Re: [PATCH 0/2] perf_counter: add tracepoint support

[Ingo Molnar]

* Peter Zijlstra <a.p.zijlstra@xxxxxxxxx> wrote:

> On Thu, 2009-07-09 at 05:10 +0200, Frederic Weisbecker wrote:

> > (RFD parenthesis: why not using the one in kernel/trace for all 
> > perf events? Now that it is lockless, it would fit nicely 
> > inside: no fear about NMI context).
> 
> because that one doesn't have a single-buffer mode you can mmap().

Yeah, the ring-buffer code could certainly be abstracted out better 
to allow this. Once that is done it would allow us to do a couple of 
nice things: such as splice support (which could perhaps be faster 
than mmap for perf record) and the sampling of tracepoints would 
become a whole lot easier a well. But it's certainly not a trivial 
job!

	Ingo
--
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/