On Thu, 24 Jul 2014 09:03:47 +0200
Sascha Hauer <s.hauer@xxxxxxxxxxxxxx> wrote:
> On Wed, Jul 23, 2014 at 09:14:08AM +0200, Sascha Hauer wrote:
> > On Thu, Jul 17, 2014 at 03:54:07PM +0400, Antony Pavlov wrote:
> > > Also drop nowhere defined CONFIG_EHCI_DESC_BIG_ENDIAN.
> > >
> > > Signed-off-by: Antony Pavlov <antonynpavlov@xxxxxxxxx>
> >
> > Applied, thanks
>
> And reverted it.
>
> > > -#if defined CONFIG_EHCI_DESC_BIG_ENDIAN
> > > -#define ehci_readl(x) (*((volatile u32 *)(x)))
> > > -#define ehci_writel(a, b) (*((volatile u32 *)(a)) = ((volatile u32)b))
> > > -#else
> > > -#define ehci_readl(x) cpu_to_le32((*((volatile u32 *)(x))))
> > > -#define ehci_writel(a, b) (*((volatile u32 *)(a)) = \
> > > - cpu_to_le32(((volatile u32)b)))
> > > -#endif
> > > +static inline void ehci_writel(const unsigned int val, __u32 __iomem *regs)
> > > +{
> > > + writel(val, regs);
> > > +}
>
> ehci_writel has the argument order (reg, val), not (val, reg)
>
> Putting a 'writel' in a function name with this argument order wasn't
> the best idea...
Sorry!
I have made this patch while I was working on AR933x USB support (big-endian EHCI).
I suppose this patch was initial buggy version and I have forgot to ammend fixup.
--
Best regards,
Antony Pavlov
_______________________________________________
barebox mailing list
barebox@xxxxxxxxxxxxxxxxxxx
http://lists.infradead.org/mailman/listinfo/barebox
