Re: can device drivers return non-ram via vm_ops->nopage?

[Benjamin Herrenschmidt]

> exists because architectures haven't defined their private
> pgprot_writecombine() implementations, preferring instead to add
> to the preprocessor junk instead.

And it's not even about writecombine... Like writecombine is an
attribute of the PCI host bridge on pmacs, not a pgprot, while
cacheability issues cannot be always abstracted the same way on
different archs. Actually, GUARDED could be used for !writecombine
on ppc, but !GUARDED would allow prefetch and out of order IOs....

Ben.


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/