linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: John Garry <john.garry@huawei.com>
To: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>, Ingo Molnar <mingo@redhat.com>,
	"Peter Zijlstra" <peterz@infradead.org>,
	Arnaldo Carvalho de Melo <acme@kernel.org>,
	Alexander Shishkin <alexander.shishkin@linux.intel.com>,
	Linuxarm <linuxarm@huawei.com>,
	"linux-arm-kernel@lists.infradead.org" 
	<linux-arm-kernel@lists.infradead.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	Namhyung Kim <namhyung@kernel.org>,
	"Will Deacon" <will.deacon@arm.com>,
	Mark Rutland <mark.rutland@arm.com>
Subject: Re: perf segmentation fault from NULL dereference
Date: Tue, 2 Oct 2018 11:20:46 +0100	[thread overview]
Message-ID: <334a114a-bb59-9582-a913-efbb01bd68c5@huawei.com> (raw)
In-Reply-To: <20180927030035.GF28040@tassilo.jf.intel.com>

On 27/09/2018 04:00, Andi Kleen wrote:
>> Please me let me know if a valid issue so we can get a fix in.
>
> If it crashes it must be a valid issue of course.
>
> But I'm not sure about your bisect. Hard to see how my patch
> could cause this. Sometimes bisects go wrong.
> You verified by just reverting the patch?

It no longer reverts cleanly. And the previous patch - 4ca69ca9db3a - 
did not have this crash:
root@localhost:~# ./perf_debug_ record -e armv8_pmuv3_0/br_mis_pred/ sleep 1
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.001 MB perf.data (6 samples) ]
root@localhost:~#

>
> First thing I would also try is to run with valgrind or ASan and see if it
> reports anything.

Here's the valgrind output:
root@localhost:~#valgrind  --leak-check=yes  ./perf_debug_ record -e 
armv8_pmuv3_0/br_mis_pred/ sleep 1
==16025== Memcheck, a memory error detector
==16025== Copyright (C) 2002-2015, and GNU GPL'd, by Julian Seward et al.
==16025== Using Valgrind-3.11.0 and LibVEX; rerun with -h for copyright info
==16025== Command: ./perf_debug_ record -e armv8_pmuv3_0/br_mis_pred/ 
sleep 1
==16025==
--16025-- WARNING: unhandled arm64-linux syscall: 168
--16025-- You may be able to write your own handler.
--16025-- Read the file README_MISSING_SYSCALL_OR_IOCTL.
--16025-- Nevertheless we consider this a bug.  Please report
--16025-- it at http://valgrind.org/support/bug_reports.html.
--16025-- WARNING: unhandled arm64-linux syscall: 241
--16025-- You may be able to write your own handler.
--16025-- Read the file README_MISSING_SYSCALL_OR_IOCTL.
--16025-- Nevertheless we consider this a bug.  Please report
--16025-- it at http://valgrind.org/support/bug_reports.html.
perf_event_open(..., PERF_FLAG_FD_CLOEXEC) failed with unexpected error 
38 (Function not implemented)
--16025-- WARNING: unhandled arm64-linux syscall: 241
--16025-- You may be able to write your own handler.
--16025-- Read the file README_MISSING_SYSCALL_OR_IOCTL.
--16025-- Nevertheless we consider this a bug.  Please report
--16025-- it at http://valgrind.org/support/bug_reports.html.
perf_event_open(..., 0) failed unexpectedly with error 38 (Function not 
implemented)
--16025-- WARNING: unhandled arm64-linux syscall: 241
--16025-- You may be able to write your own handler.
--16025-- Read the file README_MISSING_SYSCALL_OR_IOCTL.
--16025-- Nevertheless we consider this a bug.  Please report
--16025-- it at http://valgrind.org/support/bug_reports.html.
--16025-- WARNING: unhandled arm64-linux syscall: 241
--16025-- You may be able to write your own handler.
--16025-- Read the file README_MISSING_SYSCALL_OR_IOCTL.
--16025-- Nevertheless we consider this a bug.  Please report
--16025-- it at http://valgrind.org/support/bug_reports.html.
--16025-- WARNING: unhandled arm64-linux syscall: 241
--16025-- You may be able to write your own handler.
--16025-- Read the file README_MISSING_SYSCALL_OR_IOCTL.
--16025-- Nevertheless we consider this a bug.  Please report
--16025-- it at http://valgrind.org/support/bug_reports.html.
--16025-- WARNING: unhandled arm64-linux syscall: 241
--16025-- You may be able to write your own handler.
--16025-- Read the file README_MISSING_SYSCALL_OR_IOCTL.
--16025-- Nevertheless we consider this a bug.  Please report
--16025-- it at http://valgrind.org/support/bug_reports.html.
Error:
The sys_perf_event_open() syscall returned with 38 (Function not 
implemented) for event (armv8_pmuv3_0/br_mis_pred/).
/bin/dmesg | grep -i perf may provide additional information.

