Hi Tomek,
2009/7/15 Pekka Enberg <penb...@cs.helsinki.fi>:
>> On Tue, 2009-07-14 at 21:12 +0200, Tomek Grabiec wrote:
>>> verify_no_recursion() checks whether all call sites of given subroutine
>>> all outside its body. It's a requirement for inlining to be possible.
>> Does this mean we won't be able to inline all subroutines and thus need
>> support for the 'jsr' and 'ret' bytecode instructions?
Tomek Grabiec wrote:
> Yes, we can't inline _all_ possible jsr/ret combination. However we
> can inline all subroutines which are generated by
> compiling try-finally blocks because of the special characteristics of
> generated code.
OK, I think we need to either fix this or raise an issue at Lighthouse.
People can do crazy things with run-time bytecode manipulation and we
can't just roll over and die if we encounter subroutines that cannot be
inlined. So we absolutely need to *support* all jsr/ret combinations but
we don't need to generate efficient code for them (if it keeps things
simple).
Pekka
------------------------------------------------------------------------------
Enter the BlackBerry Developer Challenge
This is your chance to win up to $100,000 in prizes! For a limited time,
vendors submitting new applications to BlackBerry App World(TM) will have
the opportunity to enter the BlackBerry Developer Challenge. See full prize
details at: http://p.sf.net/sfu/Challenge
_______________________________________________
Jatovm-devel mailing list
Jatovm-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/jatovm-devel
