Re: pci-sysfs resource mmap broken

[Greg KH <greg@kroah.com>]

On Tue, Apr 26, 2005 at 04:09:41PM +1000, Benjamin Herrenschmidt wrote:
> On Tue, 2005-04-26 at 15:33 +1000, Benjamin Herrenschmidt wrote:
> 
> > In a similar vein, the "resource" is exposing directly to userland the
> > content of "struct resource". This doesn't mean anything. The kernel is
> > internally playing all sort of offset tricks on these values, so they
> > can't be used for anything useful, either via /dev/mem, or for io port
> > accesses, or whatever.
> > 
> > Shouldn't we expose the BAR values & size rather here ? That is,
> > reconsitutes non-offset'd resources, possibly with arch help, or just
> > reading BAR to get base, and apply resource size & flags ?
> 
>   .../...
> 
> Ok, after a bit more thinking, I think I'll go that way for now, please
> let me know if you think I'm wrong:
> 
> rename "resource" to "resources" and make it contain a start address
> that matches the BAR value, that is something that has at least some
> sort of meaning in userland and can be passed to pci_mmap_page_range().
> To do that "translation", I'll read the BAR value, and use it as start,
> then use the resource size & flags.

That sounds fine with me.

thanks,

greg k-h