[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <1301665694.4859.675.camel@twins>
Date: Fri, 01 Apr 2011 15:48:14 +0200
From: Peter Zijlstra <peterz@...radead.org>
To: Lin Ming <ming.m.lin@...el.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@...achi.com>,
Arnaldo Carvalho de Melo <acme@...radead.org>,
Frederic Weisbecker <fweisbec@...il.com>,
LKML <linux-kernel@...r.kernel.org>,
"2nddept-manager@....hitachi.co.jp"
<2nddept-manager@....hitachi.co.jp>
Subject: Re: [RFC PATCH] perf report: add sort by file lines
On Fri, 2011-04-01 at 21:02 +0800, Lin Ming wrote:
> On Fri, 2011-04-01 at 00:32 +0800, Peter Zijlstra wrote:
> > On Thu, 2011-03-31 at 18:28 +0200, Peter Zijlstra wrote:
> > >
> > > You could even first build the basic block tree and only follow those
> > > branches that end up covering the region IP is in.
> >
> > s/tree/directed-graph/ clearly the basic blocks don't form a tree and
> > can contain cycles.
>
> There may be multiple different paths from the beginning of the function
> to the point we are interested in.
>
> Any idea how to handle this case?
Any one path should be sufficient I think, C doesn't really have dynamic
typing so whatever path leads you to where you need to be ought to be
type invariant.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists