bpf

perform a command on an extended BPF map or program

sys_bpf(int cmd, union bpf_attr *attr, unsigned int size, struct bpf_common_attr *attr_common, unsigned int size_common)

register	value
`d0`	354, the syscall number
`d1`	int cmd
`d2`	union bpf_attr *attr
`d3`	unsigned int size
`d4`	struct bpf_common_attr *attr_common
`d5`	unsigned int size_common
`d0`	the return value

Then run trap #0 to enter the kernel.

Same syscall on other architectures