I'm experimenting a bit with CLFS2.
Unfortunately I don't own an ARM system, or any other architecture so I'm not realy cross-compiling.
The reason I'm trying this is because I want to learn more about sysroot-style builds.
I am wondering why the package order for building the cross-compile tools (Chapter 5) is chosen the way it is.
why are linux-headers in front of binutils? are they used during the build?
I noticed I can switch those 2 without errors.
also I found some other document describing the building of a toolchain which just has binutils first.
http://frank.harvard.edu/~coldwell/toolchain/
also, gcc is rebuilt after glibc to build against it, why isn't binutils rebuilt? isn't binutils still built against the host's glibc?
I'd also like to know how to handle updating toolchain stuff.
if binutils gets updated, should I rebuild glibc and gcc then?
or go gcc-static, glibc, gcc again? or doesn't binutils have much impact on the toolchain?
if gcc gets updated, should I rebuild binutils and/or glibc?
and if glibc gets updated, would it suffice to rebuild gcc afterwards or is the gcc-static, glibc, gcc cycle needed again?
I hope I'm not asking realy stupid questions, but I can't find a real answer/rationale elsewhere.
Thanks and keep up the good work,
Mathijs
_______________________________________________ Clfs-dev mailing list [email protected] http://ninja.linux-phreak.biz/mailman/listinfo/clfs-dev
