> module and worry about external interfaces. Expecting to be able to
> hack an internal function and not have to check, and if necessary
> fix everything that depends on the old behaviour is silly. ]
It's silly your otimization. What do you think to gain? Show me the
numbers. IMHO you only add a very subdle complexity and dependences to
code that currently are completly orthogonal and you gain nothing. I
consider it a design bug. It's a no-way for me and it was obvious since
the first look at your patch.
>more though.) [This is different from the previously mentioned
>scenarios (no idle cpus involved).]
The previous scenario was the same scenario but thought against my last
algorithm that I mistaken with the 2.3.x one. If you fall back on the idle
cpus only if avg_slice is relevant, then you don't need to take all CPU
busy with SCHED_OTHER tasks to make the same behaviour to happen.
>No, this does not appear to be a big issue. I'm not even convinced it
That's why I was not worried. Anyway it _make_ differences.
Andrea
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/