Re: [Qemu-ppc] [PULL 00/33] ppc-for-3.0 queue 20180612

[Greg Kurz]

On Wed, 13 Jun 2018 09:31:40 +1000
David Gibson <address@hidden> wrote:

> On Tue, Jun 12, 2018 at 08:46:03PM +0200, Greg Kurz wrote:
> > On Tue, 12 Jun 2018 16:55:08 +0100
> > Peter Maydell <address@hidden> wrote:
> >   
> > > On 12 June 2018 at 16:52, Greg Kurz <address@hidden> wrote:  
> > > > On Tue, 12 Jun 2018 15:34:22 +0100
> > > > Peter Maydell <address@hidden> wrote:
> > > >    
> > > >> On 12 June 2018 at 07:44, David Gibson <address@hidden> wrote:    
> > > >> > ----------------------------------------------------------------
> > > >> > ppc patch queue 2018-06-12
> > > >> >
> > > >> > Here's another batch of ppc patches towards the 3.0 release.  There's
> > > >> > a fair bit here, because I've been working through my mail backlog
> > > >> > after a holiday.  There's not much of a central theme, amongst other
> > > >> > things we have:
> > > >> >     * ppc440 / sam460ex improvements
> > > >> >     * logging and error cleanups
> > > >> >     * 40p (PReP) bugfixes
> > > >> >     * Macintosh fixes and cleanups
> > > >> >     * Add emulation of the new POWER9 store-forwarding barrier
> > > >> >       instruction variant
> > > >> >     * Hotplug cleanups
> > > >> >    
> > > >>
> > > >> Applied, thanks.    
> > >   
> > > > Oh... so we've broken bisect until Cedric's fix is merged.    
> > > 
> > > If you want me not to merge a pullreq you need to follow up to
> > > the cover letter saying "please don't merge this" or something
> > > similar. Otherwise I will not notice.  
> > 
> > Yeah I've only answered to the offending patch, sorry...  
> 
> And it was the middle of the night for me.
> 
> It's only with pretty old gcc, I think we can live with it, I intend
> to send another pullreq with the fixup as soon as I can.
> 

It was meeting day for me yesterday so I didn't have time to check with
other GCC versions, but I could finally do it this morning. They fail
all the same:

hw/intc/xics_kvm.c: In function ‘ics_set_kvm_state’:
hw/intc/xics_kvm.c:281:20: error: ‘ret’ may be used uninitialized in this
 function [-Werror=maybe-uninitialized]
             return ret;

Fedora28:

$ cc --version
cc (GCC) 8.1.1 20180502 (Red Hat 8.1.1-1)

Ubuntu 18.04:

$ cc --version
cc (Ubuntu 7.3.0-16ubuntu3) 7.3.0

Ubuntu 17.10:

$ cc --version
cc (Ubuntu 7.2.0-8ubuntu3.2) 7.2.0

RHEL 7.5:

$ cc --version
cc (GCC) 4.8.5 20150623 (Red Hat 4.8.5-28)

My concern is more that it requires --enable-debug to be passed to
configure. As noted by Cedric in another mail, -O2 seems to prevent
GCC to detect this obvious error...

pgp6P_96ur2xR.pgp
Description: OpenPGP digital signature