On 12.07.2013, at 22:30, Anthony Liguori wrote: > Setting it to LE forces a byte swap when host != guest endian but > this makes no sense at all. > > Herve made the suggestion upon observing that word writes/reads > were broken into byte writes/reads in such a way as to assume > devices are interpret registers as LE. > > However, even if this were a problem, marking the region as LE is > not useful because what's essentially happening here is that LE is > open coded. So by marking it LE in MemoryRegionOps, we're doing a > superflous swap. > > Now, the portio code is suspicious to begin with. The dispatch > layer really has no purpose in splitting I/O requests in the first > place... > > Cc: Hervé Poussineau <[email protected]> > Cc: Alex Graf <[email protected]> > Cc: Benjamin Herrenschmidt <[email protected]> > Signed-off-by: Anthony Liguori <[email protected]>
Acked-by: Alexander Graf <[email protected]> Let's take a look at the PReP breakage with a test case after we fixed everyone else again. Alex > --- > ioport.c | 1 - > 1 file changed, 1 deletion(-) > > diff --git a/ioport.c b/ioport.c > index 79b7f1a..89b17d6 100644 > --- a/ioport.c > +++ b/ioport.c > @@ -183,7 +183,6 @@ static void portio_write(void *opaque, hwaddr addr, > uint64_t data, > static const MemoryRegionOps portio_ops = { > .read = portio_read, > .write = portio_write, > - .endianness = DEVICE_LITTLE_ENDIAN, > .valid.unaligned = true, > .impl.unaligned = true, > }; > -- > 1.8.0 >
