Re: [GENERAL] Is float8 a reference type?

On Fri, Sep 22, 2017 at 8:05 PM, Pavel Stehule <pavel.stehule@gmail.com> wrote:
> yes, it is 8 bytes on 64-bit.

Thanks!

> I don't think so it is good idea to write 64bit only extensions.

I agree, but how about this?:

    if (FLOAT8PASSBYVAL) {
      datums = (Datum *)floats;
    } else {
      datums = palloc0(arrlen * sizeof(Datum));
      for (i = 0; i < arrlen; i++) {
        datums[i] = Float8GetDatum(floats[i]);
      }
    }

Thanks,
Paul