From d33b9f45bd24a6391bc05e2b5a13c1b5787ca9c2 Mon Sep 17 00:00:00 2001 From: Naoya Horiguchi Date: Mon, 14 Dec 2009 17:59:59 -0800 Subject: mm: hugetlb: fix hugepage memory leak in walk_page_range() Most callers of pmd_none_or_clear_bad() check whether the target page is in a hugepage or not, but walk_page_range() do not check it. So if we read /proc/pid/pagemap for the hugepage on x86 machine, the hugepage memory is leaked as shown below. This patch fixes it. Details ======= My test program (leak_pagemap) works as follows: - creat() and mmap() a file on hugetlbfs (file size is 200MB == 100 hugepages,) - read()/write() something on it, - call page-types with option -p (walk around the page tables), - munmap() and unlink() the file on hugetlbfs Without my patches ------------------ $ cat /proc/meminfo |grep "HugePage" HugePages_Total: 1000 HugePages_Free: 1000 HugePages_Rsvd: 0 HugePages_Surp: 0 $ ./leak_pagemap [snip output] $ cat /proc/meminfo |grep "HugePage" HugePages_Total: 1000 HugePages_Free: 900 HugePages_Rsvd: 0 HugePages_Surp: 0 $ ls /hugetlbfs/ $ 100 hugepages are accounted as used while there is no file on hugetlbfs. With my patches --------------- $ cat /proc/meminfo |grep "HugePage" HugePages_Total: 1000 HugePages_Free: 1000 HugePages_Rsvd: 0 HugePages_Surp: 0 $ ./leak_pagemap [snip output] $ cat /proc/meminfo |grep "HugePage" HugePages_Total: 1000 HugePages_Free: 1000 HugePages_Rsvd: 0 HugePages_Surp: 0 $ ls /hugetlbfs $ No memory leaks. Signed-off-by: Naoya Horiguchi Cc: Andi Kleen Cc: Wu Fengguang Cc: Hugh Dickins Cc: Mel Gorman Cc: Lee Schermerhorn Cc: Andy Whitcroft Cc: David Rientjes Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/pagewalk.c | 16 +++++++++++++++- 1 file changed, 15 insertions(+), 1 deletion(-) (limited to 'mm/pagewalk.c') diff --git a/mm/pagewalk.c b/mm/pagewalk.c index d5878be..a286915 100644 --- a/mm/pagewalk.c +++ b/mm/pagewalk.c @@ -1,6 +1,7 @@ #include #include #include +#include static int walk_pte_range(pmd_t *pmd, unsigned long addr, unsigned long end, struct mm_walk *walk) @@ -107,6 +108,7 @@ int walk_page_range(unsigned long addr, unsigned long end, pgd_t *pgd; unsigned long next; int err = 0; + struct vm_area_struct *vma; if (addr >= end) return err; @@ -117,11 +119,22 @@ int walk_page_range(unsigned long addr, unsigned long end, pgd = pgd_offset(walk->mm, addr); do { next = pgd_addr_end(addr, end); + + /* skip hugetlb vma to avoid hugepage PMD being cleared + * in pmd_none_or_clear_bad(). */ + vma = find_vma(walk->mm, addr); + if (vma && is_vm_hugetlb_page(vma)) { + if (vma->vm_end < next) + next = vma->vm_end; + continue; + } + if (pgd_none_or_clear_bad(pgd)) { if (walk->pte_hole) err = walk->pte_hole(addr, next, walk); if (err) break; + pgd++; continue; } if (walk->pgd_entry) @@ -131,7 +144,8 @@ int walk_page_range(unsigned long addr, unsigned long end, err = walk_pud_range(pgd, addr, next, walk); if (err) break; - } while (pgd++, addr = next, addr != end); + pgd++; + } while (addr = next, addr != end); return err; } -- cgit v1.1