Re: build system: HOSTCC -m64?

[Michael Tokarev]

02.12.2024 08:43, Nicholas Piggin wrote:
> On Sun Dec 1, 2024 at 3:54 PM AEST, Michael Tokarev wrote:

> > I stumbled across a somewhat stupid "issue", - while building skiboot
> > as part of qemu package build in debian, I tried running the build on
> > an i386 host (usually I do this on amd64 or aarch64 host).  And skiboot
> > build failed:

> > This is ${HOSTCC}, not ${CC} - which is supposed to be a compiler for
> > native architecture - it is building a helper binary to generate some
> > build-time tables, so any native/default architecture is fine.
> >
> > However, skiboot build procedure tries to force -m64 flag to the
> > default/native ${HOSTCC}:
> >
> >    Makefile.main: HOSTCFLAGS += $(call try-cflag,$(HOSTCC),-m64)
> >
> > And on i386, native gcc does support this option.  But it does no
> > necessary have all the headers/libraries to actually build any 64bit
> > executable - which is exactly the case here.
>
> Yeah, I think you have a point. Should fix it.
>
> extract-gcov.c should be fixed for 32-bit host, and -m64 flag
> removed. If you just remove -m64 I assume you hit the type
> errors that the original issue reported?
I tried to build skiboot on i386 after removing the mentioned line from
Makefile.main, - the compilation goes fine.  The commit introduced this
line says something about specific ppc variant - maybe the prob happens
only there, I dunno.  There's no warnings from current gcc, either,
besides an unrelated:

 core/hmi.c:860:25: warning: ‘pau’ may be used uninitialized 
[-Wmaybe-uninitialized]
   860 |                         pau_opencapi_dump_scoms(pau);

(which might be worth to check regardless).

I didn't try booting/testing the resulting firmware though.

Thanks,

/mjt