->adapter->eth_dev;
-	struct rte_intr_handle *intr_handle = &dev->pci_dev->intr_handle;
+	struct rte_intr_handle *intr_handle = ETH_DEV_TO_INTR_HANDLE(dev);
 	struct i40e_hw *hw = I40E_VSI_TO_HW(vsi);
 	uint16_t msix_intr, i;
 
@@ -1675,7 +1675,7 @@ i40e_dev_start(struct rte_eth_dev *dev)
 	struct i40e_hw *hw = I40E_DEV_PRIVATE_TO_HW(dev->data->dev_private);
 	struct i40e_vsi *main_vsi = pf->main_vsi;
 	int ret, i;
-	struct rte_intr_handle *intr_handle = &dev->pci_dev->intr_handle;
+	struct rte_intr_handle *intr_handle = ETH_DEV_TO_INTR_HANDLE(dev);
 	uint32_t intr_vector = 0;
 
 	hw->adapter_stopped = 0;
@@ -1808,7 +1808,7 @@ i40e_dev_stop(struct rte_eth_dev *dev)
 	struct i40e_pf *pf = I40E_DEV_PRIVATE_TO_PF(dev->data->dev_private);
 	struct i40e_vsi *main_vsi = pf->main_vsi;
 	struct i40e_mirror_rule *p_mirror;
-	struct rte_intr_handle *intr_handle = &dev->pci_dev->intr_handle;
+	struct rte_intr_handle *intr_handle = ETH_DEV_TO_INTR_HANDLE(dev);
 	int i;
 
 	/* Disable all queues */
@@ -1870,7 +1870,7 @@ i40e_dev_close(struct rte_eth_dev *dev)
 
 	/* Disable interrupt */
 	i40e_pf_disable_irq0(hw);
-	rte_intr_disable(&(dev->pci_dev->intr_handle));
+	rte_intr_disable(ETH_DEV_TO_INTR_HANDLE(dev));
 
 	/* shutdown and destroy the HMC */
 	i40e_shutdown_lan_hmc(hw);
@@ -2588,7 +2588,7 @@ i40e_dev_info_get(struct rte_eth_dev *dev, struct rte_eth_dev_info *dev_info)
 	dev_info->min_rx_bufsize = I40E_BUF_SIZE_MIN;
 	dev_info->max_rx_pktlen = I40E_FRAME_SIZE_MAX;
 	dev_info->max_mac_addrs = vsi->max_macaddrs;
-	dev_info->max_vfs = dev->pci_dev->max_vfs;
+	dev_info->max_vfs = ETH_DEV_PCI_DEV(dev)->max_vfs;
 	dev_info->rx_offload_capa =
 		DEV_RX_OFFLOAD_VLAN_STRIP |
 		DEV_RX_OFFLOAD_QINQ_STRIP |
@@ -3488,11 +3488,12 @@ i40e_get_cap(struct i40e_hw *hw)
 static int
 i40e_pf_parameter_init(struct rte_eth_dev *dev)
 {
+	struct rte_pci_device *pci_dev = ETH_DEV_PCI_DEV(dev);
 	struct i40e_pf *pf = I40E_DEV_PRIVATE_TO_PF(dev->data->dev_private);
 	struct i40e_hw *hw = I40E_PF_TO_HW(pf);
 	uint16_t qp_count = 0, vsi_count = 0;
 
-	if (dev->pci_dev->max_vfs && !hw->func_caps.sr_iov_1_1) {
+	if (pci_dev->max_vfs && !hw->func_caps.sr_iov_1_1) {
 		PMD_INIT_LOG(ERR, "HW configuration doesn't support SRIOV");
 		return -EINVAL;
 	}
@@ -3533,10 +3534,10 @@ i40e_pf_parameter_init(struct rte_eth_dev *dev)
 
 	/* VF queue/VSI allocation */
 	pf->vf_qp_offset = pf->lan_qp_offset + pf->lan_nb_qps;
-	if (hw->func_caps.sr_iov_1_1 && dev->pci_dev->max_vfs) {
+	if (hw->func_caps.sr_iov_1_1 && pci_dev->max_vfs) {
 		pf->flags |= I40E_FLAG_SRIOV;
 		pf->vf_nb_qps = RTE_LIBRTE_I40E_QUEUE_NUM_PER_VF;
-		pf->vf_num = dev->pci_dev->max_vfs;
+		pf->vf_num = pci_dev->max_vfs;
 		PMD_DRV_LOG(DEBUG, "%u VF VSIs, %u queues per VF VSI, "
 			    "in total %u queues", pf->vf_num, pf->vf_nb_qps,
 			    pf->vf_nb_qps * pf->vf_num);
@@ -5573,7 +5574,7 @@ i40e_dev_interrupt_handler(__rte_unused struct rte_intr_handle *handle,
 done:
 	/* Enable interrupt */
 	i40e_pf_enable_irq0(hw);
-	rte_intr_enable(&(dev->pci_dev->intr_handle));
+	rte_intr_enable(ETH_DEV_TO_INTR_HANDLE(dev));
 }
 
 static int
@@ -8124,10 +8125,11 @@ i40e_dev_filter_ctrl(struct rte_eth_dev *dev,
 static void
 i40e_enable_extended_tag(struct rte_eth_dev *dev)
 {
+	struct rte_pci_device *pci_dev = ETH_DEV_PCI_DEV(dev);
 	uint32_t buf = 0;
 	int ret;
 
-	ret = rte_eal_pci_read_config(dev->pci_dev, &buf, sizeof(buf),
+	ret = rte_eal_pci_read_config(pci_dev, &buf, sizeof(buf),
 				      PCI_DEV_CAP_REG);
 	if (ret < 0) {
 		PMD_DRV_LOG(ERR, "Failed to read PCI offset 0x%x",
@@ -8140,7 +8142,7 @@ i40e_enable_extended_tag(struct rte_eth_dev *dev)
 	}
 
 	buf = 0;
-	ret = rte_eal_pci_read_config(dev->pci_dev, &buf, sizeof(buf),
+	ret = rte_eal_pci_read_config(pci_dev, &buf, sizeof(buf),
 				      PCI_DEV_CTRL_REG);
 	if (ret < 0) {
 		PMD_DRV_LOG(ERR, "Failed to read PCI offset 0x%x",
@@ -8152,7 +8154,7 @@ i40e_enable_extended_tag(struct rte_eth_dev *dev)
 		return;
 	}
 	buf |= PCI_DEV_CTRL_EXT_TAG_MASK;
-	ret = rte_eal_pci_write_config(dev->pci_dev, &buf, sizeof(buf),
+	ret = rte_eal_pci_write_config(pci_dev, &buf, sizeof(buf),
 				       PCI_DEV_CTRL_REG);
 	if (ret < 0) {
 		PMD_DRV_LOG(ERR, "Failed to write PCI offset 0x%x",
@@ -9555,7 +9557,7 @@ i40e_dev_get_dcb_info(struct rte_eth_dev *dev,
 static int
 i40e_dev_rx_queue_intr_enable(struct rte_eth_dev *dev, uint16_t queue_id)
 {
-	struct rte_intr_handle *intr_handle = &dev->pci_dev->intr_handle;
+	struct rte_intr_handle *intr_handle = ETH_DEV_TO_INTR_HANDLE(dev);
 	struct i40e_hw *hw = I40E_DEV_PRIVATE_TO_HW(dev->data->dev_private);
 	uint16_t interval =
 		i40e_calc_itr_interval(RTE_LIBRTE_I40E_ITR_INTERVAL);
@@ -9580,7 +9582,7 @@ i40e_dev_rx_queue_intr_enable(struct rte_eth_dev *dev, uint16_t queue_id)
 				I40E_PFINT_DYN_CTLN_INTERVAL_SHIFT));
 
 	I40E_WRITE_FLUSH(hw);
-	rte_intr_enable(&dev->pci_dev->intr_handle);
+	rte_intr_enable(ETH_DEV_TO_INTR_HANDLE(dev));
 
 	return 0;
 }
@@ -9588,7 +9590,7 @@ i40e_dev_rx_queue_intr_enable(struct rte_eth_dev *dev, uint16_t queue_id)
 static int
 i40e_dev_rx_queue_intr_disable(struct rte_eth_dev *dev, uint16_t queue_id)
 {
-	struct rte_intr_handle *intr_handle = &dev->pci_dev->intr_handle;
+	struct rte_intr_handle *intr_handle = ETH_DEV_TO_INTR_HANDLE(dev);
 	struct i40e_hw *hw = I40E_DEV_PRIVATE_TO_HW(dev->data->dev_private);
 	uint16_t msix_intr;
 
diff --git a/drivers/net/i40e/i40e_ethdev.h b/drivers/net/i40e/i40e_ethdev.h
index 298cef4..9d4bea7 100644
--- a/drivers/net/i40e/i40e_ethdev.h
+++ b/drivers/net/i40e/i40e_ethdev.h
@@ -671,6 +671,10 @@ i40e_get_vsi_from_adapter(struct i40e_adapter *adapter)
 #define I40E_VF_TO_HW(vf) \
 	(&(((struct i40e_vf *)vf)->adapter->hw))
 
+/* ETH_DEV_TO_INTR_HANDLE */
+#define ETH_DEV_TO_INTR_HANDLE(ptr) \
+	(&(ETH_DEV_PCI_DEV(ptr)->intr_handle))
+
 static inline void
 i40e_init_adminq_parameter(struct i40e_hw *hw)
 {
diff --git a/drivers/net/i40e/i40e_ethdev_vf.c b/drivers/net/i40e/i40e_ethdev_vf.c
index aa306d6..781e658 100644
--- a/drivers/net/i40e/i40e_ethdev_vf.c
+++ b/drivers/net/i40e/i40e_ethdev_vf.c
@@ -718,7 +718,7 @@ i40evf_config_irq_map(struct rte_eth_dev *dev)
 	uint8_t cmd_buffer[sizeof(struct i40e_virtchnl_irq_map_info) + \
 		sizeof(struct i40e_virtchnl_vector_map)];
 	struct i40e_virtchnl_irq_map_info *map_info;
-	struct rte_intr_handle *intr_handle = &dev->pci_dev->intr_handle;
+	struct rte_intr_handle *intr_handle = ETH_DEV_TO_INTR_HANDLE(dev);
 	uint32_t vector_id;
 	int i, err;
 
@@ -1431,7 +1431,7 @@ i40evf_dev_interrupt_handler(__rte_unused struct rte_intr_handle *handle,
 
 done:
 	i40evf_enable_irq0(hw);
-	rte_intr_enable(&dev->pci_dev->intr_handle);
+	rte_intr_enable(ETH_DEV_TO_INTR_HANDLE(dev));
 }
 
 static int
@@ -1439,7 +1439,7 @@ i40evf_dev_init(struct rte_eth_dev *eth_dev)
 {
 	struct i40e_hw *hw = I40E_DEV_PRIVATE_TO_HW(\
 			eth_dev->data->dev_private);
-	struct rte_pci_device *pci_dev = eth_dev->pci_dev;
+	struct rte_pci_device *pci_dev = ETH_DEV_PCI_DEV(eth_dev);
 
 	PMD_INIT_FUNC_TRACE();
 
@@ -1458,15 +1458,15 @@ i40evf_dev_init(struct rte_eth_dev *eth_dev)
 		return 0;
 	}
 
-	rte_eth_copy_pci_info(eth_dev, eth_dev->pci_dev);
+	rte_eth_copy_pci_info(eth_dev, pci_dev);
 
-	hw->vendor_id = eth_dev->pci_dev->id.vendor_id;
-	hw->device_id = eth_dev->pci_dev->id.device_id;
-	hw->subsystem_vendor_id = eth_dev->pci_dev->id.subsystem_vendor_id;
-	hw->subsystem_device_id = eth_dev->pci_dev->id.subsystem_device_id;
-	hw->bus.device = eth_dev->pci_dev->addr.devid;
-	hw->bus.func = eth_dev->pci_dev->addr.function;
-	hw->hw_addr = (void *)eth_dev->pci_dev->mem_resource[0].addr;
+	hw->vendor_id = pci_dev->id.vendor_id;
+	hw->device_id = pci_dev->id.device_id;
+	hw->subsystem_vendor_id = pci_dev->id.subsystem_vendor_id;
+	hw->subsystem_device_id = pci_dev->id.subsystem_device_id;
+	hw->bus.device = pci_dev->addr.devid;
+	hw->bus.func = pci_dev->addr.function;
+	hw->hw_addr = (void *)pci_dev->mem_resource[0].addr;
 	hw->adapter_stopped = 0;
 
 	if(i40evf_init_vf(eth_dev) != 0) {
@@ -1853,7 +1853,7 @@ i40evf_enable_queues_intr(struct rte_eth_dev *dev)
 {
 	struct i40e_vf *vf = I40EVF_DEV_PRIVATE_TO_VF(dev->data->dev_private);
 	struct i40e_hw *hw = I40E_DEV_PRIVATE_TO_HW(dev->data->dev_private);
-	struct rte_intr_handle *intr_handle = &dev->pci_dev->intr_handle;
+	struct rte_intr_handle *intr_handle = ETH_DEV_TO_INTR_HANDLE(dev);
 
 	if (!rte_intr_allow_others(intr_handle)) {
 		I40E_WRITE_REG(hw,
@@ -1885,7 +1885,7 @@ i40evf_disable_queues_intr(struct rte_eth_dev *dev)
 {
 	struct i40e_vf *vf = I40EVF_DEV_PRIVATE_TO_VF(dev->data->dev_private);
 	struct i40e_hw *hw = I40E_DEV_PRIVATE_TO_HW(dev->data->dev_private);
-	struct rte_intr_handle *intr_handle = &dev->pci_dev->intr_handle;
+	struct rte_intr_handle *intr_handle = ETH_DEV_TO_INTR_HANDLE(dev);
 
 	if (!rte_intr_allow_others(intr_handle)) {
 		I40E_WRITE_REG(hw, I40E_VFINT_DYN_CTL01,
@@ -1911,7 +1911,7 @@ i40evf_disable_queues_intr(struct rte_eth_dev *dev)
 static int
 i40evf_dev_rx_queue_intr_enable(struct rte_eth_dev *dev, uint16_t queue_id)
 {
-	struct rte_intr_handle *intr_handle = &dev->pci_dev->intr_handle;
+	struct rte_intr_handle *intr_handle = ETH_DEV_TO_INTR_HANDLE(dev);
 	struct i40e_hw *hw = I40E_DEV_PRIVATE_TO_HW(dev->data->dev_private);
 	uint16_t interval =
 		i40e_calc_itr_interval(RTE_LIBRTE_I40E_ITR_INTERVAL);
@@ -1937,7 +1937,7 @@ i40evf_dev_rx_queue_intr_enable(struct rte_eth_dev *dev, uint16_t queue_id)
 
 	I40EVF_WRITE_FLUSH(hw);
 
-	rte_intr_enable(&dev->pci_dev->intr_handle);
+	rte_intr_enable(ETH_DEV_TO_INTR_HANDLE(dev));
 
 	return 0;
 }
@@ -1945,7 +1945,7 @@ i40evf_dev_rx_queue_intr_enable(struct rte_eth_dev *dev, uint16_t queue_id)
 static int
 i40evf_dev_rx_queue_intr_disable(struct rte_eth_dev *dev, uint16_t queue_id)
 {
-	struct rte_intr_handle *intr_handle = &dev->pci_dev->intr_handle;
+	struct rte_intr_handle *intr_handle = ETH_DEV_TO_INTR_HANDLE(dev);
 	struct i40e_hw *hw = I40E_DEV_PRIVATE_TO_HW(dev->data->dev_private);
 	uint16_t msix_intr;
 
@@ -2025,7 +2025,7 @@ i40evf_dev_start(struct rte_eth_dev *dev)
 {
 	struct i40e_vf *vf = I40EVF_DEV_PRIVATE_TO_VF(dev->data->dev_private);
 	struct i40e_hw *hw = I40E_DEV_PRIVATE_TO_HW(dev->data->dev_private);
-	struct rte_intr_handle *intr_handle = &dev->pci_dev->intr_handle;
+	struct rte_intr_handle *intr_handle = ETH_DEV_TO_INTR_HANDLE(dev);
 	uint32_t intr_vector = 0;
 
 	PMD_INIT_FUNC_TRACE();
@@ -2090,7 +2090,7 @@ i40evf_dev_start(struct rte_eth_dev *dev)
 static void
 i40evf_dev_stop(struct rte_eth_dev *dev)
 {
-	struct rte_intr_handle *intr_handle = &dev->pci_dev->intr_handle;
+	struct rte_intr_handle *intr_handle = ETH_DEV_TO_INTR_HANDLE(dev);
 
 	PMD_INIT_FUNC_TRACE();
 
@@ -2285,7 +2285,7 @@ static void
 i40evf_dev_close(struct rte_eth_dev *dev)
 {
 	struct i40e_hw *hw = I40E_DEV_PRIVATE_TO_HW(dev->data->dev_private);
-	struct rte_pci_device *pci_dev = dev->pci_dev;
+	struct rte_pci_device *pci_dev = ETH_DEV_PCI_DEV(dev);
 
 	i40evf_dev_stop(dev);
 	hw->adapter_stopped = 1;
diff --git a/drivers/net/ixgbe/ixgbe_ethdev.c b/drivers/net/ixgbe/ixgbe_ethdev.c
index edc9b22..f17da46 100644
--- a/drivers/net/ixgbe/ixgbe_ethdev.c
+++ b/drivers/net/ixgbe/ixgbe_ethdev.c
@@ -427,6 +427,9 @@ static int ixgbe_dev_udp_tunnel_port_del(struct rte_eth_dev *dev,
 		(r) = (h)->bitmap[idx] >> bit & 1;\
 	} while (0)
 
+#define ETH_DEV_TO_INTR_HANDLE(ptr)			\
+	(&(ETH_DEV_PCI_DEV(ptr)->intr_handle))
+
 /*
  * The set of PCI devices this driver supports
  */
@@ -1127,7 +1130,7 @@ eth_ixgbe_dev_init(struct rte_eth_dev *eth_dev)
 
 		return 0;
 	}
-	pci_dev = eth_dev->pci_dev;
+	pci_dev = ETH_DEV_PCI_DEV(eth_dev);
 
 	rte_eth_copy_pci_info(eth_dev, pci_dev);
 
@@ -1302,7 +1305,7 @@ eth_ixgbe_dev_uninit(struct rte_eth_dev *eth_dev)
 		return -EPERM;
 
 	hw = IXGBE_DEV_PRIVATE_TO_HW(eth_dev->data->dev_private);
-	pci_dev = eth_dev->pci_dev;
+	pci_dev = ETH_DEV_PCI_DEV(eth_dev);
 
 	if (hw->adapter_stopped == 0)
 		ixgbe_dev_close(eth_dev);
@@ -1419,7 +1422,7 @@ eth_ixgbevf_dev_init(struct rte_eth_dev *eth_dev)
 		return 0;
 	}
 
-	pci_dev = eth_dev->pci_dev;
+	pci_dev = ETH_DEV_PCI_DEV(eth_dev);
 
 	rte_eth_copy_pci_info(eth_dev, pci_dev);
 
@@ -1532,7 +1535,9 @@ static int
 eth_ixgbevf_dev_uninit(struct rte_eth_dev *eth_dev)
 {
 	struct ixgbe_hw *hw;
-	struct rte_pci_device *pci_dev = eth_dev->pci_dev;
+	struct rte_pci_device *pci_dev;
+
+	pci_dev = ETH_DEV_PCI_DEV(eth_dev);
 
 	PMD_INIT_FUNC_TRACE();
 
@@ -1960,7 +1965,8 @@ ixgbe_check_vf_rss_rxq_num(struct rte_eth_dev *dev, uint16_t nb_rx_q)
 	}
 
 	RTE_ETH_DEV_SRIOV(dev).nb_q_per_pool = nb_rx_q;
-	RTE_ETH_DEV_SRIOV(dev).def_pool_q_idx = dev->pci_dev->max_vfs * nb_rx_q;
+	RTE_ETH_DEV_SRIOV(dev).def_pool_q_idx =
+		ETH_DEV_PCI_DEV(dev)->max_vfs * nb_rx_q;
 
 	return 0;
 }
@@ -2191,7 +2197,8 @@ ixgbe_dev_start(struct rte_eth_dev *dev)
 		IXGBE_DEV_PRIVATE_TO_HW(dev->data->dev_private);
 	struct ixgbe_vf_info *vfinfo =
 		*IXGBE_DEV_PRIVATE_TO_P_VFDATA(dev->data->dev_private);
-	struct rte_intr_handle *intr_handle = &dev->pci_dev->intr_handle;
+	struct rte_pci_device *pci_dev = ETH_DEV_PCI_DEV(dev);
+	struct rte_intr_handle *intr_handle = &pci_dev->intr_handle;
 	uint32_t intr_vector = 0;
 	int err, link_up = 0, negotiate = 0;
 	uint32_t speed = 0;
@@ -2291,7 +2298,7 @@ ixgbe_dev_start(struct rte_eth_dev *dev)
 
