aboutsummaryrefslogtreecommitdiffstats
path: root/mm/hugetlb.c
diff options
context:
space:
mode:
authorKOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>2008-10-18 23:27:10 -0400
committerLinus Torvalds <torvalds@linux-foundation.org>2008-10-20 11:52:32 -0400
commit4b2e38ad703541f7845c2d766426148b8d1aa329 (patch)
treeaaafbec5325d15c38c382c655120fb6492c11f82 /mm/hugetlb.c
parente575f111dc0f27044e170580e7de50985ab3e011 (diff)
hugepage: support ZERO_PAGE()
Presently hugepage doesn't use zero page at all because zero page is only used for coredumping and hugepage can't core dump. However we have now implemented hugepage coredumping. Therefore we should implement the zero page of hugepage. Implementation note: o Why do we only check VM_SHARED for zero page? normal page checked as .. static inline int use_zero_page(struct vm_area_struct *vma) { if (vma->vm_flags & (VM_LOCKED | VM_SHARED)) return 0; return !vma->vm_ops || !vma->vm_ops->fault; } First, hugepages are never mlock()ed. We aren't concerned with VM_LOCKED. Second, hugetlbfs is a pseudo filesystem, not a real filesystem and it doesn't have any file backing. Thus ops->fault checking is meaningless. o Why don't we use zero page if !pte. !pte indicate {pud, pmd} doesn't exist or some error happened. So we shouldn't return zero page if any error occurred. Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Adam Litke <agl@us.ibm.com> Cc: Hugh Dickins <hugh@veritas.com> Cc: Kawai Hidehiro <hidehiro.kawai.ez@hitachi.com> Cc: Mel Gorman <mel@skynet.ie> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Diffstat (limited to 'mm/hugetlb.c')
-rw-r--r--mm/hugetlb.c22
1 files changed, 19 insertions, 3 deletions
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index ab79cd4dd23c..ce8cbb29860b 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -2071,6 +2071,14 @@ follow_huge_pud(struct mm_struct *mm, unsigned long address,
2071 return NULL; 2071 return NULL;
2072} 2072}
2073 2073
2074static int huge_zeropage_ok(pte_t *ptep, int write, int shared)
2075{
2076 if (!ptep || write || shared)
2077 return 0;
2078 else
2079 return huge_pte_none(huge_ptep_get(ptep));
2080}
2081
2074int follow_hugetlb_page(struct mm_struct *mm, struct vm_area_struct *vma, 2082int follow_hugetlb_page(struct mm_struct *mm, struct vm_area_struct *vma,
2075 struct page **pages, struct vm_area_struct **vmas, 2083 struct page **pages, struct vm_area_struct **vmas,
2076 unsigned long *position, int *length, int i, 2084 unsigned long *position, int *length, int i,
@@ -2080,6 +2088,8 @@ int follow_hugetlb_page(struct mm_struct *mm, struct vm_area_struct *vma,
2080 unsigned long vaddr = *position; 2088 unsigned long vaddr = *position;
2081 int remainder = *length; 2089 int remainder = *length;
2082 struct hstate *h = hstate_vma(vma); 2090 struct hstate *h = hstate_vma(vma);
2091 int zeropage_ok = 0;
2092 int shared = vma->vm_flags & VM_SHARED;
2083 2093
2084 spin_lock(&mm->page_table_lock); 2094 spin_lock(&mm->page_table_lock);
2085 while (vaddr < vma->vm_end && remainder) { 2095 while (vaddr < vma->vm_end && remainder) {
@@ -2092,8 +2102,11 @@ int follow_hugetlb_page(struct mm_struct *mm, struct vm_area_struct *vma,
2092 * first, for the page indexing below to work. 2102 * first, for the page indexing below to work.
2093 */ 2103 */
2094 pte = huge_pte_offset(mm, vaddr & huge_page_mask(h)); 2104 pte = huge_pte_offset(mm, vaddr & huge_page_mask(h));
2105 if (huge_zeropage_ok(pte, write, shared))
2106 zeropage_ok = 1;
2095 2107
2096 if (!pte || huge_pte_none(huge_ptep_get(pte)) || 2108 if (!pte ||
2109 (huge_pte_none(huge_ptep_get(pte)) && !zeropage_ok) ||
2097 (write && !pte_write(huge_ptep_get(pte)))) { 2110 (write && !pte_write(huge_ptep_get(pte)))) {
2098 int ret; 2111 int ret;
2099 2112
@@ -2113,8 +2126,11 @@ int follow_hugetlb_page(struct mm_struct *mm, struct vm_area_struct *vma,
2113 page = pte_page(huge_ptep_get(pte)); 2126 page = pte_page(huge_ptep_get(pte));
2114same_page: 2127same_page:
2115 if (pages) { 2128 if (pages) {
2116 get_page(page); 2129 if (zeropage_ok)
2117 pages[i] = page + pfn_offset; 2130 pages[i] = ZERO_PAGE(0);
2131 else
2132 pages[i] = page + pfn_offset;
2133 get_page(pages[i]);
2118 } 2134 }
2119 2135
2120 if (vmas) 2136 if (vmas)