lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20150813180942.GF30356@kernel.org>
Date:	Thu, 13 Aug 2015 15:09:43 -0300
From:	Arnaldo Carvalho de Melo <acme@...nel.org>
To:	Masami Hiramatsu <masami.hiramatsu.pt@...achi.com>
Cc:	Namhyung Kim <namhyung@...nel.org>, Jiri Olsa <jolsa@...hat.com>,
	Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
	David Ahern <dsahern@...il.com>,
	Brendan Gregg <brendan.d.gregg@...il.com>
Subject: Re: [BUGFIX PATCH perf/core ] perf-probe: Fix to add missed brace
 around if block

Em Thu, Aug 13, 2015 at 06:55:41AM +0900, Masami Hiramatsu escreveu:
> The commit 75186a9b09e4 (perf probe: Fix to show lines of
> sys_ functions correctly) introduced a bug by a missed brace
> around if block. This fixes to add it.
> 
> Signed-off-by: Masami Hiramatsu <masami.hiramatsu.pt@...achi.com>

Thanks! now it works:

  [root@zoo ~]# perf probe 'get_timeval=sys_select:11 tv_sec=tv.tv_sec tv_usec=tv.tv_usec'
  Added new event:
    probe:get_timeval    (on sys_select:11 with tv_sec=tv.tv_sec tv_usec=tv.tv_usec)

  You can now use it in all perf tools, such as:

	perf record -e probe:get_timeval -aR sleep 1

  [root@zoo ~]# perf probe -l
    probe:get_timeval    (on SYSC_select:11@...me/git/linux/fs/select.c with tv_sec tv_usec)
    probe:vfs_getname    (on getname_flags:72@...me/git/linux/fs/namei.c with pathname)
  [root@zoo ~]#

Now we have:

  # strace -e select -p 960
  Process 960 attached
  select(13, [4 5 7 9 10 11 12], [], [5], {10, 0}) = 0 (Timeout)
  select(13, [4 5 7 9 10 11 12], [], [5], {9, 999999}) = 0 (Timeout)
  select(13, [4 5 7 9 10 11 12], [], [5], {10, 0}) = 0 (Timeout)
  select(13, [4 5 7 9 10 11 12], [], [5], {10, 0}) = 0 (Timeout)
  select(13, [4 5 7 9 10 11 12], [], [5], {0, 3072}) = 0 (Timeout)

At the same time as:

 # trace --ev probe:get_timeval -e select -p 960
     0.322 ( 0.013 ms): select(n: 13, inp: 0x55ac425996c0, outp: 0x55ac42599750, exp: 0x55ac425997e0, tvp: 0x7ffd3371fa30) ...
     0.322 (         ): probe:get_timeval:(ffffffff81228f15) tv_sec=10 tv_usec=0)
 10022.189 (10021.880 ms):  ... [continued]: select()) = 0 Timeout
 10022.681 ( 0.043 ms): select(n: 13, inp: 0x55ac425996c0, outp: 0x55ac42599750, exp: 0x55ac425997e0, tvp: 0x7ffd3371fa30) ...
 10022.681 (         ): probe:get_timeval:(ffffffff81228f15) tv_sec=9 tv_usec=999999)
 20040.673 (10018.035 ms):  ... [continued]: select()) = 0 Timeout
 20041.174 ( 0.046 ms): select(n: 13, inp: 0x55ac425996c0, outp: 0x55ac42599750, exp: 0x55ac425997e0, tvp: 0x7ffd3371fa30) ...
 20041.174 (         ): probe:get_timeval:(ffffffff81228f15) tv_sec=10 tv_usec=0)
 30063.041 (10021.912 ms):  ... [continued]: select()) = 0 Timeout
 30063.581 ( 0.042 ms): select(n: 13, inp: 0x55ac425996c0, outp: 0x55ac42599750, exp: 0x55ac425997e0, tvp: 0x7ffd3371fa30) ...
 30063.581 (         ): probe:get_timeval:(ffffffff81228f15) tv_sec=10 tv_usec=0)
 40085.448 (10021.909 ms):  ... [continued]: select()) = 0 Timeout
 40087.983 ( 0.033 ms): select(n: 13, inp: 0x55ac425996c0, outp: 0x55ac42599750, exp: 0x55ac425997e0, tvp: 0x7ffd3371fa30) ...
 40087.983 (         ): probe:get_timeval:(ffffffff81228f15) tv_sec=0 tv_usec=3072)
 40091.202 ( 3.252 ms):  ... [continued]: select()) = 0 Timeout

Seems to be working, now its just to add support in 'perf trace' for a
"get_timeval" wannabe tracepoint in the same fashion as there is support for
another wannabe tracepoint: "vfs_getname", add more of these where timevals are
copied from the kernel, etc.

Anyway, now 'perf probe' suits my needs with syscalls, applied!

Thanks,

- Arnaldo
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