d/numam-dpdk
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

net/virtio: set default speed unknown

Browse Source 
rte_ethdev states new rule for NICs: they should return UNKNOWN
speed if speed is unknown and interface is up, in case of down
interface, NONE speed should be returned.

Signed-off-by: Ivan Dyukov <i.dyukov@samsung.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
This commit is contained in:
Ivan Dyukov 2020-09-23 00:18:01 +03:00 committed by Ferruh Yigit
parent eca9a0d6c8
commit b415b7a068
3 changed files with 10 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
doc/guides/nics/virtio.rst
View File

@ -361,7 +361,7 @@ Below devargs are supported by the PCI virtio driver:
It is used to specify link speed of virtio device. Link speed is a part of
link status structure. It could be requested by application using
rte_eth_link_get_nowait function.
(Default: 10000 (10G))
(Default: 0xffffffff (Unknown))
#. ``vectorized``:
@ -422,7 +422,7 @@ Below devargs are supported by the virtio-user vdev:
It is used to specify link speed of virtio device. Link speed is a part of
link status structure. It could be requested by application using
rte_eth_link_get_nowait function.
(Default: 10000 (10G))
(Default: 0xffffffff (Unknown))
#. ``vectorized``:

4
doc/guides/rel_notes/release_20_11.rst
View File

@ -74,6 +74,10 @@ New Features
* Added SR-IOV PF support
* **Updated Virtio driver.**
* Changed default link speed to unknown.
* **Extended flow-perf application.**
* Started supporting user order instead of bit mask:

9
drivers/net/virtio/virtio_ethdev.c
View File

@ -1670,7 +1670,6 @@ virtio_configure_intr(struct rte_eth_dev *dev)
return 0;
}
#define SPEED_UNKNOWN 0xffffffff
#define DUPLEX_UNKNOWN 0xff
/* reset device and renegotiate features if needed */
static int
@ -1727,7 +1726,7 @@ virtio_init_device(struct rte_eth_dev *eth_dev, uint64_t req_features)
hw->mac_addr[0], hw->mac_addr[1], hw->mac_addr[2],
hw->mac_addr[3], hw->mac_addr[4], hw->mac_addr[5]);
if (hw->speed == SPEED_UNKNOWN) {
if (hw->speed == ETH_SPEED_NUM_UNKNOWN) {
if (vtpci_with_feature(hw, VIRTIO_NET_F_SPEED_DUPLEX)) {
config = &local_config;
vtpci_read_dev_config(hw,
@ -1740,8 +1739,6 @@ virtio_init_device(struct rte_eth_dev *eth_dev, uint64_t req_features)
hw->duplex = config->duplex;
}
}
if (hw->speed == SPEED_UNKNOWN)
hw->speed = ETH_SPEED_NUM_10G;
if (hw->duplex == DUPLEX_UNKNOWN)
hw->duplex = ETH_LINK_FULL_DUPLEX;
PMD_INIT_LOG(DEBUG, "link speed = %d, duplex = %d",
@ -1893,7 +1890,7 @@ int
eth_virtio_dev_init(struct rte_eth_dev *eth_dev)
{
struct virtio_hw *hw = eth_dev->data->dev_private;
uint32_t speed = SPEED_UNKNOWN;
uint32_t speed = ETH_SPEED_NUM_UNKNOWN;
int vectorized = 0;
int ret;
@ -2552,6 +2549,7 @@ virtio_dev_link_update(struct rte_eth_dev *dev, __rte_unused int wait_to_complet
if (!hw->started) {
link.link_status = ETH_LINK_DOWN;
link.link_speed = ETH_SPEED_NUM_NONE;
} else if (vtpci_with_feature(hw, VIRTIO_NET_F_STATUS)) {
PMD_INIT_LOG(DEBUG, "Get link status from hw");
vtpci_read_dev_config(hw,
@ -2559,6 +2557,7 @@ virtio_dev_link_update(struct rte_eth_dev *dev, __rte_unused int wait_to_complet
&status, sizeof(status));
if ((status & VIRTIO_NET_S_LINK_UP) == 0) {
link.link_status = ETH_LINK_DOWN;
link.link_speed = ETH_SPEED_NUM_NONE;
PMD_INIT_LOG(DEBUG, "Port %d is down",
dev->data->port_id);
} else {