percpu: fix how @gfp is interpreted by the percpu allocator

From: Tejun Heo <[email protected]>
Date: Wed, 8 Oct 2014 12:01:52 -0400

When @gfp is specified, the percpu allocator is interested in whether it contains all of GFP_KERNEL or not. If it does, the normal allocation path is taken; otherwise, the atomic allocation path. Unfortunately, pcpu_alloc() was incorrectly testing for whether @gfp contains any part of GFP_KERNEL.

Fix it by testing "(gfp & GFP_KERNEL) != GFP_KERNEL" instead of "!(gfp & GFP_KERNEL)" to decide whether the allocation should be atomic or not.

---
Oops, noticed this while preparing the pull request for v3.18-rc1. Applied to percpu/for-3.18.

Thanks.

 mm/percpu.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

index e10f9f7..014bab6 100644
@@ -876,7 +876,7 @@ static void __percpu *pcpu_alloc(size_t size, size_t align, bool reserved,
 	static int warn_limit = 10;
 	struct pcpu_chunk *chunk;
 	const char *err;
-	bool is_atomic = !(gfp & GFP_KERNEL);
+	bool is_atomic = (gfp & GFP_KERNEL) != GFP_KERNEL;
 	int occ_pages = 0;
 	int slot, off, new_alloc, cpu, ret;
 	unsigned long flags;

1.9.3