[PATCH v9 00/31] file system-wide error monitoring

[PATCH v9 00/31] file system-wide error monitoring

[Gabriel Krisman Bertazi]

Hi,

This is the 9th version of this patch series.  Thank you, Amir, Jan and
Ted, for the feedback in the previous versions.

The main difference in this version is that the pool is no longer
resizeable nor limited in number of marks, even though we only
pre-allocate 32 slots.  In addition, ext4 was modified to always return
non-zero errno, and the documentation was fixed accordingly (No longer
suggests we return EXT4_ERR* values.

I also droped the Reviewed-by tags from the ext4 patch, due to the
changes above.

Please let me know what you think.

This was tested with LTP for regressions and also using the sample code
on the last patch, with a corrupted image.  I wrote a new ltp test for
this feature which is being reviewed and is available at:

  https://gitlab.collabora.com/krisman/ltp  -b fan-fs-error

In addition, I wrote a man-page that can be pulled from:

  https://gitlab.collabora.com/krisman/man-pages.git -b fan-fs-error

And is being reviewed at the list.

I also pushed this full series to:

  https://gitlab.collabora.com/krisman/linux -b fanotify-notifications-v8

Thank you

Cc: Darrick J. Wong <djwong@kernel.org>
Cc: Theodore Ts'o <tytso@mit.edu>
Cc: Dave Chinner <david@fromorbit.com>
Cc: jack@suse.com
To: amir73il@gmail.com
Cc: dhowells@redhat.com
Cc: khazhy@google.com
Cc: linux-fsdevel@vger.kernel.org
Cc: linux-ext4@vger.kernel.org
Cc: linux-api@vger.kernel.org
Cc: linux-api@vger.kernel.org

Amir Goldstein (3):
  fsnotify: pass data_type to fsnotify_name()
  fsnotify: pass dentry instead of inode data
  fsnotify: clarify contract for create event hooks

Gabriel Krisman Bertazi (28):
  fsnotify: Don't insert unmergeable events in hashtable
  fanotify: Fold event size calculation to its own function
  fanotify: Split fsid check from other fid mode checks
  inotify: Don't force FS_IN_IGNORED
  fsnotify: Add helper to detect overflow_event
  fsnotify: Add wrapper around fsnotify_add_event
  fsnotify: Retrieve super block from the data field
  fsnotify: Protect fsnotify_handle_inode_event from no-inode events
  fsnotify: Pass group argument to free_event
  fanotify: Support null inode event in fanotify_dfid_inode
  fanotify: Allow file handle encoding for unhashed events
  fanotify: Encode empty file handle when no inode is provided
  fanotify: Require fid_mode for any non-fd event
  fsnotify: Support FS_ERROR event type
  fanotify: Reserve UAPI bits for FAN_FS_ERROR
  fanotify: Pre-allocate pool of error events
  fanotify: Support enqueueing of error events
  fanotify: Support merging of error events
  fanotify: Wrap object_fh inline space in a creator macro
  fanotify: Add helpers to decide whether to report FID/DFID
  fanotify: Report fid entry even for zero-length file_handle
  fanotify: WARN_ON against too large file handles
  fanotify: Report fid info for file related file system errors
  fanotify: Emit generic error info for error event
  fanotify: Allow users to request FAN_FS_ERROR events
  ext4: Send notifications on error
  samples: Add fs error monitoring example
  docs: Document the FAN_FS_ERROR event

