leandrof/linux
1
0
Fork 0
forked from Minki/linux
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Reimplement RLIMIT_MEMLOCK on top of ucounts

Browse Source 
The rlimit counter is tied to uid in the user_namespace. This allows
rlimit values to be specified in userns even if they are already
globally exceeded by the user. However, the value of the previous
user_namespaces cannot be exceeded.

Changelog

v11:
* Fix issue found by lkp robot.

v8:
* Fix issues found by lkp-tests project.

v7:
* Keep only ucounts for RLIMIT_MEMLOCK checks instead of struct cred.

v6:
* Fix bug in hugetlb_file_setup() detected by trinity.

Reported-by: kernel test robot <oliver.sang@intel.com>
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Alexey Gladkov <legion@kernel.org>
Link: https://lkml.kernel.org/r/970d50c70c71bfd4496e0e8d2a0a32feebebb350.1619094428.git.legion@kernel.org
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
This commit is contained in:
Alexey Gladkov 2021-04-22 14:27:14 +02:00 committed by Eric W. Biederman
parent d646969055
commit d7c9e99aee
15 changed files with 53 additions and 45 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
fs/hugetlbfs/inode.c
View File

@ -1443,7 +1443,7 @@ static int get_hstate_idx(int page_size_log)
* otherwise hugetlb_reserve_pages reserves one less hugepages than intended. * otherwise hugetlb_reserve_pages reserves one less hugepages than intended.
*/ */
struct file *hugetlb_file_setup(const char *name, size_t size, struct file *hugetlb_file_setup(const char *name, size_t size,
vm_flags_t acctflag, struct user_struct **user, vm_flags_t acctflag, struct ucounts **ucounts,
int creat_flags, int page_size_log) int creat_flags, int page_size_log)
{ {
struct inode *inode; struct inode *inode;
@ -1455,20 +1455,20 @@ struct file *hugetlb_file_setup(const char *name, size_t size,
if (hstate_idx < 0) if (hstate_idx < 0)
return ERR_PTR(-ENODEV); return ERR_PTR(-ENODEV);
*user = NULL; *ucounts = NULL;
mnt = hugetlbfs_vfsmount[hstate_idx]; mnt = hugetlbfs_vfsmount[hstate_idx];
if (!mnt) if (!mnt)
return ERR_PTR(-ENOENT); return ERR_PTR(-ENOENT);
if (creat_flags == HUGETLB_SHMFS_INODE && !can_do_hugetlb_shm()) { if (creat_flags == HUGETLB_SHMFS_INODE && !can_do_hugetlb_shm()) {
*user = current_user(); *ucounts = current_ucounts();
if (user_shm_lock(size, *user)) { if (user_shm_lock(size, *ucounts)) {
task_lock(current); task_lock(current);
pr_warn_once("%s (%d): Using mlock ulimits for SHM_HUGETLB is deprecated\n", pr_warn_once("%s (%d): Using mlock ulimits for SHM_HUGETLB is deprecated\n",
current->comm, current->pid); current->comm, current->pid);
task_unlock(current); task_unlock(current);
} else { } else {
*user = NULL; *ucounts = NULL;
return ERR_PTR(-EPERM); return ERR_PTR(-EPERM);
} }
} }
@ -1495,9 +1495,9 @@ struct file *hugetlb_file_setup(const char *name, size_t size,
iput(inode); iput(inode);
out: out:
if (*user) { if (*ucounts) {
user_shm_unlock(size, *user); user_shm_unlock(size, *ucounts);
*user = NULL; *ucounts = NULL;
} }
return file; return file;
} }

4
include/linux/hugetlb.h
View File

@ -434,7 +434,7 @@ static inline struct hugetlbfs_inode_info *HUGETLBFS_I(struct inode *inode)
extern const struct file_operations hugetlbfs_file_operations; extern const struct file_operations hugetlbfs_file_operations;
extern const struct vm_operations_struct hugetlb_vm_ops; extern const struct vm_operations_struct hugetlb_vm_ops;
struct file *hugetlb_file_setup(const char *name, size_t size, vm_flags_t acct, struct file *hugetlb_file_setup(const char *name, size_t size, vm_flags_t acct,
struct user_struct **user, int creat_flags, struct ucounts **ucounts, int creat_flags,
int page_size_log); int page_size_log);
static inline bool is_file_hugepages(struct file *file) static inline bool is_file_hugepages(struct file *file)
@ -454,7 +454,7 @@ static inline struct hstate *hstate_inode(struct inode *i)
#define is_file_hugepages(file) false #define is_file_hugepages(file) false
static inline struct file * static inline struct file *
hugetlb_file_setup(const char *name, size_t size, vm_flags_t acctflag, hugetlb_file_setup(const char *name, size_t size, vm_flags_t acctflag,
struct user_struct **user, int creat_flags, struct ucounts **ucounts, int creat_flags,
int page_size_log) int page_size_log)
{ {
return ERR_PTR(-ENOSYS); return ERR_PTR(-ENOSYS);

4
include/linux/mm.h
View File

@ -1670,8 +1670,8 @@ extern bool can_do_mlock(void);
#else #else
static inline bool can_do_mlock(void) { return false; } static inline bool can_do_mlock(void) { return false; }
#endif #endif
extern int user_shm_lock(size_t, struct user_struct *); extern int user_shm_lock(size_t, struct ucounts *);
extern void user_shm_unlock(size_t, struct user_struct *); extern void user_shm_unlock(size_t, struct ucounts *);
/* /*
* Parameter block passed down to zap_pte_range in exceptional cases. * Parameter block passed down to zap_pte_range in exceptional cases.

1
include/linux/sched/user.h
View File

@ -18,7 +18,6 @@ struct user_struct {
#ifdef CONFIG_EPOLL #ifdef CONFIG_EPOLL
atomic_long_t epoll_watches; /* The number of file descriptors currently watched */ atomic_long_t epoll_watches; /* The number of file descriptors currently watched */
#endif #endif
unsigned long locked_shm; /* How many pages of mlocked shm ? */
unsigned long unix_inflight; /* How many files in flight in unix sockets */ unsigned long unix_inflight; /* How many files in flight in unix sockets */
atomic_long_t pipe_bufs; /* how many pages are allocated in pipe buffers */ atomic_long_t pipe_bufs; /* how many pages are allocated in pipe buffers */

2
include/linux/shmem_fs.h
View File

@ -65,7 +65,7 @@ extern struct file *shmem_file_setup_with_mnt(struct vfsmount *mnt,
extern int shmem_zero_setup(struct vm_area_struct *); extern int shmem_zero_setup(struct vm_area_struct *);
extern unsigned long shmem_get_unmapped_area(struct file *, unsigned long addr, extern unsigned long shmem_get_unmapped_area(struct file *, unsigned long addr,
unsigned long len, unsigned long pgoff, unsigned long flags); unsigned long len, unsigned long pgoff, unsigned long flags);
extern int shmem_lock(struct file *file, int lock, struct user_struct *user); extern int shmem_lock(struct file *file, int lock, struct ucounts *ucounts);
#ifdef CONFIG_SHMEM #ifdef CONFIG_SHMEM
extern const struct address_space_operations shmem_aops; extern const struct address_space_operations shmem_aops;
static inline bool shmem_mapping(struct address_space *mapping) static inline bool shmem_mapping(struct address_space *mapping)

1
include/linux/user_namespace.h
View File

@ -53,6 +53,7 @@ enum ucount_type {
UCOUNT_RLIMIT_NPROC, UCOUNT_RLIMIT_NPROC,
UCOUNT_RLIMIT_MSGQUEUE, UCOUNT_RLIMIT_MSGQUEUE,
UCOUNT_RLIMIT_SIGPENDING, UCOUNT_RLIMIT_SIGPENDING,
UCOUNT_RLIMIT_MEMLOCK,
UCOUNT_COUNTS, UCOUNT_COUNTS,
}; };