 	/* Restore vf rate limit */
 	if (vfinfo != NULL) {
-		for (vf = 0; vf < dev->pci_dev->max_vfs; vf++)
+		for (vf = 0; vf < pci_dev->max_vfs; vf++)
 			for (idx = 0; idx < IXGBE_MAX_QUEUE_NUM_PER_VF; idx++)
 				if (vfinfo[vf].tx_rate[idx] != 0)
 					ixgbe_set_vf_rate_limit(dev, vf,
@@ -2408,7 +2415,8 @@ ixgbe_dev_stop(struct rte_eth_dev *dev)
 	struct ixgbe_filter_info *filter_info =
 		IXGBE_DEV_PRIVATE_TO_FILTER_INFO(dev->data->dev_private);
 	struct ixgbe_5tuple_filter *p_5tuple, *p_5tuple_next;
-	struct rte_intr_handle *intr_handle = &dev->pci_dev->intr_handle;
+	struct rte_pci_device *pci_dev = ETH_DEV_PCI_DEV(dev);
+	struct rte_intr_handle *intr_handle = &pci_dev->intr_handle;
 	int vf;
 
 	PMD_INIT_FUNC_TRACE();
@@ -2424,7 +2432,7 @@ ixgbe_dev_stop(struct rte_eth_dev *dev)
 	ixgbe_stop_adapter(hw);
 
 	for (vf = 0; vfinfo != NULL &&
-		     vf < dev->pci_dev->max_vfs; vf++)
+		     vf < pci_dev->max_vfs; vf++)
 		vfinfo[vf].clear_to_send = false;
 
 	if (hw->mac.ops.get_media_type(hw) == ixgbe_media_type_copper) {
@@ -3033,6 +3041,7 @@ ixgbe_dev_info_get(struct rte_eth_dev *dev, struct rte_eth_dev_info *dev_info)
 {
 	struct ixgbe_hw *hw = IXGBE_DEV_PRIVATE_TO_HW(dev->data->dev_private);
 	struct rte_eth_conf *dev_conf = &dev->data->dev_conf;
+	struct rte_pci_device *pci_dev = ETH_DEV_PCI_DEV(dev);
 
 	dev_info->max_rx_queues = (uint16_t)hw->mac.max_rx_queues;
 	dev_info->max_tx_queues = (uint16_t)hw->mac.max_tx_queues;
@@ -3049,7 +3058,7 @@ ixgbe_dev_info_get(struct rte_eth_dev *dev, struct rte_eth_dev_info *dev_info)
 	dev_info->max_rx_pktlen = 15872; /* includes CRC, cf MAXFRS register */
 	dev_info->max_mac_addrs = hw->mac.num_rar_entries;
 	dev_info->max_hash_mac_addrs = IXGBE_VMDQ_NUM_UC_MAC;
-	dev_info->max_vfs = dev->pci_dev->max_vfs;
+	dev_info->max_vfs = pci_dev->max_vfs;
 	if (hw->mac.type == ixgbe_mac_82598EB)
 		dev_info->max_vmdq_pools = ETH_16_POOLS;
 	else
@@ -3164,6 +3173,7 @@ ixgbevf_dev_info_get(struct rte_eth_dev *dev,
 		     struct rte_eth_dev_info *dev_info)
 {
 	struct ixgbe_hw *hw = IXGBE_DEV_PRIVATE_TO_HW(dev->data->dev_private);
+	struct rte_pci_device *pci_dev = ETH_DEV_PCI_DEV(dev);
 
 	dev_info->max_rx_queues = (uint16_t)hw->mac.max_rx_queues;
 	dev_info->max_tx_queues = (uint16_t)hw->mac.max_tx_queues;
@@ -3171,7 +3181,7 @@ ixgbevf_dev_info_get(struct rte_eth_dev *dev,
 	dev_info->max_rx_pktlen = 15872; /* includes CRC, cf MAXFRS reg */
 	dev_info->max_mac_addrs = hw->mac.num_rar_entries;
 	dev_info->max_hash_mac_addrs = IXGBE_VMDQ_NUM_UC_MAC;
-	dev_info->max_vfs = dev->pci_dev->max_vfs;
+	dev_info->max_vfs = pci_dev->max_vfs;
 	if (hw->mac.type == ixgbe_mac_82598EB)
 		dev_info->max_vmdq_pools = ETH_16_POOLS;
 	else
@@ -3434,6 +3444,7 @@ static void
 ixgbe_dev_link_status_print(struct rte_eth_dev *dev)
 {
 	struct rte_eth_link link;
+	struct rte_pci_device *pci_dev = ETH_DEV_PCI_DEV(dev);
 
 	memset(&link, 0, sizeof(link));
 	rte_ixgbe_dev_atomic_read_link_status(dev, &link);
@@ -3448,10 +3459,10 @@ ixgbe_dev_link_status_print(struct rte_eth_dev *dev)
 				(int)(dev->data->port_id));
 	}
 	PMD_INIT_LOG(DEBUG, "PCI Address: " PCI_PRI_FMT,
-				dev->pci_dev->addr.domain,
-				dev->pci_dev->addr.bus,
-				dev->pci_dev->addr.devid,
-				dev->pci_dev->addr.function);
+				pci_dev->addr.domain,
+				pci_dev->addr.bus,
+				pci_dev->addr.devid,
+				pci_dev->addr.function);
 }
 
 /*
@@ -3515,7 +3526,7 @@ ixgbe_dev_interrupt_action(struct rte_eth_dev *dev)
 	} else {
 		PMD_DRV_LOG(DEBUG, "enable intr immediately");
 		ixgbe_enable_intr(dev);
-		rte_intr_enable(&(dev->pci_dev->intr_handle));
+		rte_intr_enable(ETH_DEV_TO_INTR_HANDLE(dev));
 	}
 
 
@@ -3564,7 +3575,7 @@ ixgbe_dev_interrupt_delayed_handler(void *param)
 
 	PMD_DRV_LOG(DEBUG, "enable intr in delayed handler S[%08x]", eicr);
 	ixgbe_enable_intr(dev);
-	rte_intr_enable(&(dev->pci_dev->intr_handle));
+	rte_intr_enable(ETH_DEV_TO_INTR_HANDLE(dev));
 }
 
 /**
@@ -4196,7 +4207,7 @@ ixgbevf_dev_start(struct rte_eth_dev *dev)
 	struct ixgbe_hw *hw =
 		IXGBE_DEV_PRIVATE_TO_HW(dev->data->dev_private);
 	uint32_t intr_vector = 0;
-	struct rte_intr_handle *intr_handle = &dev->pci_dev->intr_handle;
+	struct rte_intr_handle *intr_handle = ETH_DEV_TO_INTR_HANDLE(dev);
 
 	int err, mask = 0;
 
@@ -4259,7 +4270,7 @@ static void
 ixgbevf_dev_stop(struct rte_eth_dev *dev)
 {
 	struct ixgbe_hw *hw = IXGBE_DEV_PRIVATE_TO_HW(dev->data->dev_private);
-	struct rte_intr_handle *intr_handle = &dev->pci_dev->intr_handle;
+	struct rte_intr_handle *intr_handle = ETH_DEV_TO_INTR_HANDLE(dev);
 
 	PMD_INIT_FUNC_TRACE();
 
@@ -5070,7 +5081,7 @@ ixgbevf_dev_rx_queue_intr_enable(struct rte_eth_dev *dev, uint16_t queue_id)
 	RTE_SET_USED(queue_id);
 	IXGBE_WRITE_REG(hw, IXGBE_VTEIMS, mask);
 
-	rte_intr_enable(&dev->pci_dev->intr_handle);
+	rte_intr_enable(ETH_DEV_TO_INTR_HANDLE(dev));
 
 	return 0;
 }
@@ -5112,7 +5123,7 @@ ixgbe_dev_rx_queue_intr_enable(struct rte_eth_dev *dev, uint16_t queue_id)
 		mask &= (1 << (queue_id - 32));
 		IXGBE_WRITE_REG(hw, IXGBE_EIMS_EX(1), mask);
 	}
-	rte_intr_enable(&dev->pci_dev->intr_handle);
+	rte_intr_enable(ETH_DEV_TO_INTR_HANDLE(dev));
 
 	return 0;
 }
@@ -5216,7 +5227,7 @@ ixgbe_set_ivar_map(struct ixgbe_hw *hw, int8_t direction,
 static void
 ixgbevf_configure_msix(struct rte_eth_dev *dev)
 {
-	struct rte_intr_handle *intr_handle = &dev->pci_dev->intr_handle;
+	struct rte_intr_handle *intr_handle = ETH_DEV_TO_INTR_HANDLE(dev);
 	struct ixgbe_hw *hw =
 		IXGBE_DEV_PRIVATE_TO_HW(dev->data->dev_private);
 	uint32_t q_idx;
@@ -5249,7 +5260,7 @@ ixgbevf_configure_msix(struct rte_eth_dev *dev)
 static void
 ixgbe_configure_msix(struct rte_eth_dev *dev)
 {
-	struct rte_intr_handle *intr_handle = &dev->pci_dev->intr_handle;
+	struct rte_intr_handle *intr_handle = ETH_DEV_TO_INTR_HANDLE(dev);
 	struct ixgbe_hw *hw =
 		IXGBE_DEV_PRIVATE_TO_HW(dev->data->dev_private);
 	uint32_t queue_id, base = IXGBE_MISC_VEC_ID;
@@ -5381,7 +5392,8 @@ static int ixgbe_set_vf_rate_limit(struct rte_eth_dev *dev, uint16_t vf,
 		return -EINVAL;
 
 	if (vfinfo != NULL) {
-		for (vf_idx = 0; vf_idx < dev->pci_dev->max_vfs; vf_idx++) {
+		for (vf_idx = 0; vf_idx < ETH_DEV_PCI_DEV(dev)->max_vfs;
+		     vf_idx++) {
 			if (vf_idx == vf)
 				continue;
 			for (idx = 0; idx < RTE_DIM(vfinfo[vf_idx].tx_rate);
@@ -7197,12 +7209,13 @@ ixgbe_e_tag_insertion_en_dis(struct rte_eth_dev *dev,
 	int ret = 0;
 	uint32_t vmtir, vmvir;
 	struct ixgbe_hw *hw = IXGBE_DEV_PRIVATE_TO_HW(dev->data->dev_private);
+	struct rte_pci_device *pci_dev = ETH_DEV_PCI_DEV(dev);
 
-	if (l2_tunnel->vf_id >= dev->pci_dev->max_vfs) {
+	if (l2_tunnel->vf_id >= pci_dev->max_vfs) {
 		PMD_DRV_LOG(ERR,
 			    "VF id %u should be less than %u",
 			    l2_tunnel->vf_id,
-			    dev->pci_dev->max_vfs);
+			    pci_dev->max_vfs);
 		return -EINVAL;
 	}
 
diff --git a/drivers/net/ixgbe/ixgbe_pf.c b/drivers/net/ixgbe/ixgbe_pf.c
index 26395e4..139d816 100644
--- a/drivers/net/ixgbe/ixgbe_pf.c
+++ b/drivers/net/ixgbe/ixgbe_pf.c
@@ -61,7 +61,7 @@
 static inline uint16_t
 dev_num_vf(struct rte_eth_dev *eth_dev)
 {
-	return eth_dev->pci_dev->max_vfs;
+	return ETH_DEV_PCI_DEV(eth_dev)->max_vfs;
 }
 
 static inline
diff --git a/drivers/net/nfp/nfp_net.c b/drivers/net/nfp/nfp_net.c
index c6b1587..7299fad 100644
--- a/drivers/net/nfp/nfp_net.c
+++ b/drivers/net/nfp/nfp_net.c
@@ -718,10 +718,12 @@ static void
 nfp_net_close(struct rte_eth_dev *dev)
 {
 	struct nfp_net_hw *hw;
+	struct rte_pci_device *pci_dev;
 
