This change cleans up the string code in a number of ways:

- For memcpy(), fix bug in prefetch and increase distance to 3 lines;
  optimize for unaligned data; do all loads before wh64 to make memcpy
  safe for forward-overlapping calls; etc.  Performance is improved.

- Use new copy_byte() function on tilegx to spread a single byte value
  out into a full word using the shufflebytes instruction.

- Clean up header include ordering to be more canonical, and remove
  spurious #undefs of function names.
Signed-off-by: NChris Metcalf <cmetcalf@tilera.com>

Our internal process shares memcpy, memset, etc., with libc, and
we did some minor tweaking as part of moving from uclibc to glibc,
which is now reflected in the kernel versions of these files.

There are no semantic changes in this commit, just whitespace
(memcpy_32.S now properly uses tabs), naming (memmove.c instead
of memmove_32.c, since TILE-Gx shares the file with TILEPro),
and a couple of other minor tweaks.
Signed-off-by: NChris Metcalf <cmetcalf@tilera.com>

This change is the core kernel support for TILEPro and TILE64 chips.
No driver support (except the console driver) is included yet.

This includes the relevant Linux headers in asm/; the low-level
low-level "Tile architecture" headers in arch/, which are
shared with the hypervisor, etc., and are build-system agnostic;
and the relevant hypervisor headers in hv/.
Signed-off-by: NChris Metcalf <cmetcalf@tilera.com>
Acked-by: NArnd Bergmann <arnd@arndb.de>
Acked-by: NFUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Reviewed-by: NPaul Mundt <lethal@linux-sh.org>