RFC: Move forwarding ptr into the object

[Christine Flood]

I feel this is too risky for too little benefit.

Yes, it's cleaner to account for the forwarding pointer in the object, but I'd be willing to bet a beer that there are assumptions all over the place about object layout and chasing down all of them will be frustrating and time consuming.

Plus, as Andrew says, it requires more Shenandoah specific code in core hotspot and hurts our case for acceptance upstream.


Christine


----- Original Message -----
> From: "Andrew Haley" <aph at redhat.com>
> To: shenandoah-dev at openjdk.java.net
> Sent: Tuesday, September 27, 2016 9:24:53 AM
> Subject: Re: RFC: Move forwarding ptr into the object
> 
> On 27/09/16 14:18, Roman Kennke wrote:
> > Great work! Some comments...
> 
> I'm a little bit worried that changing object layout in such a
> fundamental way might lead to objections merging this, and might
> break more things.  At the moment almost everything which gets
> an OOP knows the object layout.  Maybe we're just setting ourselves
> up for more pain.
> 
> Andrew.
> 
>