Re: [qemu-s390x] [Qemu-devel] [PATCH] hw/s390x: Fix the function arguments in the pci stub file

[Cornelia Huck]

On Mon, 11 Feb 2019 12:17:26 +0100
Thomas Huth <address@hidden> wrote:

> On 2019-02-11 12:04, Cornelia Huck wrote:
> > On Mon, 11 Feb 2019 11:54:45 +0100
> > Thomas Huth <address@hidden> wrote:
> >   
> >> On 2019-02-11 11:48, Cornelia Huck wrote:  
> >>> On Fri, 1 Feb 2019 09:23:56 +0100
> >>> Cornelia Huck <address@hidden> wrote:
> >>>     
> >>>> On Fri, 1 Feb 2019 07:46:40 +0100
> >>>> Thomas Huth <address@hidden> wrote:    
> >>>     
> >>>>> So I see two options now:
> >>>>>
> >>>>> 1) Finally really make the device optional, at least for new machine
> >>>>> types, so we can really disable CONFIG_PCI and get a working executable.
> >>>>>
> >>>>> 2) Scratch the idea completely to make this optional, always link the
> >>>>> s390-pci-bus.o and s390-pci-inst.o files unconditionally, and remove the
> >>>>> s390-pci-stub.c file.
> >>>>>
> >>>>> I assume options 2 is preferred, since we likely rather want to move
> >>>>> into the PCI direction in the long run, instead of ignoring it...      
> >>>>
> >>>> I think both options are viable, but option 1 is of course more work.
> >>>> The win there is that we could disable an entire subsystem.
> >>>>
> >>>> I guess that the basic questions are: How important is it that
> >>>> subsystems can be compiled out, and do we see a use case for a pci-less
> >>>> s390 machine in the future? We really don't want to spend much time on
> >>>> something of dubious use...    
> >>>
> >>> Any thoughts on this?
> >>>
> >>> I'm currently tending towards option 2 (and can cook up a patch for
> >>> that). Unless someone is already working on option 1 :)    
> >>
> >> Since nobody currently has a need to completely disable PCI, I think we
> >> should go with option 2.  
> > 
> > Hm... I'm wondering if we also should move S390_FEAT_ZPCI from the max
> > cpu model to the qemu cpu model (is there any reason not to turn it on
> > by default in tcg?)  
> 
> Migration compatibility? Wouldn't that cause problems when migrating
> back to older versions of QEMU?

Not sure. But we can still do that in a follow-on patch, if wanted.