Re: USB mass storage and ARM cache coherency

[Catalin Marinas <catalin.marinas@arm.com>]

On Tue, 2010-02-02 at 12:07 +0000, Oliver Neukum wrote:
> Am Dienstag, 2. Februar 2010 13:01:12 schrieb Catalin Marinas:
> > On Tue, 2010-02-02 at 11:48 +0000, Oliver Neukum wrote:
> > > Am Montag, 1. Februar 2010 18:29:14 schrieb Catalin Marinas:
> > > > +       if (usb_pipein(urb->pipe) && usb_pipetype(urb->pipe) == PIPE_BULK) {
> > > > +               void *ptr;
> > > > +               for (ptr = urb->transfer_buffer;
> > > > +                    ptr < urb->transfer_buffer + urb->transfer_buffer_length;
> > > > +                    ptr += PAGE_SIZE)
> > > > +                       flush_dcache_page(virt_to_page(ptr));
> > >
> > > Is it correct to limit this to BULK pipes?
> >
> > I'm not entirely sure. The flush_dcache_page() should only be called for
> > pages that may be mapped into user space (page cache pages). We don't
> > need this for control buffers. It was my impression that what's coming
> > from the mass storage layer intended for page cache pages has the
> > PIPE_BULK type (I may be wrong though).
> 
> For storage that is correct. But what about other sources of pages,
> for example iSCSI?

In the iSCSI case, does the HCD driver write directly to a page cache
page? Or it just fills in network packets that are copied to page cache
pages by the iSCSI code (sorry, I'm not familiar with this part of the
kernel). If the latter, the cache flushing in the HCD driver would not
help and it needs to be done in the iSCSI code.

-- 
Catalin