 	PMD_INIT_LOG(DEBUG, "Close\n");
 
 	hw = NFP_NET_DEV_PRIVATE_TO_HW(dev->data->dev_private);
+	pci_dev = ETH_DEV_PCI_DEV(dev);
 
 	/*
 	 * We assume that the DPDK application is stopping all the
@@ -730,11 +732,11 @@ nfp_net_close(struct rte_eth_dev *dev)
 
 	nfp_net_stop(dev);
 
-	rte_intr_disable(&dev->pci_dev->intr_handle);
+	rte_intr_disable(&pci_dev->intr_handle);
 	nn_cfg_writeb(hw, NFP_NET_CFG_LSC, 0xff);
 
 	/* unregister callback func from eal lib */
-	rte_intr_callback_unregister(&dev->pci_dev->intr_handle,
+	rte_intr_callback_unregister(&pci_dev->intr_handle,
 				     nfp_net_dev_interrupt_handler,
 				     (void *)dev);
 
@@ -1122,6 +1124,7 @@ nfp_net_rx_queue_count(struct rte_eth_dev *dev, uint16_t queue_idx)
 static void
 nfp_net_dev_link_status_print(struct rte_eth_dev *dev)
 {
+	struct rte_pci_device *pci_dev = ETH_DEV_PCI_DEV(dev);
 	struct rte_eth_link link;
 
 	memset(&link, 0, sizeof(link));
@@ -1136,8 +1139,8 @@ nfp_net_dev_link_status_print(struct rte_eth_dev *dev)
 			(int)(dev->data->port_id));
 
 	RTE_LOG(INFO, PMD, "PCI Address: %04d:%02d:%02d:%d\n",
-		dev->pci_dev->addr.domain, dev->pci_dev->addr.bus,
-		dev->pci_dev->addr.devid, dev->pci_dev->addr.function);
+		pci_dev->addr.domain, pci_dev->addr.bus,
+		pci_dev->addr.devid, pci_dev->addr.function);
 }
 
 /* Interrupt configuration and handling */
@@ -1158,7 +1161,7 @@ nfp_net_irq_unmask(struct rte_eth_dev *dev)
 	if (hw->ctrl & NFP_NET_CFG_CTRL_MSIXAUTO) {
 		/* If MSI-X auto-masking is used, clear the entry */
 		rte_wmb();
-		rte_intr_enable(&dev->pci_dev->intr_handle);
+		rte_intr_enable(&(ETH_DEV_PCI_DEV(dev)->intr_handle));
 	} else {
 		/* Make sure all updates are written before un-masking */
 		rte_wmb();
@@ -2330,7 +2333,7 @@ nfp_net_init(struct rte_eth_dev *eth_dev)
 	if (rte_eal_process_type() != RTE_PROC_PRIMARY)
 		return 0;
 
-	pci_dev = eth_dev->pci_dev;
+	pci_dev = ETH_DEV_PCI_DEV(eth_dev);
 	rte_eth_copy_pci_info(eth_dev, pci_dev);
 
 	hw->device_id = pci_dev->id.device_id;
@@ -2400,7 +2403,6 @@ nfp_net_init(struct rte_eth_dev *eth_dev)
 		     hw->cap & NFP_NET_CFG_CTRL_LSO     ? "TSO "     : "",
 		     hw->cap & NFP_NET_CFG_CTRL_RSS     ? "RSS "     : "");
 
-	pci_dev = eth_dev->pci_dev;
 	hw->ctrl = 0;
 
 	hw->stride_rx = stride;
diff --git a/drivers/net/qede/qede_ethdev.c b/drivers/net/qede/qede_ethdev.c
index d106dd0..959ff0f 100644
--- a/drivers/net/qede/qede_ethdev.c
+++ b/drivers/net/qede/qede_ethdev.c
@@ -178,11 +178,12 @@ static void
 qede_interrupt_handler(__rte_unused struct rte_intr_handle *handle, void *param)
 {
 	struct rte_eth_dev *eth_dev = (struct rte_eth_dev *)param;
+	struct rte_pci_device *pci_dev = ETH_DEV_PCI_DEV(eth_dev);
 	struct qede_dev *qdev = eth_dev->data->dev_private;
 	struct ecore_dev *edev = &qdev->edev;
 
 	qede_interrupt_action(ECORE_LEADING_HWFN(edev));
-	if (rte_intr_enable(&eth_dev->pci_dev->intr_handle))
+	if (rte_intr_enable(&pci_dev->intr_handle))
 		DP_ERR(edev, "rte_intr_enable failed\n");
 }
 
@@ -809,6 +810,7 @@ static void qede_poll_sp_sb_cb(void *param)
 
 static void qede_dev_close(struct rte_eth_dev *eth_dev)
 {
+	struct rte_pci_device *pci_dev = ETH_DEV_PCI_DEV(eth_dev);
 	struct qede_dev *qdev = QEDE_INIT_QDEV(eth_dev);
 	struct ecore_dev *edev = QEDE_INIT_EDEV(qdev);
 	int rc;
@@ -835,9 +837,9 @@ static void qede_dev_close(struct rte_eth_dev *eth_dev)
 
 	qdev->ops->common->remove(edev);
 
-	rte_intr_disable(&eth_dev->pci_dev->intr_handle);
+	rte_intr_disable(&pci_dev->intr_handle);
 
-	rte_intr_callback_unregister(&eth_dev->pci_dev->intr_handle,
+	rte_intr_callback_unregister(&pci_dev->intr_handle,
 				     qede_interrupt_handler, (void *)eth_dev);
 
 	if (edev->num_hwfns > 1)
@@ -1403,7 +1405,8 @@ static int qede_common_dev_init(struct rte_eth_dev *eth_dev, bool is_vf)
 	/* Extract key data structures */
 	adapter = eth_dev->data->dev_private;
 	edev = &adapter->edev;
-	pci_addr = eth_dev->pci_dev->addr;
+	pci_dev = ETH_DEV_PCI_DEV(eth_dev);
+	pci_addr = pci_dev->addr;
 
 	PMD_INIT_FUNC_TRACE(edev);
 
@@ -1420,8 +1423,6 @@ static int qede_common_dev_init(struct rte_eth_dev *eth_dev, bool is_vf)
 		return 0;
 	}
 
-	pci_dev = eth_dev->pci_dev;
-
 	rte_eth_copy_pci_info(eth_dev, pci_dev);
 
 	qed_ops = qed_get_eth_ops();
@@ -1442,10 +1443,10 @@ static int qede_common_dev_init(struct rte_eth_dev *eth_dev, bool is_vf)
 
 	qede_update_pf_params(edev);
 
-	rte_intr_callback_register(&eth_dev->pci_dev->intr_handle,
+	rte_intr_callback_register(&pci_dev->intr_handle,
 				   qede_interrupt_handler, (void *)eth_dev);
 
-	if (rte_intr_enable(&eth_dev->pci_dev->intr_handle)) {
+	if (rte_intr_enable(&pci_dev->intr_handle)) {
 		DP_ERR(edev, "rte_intr_enable() failed\n");
 		return -ENODEV;
 	}
diff --git a/drivers/net/szedata2/rte_eth_szedata2.c b/drivers/net/szedata2/rte_eth_szedata2.c
index f3cd52d..6d80a5b 100644
--- a/drivers/net/szedata2/rte_eth_szedata2.c
+++ b/drivers/net/szedata2/rte_eth_szedata2.c
@@ -1357,7 +1357,7 @@ get_szedata2_index(struct rte_eth_dev *dev, uint32_t *index)
 	uint32_t tmp_index;
 	FILE *fd;
 	char pcislot_path[PATH_MAX];
-	struct rte_pci_addr pcislot_addr = dev->pci_dev->addr;
+	struct rte_pci_addr pcislot_addr = ETH_DEV_PCI_DEV(dev)->addr;
 	uint32_t domain;
 	uint32_t bus;
 	uint32_t devid;
@@ -1415,9 +1415,10 @@ rte_szedata2_eth_dev_init(struct rte_eth_dev *dev)
 	struct szedata *szedata_temp;
 	int ret;
 	uint32_t szedata2_index;
-	struct rte_pci_addr *pci_addr = &dev->pci_dev->addr;
+	struct rte_pci_device *pci_dev = ETH_DEV_PCI_DEV(dev);
+	struct rte_pci_addr *pci_addr = &pci_dev->addr;
 	struct rte_mem_resource *pci_rsc =
-		&dev->pci_dev->mem_resource[PCI_RESOURCE_NUMBER];
+		&pci_dev->mem_resource[PCI_RESOURCE_NUMBER];
 	char rsc_filename[PATH_MAX];
 	void *pci_resource_ptr = NULL;
 	int fd;
@@ -1471,10 +1472,10 @@ rte_szedata2_eth_dev_init(struct rte_eth_dev *dev)
 	/* Set function callbacks for Ethernet API */
 	dev->dev_ops = &ops;
 
-	rte_eth_copy_pci_info(dev, dev->pci_dev);
+	rte_eth_copy_pci_info(dev, pci_dev);
 
 	/* mmap pci resource0 file to rte_mem_resource structure */
-	if (dev->pci_dev->mem_resource[PCI_RESOURCE_NUMBER].phys_addr ==
+	if (pci_dev->mem_resource[PCI_RESOURCE_NUMBER].phys_addr ==
 			0) {
 		RTE_LOG(ERR, PMD, "Missing resource%u file\n",
 				PCI_RESOURCE_NUMBER);
@@ -1491,7 +1492,7 @@ rte_szedata2_eth_dev_init(struct rte_eth_dev *dev)
 	}
 
 	pci_resource_ptr = mmap(0,
-			dev->pci_dev->mem_resource[PCI_RESOURCE_NUMBER].len,
+			pci_dev->mem_resource[PCI_RESOURCE_NUMBER].len,
 			PROT_READ | PROT_WRITE, MAP_SHARED, fd, 0);
 	close(fd);
 	if (pci_resource_ptr == NULL) {
@@ -1499,8 +1500,7 @@ rte_szedata2_eth_dev_init(struct rte_eth_dev *dev)
 				rsc_filename, fd);
 		return -EINVAL;
 	}
-	dev->pci_dev->mem_resource[PCI_RESOURCE_NUMBER].addr =
-		pci_resource_ptr;
+	pci_dev->mem_resource[PCI_RESOURCE_NUMBER].addr = pci_resource_ptr;
 
