Instrumenting call sites.

[Remi Forax]

On 11/29/2012 08:15 PM, MacGregor, Duncan (GE Energy Management) wrote:
> I've been doing a little project in my spare time to instrument our call sites and check the number of receivers being seen by each site in a running system. As expected most sites only see one or two classes and so go through the fast path. There are however some sites which see slightly more receivers than our limit for the fast paths but are still getting called an awful lot. So I've got two questions.
>
>
>    1.  What is the recommended maximum tree size and longest path for composed MethodHandles (can we push our fast path limit a little higher)?

The real limit is the number of jumps that you have to do before finding 
the target, 3 to 5 is the maximum.
After you can still use a kind of vtable local to a callsite (see the 
dispatch map in [1]).

>    2.  What's my best option for connecting call sites to specific points in byte code and hence to their source locations?

use ASM :)
collect the source with visitSource() and the line with visitLineNumber 
to get the user readable info about the callsite.
rewrite your invokedynamic to call only one bootstrap method that acts 
as a tranpoline to call the runtime bootstrap methods.
With that you can add method hande before and after [2] the call to 
collect receiver data.

>
> Thanks for any help you can give.,
>
> Duncan.

Rémi

[1] 
https://code.google.com/p/jsr292-cookbook/source/browse/trunk/bimorphic-cache/src/jsr292/cookbook/bicache/RT.java
[2] https://code.google.com/p/jsr292-cookbook/ (see before and after)