Re: [Emacs-diffs] master d826037 3/3: Remove the need for temacs.in

[Stefan Monnier <monnier@iro.umontreal.ca>]

>  The fingerprint is not guaranteed foolproof when computed from the *.o
>  files and is not guaranteed foolproof when made from the temacs.in
>  file either.
>
> Isn't it?

I don't see anything in the semantics of `ld` which makes the guarantees
we'd need, no.  Maybe current `ld` does, in practice, of course.

> The two approaches are not equally robust.

No, indeed.

> The temacs.in approach is as close as you're going to get to foolproof
> and future proof.

Might be.  Actually backpatching the fingerprint into `temacs` might be
in some cases more robust (it doesn't assume that both runs of `ld`
produce the same result).

But my point is just that it's a question of degree.
Just like the compilation time is a question of degree.

Noone is wrong or right here, these are just personal preferences.
I use machines whose age spans between 16 and 5 years (my main work
laptop is more than 10 years old because it's more or less the most
recent 4:3 I could find) so I value speed ups in compilation time
probably more than others on this list.


        Stefan