<AWT Dev> RFR(XS): 7196866: CTW fails on Solaris

[Phil Race]

I agree, that I don't think we need to go so far as that. I've concluded 
that the
original proposed fix appears harmless and is very probably OK and the
low incidence of such CTW issues over the years is surprising but 
reassuring.
I've been waiting (all afternoon!) for my fastdebug (*) build to complete so
I can see for sure how we enter the code that crashes in case there is 
anything
that we are missing that goes beyond what is needed by CTW. But you can
go ahead with your fix. Anything else I find we can deal with separately.

(*) fastdebug clearly refers solely to its runtime performance 
characterisrics,
not its build speed.

-phil.

On 5/16/2013 3:48 PM, John Rose wrote:
> On May 16, 2013, at 3:28 PM, Morris Meyer <morris.meyer at oracle.com 
> <mailto:morris.meyer at oracle.com>> wrote:
>
>> Perhaps I should throw an ExceptionInInitializerError w a 
>> CompileTheWorld message?
>
> That goes farther beyond native code hardening than I was thinking. 
>  If I were Phil I'd be uncomfortable with that, because it appears to 
> create more of a contract than simply "don't crash".
>
> — John