PF queues break the 4 Gbps barrier
by from OpenBSD Journal on (#74BN6)
OpenBSD'sPFpacket filter has long supported HFSC traffic shapingwith the queuerules inpf.conf(5).However, an internal 32-bit limitation in the HFSCservice curve structure (struct hfsc_sc) meant that bandwidth valueswere silently capped at approximately 4.29 Gbps," the maximum value of a u_int ".
With 10G, 25G, and 100Gnetwork interfaces now commonplace,OpenBSD devs making huge progress unlocking the kernel for SMP,and adding drivers for cards supporting some of these speeds,this limitation started to get in the way.Configuring bandwidth 10G on a queue would silently wrap around,producing incorrect and unpredictable scheduling behaviour.
A new patchwidens the bandwidth fields in the kernel's HFSC schedulerfrom 32-bit to 64-bit integers, removing this bottleneck entirely.The diff also fixes a pre-existing display bug inpftop(1)where bandwidth values above 4 Gbps would be shown incorrectly.