 	RTE_LOG(DEBUG, PMD, "resource%u phys_addr = 0x%llx len = %llu "
 			"virt addr = %llx\n", PCI_RESOURCE_NUMBER,
@@ -1516,8 +1516,8 @@ rte_szedata2_eth_dev_init(struct rte_eth_dev *dev)
 			RTE_CACHE_LINE_SIZE);
 	if (data->mac_addrs == NULL) {
 		RTE_LOG(ERR, PMD, "Could not alloc space for MAC address!\n");
-		munmap(dev->pci_dev->mem_resource[PCI_RESOURCE_NUMBER].addr,
-			dev->pci_dev->mem_resource[PCI_RESOURCE_NUMBER].len);
+		munmap(pci_dev->mem_resource[PCI_RESOURCE_NUMBER].addr,
+		       pci_dev->mem_resource[PCI_RESOURCE_NUMBER].len);
 		return -EINVAL;
 	}
 
@@ -1537,12 +1537,13 @@ rte_szedata2_eth_dev_init(struct rte_eth_dev *dev)
 static int
 rte_szedata2_eth_dev_uninit(struct rte_eth_dev *dev)
 {
-	struct rte_pci_addr *pci_addr = &dev->pci_dev->addr;
+	struct rte_pci_device *pci_dev = ETH_DEV_PCI_DEV(dev);
+	struct rte_pci_addr *pci_addr = &pci_dev->addr;
 
 	rte_free(dev->data->mac_addrs);
 	dev->data->mac_addrs = NULL;
-	munmap(dev->pci_dev->mem_resource[PCI_RESOURCE_NUMBER].addr,
-		dev->pci_dev->mem_resource[PCI_RESOURCE_NUMBER].len);
+	munmap(pci_dev->mem_resource[PCI_RESOURCE_NUMBER].addr,
+	       pci_dev->mem_resource[PCI_RESOURCE_NUMBER].len);
 
 	RTE_LOG(INFO, PMD, "szedata2 device ("
 			PCI_PRI_FMT ") successfully uninitialized\n",
diff --git a/drivers/net/szedata2/rte_eth_szedata2.h b/drivers/net/szedata2/rte_eth_szedata2.h
index 522cf47..1c28c42 100644
--- a/drivers/net/szedata2/rte_eth_szedata2.h
+++ b/drivers/net/szedata2/rte_eth_szedata2.h
@@ -120,8 +120,8 @@ static inline uint8_t
 pci_resource_read8(struct rte_eth_dev *dev, uint32_t offset)
 {
 	return *((uint8_t *)((uint8_t *)
-			dev->pci_dev->mem_resource[PCI_RESOURCE_NUMBER].addr +
-			offset));
+		ETH_DEV_PCI_DEV(dev)->mem_resource[PCI_RESOURCE_NUMBER].addr +
+		offset));
 }
 
 /*
@@ -131,8 +131,8 @@ static inline uint16_t
 pci_resource_read16(struct rte_eth_dev *dev, uint32_t offset)
 {
 	return rte_le_to_cpu_16(*((uint16_t *)((uint8_t *)
-			dev->pci_dev->mem_resource[PCI_RESOURCE_NUMBER].addr +
-			offset)));
+		ETH_DEV_PCI_DEV(dev)->mem_resource[PCI_RESOURCE_NUMBER].addr +
+		offset)));
 }
 
 /*
@@ -142,8 +142,8 @@ static inline uint32_t
 pci_resource_read32(struct rte_eth_dev *dev, uint32_t offset)
 {
 	return rte_le_to_cpu_32(*((uint32_t *)((uint8_t *)
-			dev->pci_dev->mem_resource[PCI_RESOURCE_NUMBER].addr +
-			offset)));
+		ETH_DEV_PCI_DEV(dev)->mem_resource[PCI_RESOURCE_NUMBER].addr +
+		offset)));
 }
 
 /*
@@ -153,8 +153,8 @@ static inline uint64_t
 pci_resource_read64(struct rte_eth_dev *dev, uint32_t offset)
 {
 	return rte_le_to_cpu_64(*((uint64_t *)((uint8_t *)
-			dev->pci_dev->mem_resource[PCI_RESOURCE_NUMBER].addr +
-			offset)));
+		ETH_DEV_PCI_DEV(dev)->mem_resource[PCI_RESOURCE_NUMBER].addr +
+		offset)));
 }
 
 /*
@@ -164,8 +164,8 @@ static inline void
 pci_resource_write8(struct rte_eth_dev *dev, uint32_t offset, uint8_t val)
 {
 	*((uint8_t *)((uint8_t *)
-			dev->pci_dev->mem_resource[PCI_RESOURCE_NUMBER].addr +
-			offset)) = val;
+		ETH_DEV_PCI_DEV(dev)->mem_resource[PCI_RESOURCE_NUMBER].addr +
+		offset)) = val;
 }
 
 /*
@@ -175,8 +175,8 @@ static inline void
 pci_resource_write16(struct rte_eth_dev *dev, uint32_t offset, uint16_t val)
 {
 	*((uint16_t *)((uint8_t *)
-			dev->pci_dev->mem_resource[PCI_RESOURCE_NUMBER].addr +
-			offset)) = rte_cpu_to_le_16(val);
+		ETH_DEV_PCI_DEV(dev)->mem_resource[PCI_RESOURCE_NUMBER].addr +
+		offset)) = rte_cpu_to_le_16(val);
 }
 
 /*
@@ -186,8 +186,8 @@ static inline void
 pci_resource_write32(struct rte_eth_dev *dev, uint32_t offset, uint32_t val)
 {
 	*((uint32_t *)((uint8_t *)
-			dev->pci_dev->mem_resource[PCI_RESOURCE_NUMBER].addr +
-			offset)) = rte_cpu_to_le_32(val);
+		ETH_DEV_PCI_DEV(dev)->mem_resource[PCI_RESOURCE_NUMBER].addr +
+		offset)) = rte_cpu_to_le_32(val);
 }
 
 /*
@@ -197,13 +197,13 @@ static inline void
 pci_resource_write64(struct rte_eth_dev *dev, uint32_t offset, uint64_t val)
 {
 	*((uint64_t *)((uint8_t *)
-			dev->pci_dev->mem_resource[PCI_RESOURCE_NUMBER].addr +
-			offset)) = rte_cpu_to_le_64(val);
+		ETH_DEV_PCI_DEV(dev)->mem_resource[PCI_RESOURCE_NUMBER].addr +
+		offset)) = rte_cpu_to_le_64(val);
 }
 
 #define SZEDATA2_PCI_RESOURCE_PTR(dev, offset, type) \
 	((type)((uint8_t *) \
-	((dev)->pci_dev->mem_resource[PCI_RESOURCE_NUMBER].addr) \
+	(ETH_DEV_PCI_DEV(dev)->mem_resource[PCI_RESOURCE_NUMBER].addr) \
 	+ (offset)))
 
 enum szedata2_link_speed {
diff --git a/drivers/net/thunderx/nicvf_ethdev.c b/drivers/net/thunderx/nicvf_ethdev.c
index 466e49c..099b252 100644
--- a/drivers/net/thunderx/nicvf_ethdev.c
+++ b/drivers/net/thunderx/nicvf_ethdev.c
@@ -1335,6 +1335,7 @@ static void
 nicvf_dev_info_get(struct rte_eth_dev *dev, struct rte_eth_dev_info *dev_info)
 {
 	struct nicvf *nic = nicvf_pmd_priv(dev);
+	struct rte_pci_device *pci_dev = ETH_DEV_PCI_DEV(dev);
 
 	PMD_INIT_FUNC_TRACE();
 
@@ -1345,7 +1346,7 @@ nicvf_dev_info_get(struct rte_eth_dev *dev, struct rte_eth_dev_info *dev_info)
 	dev_info->max_tx_queues =
 			(uint16_t)MAX_SND_QUEUES_PER_QS * (MAX_SQS_PER_VF + 1);
 	dev_info->max_mac_addrs = 1;
-	dev_info->max_vfs = dev->pci_dev->max_vfs;
+	dev_info->max_vfs = pci_dev->max_vfs;
 
 	dev_info->rx_offload_capa = DEV_RX_OFFLOAD_VLAN_STRIP;
 	dev_info->tx_offload_capa =
@@ -1975,7 +1976,7 @@ nicvf_eth_dev_init(struct rte_eth_dev *eth_dev)
 		}
 	}
 
-	pci_dev = eth_dev->pci_dev;
+	pci_dev = ETH_DEV_PCI_DEV(eth_dev);
 	rte_eth_copy_pci_info(eth_dev, pci_dev);
 
 	nic->device_id = pci_dev->id.device_id;
diff --git a/drivers/net/vmxnet3/vmxnet3_ethdev.c b/drivers/net/vmxnet3/vmxnet3_ethdev.c
index 8bb13e5..bcb3751 100644
--- a/drivers/net/vmxnet3/vmxnet3_ethdev.c
+++ b/drivers/net/vmxnet3/vmxnet3_ethdev.c
@@ -138,7 +138,7 @@ gpa_zone_reserve(struct rte_eth_dev *dev, uint32_t size,
 	const struct rte_memzone *mz;
 
 	snprintf(z_name, sizeof(z_name), "%s_%d_%s",
-		 dev->driver->pci_drv.driver.name, dev->data->port_id, post_string);
+		 dev->data->drv_name, dev->data->port_id, post_string);
 
 	mz = rte_memzone_lookup(z_name);
 	if (!reuse) {
@@ -237,7 +237,7 @@ eth_vmxnet3_dev_init(struct rte_eth_dev *eth_dev)
 	eth_dev->dev_ops = &vmxnet3_eth_dev_ops;
 	eth_dev->rx_pkt_burst = &vmxnet3_recv_pkts;
 	eth_dev->tx_pkt_burst = &vmxnet3_xmit_pkts;
-	pci_dev = eth_dev->pci_dev;
+	pci_dev = ETH_DEV_PCI_DEV(eth_dev);
 
 	/*
 	 * for secondary processes, we don't initialize any further as primary
-- 
2.7.4

^ permalink raw reply related

* [PATCH v3 2/9] ethdev: Helper to convert to struct rte_pci_device
From: Jan Blunck @ 2016-12-20 11:11 UTC (permalink / raw)
  To: dev; +Cc: shreyansh.jain, david.marchand, stephen
In-Reply-To: <1482232315-21626-1-git-send-email-jblunck@infradead.org>

Signed-off-by: Jan Blunck <jblunck@infradead.org>
Acked-by: Shreyansh Jain <shreyansh.jain@nxp.com>
---
 lib/librte_ether/rte_ethdev.h | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/lib/librte_ether/rte_ethdev.h b/lib/librte_ether/rte_ethdev.h
index 9678179..3adbb2b 100644
--- a/lib/librte_ether/rte_ethdev.h
+++ b/lib/librte_ether/rte_ethdev.h
@@ -1644,6 +1644,12 @@ struct rte_eth_dev {
 	uint8_t attached; /**< Flag indicating the port is attached */
 } __rte_cache_aligned;
 
+/**
+ * @internal
+ * Helper for drivers that need to convert from rte_eth_dev to rte_pci_device.
+ */
+#define ETH_DEV_PCI_DEV(ptr) ((ptr)->pci_dev)
+
 struct rte_eth_dev_sriov {
 	uint8_t active;               /**< SRIOV is active with 16, 32 or 64 pools */
 	uint8_t nb_q_per_pool;        /**< rx queue number per pool */
-- 
2.7.4

^ permalink raw reply related

* [PATCH v3 1/9] eal: define container_of macro
From: Jan Blunck @ 2016-12-20 11:11 UTC (permalink / raw)
  To: dev; +Cc: shreyansh.jain, david.marchand, stephen, Jan Viktorin
In-Reply-To: <1482232315-21626-1-git-send-email-jblunck@infradead.org>

This macro is based on Jan Viktorin's original patch but also checks the
type of the passed pointer against the type of the member.

