Andreas' 2013-02-08 change reverted some breakage for struct return

values from 2011-11-12, but in so doing reintroduced string
instructions to sysv.S that are not supported on all powerpc variants.
This patch properly copies the bounce buffer to destination in C code
rather than in asm.

I have tested this on powerpc64-linux, powerpc-linux and
powerpc-freebsd.  Well, the last on powerpc-linux by lying to
configure with

CC="gcc -m32 -msvr4-struct-return -mlong-double-64" \
CXX="g++ -m32 -msvr4-struct-return -mlong-double-64" \
/src/libffi-current/configure --build=powerpc-freebsd

and then

make && make CC="gcc -m32" CXX="g++ -m32" \
RUNTESTFLAGS=--target_board=unix/-m32/-msvr4-struct-return/-mlong-double-64\
 check

src/powerpc/sysv.S

@@ -142,19 +142,14 @@ L(float_return_value):
 #endif
 
 L(small_struct_return_value):
-	extrwi	%r6,%r31,2,19         /* number of bytes padding = shift/8 */
-	mtcrf	0x02,%r31	      /* copy flags to cr[24:27] (cr6) */
-	extrwi	%r5,%r31,5,19         /* r5 <- number of bits of padding */
-	subfic  %r6,%r6,4             /* r6 <- number of useful bytes in r3 */
-	bf-	25,L(done_return_value) /* struct in r3 ? if not, done. */
-/* smst_one_register: */
-	slw	%r3,%r3,%r5           /* Left-justify value in r3 */
-	mtxer	%r6                   /* move byte count to XER ... */
-	stswx	%r3,0,%r30            /* ... and store that many bytes */
-	bf+	26,L(done_return_value)  /* struct in r3:r4 ? */
-	add	%r6,%r6,%r30          /* adjust pointer */
-	stswi	%r4,%r6,4             /* store last four bytes */
-	b	L(done_return_value)
+	/*
+	 * The C code always allocates a properly-aligned 8-byte bounce
+	 * buffer to make this assembly code very simple.  Just write out
+	 * r3 and r4 to the buffer to allow the C code to handle the rest.
+	 */
+	stw %r3, 0(%r30)
+	stw %r4, 4(%r30)
+	b L(done_return_value)
 
 .LFE1:
 END(ffi_call_SYSV)