Revision 6f6acb00514c10be35529402f36ad7a288f08c2e authored by Michal Hocko on 22 May 2014, 18:54:19 UTC, committed by Linus Torvalds on 23 May 2014, 16:37:29 UTC
Commit 284f39afeaa4 ("mm: memcg: push !mm handling out to page cache charge function") explicitly checks for page cache charges without any mm context (from kernel thread context[1]). This seemed to be the only possible case where memory could be charged without mm context so commit 03583f1a631c ("memcg: remove unnecessary !mm check from try_get_mem_cgroup_from_mm()") removed the mm check from get_mem_cgroup_from_mm(). This however caused another NULL ptr dereference during early boot when loopback kernel thread splices to tmpfs as reported by Stephan Kulow: BUG: unable to handle kernel NULL pointer dereference at 0000000000000360 IP: get_mem_cgroup_from_mm.isra.42+0x2b/0x60 Oops: 0000 [#1] SMP Modules linked in: btrfs dm_multipath dm_mod scsi_dh multipath raid10 raid456 async_raid6_recov async_memcpy async_pq raid6_pq async_xor xor async_tx raid1 raid0 md_mod parport_pc parport nls_utf8 isofs usb_storage iscsi_ibft iscsi_boot_sysfs arc4 ecb fan thermal nfs lockd fscache nls_iso8859_1 nls_cp437 sg st hid_generic usbhid af_packet sunrpc sr_mod cdrom ata_generic uhci_hcd virtio_net virtio_blk ehci_hcd usbcore ata_piix floppy processor button usb_common virtio_pci virtio_ring virtio edd squashfs loop ppa] CPU: 0 PID: 97 Comm: loop1 Not tainted 3.15.0-rc5-5-default #1 Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011 Call Trace: __mem_cgroup_try_charge_swapin+0x40/0xe0 mem_cgroup_charge_file+0x8b/0xd0 shmem_getpage_gfp+0x66b/0x7b0 shmem_file_splice_read+0x18f/0x430 splice_direct_to_actor+0xa2/0x1c0 do_lo_receive+0x5a/0x60 [loop] loop_thread+0x298/0x720 [loop] kthread+0xc6/0xe0 ret_from_fork+0x7c/0xb0 Also Branimir Maksimovic reported the following oops which is tiggered for the swapcache charge path from the accounting code for kernel threads: CPU: 1 PID: 160 Comm: kworker/u8:5 Tainted: P OE 3.15.0-rc5-core2-custom #159 Hardware name: System manufacturer System Product Name/MAXIMUSV GENE, BIOS 1903 08/19/2013 task: ffff880404e349b0 ti: ffff88040486a000 task.ti: ffff88040486a000 RIP: get_mem_cgroup_from_mm.isra.42+0x2b/0x60 Call Trace: __mem_cgroup_try_charge_swapin+0x45/0xf0 mem_cgroup_charge_file+0x9c/0xe0 shmem_getpage_gfp+0x62c/0x770 shmem_write_begin+0x38/0x40 generic_perform_write+0xc5/0x1c0 __generic_file_aio_write+0x1d1/0x3f0 generic_file_aio_write+0x4f/0xc0 do_sync_write+0x5a/0x90 do_acct_process+0x4b1/0x550 acct_process+0x6d/0xa0 do_exit+0x827/0xa70 kthread+0xc3/0xf0 This patch fixes the issue by reintroducing mm check into get_mem_cgroup_from_mm. We could do the same trick in __mem_cgroup_try_charge_swapin as we do for the regular page cache path but it is not worth troubles. The check is not that expensive and it is better to have get_mem_cgroup_from_mm more robust. [1] - http://marc.info/?l=linux-mm&m=139463617808941&w=2 Fixes: 03583f1a631c ("memcg: remove unnecessary !mm check from try_get_mem_cgroup_from_mm()") Reported-and-tested-by: Stephan Kulow <coolo@suse.com> Reported-by: Branimir Maksimovic <branimir.maksimovic@gmail.com> Signed-off-by: Michal Hocko <mhocko@suse.cz> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Hugh Dickins <hughd@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
1 parent 55231e5
audit.h
/* audit -- definition of audit_context structure and supporting types
*
* Copyright 2003-2004 Red Hat, Inc.
* Copyright 2005 Hewlett-Packard Development Company, L.P.
* Copyright 2005 IBM Corporation
*
* 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; either version 2 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
*/
#include <linux/fs.h>
#include <linux/audit.h>
#include <linux/skbuff.h>
#include <uapi/linux/mqueue.h>
/* 0 = no checking
1 = put_count checking
2 = verbose put_count checking
*/
#define AUDIT_DEBUG 0
/* AUDIT_NAMES is the number of slots we reserve in the audit_context
* for saving names from getname(). If we get more names we will allocate
* a name dynamically and also add those to the list anchored by names_list. */
#define AUDIT_NAMES 5
/* At task start time, the audit_state is set in the audit_context using
a per-task filter. At syscall entry, the audit_state is augmented by
the syscall filter. */
enum audit_state {
AUDIT_DISABLED, /* Do not create per-task audit_context.
