FRV: work around a possible compiler bug

Work around a possible bug in the FRV compiler.

What appears to be happening is that gcc resolves the
__builtin_constant_p() in kmalloc() to true, but then fails to reduce the
therefore constant conditions in the if-statements it guards to constant
results.

When compiling with -O2 or -Os, one single spurious error crops up in
cpuup_callback() in mm/slab.c.  This can be avoided by making the memsize
variable const.

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

arch/frv/Makefile

@@ -88,7 +88,7 @@ ASFLAGS		+= -mno-fdpic
 # make sure the .S files get compiled with debug info
 # and disable optimisations that are unhelpful whilst debugging
 ifdef CONFIG_DEBUG_INFO
-CFLAGS		+= -O1
+#CFLAGS		+= -O1
 AFLAGS		+= -Wa,--gdwarf2
 ASFLAGS		+= -Wa,--gdwarf2
 endif

mm/slab.c

@@ -1163,7 +1163,7 @@ static int __cpuinit cpuup_callback(struct notifier_block *nfb,
 	struct kmem_cache *cachep;
 	struct kmem_list3 *l3 = NULL;
 	int node = cpu_to_node(cpu);
-	int memsize = sizeof(struct kmem_list3);
+	const int memsize = sizeof(struct kmem_list3);
 
 	switch (action) {
 	case CPU_LOCK_ACQUIRE: