Re: [PATCH v2 2/4] KVM: s390: vsie: Fix length of facility list shadowed

[Heiko Carstens]

On Wed, Nov 08, 2023 at 11:30:09AM +0100, Nina Schoetterl-Glausch wrote:
> On Tue, 2023-11-07 at 18:11 +0100, Claudio Imbrenda wrote:
> > On Tue,  7 Nov 2023 13:31:16 +0100
> > Nina Schoetterl-Glausch <nsg@xxxxxxxxxxxxx> wrote:
> > 
> > [...]
> > 
> > > -obj-y	+= smp.o text_amode31.o stacktrace.o abs_lowcore.o
> > > +obj-y	+= smp.o text_amode31.o stacktrace.o abs_lowcore.o facility.o
> > >  
> > >  extra-y				+= vmlinux.lds
> > >  
> > > diff --git a/arch/s390/kernel/facility.c b/arch/s390/kernel/facility.c
> > > new file mode 100644
> > > index 000000000000..5e80a4f65363
> > > --- /dev/null
> > > +++ b/arch/s390/kernel/facility.c
> > 
> > I wonder if this is the right place for this?
> 
> I've wondered the same :D
> > 
> > This function seems to be used only for vsie, maybe you can just move
> > it to vsie.c? or do you think it will be used elsewhere too?
> 
> It's a general STFLE function and if I put it into vsie.c I'm not sure
> that, if the same functionality was required somewhere else, it would be
> found and moved to a common location.
> 
> I was also somewhat resistant to calling a double underscore function from
> vsie.c. Of course I could implement it with my own inline asm...
> 
> The way I did it seemed nicest, but if someone else has a strong opinion
> I'll defer to that.

I think it is ok to have new file for just this. It is better than what
we've done too often in the past: dump new functionality to some more or
less random file instead. The usual victim would have been setup.c.

So I prefer a new file, even if we end up with only one function there.