Request for review (S): 6589834 deoptimization problem with -XX:+DeoptimizeALot

[Changpeng Fang]

Thanks, John:

On 04/22/09 15:53, John Rose wrote:
> To avoid those uncertainties, I suggest passing the absolute value of 
> GraphKit::_sp, not a relative adjustment.  -- John
At the time of processing inline_intrinsic, all arguments have already 
been temporarily popped off the stack. So we still need to
calculate the original GraphKit::_sp using the number of arguments. In 
these cases, we have to pass GraphKit::_sp + nargs
which is essentially the same as that we pass the nargs and thus adjust 
_sp before the uncommon trap.

Thanks,

Changpeng

*  *




> On Apr 22, 2009, at 3:42 PM, Vladimir Kozlov wrote:
>
>> For multianewarray passing ndimensions parameter is incorrect since
>> it is decremented for recursive calls. You have to add an additional
>> parameter (nargs ?) for expand_multianewarray() to pass original
>> ndimensions values to all recursive calls.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.openjdk.java.net/pipermail/hotspot-compiler-dev/attachments/20090422/4198e166/attachment.html