* No syscall-specific audit records can
* be generated. */
AUDIT_BUILD_CONTEXT, /* Create the per-task audit_context,
* and fill it in at syscall
* entry time. This makes a full
* syscall record available if some
* other part of the kernel decides it
* should be recorded. */
AUDIT_RECORD_CONTEXT /* Create the per-task audit_context,
* always fill it in at syscall entry
* time, and always write out the audit
* record at syscall exit time. */
};
/* Rule lists */
struct audit_watch;
struct audit_tree;
struct audit_chunk;
struct audit_entry {
struct list_head list;
struct rcu_head rcu;
struct audit_krule rule;
};
struct audit_cap_data {
kernel_cap_t permitted;
kernel_cap_t inheritable;
union {
unsigned int fE; /* effective bit of file cap */
kernel_cap_t effective; /* effective set of process */
};
};
/* When fs/namei.c:getname() is called, we store the pointer in name and
* we don't let putname() free it (instead we free all of the saved
* pointers at syscall exit time).
*
* Further, in fs/namei.c:path_lookup() we store the inode and device.
*/
struct audit_names {
struct list_head list; /* audit_context->names_list */
struct filename *name;
int name_len; /* number of chars to log */
bool hidden; /* don't log this record */
bool name_put; /* call __putname()? */
unsigned long ino;
dev_t dev;
umode_t mode;
kuid_t uid;
kgid_t gid;
dev_t rdev;
u32 osid;
struct audit_cap_data fcap;
unsigned int fcap_ver;
unsigned char type; /* record type */
/*
* This was an allocated audit_names and not from the array of
* names allocated in the task audit context. Thus this name
* should be freed on syscall exit.
*/
bool should_free;
};
struct audit_proctitle {
int len; /* length of the cmdline field. */
char *value; /* the cmdline field */
};
/* The per-task audit context. */
struct audit_context {
int dummy; /* must be the first element */
int in_syscall; /* 1 if task is in a syscall */
enum audit_state state, current_state;
unsigned int serial; /* serial number for record */
int major; /* syscall number */
struct timespec ctime; /* time of syscall entry */
unsigned long argv[4]; /* syscall arguments */
long return_code;/* syscall return code */
u64 prio;
int return_valid; /* return code is valid */
/*
* The names_list is the list of all audit_names collected during this
* syscall. The first AUDIT_NAMES entries in the names_list will
* actually be from the preallocated_names array for performance
* reasons. Except during allocation they should never be referenced
* through the preallocated_names array and should only be found/used
* by running the names_list.
*/
struct audit_names preallocated_names[AUDIT_NAMES];
int name_count; /* total records in names_list */
struct list_head names_list; /* struct audit_names->list anchor */
char *filterkey; /* key for rule that triggered record */
struct path pwd;
struct audit_aux_data *aux;
struct audit_aux_data *aux_pids;
struct sockaddr_storage *sockaddr;
size_t sockaddr_len;
/* Save things to print about task_struct */
pid_t pid, ppid;
kuid_t uid, euid, suid, fsuid;
kgid_t gid, egid, sgid, fsgid;
unsigned long personality;
int arch;
pid_t target_pid;
kuid_t target_auid;
kuid_t target_uid;
unsigned int target_sessionid;
u32 target_sid;
char target_comm[TASK_COMM_LEN];
struct audit_tree_refs *trees, *first_trees;
struct list_head killed_trees;
int tree_count;
int type;
union {
struct {
int nargs;
long args[6];
} socketcall;
struct {
kuid_t uid;
kgid_t gid;
umode_t mode;
u32 osid;
int has_perm;
uid_t perm_uid;
gid_t perm_gid;
umode_t perm_mode;
unsigned long qbytes;
} ipc;
struct {
mqd_t mqdes;
struct mq_attr mqstat;
} mq_getsetattr;
struct {
mqd_t mqdes;
int sigev_signo;
} mq_notify;
struct {
mqd_t mqdes;
size_t msg_len;
unsigned int msg_prio;
struct timespec abs_timeout;
} mq_sendrecv;
struct {
int oflag;
umode_t mode;
struct mq_attr attr;
} mq_open;
struct {
pid_t pid;
struct audit_cap_data cap;
} capset;
struct {
int fd;
int flags;
} mmap;
struct {
int argc;
} execve;
};
int fds[2];
struct audit_proctitle proctitle;
#if AUDIT_DEBUG
int put_count;
int ino_count;
#endif
};
extern u32 audit_ever_enabled;
extern void audit_copy_inode(struct audit_names *name,
const struct dentry *dentry,
