I shouldn't assume (you know what happens)... But I assume they are asking
how to compile GCC from source on a machine running (for example) some ARM
chip running Linux to get an executable GCC that will run on (for example)
an Intel x64 chip running windows.

I assume (there it is again) that they are compiling on one well known
"architecture" targeting another well known (but different) "architecture".

I'm a GCC noob, but I think this basically amounts to them needing a cross
compiler. Beyond that... Is probably a question for Jonathan the great.

Jackson Pfeffer

On Fri, Sep 16, 2022, 11:28 Jonathan Wakely via Gcc-help <
gcc-help@gcc.gnu.org> wrote:

> On Fri, 16 Sept 2022 at 17:05, Chengyin Yao via Gcc-help
> <gcc-help@gcc.gnu.org> wrote:
> >
> > Yea so I'm just wondering, how to compile gcc to (my own) architecture
> cpu that has a completely different architecture?
>
> I'm not sure what you mean by "move gcc", and what is this "completely
> different architecture"? The best answers depend on what you mean.
>
> Have you created your own CPU with its own instruction set? Then you
> want to "port" GCC (and binutils for the assembler and linker) to your
> architecture. That's quite a lot of work. You could start by reading
> https://gcc.gnu.org/onlinedocs/gccint/Back-End.html and
> https://wiki.osdev.org/Porting_GCC_to_your_OS
>