Signed-off-by: Jan Viktorin <viktorin@rehivetech.com>
Signed-off-by: Shreyansh Jain <shreyansh.jain@nxp.com>
[jblunck@infradead.org: add type checking and __extension__]
Signed-off-by: Jan Blunck <jblunck@infradead.org>
---
 lib/librte_eal/common/include/rte_common.h | 20 ++++++++++++++++++++
 1 file changed, 20 insertions(+)

diff --git a/lib/librte_eal/common/include/rte_common.h b/lib/librte_eal/common/include/rte_common.h
index db5ac91..8dda3e2 100644
--- a/lib/librte_eal/common/include/rte_common.h
+++ b/lib/librte_eal/common/include/rte_common.h
@@ -331,6 +331,26 @@ rte_bsf32(uint32_t v)
 #define offsetof(TYPE, MEMBER)  __builtin_offsetof (TYPE, MEMBER)
 #endif
 
+/**
+ * Return pointer to the wrapping struct instance.
+ *
+ * Example:
+ *
+ *  struct wrapper {
+ *      ...
+ *      struct child c;
+ *      ...
+ *  };
+ *
+ *  struct child *x = obtain(...);
+ *  struct wrapper *w = container_of(x, struct wrapper, c);
+ */
+#ifndef container_of
+#define container_of(ptr, type, member)	__extension__ ({		\
+			typeof(((type *)0)->member) *_ptr = (ptr);	\
+			(type *)(((char *)_ptr) - offsetof(type, member)); })
+#endif
+
 #define _RTE_STR(x) #x
 /** Take a macro value and get a string version of it */
 #define RTE_STR(x) _RTE_STR(x)
-- 
2.7.4

^ permalink raw reply related

* [PATCH v3 0/9] Decouple ethdev from PCI device
From: Jan Blunck @ 2016-12-20 11:11 UTC (permalink / raw)
  To: dev; +Cc: shreyansh.jain, david.marchand, stephen

This is a repost of the series I sent in November. I've addressed Shreyansh's
review comments about the PCI device users I've missed.

Jan Blunck (9):
  eal: define container_of macro
  ethdev: Helper to convert to struct rte_pci_device
  drivers: Use ETH_DEV_PCI_DEV() helper
  virtio: Don't fill dev_info->driver_name
  virtio: Add vtpci_intr_handle() helper to get rte_intr_handle
  virtio: Don't depend on struct rte_eth_dev's pci_dev
  ethdev: Move filling of rte_eth_dev_info->pci_dev to dev_infos_get()
  ethdev: Decouple interrupt handling from PCI device
  ethdev: Decouple struct rte_eth_dev from struct rte_pci_device

 app/test/virtual_pmd.c                     |  4 +-
 drivers/net/bnx2x/bnx2x_ethdev.c           | 13 +++---
 drivers/net/bnxt/bnxt_ethdev.c             | 21 ++++++----
 drivers/net/bnxt/bnxt_ring.c               | 11 ++---
 drivers/net/bonding/rte_eth_bond_args.c    | 12 +++++-
 drivers/net/cxgbe/cxgbe_ethdev.c           |  4 +-
 drivers/net/cxgbe/cxgbe_main.c             |  4 +-
 drivers/net/e1000/em_ethdev.c              | 21 ++++++----
 drivers/net/e1000/igb_ethdev.c             | 52 +++++++++++++----------
 drivers/net/e1000/igb_pf.c                 |  3 +-
 drivers/net/ena/ena_ethdev.c               |  4 +-
 drivers/net/enic/enic_ethdev.c             |  3 +-
 drivers/net/fm10k/fm10k_ethdev.c           | 56 +++++++++++++------------
 drivers/net/i40e/i40e_ethdev.c             | 45 ++++++++++----------
 drivers/net/i40e/i40e_ethdev.h             |  4 ++
 drivers/net/i40e/i40e_ethdev_vf.c          | 39 ++++++++---------
 drivers/net/ixgbe/ixgbe_ethdev.c           | 67 ++++++++++++++++++------------
 drivers/net/ixgbe/ixgbe_pf.c               |  2 +-
 drivers/net/mlx4/mlx4.c                    |  2 +
 drivers/net/mlx5/mlx5_ethdev.c             |  2 +
 drivers/net/nfp/nfp_net.c                  | 17 ++++----
 drivers/net/qede/qede_ethdev.c             | 18 ++++----
 drivers/net/szedata2/rte_eth_szedata2.c    | 28 +++++++------
 drivers/net/szedata2/rte_eth_szedata2.h    | 34 +++++++--------
 drivers/net/thunderx/nicvf_ethdev.c        |  7 +++-
 drivers/net/virtio/virtio_ethdev.c         | 38 ++++++++---------
 drivers/net/virtio/virtio_pci.h            |  6 +++
 drivers/net/virtio/virtio_user_ethdev.c    |  1 -
 drivers/net/vmxnet3/vmxnet3_ethdev.c       |  6 ++-
 lib/librte_eal/common/include/rte_common.h | 20 +++++++++
 lib/librte_eal/common/include/rte_pci.h    |  6 +++
 lib/librte_ether/rte_ethdev.c              | 25 ++++++++---
 lib/librte_ether/rte_ethdev.h              |  9 +++-
 33 files changed, 355 insertions(+), 229 deletions(-)

-- 
2.7.4

^ permalink raw reply

* Re: [PATCH] nfp: extend speed capabilities advertised
From: Alejandro Lucero @ 2016-12-20 11:02 UTC (permalink / raw)
  To: Ferruh Yigit; +Cc: dev
In-Reply-To: <CAD+H992=T2bgGiNhC-Lcir8a8pjdwpd1icvtJVif1_oZv4Q_sw@mail.gmail.com>

On Tue, Dec 20, 2016 at 10:29 AM, Alejandro Lucero <
alejandro.lucero@netronome.com> wrote:

>
>
> On Tue, Dec 20, 2016 at 10:25 AM, Ferruh Yigit <ferruh.yigit@intel.com>
> wrote:
>
>> On 12/19/2016 6:00 PM, Alejandro Lucero wrote:
>> > I forgot one thing: to update the features file with this new one.
>> >
>> > I will wait for your feedback regarding the discussed problem for
>> > sending another version.
>>
>> I think it is good to go, please send updated version.
>>
>>
> OK
>
>
>> >
>>
>> <...>
>>
>> >
>> >         Sorry, confused. Is it like following:
>> >
>> >         "
>> >         For new FW, there is no problem, it supports the range between
>> >         1G - 50G,
>> >         and reports correct current speed.
>> >
>> >         With old FW, device still can be set to speed range between 1G -
>> >         50G,
>> >         but it doesn't report current speed correct, and DPDK driver
>> reports
>> >         back to application as device current speed is 40G, without
>> really
>> >         knowing the current speed.
>> >         "
>> >
>> >
>> >     Yes, that is. Should then I do anything else or the patch is right
>> >     for you now?
>>
>> So, this patch is correct.
>>
>> But for the previous patch "net/nfp: report link speed using hardware
>> info", it would be nice to add above information into source as comment
>> and into commit log.
>> If you can, would you mind sending a new version of that patch?
>>
>>
>
Would it be better to report ETH_SPEED_NUM_NONE instead of
ETH_SPEED_NUM_40G for old firmware?
I will add a comment about why, but I think this is better for the user not
getting (sometimes) the right speed.


> I'll do it.
>
> Thanks!
>
>
>> Thanks,
>> ferruh
>>
>> >
>> <...>
>>
>
>

^ permalink raw reply

* Re: [PATCH v3 25/25] doc: describe testpmd flow command
From: Adrien Mazarguil @ 2016-12-20 10:51 UTC (permalink / raw)
  To: Mcnamara, John; +Cc: dev@dpdk.org, Kevin Traynor
In-Reply-To: <B27915DBBA3421428155699D51E4CFE2026896B8@IRSMSX103.ger.corp.intel.com>

On Mon, Dec 19, 2016 at 08:44:07PM +0000, Mcnamara, John wrote:
> 
> 
> > -----Original Message-----
> > From: dev [mailto:dev-bounces@dpdk.org] On Behalf Of Adrien Mazarguil
> > Sent: Monday, December 19, 2016 5:49 PM
> > To: dev@dpdk.org
> > Subject: [dpdk-dev] [PATCH v3 25/25] doc: describe testpmd flow command
> > 
> > Document syntax, interaction with rte_flow and provide usage examples.
> > 
> > Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
> > 
> > ...
> >
> > +
> > +- Check whether a flow rule can be created::
> > +
> > +   flow validate {port_id}
> > +       [group {group_id}] [priority {level}] [ingress] [egress]
> > +       pattern {item} [/ {item} [...]] / end
> > +       actions {action} [/ {action} [...]] / end
> > +
> > +- Create a flow rule::
> > +
> > +   flow create {port_id}
> > +       [group {group_id}] [priority {level}] [ingress] [egress]
> > +       pattern {item} [/ {item} [...]] / end
> > +       actions {action} [/ {action} [...]] / end
> > +
> > +- Destroy specific flow rules::
> > +
> > +   flow destroy {port_id} rule {rule_id} [...]
> > +
> > +- Destroy all flow rules::
> > +
> > +   flow flush {port_id}
> > +
> 
> Just a note:
> 
> The verbs destroy and flush don't sound right here. Create/destroy are common
> verbs pairs for objects but these actions are more like add/remove. I guess the
> names come from the underlying APIs which possibly are creating/freeing
> objects/structures but maybe they should be called add/remove as well.
> 
> And flush generally applies to a pipeline or a queue. The action here is closer
> to "remove all".
> 
> Probably not worth reworking at this stage if it hasn't bothered anyone else.

Well, Kevin Traynor made a similar suggestion to which I replied that the
name would be modified if enough people complained [1].

I understand your point but for some reason I keep hearing a flushing noise
every time all rules are removed at once, hence the name.

Problem is also that we now have 3 PMD series floating on the ML that depend
on the current definition. If we decided to change it, I suggest doing so in
a separate fix. A few more complaints from developers are needed before it's
too late for 17.02.

