On Wed, 31 Jul 2013 16:50:02 +0800, David Crayford wrote:

>On 31/07/2013 12:25 PM, Paul Gilmartin wrote:
>> (You mean IEEE *isn't* the default!? I guess I'll put it in my Makefile.)
>
>Heads up if you want to make FLOAT(IEEE) the default. I got burned by
>the DB2 ODBC library which only supports hex floating point. The good
>news is there are runtime functions for conversion  __fp_htob() and
>__fp_btoh().
> 
IBM feels compelled to do things differently from the rest of the world.
(Yah, I know; "diachronic".)

I understand that the Perl validation suite fails because it presumes IEEE.
And Perl suffers from a Unicode-EBCDIC conflict.  It would be interesting
to see a port of Perl in Ehanced ASCII and IEEE flavor.

How does printf() handle it?  I know it's ASCII-savvy; I haven't tried IEEE.
Is there a single subroutine with multiple paths for ASCII/EBCDIC and
HFP/IEEE (DFP?) and AMODE 31/64 and ...?  Or are there (Cartesian
product) different subroutines selected by the compiler according to
feature test macros?

Is the pseudo-assembler listing generated by the compiler using the
output from the optimizing backend, somehow fed back to the compiler?

-- gil

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Reply via email to