On Wed, 8 Feb 2023 16:39:05 +0100 Pawel Chmielewski wrote: > With the introduction of sched_numa_hop_mask() and > for_each_numa_hop_mask(), the affinity masks for queue vectors can be > conveniently set by preferring the CPUs that are closest to the NUMA node > of the parent PCI device. Damn, you had this one locked and ready, didn't you.. :)
