[MLton] Re: Porting MLton to Mac OS X

[Stephen Weeks]

> One issue will be that the Power PC chip is big-endian where as x86
> is little-endian. So e.g. the order of the 4 bytes in a 32bit word is
> reversed - which will definately have importance in a compiler.

This should be no problem, since we've already ported to Sparc which
is also big endian.  There is a single boolean in the basis library,
MLton.Platform.Arch.isBigEndian, which controls the endianness.  And
there are a few, but not many, places in the runtime sources where
endianness matters.  Look for "defined (__sparc__)".

> For what it's worth I'll gladly answer questions (but I don't always
> pay a lot of attention to this list and you can send to me directly).

Please send (or cc) all questions to the list.  That way, it will be
archived, and you will get more eyeballs looking at the problem.
Since Anoq reads this list digested, you could include him if you want
his immediate attention.