Date: Thu, 26 Feb 2015 17:44:27 -0300
From: Arnaldo Carvalho de Melo <acme@kernel.org>
To: Andi Kleen <ak@linux.intel.com>
Cc: kan.liang@intel.com, jolsa@redhat.com, namhyung@kernel.org,
        linux-kernel@vger.kernel.org
Subject: Re: [PATCH V7 1/1] perf tool:perf diff support for different binaries
Message-ID: <20150226204427.GE13373@kernel.org>
References: <1423460384-11645-1-git-send-email-kan.liang@intel.com>
 <20150226151751.GD13373@kernel.org>
 <20150226203404.GF27767@tassilo.jf.intel.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20150226203404.GF27767@tassilo.jf.intel.com>
User-Agent: Mutt/1.5.23 (2014-03-12)
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1207
Lines: 33

Em Thu, Feb 26, 2015 at 12:34:04PM -0800, Andi Kleen escreveu:
> > Having a nice explanation of the problem helps, as my first reaction to
> > this patch was: "What? This is what this tool is supposed to do, to
> > compare two versions of a binary, one that is being developed from the
> > same source, the other with slight modifications, etc", while the
> > description of the patch made it look as this was a feature that was now
> > being introduced.
> 
> AFAIK it was actually to compare multiple runs of the same binary
> with different setups: e.g. varying number of threads.
> 
> However in the Linux kernel development world it turns out comparing
> multiple similar binaries is much more useful.

Surely I am getting old and forgetting things, but when I wrote it, my
intent was to do:

vi a.c
build it
perf record a
vi a.c # change it
build it
perf record a
perf diff

And see if what I did while vi'ing it matched what I thought it would.

- Arnaldo
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/