On Wed, 2010-02-17 at 20:04 +0000, Russell King wrote: > > That would be fine if we weren't already using PG_arch_1 for delaying > D-cache flushes for pages which aren't mapped, in the same way that > Sparc64 does. Interesting. Maybe we need more arch bits in there :-) > That doesn't cover this case though - the problem is not I/D cache > coherency - the problem has manifested itself as data corruption when > userspace reads the page. Yup, let's stick to the other thread. I still don't see why a PIO mapping API is needed, from what I've read so far, it's a case of mis-using the existing DMA API that could be fixed at the driver level. No ? Cheers, Ben. -- To unsubscribe from this list: send the line "unsubscribe linux-arch" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html