From mboxrd@z Thu Jan 1 00:00:00 1970 From: fche@redhat.com (Frank Ch. Eigler) Subject: Re: Size of perf data files Date: Thu, 08 Jan 2015 21:06:37 -0500 Message-ID: References: <87ppas31sq.fsf@tassilo.jf.intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from mx1.redhat.com ([209.132.183.28]:42558 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751129AbbAICIL (ORCPT ); Thu, 8 Jan 2015 21:08:11 -0500 In-Reply-To: (Yale Zhang's message of "Tue, 6 Jan 2015 13:02:23 -0800") Sender: linux-perf-users-owner@vger.kernel.org List-ID: To: Yale Zhang Cc: Andi Kleen , linux-perf-users@vger.kernel.org Yale Zhang writes: > [...] > I've already tried lowering the size of each sample from 8KiB to 512, > but that only allows storing 2 or 3 parent function calls, which isn't > enough for me. > > "don't use dwarf unwinding." > The main reason I'm trying to switch from Zoom to perf is because it > supports dwarf unwinding! [...] You may wish to try systemtap. Its backtracing uses in-situ dwarf unwinding, and is pretty fast. Each sample costs not 4K+ of stack snapshots, but a hexadecimal pc-list or optionally symbolic backtrace string. The downside is that the programs that you may be backtracing need to be identified at stap invocation - ahead of time - via something like: % stap -d /bin/foo -d /usr/lib64/libbar.so --ldd --all-modules SCRIPT.stp See e.g. https://sourceware.org/systemtap/examples/#profiling/pf4.stp - FChE