RFR 8226800 [lworld][c1] SharedRuntime::fixup_callers_callsite is wrong when caller is C1

[Ioi Lam]

Hi Tobias,

Thanks for the review. I've fixed the comment as you suggested and pushed.

- Ioi

On 6/26/19 4:35 AM, Tobias Hartmann wrote:
> Hi Ioi,
>
> looks good to me. Maybe change "may be passed as unpacked fields" to "may be passed as fields".
>
> Best regards,
> Tobias
>
> On 26.06.19 07:47, Ioi Lam wrote:
>> https://bugs.openjdk.java.net/browse/JDK-8226800
>> http://cr.openjdk.java.net/~iklam/valhalla/8226800-fixup_callers_callsite-c1.v01/
>>
>> SharedRuntime::fixup_callers_callsite() needs to know whether the caller
>> is compiled by C1 or C2, and use the appropriate entry point, due to the
>> difference in how C1 and C2 handle calling convention for value parameters.
>>
>> I think there are probably similar bugs lurking around. E.g.,
>> Method::from_compiled_entry() probably returns the wrong value when
>> we have a C1-compiled caller. I'll investigate more.
>>
>> I also updated TestCallingConventionC1.java to test more combinations
>> of the interaction between C1 and C2.
>>
>>
>> Thanks
>> - Ioi
>>
>>
>>
>>
>>
>>