Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752497AbdGACUS (ORCPT ); Fri, 30 Jun 2017 22:20:18 -0400 Received: from sub5.mail.dreamhost.com ([208.113.200.129]:42855 "EHLO homiemail-a116.g.dreamhost.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752145AbdGACTh (ORCPT ); Fri, 30 Jun 2017 22:19:37 -0400 From: Krister Johansen To: Peter Zijlstra , Ingo Molnar , Arnaldo Carvalho de Melo Cc: Alexander Shishkin , linux-kernel@vger.kernel.org, Krister Johansen Subject: [PATCH tip/perf/core 5/7] perf top: support lookup of symbols in other mount namespaces. Date: Fri, 30 Jun 2017 19:18:57 -0700 Message-Id: <1498875539-4200-6-git-send-email-kjlx@templeofstupid.com> X-Mailer: git-send-email 2.7.4 In-Reply-To: <1498875539-4200-1-git-send-email-kjlx@templeofstupid.com> References: <1498875539-4200-1-git-send-email-kjlx@templeofstupid.com> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1779 Lines: 52 The perf top command needs to unshare its fs from the helper threads in order to successfully setns(2) during its symbol lookup. It also needs to impelement a force flag to ignore ownership of perf-.map files. Signed-off-by: Krister Johansen --- tools/perf/builtin-top.c | 15 +++++++++++++++ 1 file changed, 15 insertions(+) diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c index 2bcfa46..c918fef 100644 --- a/tools/perf/builtin-top.c +++ b/tools/perf/builtin-top.c @@ -586,6 +586,13 @@ static void *display_thread_tui(void *arg) .refresh = top->delay_secs, }; + /* In order to read symbols from other namespaces perf to needs to call + * setns(2). This isn't permitted if the struct_fs has multiple users. + * unshare(2) the fs so that we may continue to setns into namespaces + * that we're observing. + */ + (void) unshare(CLONE_FS); + perf_top__sort_new_samples(top); /* @@ -627,6 +634,13 @@ static void *display_thread(void *arg) struct perf_top *top = arg; int delay_msecs, c; + /* In order to read symbols from other namespaces perf to needs to call + * setns(2). This isn't permitted if the struct_fs has multiple users. + * unshare(2) the fs so that we may continue to setns into namespaces + * that we're observing. + */ + (void) unshare(CLONE_FS); + display_setup_sig(); pthread__unblock_sigwinch(); repeat: @@ -1205,6 +1219,7 @@ int cmd_top(int argc, const char **argv) "Show raw trace event output (do not use print fmt or plugins)"), OPT_BOOLEAN(0, "hierarchy", &symbol_conf.report_hierarchy, "Show entries in a hierarchy"), + OPT_BOOLEAN(0, "force", &symbol_conf.force, "don't complain, do it"), OPT_END() }; const char * const top_usage[] = { -- 2.7.4