From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932636AbbCEJq7 (ORCPT ); Thu, 5 Mar 2015 04:46:59 -0500 Received: from mail-wi0-f176.google.com ([209.85.212.176]:46984 "EHLO mail-wi0-f176.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932592AbbCEJqy (ORCPT ); Thu, 5 Mar 2015 04:46:54 -0500 Date: Thu, 5 Mar 2015 10:46:50 +0100 From: Ingo Molnar To: Borislav Petkov , Jiri Olsa , Arnaldo Carvalho de Melo , Adrian Hunter Cc: X86 ML , Andy Lutomirski , LKML , Linus Torvalds Subject: Re: [PATCH v2 07/15] x86/lib/copy_user_64.S: Convert to ALTERNATIVE_2 Message-ID: <20150305094650.GA29140@gmail.com> References: <1424776497-3180-1-git-send-email-bp@alien8.de> <1424776497-3180-8-git-send-email-bp@alien8.de> <20150304062552.GA16111@gmail.com> <20150304090038.GB3233@pd.tnic> <20150305003249.GB20449@gmail.com> <20150305083519.GB3915@pd.tnic> <20150305093457.GA25419@gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20150305093457.GA25419@gmail.com> User-Agent: Mutt/1.5.23 (2014-03-12) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org * Ingo Molnar wrote: > Ha! There's a neat alternatives debugging trick with perf that you > might not know about: if you run 'perf top' as root then perf will > use /proc/kcore to disassemble the live kernel image and if you look > at the assembly output of hot functions then you'll see the real, > patched instructions on the live kernel, not the vmlinux > instructions. > > I have two enhancement suggestions to the perf tooling developers for > this usecase: and I've got a bugreport as well, when I use the kcore annotations on 'fput' symbol (i.e. running 'perf top' as root), go into the disassembly window and then try to exit that window, then I get this segfault: Program received signal SIGSEGV, Segmentation fault. [Switching to Thread 0x7ffff3c41700 (LWP 29134)] lock__delete (ops=0x7fffec001818) at util/annotate.c:219 219 ins__delete(ops->locked.ops); (gdb) bt #0 lock__delete (ops=0x7fffec001818) at util/annotate.c:219 #1 0x000000000046e568 in disasm_line__free (dl=0x7fffec0017e0) at util/annotate.c:619 #2 0x00000000004f0743 in symbol__tui_annotate (sym=, map=, evsel=evsel@entry=0x83d770, hbt=hbt@entry=0x7ffff3c40e60) at ui/browsers/annotate.c:976 #3 0x00000000004f0973 in hist_entry__tui_annotate (he=he@entry=0x151ba40, evsel=evsel@entry=0x83d770, hbt=hbt@entry=0x7ffff3c40e60) at ui/browsers/annotate.c:835 #4 0x00000000004f5685 in perf_evsel__hists_browse (evsel=0x83d770, nr_events=nr_events@entry=1, helpline=helpline@entry=0x51d350 "For a higher level overview, try: perf top --sort comm,dso", left_exits=left_exits@entry=false, hbt=hbt@entry=0x7ffff3c40e60, min_pcnt=, env=0x83de40) at ui/browsers/hists.c:1716 #5 0x00000000004f7ae4 in perf_evlist__tui_browse_hists (evlist=0x7cbb90, help=help@entry=0x51d350 "For a higher level overview, try: perf top --sort comm,dso", hbt=hbt@entry=0x7ffff3c40e60, min_pcnt=, env=) at ui/browsers/hists.c:2006 #6 0x0000000000435850 in display_thread_tui (arg=0x7fffffffa580) at builtin-top.c:582 #7 0x00007ffff7bc40a5 in start_thread (arg=0x7ffff3c41700) at pthread_create.c:309 #8 0x00007ffff5fe4cfd in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:111 using latestest tools/perf. Thanks, Ingo