Re: moonjelly vs tsearch

>> News from Fabien's bleeding edge compiler zoo: apparently GCC 10 
>> 20190831 thinks the tsearch test should produce different output than 
>> 20190824 did. It looks like the parsing of question marks change...
>
> Indeed, that looks pretty strange.
>
>> I can't wait for next week's installment.
>
> I'll relaunch gcc trunk compilation before the end of the week.

It did not do anything then, but this Saturday update fixed the issue.

-- 
Fabien.