> > > I solved a problem with the pci_alloc_consistent function for my
> > > architecture. Now the pci_alloc_consistent function allocates uncachable
> > > memory (from a region starting from b7000000 address as you can see from
> > > the following debug messages pci_alloc_consistent: ).
> > >
> > > When I run I get a timeout: [ ... deletia ... ]
> > >
> > > Could you give me any helps about this problem please?
> >
> > Did you provide solutions for pci_{map,unmap}_single() too?
> > As I understand, SH is like MIPS ... you should be able to
> > steal a lot of code from there (that b7000000 address is much
> > like what the MIPS folk use, I seem to recall).
>
> What kind of solutions should I support for pci_{map,unmap}_single()?
Whatever is necessary on your platform ... :)
> Why could these functions give problems about timeout?
Lack of them, that is.
I've seen those sorts of timeouts in some cases when a host
controller (OHCI for example :) is given the wrong address,
since the devices will then ignore the data they're sent. If
the right address was given, but the data wasn't flushed,
then the same symptoms could occur.
> Now I have these functions:
If those work, then as far as I can tell they'd be OK. I don't
know the SH (except it has a MIPS instruction set :), but
I can imagine hardware where that approach suffices.
- Dave
_______________________________________________
[EMAIL PROTECTED]
To unsubscribe, use the last form field at:
http://lists.sourceforge.net/lists/listinfo/linux-usb-devel