const struct inode *inode);
extern void audit_log_cap(struct audit_buffer *ab, char *prefix,
kernel_cap_t *cap);
extern void audit_log_fcaps(struct audit_buffer *ab, struct audit_names *name);
extern void audit_log_name(struct audit_context *context,
struct audit_names *n, struct path *path,
int record_num, int *call_panic);
extern int audit_pid;
#define AUDIT_INODE_BUCKETS 32
extern struct list_head audit_inode_hash[AUDIT_INODE_BUCKETS];
static inline int audit_hash_ino(u32 ino)
{
return (ino & (AUDIT_INODE_BUCKETS-1));
}
/* Indicates that audit should log the full pathname. */
#define AUDIT_NAME_FULL -1
extern int audit_match_class(int class, unsigned syscall);
extern int audit_comparator(const u32 left, const u32 op, const u32 right);
extern int audit_uid_comparator(kuid_t left, u32 op, kuid_t right);
extern int audit_gid_comparator(kgid_t left, u32 op, kgid_t right);
extern int parent_len(const char *path);
extern int audit_compare_dname_path(const char *dname, const char *path, int plen);
extern struct sk_buff *audit_make_reply(__u32 portid, int seq, int type,
int done, int multi,
const void *payload, int size);
extern void audit_panic(const char *message);
struct audit_netlink_list {
__u32 portid;
struct net *net;
struct sk_buff_head q;
};
int audit_send_list(void *);
struct audit_net {
struct sock *nlsk;
};
extern int selinux_audit_rule_update(void);
extern struct mutex audit_filter_mutex;
extern void audit_free_rule_rcu(struct rcu_head *);
extern struct list_head audit_filter_list[];
extern struct audit_entry *audit_dupe_rule(struct audit_krule *old);
/* audit watch functions */
#ifdef CONFIG_AUDIT_WATCH
extern void audit_put_watch(struct audit_watch *watch);
extern void audit_get_watch(struct audit_watch *watch);
extern int audit_to_watch(struct audit_krule *krule, char *path, int len, u32 op);
extern int audit_add_watch(struct audit_krule *krule, struct list_head **list);
extern void audit_remove_watch_rule(struct audit_krule *krule);
extern char *audit_watch_path(struct audit_watch *watch);
extern int audit_watch_compare(struct audit_watch *watch, unsigned long ino, dev_t dev);
#else
#define audit_put_watch(w) {}
#define audit_get_watch(w) {}
#define audit_to_watch(k, p, l, o) (-EINVAL)
#define audit_add_watch(k, l) (-EINVAL)
#define audit_remove_watch_rule(k) BUG()
#define audit_watch_path(w) ""
#define audit_watch_compare(w, i, d) 0
#endif /* CONFIG_AUDIT_WATCH */
#ifdef CONFIG_AUDIT_TREE
extern struct audit_chunk *audit_tree_lookup(const struct inode *);
extern void audit_put_chunk(struct audit_chunk *);
extern int audit_tree_match(struct audit_chunk *, struct audit_tree *);
extern int audit_make_tree(struct audit_krule *, char *, u32);
extern int audit_add_tree_rule(struct audit_krule *);
extern int audit_remove_tree_rule(struct audit_krule *);
extern void audit_trim_trees(void);
extern int audit_tag_tree(char *old, char *new);
extern const char *audit_tree_path(struct audit_tree *);
extern void audit_put_tree(struct audit_tree *);
extern void audit_kill_trees(struct list_head *);
#else
#define audit_remove_tree_rule(rule) BUG()
#define audit_add_tree_rule(rule) -EINVAL
#define audit_make_tree(rule, str, op) -EINVAL
#define audit_trim_trees() (void)0
#define audit_put_tree(tree) (void)0
#define audit_tag_tree(old, new) -EINVAL
#define audit_tree_path(rule) "" /* never called */
#define audit_kill_trees(list) BUG()
#endif
extern char *audit_unpack_string(void **, size_t *, size_t);
extern pid_t audit_sig_pid;
extern kuid_t audit_sig_uid;
extern u32 audit_sig_sid;
#ifdef CONFIG_AUDITSYSCALL
extern int __audit_signal_info(int sig, struct task_struct *t);
static inline int audit_signal_info(int sig, struct task_struct *t)
{
if (unlikely((audit_pid && t->tgid == audit_pid) ||
(audit_signals && !audit_dummy_context())))
return __audit_signal_info(sig, t);
return 0;
}
extern void audit_filter_inodes(struct task_struct *, struct audit_context *);
extern struct list_head *audit_killed_trees(void);
#else
#define audit_signal_info(s,t) AUDIT_DISABLED
#define audit_filter_inodes(t,c) AUDIT_DISABLED
#endif
extern struct mutex audit_cmd_mutex;
![swh spinner](/static/img/swh-spinner.gif)
Computing file changes ...