Revision c9d398fa237882ea07167e23bcfc5e6847066518 authored by Naoya Horiguchi on 31 March 2017, 22:11:55 UTC, committed by Linus Torvalds on 01 April 2017, 00:13:30 UTC
I found the race condition which triggers the following bug when move_pages() and soft offline are called on a single hugetlb page concurrently. Soft offlining page 0x119400 at 0x700000000000 BUG: unable to handle kernel paging request at ffffea0011943820 IP: follow_huge_pmd+0x143/0x190 PGD 7ffd2067 PUD 7ffd1067 PMD 0 [61163.582052] Oops: 0000 [#1] SMP Modules linked in: binfmt_misc ppdev virtio_balloon parport_pc pcspkr i2c_piix4 parport i2c_core acpi_cpufreq ip_tables xfs libcrc32c ata_generic pata_acpi virtio_blk 8139too crc32c_intel ata_piix serio_raw libata virtio_pci 8139cp virtio_ring virtio mii floppy dm_mirror dm_region_hash dm_log dm_mod [last unloaded: cap_check] CPU: 0 PID: 22573 Comm: iterate_numa_mo Tainted: P OE 4.11.0-rc2-mm1+ #2 Hardware name: Red Hat KVM, BIOS 0.5.1 01/01/2011 RIP: 0010:follow_huge_pmd+0x143/0x190 RSP: 0018:ffffc90004bdbcd0 EFLAGS: 00010202 RAX: 0000000465003e80 RBX: ffffea0004e34d30 RCX: 00003ffffffff000 RDX: 0000000011943800 RSI: 0000000000080001 RDI: 0000000465003e80 RBP: ffffc90004bdbd18 R08: 0000000000000000 R09: ffff880138d34000 R10: ffffea0004650000 R11: 0000000000c363b0 R12: ffffea0011943800 R13: ffff8801b8d34000 R14: ffffea0000000000 R15: 000077ff80000000 FS: 00007fc977710740(0000) GS:ffff88007dc00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: ffffea0011943820 CR3: 000000007a746000 CR4: 00000000001406f0 Call Trace: follow_page_mask+0x270/0x550 SYSC_move_pages+0x4ea/0x8f0 SyS_move_pages+0xe/0x10 do_syscall_64+0x67/0x180 entry_SYSCALL64_slow_path+0x25/0x25 RIP: 0033:0x7fc976e03949 RSP: 002b:00007ffe72221d88 EFLAGS: 00000246 ORIG_RAX: 0000000000000117 RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007fc976e03949 RDX: 0000000000c22390 RSI: 0000000000001400 RDI: 0000000000005827 RBP: 00007ffe72221e00 R08: 0000000000c2c3a0 R09: 0000000000000004 R10: 0000000000c363b0 R11: 0000000000000246 R12: 0000000000400650 R13: 00007ffe72221ee0 R14: 0000000000000000 R15: 0000000000000000 Code: 81 e4 ff ff 1f 00 48 21 c2 49 c1 ec 0c 48 c1 ea 0c 4c 01 e2 49 bc 00 00 00 00 00 ea ff ff 48 c1 e2 06 49 01 d4 f6 45 bc 04 74 90 <49> 8b 7c 24 20 40 f6 c7 01 75 2b 4c 89 e7 8b 47 1c 85 c0 7e 2a RIP: follow_huge_pmd+0x143/0x190 RSP: ffffc90004bdbcd0 CR2: ffffea0011943820 ---[ end trace e4f81353a2d23232 ]--- Kernel panic - not syncing: Fatal exception Kernel Offset: disabled This bug is triggered when pmd_present() returns true for non-present hugetlb, so fixing the present check in follow_huge_pmd() prevents it. Using pmd_present() to determine present/non-present for hugetlb is not correct, because pmd_present() checks multiple bits (not only _PAGE_PRESENT) for historical reason and it can misjudge hugetlb state. Fixes: e66f17ff7177 ("mm/hugetlb: take page table lock in follow_huge_pmd()") Link: http://lkml.kernel.org/r/1490149898-20231-1-git-send-email-n-horiguchi@ah.jp.nec.com Signed-off-by: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com> Acked-by: Hillf Danton <hillf.zj@alibaba-inc.com> Cc: Hugh Dickins <hughd@google.com> Cc: Michal Hocko <mhocko@kernel.org> Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com> Cc: Mike Kravetz <mike.kravetz@oracle.com> Cc: Christian Borntraeger <borntraeger@de.ibm.com> Cc: Gerald Schaefer <gerald.schaefer@de.ibm.com> Cc: <stable@vger.kernel.org> [4.0+] Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
1 parent 0cefabd
ucount.c
/*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License as
* published by the Free Software Foundation, version 2 of the
* License.
