Naming the ARM(64) binaries for Windows

[Johannes Hielscher]

Am Wed, 02 Dec 2020 09:42:28 +0100
schrieb Henri Menke via tex-live <tex-live at tug.org>:

> > I just wanted to ask what TeX Live's official name would be for that
> > architecture in case it gets supported one day (in order to at least
> > try to be somewhat compatible in naming).  
> 
> It should be named `aarch64`, because the GNU triplet for this
> architecture is `aarch64-linux-gnu`

I second this. Back then (October 2017) when I suggested inclusion of
the aarch64-linux port into TL, I did some (superficial) research on
where the different names for that ISA come from, and concluded that
“aarch64” is the least troublesome one (wrt arm64, armv8h, armv8-a and
others).
This is reflected in the installer/tlmgr info text 
> GNU/Linux on ARM64 (aarch64-linux)
http://tug.org/svn/texlive/trunk/Master/tlpkg/TeXLive/TLUtils.pm?r1=46207&r2=46208

Unfortunately, the naming isn't always consistent across OS/distro
boundaries (cf. ppc64el|ppc64le, x86_64|x86-64|x64|amd64), and at least
some of this Babylonian confusion carried over into TL in the form of
x86_64-{linux,solaris,…} vs. amd64-{free,net}bsd.

Naming is balancing uniformity among our own naming vs. respect for
varying conceptions and history of OS platforms out there. If I were to
decide, I'd try to avoid multiple names for the same thing, and go with
what we have (here: aarch64) as far as no grave objections are raised
from the “new” side.

To add some perspective: Supposed we chose the “wrong” name here (i. e.
Microsoft decides to refuse allegiance with TL). Then it'll be at least
as much of an issue as explaining users of recent Windows “x64”
machines why to go with “win32” or “x86_64-cygwin”.

> > Naming the ARM(64) binaries for Windows
Just to be sure: we are talking about pure native 64-bit ARM64 aka ARMv8
here? Or has MS decided to also include some 32-bit legacy, not only
with their x86 emulation, but also native ARM (NB: ARMv8 also supports
32-bit mode, coined AArch32)?

Best,
Johannes