Re: [PATCH v3 11/17] rv: Retry when da monitor detects race conditions

On Wed, Jul 16, 2025 at 10:38:11AM +0200, Gabriele Monaco wrote:

On Wed, 2025-07-16 at 10:27 +0200, Nam Cao wrote:

quoted

Can't you bring the monitor back to the init state, and start over
again?

I think "da_mon->monitoring = 0;" does the trick?

Yes you can, but I wouldn't do so silently.

Why not? The absolute worst that we get, is the rare case where a bug
appears at the exact same time. In that case, we would get a false
negative.

And I think that is really really rare.

I'd say the cleanest approach without reaction is to still return false
for the system to do all the cleanup but trace the event or, at the
very least, print a warning.

But you're right, this is more relevant for who develops the monitor
rather than for the user, so should probably be tracked separately.

Yes, if you really want to emit some sort of warning here, it should be
absolutely clear that the monitor itself is having a hiccup, not the
monitored kernel.

Nam