Re: [PATCH 4/4] perf, x86: Fix event scheduler to solve complex scheduling problems

[Robert Richter <robert.richter@amd.com>]

On 16.04.11 04:51:17, Peter Zijlstra wrote:
> Argh, crap. That's because AMD is now the first with overlapping
> constraints. Be sure to let your hardware guys know that they went from
> top to bottom om my appreciation list. AMD used to have no constraints
> and now they have the absolute worst.

Yes, the overlapping constraints are the problem.

> I'd really prefer not to do this for .39, and I'll have to sit down and
> actually read this code. It looks like we went from O(n^2) to O(n!) or
> somesuch, also not much of an improvement. I'll have to analyze the
> solver to see what it does for 'simple' constraints set to see if it
> will indeed be more expensive than the O(n^2) solver we had.

It wont be more expensive, if there is a solution. But if there is no
one we walk all possible ways now which is something like O(n!).

Yes, we can shift this general solution after .39. Will try to find a
solution that handles the special case for family 15h as a fix for
.39.

> Also, I think this code could do with a tiny bit of comments ;-)

Will comment on the code inline in the patch.

-Robert

-- 
Advanced Micro Devices, Inc.
Operating System Research Center