On Sat, 2015-05-30 at 11:18 +0200, Geert Uytterhoeven wrote:
> On Sat, May 30, 2015 at 12:59 AM, Toshi Kani <toshi.kani@xxxxxx> wrote:
> > --- a/include/asm-generic/io.h
> > +++ b/include/asm-generic/io.h
> > @@ -785,8 +785,17 @@ static inline void __iomem *ioremap_wc(phys_addr_t offset, size_t size)
> > }
> > #endif
> >
> > +#ifndef ioremap_wt
> > +#define ioremap_wt ioremap_wt
> > +static inline void __iomem *ioremap_wt(phys_addr_t offset, size_t size)
> > +{
> > + return ioremap_nocache(offset, size);
> > +}
> > +#endif
> > +
> > #ifndef iounmap
> > #define iounmap iounmap
> > +
> > static inline void iounmap(void __iomem *addr)
> > {
> > }
> > diff --git a/include/asm-generic/iomap.h b/include/asm-generic/iomap.h
> > index 1b41011..d8f8622 100644
> > --- a/include/asm-generic/iomap.h
> > +++ b/include/asm-generic/iomap.h
> > @@ -66,6 +66,10 @@ extern void ioport_unmap(void __iomem *);
> > #define ioremap_wc ioremap_nocache
> > #endif
> >
> > +#ifndef ARCH_HAS_IOREMAP_WT
> > +#define ioremap_wt ioremap_nocache
> > +#endif
>
> Defining ioremap_wt in two different places in asm-generic looks fishy to me.
>
> If <asm/io.h> already provides it (either through asm-generic/io.h or
> arch/<arch>/include/asm/io.h), why does asm-generic/iomap.h need to define
> its own version?
>
> I see this pattern already exists for ioremap_wc...
Yes, this patchset follows the model of ioremap_wc. This duplication
was introduced by 9216efafc52 "asm-generic/io.h: Reconcile I/O accessor
overrides", while the original ioremap_wc support changed
asm-generic/iomap.h (1526a756fba). As described in patch 07, some
architectures define ioremap_xxx() locally as well.
It is too risky to do everything in one short. I will look into the
duplication issue as a separate item after this patchset is settled.
Thanks,
-Toshi
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxx. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>