testPMD error with valid CPUSET (EAL: invalid core list syntax)

Question

In my system we have two AMD sockets. In NPS4 set-up (4 NUMA per socket) I can see following cores for NUMA-1

node 1 cpus: 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159

My NIC is connected to NUMA-1

cat /sys/bus/pci/devices/0000\:41\:00.0/numa_node
1
cat /sys/bus/pci/devices/0000\:41\:00.1/numa_node
1

In testPMD application (built over dpdk 19.11), if I use CPU-SET 16-31, I do not see any error. But if I use CPU-SET 144-159, I am seeing error " EAL: invalid core list syntax ".

I would like to know why is such error and what is the solution for this. I need to use more ncores from same NUMA in testPMD, so like to use cores from both CPU-SET 16-31 and 144-159.

cat /sys/devices/system/node/node1/cpulist
16-31,144-159

sudo ./testpmd -l144-159 -w 0000:41:00.0 -w 0000:41:00.1 -- -i --rxd=512 --txd=512 --rxq=4 --txq=4 --nb-cores=8
EAL: Detected 128 lcore(s)
EAL: Detected 8 NUMA nodes
EAL: invalid core list syntax

 sudo ./testpmd -l16-31 -w 0000:41:00.0 -w 0000:41:00.1 -- -i --rxd=512 --txd=512 --rxq=4 --txq=4 --nb-cores=8
EAL: Detected 128 lcore(s)
EAL: Detected 8 NUMA nodes
EAL: Multi-process socket /var/run/dpdk/rte/mp_socket
EAL: Selected IOVA mode 'VA'
EAL: Probing VFIO support...
EAL: VFIO support initialized
EAL: PCI device 0000:41:00.0 on NUMA socket 1
EAL:   probe driver: 15b3:101d net_mlx5
EAL: PCI device 0000:41:00.1 on NUMA socket 1
EAL:   probe driver: 15b3:101d net_mlx5
Interactive-mode selected
...

Answer 1

from the logs it is evident the max value for CONFIG_RTE_MAX_LCORE is 128 scores. If you want to use core from 129-160 update the value in $RTE_TARGET/.config and rebuild DPDK libraries and applications