Il 24/01/2014 15:35, Peter Maydell ha scritto:
> (1) decide that the Xen ring buffers are little-endian even on big-endian
> CPUs
>
> (2) communicate the endianness of the Xen ring buffers via Xenstore, just
> like we do for sizeof(long), and let the guest use either endianness on any
> architecture.
You still have to make a choice about what you think
TARGET_WORDS_BIGENDIAN should be, and it's still going
to be wrong half the time and horribly confusing.
I just think this is completely the wrong solution to
the problem.
Theoretically the xenpv-softmmu machine shouldn't need any code that
depends on TARGET_WORDS_BIGENDIAN.
If we changed every #ifdef TARGET_WORDS_BIGENDIAN to if(), we could
compile it with "#define TARGET_WORDS_BIGENDIAN abort()".
Paolo
