lib/kstrtox.c: smaller _parse_integer()

Set "overflow" bit upon encountering it instead of postponing to the end
of the conversion. Somehow gcc unwedges itself and generates better code:

	$ ./scripts/bloat-o-meter ../vmlinux-000 ../obj/vmlinux
	_parse_integer                      177     139     -38

Inspired by patch from Zhaoxiu Zeng.

Link: http://lkml.kernel.org/r/20160826221920.GA1909@p183.telecom.by
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

lib/kstrtox.c

@@ -48,11 +48,9 @@ unsigned int _parse_integer(const char *s, unsigned int base, unsigned long long
 {
 	unsigned long long res;
 	unsigned int rv;
-	int overflow;
 
 	res = 0;
 	rv = 0;
-	overflow = 0;
 	while (*s) {
 		unsigned int val;
 
@@ -71,15 +69,13 @@ unsigned int _parse_integer(const char *s, unsigned int base, unsigned long long
 		 */
 		if (unlikely(res & (~0ull << 60))) {
 			if (res > div_u64(ULLONG_MAX - val, base))
-				overflow = 1;
+				rv |= KSTRTOX_OVERFLOW;
 		}
 		res = res * base + val;
 		rv++;
 		s++;
 	}
 	*p = res;
-	if (overflow)
-		rv |= KSTRTOX_OVERFLOW;
 	return rv;
 }