[9] RFR (S): 8160527: Check for final instance field updates can be omitted

Zoltán Majó zoltan.majo at oracle.com
Thu Jun 30 10:07:20 UTC 2016


Hi John,


On 06/29/2016 11:20 PM, John Rose wrote:
> [...]
> Reviewed.  I'm glad it was a simple tweak; I was afraid you'd have to 
> muck with assembly code.
>
> Kudos to the original designers of the CP cache mechanism, which 
> allows a CPCE to be (a) resolved for use by assembly code, and (b) 
> force re-resolution (because a "byte" in the CPCE indices is still 
> zero).  We use this trick for some corner cases with interface 
> invocation, IIRC.

thank you for the review!

Yes, the CP cache mechanism is nicely designed -- I am glad it was 
possible to solve the problem only by existing functionality provided by 
the CP cache mechanism.

Thank you!

Best regards,


Zoltan

>
> — John



More information about the hotspot-runtime-dev mailing list