*/
#include <linux/stat.h>
#include <linux/sysctl.h>
#include <linux/slab.h>
#include <linux/cred.h>
#include <linux/hash.h>
#include <linux/user_namespace.h>
#define UCOUNTS_HASHTABLE_BITS 10
static struct hlist_head ucounts_hashtable[(1 << UCOUNTS_HASHTABLE_BITS)];
static DEFINE_SPINLOCK(ucounts_lock);
#define ucounts_hashfn(ns, uid) \
hash_long((unsigned long)__kuid_val(uid) + (unsigned long)(ns), \
UCOUNTS_HASHTABLE_BITS)
#define ucounts_hashentry(ns, uid) \
(ucounts_hashtable + ucounts_hashfn(ns, uid))
#ifdef CONFIG_SYSCTL
static struct ctl_table_set *
set_lookup(struct ctl_table_root *root)
{
return ¤t_user_ns()->set;
}
static int set_is_seen(struct ctl_table_set *set)
{
return ¤t_user_ns()->set == set;
}
static int set_permissions(struct ctl_table_header *head,
struct ctl_table *table)
{
struct user_namespace *user_ns =
container_of(head->set, struct user_namespace, set);
int mode;
/* Allow users with CAP_SYS_RESOURCE unrestrained access */
if (ns_capable(user_ns, CAP_SYS_RESOURCE))
mode = (table->mode & S_IRWXU) >> 6;
else
/* Allow all others at most read-only access */
mode = table->mode & S_IROTH;
return (mode << 6) | (mode << 3) | mode;
}
static struct ctl_table_root set_root = {
.lookup = set_lookup,
.permissions = set_permissions,
};
static int zero = 0;
static int int_max = INT_MAX;
#define UCOUNT_ENTRY(name) \
{ \
.procname = name, \
.maxlen = sizeof(int), \
.mode = 0644, \
.proc_handler = proc_dointvec_minmax, \
.extra1 = &zero, \
.extra2 = &int_max, \
}
static struct ctl_table user_table[] = {
UCOUNT_ENTRY("max_user_namespaces"),
UCOUNT_ENTRY("max_pid_namespaces"),
UCOUNT_ENTRY("max_uts_namespaces"),
UCOUNT_ENTRY("max_ipc_namespaces"),
UCOUNT_ENTRY("max_net_namespaces"),
UCOUNT_ENTRY("max_mnt_namespaces"),
UCOUNT_ENTRY("max_cgroup_namespaces"),
#ifdef CONFIG_INOTIFY_USER
UCOUNT_ENTRY("max_inotify_instances"),
UCOUNT_ENTRY("max_inotify_watches"),
#endif
{ }
};
#endif /* CONFIG_SYSCTL */
bool setup_userns_sysctls(struct user_namespace *ns)
{
#ifdef CONFIG_SYSCTL
struct ctl_table *tbl;
setup_sysctl_set(&ns->set, &set_root, set_is_seen);
tbl = kmemdup(user_table, sizeof(user_table), GFP_KERNEL);
if (tbl) {
int i;
for (i = 0; i < UCOUNT_COUNTS; i++) {
tbl[i].data = &ns->ucount_max[i];
}
ns->sysctls = __register_sysctl_table(&ns->set, "user", tbl);
}
if (!ns->sysctls) {
kfree(tbl);
retire_sysctl_set(&ns->set);
return false;
}
#endif
return true;
}
void retire_userns_sysctls(struct user_namespace *ns)
{
#ifdef CONFIG_SYSCTL
struct ctl_table *tbl;
tbl = ns->sysctls->ctl_table_arg;
unregister_sysctl_table(ns->sysctls);
retire_sysctl_set(&ns->set);
kfree(tbl);
#endif
}
static struct ucounts *find_ucounts(struct user_namespace *ns, kuid_t uid, struct hlist_head *hashent)
{
struct ucounts *ucounts;
hlist_for_each_entry(ucounts, hashent, node) {
if (uid_eq(ucounts->uid, uid) && (ucounts->ns == ns))
return ucounts;
}
return NULL;
}
static struct ucounts *get_ucounts(struct user_namespace *ns, kuid_t uid)
{
struct hlist_head *hashent = ucounts_hashentry(ns, uid);
struct ucounts *ucounts, *new;
spin_lock_irq(&ucounts_lock);
ucounts = find_ucounts(ns, uid, hashent);
if (!ucounts) {
spin_unlock_irq(&ucounts_lock);
new = kzalloc(sizeof(*new), GFP_KERNEL);
if (!new)
return NULL;
new->ns = ns;
new->uid = uid;
new->count = 0;
spin_lock_irq(&ucounts_lock);
ucounts = find_ucounts(ns, uid, hashent);
if (ucounts) {
kfree(new);
} else {
hlist_add_head(&new->node, hashent);
ucounts = new;
}
}
if (ucounts->count == INT_MAX)
ucounts = NULL;
else
ucounts->count += 1;
spin_unlock_irq(&ucounts_lock);
return ucounts;
}
static void put_ucounts(struct ucounts *ucounts)
{
unsigned long flags;
spin_lock_irqsave(&ucounts_lock, flags);
ucounts->count -= 1;
if (!ucounts->count)
hlist_del_init(&ucounts->node);
else
ucounts = NULL;
spin_unlock_irqrestore(&ucounts_lock, flags);
kfree(ucounts);
}
static inline bool atomic_inc_below(atomic_t *v, int u)
{
int c, old;
c = atomic_read(v);
for (;;) {
if (unlikely(c >= u))
return false;
old = atomic_cmpxchg(v, c, c+1);
if (likely(old == c))
return true;
c = old;
}
}
struct ucounts *inc_ucount(struct user_namespace *ns, kuid_t uid,
enum ucount_type type)
{
struct ucounts *ucounts, *iter, *bad;
struct user_namespace *tns;
ucounts = get_ucounts(ns, uid);
for (iter = ucounts; iter; iter = tns->ucounts) {
int max;
tns = iter->ns;
max = READ_ONCE(tns->ucount_max[type]);
if (!atomic_inc_below(&iter->ucount[type], max))
goto fail;
}
return ucounts;
fail:
bad = iter;
for (iter = ucounts; iter != bad; iter = iter->ns->ucounts)
atomic_dec(&iter->ucount[type]);
put_ucounts(ucounts);
return NULL;
}
void dec_ucount(struct ucounts *ucounts, enum ucount_type type)
{
struct ucounts *iter;
for (iter = ucounts; iter; iter = iter->ns->ucounts) {
int dec = atomic_dec_if_positive(&iter->ucount[type]);
WARN_ON_ONCE(dec < 0);
}
put_ucounts(ucounts);
}
static __init int user_namespace_sysctl_init(void)
{
#ifdef CONFIG_SYSCTL
static struct ctl_table_header *user_header;
static struct ctl_table empty[1];
/*
* It is necessary to register the user directory in the
* default set so that registrations in the child sets work
* properly.
*/
user_header = register_sysctl("user", empty);
kmemleak_ignore(user_header);
BUG_ON(!user_header);
BUG_ON(!setup_userns_sysctls(&init_user_ns));
#endif
return 0;
}
subsys_initcall(user_namespace_sysctl_init);
![swh spinner](/static/img/swh-spinner.gif)
Computing file changes ...