[PATCH] 2.3.41 scheduler change

[Alex Khripin]

Rik,
    I'll start with just one specific comment:

This part here: why bother with a goto when you can simply "return weight" from that point?
+       if (!p->counter)
                goto out;

    However, this patch seems to be a good idea in general. But I don't think the 20 percent bonus is very
realistic-sounding. Maybe set up a specific test environment to benchmark this:
2 calculation-intensive threads running in tandem. Say some sort of crypto or maybe encoders of some sort. No device I/O,
just a continuous loop.
2 small programs that bounce a packet to each other with a delay. These will function as the programs that will interrupt
the time slice and then go back to sleep.
1 program that polls the CPU usage of the 2 primary programs, for information purposes.
Regards,
    Alex K.


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo en vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/