linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Milian Wolff <milian.wolff@kdab.com>
To: acme@kernel.org, jolsa@kernel.org, Jin Yao <yao.jin@linux.intel.com>
Cc: Linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org,
	Milian Wolff <milian.wolff@kdab.com>,
	Arnaldo Carvalho de Melo <acme@redhat.com>,
	David Ahern <dsahern@gmail.com>,
	Namhyung Kim <namhyung@kernel.org>,
	Peter Zijlstra <a.p.zijlstra@chello.nl>
Subject: [PATCH v5 09/16] perf report: compare symbol name for inlined frames when matching
Date: Mon,  9 Oct 2017 22:33:03 +0200	[thread overview]
Message-ID: <20171009203310.17362-10-milian.wolff@kdab.com> (raw)
In-Reply-To: <20171009203310.17362-1-milian.wolff@kdab.com>

The fake symbols we create for inlined frames will represent
different functions but can use the symbol start address. This leads
to issues when different inline branches all lead to the same
function.

Before:
~~~~~
$ perf report -s sym -i perf.inlining.data --inline --stdio -g function
...
             --38.86%--_start
                       __libc_start_main
                       main
                       |
                        --37.57%--std::norm<double> (inlined)
                                  std::_Norm_helper<true>::_S_do_it<double> (inlined)
                                  |
                                   --36.36%--std::abs<double> (inlined)
                                             std::__complex_abs (inlined)
                                             |
                                              --12.24%--std::linear_congruential_engine<unsigned long, 16807ul, 0ul, 2147483647ul>::operator() (inlined)
                                                        std::__detail::__mod<unsigned long, 2147483647ul, 16807ul, 0ul> (inlined)
                                                        std::__detail::_Mod<unsigned long, 2147483647ul, 16807ul, 0ul, true, true>::__calc (inlined)
~~~~~

Note that this backtrace representation is completely bogus.
Complex abs does not call the linear congruential engine! It
is just a side-effect of a longer inlined stack being appended
to a shorter, different inlined stack, both of which originate
in the same function (main).

This patch fixes the issue:

~~~~~
$ perf report -s sym -i perf.inlining.data --inline --stdio -g function
...
             --38.86%--_start
                       __libc_start_main
                       main
                       |
                       |--35.59%--std::uniform_real_distribution<double>::operator()<std::linear_congruential_engine<unsigned long, 16807ul, 0ul, 2147483647ul> > (inlined)
                       |          std::uniform_real_distribution<double>::operator()<std::linear_congruential_engine<unsigned long, 16807ul, 0ul, 2147483647ul> > (inlined)
                       |          |
                       |           --34.37%--std::__detail::_Adaptor<std::linear_congruential_engine<unsigned long, 16807ul, 0ul, 2147483647ul>, double>::operator() (inlined)
                       |                     std::generate_canonical<double, 53ul, std::linear_congruential_engine<unsigned long, 16807ul, 0ul, 2147483647ul> > (inlined)
                       |                     |
                       |                      --12.24%--std::linear_congruential_engine<unsigned long, 16807ul, 0ul, 2147483647ul>::operator() (inlined)
                       |                                std::__detail::__mod<unsigned long, 2147483647ul, 16807ul, 0ul> (inlined)
                       |                                std::__detail::_Mod<unsigned long, 2147483647ul, 16807ul, 0ul, true, true>::__calc (inlined)
                       |
                        --1.99%--std::norm<double> (inlined)
                                  std::_Norm_helper<true>::_S_do_it<double> (inlined)
                                  std::abs<double> (inlined)
                                  std::__complex_abs (inlined)
~~~~~

Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Yao Jin <yao.jin@linux.intel.com>
Signed-off-by: Milian Wolff <milian.wolff@kdab.com>
---
 tools/perf/util/callchain.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/tools/perf/util/callchain.c b/tools/perf/util/callchain.c
index b0ff56332134..3f1431bf71bd 100644
--- a/tools/perf/util/callchain.c
+++ b/tools/perf/util/callchain.c
@@ -670,11 +670,11 @@ static enum match_result match_chain(struct callchain_cursor_node *node,
 				     struct callchain_list *cnode)
 {
 	struct symbol *sym = node->sym;
+	enum match_result match;
 	u64 left, right;
 
 	if (callchain_param.key == CCKEY_SRCLINE) {
-		enum match_result match = match_chain_strings(cnode->srcline,
-							      node->srcline);
+		match = match_chain_strings(cnode->srcline, node->srcline);
 
 		/* if no srcline is available, fallback to symbol name */
 		if (match == MATCH_ERROR && cnode->ms.sym && node->sym)
@@ -688,6 +688,13 @@ static enum match_result match_chain(struct callchain_cursor_node *node,
 	}
 
 	if (cnode->ms.sym && sym && callchain_param.key == CCKEY_FUNCTION) {
+		/* compare inlined frames based on their symbol name because
+		 * different inlined frames will have the same symbol start
+		 */
+		if (cnode->ms.sym->inlined || node->sym->inlined)
+			return match_chain_strings(cnode->ms.sym->name,
+						   node->sym->name);
+
 		left = cnode->ms.sym->start;
 		right = sym->start;
 	} else {
-- 
2.14.2

  parent reply	other threads:[~2017-10-09 20:35 UTC|newest]

