Merge tag 'fuse-update-6.2' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse

Pull fuse update from Miklos Szeredi:

 - Allow some write requests to proceed in parallel

 - Fix a performance problem with allow_sys_admin_access

 - Add a special kind of invalidation that doesn't immediately purge
   submounts

 - On revalidation treat the target of rename(RENAME_NOREPLACE) the same
   as open(O_EXCL)

 - Use type safe helpers for some mnt_userns transformations

* tag 'fuse-update-6.2' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse:
  fuse: Rearrange fuse_allow_current_process checks
  fuse: allow non-extending parallel direct writes on the same file
  fuse: remove the unneeded result variable
  fuse: port to vfs{g,u}id_t and associated helpers
  fuse: Remove user_ns check for FUSE_DEV_IOC_CLONE
  fuse: always revalidate rename target dentry
  fuse: add "expire only" mode to FUSE_NOTIFY_INVAL_ENTRY
  fs/fuse: Replace kmap() with kmap_local_page()

7 files changed, 87 insertions, 35 deletions

diff --git a/fs/fuse/cuse.c b/fs/fuse/cuse.c
index c7d882a9fe33..a06fbb1a8a5b 100644
--- a/fs/fuse/cuse.c
+++ b/fs/fuse/cuse.c
@@ -545,7 +545,6 @@ static int cuse_channel_release(struct inode *inode, struct file *file)
 {
 	struct fuse_dev *fud = file->private_data;
 	struct cuse_conn *cc = fc_to_cc(fud->fc);
-	int rc;
 
 	/* remove from the conntbl, no more access from this point on */
 	mutex_lock(&cuse_lock);
@@ -560,9 +559,7 @@ static int cuse_channel_release(struct inode *inode, struct file *file)
 		cdev_del(cc->cdev);
 	}
 
-	rc = fuse_dev_release(inode, file);	/* puts the base reference */
-
-	return rc;
+	return fuse_dev_release(inode, file);
 }
 
 static struct file_operations cuse_channel_fops; /* initialized during init */
diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c
index b4a6e0a1b945..c73d9c4132f6 100644
--- a/fs/fuse/dev.c
+++ b/fs/fuse/dev.c
@@ -1498,7 +1498,7 @@ static int fuse_notify_inval_entry(struct fuse_conn *fc, unsigned int size,
 	buf[outarg.namelen] = 0;
 
 	down_read(&fc->killsb);
-	err = fuse_reverse_inval_entry(fc, outarg.parent, 0, &name);
+	err = fuse_reverse_inval_entry(fc, outarg.parent, 0, &name, outarg.flags);
 	up_read(&fc->killsb);
 	kfree(buf);
 	return err;
@@ -1546,7 +1546,7 @@ static int fuse_notify_delete(struct fuse_conn *fc, unsigned int size,
 	buf[outarg.namelen] = 0;
 
 	down_read(&fc->killsb);
-	err = fuse_reverse_inval_entry(fc, outarg.parent, outarg.child, &name);
+	err = fuse_reverse_inval_entry(fc, outarg.parent, outarg.child, &name, 0);
 	up_read(&fc->killsb);
 	kfree(buf);
 	return err;
@@ -2267,8 +2267,7 @@ static long fuse_dev_ioctl(struct file *file, unsigned int cmd,
 				 * Check against file->f_op because CUSE
 				 * uses the same ioctl handler.
 				 */
-				if (old->f_op == file->f_op &&
-				    old->f_cred->user_ns == file->f_cred->user_ns)
+				if (old->f_op == file->f_op)
 					fud = fuse_get_dev(old);
 
 				if (fud) {
diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c
index 25e6b0f7e73d..cd1a071b625a 100644
--- a/fs/fuse/dir.c
+++ b/fs/fuse/dir.c
@@ -214,7 +214,7 @@ static int fuse_dentry_revalidate(struct dentry *entry, unsigned int flags)
 	if (inode && fuse_is_bad(inode))
 		goto invalid;
 	else if (time_before64(fuse_dentry_time(entry), get_jiffies_64()) ||
-		 (flags & (LOOKUP_EXCL | LOOKUP_REVAL))) {
+		 (flags & (LOOKUP_EXCL | LOOKUP_REVAL | LOOKUP_RENAME_TARGET))) {
 		struct fuse_entry_out outarg;
 		FUSE_ARGS(args);
 		struct fuse_forget_link *forget;
@@ -1170,7 +1170,7 @@ int fuse_update_attributes(struct inode *inode, struct file *file, u32 mask)
 }
 
 int fuse_reverse_inval_entry(struct fuse_conn *fc, u64 parent_nodeid,
-			     u64 child_nodeid, struct qstr *name)
+			     u64 child_nodeid, struct qstr *name, u32 flags)
 {
 	int err = -ENOTDIR;
 	struct inode *parent;
@@ -1197,7 +1197,9 @@ int fuse_reverse_inval_entry(struct fuse_conn *fc, u64 parent_nodeid,
 		goto unlock;
 
 	fuse_dir_changed(parent);
-	fuse_invalidate_entry(entry);
+	if (!(flags & FUSE_EXPIRE_ONLY))
+		d_invalidate(entry);
+	fuse_invalidate_entry_cache(entry);
 
 	if (child_nodeid != 0 && d_really_is_positive(entry)) {
 		inode_lock(d_inode(entry));
@@ -1235,6 +1237,18 @@ int fuse_reverse_inval_entry(struct fuse_conn *fc, u64 parent_nodeid,
 	return err;
 }
 
