[9] RFR (XS): 8153167: Allow stack walking pass through method handle intrinsic frames

[Zoltán Majó]

Hi,


let's put this RFR on hold -- I've just got an error message after my 
80th run with the Sun Studio Performance Analyzer. Sorry for the noise.

Best regards,


Zoltan

On 03/31/2016 03:02 PM, Zoltán Majó wrote:
> Hi,
>
>
> please review the patch for 8153167.
>
> https://bugs.openjdk.java.net/browse/JDK-8153167
>
> Problem: Currently, stack walking is stopped when frame of a method 
> handle intrinsic is encountered (i.e., a method handle intrinsic is 
> not considered as a "safe" sender) [1]. (Tobias drew my attention to 
> this issue.)
>
> The restriction was added to the frame walking code by JDK-8068945 [2] 
> and is actually not necessary: Since JDK-8068945 frames belonging to 
> method handle intrinsics are also treated as usual compiled frames [3].
>
>
> Solution: Remove the unnecessary condition.
>
> Webrev:
> http://cr.openjdk.java.net/~zmajo/8153167/webrev.00/
>
> Testing:
> - JPRT;
> - several runs of jstack, pstack, jfr, and Sun Studio Performance 
> Analyzer tracing a test generating long chains of method handle calls 
> (with -Xcomp +/-PreserveFramePointer);
> - RBT testing is in progress (-Xmixed/-Xcomp +-PreserveFramePointer) 
> and has shown no failures so far.
>
> Thank you!
>
> Best regards,
>
>
> Zoltan
>
> [1]: 
> http://hg.openjdk.java.net/jdk9/hs-comp/hotspot/file/8911d94e888b/src/cpu/x86/vm/frame_x86.cpp#l227
> [2]: http://hg.openjdk.java.net/jdk9/jdk9/hotspot/rev/382e9e4b3b71#l9.9
> [3]: http://hg.openjdk.java.net/jdk9/jdk9/hotspot/rev/382e9e4b3b71#l9.41