From: Nadav Amit namit@vmware.com
stable inclusion from stable-5.10.67 commit 6afd1e053d9bd1f39cf03f2b0702ad0b06b647bd bugzilla: 182619 https://gitee.com/openeuler/kernel/issues/I4EWO7
Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=...
--------------------------------
[ Upstream commit 22e5fe2a2a279d9a6fcbdfb4dffe73821bef1c90 ]
userfaultfd assumes that the enabled features are set once and never changed after UFFDIO_API ioctl succeeded.
However, currently, UFFDIO_API can be called concurrently from two different threads, succeed on both threads and leave userfaultfd's features in non-deterministic state. Theoretically, other uffd operations (ioctl's and page-faults) can be dispatched while adversely affected by such changes of features.
Moreover, the writes to ctx->state and ctx->features are not ordered, which can - theoretically, again - let userfaultfd_ioctl() think that userfaultfd API completed, while the features are still not initialized.
To avoid races, it is arguably best to get rid of ctx->state. Since there are only 2 states, record the API initialization in ctx->features as the uppermost bit and remove ctx->state.
Link: https://lkml.kernel.org/r/20210808020724.1022515-3-namit@vmware.com Fixes: 9cd75c3cd4c3d ("userfaultfd: non-cooperative: add ability to report non-PF events from uffd descriptor") Signed-off-by: Nadav Amit namit@vmware.com Cc: Alexander Viro viro@zeniv.linux.org.uk Cc: Andrea Arcangeli aarcange@redhat.com Cc: Axel Rasmussen axelrasmussen@google.com Cc: Jens Axboe axboe@kernel.dk Cc: Mike Rapoport rppt@linux.vnet.ibm.com Cc: Peter Xu peterx@redhat.com Signed-off-by: Andrew Morton akpm@linux-foundation.org Signed-off-by: Linus Torvalds torvalds@linux-foundation.org Signed-off-by: Sasha Levin sashal@kernel.org Signed-off-by: Chen Jun chenjun102@huawei.com Acked-by: Weilong Chen chenweilong@huawei.com
Signed-off-by: Chen Jun chenjun102@huawei.com --- fs/userfaultfd.c | 91 +++++++++++++++++++++++------------------------- 1 file changed, 44 insertions(+), 47 deletions(-)
diff --git a/fs/userfaultfd.c b/fs/userfaultfd.c index 6a1d53f62d5a..4dadb641efd1 100644 --- a/fs/userfaultfd.c +++ b/fs/userfaultfd.c @@ -35,11 +35,6 @@ static struct kmem_cache *userfaultfd_ctx_cachep __read_mostly; int enable_userswap; #endif
-enum userfaultfd_state { - UFFD_STATE_WAIT_API, - UFFD_STATE_RUNNING, -}; - /* * Start with fault_pending_wqh and fault_wqh so they're more likely * to be in the same cacheline. @@ -71,8 +66,6 @@ struct userfaultfd_ctx { unsigned int flags; /* features requested from the userspace */ unsigned int features; - /* state machine */ - enum userfaultfd_state state; /* released */ bool released; /* memory mappings are changing because of non-cooperative event */ @@ -106,6 +99,14 @@ struct userfaultfd_wake_range { unsigned long len; };
+/* internal indication that UFFD_API ioctl was successfully executed */ +#define UFFD_FEATURE_INITIALIZED (1u << 31) + +static bool userfaultfd_is_initialized(struct userfaultfd_ctx *ctx) +{ + return ctx->features & UFFD_FEATURE_INITIALIZED; +} + static int userfaultfd_wake_function(wait_queue_entry_t *wq, unsigned mode, int wake_flags, void *key) { @@ -666,7 +667,6 @@ int dup_userfaultfd(struct vm_area_struct *vma, struct list_head *fcs)
refcount_set(&ctx->refcount, 1); ctx->flags = octx->flags; - ctx->state = UFFD_STATE_RUNNING; ctx->features = octx->features; ctx->released = false; ctx->mmap_changing = false; @@ -948,38 +948,33 @@ static __poll_t userfaultfd_poll(struct file *file, poll_table *wait)
poll_wait(file, &ctx->fd_wqh, wait);