+static inline bool fuse_permissible_uidgid(struct fuse_conn *fc)
+{
+	const struct cred *cred = current_cred();
+
+	return (uid_eq(cred->euid, fc->user_id) &&
+		uid_eq(cred->suid, fc->user_id) &&
+		uid_eq(cred->uid,  fc->user_id) &&
+		gid_eq(cred->egid, fc->group_id) &&
+		gid_eq(cred->sgid, fc->group_id) &&
+		gid_eq(cred->gid,  fc->group_id));
+}
+
 /*
  * Calling into a user-controlled filesystem gives the filesystem
  * daemon ptrace-like capabilities over the current process.  This
@@ -1248,26 +1262,19 @@ int fuse_reverse_inval_entry(struct fuse_conn *fc, u64 parent_nodeid,
  * for which the owner of the mount has ptrace privilege.  This
  * excludes processes started by other users, suid or sgid processes.
  */
-int fuse_allow_current_process(struct fuse_conn *fc)
+bool fuse_allow_current_process(struct fuse_conn *fc)
 {
-	const struct cred *cred;
-
-	if (allow_sys_admin_access && capable(CAP_SYS_ADMIN))
-		return 1;
+	bool allow;
 
 	if (fc->allow_other)
-		return current_in_userns(fc->user_ns);
+		allow = current_in_userns(fc->user_ns);
+	else
+		allow = fuse_permissible_uidgid(fc);
 
-	cred = current_cred();
-	if (uid_eq(cred->euid, fc->user_id) &&
-	    uid_eq(cred->suid, fc->user_id) &&
-	    uid_eq(cred->uid,  fc->user_id) &&
-	    gid_eq(cred->egid, fc->group_id) &&
-	    gid_eq(cred->sgid, fc->group_id) &&
-	    gid_eq(cred->gid,  fc->group_id))
-		return 1;
+	if (!allow && allow_sys_admin_access && capable(CAP_SYS_ADMIN))
+		allow = true;
 
-	return 0;
+	return allow;
 }
 
 static int fuse_access(struct inode *inode, int mask)
diff --git a/fs/fuse/file.c b/fs/fuse/file.c
index c996c0ef8c63..875314ee6f59 100644
--- a/fs/fuse/file.c
+++ b/fs/fuse/file.c
@@ -1563,14 +1563,47 @@ static ssize_t fuse_direct_read_iter(struct kiocb *iocb, struct iov_iter *to)
 	return res;
 }
 
+static bool fuse_direct_write_extending_i_size(struct kiocb *iocb,
+					       struct iov_iter *iter)
+{
+	struct inode *inode = file_inode(iocb->ki_filp);
+
+	return iocb->ki_pos + iov_iter_count(iter) > i_size_read(inode);
+}
+
 static ssize_t fuse_direct_write_iter(struct kiocb *iocb, struct iov_iter *from)
 {
 	struct inode *inode = file_inode(iocb->ki_filp);
+	struct file *file = iocb->ki_filp;
+	struct fuse_file *ff = file->private_data;
 	struct fuse_io_priv io = FUSE_IO_PRIV_SYNC(iocb);
 	ssize_t res;
+	bool exclusive_lock =
+		!(ff->open_flags & FOPEN_PARALLEL_DIRECT_WRITES) ||
+		iocb->ki_flags & IOCB_APPEND ||
+		fuse_direct_write_extending_i_size(iocb, from);
+
+	/*
+	 * Take exclusive lock if
+	 * - Parallel direct writes are disabled - a user space decision
+	 * - Parallel direct writes are enabled and i_size is being extended.
+	 *   This might not be needed at all, but needs further investigation.
+	 */
+	if (exclusive_lock)
+		inode_lock(inode);
+	else {
+		inode_lock_shared(inode);
+
+		/* A race with truncate might have come up as the decision for
+		 * the lock type was done without holding the lock, check again.
+		 */
+		if (fuse_direct_write_extending_i_size(iocb, from)) {
+			inode_unlock_shared(inode);
+			inode_lock(inode);
+			exclusive_lock = true;
+		}
+	}
 
