RE: [PATCH v2 01/26] iommu/fsl: Sprinkle some __init* annotations

[Varun Sethi <Varun.Sethi-KZfg59tc24xl57MIdRCFDg@public.gmane.org>]

Hi Emil,

> -----Original Message-----
> From: Emil Medve [mailto:Emilian.Medve-eDlz3WWmN0ll57MIdRCFDg@public.gmane.org]
> Sent: Friday, January 30, 2015 7:54 PM
> To: Sethi Varun-B16395; iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org;
> joro-zLv9SwRftAIdnm+yROfE0A@public.gmane.org; jroedel-l3A5Bk7waGM@public.gmane.org
> Subject: Re: [PATCH v2 01/26] iommu/fsl: Sprinkle some __init* annotations
> 
> Hello Varun,
> 
> 
> On 01/30/2015 03:20 AM, Sethi Varun-B16395 wrote:
> > Hi Emil,
> >
> >> -----Original Message-----
> >> From: Emil Medve [mailto:Emilian.Medve-eDlz3WWmN0ll57MIdRCFDg@public.gmane.org]
> >> Sent: Friday, January 30, 2015 12:26 PM
> >> To: Sethi Varun-B16395; iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org;
> >> joro-zLv9SwRftAIdnm+yROfE0A@public.gmane.org; jroedel-l3A5Bk7waGM@public.gmane.org
> >> Subject: Re: [PATCH v2 01/26] iommu/fsl: Sprinkle some __init*
> >> annotations
> >>
> >> Hello Varun,
> >>
> >>
> >> On 01/30/2015 12:46 AM, Sethi Varun-B16395 wrote:
> >>> Hi Emil,
> >>>
> >>>> -----Original Message-----
> >>>> From: Emil Medve [mailto:Emilian.Medve-eDlz3WWmN0ll57MIdRCFDg@public.gmane.org]
> >>>> Sent: Wednesday, January 28, 2015 8:05 PM
> >>>> To: iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org; joro-zLv9SwRftAIdnm+yROfE0A@public.gmane.org;
> >>>> jroedel-l3A5Bk7waGM@public.gmane.org; Sethi Varun-B16395
> >>>> Cc: Medve Emilian-EMMEDVE1
> >>>> Subject: [PATCH v2 01/26] iommu/fsl: Sprinkle some __init*
> >>>> annotations
> >>>>
> >>>> Signed-off-by: Emil Medve <Emilian.Medve-eDlz3WWmN0ll57MIdRCFDg@public.gmane.org>
> >>>> ---
> >>>>  drivers/iommu/fsl_pamu.c        | 12 ++++++------
> >>>>  drivers/iommu/fsl_pamu_domain.c |  2 +-
> >>>>  2 files changed, 7 insertions(+), 7 deletions(-)
> >>>>
> >>>> diff --git a/drivers/iommu/fsl_pamu.c b/drivers/iommu/fsl_pamu.c
> >>>> index d958e65..652c34d 100644
> >>>> --- a/drivers/iommu/fsl_pamu.c
> >>>> +++ b/drivers/iommu/fsl_pamu.c
> >>>> @@ -50,7 +50,7 @@ struct pamu_isr_data {
> >>>>
> >>>>  static struct paace *ppaact;
> >>>>  static struct paace *spaact;
> >>>> -static struct ome *omt;
> >>>> +static struct ome *omt __initdata;
> >>>
> >>> [varun] omt shouldn't be initdata.
> >>
> >> It seems to be used only from probe()'s call tree: fsl_pamu_probe()
> >> and setup_omt(); both __init functions. Am I missing something?
> >>
> >>
> > Omt corresponds to PAMU operation mapping table, which is pointed to
> > by the PAMU OMT base registers. We are using operation mapping table
> > for operation translation.
> 
> Yes, but this 'omt' is just a pointer (should it be annotated as
> __iomem?) to the table and not used past the call sequence
> fsl_pamu_probe()->setup_omt(). For that matter 'omt' can be local to
> fsl_pamu_probe(). The hardware table/storage is used by the PAMU past
> the setup, but this pointer is not
Yes, you are right we won't use the pointer after the OMT table has been setup. We would just need the operation mapping entry index.

Acked-by: Varun Sethi <Varun.Sethi-KZfg59tc24xl57MIdRCFDg@public.gmane.org>