Re: Suggestions for how ARM Ltd could help Debian?
In message <3FE19783.firstname.lastname@example.org>
Lauri Leukkunen <email@example.com> wrote:
> Umm... How do you define "without incident"? As in "zero modification to
> the build system"?
That's right. Or, in some cases, changes can be automated.
> If it really is that easy to cross compile everything, how come there
> are all these ARM based build machines used to build Debian and
> Familiar? Wouldn't it be so much easier to just have one really fast
> powerpc or x86 machine doing all the work those netwinder/ipaq/etc
> machines have been doing all these years?
I think the answer is quite complex. Firstly, as I'll be the first to
point out, some things _do_ need modification. I'd guess around 5% or
so programs, maybe a bit more. For a few dozen base packages for a
bootstrap system (or even the basic Debian packages), this is more than
realistic. You could argue that patching these packages/encouraging
the maintainers to make them cross compile might be rather less effort
(and pay dividends later) than building a system like scratchbox.
As for Debian, well, we are talking > 10000 packages, plus the Debian
build system sits on top of configure so it is homogenized, so isn't
quite the same techinical problem. There has been work for cross
compiling of Debian packages, and I expect it might one day work
properly, but there's also Debian policy considerations to take into
Finally, as has been pointed out, the ARM build machines are doing ok
as is, and aren't in desperate need of lots of CPU power. And you
still need to have an ARM machine to debug things on if things go wrong.
The answer might be different for the 68K port, however.
> > I don't doubt that your requirements may justify the project, but I do
> > object to the implications made by this project to someone who might be
> > relatively ignorant on the subject.
> I'm somewhat curious to know what you mean with the implications? You
> make it sound like it is something bad?
My main problem with it is that it suggests that cross compiling is
hard, when in most cases, it isn't, and the methods for doing it are
quite straight forward. In other words, it doesn't seem like it has
fully explored or discussed the alternatives fairly.
Peter Naulls - firstname.lastname@example.org | http://www.chocky.org/
The RISC OS Browser Issue - http://www.chocky.org/unix/browser.html