[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20080725134159.GA7493@amd.com>
Date: Fri, 25 Jul 2008 15:41:59 +0200
From: Joerg Roedel <joerg.roedel@....com>
To: FUJITA Tomonori <fujita.tomonori@....ntt.co.jp>
CC: mingo@...hat.com, tglx@...utronix.de, linux-kernel@...r.kernel.org,
iommu@...ts.linux-foundation.org, bhavna.sarathy@....com,
robert.richter@....com
Subject: Re: [PATCH] add iommu_num_pages helper function
On Fri, Jul 25, 2008 at 10:34:57PM +0900, FUJITA Tomonori wrote:
> On Fri, 25 Jul 2008 15:21:21 +0200
> Joerg Roedel <joerg.roedel@....com> wrote:
>
> > On Fri, Jul 25, 2008 at 10:15:06PM +0900, FUJITA Tomonori wrote:
> > > On Fri, 25 Jul 2008 14:57:58 +0200
> > > Joerg Roedel <joerg.roedel@....com> wrote:
> > >
> > > > Calculating the number of pages from given address and length numbers is a task
> > > > required in multiple IOMMU implementations. So implement this as a generic
> > > > function into the IOMMU helper code.
> > > >
> > > > Cc: FUJITA Tomonori <fujita.tomonori@....ntt.co.jp>
> > > > Signed-off-by: Joerg Roedel <joerg.roedel@....com>
> > > > ---
> > > > include/linux/iommu-helper.h | 1 +
> > > > lib/iommu-helper.c | 8 ++++++++
> > > > 2 files changed, 9 insertions(+), 0 deletions(-)
> > >
> > > Thanks. I planed to work on this after rc1. You beat me. :)
> > >
> > >
> > > > diff --git a/include/linux/iommu-helper.h b/include/linux/iommu-helper.h
> > > > index c975caf..f8598f5 100644
> > > > --- a/include/linux/iommu-helper.h
> > > > +++ b/include/linux/iommu-helper.h
> > > > @@ -8,3 +8,4 @@ extern unsigned long iommu_area_alloc(unsigned long *map, unsigned long size,
> > > > unsigned long align_mask);
> > > > extern void iommu_area_free(unsigned long *map, unsigned long start,
> > > > unsigned int nr);
> > > > +extern unsigned long iommu_num_pages(unsigned long addr, unsigned long len);
> > > > diff --git a/lib/iommu-helper.c b/lib/iommu-helper.c
> > > > index a3b8d4c..889ddce 100644
> > > > --- a/lib/iommu-helper.c
> > > > +++ b/lib/iommu-helper.c
> > > > @@ -80,3 +80,11 @@ void iommu_area_free(unsigned long *map, unsigned long start, unsigned int nr)
> > > > }
> > > > }
> > > > EXPORT_SYMBOL(iommu_area_free);
> > > > +
> > > > +unsigned long iommu_num_pages(unsigned long addr, unsigned long len)
> > > > +{
> > > > + unsigned long size = roundup((addr & ~PAGE_MASK) + len, PAGE_SIZE);
> > > > +
> > > > + return size >> PAGE_SHIFT;
> > > > +}
> > > > +EXPORT_SYMBOL(iommu_num_pages);
> > >
> > > This doesn't work since PAGE_SIZE != IOMMU_PAGE_SIZE on the majority
> > > of architectures.
> >
> > A quick grep in include/ tells me that only powerpc is declaring
> > IOMMU_PAGE_SIZE by now.
>
> Ah, I just meant the I/O size of IOMMUs.
>
> This macro doesn't work for IA64, POWER, and SPARC.
Is there any other architecture independent way to find out the page
size the IOMMU is using? If yes I can rework that patchset. If not, I
suggest that we (you?) introduce one in a follow-on patch series and fix
the iommu_num_pages function then.
Joerg
--
| AMD Saxony Limited Liability Company & Co. KG
Operating | Wilschdorfer Landstr. 101, 01109 Dresden, Germany
System | Register Court Dresden: HRA 4896
Research | General Partner authorized to represent:
Center | AMD Saxony LLC (Wilmington, Delaware, US)
| General Manager of AMD Saxony LLC: Dr. Hans-R. Deppe, Thomas McCoy
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists