Re: [PATCH v4 0/2] perf, x86: handle overlapping counters

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

On 18.11.11 16:10:17, Peter Zijlstra wrote:
> OK, I've queued these and tested them using something like:

> +	EVENT_CONSTRAINT_OVERLAP(0x003c, 0x1|0x2, INTEL_ARCH_EVENT_MASK),
> +	EVENT_CONSTRAINT_OVERLAP(0x00c0, 0x2|0x4, INTEL_ARCH_EVENT_MASK),
> +	EVENT_CONSTRAINT_OVERLAP(0x4f2e, 0x4|0x8, INTEL_ARCH_EVENT_MASK),
> +	EVENT_CONSTRAINT_OVERLAP(0x412e, 0x8|0x1, INTEL_ARCH_EVENT_MASK),
> +
>  	EVENT_CONSTRAINT_END
>  };
>  
> To stress the thing, you can quite easily manage to hit that
> WARN_ON_ONCE() in there, but I haven't seen things explode yet.

Yes, the above would require a stack depth of 4 in the worst case
ending up in 2^4 max loops for the above. It would become more with
higher weights.

> > Robert Richter (2):
> >   configs: updating misc-x86_64-erda to v3.0.8
> >   config: updating misc-x86_64-erda.config for ASUS M4A89GTD PRO
> > 
> >  configs/misc-x86_64-erda.config |  517 ++++++++++++++++++++++++++++-----------
> >  1 files changed, 379 insertions(+), 138 deletions(-)
> 
> You sure about that? :-)

Yeah, wrong terminal... Luckily the patches are correct. ;)

Thanks,

-Robert

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