> > +underlying device in its current state but stops short of creating it.
> > +It is bound to ``rte_flow_validate()``::
> > +
> > + flow validate {port_id}
> > +     [group {group_id}] [priority {level}] [ingress] [egress]
> > +     pattern {item} [/ {item} [...]] / end
> > +     actions {action} [/ {action} [...]] / end
> > +
> 
> Here and elsewhere the indentation should be the RST standard 3 spaces,
> similar to the rest of the doc. This is only worth changing if you
> do some other revision of this doc.
> 
> Otherwise very good documentation.
> 
> Acked-by: John McNamara <john.mcnamara@intel.com>

Thanks, I'll make those changes if anything else warrants a v4.

[1] http://dpdk.org/ml/archives/dev/2016-December/050973.html

-- 
Adrien Mazarguil
6WIND

^ permalink raw reply

* Re: [PATCH v2 8/8] ethdev: Decouple interrupt handling from PCI device
From: Jan Blunck @ 2016-12-20 10:51 UTC (permalink / raw)
  To: Shreyansh Jain; +Cc: dev, David Marchand
In-Reply-To: <84460a6f-3574-8da2-3ccc-6f1e23fb50e7@nxp.com>

On Tue, Nov 22, 2016 at 1:57 PM, Shreyansh Jain <shreyansh.jain@nxp.com> wrote:
> On Monday 21 November 2016 10:25 PM, Jan Blunck wrote:
>>
>> The struct rte_intr_handle is an abstraction layer for different types of
>> interrupt mechanisms. It is embedded in the low-level device (e.g. PCI).
>> On allocation of a struct rte_eth_dev a reference to the intr_handle
>> should be stored for devices supporting interrupts.
>>
>> Signed-off-by: Jan Blunck <jblunck@infradead.org>
>> ---
>>  lib/librte_ether/rte_ethdev.c | 18 ++++++++++++++++--
>>  lib/librte_ether/rte_ethdev.h |  1 +
>>  2 files changed, 17 insertions(+), 2 deletions(-)
>>
>> diff --git a/lib/librte_ether/rte_ethdev.c b/lib/librte_ether/rte_ethdev.c
>> index 4288577..4ecea50 100644
>> --- a/lib/librte_ether/rte_ethdev.c
>> +++ b/lib/librte_ether/rte_ethdev.c
>> @@ -258,6 +258,7 @@ rte_eth_dev_pci_probe(struct rte_pci_driver *pci_drv,
>>                         rte_panic("Cannot allocate memzone for private
>> port data\n");
>>         }
>>         eth_dev->pci_dev = pci_dev;
>> +       eth_dev->intr_handle = &pci_dev->intr_handle;
>>         eth_dev->driver = eth_drv;
>>         eth_dev->data->rx_mbuf_alloc_failed = 0;
>>
>> @@ -2543,7 +2544,13 @@ rte_eth_dev_rx_intr_ctl(uint8_t port_id, int epfd,
>> int op, void *data)
>>         RTE_ETH_VALID_PORTID_OR_ERR_RET(port_id, -ENODEV);
>>
>>         dev = &rte_eth_devices[port_id];
>> -       intr_handle = &dev->pci_dev->intr_handle;
>> +
>> +       if (!dev->intr_handle) {
>> +               RTE_PMD_DEBUG_TRACE("RX Intr handle unset\n");
>> +               return -ENOTSUP;
>> +       }
>> +
>> +       intr_handle = dev->intr_handle;
>>         if (!intr_handle->intr_vec) {
>>                 RTE_PMD_DEBUG_TRACE("RX Intr vector unset\n");
>>                 return -EPERM;
>> @@ -2603,7 +2610,12 @@ rte_eth_dev_rx_intr_ctl_q(uint8_t port_id, uint16_t
>> queue_id,
>>                 return -EINVAL;
>>         }
>>
>> -       intr_handle = &dev->pci_dev->intr_handle;
>> +       if (!dev->intr_handle) {
>> +               RTE_PMD_DEBUG_TRACE("RX Intr handle unset\n");
>> +               return -ENOTSUP;
>> +       }
>> +
>> +       intr_handle = dev->intr_handle;
>>         if (!intr_handle->intr_vec) {
>>                 RTE_PMD_DEBUG_TRACE("RX Intr vector unset\n");
>>                 return -EPERM;
>> @@ -3205,6 +3217,8 @@ rte_eth_copy_pci_info(struct rte_eth_dev *eth_dev,
>> struct rte_pci_device *pci_de
>>                 return;
>>         }
>>
>> +       eth_dev->intr_handle = &pci_dev->intr_handle;
>> +
>>         eth_dev->data->dev_flags = 0;
>>         if (pci_dev->driver->drv_flags & RTE_PCI_DRV_INTR_LSC)
>>                 eth_dev->data->dev_flags |= RTE_ETH_DEV_INTR_LSC;
>> diff --git a/lib/librte_ether/rte_ethdev.h b/lib/librte_ether/rte_ethdev.h
>> index 3adbb2b..f1f656a 100644
>> --- a/lib/librte_ether/rte_ethdev.h
>> +++ b/lib/librte_ether/rte_ethdev.h
>> @@ -1629,6 +1629,7 @@ struct rte_eth_dev {
>>         const struct eth_driver *driver;/**< Driver for this device */
>>         const struct eth_dev_ops *dev_ops; /**< Functions exported by PMD
>> */
>>         struct rte_pci_device *pci_dev; /**< PCI info. supplied by probing
>> */
>> +       struct rte_intr_handle *intr_handle; /**< Device interrupt handle
>> */
>>         /** User application callbacks for NIC interrupts */
>>         struct rte_eth_dev_cb_list link_intr_cbs;
>>         /**
>>
>
> Is there another patch which replaces all uses of
> eth_dev->pci_dev->intr_handle with that of eth_dev->intr_handle?
>
> Now that eth_dev has a reference which is initialized as early as probe, we
> should use that rather than pci_dev->intr_handle for PCI PMDs.
>

I've added this indirection because it is required for the ethdev
function. The drivers shouldn't use the indirection through ethdev to
access the intr_handle because they do have direct access to the
device.


> OR maybe, ETH_DEV_INTR_HANDLE() like macro which you have introduced in
> i40e_ethdev.h.
>
> -
> Shreyansh

^ permalink raw reply

* Re: [PATCH v2] doc: fix required tools list layout
From: Mcnamara, John @ 2016-12-20 10:35 UTC (permalink / raw)
  To: Thomas Monjalon, Baruch Siach; +Cc: dev@dpdk.org
In-Reply-To: <1982293.fdVZR7yPaX@xps13>

> -----Original Message-----
> From: Thomas Monjalon [mailto:thomas.monjalon@6wind.com]
> Sent: Tuesday, December 20, 2016 10:27 AM
> To: Baruch Siach <baruch@tkos.co.il>
> Cc: dev@dpdk.org; Mcnamara, John <john.mcnamara@intel.com>
> Subject: Re: [dpdk-dev] [PATCH v2] doc: fix required tools list layout
> 
> > > The Python requirement should appear in the bullet list.
> > >
> > > Also, indent the x32 note, since it is related to the previous bullet.
> > >
> > > Signed-off-by: Baruch Siach <baruch@tkos.co.il>
> >
> > Acked-by: John McNamara <john.mcnamara@intel.com>
> 
> Applied, thanks

Thanks. 

I'll rebase my other patch that touches this file:

    http://dpdk.org/dev/patchwork/patch/18152/

John

^ permalink raw reply

* Re: [PATCH] doc: simplify L3fwd user guide examples
From: Mcnamara, John @ 2016-12-20 10:31 UTC (permalink / raw)
  To: De Lara Guarch, Pablo; +Cc: dev@dpdk.org
In-Reply-To: <1482165252-230968-1-git-send-email-pablo.de.lara.guarch@intel.com>



> -----Original Message-----
> From: De Lara Guarch, Pablo
> Sent: Monday, December 19, 2016 4:34 PM
> To: Mcnamara, John <john.mcnamara@intel.com>
> Cc: dev@dpdk.org; De Lara Guarch, Pablo <pablo.de.lara.guarch@intel.com>
> Subject: [PATCH] doc: simplify L3fwd user guide examples
> 
> L3 Forwarding sample app user guides have some inconsistencies between the
> example command line and the configuration table.
> Also, they were showing too complicated configuration, using two different
> NUMA nodes for two ports, which will probably lead to performance drop due
> to use cross-socket channel.
> 
> This patch simplifies the configuration of these examples, by using a
> single NUMA node and a single queue per port.
> 
> Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>

Acked-by: John McNamara <john.mcnamara@intel.com>

^ permalink raw reply

* Re: [PATCH] doc: correct source extract command
From: Thomas Monjalon @ 2016-12-20 10:31 UTC (permalink / raw)
  To: Baruch Siach; +Cc: dev, Mcnamara, John, David Marchand
In-Reply-To: <B27915DBBA3421428155699D51E4CFE202686057@IRSMSX103.ger.corp.intel.com>

> > DPDK source archives are .tar.xz or .tar.gz, not .zip. Use .tar.xz in the
> > instructions, since that is what the main download page links to.
> > 
> > Also, correct the archive file and directory name capitalization.
> > 
> > Signed-off-by: Baruch Siach <baruch@tkos.co.il>
> 
> Acked-by: John McNamara <john.mcnamara@intel.com>

Applied, thanks

^ permalink raw reply

* Re: [PATCH] nfp: extend speed capabilities advertised
From: Alejandro Lucero @ 2016-12-20 10:29 UTC (permalink / raw)
  To: Ferruh Yigit; +Cc: dev
In-Reply-To: <d25cdbc7-b4c5-3709-2d1f-0054f5071fa8@intel.com>

On Tue, Dec 20, 2016 at 10:25 AM, Ferruh Yigit <ferruh.yigit@intel.com>
wrote:

> On 12/19/2016 6:00 PM, Alejandro Lucero wrote:
> > I forgot one thing: to update the features file with this new one.
> >
> > I will wait for your feedback regarding the discussed problem for
> > sending another version.
>
> I think it is good to go, please send updated version.
>
>
OK


> >
>
> <...>
>
> >
> >         Sorry, confused. Is it like following:
> >
> >         "
> >         For new FW, there is no problem, it supports the range between
> >         1G - 50G,
> >         and reports correct current speed.
> >
> >         With old FW, device still can be set to speed range between 1G -
> >         50G,
> >         but it doesn't report current speed correct, and DPDK driver
> reports
> >         back to application as device current speed is 40G, without
> really
> >         knowing the current speed.
> >         "
> >
> >
> >     Yes, that is. Should then I do anything else or the patch is right
> >     for you now?
>
> So, this patch is correct.
>
> But for the previous patch "net/nfp: report link speed using hardware
> info", it would be nice to add above information into source as comment
> and into commit log.
> If you can, would you mind sending a new version of that patch?
>
>
I'll do it.

Thanks!


