On Thu, 2026-10-08 at 13:16 +0200, Nam Cao wrote: > Gabriele Monaco <[email protected]> writes: > > Mmh, I'm not too familiar with them, but they wouldn't be drop-in > > replacements as far as I understand. I'd need to use C11 atomic types > > and that would diverge further from the kernel implementation. > > I am thinking we should migrate to BPF entirely. Keeping both in-kernel > and BPF will be a bit painful.
Yeah that's a valid point, though I'm considering it more like a longer term goal. I'd rather avoid rushing the migration until we're sure BPF is perfectly capable to handle absolutely everything and we can cover every usecase. There's a common belief considering BPF a prototyping playground, where "real" things should eventually end up in the kernel. Now, I don't agree with it (look at the many sched_ext policies used in production), but I'd want to assess if dropping kernel support could drive users away (e.g. industry/automotive which keeps popping out) and understand the various not purely technical reasons. Gabriele
