* Borislav Petkov <bp@amd64.org> wrote:

quoted text
> On Thu, Nov 11, 2010 at 10:38:50AM -0500, Peter Zijlstra wrote:
> > On Thu, 2010-11-11 at 14:34 +0100, Borislav Petkov wrote:
> > > 3) enable the persistent event and stash its fd in some /sysfs member
> > > which the RAS daemon reads out and mmaps. This last one actually sounds
> > > pretty neat. 
> > 
> > Right, stashing the thing in sysfs, which you can open()+mmap() seems
> > the best way. Not sure you want to make it a RAS specific place, maybe a
> > generic persistent buffer place.
> > 
> > You'll get multiple files anyway, one per cpu, so maybe something like:
> > 
> > /sysfs/persistent_buffer/ras[0..n]
> 
> Yeah, or even integrate it even further into the events hierarchy like
> this:
> 
> /sys/.../cpu/events/mce		-> this is the trace_mce_record tracepoint
> /sys/.../cpu/events/mce_pers	-> persistent buffer fd
> 
> or an additional level further down
> 
> /sys/.../cpu/events/mce/persistent/...
> 
> or whatever.
> 
> Generally I don't care too much where we put it at the end.

It should be in a nice logical place though :-)

Thanks,

	Ingo
--
unsubscribe notice
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/