summaryrefslogtreecommitdiffstats
path: root/fs/userfaultfd.c
diff options
context:
space:
mode:
Diffstat (limited to '')
-rw-r--r--fs/userfaultfd.c78
1 files changed, 75 insertions, 3 deletions
diff --git a/fs/userfaultfd.c b/fs/userfaultfd.c
index e8af40b055..959551ff9a 100644
--- a/fs/userfaultfd.c
+++ b/fs/userfaultfd.c
@@ -45,7 +45,6 @@ static struct ctl_table vm_userfaultfd_table[] = {
.extra1 = SYSCTL_ZERO,
.extra2 = SYSCTL_ONE,
},
- { }
};
#endif
@@ -1033,7 +1032,7 @@ static int resolve_userfault_fork(struct userfaultfd_ctx *new,
{
int fd;
- fd = anon_inode_getfd_secure("[userfaultfd]", &userfaultfd_fops, new,
+ fd = anon_inode_create_getfd("[userfaultfd]", &userfaultfd_fops, new,
O_RDONLY | (new->flags & UFFD_SHARED_FCNTL_FLAGS), inode);
if (fd < 0)
return fd;
@@ -2005,6 +2004,75 @@ static inline unsigned int uffd_ctx_features(__u64 user_features)
return (unsigned int)user_features | UFFD_FEATURE_INITIALIZED;
}
+static int userfaultfd_move(struct userfaultfd_ctx *ctx,
+ unsigned long arg)
+{
+ __s64 ret;
+ struct uffdio_move uffdio_move;
+ struct uffdio_move __user *user_uffdio_move;
+ struct userfaultfd_wake_range range;
+ struct mm_struct *mm = ctx->mm;
+
+ user_uffdio_move = (struct uffdio_move __user *) arg;
+
+ if (atomic_read(&ctx->mmap_changing))
+ return -EAGAIN;
+
+ if (copy_from_user(&uffdio_move, user_uffdio_move,
+ /* don't copy "move" last field */
+ sizeof(uffdio_move)-sizeof(__s64)))
+ return -EFAULT;
+
+ /* Do not allow cross-mm moves. */
+ if (mm != current->mm)
+ return -EINVAL;
+
+ ret = validate_range(mm, uffdio_move.dst, uffdio_move.len);
+ if (ret)
+ return ret;
+
+ ret = validate_range(mm, uffdio_move.src, uffdio_move.len);
+ if (ret)
+ return ret;
+
+ if (uffdio_move.mode & ~(UFFDIO_MOVE_MODE_ALLOW_SRC_HOLES|
+ UFFDIO_MOVE_MODE_DONTWAKE))
+ return -EINVAL;
+
+ if (mmget_not_zero(mm)) {
+ mmap_read_lock(mm);
+
+ /* Re-check after taking mmap_lock */
+ if (likely(!atomic_read(&ctx->mmap_changing)))
+ ret = move_pages(ctx, mm, uffdio_move.dst, uffdio_move.src,
+ uffdio_move.len, uffdio_move.mode);
+ else
+ ret = -EINVAL;
+
+ mmap_read_unlock(mm);
+ mmput(mm);
+ } else {
+ return -ESRCH;
+ }
+
+ if (unlikely(put_user(ret, &user_uffdio_move->move)))
+ return -EFAULT;
+ if (ret < 0)
+ goto out;
+
+ /* len == 0 would wake all */
+ VM_WARN_ON(!ret);
+ range.len = ret;
+ if (!(uffdio_move.mode & UFFDIO_MOVE_MODE_DONTWAKE)) {
+ range.start = uffdio_move.dst;
+ wake_userfault(ctx, &range);
+ }
+ ret = range.len == uffdio_move.len ? 0 : -EAGAIN;
+
+out:
+ return ret;
+}
+
/*
* userland asks for a certain API version and we return which bits
* and ioctl commands are implemented in this kernel for such API
@@ -2097,6 +2165,9 @@ static long userfaultfd_ioctl(struct file *file, unsigned cmd,
case UFFDIO_ZEROPAGE:
ret = userfaultfd_zeropage(ctx, arg);
break;
+ case UFFDIO_MOVE:
+ ret = userfaultfd_move(ctx, arg);
+ break;
case UFFDIO_WRITEPROTECT:
ret = userfaultfd_writeprotect(ctx, arg);
break;
@@ -2189,7 +2260,8 @@ static int new_userfaultfd(int flags)
/* prevent the mm struct to be freed */
mmgrab(ctx->mm);
- fd = anon_inode_getfd_secure("[userfaultfd]", &userfaultfd_fops, ctx,
+ /* Create a new inode so that the LSM can block the creation. */
+ fd = anon_inode_create_getfd("[userfaultfd]", &userfaultfd_fops, ctx,
O_RDONLY | (flags & UFFD_SHARED_FCNTL_FLAGS), NULL);
if (fd < 0) {
mmdrop(ctx->mm);