From mboxrd@z Thu Jan 1 00:00:00 1970 From: Alex Elsayed Subject: Re: -fno-omit-frame-pointer Date: Sat, 21 Feb 2015 14:09:50 -0800 Message-ID: References: Mime-Version: 1.0 Content-Type: text/plain; charset="ISO-8859-1" Content-Transfer-Encoding: 7Bit Return-path: Received: from plane.gmane.org ([80.91.229.3]:53816 "EHLO plane.gmane.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751288AbbBUWKL (ORCPT ); Sat, 21 Feb 2015 17:10:11 -0500 Received: from list by plane.gmane.org with local (Exim 4.69) (envelope-from ) id 1YPIFT-0001o0-J8 for ceph-devel@vger.kernel.org; Sat, 21 Feb 2015 23:10:07 +0100 Received: from 66-87-139-69.pools.spcsdns.net ([66-87-139-69.pools.spcsdns.net]) by main.gmane.org with esmtp (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Sat, 21 Feb 2015 23:10:07 +0100 Received: from eternaleye by 66-87-139-69.pools.spcsdns.net with local (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Sat, 21 Feb 2015 23:10:07 +0100 Sender: ceph-devel-owner@vger.kernel.org List-ID: To: ceph-devel@vger.kernel.org Sage Weil wrote: > Hey Mark, > > I just watched Brendan Gregg's talk at SCALE (which was extremely good) > and one of the things he mentioned is the importance of > -fno-omit-frame-pointer. I notice we are setting it via do_autogen.sh > (i.e., for developers) but not in the production builds. Should we be? > I think it's important to be able to get meaningful callgraph output from > production deployments and I suspect there is minimal performance impact. > > http://www.slideshare.net/brendangregg/scale2015-linux-perfprofiling I personally favor DWARF-based unwinding - for one, on register-starved architectures such as 32-bit x86, the frame pointer itself can skew performance. With regard to it inflating instance size, that's something that can be mitigated - via split debuginfo (only pull in the ones you actually need), compressed debug info (zlib -9 shrinks it a good bit), and semantic deduplication/compression with dwz.