https://bugs.dpdk.org/show_bug.cgi?id=350
Bug ID: 350
Summary: ixgbe: incorrect speed capabilities advertised for
X553 devices
Product: DPDK
Version: 19.08
Hardware: All
OS: Linux
Status: UNCONFIRMED
Severity: normal
Priority: Normal
Component: ethdev
Assignee: dev@dpdk.org
Reporter: mgsm...@netgate.com
Target Milestone: ---
Overview:
X553 copper 1Gbps devices (vendor ID 0x8086, device IDs 0x15e4 and 0x15e5) have
speed_capa set to ETH_LINK_SPEED_1G | ETH_LINK_SPEED_10G when device info is
retrieved with rte_eth_dev_info_get(). These devices are not capable of
supporting 10G, but they are probably capable of supporting 100M and 10M, which
are omitted from speed_capa.
Steps to Reproduce:
1. Install FD.io VPP packages (available at
https://packagecloud.io/fdio/release) on a system with X553 copper ports
attached which has a recent release of Ubuntu or CentOS 7 installed.
2. If the X553 ports are attached to the kernel ixgbe driver, bring those ports
down with a command like 'sudo ip link set dev <interface_name> down'
3. Start VPP with 'sudo systemctl start vpp'. It should bind the ports to a uio
driver and take control of them.
4. Check the interfaces VPP is managing with 'sudo vppctl show int'. The X553
ports will have names like TenGigabitEthernetX/Y/Z (X/Y/Z are filled in with
PCI address). The 'TenGigabitEthernet' portion of the name was selected based
on the speeds advertised.
Actual Results:
If gdb is attached and used to inspect contents of a struct rte_eth_dev_info
that was filled in by rte_eth_dev_info_get(), speed_capa has a value of 288
(ETH_LINK_SPEED_1G | ETH_LINK_SPEED_10G).
Expected Results:
rte_eth_dev_info_get(), via ixgbe_dev_info_get(), should set speed_capa to
ETH_LINK_SPEED_1G | ETH_LINK_SPEED_100M | ETH_LINK_SPEED_10M for devices with
IDs 0x15E4 (IXGBE_DEV_ID_X550EM_A_1G_T) and 0x15E5
(IXGBE_DEV_ID_X550EM_A_1G_T_L)
Build date and hardware:
Problem observed in DPDK 19.05. Appears to still be present in current
Intel C3558 SoC with 4 copper ports - 2 with device ID 0x15e4, 2 with device ID
0x15e5
--
You are receiving this mail because:
You are the assignee for the bug.
