Re: [PATCH v0 00/71] perf: Add support for Intel Processor Trace

[Frederic Weisbecker]

On Mon, Dec 16, 2013 at 07:45:27AM -0800, Andi Kleen wrote:
> > You're right it's extremely slow. But it can still be relevant for debugging,
> > at least for apps that don't do too much CPU bound stuffs.
> 
> There are patches from Markus already for gdb to use it (using the old
> BTS perf interface). I'm not sure they have been merged into gdb
> mainline yet though.

Ok.

> 
> > My hope has always been that we can make a userspace function graph tracer
> > out of its dumps. And I think we can, I'm pretty sure that would be a useful tool.
> 
> I wrote one, based on the __fentry__, like the kernel:
> http://github.com/andikleen/ftracer

Sounds like nice stuff, but that implies building with the gcc option I think.

> 
> BTS has no timing information, so you could at best do a function tracer
> without timing.

Right, now although the function timing was the initial purpose of the function graph
tracer, the graph itself proved to be much more useful :)

But yeah the timing is nice too when we chase hotspot though probably perf report deprecated it.

> 
> -Andi
> -- 
> ak@xxxxxxxxxxxxxxx -- Speaking for myself only
--
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/