nommu: Provide is_vmalloc_addr() stub.

Introduced in commit-id 9e2779f and
ifdef'ed out for nommu in 8ca3ed8, both
approaches end up breaking the nommu build in different ways. An
impressive feat for a 2-liner.

Current is_vmalloc_addr() users fall in to two camps:

	- Determining whether to use vfree()/kfree()
	- Whether to do vmlist traversal (only /proc/kcore).

Since we don't support /proc/kcore on nommu, that leaves the
vfree()/kfree() determination use cases. nommu vfree() happens to be a
wrapper to kfree() anyways, so is_vmalloc_addr() can always return 0
and end up with the right behaviour.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

include/linux/mm.h

@@ -235,15 +235,22 @@ static inline int get_page_unless_zero(struct page *page)
 struct page *vmalloc_to_page(const void *addr);
 unsigned long vmalloc_to_pfn(const void *addr);
 
-#ifdef CONFIG_MMU
-/* Determine if an address is within the vmalloc range */
+/*
+ * Determine if an address is within the vmalloc range
+ *
+ * On nommu, vmalloc/vfree wrap through kmalloc/kfree directly, so there
+ * is no special casing required.
+ */
 static inline int is_vmalloc_addr(const void *x)
 {
+#ifdef CONFIG_MMU
 	unsigned long addr = (unsigned long)x;
 
 	return addr >= VMALLOC_START && addr < VMALLOC_END;
-}
+#else
+	return 0;
 #endif
+}
 
 static inline struct page *compound_head(struct page *page)
 {