This sounds like a question that could be answered by the folks on
hotspot-compiler-...@openjdk.java.net.
... peter
On 06/09/14 10:32, Brendan Gregg via dtrace-discuss wrote:
G'Day dtrace-discuss,
I saw an opportunity to demo DTrace recently, but jstack() for the Oracle JVM
keeps returning incomplete stacks. I've tried a number of different JDK
versions from 6 to 8, and options including libdtrace_forceload.so. Anyone know
of a JDK version or workaround that does jstack() reliably?
# dtrace -n 'profile-99 /execname == "java"/ { @[jstack(100, 8000)] = count();
}'
[...]
org/mozilla/javascript/ScriptableObject.createSlot(Ljava/lang/String;II)Lorg/mozilla/javascript/ScriptableObject$Slot;*
0x884acce8200002da
1
sun/nio/ch/SocketChannelImpl.read(Ljava/nio/ByteBuffer;)I*
0xffffffff20007f4b
1
org/mozilla/javascript/ScriptRuntime.newObjectLiteral([Ljava/lang/Object;[Ljava/lang/Object;[ILorg/mozilla/javascript/Context;Lorg/mozilla/javascript/Scriptable;)Lorg/mozilla/javascript/Scriptable;*
0xa20000041
1
[...]
A flame graph based on this output is more than 50% stacks like this. It looks
a lot like omit frame pointer optimization, but there is no
-XX:-omit-frame-pointer option I can find. :-)
I can use other profilers, but wanted to give DTrace a shot at this. Thanks,
Brendan
--
http://www.brendangregg.com
*dtrace-discuss* | Archives <https://www.listbox.com/member/archive/184261/=now>
<https://www.listbox.com/member/archive/rss/184261/21483692-e4c1a27c> | Modify
<https://www.listbox.com/member/?&;> Your Subscription [Powered by Listbox]
<http://www.listbox.com>
-------------------------------------------
dtrace-discuss
Archives: https://www.listbox.com/member/archive/184261/=now
RSS Feed: https://www.listbox.com/member/archive/rss/184261/25769126-e243886f
Modify Your Subscription:
https://www.listbox.com/member/?member_id=25769126&id_secret=25769126-8d47a7b2
Powered by Listbox: http://www.listbox.com
