mm/hugetlb: don't crash when HPAGE_SHIFT is 0
authorBenjamin Herrenschmidt <benh@kernel.crashing.org>
Thu, 31 Jul 2008 07:07:30 +0000 (00:07 -0700)
committerLinus Torvalds <torvalds@linux-foundation.org>
Fri, 1 Aug 2008 19:46:41 +0000 (12:46 -0700)
Some platform decide whether they support huge pages at boot time.  On
these, such as powerpc, HPAGE_SHIFT is a variable, not a constant, and is
set to 0 when there is no such support.

The patches to introduce multiple huge pages support broke that causing
the kernel to crash at boot time on machines such as POWER3 which lack
support for multiple page sizes.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mm/hugetlb.c

index d237a02..28a2980 100644 (file)
@@ -1283,7 +1283,12 @@ module_exit(hugetlb_exit);
 
 static int __init hugetlb_init(void)
 {
-       BUILD_BUG_ON(HPAGE_SHIFT == 0);
+       /* Some platform decide whether they support huge pages at boot
+        * time. On these, such as powerpc, HPAGE_SHIFT is set to 0 when
+        * there is no such support
+        */
+       if (HPAGE_SHIFT == 0)
+               return 0;
 
        if (!size_to_hstate(default_hstate_size)) {
                default_hstate_size = HPAGE_SIZE;