Re: [PATCH] perf_events: fix remapped count support

[Stephane Eranian <eranian@google.com>]

On Tue, Mar 23, 2010 at 6:41 PM, Peter Zijlstra <peterz@infradead.org> wrote:
> On Tue, 2010-03-23 at 18:25 +0200, Stephane Eranian wrote:
>>         This patch fixes the remapped counter support such that it now works
>>         on X86 processors.
>
> (could you please not add all this whitespace in front? and make sure
>  it's no wider than 70 chars)
>
I will fix that.

> Also, I wouldn't say it fixes it, it's currently not broken, its plain
> not implemented. So this patch adds support for rdpmc.
>
There you go, then.

>>         For other architectures, e.g., PPC, SPARC, assuming they do offer the ability
>>         to read counts directly from user space, all that is needed is a couple of new
>>         arch specific functions:
>
> Power already supports this, so if you just broke this Paul is going to
> be unhappy.
>
I certainly changed the content of hdr->offset. But I don't see how
PPC could have this working otherwise in the presence of multiplex
and thus when you need the timing information to scale.

I'll wait for Paul's comments, then.