On Wed, Mar 5, 2025 at 9:08 AM Alexei Starovoitov <alexei.starovoitov@xxxxxxxxx> wrote: [...] > My preference is to go via bpf-next, since changes are bigger > on bpf side than on lsm side. > > Re: selftest. > > Why change them at all if 'bool kernel' attribute is unused ? > Addition of the attr should be backward compatible change, > so all tests should still pass as-is. I was thinking of keeping the argument list in the selftests up to date, so that the users can use selftests as examples when writing their BPF programs. OTOH, with the "bool kernel" at the end of the argument list, it is backward compatible. > You probably should add a new test where 'kernel' arg is actually > used for something. That would be patch 2. +1. This is a great idea. Thanks, Song
