[m-rev.] for post-commit review: group *int_least* types together

[Peter Wang]

On Tue, 20 Feb 2018 17:05:32 +1100 (AEDT), "Zoltan Somogyi" <zoltan.somogyi at runbox.com> wrote:
> 
> 
> On Tue, 20 Feb 2018 16:13:37 +1100, Peter Wang <novalazy at gmail.com> wrote:
> 
> > On Tue, 20 Feb 2018 15:39:33 +1100 (AEDT), "Zoltan Somogyi" <zoltan.somogyi at runbox.com> wrote:
> > > 
> > > not needed.) I cannot test it, since I don't have
> > > any 32 bit machines. I asked you because you were
> > > the one to last touch the relevant parts of the compiler,
> > > as part of adding support for 64 bit unboxed integers.
> > 
> > Since you are going to be working on data representations, it's
> > worth setting up a 32-bit virtual machine for testing.
> 
> I agree: that is a good idea. Does anyone already have such a VM?
> If noone does, who can set one up most easily?

I use VirtualBox (point-and-click) but disk space might be a concern for you.

> > Alternatively,
> > unpack a 32-bit root filesystem of a Linux distribution somewhere, then
> > chroot/proot[1] into it from your 64-bit Linux host. You will know about
> > VMs/chroot, but let me know if you want detailed instructions for proot.
> 
> Thank you, but I don't think that would work for me. I do my Mercury work
> on an old laptop (bought in nov 2013), and it is usually only a few Gb away
> from running out of disk space. I don't think it has enough space
> for a complete Linux filesystem, however cut down it may be.

Alpine Linux takes < 400 MB here including development tools,
but not including the Mercury bootstrap compiler and workspace.
It could be tight, but let me know.

Peter