The computation for pgoff is incorrect, at least with (vma->vm_pgoff >> PAGE_SHIFT) involved. It is fixed with the available method if HPAGE_SIZE is concerned in page cache lookup. Cc: Michal Hocko <mhocko@xxxxxxx> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@xxxxxxxxxxxxxx> Cc: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> Signed-off-by: Hillf Danton <dhillf@xxxxxxxxx> --- --- a/mm/hugetlb.c Tue Dec 20 21:26:30 2011 +++ b/mm/hugetlb.c Tue Dec 20 21:40:44 2011 @@ -2315,8 +2315,7 @@ static int unmap_ref_private(struct mm_s * from page cache lookup which is in HPAGE_SIZE units. */ address = address & huge_page_mask(h); - pgoff = ((address - vma->vm_start) >> PAGE_SHIFT) - + (vma->vm_pgoff >> PAGE_SHIFT); + pgoff = linear_hugepage_index(vma, address); mapping = (struct address_space *)page_private(page); /* -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@xxxxxxxxx. For more info on Linux MM, see: http://www.linux-mm.org/ . Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/ Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>