RFR(S): 8029668: Kithcensink crashed with guarantee(Assembler::is_simm13(disp)) failed: Do not match large constant offsets

[Igor Veresov]

Thanks, Chris!

igor

On Dec 10, 2013, at 9:20 AM, Christian Thalinger <christian.thalinger at oracle.com> wrote:

> Thanks!  Looks good.
> 
> On Dec 9, 2013, at 11:26 PM, Igor Veresov <igor.veresov at oracle.com> wrote:
> 
>> Sure, also fixed the formatting in the functions I’ve touched.
>> 
>> http://cr.openjdk.java.net/~iveresov/8029668/webrev.01/
>> 
>> On Dec 9, 2013, at 7:11 PM, Christian Thalinger <christian.thalinger at oracle.com> wrote:
>> 
>>> Nit:  PhaseRegAlloc *ra vs. PhaseRegAlloc* ra
>>> 
>>> Looks good.
>>> 
>>> On Dec 9, 2013, at 4:22 PM, Igor Veresov <igor.veresov at oracle.com> wrote:
>>> 
>>>> A quick fix for the problem, since we’re under time constraints. Bailout if we try to reference a stack location that we can't encode.  Note that, MachSpillCopy emission already behaved this way.
>>>> 
>>>> Webrev: http://cr.openjdk.java.net/~iveresov/8029668/webrev.00/
>>>> 
>>>> Thanks!
>>>> igor
>>> 
>> 
>