From: Larry Woodman <lwoodman@redhat.com> Date: Thu, 17 Jan 2008 10:20:26 -0500 Subject: [mm] hugepages: leak due to pagetable page sharing Message-id: 478F723A.2010609@redhat.com O-Subject: [RHEL5-U2 patch] fix hugepages leak due to pagetable page sharing. Bugzilla: 428612 The shared page table patch for hugetlb memory on x86 and x86_64 is causing a leak. When a user of hugepages exits using this code the system leaks some of the hugepages. ------------------------------------------------------- Part of /proc/meminfo just before database startup: HugePages_Total: 5500 HugePages_Free: 5500 HugePages_Rsvd: 0 Hugepagesize: 2048 kB Just before shutdown: HugePages_Total: 5500 HugePages_Free: 4475 HugePages_Rsvd: 0 Hugepagesize: 2048 kB After shutdown: HugePages_Total: 5500 HugePages_Free: 4988 HugePages_Rsvd: 0 Hugepagesize: 2048 kB ---------------------------------------------------------- The problem occurs durring a fork, in copy_hugetlb_page_range(). It locates the dst_pte using huge_pte_alloc(). Since huge_pte_alloc() calls huge_pmd_share() it will share the pmd page if can, yet the main loop in copy_hugetlb_page_range() does a get_page() on every hugepage. This is a violation of the shared hugepmd pagetable protocol and creates additional referenced to the hugepages causing a leak when the unmap of the VMA occurs. We can skip the entire replication of the ptes when the hugepage pagetables are shared. The attached patch skips copying the ptes and the get_page() calls if the hugetlbpage pagetable is shared. Fixes BZ 428612 and this patch was send upstream and ACK'd there. diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 0a80f51..e4157ca 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -369,6 +369,9 @@ int copy_hugetlb_page_range(struct mm_struct *dst, struct mm_struct *src, dst_pte = huge_pte_alloc(dst, addr); if (!dst_pte) goto nomem; + /* if the page table is shared dont copy or take references */ + if (dst_pte == src_pte) + continue; spin_lock(&dst->page_table_lock); spin_lock(&src->page_table_lock); if (!pte_none(*src_pte)) {