- switch (ctx->state) { - case UFFD_STATE_WAIT_API: + if (!userfaultfd_is_initialized(ctx)) return EPOLLERR; - case UFFD_STATE_RUNNING: - /* - * poll() never guarantees that read won't block. - * userfaults can be waken before they're read(). - */ - if (unlikely(!(file->f_flags & O_NONBLOCK))) - return EPOLLERR; - /* - * lockless access to see if there are pending faults - * __pollwait last action is the add_wait_queue but - * the spin_unlock would allow the waitqueue_active to - * pass above the actual list_add inside - * add_wait_queue critical section. So use a full - * memory barrier to serialize the list_add write of - * add_wait_queue() with the waitqueue_active read - * below. - */ - ret = 0; - smp_mb(); - if (waitqueue_active(&ctx->fault_pending_wqh)) - ret = EPOLLIN; - else if (waitqueue_active(&ctx->event_wqh)) - ret = EPOLLIN;
- return ret; - default: - WARN_ON_ONCE(1); + /* + * poll() never guarantees that read won't block. + * userfaults can be waken before they're read(). + */ + if (unlikely(!(file->f_flags & O_NONBLOCK))) return EPOLLERR; - } + /* + * lockless access to see if there are pending faults + * __pollwait last action is the add_wait_queue but + * the spin_unlock would allow the waitqueue_active to + * pass above the actual list_add inside + * add_wait_queue critical section. So use a full + * memory barrier to serialize the list_add write of + * add_wait_queue() with the waitqueue_active read + * below. + */ + ret = 0; + smp_mb(); + if (waitqueue_active(&ctx->fault_pending_wqh)) + ret = EPOLLIN; + else if (waitqueue_active(&ctx->event_wqh)) + ret = EPOLLIN; + + return ret; }
static const struct file_operations userfaultfd_fops; @@ -1173,7 +1168,7 @@ static ssize_t userfaultfd_read(struct file *file, char __user *buf, struct uffd_msg msg; int no_wait = file->f_flags & O_NONBLOCK;
- if (ctx->state == UFFD_STATE_WAIT_API) + if (!userfaultfd_is_initialized(ctx)) return -EINVAL;
for (;;) { @@ -1850,9 +1845,10 @@ static int userfaultfd_writeprotect(struct userfaultfd_ctx *ctx, static inline unsigned int uffd_ctx_features(__u64 user_features) { /* - * For the current set of features the bits just coincide + * For the current set of features the bits just coincide. Set + * UFFD_FEATURE_INITIALIZED to mark the features as enabled. */ - return (unsigned int)user_features; + return (unsigned int)user_features | UFFD_FEATURE_INITIALIZED; }
/* @@ -1865,12 +1861,10 @@ static int userfaultfd_api(struct userfaultfd_ctx *ctx, { struct uffdio_api uffdio_api; void __user *buf = (void __user *)arg; + unsigned int ctx_features; int ret; __u64 features;
- ret = -EINVAL; - if (ctx->state != UFFD_STATE_WAIT_API) - goto out; ret = -EFAULT; if (copy_from_user(&uffdio_api, buf, sizeof(uffdio_api))) goto out; @@ -1887,9 +1881,13 @@ static int userfaultfd_api(struct userfaultfd_ctx *ctx, ret = -EFAULT; if (copy_to_user(buf, &uffdio_api, sizeof(uffdio_api))) goto out; - ctx->state = UFFD_STATE_RUNNING; + /* only enable the requested features for this uffd context */ - ctx->features = uffd_ctx_features(features); + ctx_features = uffd_ctx_features(features); + ret = -EINVAL; + if (cmpxchg(&ctx->features, 0, ctx_features) != 0) + goto err_out; + ret = 0; out: return ret; @@ -1906,7 +1904,7 @@ static long userfaultfd_ioctl(struct file *file, unsigned cmd, int ret = -EINVAL; struct userfaultfd_ctx *ctx = file->private_data;
- if (cmd != UFFDIO_API && ctx->state == UFFD_STATE_WAIT_API) + if (cmd != UFFDIO_API && !userfaultfd_is_initialized(ctx)) return -EINVAL;
switch(cmd) { @@ -2010,7 +2008,6 @@ SYSCALL_DEFINE1(userfaultfd, int, flags) refcount_set(&ctx->refcount, 1); ctx->flags = flags; ctx->features = 0; - ctx->state = UFFD_STATE_WAIT_API; ctx->released = false; ctx->mmap_changing = false; ctx->mm = current->mm;