26
ipc/shm.c
View File

@ -60,7 +60,7 @@ struct shmid_kernel /* private to the kernel */
time64_t shm_ctim; time64_t shm_ctim;
struct pid *shm_cprid; struct pid *shm_cprid;
struct pid *shm_lprid; struct pid *shm_lprid;
struct user_struct *mlock_user; struct ucounts *mlock_ucounts;
/* The task created the shm object. NULL if the task is dead. */ /* The task created the shm object. NULL if the task is dead. */
struct task_struct *shm_creator; struct task_struct *shm_creator;
@ -286,10 +286,10 @@ static void shm_destroy(struct ipc_namespace *ns, struct shmid_kernel *shp)
shm_rmid(ns, shp); shm_rmid(ns, shp);
shm_unlock(shp); shm_unlock(shp);
if (!is_file_hugepages(shm_file)) if (!is_file_hugepages(shm_file))
shmem_lock(shm_file, 0, shp->mlock_user); shmem_lock(shm_file, 0, shp->mlock_ucounts);
else if (shp->mlock_user) else if (shp->mlock_ucounts)
user_shm_unlock(i_size_read(file_inode(shm_file)), user_shm_unlock(i_size_read(file_inode(shm_file)),
shp->mlock_user); shp->mlock_ucounts);
fput(shm_file); fput(shm_file);
ipc_update_pid(&shp->shm_cprid, NULL); ipc_update_pid(&shp->shm_cprid, NULL);
ipc_update_pid(&shp->shm_lprid, NULL); ipc_update_pid(&shp->shm_lprid, NULL);
@ -625,7 +625,7 @@ static int newseg(struct ipc_namespace *ns, struct ipc_params *params)
shp->shm_perm.key = key; shp->shm_perm.key = key;
shp->shm_perm.mode = (shmflg & S_IRWXUGO); shp->shm_perm.mode = (shmflg & S_IRWXUGO);
shp->mlock_user = NULL; shp->mlock_ucounts = NULL;
shp->shm_perm.security = NULL; shp->shm_perm.security = NULL;
error = security_shm_alloc(&shp->shm_perm); error = security_shm_alloc(&shp->shm_perm);
@ -650,7 +650,7 @@ static int newseg(struct ipc_namespace *ns, struct ipc_params *params)
if (shmflg & SHM_NORESERVE) if (shmflg & SHM_NORESERVE)
acctflag = VM_NORESERVE; acctflag = VM_NORESERVE;
file = hugetlb_file_setup(name, hugesize, acctflag, file = hugetlb_file_setup(name, hugesize, acctflag,
&shp->mlock_user, HUGETLB_SHMFS_INODE, &shp->mlock_ucounts, HUGETLB_SHMFS_INODE,
(shmflg >> SHM_HUGE_SHIFT) & SHM_HUGE_MASK); (shmflg >> SHM_HUGE_SHIFT) & SHM_HUGE_MASK);
} else { } else {
/* /*
@ -698,8 +698,8 @@ static int newseg(struct ipc_namespace *ns, struct ipc_params *params)
no_id: no_id:
ipc_update_pid(&shp->shm_cprid, NULL); ipc_update_pid(&shp->shm_cprid, NULL);
ipc_update_pid(&shp->shm_lprid, NULL); ipc_update_pid(&shp->shm_lprid, NULL);
if (is_file_hugepages(file) && shp->mlock_user) if (is_file_hugepages(file) && shp->mlock_ucounts)
user_shm_unlock(size, shp->mlock_user); user_shm_unlock(size, shp->mlock_ucounts);
fput(file); fput(file);
ipc_rcu_putref(&shp->shm_perm, shm_rcu_free); ipc_rcu_putref(&shp->shm_perm, shm_rcu_free);
return error; return error;
@ -1105,12 +1105,12 @@ static int shmctl_do_lock(struct ipc_namespace *ns, int shmid, int cmd)
goto out_unlock0; goto out_unlock0;
if (cmd == SHM_LOCK) { if (cmd == SHM_LOCK) {
struct user_struct *user = current_user(); struct ucounts *ucounts = current_ucounts();
err = shmem_lock(shm_file, 1, user); err = shmem_lock(shm_file, 1, ucounts);
if (!err && !(shp->shm_perm.mode & SHM_LOCKED)) { if (!err && !(shp->shm_perm.mode & SHM_LOCKED)) {
shp->shm_perm.mode |= SHM_LOCKED; shp->shm_perm.mode |= SHM_LOCKED;
shp->mlock_user = user; shp->mlock_ucounts = ucounts;
} }
goto out_unlock0; goto out_unlock0;
} }
@ -1118,9 +1118,9 @@ static int shmctl_do_lock(struct ipc_namespace *ns, int shmid, int cmd)
/* SHM_UNLOCK */ /* SHM_UNLOCK */
if (!(shp->shm_perm.mode & SHM_LOCKED)) if (!(shp->shm_perm.mode & SHM_LOCKED))
goto out_unlock0; goto out_unlock0;
shmem_lock(shm_file, 0, shp->mlock_user); shmem_lock(shm_file, 0, shp->mlock_ucounts);
shp->shm_perm.mode &= ~SHM_LOCKED; shp->shm_perm.mode &= ~SHM_LOCKED;
shp->mlock_user = NULL; shp->mlock_ucounts = NULL;
get_file(shm_file); get_file(shm_file);
ipc_unlock_object(&shp->shm_perm); ipc_unlock_object(&shp->shm_perm);
rcu_read_unlock(); rcu_read_unlock();

1
kernel/fork.c
View File

@ -825,6 +825,7 @@ void __init fork_init(void)
init_user_ns.ucount_max[UCOUNT_RLIMIT_NPROC] = task_rlimit(&init_task, RLIMIT_NPROC); init_user_ns.ucount_max[UCOUNT_RLIMIT_NPROC] = task_rlimit(&init_task, RLIMIT_NPROC);
init_user_ns.ucount_max[UCOUNT_RLIMIT_MSGQUEUE] = task_rlimit(&init_task, RLIMIT_MSGQUEUE); init_user_ns.ucount_max[UCOUNT_RLIMIT_MSGQUEUE] = task_rlimit(&init_task, RLIMIT_MSGQUEUE);
init_user_ns.ucount_max[UCOUNT_RLIMIT_SIGPENDING] = task_rlimit(&init_task, RLIMIT_SIGPENDING); init_user_ns.ucount_max[UCOUNT_RLIMIT_SIGPENDING] = task_rlimit(&init_task, RLIMIT_SIGPENDING);
init_user_ns.ucount_max[UCOUNT_RLIMIT_MEMLOCK] = task_rlimit(&init_task, RLIMIT_MEMLOCK);
#ifdef CONFIG_VMAP_STACK #ifdef CONFIG_VMAP_STACK
cpuhp_setup_state(CPUHP_BP_PREPARE_DYN, "fork:vm_stack_cache", cpuhp_setup_state(CPUHP_BP_PREPARE_DYN, "fork:vm_stack_cache",

1
kernel/ucount.c
View File

@ -83,6 +83,7 @@ static struct ctl_table user_table[] = {
{ }, { },
{ }, { },
{ }, { },
{ },
{ } { }
}; };
#endif /* CONFIG_SYSCTL */ #endif /* CONFIG_SYSCTL */

1
kernel/user.c
View File

@ -98,7 +98,6 @@ static DEFINE_SPINLOCK(uidhash_lock);
/* root_user.__count is 1, for init task cred */ /* root_user.__count is 1, for init task cred */
struct user_struct root_user = { struct user_struct root_user = {
.__count = REFCOUNT_INIT(1), .__count = REFCOUNT_INIT(1),
.locked_shm = 0,
.uid = GLOBAL_ROOT_UID, .uid = GLOBAL_ROOT_UID,
.ratelimit = RATELIMIT_STATE_INIT(root_user.ratelimit, 0, 0), .ratelimit = RATELIMIT_STATE_INIT(root_user.ratelimit, 0, 0),
}; };

1
kernel/user_namespace.c
View File

@ -125,6 +125,7 @@ int create_user_ns(struct cred *new)
ns->ucount_max[UCOUNT_RLIMIT_NPROC] = rlimit(RLIMIT_NPROC); ns->ucount_max[UCOUNT_RLIMIT_NPROC] = rlimit(RLIMIT_NPROC);
ns->ucount_max[UCOUNT_RLIMIT_MSGQUEUE] = rlimit(RLIMIT_MSGQUEUE); ns->ucount_max[UCOUNT_RLIMIT_MSGQUEUE] = rlimit(RLIMIT_MSGQUEUE);
ns->ucount_max[UCOUNT_RLIMIT_SIGPENDING] = rlimit(RLIMIT_SIGPENDING); ns->ucount_max[UCOUNT_RLIMIT_SIGPENDING] = rlimit(RLIMIT_SIGPENDING);
ns->ucount_max[UCOUNT_RLIMIT_MEMLOCK] = rlimit(RLIMIT_MEMLOCK);
ns->ucounts = ucounts; ns->ucounts = ucounts;
/* Inherit USERNS_SETGROUPS_ALLOWED from our parent */ /* Inherit USERNS_SETGROUPS_ALLOWED from our parent */

4
mm/memfd.c
View File

@ -297,9 +297,9 @@ SYSCALL_DEFINE2(memfd_create,
} }
if (flags & MFD_HUGETLB) { if (flags & MFD_HUGETLB) {
struct user_struct *user = NULL; struct ucounts *ucounts = NULL;
file = hugetlb_file_setup(name, 0, VM_NORESERVE, &user, file = hugetlb_file_setup(name, 0, VM_NORESERVE, &ucounts,
HUGETLB_ANONHUGE_INODE, HUGETLB_ANONHUGE_INODE,
(flags >> MFD_HUGE_SHIFT) & (flags >> MFD_HUGE_SHIFT) &
MFD_HUGE_MASK); MFD_HUGE_MASK);

22
mm/mlock.c
View File

@ -817,9 +817,10 @@ SYSCALL_DEFINE0(munlockall)
*/ */
static DEFINE_SPINLOCK(shmlock_user_lock); static DEFINE_SPINLOCK(shmlock_user_lock);
int user_shm_lock(size_t size, struct user_struct *user) int user_shm_lock(size_t size, struct ucounts *ucounts)
{ {
unsigned long lock_limit, locked; unsigned long lock_limit, locked;
long memlock;
int allowed = 0; int allowed = 0;
locked = (size + PAGE_SIZE - 1) >> PAGE_SHIFT; locked = (size + PAGE_SIZE - 1) >> PAGE_SHIFT;
@ -828,21 +829,26 @@ int user_shm_lock(size_t size, struct user_struct *user)
allowed = 1; allowed = 1;
lock_limit >>= PAGE_SHIFT; lock_limit >>= PAGE_SHIFT;
spin_lock(&shmlock_user_lock); spin_lock(&shmlock_user_lock);
if (!allowed && memlock = inc_rlimit_ucounts(ucounts, UCOUNT_RLIMIT_MEMLOCK, locked);
locked + user->locked_shm > lock_limit && !capable(CAP_IPC_LOCK))
if (!allowed && (memlock == LONG_MAX || memlock > lock_limit) && !capable(CAP_IPC_LOCK)) {
dec_rlimit_ucounts(ucounts, UCOUNT_RLIMIT_MEMLOCK, locked);
goto out; goto out;
get_uid(user); }
user->locked_shm += locked; if (!get_ucounts(ucounts)) {
dec_rlimit_ucounts(ucounts, UCOUNT_RLIMIT_MEMLOCK, locked);
goto out;
}
allowed = 1; allowed = 1;
out: out:
spin_unlock(&shmlock_user_lock); spin_unlock(&shmlock_user_lock);
return allowed; return allowed;
} }
void user_shm_unlock(size_t size, struct user_struct *user) void user_shm_unlock(size_t size, struct ucounts *ucounts)
{ {
spin_lock(&shmlock_user_lock); spin_lock(&shmlock_user_lock);
user->locked_shm -= (size + PAGE_SIZE - 1) >> PAGE_SHIFT; dec_rlimit_ucounts(ucounts, UCOUNT_RLIMIT_MEMLOCK, (size + PAGE_SIZE - 1) >> PAGE_SHIFT);
spin_unlock(&shmlock_user_lock); spin_unlock(&shmlock_user_lock);
free_uid(user); put_ucounts(ucounts);
} }

4
mm/mmap.c
View File

@ -1605,7 +1605,7 @@ unsigned long ksys_mmap_pgoff(unsigned long addr, unsigned long len,
goto out_fput; goto out_fput;
} }
} else if (flags & MAP_HUGETLB) { } else if (flags & MAP_HUGETLB) {
struct user_struct *user = NULL; struct ucounts *ucounts = NULL;
struct hstate *hs; struct hstate *hs;
hs = hstate_sizelog((flags >> MAP_HUGE_SHIFT) & MAP_HUGE_MASK); hs = hstate_sizelog((flags >> MAP_HUGE_SHIFT) & MAP_HUGE_MASK);
@ -1621,7 +1621,7 @@ unsigned long ksys_mmap_pgoff(unsigned long addr, unsigned long len,
*/ */
file = hugetlb_file_setup(HUGETLB_ANON_FILE, len, file = hugetlb_file_setup(HUGETLB_ANON_FILE, len,
VM_NORESERVE, VM_NORESERVE,
&user, HUGETLB_ANONHUGE_INODE, &ucounts, HUGETLB_ANONHUGE_INODE,
(flags >> MAP_HUGE_SHIFT) & MAP_HUGE_MASK); (flags >> MAP_HUGE_SHIFT) & MAP_HUGE_MASK);
if (IS_ERR(file)) if (IS_ERR(file))
return PTR_ERR(file); return PTR_ERR(file);

10
mm/shmem.c
View File

@ -2227,7 +2227,7 @@ static struct mempolicy *shmem_get_policy(struct vm_area_struct *vma,
} }
#endif #endif
int shmem_lock(struct file *file, int lock, struct user_struct *user) int shmem_lock(struct file *file, int lock, struct ucounts *ucounts)
{ {
struct inode *inode = file_inode(file); struct inode *inode = file_inode(file);
struct shmem_inode_info *info = SHMEM_I(inode); struct shmem_inode_info *info = SHMEM_I(inode);
@ -2239,13 +2239,13 @@ int shmem_lock(struct file *file, int lock, struct user_struct *user)
* no serialization needed when called from shm_destroy(). * no serialization needed when called from shm_destroy().
*/ */
if (lock && !(info->flags & VM_LOCKED)) { if (lock && !(info->flags & VM_LOCKED)) {
if (!user_shm_lock(inode->i_size, user)) if (!user_shm_lock(inode->i_size, ucounts))
goto out_nomem; goto out_nomem;
info->flags |= VM_LOCKED; info->flags |= VM_LOCKED;
mapping_set_unevictable(file->f_mapping); mapping_set_unevictable(file->f_mapping);
} }
if (!lock && (info->flags & VM_LOCKED) && user) { if (!lock && (info->flags & VM_LOCKED) && ucounts) {
user_shm_unlock(inode->i_size, user); user_shm_unlock(inode->i_size, ucounts);
info->flags &= ~VM_LOCKED; info->flags &= ~VM_LOCKED;
mapping_clear_unevictable(file->f_mapping); mapping_clear_unevictable(file->f_mapping);
} }
@ -4093,7 +4093,7 @@ int shmem_unuse(unsigned int type, bool frontswap,
return 0; return 0;
} }
int shmem_lock(struct file *file, int lock, struct user_struct *user) int shmem_lock(struct file *file, int lock, struct ucounts *ucounts)
{ {
return 0; return 0;
} }