Re: [PATCH 7/7] Simple Performance Counters: SLUB instrumentation

2007-08-20 Thread Christoph Lameter
On Fri, 17 Aug 2007, Mathieu Desnoyers wrote: > > The cpu will definitely not be idle during these measurements and no > > frequency scaling is active. > > The problem is that if the cpu is idle _before_ the measurements, the > frequency will change differently from one cpu to another.

Re: [PATCH 7/7] Simple Performance Counters: SLUB instrumentation

2007-08-20 Thread Christoph Lameter
On Fri, 17 Aug 2007, Mathieu Desnoyers wrote: The cpu will definitely not be idle during these measurements and no frequency scaling is active. The problem is that if the cpu is idle _before_ the measurements, the frequency will change differently from one cpu to another. Therefore, the

Re: [PATCH 7/7] Simple Performance Counters: SLUB instrumentation

2007-08-17 Thread Mathieu Desnoyers
* Christoph Lameter ([EMAIL PROTECTED]) wrote: > On Fri, 17 Aug 2007, Mathieu Desnoyers wrote: > > > Why do you printk inside the timing period ? Filling the printk buffers > > or outputting on things such as serial console could really hurt your > > results. > > It was easier to code? > > > I

Re: [PATCH 7/7] Simple Performance Counters: SLUB instrumentation

2007-08-17 Thread Christoph Lameter
On Fri, 17 Aug 2007, Mathieu Desnoyers wrote: > Why do you printk inside the timing period ? Filling the printk buffers > or outputting on things such as serial console could really hurt your > results. It was easier to code? > I hope you run your system with idle=poll and without frequency

Re: [PATCH 7/7] Simple Performance Counters: SLUB instrumentation

2007-08-17 Thread Mathieu Desnoyers
Hi Christoph, A few remarks on these tests: Why do you printk inside the timing period ? Filling the printk buffers or outputting on things such as serial console could really hurt your results. I hope you run your system with idle=poll and without frequency scaling at all, because otherwise

Re: [PATCH 7/7] Simple Performance Counters: SLUB instrumentation

2007-08-17 Thread Mathieu Desnoyers
Hi Christoph, A few remarks on these tests: Why do you printk inside the timing period ? Filling the printk buffers or outputting on things such as serial console could really hurt your results. I hope you run your system with idle=poll and without frequency scaling at all, because otherwise

Re: [PATCH 7/7] Simple Performance Counters: SLUB instrumentation

2007-08-17 Thread Christoph Lameter
On Fri, 17 Aug 2007, Mathieu Desnoyers wrote: Why do you printk inside the timing period ? Filling the printk buffers or outputting on things such as serial console could really hurt your results. It was easier to code? I hope you run your system with idle=poll and without frequency scaling

Re: [PATCH 7/7] Simple Performance Counters: SLUB instrumentation

2007-08-17 Thread Mathieu Desnoyers
* Christoph Lameter ([EMAIL PROTECTED]) wrote: On Fri, 17 Aug 2007, Mathieu Desnoyers wrote: Why do you printk inside the timing period ? Filling the printk buffers or outputting on things such as serial console could really hurt your results. It was easier to code? I hope you run

[PATCH 7/7] Simple Performance Counters: SLUB instrumentation

2007-07-31 Thread Christoph Lameter
With this patch SLUB will perform tests on bootup and display results. Signed-off-by: Christoph Lameter <[EMAIL PROTECTED]> --- mm/slub.c | 97 +--- 1 files changed, 92 insertions(+), 5 deletions(-) diff --git a/mm/slub.c b/mm/slub.c

[PATCH 7/7] Simple Performance Counters: SLUB instrumentation

2007-07-31 Thread Christoph Lameter
With this patch SLUB will perform tests on bootup and display results. Signed-off-by: Christoph Lameter [EMAIL PROTECTED] --- mm/slub.c | 97 +--- 1 files changed, 92 insertions(+), 5 deletions(-) diff --git a/mm/slub.c b/mm/slub.c index