> Thanks,
> ferruh
>
> >
> <...>
>

^ permalink raw reply

* Re: [PATCH v2] doc: fix required tools list layout
From: Thomas Monjalon @ 2016-12-20 10:26 UTC (permalink / raw)
  To: Baruch Siach; +Cc: dev, Mcnamara, John
In-Reply-To: <B27915DBBA3421428155699D51E4CFE202689C6D@IRSMSX103.ger.corp.intel.com>

> > The Python requirement should appear in the bullet list.
> > 
> > Also, indent the x32 note, since it is related to the previous bullet.
> > 
> > Signed-off-by: Baruch Siach <baruch@tkos.co.il>
> 
> Acked-by: John McNamara <john.mcnamara@intel.com>

Applied, thanks

^ permalink raw reply

* Re: [PATCH] nfp: extend speed capabilities advertised
From: Ferruh Yigit @ 2016-12-20 10:25 UTC (permalink / raw)
  To: Alejandro Lucero; +Cc: dev
In-Reply-To: <CAD+H991y+h4T6OdCtYJNV4cq3wndGLxK9kkVqyGs5e_TN8Hiqg@mail.gmail.com>

On 12/19/2016 6:00 PM, Alejandro Lucero wrote:
> I forgot one thing: to update the features file with this new one.
> 
> I will wait for your feedback regarding the discussed problem for
> sending another version.

I think it is good to go, please send updated version.

> 

<...>

> 
>         Sorry, confused. Is it like following:
> 
>         "
>         For new FW, there is no problem, it supports the range between
>         1G - 50G,
>         and reports correct current speed.
> 
>         With old FW, device still can be set to speed range between 1G -
>         50G,
>         but it doesn't report current speed correct, and DPDK driver reports
>         back to application as device current speed is 40G, without really
>         knowing the current speed.
>         "
> 
> 
>     Yes, that is. Should then I do anything else or the patch is right
>     for you now?

So, this patch is correct.

But for the previous patch "net/nfp: report link speed using hardware
info", it would be nice to add above information into source as comment
and into commit log.
If you can, would you mind sending a new version of that patch?

Thanks,
ferruh

>      
<...>

^ permalink raw reply

* Re: [PATCH v2] doc: fix mistakes in contribution guide
From: Thomas Monjalon @ 2016-12-20 10:22 UTC (permalink / raw)
  To: Yong Wang; +Cc: dev, Mcnamara, John
In-Reply-To: <B27915DBBA3421428155699D51E4CFE202671B27@IRSMSX103.ger.corp.intel.com>

> > Signed-off-by: Yong Wang <wang.yong19@zte.com.cn>
> > ---
> > v2:
> > * modify some redundant descriptions.
> 
> 
> Thanks,
> 
> Acked-by: John McNamara <john.mcnamara@intel.com>

Applied, thanks

^ permalink raw reply

* Re: [PATCH v2] doc: add details of sub-trees and maintainers
From: Thomas Monjalon @ 2016-12-20 10:20 UTC (permalink / raw)
  To: John McNamara; +Cc: dev
In-Reply-To: <1481910634-19445-1-git-send-email-john.mcnamara@intel.com>

2016-12-16 17:50, John McNamara:
> Add a new section to the Code Contributors Guide to outline
> the roles of tree and component maintainers and how they
> can be added and removed.
> 
> Signed-off-by: John McNamara <john.mcnamara@intel.com>

Applied, thanks

^ permalink raw reply

* [PATCH] nfp: add tso support
From: Alejandro Lucero @ 2016-12-20 10:18 UTC (permalink / raw)
  To: dev

This patch implements NFP PMD support for TSO but it also requires
a firmware advertising the capability.

Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
---
 doc/guides/nics/features/nfp.ini |  1 +
 drivers/net/nfp/nfp_net.c        | 40 ++++++++++++++++++++++++++++++++--------
 2 files changed, 33 insertions(+), 8 deletions(-)

diff --git a/doc/guides/nics/features/nfp.ini b/doc/guides/nics/features/nfp.ini
index 7ac0d34..c04a738 100644
--- a/doc/guides/nics/features/nfp.ini
+++ b/doc/guides/nics/features/nfp.ini
@@ -11,6 +11,7 @@ Queue start/stop     = Y
 MTU update           = Y
 Jumbo frame          = Y
 Promiscuous mode     = Y
+TSO                  = Y
 RSS hash             = Y
 RSS key update       = Y
 RSS reta update      = Y
diff --git a/drivers/net/nfp/nfp_net.c b/drivers/net/nfp/nfp_net.c
index ace9583..cee8f63 100644
--- a/drivers/net/nfp/nfp_net.c
+++ b/drivers/net/nfp/nfp_net.c
@@ -1147,6 +1147,9 @@ static void nfp_net_read_mac(struct nfp_net_hw *hw)
 	dev_info->speed_capa = ETH_SPEED_NUM_1G | ETH_LINK_SPEED_10G |
 			       ETH_SPEED_NUM_25G | ETH_SPEED_NUM_40G |
 			       ETH_SPEED_NUM_50G | ETH_LINK_SPEED_100G;
+
+	if (hw->cap & NFP_NET_CFG_CTRL_LSO)
+		dev_info->tx_offload_capa |= DEV_TX_OFFLOAD_TCP_TSO;
 }
 
 static const uint32_t *
@@ -1641,6 +1644,27 @@ static void nfp_net_read_mac(struct nfp_net_hw *hw)
 	return 0;
 }
 
+/* nfp_net_tx_tso - Set TX descriptor for TSO */
+static inline void
+nfp_net_tx_tso(struct nfp_net_txq *txq, struct nfp_net_tx_desc *txd,
+	       struct rte_mbuf *mb)
+{
+	uint64_t ol_flags;
+	struct nfp_net_hw *hw = txq->hw;
+
+	if (!(hw->cap & NFP_NET_CFG_CTRL_LSO))
+		return;
+
+	ol_flags = mb->ol_flags;
+
+	if (!(ol_flags & PKT_TX_TCP_SEG))
+		return;
+
+	txd->l4_offset = mb->l2_len + mb->l3_len + mb->l4_len;
+	txd->lso = rte_cpu_to_le_16(mb->tso_segsz);
+	txd->flags |= PCIE_DESC_TX_LSO;
+}
+
 /* nfp_net_tx_cksum - Set TX CSUM offload flags in TX descriptor */
 static inline void
 nfp_net_tx_cksum(struct nfp_net_txq *txq, struct nfp_net_tx_desc *txd,
@@ -2009,7 +2033,7 @@ uint32_t nfp_net_txq_full(struct nfp_net_txq *txq)
 {
 	struct nfp_net_txq *txq;
 	struct nfp_net_hw *hw;
-	struct nfp_net_tx_desc *txds;
+	struct nfp_net_tx_desc *txds, txd;
 	struct rte_mbuf *pkt;
 	uint64_t dma_addr;
 	int pkt_size, dma_size;
@@ -2058,19 +2082,17 @@ uint32_t nfp_net_txq_full(struct nfp_net_txq *txq)
 
 		/*
 		 * Checksum and VLAN flags just in the first descriptor for a
-		 * multisegment packet
+		 * multisegment packet, but TSO info needs to be in all of them.
 		 */
-		nfp_net_tx_cksum(txq, txds, pkt);
+		nfp_net_tx_tso(txq, &txd, pkt);
+		nfp_net_tx_cksum(txq, &txd, pkt);
 
 		if ((pkt->ol_flags & PKT_TX_VLAN_PKT) &&
 		    (hw->cap & NFP_NET_CFG_CTRL_TXVLAN)) {
-			txds->flags |= PCIE_DESC_TX_VLAN;
-			txds->vlan = pkt->vlan_tci;
+			txd.flags |= PCIE_DESC_TX_VLAN;
+			txd.vlan = pkt->vlan_tci;
 		}
 
-		if (pkt->ol_flags & PKT_TX_TCP_SEG)
-			rte_panic("TSO is not supported\n");
-
 		/*
 		 * mbuf data_len is the data in one segment and pkt_len data
 		 * in the whole packet. When the packet is just one segment,
@@ -2088,6 +2110,8 @@ uint32_t nfp_net_txq_full(struct nfp_net_txq *txq)
 		*lmbuf = pkt;
 
 		while (pkt_size) {
+			/* Copying TSO, VLAN and cksum info */
+			*txds = txd;
 			dma_size = pkt->data_len;
 			dma_addr = rte_mbuf_data_dma_addr(pkt);
 			PMD_TX_LOG(DEBUG, "Working with mbuf at dma address:"
-- 
1.9.1

^ permalink raw reply related

* Re: [PATCH v2] doc: introduce PVP reference benchmark
From: Thomas Monjalon @ 2016-12-20 10:03 UTC (permalink / raw)
  To: Maxime Coquelin
  Cc: Mcnamara, John, yuanhan.liu@linux.intel.com, Yang, Zhiyong,
	ktraynor@redhat.com, dev
In-Reply-To: <B27915DBBA3421428155699D51E4CFE202671A59@IRSMSX103.ger.corp.intel.com>

> > >> Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
> > >
> > > There is one trailing whitespace warning but apart from that:
> > >
> > > Acked-by: John McNamara <john.mcnamara@intel.com>
> > 
> > Thanks John,
> > 
> > Do you want me to send a v3, fixing the trailing whitespace & collecting
> > the acks?
> 
> No need (unless the tree maintainer says otherwise).
> It was one trailing whitespace. Just something to look out for in future. 

Removed when applying.
Fixed some heading issues also:

--- a/doc/guides/howto/pvp_reference_benchmark.rst
+++ b/doc/guides/howto/pvp_reference_benchmark.rst
@@ -160,7 +160,7 @@ Testpmd launch
           --portmask=f --disable-hw-vlan -i --rxq=1 --txq=1
           --nb-cores=4 --forward-mode=io
 
-With this command, isolated CPUs 2 to 5 will be used as lcores for PMD threads.
+   With this command, isolated CPUs 2 to 5 will be used as lcores for PMD threads.
 
 #. In testpmd interactive mode, set the portlist to obtain the correct port
    chaining:
@@ -176,7 +176,8 @@ VM launch
 
 The VM may be launched either by calling QEMU directly, or by using libvirt.
 
-#. Qemu way:
+Qemu way
+^^^^^^^^
 
 Launch QEMU with two Virtio-net devices paired to the vhost-user sockets
 created by testpmd. Below example uses default Virtio-net options, but options
@@ -210,7 +211,8 @@ where isolated CPUs 6 and 7 will be used as lcores for Virtio PMDs:
       export PYTHONPATH=$PYTHONPATH:<QEMU path>/scripts/qmp
       ./qmp-vcpu-pin -s /tmp/qmp.socket 1 6 7
 
-#. Libvirt way:
+Libvirt way
+^^^^^^^^^^^
 
 Some initial steps are required for libvirt to be able to connect to testpmd's
 sockets.


Thanks

^ permalink raw reply


This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox