From 58493110046c7217f6171aa5f73cbfe30b3e7a43 Mon Sep 17 00:00:00 2001 From: Naoya Horiguchi Date: Mon, 14 Dec 2009 17:59:58 -0800 Subject: [PATCH] mm: hugetlb: fix hugepage memory leak in mincore() commit 4f16fc107d9c9b8a72aa19b189a9216e90a7aaef upstream. Most callers of pmd_none_or_clear_bad() check whether the target page is in a hugepage or not, but mincore() and walk_page_range() do not check it. So if we use mincore() on a hugepage on x86 machine, the hugepage memory is leaked as shown below. This patch fixes it by extending mincore() system call to support hugepages. Details ======= My test program (leak_mincore) works as follows: - creat() and mmap() a file on hugetlbfs (file size is 200MB == 100 hugepages,) - read()/write() something on it, - call mincore() for first ten pages and printf() the values of *vec - munmap() and unlink() the file on hugetlbfs Without my patch ---------------- $ cat /proc/meminfo| grep "HugePage" HugePages_Total: 1000 HugePages_Free: 1000 HugePages_Rsvd: 0 HugePages_Surp: 0 $ ./leak_mincore vec[0] 0 vec[1] 0 vec[2] 0 vec[3] 0 vec[4] 0 vec[5] 0 vec[6] 0 vec[7] 0 vec[8] 0 vec[9] 0 $ cat /proc/meminfo |grep "HugePage" HugePages_Total: 1000 HugePages_Free: 999 HugePages_Rsvd: 0 HugePages_Surp: 0 $ ls /hugetlbfs/ $ Return values in *vec from mincore() are set to 0, while the hugepage should be in memory, and 1 hugepage is still accounted as used while there is no file on hugetlbfs. With my patch ------------- $ cat /proc/meminfo| grep "HugePage" HugePages_Total: 1000 HugePages_Free: 1000 HugePages_Rsvd: 0 HugePages_Surp: 0 $ ./leak_mincore vec[0] 1 vec[1] 1 vec[2] 1 vec[3] 1 vec[4] 1 vec[5] 1 vec[6] 1 vec[7] 1 vec[8] 1 vec[9] 1 $ cat /proc/meminfo |grep "HugePage" HugePages_Total: 1000 HugePages_Free: 1000 HugePages_Rsvd: 0 HugePages_Surp: 0 $ ls /hugetlbfs/ $ Return value in *vec set to 1 and no memory leaks. [akpm@linux-foundation.org: cleanup] [akpm@linux-foundation.org: build fix] 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 Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds Signed-off-by: Greg Kroah-Hartman --- mm/mincore.c | 37 +++++++++++++++++++++++++++++++++++++ 1 file changed, 37 insertions(+) diff --git a/mm/mincore.c b/mm/mincore.c index 8cb508f84ea4..7a3436ef39eb 100644 --- a/mm/mincore.c +++ b/mm/mincore.c @@ -14,6 +14,7 @@ #include #include #include +#include #include #include @@ -72,6 +73,42 @@ static long do_mincore(unsigned long addr, unsigned char *vec, unsigned long pag if (!vma || addr < vma->vm_start) return -ENOMEM; +#ifdef CONFIG_HUGETLB_PAGE + if (is_vm_hugetlb_page(vma)) { + struct hstate *h; + unsigned long nr_huge; + unsigned char present; + + i = 0; + nr = min(pages, (vma->vm_end - addr) >> PAGE_SHIFT); + h = hstate_vma(vma); + nr_huge = ((addr + pages * PAGE_SIZE - 1) >> huge_page_shift(h)) + - (addr >> huge_page_shift(h)) + 1; + nr_huge = min(nr_huge, + (vma->vm_end - addr) >> huge_page_shift(h)); + while (1) { + /* hugepage always in RAM for now, + * but generally it needs to be check */ + ptep = huge_pte_offset(current->mm, + addr & huge_page_mask(h)); + present = !!(ptep && + !huge_pte_none(huge_ptep_get(ptep))); + while (1) { + vec[i++] = present; + addr += PAGE_SIZE; + /* reach buffer limit */ + if (i == nr) + return nr; + /* check hugepage border */ + if (!((addr & ~huge_page_mask(h)) + >> PAGE_SHIFT)) + break; + } + } + return nr; + } +#endif + /* * Calculate how many pages there are left in the last level of the * PTE array for our address. -- 2.39.5