Planning for Cross-LFS/Multi-Architecture 7.x Release

Ryan.Oliver at pha.com.au Ryan.Oliver at pha.com.au
Mon Apr 18 21:49:50 PDT 2005







>Ryan.Oliver at pha.com.au wrote these words on 04/18/05 21:55 CST:
>
>> Please folks, have a bit of a think about how things actually
>work,
>> or actually take a look at how it hangs together, before going off
>> the deep end
>
>Nobody's gone off the deep end, Ryan. We're just asking questions.
>The LFS Editor who announced the proposed changes said there was
>2 pitfalls, I mentioned one more and Bruce mentioned another.

Neither are unresolvable

>
>The LFS editor said the solution to *one* of the pitfalls is to
>install a couple of BLFS packages during the LFS installation.
>
>I commented that using the new proposed methodology would take some
>of the zest out of LFS, for me. I'm entitled to an opinion. You
>somehow took this personally.

No I phrased it in such a way to get everyones attention, and to
immediately squish wrong assumptions.

> It isn't a reflection on your work,
>Ryan, but an opinion that I offered (mainly that we somehow keep
>the existing build method alive), as Jim mentioned that LFS would
>*only* offer the new build method.

We can incorporate a "cross-lfs lite" (ie: wiki instructions) to
current book with minimal impact to achieve that aim...

>As far as the *CURRENT* builders and readers of the book, do you
>expect an influx of new readers because we offer cross-build?

Yes. lots of folks buying x86_64's , G5's etc out there and
we do not cater for them. x86_64 is probably the most asked for.
No way to build x86_64 bi-arch from, say, a knoppix cd unless you
cross-build it...

> I
>guess what I mean is, will this be the new target audience, and
>not the x86 "bitty" (whatever that is) builders?
>
>Your message did not help me understand the changes, Ryan. It came
>off as a defensive, poorly phrased way of saying that building LFS
>on the x86 platform is for sissys.

Heh ;-) it is the path most travelled, not much interesting to see ;-)

A none too subtle posting on my part to try to get folks thinking
outside of our current limitations and x86-centricness. True, should
have been phrased better instead of going into clue-bat mode...

I can understand the concerns, it is a fair bit of change
to have to cope with (but hey, plfs was equally as disruptive), but
overall the concerns listed are easily addressed, once you spend some
time playing with the build and comparing it to current lfs you will
see... and have a whole new bag of tricks to take away with you

 Regards
[R]




More information about the lfs-dev mailing list