[PATCH] ethdev: fix device init without socket-local memory

[Bruce Richardson]

When allocating memory for an ethdev, the rte_malloc_socket call used
only allocates memory on the NUMA node/socket local to the device. This
means that even if the user wanted to, they could never use a remote NIC
without also having memory on that NIC's socket.

For example, if we change examples/skeleton/basicfwd.c to have
SOCKET_ID_ANY as the socket_id parameter for Rx and Tx rings, we should
be able to run the app cross-numa e.g. as below, where the two PCI
devices are on socket 1, and core 1 is on socket 0:

 ./build/examples/dpdk-skeleton -l 1 --legacy-mem --socket-mem=1024,0 \
		-a a8:00.0 -a b8:00.0

This fails however, with the error:

  ETHDEV: failed to allocate private data
  PCI_BUS: Requested device 0000:a8:00.0 cannot be used

We can remove this restriction by doing a fallback call to general
rte_malloc after a call to rte_malloc_socket fails. This should be safe
to do because the later ethdev calls to setup Rx/Tx queues all take a
socket_id parameter, which can be used by applications to enforce the
requirement for local-only memory for a device, if so desired. [If
device-local memory is present it will be used as before, while if not
present the rte_eth_dev_configure call will now pass, but the subsequent
queue setup calls requesting local memory will fail].

Fixes: e489007a411c ("ethdev: add generic create/destroy ethdev APIs")
Fixes: dcd5c8112bc3 ("ethdev: add PCI driver helpers")
Cc: stable at dpdk.org

Signed-off-by: Bruce Richardson <bruce.richardson at intel.com>
Signed-off-by: Padraig Connolly <padraig.j.connolly at intel.com>
---
 lib/ethdev/ethdev_driver.c |  7 ++++++-
 lib/ethdev/ethdev_pci.h    | 11 ++++++++++-
 2 files changed, 16 insertions(+), 2 deletions(-)

diff --git a/lib/ethdev/ethdev_driver.c b/lib/ethdev/ethdev_driver.c
index f48c0eb8bc..f9ce7ec348 100644
--- a/lib/ethdev/ethdev_driver.c
+++ b/lib/ethdev/ethdev_driver.c
@@ -303,11 +303,16 @@ rte_eth_dev_create(struct rte_device *device, const char *name,
 			return -ENODEV;
 
 		if (priv_data_size) {
+			/* try alloc private data on device-local node. */
 			ethdev->data->dev_private = rte_zmalloc_socket(
 				name, priv_data_size, RTE_CACHE_LINE_SIZE,
 				device->numa_node);
+			/* fall back to alloc on any socket on failure */
+			if (ethdev->data->dev_private == NULL)
+				ethdev->data->dev_private = rte_zmalloc(name,
+						priv_data_size, RTE_CACHE_LINE_SIZE);
 
-			if (!ethdev->data->dev_private) {
+			if (ethdev->data->dev_private == NULL) {
 				RTE_ETHDEV_LOG_LINE(ERR,
 					"failed to allocate private data");
 				retval = -ENOMEM;
diff --git a/lib/ethdev/ethdev_pci.h b/lib/ethdev/ethdev_pci.h
index 737fff1833..d600d9acbb 100644
--- a/lib/ethdev/ethdev_pci.h
+++ b/lib/ethdev/ethdev_pci.h
@@ -93,10 +93,19 @@ rte_eth_dev_pci_allocate(struct rte_pci_device *dev, size_t private_data_size)
 			return NULL;
 
 		if (private_data_size) {
+			/* Try and alloc the private-data structure on socket local to the device */
 			eth_dev->data->dev_private = rte_zmalloc_socket(name,
 				private_data_size, RTE_CACHE_LINE_SIZE,
 				dev->device.numa_node);
-			if (!eth_dev->data->dev_private) {
+
+			/* if cannot allocate memory on the socket local to the device
+			 * use rte_malloc to allocate memory on some other socket, if available.
+			 */
+			if (eth_dev->data->dev_private == NULL)
+				eth_dev->data->dev_private = rte_zmalloc(name,
+					private_data_size, RTE_CACHE_LINE_SIZE);
+
+			if (eth_dev->data->dev_private == NULL) {
 				rte_eth_dev_release_port(eth_dev);
 				return NULL;
 			}
-- 
2.43.0