Thread overview: 40+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-10-09 20:32 [PATCH v5 00/16] generate full callchain cursor entries for inlined frames Milian Wolff
2017-10-09 20:32 ` [PATCH v5 01/16] perf report: remove code to handle inline frames from browsers Milian Wolff
2017-10-25 17:15   ` [tip:perf/core] perf report: Remove " tip-bot for Milian Wolff
2017-10-09 20:32 ` [PATCH v5 02/16] perf util: store srcline in callchain_cursor_node Milian Wolff
2017-10-25 17:16   ` [tip:perf/core] perf callchain: Store " tip-bot for Milian Wolff
2017-10-09 20:32 ` [PATCH v5 03/16] perf util: refactor inline_list to operate on symbols Milian Wolff
2017-10-25 17:16   ` [tip:perf/core] perf callchain: Refactor " tip-bot for Milian Wolff
2017-10-09 20:32 ` [PATCH v5 04/16] perf util: refactor inline_list to store srcline string directly Milian Wolff
2017-10-25 17:17   ` [tip:perf/core] perf callchain: Refactor " tip-bot for Milian Wolff
2017-10-09 20:32 ` [PATCH v5 05/16] perf report: create real callchain entries for inlined frames Milian Wolff
2017-10-25 17:17   ` [tip:perf/core] perf callchain: Create " tip-bot for Milian Wolff
2017-10-09 20:33 ` [PATCH v5 06/16] perf report: fall-back to function name comparison for -g srcline Milian Wolff
2017-10-25 17:18   ` [tip:perf/core] perf report: Fall-back " tip-bot for Milian Wolff
2017-10-09 20:33 ` [PATCH v5 07/16] perf report: mark inlined frames in output by " (inlined)" suffix Milian Wolff
2017-10-25 17:18   ` [tip:perf/core] perf callchain: Mark " tip-bot for Milian Wolff
2017-10-09 20:33 ` [PATCH v5 08/16] perf script: mark inlined frames and do not print DSO for them Milian Wolff
2017-10-25 17:18   ` [tip:perf/core] perf script: Mark " tip-bot for Milian Wolff
2017-10-09 20:33 ` Milian Wolff [this message]
2017-10-13 13:28   ` [PATCH v5 09/16] perf report: compare symbol name for inlined frames when matching Arnaldo Carvalho de Melo
2017-10-25 17:19   ` [tip:perf/core] perf callchain: Compare " tip-bot for Milian Wolff
2017-10-09 20:33 ` [PATCH v5 10/16] perf report: compare symbol name for inlined frames when sorting Milian Wolff
2017-10-25 17:19   ` [tip:perf/core] perf report: Compare " tip-bot for Milian Wolff
2017-10-09 20:33 ` [PATCH v5 11/16] perf report: properly handle branch count in match_chain Milian Wolff
2017-10-13 13:39   ` Arnaldo Carvalho de Melo
2017-10-13 14:08     ` Arnaldo Carvalho de Melo
2017-10-14 19:30       ` Milian Wolff
2017-10-16 14:17         ` Arnaldo Carvalho de Melo
2017-10-16  4:18       ` ravi
2017-10-16  8:27         ` Milian Wolff
2017-10-16 14:19         ` Arnaldo Carvalho de Melo
2017-10-09 20:33 ` [PATCH v5 12/16] perf report: cache failed lookups of inlined frames Milian Wolff
2017-10-09 20:33 ` [PATCH v5 13/16] perf report: cache srclines for callchain nodes Milian Wolff
2017-10-09 20:33 ` [PATCH v5 14/16] perf report: use srcline from callchain for hist entries Milian Wolff
2017-10-09 20:33 ` [PATCH v5 15/16] perf util: enable handling of inlined frames by default Milian Wolff
2017-10-09 20:33 ` [PATCH v5 16/16] perf util: use correct IP mapping to find srcline for hist entry Milian Wolff
2017-10-10  4:49   ` Namhyung Kim
2017-10-12 18:22     ` Milian Wolff
2017-10-12 18:52       ` Jiri Olsa
2017-10-13 11:03         ` Jiri Olsa
2017-10-13  1:19       ` Namhyung Kim

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20171009203310.17362-10-milian.wolff@kdab.com \
    --to=milian.wolff@kdab.com \
    --cc=Linux-kernel@vger.kernel.org \
    --cc=a.p.zijlstra@chello.nl \
    --cc=acme@kernel.org \
    --cc=acme@redhat.com \
    --cc=dsahern@gmail.com \
    --cc=jolsa@kernel.org \
    --cc=linux-perf-users@vger.kernel.org \
    --cc=namhyung@kernel.org \
    --cc=yao.jin@linux.intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).