Re: float4/float8/int64 passed by value with tsearchfixup

Gregory Stark írta:
> "Tom Lane" <tgl(at)sss(dot)pgh(dot)pa(dot)us> writes:
>
>
>> BTW, I trolled the contrib files for other v0 functions taking or
>> returning float4 or float8. I found seg_size (fixed it) and a whole
>> bunch of functions in earthdistance. Those use float8 not float4,
>> so they are not broken by this patch, but that module will have to
>> be v1-ified before we can consider applying the other part of the
>> patch.
>>
>
> So are you killing V0 for non-integral types? Because if not we should keep
> some sacrificial module to the regression tests to use to test for this
> problem.
>
> I don't see any way not to kill v0 for non-integral types if we want to make
> float4 and float8 pass-by-value. We could leave those pass-by-reference and
> just make bigint pass-by-value.
>

Just a note: time[stamp[tz]] also depend on either float8 or int64 and
they have to be the same pass-by-value or pass-by-reference as their
base storage types. There were crashes or regression failures if not.

--
----------------------------------
Zoltán Böszörményi
Cybertec Schönig & Schönig GmbH
http://www.postgresql.at/