Reference.reachabilityFence

[Paul Sandoz]

> On 30 Nov 2015, at 12:48, Chris Hegarty <chris.hegarty at oracle.com> wrote:
> 
> I looked at the jdk webrev, looks fine.
> 
> Out of curiosity, what is the point of running the test with
> -Xint? Does it verify anything useful?
> 

It tests our assumptions are correct about the interpreter that finalisation should not occur (more specifically correct for the number of iterations). If the interpreter test passes we have more confidence that if the C1/C2 tests fail it has something to do with compilation and not their interpreted execution when warming up. FWIW it does not increase the execution time by any significant amount.

Paul.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 841 bytes
Desc: Message signed with OpenPGP using GPGMail
URL: <http://mail.openjdk.java.net/pipermail/hotspot-compiler-dev/attachments/20151130/6f797a42/signature.asc>