Re: REGRESSION: Performance regressions from switchinganon_vma->lock to mutex

[Tim Chen]

On Wed, 2011-06-15 at 18:45 +0200, Peter Zijlstra wrote:
> On Wed, 2011-06-15 at 09:18 -0700, Andi Kleen wrote:
> 
> > And in general it looks like blind conversion from spinlock to mutex
> > is a bad idea right now.
> 
> For 4 socket machines, maybe. On 2 sockets I cannot reproduce anything.
> 
> I wonder if its the fairness thing, the mutex spinners aren't fifo fair
> like the ticket locks are. It could be significant with larger socket
> count since their cacheline arbitration is more sucky.
> 
> 
> 

Peter,

Wonder if you can provide the profile on your run so I can compare with
what I got on 4 sockets?

Thanks.

Tim

--
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/