==16059==
==16059== Process terminating with default action of signal 15 (SIGTERM)
==16059==    at 0x486F974: __read_nocancel (syscall-template.S:84)
==16059==    by 0x48D02F: read (unistd.h:44)
==16059==    by 0x48D02F: perf_evlist__prepare_workload (evlist.c:1471)
==16059==    by 0x41AB0F: __cmd_record (builtin-record.c:898)
==16059==    by 0x41AB0F: cmd_record (builtin-record.c:1873)
==16059==    by 0x476C7F: run_builtin (perf.c:302)
==16059==    by 0x476F37: handle_internal_command (perf.c:354)
==16059==    by 0x407093: run_argv (perf.c:398)
==16059==    by 0x407093: main (perf.c:520)
==16059==
==16059== HEAP SUMMARY:
==16059==     in use at exit: 56,239 bytes in 226 blocks
==16059==   total heap usage: 1,164 allocs, 938 frees, 2,238,979 bytes 
allocated
==16059==
==16059== 12 bytes in 1 blocks are definitely lost in loss record 1 of 6
==16059==    at 0x4844B88: malloc (in 
/usr/lib/valgrind/vgpreload_memcheck-arm64-linux.so)
==16059==
==16059== 344 bytes in 5 blocks are possibly lost in loss record 3 of 6
==16059==    at 0x4846CFC: calloc (in 
/usr/lib/valgrind/vgpreload_memcheck-arm64-linux.so)
==16059==
==16059== 5,736 bytes in 125 blocks are possibly lost in loss record 4 of 6
==16059==    at 0x4844B88: malloc (in 
/usr/lib/valgrind/vgpreload_memcheck-arm64-linux.so)
==16059==
==16059== LEAK SUMMARY:
==16059==    definitely lost: 12 bytes in 1 blocks
==16059==    indirectly lost: 0 bytes in 0 blocks
==16059==      possibly lost: 6,080 bytes in 130 blocks
==16059==    still reachable: 50,147 bytes in 95 blocks
==16059==         suppressed: 0 bytes in 0 blocks
==16059== Reachable blocks (those to which a pointer was found) are not 
shown.
==16059== To see them, rerun with: --leak-check=full --show-leak-kinds=all
==16059==
==16059== For counts of detected and suppressed errors, rerun with: -v
==16059== ERROR SUMMARY: 3 errors from 3 contexts (suppressed: 0 from 0)
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025== Warning: invalid file descriptor -1 in syscall close()
==16025==
==16025== HEAP SUMMARY:
==16025==     in use at exit: 26,640 bytes in 209 blocks
==16025==   total heap usage: 1,202 allocs, 993 frees, 2,455,112 bytes 
allocated
==16025==
==16025== 328 bytes in 7 blocks are definitely lost in loss record 2 of 6
==16025==    at 0x4844B88: malloc (in 
/usr/lib/valgrind/vgpreload_memcheck-arm64-linux.so)
==16025==
==16025== 344 bytes in 5 blocks are possibly lost in loss record 3 of 6
==16025==    at 0x4846CFC: calloc (in 
/usr/lib/valgrind/vgpreload_memcheck-arm64-linux.so)
==16025==
==16025== 6,000 bytes in 126 blocks are possibly lost in loss record 5 of 6
==16025==    at 0x4844B88: malloc (in 
/usr/lib/valgrind/vgpreload_memcheck-arm64-linux.so)
==16025==
==16025== LEAK SUMMARY:
==16025==    definitely lost: 328 bytes in 7 blocks
==16025==    indirectly lost: 0 bytes in 0 blocks
==16025==      possibly lost: 6,344 bytes in 131 blocks
==16025==    still reachable: 19,968 bytes in 71 blocks
==16025==         suppressed: 0 bytes in 0 blocks
==16025== Reachable blocks (those to which a pointer was found) are not 
shown.
==16025== To see them, rerun with: --leak-check=full --show-leak-kinds=all
==16025==
==16025== For counts of detected and suppressed errors, rerun with: -v
==16025== ERROR SUMMARY: 3 errors from 3 contexts (suppressed: 0 from 0)
root@localhost:~#

(pretty much the same as previous patch)

Cheers,
John

>
> -Andi
>
> .
>



  reply	other threads:[~2018-10-02 10:21 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-09-25 15:53 perf segmentation fault from NULL dereference John Garry
2018-09-27  3:00 ` Andi Kleen
2018-10-02 10:20   ` John Garry [this message]
2018-09-27 16:02 ` Jiri Olsa
2018-10-02 10:41   ` John Garry
2018-10-02 11:16     ` Jiri Olsa
2018-10-03 11:36       ` [PATCH] perf tools: Allocate id array in perf_event__synthesize_event_update_cpus Jiri Olsa
2018-10-03 14:08         ` John Garry
2018-10-03 14:16           ` Jiri Olsa
2018-10-03 21:20             ` [PATCH] perf tools: Store ids for events with their own cpus perf_event__synthesize_event_update_cpus Jiri Olsa
2018-10-04  9:20               ` John Garry
2018-10-09 10:00                 ` Jiri Olsa
2018-10-12 13:25                   ` John Garry
2018-10-15 19:15                     ` Arnaldo Carvalho de Melo
2018-10-16  9:10                       ` John Garry
2018-10-16 10:47                         ` Jiri Olsa
2018-10-18  6:18               ` [tip:perf/urgent] perf evsel: " tip-bot for Jiri Olsa

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=334a114a-bb59-9582-a913-efbb01bd68c5@huawei.com \
    --to=john.garry@huawei.com \
    --cc=acme@kernel.org \
    --cc=ak@linux.intel.com \
    --cc=alexander.shishkin@linux.intel.com \
    --cc=jolsa@redhat.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linuxarm@huawei.com \
    --cc=mark.rutland@arm.com \
    --cc=mingo@redhat.com \
    --cc=namhyung@kernel.org \
    --cc=peterz@infradead.org \
    --cc=will.deacon@arm.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).