From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752621AbdLLPDA (ORCPT ); Tue, 12 Dec 2017 10:03:00 -0500 Received: from mail.kernel.org ([198.145.29.99]:36952 "EHLO mail.kernel.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751234AbdLLPC6 (ORCPT ); Tue, 12 Dec 2017 10:02:58 -0500 DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org D749C20740 Authentication-Results: mail.kernel.org; dmarc=none (p=none dis=none) header.from=kernel.org Authentication-Results: mail.kernel.org; spf=none smtp.mailfrom=mhiramat@kernel.org Date: Wed, 13 Dec 2017 00:02:55 +0900 From: Masami Hiramatsu To: Arnaldo Carvalho de Melo Cc: Paul Clarke , Masami Hiramatsu , bhargavb , linux-kernel@vger.kernel.org, Ravi Bangoria , linux-rt-users@vger.kernel.org, linux-perf-users@vger.kernel.org Subject: Re: [PATCH v2 2/5] perf-probe: Cut off the version suffix from event name Message-Id: <20171213000255.8cc0e9a698eb97d6f7b0c579@kernel.org> In-Reply-To: <20171211182547.GH3958@kernel.org> References: <151263115609.13843.6362262297053841418.stgit@devbox> <151263122864.13843.10998234736675352577.stgit@devbox> <20171208120148.63dcf5fb271cb8654e70b5cf@kernel.org> <3cfc7d95-0956-3e7b-9d36-db5bda9dc753@us.ibm.com> <20171211182547.GH3958@kernel.org> X-Mailer: Sylpheed 3.5.1 (GTK+ 2.24.31; x86_64-redhat-linux-gnu) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon, 11 Dec 2017 15:25:47 -0300 Arnaldo Carvalho de Melo wrote: > Em Fri, Dec 08, 2017 at 08:49:46AM -0600, Paul Clarke escreveu: > > > > > > On 12/07/2017 09:01 PM, Masami Hiramatsu wrote: > > > On Thu, 7 Dec 2017 10:34:51 -0600 > > > Paul Clarke wrote: > > >> On 12/07/2017 01:20 AM, Masami Hiramatsu wrote: > > >>> Cut off the version suffix (e.g. @GLIBC_2.2.5 etc.) from > > >>> automatic generated event name. This fixes wildcard event > > >>> adding like below case; > > >>> > > >>> ===== > > >>> # perf probe -x /lib64/libc-2.25.so malloc* > > >>> Internal error: "malloc_get_state@GLIBC_2" is wrong event name. > > >>> Error: Failed to add events. > > >>> ===== > > >>> > > >>> This failure was caused by a versioned suffix symbol. > > >>> With this fix, perf probe automatically cuts the > > >>> suffix after @ as below. > > >>> > > >>> ===== > > >>> # ./perf probe -x /lib64/libc-2.25.so malloc* > > >>> Added new events: > > >>> probe_libc:malloc_printerr (on malloc* in /usr/lib64/libc-2.25.so) > > >>> probe_libc:malloc_consolidate (on malloc* in /usr/lib64/libc-2.25.so) > > >>> probe_libc:malloc_check (on malloc* in /usr/lib64/libc-2.25.so) > > >>> probe_libc:malloc_hook_ini (on malloc* in /usr/lib64/libc-2.25.so) > > >>> probe_libc:malloc (on malloc* in /usr/lib64/libc-2.25.so) > > >>> probe_libc:malloc_trim (on malloc* in /usr/lib64/libc-2.25.so) > > >>> probe_libc:malloc_usable_size (on malloc* in /usr/lib64/libc-2.25.so) > > >>> probe_libc:malloc_stats (on malloc* in /usr/lib64/libc-2.25.so) > > >>> probe_libc:malloc_info (on malloc* in /usr/lib64/libc-2.25.so) > > >>> probe_libc:mallochook (on malloc* in /usr/lib64/libc-2.25.so) > > >>> probe_libc:malloc_get_state (on malloc* in /usr/lib64/libc-2.25.so) > > >>> probe_libc:malloc_set_state (on malloc* in /usr/lib64/libc-2.25.so) > > >>> > > >>> You can now use it in all perf tools, such as: > > >>> > > >>> perf record -e probe_libc:malloc_set_state -aR sleep 1 > > >>> > > >>> ===== > > >> > > >> I think this will still fall over for a case where there are multiple versions of the same base symbol, like: > > >> > > >> $ nm /lib64/libc.so.6 | egrep ' sched_getaffinity' > > >> 0000000000134430 T sched_getaffinity@GLIBC_2.3.3 > > >> 00000000000dcf00 T sched_getaffinity@@GLIBC_2.3.4 > > > > > > No, in that case perf probe adds number suffix (_1, _2...) for new events :) > > > > > > This feature (number suffix) is anyway required for the case that inlined > > > function is spreaded in multiple places. I think this is natural for perf probe. > > > > I can live with that. > > > > Is the versioning tag retained for display with "perf probe -l" display, so the probe points can be distinguished? > > > > In other words, will one see: > > # perf probe -l > > probe_libc:sched_getaffinity (on sched_getaffinity@GLIBC_2.3.3 in /usr/lib64/libc-2.17.so) > > probe_libc:sched_getaffinity_1 (on sched_getaffinity@@GLIBC_2.3.4 in /usr/lib64/libc-2.17.so) > > or > > # perf probe -l > > probe_libc:sched_getaffinity (on sched_getaffinity in /usr/lib64/libc-2.17.so) > > probe_libc:sched_getaffinity_1 (on sched_getaffinity in /usr/lib64/libc-2.17.so) > > [root@jouet ~]# perf probe -x /lib64/libc-2.25.so malloc* > Added new events: > probe_libc:malloc_printerr (on malloc* in /usr/lib64/libc-2.25.so) > probe_libc:malloc_consolidate (on malloc* in /usr/lib64/libc-2.25.so) > probe_libc:malloc_check (on malloc* in /usr/lib64/libc-2.25.so) > probe_libc:malloc_hook_ini (on malloc* in /usr/lib64/libc-2.25.so) > probe_libc:malloc (on malloc* in /usr/lib64/libc-2.25.so) > probe_libc:malloc_trim (on malloc* in /usr/lib64/libc-2.25.so) > probe_libc:malloc_usable_size (on malloc* in /usr/lib64/libc-2.25.so) > probe_libc:malloc_stats (on malloc* in /usr/lib64/libc-2.25.so) > probe_libc:malloc_info (on malloc* in /usr/lib64/libc-2.25.so) > probe_libc:mallochook (on malloc* in /usr/lib64/libc-2.25.so) > probe_libc:malloc_get_state (on malloc* in /usr/lib64/libc-2.25.so) > probe_libc:malloc_set_state (on malloc* in /usr/lib64/libc-2.25.so) > > You can now use it in all perf tools, such as: > > perf record -e probe_libc:malloc_set_state -aR sleep 1 > > [root@jouet ~]# perf probe -l > probe_libc:malloc (on 0xd22a6973 in /usr/lib64/libc-2.25.so) > probe_libc:malloc_check (on 0x25db6f5a in /usr/lib64/libc-2.25.so) > probe_libc:malloc_consolidate (on 0x88f7c44a in /usr/lib64/libc-2.25.so) > probe_libc:malloc_get_state (on 0xdd38eca7 in /usr/lib64/libc-2.25.so) > probe_libc:malloc_hook_ini (on 0x8de2b533 in /usr/lib64/libc-2.25.so) > probe_libc:malloc_info (on 0x8b327bde in /usr/lib64/libc-2.25.so) > probe_libc:malloc_printerr (on 0x8fff087c in /usr/lib64/libc-2.25.so) > probe_libc:malloc_set_state (on 0x861940e6 in /usr/lib64/libc-2.25.so) > probe_libc:malloc_stats (on 0x51fa3cdc in /usr/lib64/libc-2.25.so) > probe_libc:malloc_trim (on 0xa9fa4e43 in /usr/lib64/libc-2.25.so) > probe_libc:malloc_usable_size (on 0xceb5e090 in /usr/lib64/libc-2.25.so) > probe_libc:mallochook (on 0x6ebc00f5 in /usr/lib64/libc-2.25.so) This looks something wrong. perf-probe should try to get symbol at least. What happened on that ? Could you run perf probe -l with -vv ? On my Fedora26, $ sudo ./perf probe -l probe_libc:malloc (on __malloc in /usr/lib64/libc-2.25.so) probe_libc:malloc_check (on malloc_check in /usr/lib64/libc-2.25.so) probe_libc:malloc_consolidate (on malloc_consolidate.part.1 in /usr/lib64/libc-2.25.so) probe_libc:malloc_get_state (on malloc_get_state@GLIBC_2.2.5 in /usr/lib64/libc-2.25.so) probe_libc:malloc_hook_ini (on malloc_hook_ini in /usr/lib64/libc-2.25.so) probe_libc:malloc_info (on __malloc_info in /usr/lib64/libc-2.25.so) probe_libc:malloc_printerr (on malloc_printerr in /usr/lib64/libc-2.25.so) probe_libc:malloc_set_state (on malloc_set_state@GLIBC_2.2.5 in /usr/lib64/libc-2.25.so) probe_libc:malloc_stats (on __malloc_stats in /usr/lib64/libc-2.25.so) probe_libc:malloc_trim (on __malloc_trim in /usr/lib64/libc-2.25.so) probe_libc:malloc_usable_size (on __malloc_usable_size in /usr/lib64/libc-2.25.so) probe_libc:mallochook (on mallochook in /usr/lib64/libc-2.25.so) I can see the symbols with version suffixes. > [root@jouet ~]# > > [acme@jouet perf]$ nm /lib64/libc-2.25.so | grep malloc_get_state > 0000000000151ba0 T malloc_get_state@GLIBC_2.2.5 > [acme@jouet perf]$ > > For your example it finds two symbols, adds two but I couldn't find a > simple way of mapping one to the other, would be nice if 'perf probe -l' > showed them like 'nm' does. > > [root@jouet ~]# nm /lib64/libc-2.25.so | grep ' sched_getaffinity' > 0000000000151cd0 T sched_getaffinity@GLIBC_2.3.3 > 00000000000f3c10 T sched_getaffinity@@GLIBC_2.3.4 > [root@jouet ~]# > > [root@jouet ~]# perf probe -x /lib64/libc-2.25.so sched_getaffinity* > Added new events: > probe_libc:sched_getaffinity (on sched_getaffinity* in /usr/lib64/libc-2.25.so) > probe_libc:sched_getaffinity_1 (on sched_getaffinity* in /usr/lib64/libc-2.25.so) > > You can now use it in all perf tools, such as: > > perf record -e probe_libc:sched_getaffinity_1 -aR sleep 1 > > [root@jouet ~]# perf probe -l > probe_libc:sched_getaffinity (on 0x5fd95e7c in /usr/lib64/libc-2.25.so) > probe_libc:sched_getaffinity_1 (on 0xd8d51601 in /usr/lib64/libc-2.25.so) > [root@jouet ~]# This case also shows $ sudo ./perf probe -l probe_libc:sched_getaffinity (on __sched_getaffinity_new in /usr/lib64/libc-2.25.so) probe_libc:sched_getaffinity_1 (on sched_getaffinity@GLIBC_2.3.3 in /usr/lib64/libc-2.25.so) (but the first one seems a bit different, maybe it is an alias of sched_getaffinity@.) Thank you, -- Masami Hiramatsu