From mboxrd@z Thu Jan 1 00:00:00 1970 From: Andi Kleen Subject: Re: Announcing simple-pt -- a simple Processor Trace implementation for Linux Date: Mon, 17 Aug 2015 20:21:04 +0200 Message-ID: <20150817182104.GJ1747@two.firstfloor.org> References: <20150817043130.GI1747@two.firstfloor.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from one.firstfloor.org ([193.170.194.197]:49470 "EHLO one.firstfloor.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751159AbbHQSVG (ORCPT ); Mon, 17 Aug 2015 14:21:06 -0400 Content-Disposition: inline In-Reply-To: Sender: linux-perf-users-owner@vger.kernel.org List-ID: To: Frederic Weisbecker Cc: Andi Kleen , LKML , perf group > > % sptdecode --sideband ptout.sideband --pt ptout.0 | less > > TIME DELTA INSNs OPERATION > > frequency 32 > > 0 [+0] [+ 1] _dl_aux_init+436 > > [+ 6] __libc_start_main+455 -> _dl_discover_osversion > > ... > > [+ 13] __libc_start_main+446 -> main > > [+ 9] main+22 -> f1 > > [+ 4] f1+9 -> f2 > > [+ 2] f1+19 -> f2 > > [+ 5] main+22 -> f1 > > [+ 4] f1+9 -> f2 > > [+ 2] f1+19 -> f2 > > [+ 5] main+22 -> f1 > > Nice. So I guess +x is the address offset. How hard would it be to > translate to file lines? Yes it's the address offset. Translating to lines wouldn't be too hard, just needs to be implemented with a dwarf reader. BTW the PT trace has all branches, just not the calls, but it's more difficult to display them all in a nice way. -Andi