Re: [tip:sched/core] sched: Lower chances of cputime scalingoverflow

From: Peter Zijlstra
Date: Fri Apr 12 2013 - 03:56:19 EST


On Thu, 2013-04-11 at 08:38 -0700, Linus Torvalds wrote:
> So *now*, once we are in the uncommon case, let's start counting bits.
> Like this:
>
> /* We know one of the values has a bit set in the high 32 bits */
> for (;;) {
> /* Make sure "stime" is the bigger of stime/rtime */
> if (rtime > stime) {
> u64 tmp = stime; stime = rtime; rtime = tmp;
> }
>
> /* Do we need to balance stime/rtime bits? */
> if (stime >> 32) {
> if (rtime >> 31)
> goto drop_precision;
>
> /* We can grow rtime and shrink stime and try to make them
> both fit */
> rtime <<= 1;
> stime >>= 1;
> continue;
> }
>
> /* stime/rtime fits in 32 bits, how about total? */
> if (!(total >> 32))
> break;
>
> drop_precision:
> /* We drop from stime, it has more bits than rtime */
> stime >>= 1;
> total >>= 1;
> }
>
> The above is totally untested, but each step is pretty damn simple and
> fairly cheap. Sure, it's a loop, but it's bounded to 32 (cheap)
> iterations, and the normal case is that it's not done at all, or done
> only a few times.

Right it gets gradually heavier the bigger the numbers get; which is
more and more unlikely.

> And the advantage is that the end result is always that simple
> 32x32/32 case that we started out with as the common case.
>
> I dunno. Maybe I'm overlooking something, and the above is horrible,
> but the above seems reasonably efficient if not optimal, and
> *understandable*.

I suppose that entirely matters on what one is used to ;-) I had to
stare rather hard at it for a little while.

But yes, you take it one step further and are willing to ditch rtime
bits too and I suppose that's fine.

Should work,.. Stanislaw could you stick this into your userspace
thingy and verify the numbers are sane enough?


--
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/