On Sat, Jan 24, 2015 at 08:36:08PM -0800, Fengguang Wu wrote: > [ 17.687075] Freeing unused kernel memory: 1716K (c190d000 - c1aba000) > [ 17.808897] random: init urandom read with 5 bits of entropy available > [ 17.828360] ------------[ cut here ]------------ > [ 17.828989] WARNING: CPU: 1 PID: 681 at mm/mmap.c:2858 exit_mmap+0x197/0x1ad() > [ 17.830086] Modules linked in: > [ 17.830549] CPU: 1 PID: 681 Comm: init Not tainted 3.19.0-rc5-gf7a7b53 #19 > [ 17.831339] 00000001 00000000 00000001 d388bd4c c14341a1 00000000 00000001 c16ebf08 > [ 17.832421] d388bd68 c1056987 00000b2a c1150db8 00000001 00000001 00000000 d388bd78 > [ 17.833488] c1056a11 00000009 00000000 d388bdd0 c1150db8 d3858380 ffffffff ffffffff > [ 17.841323] Call Trace: > [ 17.844215] [<c14341a1>] dump_stack+0x78/0xa8 > [ 17.844700] [<c1056987>] warn_slowpath_common+0xb7/0xce > [ 17.847797] [<c1150db8>] ? exit_mmap+0x197/0x1ad > [ 17.850955] [<c1056a11>] warn_slowpath_null+0x14/0x18 > [ 17.854131] [<c1150db8>] exit_mmap+0x197/0x1ad > [ 17.854629] [<c10537ff>] mmput+0x52/0xef > [ 17.857584] [<c1175602>] flush_old_exec+0x923/0x99d > [ 17.860806] [<c11aea1e>] load_elf_binary+0x430/0x11af > [ 17.861378] [<c108559f>] ? local_clock+0x2f/0x39 > [ 17.865327] [<c109817f>] ? lock_release_holdtime+0x60/0x6d > [ 17.866002] [<c1174159>] search_binary_handler+0x9c/0x20f > [ 17.866588] [<c11ac7e5>] load_script+0x339/0x355 > [ 17.874149] [<c108550c>] ? sched_clock_cpu+0x188/0x1a3 > [ 17.874718] [<c108559f>] ? local_clock+0x2f/0x39 > [ 17.878580] [<c109817f>] ? lock_release_holdtime+0x60/0x6d > [ 17.879355] [<c109c1bf>] ? do_raw_read_unlock+0x28/0x53 > [ 17.879997] [<c1174159>] search_binary_handler+0x9c/0x20f > [ 17.887644] [<c1176054>] do_execveat_common+0x6d6/0x954 > [ 17.890904] [<c11762eb>] do_execve+0x19/0x1b > [ 17.891389] [<c1176586>] SyS_execve+0x21/0x25 > [ 17.895168] [<c143be92>] syscall_call+0x7/0x7 > [ 17.895653] ---[ end trace 6a7094e9a1d04ce0 ]--- > [ 17.909585] ------------[ cut here ]------------ Thanks for report. The patch below should fix this. >From 00fc3107c8c62c4592a64293faff7e5643a068c0 Mon Sep 17 00:00:00 2001 From: "Kirill A. Shutemov" <kirill.shutemov@xxxxxxxxxxxxxxx> Date: Mon, 26 Jan 2015 13:50:58 +0200 Subject: [PATCH 2/2] mm, x86: fix pmd accounting for PAE Add missed mm->nr_pmds decrement on cleanup prepopulated PMDs. Signed-off-by: Kirill A. Shutemov <kirill.shutemov@xxxxxxxxxxxxxxx> Reported-by: Fengguang Wu <fengguang.wu@xxxxxxxxx> --- arch/x86/mm/pgtable.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/x86/mm/pgtable.c b/arch/x86/mm/pgtable.c index a7d36de0bd30..7b22adaad4f1 100644 --- a/arch/x86/mm/pgtable.c +++ b/arch/x86/mm/pgtable.c @@ -249,6 +249,7 @@ static void pgd_mop_up_pmds(struct mm_struct *mm, pgd_t *pgdp) paravirt_release_pmd(pgd_val(pgd) >> PAGE_SHIFT); pmd_free(mm, pmd); + mm_dec_nr_pmds(mm); } } } -- Kirill A. Shutemov -- 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/ . Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>