Re: [PATCH 0/4] perf: memory load/store events generalization

From: Stephane Eranian
Date: Fri Jul 22 2011 - 17:59:58 EST


Hi,

On Fri, Jul 22, 2011 at 2:43 PM, Andi Kleen <andi@xxxxxxxxxxxxxx> wrote:
> On Fri, Jul 22, 2011 at 02:14:26PM -0700, Stephane Eranian wrote:
>> On Fri, Jul 22, 2011 at 2:01 PM, Andi Kleen <andi@xxxxxxxxxxxxxx> wrote:
>> >> That looks okay as a first approach tool. But what people are most
>> >> often interested in is to see where the misses occur, i.e., you need
>> >> to display load/store addresses somehow, especially for the more
>> >
>> > But that's what it already does? Â(for loads, stores
>> > are not in there yet) Did you try it?
>> >
>> I meant displaying the load + data addresses.
>
> That's what it does already.
>
Are you talking about dump_load_data()?

You get:
- load instruction addr
- load data address
- latency
- data source

How do you display the load instruction address? Just using plain perf report?
How do you display the load data addresses?

But what's certainly more interesting if to get an output that shows
the correlation between the load and data addresses.


> Or did you mean the instructions that caused them?
>
yes.

> -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/