the below details is about Intel, what’s about mlx5 drivers ???
+/* How rings are layed out when accessing virtual functions or
- offloaded queues is device specific. To allow users to flow
- steering and specify these queues though break the ring cookie
- into a 32bit queue index with an 8 bit virtual function "
VF bind to a VM / KVM hypervisor (Linux Only)
By default the number of queues available are related to the number of cores allocated to your VM/VF
But there’s a way to increase this number by allocation a number of MSI-X interrupt line per VF with NUM_VF_MSIX where a number of lines maximum is 1144 shared to PF and VFs
Theses settings are avaialble at the FW level and can be configured by mlxconfig.
A new mechanism is also implemented in our latest fw and can be enable at the FW level DYNAMIC_VF_MSIX_TABLE
Allocation of number of queues is also dependant on your system, VM on top Esxi doesn’t have the same mechanism.