Re: [Gcl-devel] GCL work with Win7 VM, 11 October 2013

[Donald Winiecki]

On Sat, Oct 12, 2013 at 8:29 AM, Camm Maguire <address@hidden> wrote:

<snip>

> I fixed the maxima segfault -- the mingw compiler fails to tail recurse
> equal1 (see predicate.c) and blew out the stack, so handcoded the call.
> Would be great if someone could file a bug and get this fixed someday.
>
> maxima works, and am checking hol88 (apparently needs work).  acl2 has a
> non-gcl related issue I'm writing Matt about, and we don't have the
> other prerequisites to try axiom.  Would like to chase down the latter
> at some point.  For now don't see any mingw reason to delay release,
> though I would like to get working builds of acl2, axiom and hol88 in
> general.
>

I've been observing the MinGW list for awhile trying to get a bead on
what issues are dominant for them. I'll get the tail-recursion problem
reported soon. First, I have to get the current MinGW setup and
supporting tools installed and confirm the same behaviour occurs.
Unfortunately, that is taking more time to work through than I had at
first imagined.

For now, fortunately, the currently specified setup successfully still
works and builds everything up to GCL 2.6.10pre on Win32 (MinGW
3.1.0-1, MSYS 1.0.9, msysDTK 1.0.1, binutils 2.14.90, GCC 3.3.1,
mingw-runtime 3.14, w32api 3.11). (While it's not to worry over, the
only exception to this is that GCL 2.6.9 can only be built and run on
WinXP -- it will not build or run on Win7.)

I'll get myself prepared for the set of CAS tools, for when the time comes.

But on a much less-pressing note, I know I can soon finish a new
version of the job aid for building
Win32 using git instead of CVS. Look for it here in the coming week.

_don