Thanks for all the help on that, it works-- I just compiled my first
arm "hello world"....I guess that stuff on the intel website is suspect.
Which brings me to another question. After the binutils/egcs/glibc
builds, the instructions call for building the kernel image to be
downloaded.
It is a little bit ambiguous as to which compiler should be used to build
this image-- the gcc native to the host or the arm-linux-gcc we just
built. Intel calls for something like this:
Download the source and unzip/untar:
%cd /home/~user~/armbuild/
%gzip -dc /home/~user~/gz/linux-2.2.2.tar.gz | tar xvf -
%cd linux
%gzip -dc /home/~user~/gz/patch-2.2.2-rmk5.gz | patch -p1
%gzip -dc /home/~user~/gz/diff-2.2.2-rmk5-np3.gz | patch -p1
Build the kernel and choose defaults when prompted:
%make config
%make zImage
This seems wrong to me-- it will build a kernel
for the i686 using the native gcc compiler.
Don't we need to edit or configure the make to
build an arm target and produce a zImage in linux/arch/arm/boot?
THanks,
Dave Borja
MMT Corp
unsubscribe: body of `unsubscribe linux-arm' to [EMAIL PROTECTED]
