Gabriele Monaco [off-list ref] writes:

Wait, by /works properly/ you mean it reports a violation. I just
noticed you mention it in the description.

It's reasonable to request RT throttling disabled on sanely configured
RT systems. But throttling is a /valid/ kernel feature, I get you mark
it as /unwanted/ though.

True.

I guess if that's the case, this monitor doesn't belong in the sched
collection because it's meant to validate the kernel behaviour, not its
configuration for a specific purpose (RT).
Isn't it better off with the rtapp ones (which validate the system
configuration to run in an RT scenario).

Does it make sense to you?

Yeah I was a bit unsure where to put this monitor. But under rtapp makes
sense, if you prefer it there.

I still want to give it a run when I have a bit more time, besides with
RT throttle, can the monitor really fail on a working system?

RT throttling and fair deadline server are the only two known mechanisms
which would fail the monitor.

In the future, there may also be sched_ext deadline server:
https://lore.kernel.org/all/20250702232944.3221001-1-joelagnelf@nvidia.com/#t (local)

They exist for good reasons, but they are also a problem to real-time
tasks. I am posting this monitor because we did a cyclic test the other
day and observed some big latencies, and we had no idea why. It turned
out it was the fair deadline server. So we need this monitor to tell us
if some similar mechanisms exist or will appear in the future.

If you try the monitor and see problems, let me know. Most likely it
would be a flaw in the monitor, but there is also a chance there is
another throttling mechanism we are not yet aware of.

Nam