-	/* Don't allow parallel writes to the same file */
-	inode_lock(inode);
 	res = generic_write_checks(iocb, from);
 	if (res > 0) {
 		if (!is_sync_kiocb(iocb) && iocb->ki_flags & IOCB_DIRECT) {
@@ -1581,7 +1614,10 @@ static ssize_t fuse_direct_write_iter(struct kiocb *iocb, struct iov_iter *from)
 			fuse_write_update_attr(inode, iocb->ki_pos, res);
 		}
 	}
-	inode_unlock(inode);
+	if (exclusive_lock)
+		inode_unlock(inode);
+	else
+		inode_unlock_shared(inode);
 
 	return res;
 }
@@ -2931,6 +2967,7 @@ fuse_direct_IO(struct kiocb *iocb, struct iov_iter *iter)
 
 	if (iov_iter_rw(iter) == WRITE) {
 		fuse_write_update_attr(inode, pos, ret);
+		/* For extending writes we already hold exclusive lock */
 		if (ret < 0 && offset + count > i_size)
 			fuse_do_truncate(file);
 	}
diff --git a/fs/fuse/fuse_i.h b/fs/fuse/fuse_i.h
index 26a7c524eb70..c673faefdcb9 100644
--- a/fs/fuse/fuse_i.h
+++ b/fs/fuse/fuse_i.h
@@ -1179,7 +1179,7 @@ bool fuse_invalid_attr(struct fuse_attr *attr);
 /**
  * Is current process allowed to perform filesystem operation?
  */
-int fuse_allow_current_process(struct fuse_conn *fc);
+bool fuse_allow_current_process(struct fuse_conn *fc);
 
 u64 fuse_lock_owner_id(struct fuse_conn *fc, fl_owner_t id);
 
@@ -1220,7 +1220,7 @@ int fuse_reverse_inval_inode(struct fuse_conn *fc, u64 nodeid,
  * then the dentry is unhashed (d_delete()).
  */
 int fuse_reverse_inval_entry(struct fuse_conn *fc, u64 parent_nodeid,
-			     u64 child_nodeid, struct qstr *name);
+			     u64 child_nodeid, struct qstr *name, u32 flags);
 
 int fuse_do_open(struct fuse_mount *fm, u64 nodeid, struct file *file,
 		 bool isdir);
diff --git a/fs/fuse/readdir.c b/fs/fuse/readdir.c
index e8deaacf1832..dc603479b30e 100644
--- a/fs/fuse/readdir.c
+++ b/fs/fuse/readdir.c
@@ -547,9 +547,9 @@ retry_locked:
 	 * Contents of the page are now protected against changing by holding
 	 * the page lock.
 	 */
-	addr = kmap(page);
+	addr = kmap_local_page(page);
 	res = fuse_parse_cache(ff, addr, size, ctx);
-	kunmap(page);
+	kunmap_local(addr);
 	unlock_page(page);
 	put_page(page);
 
diff --git a/include/uapi/linux/fuse.h b/include/uapi/linux/fuse.h
index 76ee8f9e024a..e3c54109bae9 100644
--- a/include/uapi/linux/fuse.h
+++ b/include/uapi/linux/fuse.h
@@ -197,6 +197,10 @@
  *
  *  7.37
  *  - add FUSE_TMPFILE
+ *
+ *  7.38
+ *  - add FUSE_EXPIRE_ONLY flag to fuse_notify_inval_entry
+ *  - add FOPEN_PARALLEL_DIRECT_WRITES
  */
 
 #ifndef _LINUX_FUSE_H
@@ -232,7 +236,7 @@
 #define FUSE_KERNEL_VERSION 7
 
 /** Minor version number of this interface */
-#define FUSE_KERNEL_MINOR_VERSION 37
+#define FUSE_KERNEL_MINOR_VERSION 38
 
 /** The node ID of the root inode */
 #define FUSE_ROOT_ID 1
@@ -304,6 +308,7 @@ struct fuse_file_lock {
  * FOPEN_CACHE_DIR: allow caching this directory
  * FOPEN_STREAM: the file is stream-like (no file position at all)
  * FOPEN_NOFLUSH: don't flush data cache on close (unless FUSE_WRITEBACK_CACHE)
+ * FOPEN_PARALLEL_DIRECT_WRITES: Allow concurrent direct writes on the same inode
  */
 #define FOPEN_DIRECT_IO		(1 << 0)
 #define FOPEN_KEEP_CACHE	(1 << 1)
@@ -311,6 +316,7 @@ struct fuse_file_lock {
 #define FOPEN_CACHE_DIR		(1 << 3)
 #define FOPEN_STREAM		(1 << 4)
 #define FOPEN_NOFLUSH		(1 << 5)
+#define FOPEN_PARALLEL_DIRECT_WRITES	(1 << 6)
 
 /**
  * INIT request/reply flags
@@ -491,6 +497,12 @@ struct fuse_file_lock {
  */
 #define FUSE_SETXATTR_ACL_KILL_SGID	(1 << 0)
 
+/**
+ * notify_inval_entry flags
+ * FUSE_EXPIRE_ONLY
+ */
+#define FUSE_EXPIRE_ONLY		(1 << 0)
+
 enum fuse_opcode {
 	FUSE_LOOKUP		= 1,
 	FUSE_FORGET		= 2,  /* no reply */
@@ -919,7 +931,7 @@ struct fuse_notify_inval_inode_out {
 struct fuse_notify_inval_entry_out {
 	uint64_t	parent;
 	uint32_t	namelen;
-	uint32_t	padding;
+	uint32_t	flags;
 };
 
 struct fuse_notify_delete_out {