Re: [PATCH 0/3] perf_event: enable overflow interrupts on KNC

[Ingo Molnar <mingo@kernel.org>]

* Vince Weaver <vincent.weaver@maine.edu> wrote:

> On Wed, 17 Oct 2012, Ingo Molnar wrote:
> 
> > How well tested is this on real hardware and how robust is the 
> > hardware with this? Since it's a new PMU driver for v3.7, and if 
> > these are reasonably well tested, then we could send these to 
> > Linus via perf/urgent, so that they don't miss and have to wait 
> > all the way to v3.8.
> 
> You probably won't like this answer.
> 
> The patches have been tested against the 2.6.34 kernel that 
> Intel provides for KNC (found here 
> http://software.intel.com/en-us/forums/topic/278102 )
> 
> perf stat, perf record, and PAPI have all been tested and 
> work.
>
> The patches I've sent are the 2.6.34 ones forward ported to 
> the 3.7-rc1 kernel.  I haven't been able to run a 3.7-rc 
> kernel on the KNC board, because there are other patches to 
> mainline needed to get KNC running and they'd all have to be 
> forward-ported from 2.6.34.
> 
> Since the knc pmu driver closely models the p6 one, it was 
> straightforward to use it as a basis for the forward port.  
> Everything _should_ work but I've only compile tested.

Ok. As long as it's all identical I'd still be inclined to 
include it in v3.7, to reduce version skew. If the v3.7-rc1 
kernel won't even boot it's not like we'll be able to further 
regress it.

Thanks,

	Ingo