 .../admin-guide/filesystem-monitoring.rst     |  74 +++++++++
 Documentation/admin-guide/index.rst           |   1 +
 fs/ext4/super.c                               |   8 +
 fs/nfsd/filecache.c                           |   3 +
 fs/notify/fanotify/fanotify.c                 | 117 +++++++++++--
 fs/notify/fanotify/fanotify.h                 |  54 +++++-
 fs/notify/fanotify/fanotify_user.c            | 156 +++++++++++++-----
 fs/notify/fsnotify.c                          |  10 +-
 fs/notify/group.c                             |   2 +-
 fs/notify/inotify/inotify_fsnotify.c          |   5 +-
 fs/notify/inotify/inotify_user.c              |   6 +-
 fs/notify/notification.c                      |  14 +-
 include/linux/fanotify.h                      |   9 +-
 include/linux/fsnotify.h                      |  58 +++++--
 include/linux/fsnotify_backend.h              |  96 ++++++++++-
 include/uapi/linux/fanotify.h                 |   8 +
 kernel/audit_fsnotify.c                       |   3 +-
 kernel/audit_watch.c                          |   3 +-
 samples/Kconfig                               |   9 +
 samples/Makefile                              |   1 +
 samples/fanotify/Makefile                     |   5 +
 samples/fanotify/fs-monitor.c                 | 142 ++++++++++++++++
 22 files changed, 685 insertions(+), 99 deletions(-)
 create mode 100644 Documentation/admin-guide/filesystem-monitoring.rst
 create mode 100644 samples/fanotify/Makefile
 create mode 100644 samples/fanotify/fs-monitor.c

-- 
2.33.0

[PATCH v9 01/31] fsnotify: pass data_type to fsnotify_name()

[Gabriel Krisman Bertazi]

From: Amir Goldstein <amir73il@gmail.com>

Align the arguments of fsnotify_name() to those of fsnotify().

Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>
---
 include/linux/fsnotify.h | 22 +++++++++++++---------
 1 file changed, 13 insertions(+), 9 deletions(-)

diff --git a/include/linux/fsnotify.h b/include/linux/fsnotify.h
index 12d3a7d308ab..d1144d7c3536 100644
--- a/include/linux/fsnotify.h
+++ b/include/linux/fsnotify.h
@@ -26,20 +26,21 @@
  * FS_EVENT_ON_CHILD mask on the parent inode and will not be reported if only
  * the child is interested and not the parent.
  */
-static inline void fsnotify_name(struct inode *dir, __u32 mask,
-				 struct inode *child,
-				 const struct qstr *name, u32 cookie)
+static inline int fsnotify_name(__u32 mask, const void *data, int data_type,
+				struct inode *dir, const struct qstr *name,
+				u32 cookie)
 {
 	if (atomic_long_read(&dir->i_sb->s_fsnotify_connectors) == 0)
-		return;
+		return 0;
 
-	fsnotify(mask, child, FSNOTIFY_EVENT_INODE, dir, name, NULL, cookie);
+	return fsnotify(mask, data, data_type, dir, name, NULL, cookie);
 }
 
 static inline void fsnotify_dirent(struct inode *dir, struct dentry *dentry,
 				   __u32 mask)
 {
-	fsnotify_name(dir, mask, d_inode(dentry), &dentry->d_name, 0);
+	fsnotify_name(mask, d_inode(dentry), FSNOTIFY_EVENT_INODE,
+		      dir, &dentry->d_name, 0);
 }
 
 static inline void fsnotify_inode(struct inode *inode, __u32 mask)
@@ -154,8 +155,10 @@ static inline void fsnotify_move(struct inode *old_dir, struct inode *new_dir,
 		new_dir_mask |= FS_ISDIR;
 	}
 
-	fsnotify_name(old_dir, old_dir_mask, source, old_name, fs_cookie);
-	fsnotify_name(new_dir, new_dir_mask, source, new_name, fs_cookie);
+	fsnotify_name(old_dir_mask, source, FSNOTIFY_EVENT_INODE,
+		      old_dir, old_name, fs_cookie);
+	fsnotify_name(new_dir_mask, source, FSNOTIFY_EVENT_INODE,
+		      new_dir, new_name, fs_cookie);
 
 	if (target)
 		fsnotify_link_count(target);
@@ -209,7 +212,8 @@ static inline void fsnotify_link(struct inode *dir, struct inode *inode,
 	fsnotify_link_count(inode);
 	audit_inode_child(dir, new_dentry, AUDIT_TYPE_CHILD_CREATE);
 
-	fsnotify_name(dir, FS_CREATE, inode, &new_dentry->d_name, 0);
+	fsnotify_name(FS_CREATE, inode, FSNOTIFY_EVENT_INODE,
+		      dir, &new_dentry->d_name, 0);
 }
 
 /*
-- 
2.33.0

[PATCH v9 02/31] fsnotify: pass dentry instead of inode data

[Gabriel Krisman Bertazi]

From: Amir Goldstein <amir73il@gmail.com>

Define a new data type to pass for event - FSNOTIFY_EVENT_DENTRY.
Use it to pass the dentry instead of it's ->d_inode where available.

