On Mon, 2017-10-09 at 13:40 -0700, Dan Williams wrote:
> On Mon, Oct 9, 2017 at 1:25 PM, Verma, Vishal L
> <vishal.l.ve...@intel.com> wrote:
> > On Sat, 2017-10-07 at 10:07 -0700, Dan Williams wrote:
> > > On Thu, Oct 5, 2017 at 6:53 PM, Vishal Verma <vishal.l.verma@intel
> > > .com
> > > > wrote:
> >
> > []
> >
> > > > NFIT_ARS_TIMEOUT = 90,
> > > > + NFIT_ARS_INJECT_INVALID = 2,
> > >
> > > If there is no usage of this define outside of nfit_test then it
> > > should be defined in tools/testing/nvdimm/test/nfit.h.
> >
> > Yes moved.
> >
> > >
> > > > };
> > > >
> > > > enum nfit_root_notifiers {
> > > > diff --git a/tools/testing/nvdimm/Kbuild
> > > > b/tools/testing/nvdimm/Kbuild
> > > > index d870520..3272ab5 100644
> > > > --- a/tools/testing/nvdimm/Kbuild
> > > > +++ b/tools/testing/nvdimm/Kbuild
> > > > @@ -69,6 +69,7 @@ libnvdimm-y += $(NVDIMM_SRC)/region_devs.o
> > > > libnvdimm-y += $(NVDIMM_SRC)/region.o
> > > > libnvdimm-y += $(NVDIMM_SRC)/namespace_devs.o
> > > > libnvdimm-y += $(NVDIMM_SRC)/label.o
> > > > +libnvdimm-y += $(NVDIMM_SRC)/badrange.o
> > >
> > > This hunk should go in the same patch that changes
> > > drivers/nvdimm/Makefile to add badrange.o so we don't have an
> > > nfit_test compile breakage gap between commits.
> >
> > The kernel's normal usage already has the kbuild change in the right
> > place (patch 1). This adds the use of badrange functions to
> > nfit_test
> > for the first time, and hence the kbuild addition here.
>
> It's not the nfit_test usages I'm worried about. The internal calls in
> libnvdimm will break. Try building tools/testing/nvdimm/libnvdimm.ko
> with only patch1 applied.
Ah I see what the problem is, I'll send a v3
_______________________________________________
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm
