[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <CAPcyv4hSjL+kjR1is3O5BPp8K4763SvDhuUzSTc6O6JNtfW2ew@mail.gmail.com>
Date: Mon, 23 Sep 2019 14:27:52 -0700
From: Dan Williams <dan.j.williams@...el.com>
To: Michael Ellerman <michael@...erman.id.au>
Cc: Qian Cai <cai@....pw>,
"Aneesh Kumar K.V" <aneesh.kumar@...ux.ibm.com>,
Michael Ellerman <mpe@...erman.id.au>,
linuxppc-dev <linuxppc-dev@...ts.ozlabs.org>,
Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: "Pick the right alignment default when creating dax devices"
failed to build on powerpc
On Sun, Sep 22, 2019 at 5:04 AM Michael Ellerman <michael@...erman.id.au> wrote:
>
>
>
> On 21 September 2019 4:31:16 am AEST, Dan Williams <dan.j.williams@...el.com> wrote:
> >On Fri, Sep 20, 2019 at 11:18 AM Qian Cai <cai@....pw> wrote:
> >>
> >> On Fri, 2019-09-20 at 19:55 +0530, Aneesh Kumar K.V wrote:
> >> > Qian Cai <cai@....pw> writes:
> >> >
> >> > > The linux-next commit "libnvdimm/dax: Pick the right alignment
> >default when
> >> > > creating dax devices" causes powerpc failed to build with this
> >config. Reverted
> >> > > it fixed the issue.
> >> > >
> >> > > ERROR: "hash__has_transparent_hugepage"
> >[drivers/nvdimm/libnvdimm.ko] undefined!
> >> > > ERROR: "radix__has_transparent_hugepage"
> >[drivers/nvdimm/libnvdimm.ko]
> >> > > undefined!
> >> > > make[1]: *** [scripts/Makefile.modpost:93: __modpost] Error 1
> >> > > make: *** [Makefile:1305: modules] Error 2
> >> > >
> >> > > [1] https://patchwork.kernel.org/patch/11133445/
> >> > > [2]
> >https://raw.githubusercontent.com/cailca/linux-mm/master/powerpc.config
> >> >
> >> > Sorry for breaking the build. How about?
> >>
> >> It works fine.
> >
> >Thanks, but let's delay "libnvdimm/dax: Pick the right alignment
> >default when creating dax devices" until after -rc1 to allow Michael
> >time to ack/nak this new export.
>
> Thanks Dan. It looks fine to me:
>
> Acked-by: Michael Ellerman <mpe@...erman.id.au>
Thanks Michael. Aneesh, care to resend with this ack, but also reword
the changelog to say that this patch is a pre-requisite for the follow
on patch to pick dax device alignment? In other words, lets hide this
compile breakage from git-bisect by merging this export before the
consumer patch.
Powered by blists - more mailing lists