Re: [PATCH] perfcounters: IRQ and NMI support on AMD CPUs

[Peter Zijlstra]

On Fri, 2009-03-06 at 00:02 +0530, Jaswinder Singh Rajput wrote:
> Hello Peter,
> 
> Ahh, so that day you was getting triple fault in this ?
> 
> On Thu, 2009-03-05 at 18:20 +0100, Peter Zijlstra wrote:
> > Subject: perfcounters: IRQ and NMI support on AMD CPUs
> > From: Peter Zijlstra <a.p.zijlstra@xxxxxxxxx>
> > Date: Thu Mar 05 18:08:27 CET 2009
> > 
> > The below completes the K7+ performance counter support.
> > 
> > Signed-off-by: Peter Zijlstra <a.p.zijlstra@xxxxxxxxx>
> > ---
> >  arch/x86/kernel/cpu/perf_counter.c |  272 +++++++++++++++++++++++++++++++------
> >  1 file changed, 228 insertions(+), 44 deletions(-)
> > 
> 
> It was in my TODO list, but I am glad that you finished it :-)
> 
> Any how with this patch on my 64 bit AMD box I am getting:
> 
> [jaswinder@hpdv5 perfcounters]$ ./perfstat -e 0,1,2,3 ls > /dev/null 
> 
>  Performance counter stats for 'ls':
> 
> perfstat: perfstat.c:415: main: Assertion `res == sizeof(single_count)'
> failed.
> Aborted
> [jaswinder@hpdv5 perfcounters]$ 
> 
> [root@hpdv5 perfcounters]# ./kerneltop 
> KernelTop refresh period: 2 seconds
> kerneltop: kerneltop.c:883: main: Assertion `res == sizeof(ip)' failed.
> Aborted
> [root@hpdv5 perfcounters]#

Ah, I think you will need a fresh copy of both utilities, Paul recently
changed the ABI.

http://people.redhat.com/mingo/perfcounters/

Contains the latest IIRC

--
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/