Re: [Xen-devel] [PATCH 09/29] libxc: check for xc_vcpu_setcontext failure in xc_domain_resume_any

[Ian Campbell <Ian.Campbell@xxxxxxxxxx>]

On Fri, 2013-11-01 at 16:34 +0000, Ian Jackson wrote:
> Ian Campbell writes ("Re: [PATCH 09/29] libxc: check for xc_vcpu_setcontext 
> failure in xc_domain_resume_any"):
> > On Wed, 2013-10-30 at 20:51 +1300, Matthew Daley wrote:
> > > Coverity-ID: 1090352
> > > Signed-off-by: Matthew Daley <mattjd@xxxxxxxxx>
> > > ---
> > >  tools/libxc/xc_resume.c |    6 +++++-
> > >  1 file changed, 5 insertions(+), 1 deletion(-)
> > > 
> > > diff --git a/tools/libxc/xc_resume.c b/tools/libxc/xc_resume.c
> > > index 50724f2..a4c0f53 100644
> > > --- a/tools/libxc/xc_resume.c
> > > +++ b/tools/libxc/xc_resume.c
> > > @@ -211,7 +211,11 @@ static int xc_domain_resume_any(xc_interface *xch, 
> > > uint32_t domid)
> > >  
> > >      /* Reset all secondary CPU states. */
> > >      for ( i = 1; i <= info.max_vcpu_id; i++ )
> > > -        xc_vcpu_setcontext(xch, domid, i, NULL);
> > > +        if ( xc_vcpu_setcontext(xch, domid, i, NULL) != 0 )
> > > +        {
> > > +            ERROR("Couldn't reset vcpu state");
> > > +            goto out;
> > 
> > The label is inside an x86 specific ifdef and therefore this breaks on
> > ARM.
> 
> Surely this is a mistake.  I mean, the label should be outside the
> ifdef.  Anything else is too tangled.

Yes, which is what Matthew did in v2, which I already applied.

Ian.



_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel