prep_zero_page: remove bogus BUG_ON

2.6.11 gave __GFP_ZERO's prep_zero_page a bogus "highmem may have to wait"
assertion.  Presumably added under the misconception that clear_highpage
uses nonatomic kmap; but then and now it uses kmap_atomic, so no problem.
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/page_alloc.c

@@ -305,7 +305,6 @@ static inline void prep_zero_page(struct page *page, int order, gfp_t gfp_flags)
 {
 	int i;
 
-	VM_BUG_ON((gfp_flags & (__GFP_WAIT | __GFP_HIGHMEM)) == __GFP_HIGHMEM);
 	/*
 	 * clear_highpage() will use KM_USER0, so it's a bug to use __GFP_ZERO
 	 * and __GFP_HIGHMEM from hard or soft interrupt context.