Re: FOR COMMENT: void __iomem * and similar casts are Bad News

[Tony Lindgren <tony@atomide.com>]

* Russell King - ARM Linux <linux@arm.linux.org.uk> [080903 12:34]:
> On Wed, Sep 03, 2008 at 09:40:08AM -0700, Tony Lindgren wrote:
> > * Russell King - ARM Linux <linux@arm.linux.org.uk> [080903 00:56]:
> > > > - Use io_p2v() for initializing dynamic stuff as it can be a function
> > > >   for non-optimized multiboot binaries.
> > > 
> > > It can't become a function - it's used in structure initialization so
> > > must be constant.
> > 
> > Outside the drivers we can use XXX_IO_ADDRESS() or set it during resource
> > init with io_p2v().
> 
> The question is why do we need it?  If the correct physical address
> is passed, then things should work out just fine anyway, especially
> if drivers start to use ioremap rather than relying on all these fixed
> translations.

Hmm, that means fixing up resource init a bit in few places to avoid
sprinkling tests for cpu_class_is_omap1() to set the physical address.
Anyways, your approach sounds cleaner in the long run.

Eventually the whole io_p2v() can be removed as it's redundant and all
the arch stuff could use XXX_IO_ADDRESS().

> > > I still want to hear on the other build fix in the patch, and there's
> > > also a missing function for mmc stuff which I've not looked into yet.
> > 
> > The mpuio build fix? That is a correct fix.
> 
> No, see drivers/usb/host/ohci-omap.c:
> 
> -#ifdef CONFIG_ARCH_OMAP_OTG
> +#ifdef CONFIG_USB_OTG /* CHECKME */
> 

That's a Dave question.

Tony