On Thu, 13 Mar 2025 20:06:58 +0100
Rainer Orth <r...@cebitec.uni-bielefeld.de> wrote:

> Hi James,
> 
> > Our intention, tell me if you disagree, is that cobol is enabled if
> >
> > 1.  --enable-languages=all, and
> > 2.  the host and target are "known good", x86_64 or aarch64
> 
> you tend to forget there's a world outside of Linux, actually: as you
> can see at least in PRs cobol/119217 and cobol/119218, the code is
> currently riddled with lots of Linux-specific assumptions which break
> left and right on e.g. Darwin/x86_64 and Solaris/amd64.  

Hi Rainer, 

I'm sure to someone using another OS it looks like I forget they
exist.  I am sorry for any inconvenience we caused.  In my defense, I
tried to be clear about what we were bringing to the table.  Portabilty
has a cost, as you know. It was more important to us to have a working
compiler on one platform than an incomplete compiler that could be
built on 10.  

I cut my teeth on NetBSD, but I think choosy users choose Linux today,
in the main.  (Not so choosy users choose Windows; I'm well aware that
running on Windows will broaden our appeal significantly.)  

I'm grateful for the bug reports. I think the potential users of
COBOL on Darwin or Solaris is vanishingly small, but in principle
(believe it or not) I too prefer the code to be Posix, not Linux.  I
myself would be Darwin user, so there's that.  

I guess the most controversial engineering choice was to rely on
__int128 and _Float128.  Those gave us native processing and binary <->
string conversion.  Having been advised of real.cc, we're taking a
look.  If we can get 128-bit hardware computation more portably, that's
all to the good.  

But I don't see the point, in 2025, of gmp.  it's more important that 
COBOL take advantage of current hardware than run on a VAX.
That's where users are. Hardware computation has real benefits, and
software math has measurable impact.  When the IT budget has a line
item for electricity, that's what matters.  

I hope you'll agree with me that in the end, that's all that matters.
Users matter.  Otherwise, what's a compiler for?  

Kind regards, 

--jkl

Reply via email to