[Openmcl-devel] 1.12 release plans

[Andrew Shalit]

I don’t think so.  The current version will continue to work on 32-bit x86 targets.  If there are critical changes in future versions, these can be back-ported.  I see any reason to put work into separating out the 32-bit back end for an ongoing life of its own.  Just use it as it currently exists.

> On Feb 2, 2018, at 3:04 AM, Matthew Stickney <mtstickney at gmail.com> wrote:
> 
> On 2/2/2018 12:36 AM, Hans Hübner wrote:
>> It makes spend to have the resources that are available to improve CCL not waste their precious time maintaining compatibility with an architecture that is obsolete.
> 
> Isn't that an argument for separating the backends, though? To paraphrase a message I mistakenly sent off-list, I've still got x86 machines that I use, and at work we've still got customers running x86 boxes, so they're definitely still around. Why throw away a working backend if the target hardware is still being used? Splitting the x86 and x86_64 backends isn't free, but it ought to be a one-time cost that minimizes the amount of time the maintainers have to spend dealing with x86.
> 
> -Matt Stickney
> _______________________________________________
> Openmcl-devel mailing list
> Openmcl-devel at clozure.com
> https://lists.clozure.com/mailman/listinfo/openmcl-devel