Re: [PATCH 01/12] sb_edac: move PCI IDs to pci_ids.h

[Aristeu Rozanski <arozansk@redhat.com>]

On Wed, Oct 30, 2013 at 11:21:49AM -0600, Bjorn Helgaas wrote:
> On Wed, Oct 30, 2013 at 11:19 AM, Mauro Carvalho Chehab
> <m.chehab@samsung.com> wrote:
> > Em Wed, 30 Oct 2013 17:40:20 +0100
> > Borislav Petkov <bp@alien8.de> escreveu:
> >
> >> On Wed, Oct 30, 2013 at 12:26:55PM -0400, Aristeu Rozanski wrote:
> >> > According to the comment, it should be done before submitting upstream.
> >> >
> >> > Signed-off-by: Aristeu Rozanski <arozansk@redhat.com>
> >> > ---
> >> >  drivers/edac/sb_edac.c  |   21 ++-------------------
> >> >  include/linux/pci_ids.h |   11 +++++++++++
> >> >  2 files changed, 13 insertions(+), 19 deletions(-)
> >> >
> >> > diff --git a/drivers/edac/sb_edac.c b/drivers/edac/sb_edac.c
> >> > index e04462b..4cdd948 100644
> >> > --- a/drivers/edac/sb_edac.c
> >> > +++ b/drivers/edac/sb_edac.c
> >> > @@ -57,26 +57,9 @@ static int probed;
> >> >   */
> >> >
> >> >  /*
> >> > - * FIXME: For now, let's order by device function, as it makes
> >> > - * easier for driver's development process. This table should be
> >> > - * moved to pci_id.h when submitted upstream
> >>
> >> Why, is anything else besides sb_edac.c using those?
> >>
> > I think that this patch makes sense, as the other Sandy Bridge registers
> > are at pci_ids.h. So, it is a matter of consistency.
> >
> > Also, I won't doubt that other drivers could need to access those, as
> > there are other things there on a few of those PCI IDs.
> >
> > Yet, on patch 12/12, you're adding the Ivy Bridge specific PCI IDs
> > inside the driver.
> >
> > So, IMHO, you should either move all to pci_ids.h or to keep them inside
> > the driver.
> >
> > My personal taste would be to move them all to pci_ids.h.
> 
> As long as it follows the guideline at the top of pci_ids.h (add them
> there only if they are shared between multiple drivers), I'm fine with
> putting them in pci_ids.h.

Fair enough, I think we can drop the first patch. I checked and all the
patches apply without the first one and it builds fine. Mauro, want me
to resubmit the patchset anyway?

-- 
Aristeu