Do you have any more thoughts on this? For my 2c: - It looks like the kernel is making the map BPF_F_RDONLY_PROG because "Lookup returns a pointer straight to dev->ifindex", so I don't know if we have to worry about the kernel allowing write access to devmaps - I like having the guardrails in libbpf, but I'm not sure what the exact contract is for what libbpf will let me reload I'll spin out a V2 similar to what I had if nobody has any strong opinions on the exact behavior of this check - Pramukh