From: Brendan Gregg <brendan.d.gregg@gmail.com>
To: Hemant Kumar <hemant@linux.vnet.ibm.com>
Cc: "linux-perf-use." <linux-perf-users@vger.kernel.org>
Subject: Re: perf user SDT markers
Date: Mon, 13 Apr 2015 10:19:50 -0700 [thread overview]
Message-ID: <CAE40pde4qwGdz0pGocxMkieF6xaS2nizyehc-GDQ4J1+K_KeyQ@mail.gmail.com> (raw)
In-Reply-To: <552B8C1A.8030606@linux.vnet.ibm.com>
G'Day Hemant,
Thanks for testing; here's how I compiled node.js:
# apt-get install -y systemtap-sdt-dev # provides "dtrace"
# git clone --depth 1 https://github.com/joyent/node
# cd node
# ./configure --with-dtrace
# make
node readelf:
# readelf -n node
Notes at offset 0x00000254 with length 0x00000020:
Owner Data size Description
GNU 0x00000010 NT_GNU_ABI_TAG (ABI version tag)
OS: Linux, ABI: 2.6.24
Notes at offset 0x00000274 with length 0x00000024:
Owner Data size Description
GNU 0x00000014 NT_GNU_BUILD_ID (unique build ID bitstring)
Build ID: 85dd5ef00ffde124f220b94c8a2b552d9274a89f
Notes at offset 0x00c45fd8 with length 0x000004e4:
Owner Data size Description
stapsdt 0x0000003c NT_STAPSDT (SystemTap probe descriptors)
Provider: node
Name: gc__start
Location: 0x0000000000bf7894, Base: 0x0000000000f252de, Semaphore:
0x0000000001245fa8
Arguments: 4@%esi 4@%edx 8@%rdi
stapsdt 0x0000003a NT_STAPSDT (SystemTap probe descriptors)
Provider: node
Name: gcstart
Location: 0x0000000000bf7895, Base: 0x0000000000f252de, Semaphore:
0x0000000001245faa
Arguments: 4@%esi 4@%edx 8@%rdi
stapsdt 0x0000003b NT_STAPSDT (SystemTap probe descriptors)
Provider: node
Name: gc__done
Location: 0x0000000000bf78a4, Base: 0x0000000000f252de, Semaphore:
0x0000000001245fac
Arguments: 4@%esi 4@%edx 8@%rdi
stapsdt 0x0000005c NT_STAPSDT (SystemTap probe descriptors)
[...etc...]
(the "gcstart" probe is one I added to test the underscore theory.)
Here's also the event files:
# ls /sys/kernel/debug/tracing/events/node/
enable gc__start_11 gc__start_2 gcstart_5 http__server__request
filter gcstart_1_1 gcstart_2 gc__start_6 http__server__request_1
gcstart gc__start_12 gc__start_3 gcstart_6 http__server__request_2
gc__start gc__start_13 gcstart_3 gc__start_7 http__server__request_3
gc__start_1 gc__start_14 gc__start_4 gcstart_7 http__server__request_4
gcstart_1 gc__start_15 gcstart_4 gc__start_8 net__server__connection
gc__start_10 gc__start_16 gc__start_5 gc__start_9 net__server__connection_1
This is after several iterations of --add and --del. Something seems
to be going wrong there. Thanks,
Brendan
On Mon, Apr 13, 2015 at 2:27 AM, Hemant Kumar <hemant@linux.vnet.ibm.com> wrote:
> Hi Brendan,
>
>
> On 04/13/2015 01:35 AM, Brendan Gregg wrote:
>>
>> On Sat, Apr 11, 2015 at 7:13 PM, Brendan Gregg
>> <brendan.d.gregg@gmail.com> wrote:
>>>
>>> G'Day,
>>>
>>> I was just trying Hemant Kumar's user SDT patch
>>> (http://lwn.net/Articles/618956/); anyone else tried it recently? I
>>> applied it to 4.0.0-rc6, and it creates instrumentation, but doesn't
>>> record the probes.
>>>
>>> This works (it finds the Node.js probes):
>>>
>>> # ./perf sdt-cache --dump
>>> /home/bgregg-testtest/node-standard/out/Release/node:
>>> %node:net__server__connection
>>> %node:net__stream__end
>>> %node:net__socket__read
>>> %node:net__socket__write
>>> %node:http__server__request
>>> %node:http__server__response
>>> %node:http__client__request
>>> %node:http__client__response
>>> %node:gc__done
>>> %node:gc__start
>>>
>>> But I can't "perf list" or record them:
>>>
>>> # ./perf record -e '%node:http__server__request' -aR
>>> invalid or unsupported event: '%node:http__server__request'
>>> Run 'perf list' for a list of valid events
>>> [...]
>>>
>>> stap works:
>>>
>>> # /root/systemtap-2.8/bin/stap -e 'probe
>>>
>>> process("/home/bgregg-testtest/node-standard/out/Release/node").mark("http__server__request")
>>> { println("hit"); }'
>>> hit
>>> hit
>>>
>>> Brendan
>>
>> Some more digging... So a simple C program with a DTRACE_PROBE1()
>> works, where the marker ends up being "%tick:loop":
>>
>> # ./perf record -e '%tick:loop' -a -g
>> [...]
>> ^C
>> [ perf record: Woken up 1 times to write data ]
>> [ perf record: Captured and wrote 0.334 MB perf.data (2 samples) ]
>>
>> But the %node markers don't work (eg, "%node:gc__start"). Maybe the
>> presence of underscores is breaking it? By adding some debug
>> statements, I dug this message out of yyparse(), which was failing:
>>
>> yyparse() yymsgbuf: syntax error, unexpected PE_NAME, expecting $end
>
>
> I rebased and tried the patches with perf for the latest tip:
>
> # ./perf sdt-cache --add /lib64/libc.so.6
> 8 events added for /lib64/libc.so.6
>
> # ./perf sdt-cache --dump
> /usr/lib64/libc-2.16.so:
> %libc:setjmp
> %libc:longjmp
> %libc:longjmp_target
> %libc:lll_futex_wake
> %libc:lll_lock_wait_private
> %libc:longjmp
> %libc:longjmp_target
> %libc:lll_futex_wake
>
>
> # ./perf record -e %libc:longjmp_target -aR sleep 1
> [ perf record: Woken up 1 times to write data ]
> [ perf record: Captured and wrote 0.286 MB perf.data ]
>
> Its able to probe and record on %libc:longjmp_target with an underscore.
>
> Also, can you please point me to the node.js with SDT markers you are using
> and can you dump me the readelf -n /path/to/node.js o/p?
>
> --
> Thanks,
> Hemant Kumar
>
next prev parent reply other threads:[~2015-04-13 17:20 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-04-12 2:13 perf user SDT markers Brendan Gregg
2015-04-12 20:05 ` Brendan Gregg
2015-04-13 9:27 ` Hemant Kumar
2015-04-13 17:19 ` Brendan Gregg [this message]
2015-04-14 10:25 ` Hemant Kumar
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=CAE40pde4qwGdz0pGocxMkieF6xaS2nizyehc-GDQ4J1+K_KeyQ@mail.gmail.com \
--to=brendan.d.gregg@gmail.com \
--cc=hemant@linux.vnet.ibm.com \
--cc=linux-perf-users@vger.kernel.org \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.