Re: [qemu-s390x] [PULL SUBSYSTEM s390x 2/3] s390-bios: Skip bootmap signature entries

[Cornelia Huck]

On Mon, 13 May 2019 09:44:37 +0200
Christian Borntraeger <address@hidden> wrote:

> On 13.05.19 09:42, Cornelia Huck wrote:
> > On Sat, 11 May 2019 08:15:21 +0200
> > Thomas Huth <address@hidden> wrote:
> >   
> >> On 10/05/2019 15.59, Christian Borntraeger wrote:  
> >>> Shall we cc stable this?    
> >>
> >> I think I'd rather not do it unless someone really speaks up that they
> >> urgently need it. If we could use the binary from the master branch, I'd
> >> say go for it, but in this case we'd need to build a separate
> >> s390-ccw.img for this (without the DASD passthrough patches), and since
> >> the stable branch does not get that much testing attention from all the
> >> s390x developers, you'd end up with a firmware binary in the stable
> >> branch that is not very well tested... This does not sound very
> >> appealing to me.  
> > 
> > FWIW, I have rebuilt the bios for the stable tree in the past, when a
> > bios patch had been picked. In this case, however, I would need to rely
> > on someone else to sanity-check the binary.
> > 
> > How likely are folks to run -stable QEMU with a bootmap containing
> > signatures? It would be one more QEMU version with toleration for this,
> > but I expect distros to pick up this one anyway?  
> 
> Yes, I will try to push this into distros. I usually try to push things also
> to stable, but this might be more important for the kernel.
> 

Yes, while the QEMU stable branch is useful, the kernel stable
backports cover a lot more.