Re: [SCHED] wrong priority calc - SIMPLE test case

[Mike Galbraith]

At 04:11 PM 12/31/2005 +0100, Paolo Ornati wrote:
> On Sat, 31 Dec 2005 09:13:24 +0100
> Mike Galbraith <efault@xxxxxx> wrote:
>
> > Ingo seems to have done something in 2.6.15-rc7-rt1 which defeats your
> > little proggy.  Taking a quick peek at the rt scheduler changes, nothing
> > poked me in the eye, but by golly, I can't get this kernel to act up,
> > whereas 2.6.14-virgin does.
>
> Ok, I've sucessfully booted 2.6.15-rc7-rt1 (I think that I was
> having troubles with Thread Softirqs and/or Thread Hardirqs).
>
> First thing: I've preemption disabled, but it shouldn't matter too much
> since we are talking about priority calculation...

Mine is fully preemptible.

> 1) My program isn't defeated at all. If I start it with the same args
> of the previous examples it "seems" defeated, but it isn't.
>
> Lowering the "cpu burn argument" I can reproduce the problem again:
>
> "./a.out 200 & ./a.out 333"
>
>   PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND
>  5607 paolo     15   0  2396  320  252 R 56.1  0.1   0:06.79 a.out
>  5606 paolo     15   0  2396  324  252 R 38.7  0.1   0:04.55 a.out
>     1 root      16   0  2556  552  468 S  0.0  0.1   0:00.28 init

Strange.  Using the exact same arguments, I do see some odd bouncing up to 
high priorities, but they spend the vast majority of their time down at 25.

        -Mike 

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