This is needed in preparation to the refactor to retrieve the super
block from the data field.  In some cases (i.e. mkdir in kernfs), the
data inode comes from a negative dentry, such that no super block
information would be available. By receiving the dentry itself, instead
of the inode, fsnotify can derive the super block even on these cases.

Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
[Expand explanation in commit message]
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>

---
Changes since v7:
  - Improve commit message (Jan)
---
 include/linux/fsnotify.h         |  5 ++---
 include/linux/fsnotify_backend.h | 16 ++++++++++++++++
 2 files changed, 18 insertions(+), 3 deletions(-)

diff --git a/include/linux/fsnotify.h b/include/linux/fsnotify.h
index d1144d7c3536..df0fa4687a18 100644
--- a/include/linux/fsnotify.h
+++ b/include/linux/fsnotify.h
@@ -39,8 +39,7 @@ static inline int fsnotify_name(__u32 mask, const void *data, int data_type,
 static inline void fsnotify_dirent(struct inode *dir, struct dentry *dentry,
 				   __u32 mask)
 {
-	fsnotify_name(mask, d_inode(dentry), FSNOTIFY_EVENT_INODE,
-		      dir, &dentry->d_name, 0);
+	fsnotify_name(mask, dentry, FSNOTIFY_EVENT_DENTRY, dir, &dentry->d_name, 0);
 }
 
 static inline void fsnotify_inode(struct inode *inode, __u32 mask)
@@ -87,7 +86,7 @@ static inline int fsnotify_parent(struct dentry *dentry, __u32 mask,
  */
 static inline void fsnotify_dentry(struct dentry *dentry, __u32 mask)
 {
-	fsnotify_parent(dentry, mask, d_inode(dentry), FSNOTIFY_EVENT_INODE);
+	fsnotify_parent(dentry, mask, dentry, FSNOTIFY_EVENT_DENTRY);
 }
 
 static inline int fsnotify_file(struct file *file, __u32 mask)
diff --git a/include/linux/fsnotify_backend.h b/include/linux/fsnotify_backend.h
index 1ce66748a2d2..a2db821e8a8f 100644
--- a/include/linux/fsnotify_backend.h
+++ b/include/linux/fsnotify_backend.h
@@ -248,6 +248,7 @@ enum fsnotify_data_type {
 	FSNOTIFY_EVENT_NONE,
 	FSNOTIFY_EVENT_PATH,
 	FSNOTIFY_EVENT_INODE,
+	FSNOTIFY_EVENT_DENTRY,
 };
 
 static inline struct inode *fsnotify_data_inode(const void *data, int data_type)
@@ -255,6 +256,8 @@ static inline struct inode *fsnotify_data_inode(const void *data, int data_type)
 	switch (data_type) {
 	case FSNOTIFY_EVENT_INODE:
 		return (struct inode *)data;
+	case FSNOTIFY_EVENT_DENTRY:
+		return d_inode(data);
 	case FSNOTIFY_EVENT_PATH:
 		return d_inode(((const struct path *)data)->dentry);
 	default:
@@ -262,6 +265,19 @@ static inline struct inode *fsnotify_data_inode(const void *data, int data_type)
 	}
 }
 
+static inline struct dentry *fsnotify_data_dentry(const void *data, int data_type)
+{
+	switch (data_type) {
+	case FSNOTIFY_EVENT_DENTRY:
+		/* Non const is needed for dget() */
+		return (struct dentry *)data;
+	case FSNOTIFY_EVENT_PATH:
+		return ((const struct path *)data)->dentry;
+	default:
+		return NULL;
+	}
+}
+
 static inline const struct path *fsnotify_data_path(const void *data,
 						    int data_type)
 {
-- 
2.33.0

[PATCH v9 03/31] fsnotify: clarify contract for create event hooks

[Gabriel Krisman Bertazi]

From: Amir Goldstein <amir73il@gmail.com>

Clarify argument names and contract for fsnotify_create() and
fsnotify_mkdir() to reflect the anomaly of kernfs, which leaves dentries
negavite after mkdir/create.

Remove the WARN_ON(!inode) in audit code that were added by the Fixes
commit under the wrong assumption that dentries cannot be negative after
mkdir/create.

Fixes: aa93bdc5500c ("fsnotify: use helpers to access data by data_type")
Link: https://lore.kernel.org/linux-fsdevel/87mtp5yz0q.fsf@collabora.com/
Reviewed-by: Jan Kara <jack@suse.cz>
Reported-by: Gabriel Krisman Bertazi <krisman@collabora.com>
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>
---
 include/linux/fsnotify.h | 22 ++++++++++++++++------
 kernel/audit_fsnotify.c  |  3 +--
 kernel/audit_watch.c     |  3 +--
 3 files changed, 18 insertions(+), 10 deletions(-)

diff --git a/include/linux/fsnotify.h b/include/linux/fsnotify.h
index df0fa4687a18..1e5f7435a4b5 100644
--- a/include/linux/fsnotify.h
+++ b/include/linux/fsnotify.h
@@ -192,16 +192,22 @@ static inline void fsnotify_inoderemove(struct inode *inode)
 
 /*
  * fsnotify_create - 'name' was linked in
+ *
+ * Caller must make sure that dentry->d_name is stable.
+ * Note: some filesystems (e.g. kernfs) leave @dentry negative and instantiate
+ * ->d_inode later
  */
-static inline void fsnotify_create(struct inode *inode, struct dentry *dentry)
+static inline void fsnotify_create(struct inode *dir, struct dentry *dentry)
 {
-	audit_inode_child(inode, dentry, AUDIT_TYPE_CHILD_CREATE);
+	audit_inode_child(dir, dentry, AUDIT_TYPE_CHILD_CREATE);
 
-	fsnotify_dirent(inode, dentry, FS_CREATE);
+	fsnotify_dirent(dir, dentry, FS_CREATE);
 }
 
 /*
  * fsnotify_link - new hardlink in 'inode' directory
+ *
+ * Caller must make sure that new_dentry->d_name is stable.
  * Note: We have to pass also the linked inode ptr as some filesystems leave
  *   new_dentry->d_inode NULL and instantiate inode pointer later
  */
@@ -230,12 +236,16 @@ static inline void fsnotify_unlink(struct inode *dir, struct dentry *dentry)
 
 /*
  * fsnotify_mkdir - directory 'name' was created
+ *
+ * Caller must make sure that dentry->d_name is stable.
+ * Note: some filesystems (e.g. kernfs) leave @dentry negative and instantiate
+ * ->d_inode later
  */
-static inline void fsnotify_mkdir(struct inode *inode, struct dentry *dentry)
+static inline void fsnotify_mkdir(struct inode *dir, struct dentry *dentry)
 {
-	audit_inode_child(inode, dentry, AUDIT_TYPE_CHILD_CREATE);
+	audit_inode_child(dir, dentry, AUDIT_TYPE_CHILD_CREATE);
 
-	fsnotify_dirent(inode, dentry, FS_CREATE | FS_ISDIR);
+	fsnotify_dirent(dir, dentry, FS_CREATE | FS_ISDIR);
 }
 
 /*
diff --git a/kernel/audit_fsnotify.c b/kernel/audit_fsnotify.c
index 60739d5e3373..02348b48447c 100644
--- a/kernel/audit_fsnotify.c
+++ b/kernel/audit_fsnotify.c
@@ -160,8 +160,7 @@ static int audit_mark_handle_event(struct fsnotify_mark *inode_mark, u32 mask,
 
 	audit_mark = container_of(inode_mark, struct audit_fsnotify_mark, mark);
 
-	if (WARN_ON_ONCE(inode_mark->group != audit_fsnotify_group) ||
-	    WARN_ON_ONCE(!inode))
+	if (WARN_ON_ONCE(inode_mark->group != audit_fsnotify_group))
 		return 0;
 
 	if (mask & (FS_CREATE|FS_MOVED_TO|FS_DELETE|FS_MOVED_FROM)) {
diff --git a/kernel/audit_watch.c b/kernel/audit_watch.c
index 2acf7ca49154..223eed7b39cd 100644
--- a/kernel/audit_watch.c
+++ b/kernel/audit_watch.c
@@ -472,8 +472,7 @@ static int audit_watch_handle_event(struct fsnotify_mark *inode_mark, u32 mask,
 
 	parent = container_of(inode_mark, struct audit_parent, mark);
 
-	if (WARN_ON_ONCE(inode_mark->group != audit_watch_group) ||
-	    WARN_ON_ONCE(!inode))
+	if (WARN_ON_ONCE(inode_mark->group != audit_watch_group))
 		return 0;
 
 	if (mask & (FS_CREATE|FS_MOVED_TO) && inode)
-- 
2.33.0

[PATCH v9 04/31] fsnotify: Don't insert unmergeable events in hashtable

[Gabriel Krisman Bertazi]

Some events, like the overflow event, are not mergeable, so they are not
hashed.  But, when failing inside fsnotify_add_event for lack of space,
fsnotify_add_event() still calls the insert hook, which adds the
overflow event to the merge list.  Add a check to prevent any kind of
unmergeable event to be inserted in the hashtable.

Fixes: 94e00d28a680 ("fsnotify: use hash table for faster events merge")
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>

---
Changes since v2:
  - Do check for hashed events inside the insert hook (Amir)
---
 fs/notify/fanotify/fanotify.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/fs/notify/fanotify/fanotify.c b/fs/notify/fanotify/fanotify.c
index 057abd2cf887..310246f8d3f1 100644
--- a/fs/notify/fanotify/fanotify.c
+++ b/fs/notify/fanotify/fanotify.c
@@ -702,6 +702,9 @@ static void fanotify_insert_event(struct fsnotify_group *group,
 
 	assert_spin_locked(&group->notification_lock);
 
+	if (!fanotify_is_hashed_event(event->mask))
+		return;
+
 	pr_debug("%s: group=%p event=%p bucket=%u\n", __func__,
 		 group, event, bucket);
 
@@ -779,8 +782,7 @@ static int fanotify_handle_event(struct fsnotify_group *group, u32 mask,
 
 	fsn_event = &event->fse;
 	ret = fsnotify_add_event(group, fsn_event, fanotify_merge,
-				 fanotify_is_hashed_event(mask) ?
-				 fanotify_insert_event : NULL);
+				 fanotify_insert_event);
 	if (ret) {
 		/* Permission events shouldn't be merged */
 		BUG_ON(ret == 1 && mask & FANOTIFY_PERM_EVENTS);
-- 
2.33.0

[PATCH v9 05/31] fanotify: Fold event size calculation to its own function

[Gabriel Krisman Bertazi]

Every time this function is invoked, it is immediately added to
FAN_EVENT_METADATA_LEN, since there is no need to just calculate the
length of info records. This minor clean up folds the rest of the
calculation into the function, which now operates in terms of events,
returning the size of the entire event, including metadata.

Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>

---
Changes since v6:
  - Rebase on top of pidfd patches
Changes since v1:
  - rebased on top of hashing patches
---
 fs/notify/fanotify/fanotify_user.c | 35 +++++++++++++++++-------------
 1 file changed, 20 insertions(+), 15 deletions(-)

diff --git a/fs/notify/fanotify/fanotify_user.c b/fs/notify/fanotify/fanotify_user.c
index 6facdf476255..6895ec310b5d 100644
--- a/fs/notify/fanotify/fanotify_user.c
+++ b/fs/notify/fanotify/fanotify_user.c
@@ -126,17 +126,24 @@ static int fanotify_fid_info_len(int fh_len, int name_len)
 		       FANOTIFY_EVENT_ALIGN);
 }
 
-static int fanotify_event_info_len(unsigned int info_mode,
-				   struct fanotify_event *event)
+static size_t fanotify_event_len(unsigned int info_mode,
+				 struct fanotify_event *event)
 {
-	struct fanotify_info *info = fanotify_event_info(event);
-	int dir_fh_len = fanotify_event_dir_fh_len(event);
-	int fh_len = fanotify_event_object_fh_len(event);
-	int info_len = 0;
+	size_t event_len = FAN_EVENT_METADATA_LEN;
+	struct fanotify_info *info;
+	int dir_fh_len;
+	int fh_len;
 	int dot_len = 0;
 
+	if (!info_mode)
+		return event_len;
+
+	info = fanotify_event_info(event);
+	dir_fh_len = fanotify_event_dir_fh_len(event);
+	fh_len = fanotify_event_object_fh_len(event);
+
 	if (dir_fh_len) {
-		info_len += fanotify_fid_info_len(dir_fh_len, info->name_len);
+		event_len += fanotify_fid_info_len(dir_fh_len, info->name_len);
 	} else if ((info_mode & FAN_REPORT_NAME) &&
 		   (event->mask & FAN_ONDIR)) {
 		/*
@@ -147,12 +154,12 @@ static int fanotify_event_info_len(unsigned int info_mode,
 	}
 
 	if (info_mode & FAN_REPORT_PIDFD)
-		info_len += FANOTIFY_PIDFD_INFO_HDR_LEN;
+		event_len += FANOTIFY_PIDFD_INFO_HDR_LEN;
 
 	if (fh_len)
-		info_len += fanotify_fid_info_len(fh_len, dot_len);
+		event_len += fanotify_fid_info_len(fh_len, dot_len);
 
-	return info_len;
+	return event_len;
 }
 
 /*
@@ -181,7 +188,7 @@ static void fanotify_unhash_event(struct fsnotify_group *group,
 static struct fanotify_event *get_one_event(struct fsnotify_group *group,
 					    size_t count)
 {
-	size_t event_size = FAN_EVENT_METADATA_LEN;
+	size_t event_size;
 	struct fanotify_event *event = NULL;
 	struct fsnotify_event *fsn_event;
 	unsigned int info_mode = FAN_GROUP_FLAG(group, FANOTIFY_INFO_MODES);
@@ -194,8 +201,7 @@ static struct fanotify_event *get_one_event(struct fsnotify_group *group,
 		goto out;
 
 	event = FANOTIFY_E(fsn_event);
-	if (info_mode)
-		event_size += fanotify_event_info_len(info_mode, event);
+	event_size = fanotify_event_len(info_mode, event);
 
 	if (event_size > count) {
 		event = ERR_PTR(-EINVAL);
@@ -537,8 +543,7 @@ static ssize_t copy_event_to_user(struct fsnotify_group *group,
 
 	pr_debug("%s: group=%p event=%p\n", __func__, group, event);
 
-	metadata.event_len = FAN_EVENT_METADATA_LEN +
-				fanotify_event_info_len(info_mode, event);
+	metadata.event_len = fanotify_event_len(info_mode, event);
 	metadata.metadata_len = FAN_EVENT_METADATA_LEN;
 	metadata.vers = FANOTIFY_METADATA_VERSION;
 	metadata.reserved = 0;
-- 
2.33.0

[PATCH v9 06/31] fanotify: Split fsid check from other fid mode checks

[Gabriel Krisman Bertazi]

FAN_FS_ERROR will require fsid, but not necessarily require the
filesystem to expose a file handle.  Split those checks into different
functions, so they can be used separately when setting up an event.

While there, update a comment about tmpfs having 0 fsid, which is no
longer true.

Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>

---
Changes since v2:
  - FAN_ERROR -> FAN_FS_ERROR (Amir)
  - Update comment (Amir)

Changes since v1:
  (Amir)
  - Sort hunks to simplify diff.
Changes since RFC:
  (Amir)
  - Rename fanotify_check_path_fsid -> fanotify_test_fsid.
  - Use dentry directly instead of path.
---
 fs/notify/fanotify/fanotify_user.c | 27 ++++++++++++++++++---------
 1 file changed, 18 insertions(+), 9 deletions(-)

diff --git a/fs/notify/fanotify/fanotify_user.c b/fs/notify/fanotify/fanotify_user.c
index 6895ec310b5d..adeae6d65e35 100644
--- a/fs/notify/fanotify/fanotify_user.c
+++ b/fs/notify/fanotify/fanotify_user.c
@@ -1300,16 +1300,15 @@ SYSCALL_DEFINE2(fanotify_init, unsigned int, flags, unsigned int, event_f_flags)
 	return fd;
 }
 
