Re: KPROBES: Instrumenting a function's call site

From: Keshavamurthy, Anil S
Date: Wed Sep 26 2007 - 13:33:32 EST

On Wed, Sep 26, 2007 at 10:09:33AM +0530, Ananth N Mavinakayanahalli wrote:
> On Tue, Sep 25, 2007 at 06:12:38PM -0400, Avishay Traeger wrote:
> > Hello,
> > I am trying to use kprobes to measure the latency of a function by
> > instrumenting its call site. Basically, I find the call instruction,
> > and insert a kprobe with a pre-handler and post-handler at that point.
> > The pre-handler measures the latency (reads the TSC counter). The
> > post-handler measures the latency again, and subtracts the value that
> > was read in the pre-handler to compute the total latency of the called
> > function.
> This sounds ok...

So what you are really measuring is the latency of just that single
instruction where you have inserted the probe i.e. because your
pre-handler is called just before the probed instruction is executed and
your post-handler is called right after you probed instruction is single-stepped.

> > So to measure the latency of foo(), I basically want kprobes to do this:
> > pre_handler();
> > foo();
When you insert a probe, you are inserting probe on an instruction boundary
and not at function level.
> > post_handler();

Hence the above looks like

Probed-instruction; // most likely the first instruction in the foo();

> >
> > The problem is that the latencies that I am getting are consistently low
> > (~10,000 cycles). When I manually instrument the functions, the latency
> > is about 20,000,000 cycles. Clearly something is not right here.
As I mentioned above what you are seeing is the latency of just the
probed instruction and hence it is very very low compared to
the latency of the function foo().

> You could try a a couple of approaches for starters.
I agree with Ananth, you can try the below approaches
for your measurements.

> a. As you mention above, a kprobe on the function invocation and the
> other on the instruction following the call; both need just pre_handlers.
> b.
> - Insert a kprobe and a kretprobe on foo()
> - The kprobe needs to have only a pre_handler that'll measure the latency
> - A similar handler for the kretprobe handler can measure the latency
> again and their difference will give you foo()'s latency.
> <b> though will require you to do some housekeeping in case foo() is
> reentrant to track which return instance corresponds to which call.
> Ananth
> PS: There was a thought of providing a facility to run a handler at
> function entry even when just a kretprobe is used. Maybe we need to
> relook at that; it'd have been useful in this case.
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at
Please read the FAQ at