On Wed, Apr 29, 2009 at 06:23:32PM +0200, Oliver Neukum wrote:
> Am Mittwoch 29 April 2009 17:43:12 schrieb Sarah Sharp:
> > On Tue, Apr 28, 2009 at 09:36:06AM +0200, Oliver Neukum wrote:
> > > Am Dienstag 28 April 2009 04:52:22 schrieb Sarah Sharp:
> > > > +/* TODO: copied from ehci.h - can be refactored? */
> > > > +/* xHCI spec says all registers are little endian */
> > > > +static inline unsigned int xhci_readl(const struct xhci_hcd *xhci,
> > > > + __u32 __iomem *regs)
> > > > +{
> > > > + return readl(regs);
> > > > +}
>
> > > No 64 bit version?
> >
> > I was told by Greg that 64-bit reads on the PCIe bus had issues on some
> > platforms. Greg, is that still true?
>
> Especially if that is true you should have a version that safely splits
> the read in two.
But I don't have any u64 register variables in the code, so why do I
need a 64-bit version of xhci_readl()? Or are you saying I should turn
all the 64-bit addresses from two u32s to one u64 and have a 64-bit
version of xhci_readl() and xhci_writel()? I'm not sure there's any
benefit to that.
Sarah
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
