[15] RFR: 8237894: CTW: C1 compilation fails with assert(x->type()->tag() == f->type()->tag()) failed: should have same type

[Jamsheed C M]

Thanks a lot for the feedback.

On 12/02/2020 16:44, Aleksey Shipilev wrote:
> Looks good! I have not run any extensive testing on this.
i ran some tests tier1-5, and hs-precheckin-comp, results were ok, 
forgot to put status here (had updated in JBS as confidential)
> But now I have second thoughts. Do we foresee any performance troubles from not GVN'ing null
> constants in C1, though?

i am not sure about all use case, i think only static finals are 
affected, service loader based initialization is an use case probably, 
but it is interface.

these error never occurred in any other testing before , so use cases 
were limited ?

> Maybe we should err on the side of caution and hash in the type as well?
hmm. yes
> That would probably be awkward for subtypes. Maybe we should relax the assert for null constants
> instead? This assumes the type mismatch is not catastrophic.

i get many asserts one after the other when i start suppressing it.

Best regards,

Jamsheed