-/* Check if filesystem can encode a unique fid */
-static int fanotify_test_fid(struct path *path, __kernel_fsid_t *fsid)
+static int fanotify_test_fsid(struct dentry *dentry, __kernel_fsid_t *fsid)
 {
 	__kernel_fsid_t root_fsid;
 	int err;
 
 	/*
-	 * Make sure path is not in filesystem with zero fsid (e.g. tmpfs).
+	 * Make sure dentry is not of a filesystem with zero fsid (e.g. fuse).
 	 */
-	err = vfs_get_fsid(path->dentry, fsid);
+	err = vfs_get_fsid(dentry, fsid);
 	if (err)
 		return err;
 
@@ -1317,10 +1316,10 @@ static int fanotify_test_fid(struct path *path, __kernel_fsid_t *fsid)
 		return -ENODEV;
 
 	/*
-	 * Make sure path is not inside a filesystem subvolume (e.g. btrfs)
+	 * Make sure dentry is not of a filesystem subvolume (e.g. btrfs)
 	 * which uses a different fsid than sb root.
 	 */
-	err = vfs_get_fsid(path->dentry->d_sb->s_root, &root_fsid);
+	err = vfs_get_fsid(dentry->d_sb->s_root, &root_fsid);
 	if (err)
 		return err;
 
@@ -1328,6 +1327,12 @@ static int fanotify_test_fid(struct path *path, __kernel_fsid_t *fsid)
 	    root_fsid.val[1] != fsid->val[1])
 		return -EXDEV;
 
+	return 0;
+}
+
+/* Check if filesystem can encode a unique fid */
+static int fanotify_test_fid(struct dentry *dentry)
+{
 	/*
 	 * We need to make sure that the file system supports at least
 	 * encoding a file handle so user can use name_to_handle_at() to
@@ -1335,8 +1340,8 @@ static int fanotify_test_fid(struct path *path, __kernel_fsid_t *fsid)
 	 * objects. However, name_to_handle_at() requires that the
 	 * filesystem also supports decoding file handles.
 	 */
-	if (!path->dentry->d_sb->s_export_op ||
-	    !path->dentry->d_sb->s_export_op->fh_to_dentry)
+	if (!dentry->d_sb->s_export_op ||
+	    !dentry->d_sb->s_export_op->fh_to_dentry)
 		return -EOPNOTSUPP;
 
 	return 0;
@@ -1487,7 +1492,11 @@ static int do_fanotify_mark(int fanotify_fd, unsigned int flags, __u64 mask,
 	}
 
 	if (fid_mode) {
-		ret = fanotify_test_fid(&path, &__fsid);
+		ret = fanotify_test_fsid(path.dentry, &__fsid);
+		if (ret)
+			goto path_put_and_out;
+
+		ret = fanotify_test_fid(path.dentry);
 		if (ret)
 			goto path_put_and_out;
 
-- 
2.33.0

[PATCH v9 07/31] inotify: Don't force FS_IN_IGNORED

[Gabriel Krisman Bertazi]

According to Amir:

"FS_IN_IGNORED is completely internal to inotify and there is no need
to set it in i_fsnotify_mask at all, so if we remove the bit from the
output of inotify_arg_to_mask() no functionality will change and we will
be able to overload the event bit for FS_ERROR."

This is done in preparation to overload FS_ERROR with the notification
mechanism in fanotify.

Suggested-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>
---
 fs/notify/inotify/inotify_user.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/fs/notify/inotify/inotify_user.c b/fs/notify/inotify/inotify_user.c
index 62051247f6d2..29fca3284bb5 100644
--- a/fs/notify/inotify/inotify_user.c
+++ b/fs/notify/inotify/inotify_user.c
@@ -94,10 +94,10 @@ static inline __u32 inotify_arg_to_mask(struct inode *inode, u32 arg)
 	__u32 mask;
 
 	/*
-	 * Everything should accept their own ignored and should receive events
-	 * when the inode is unmounted.  All directories care about children.
+	 * Everything should receive events when the inode is unmounted.
+	 * All directories care about children.
 	 */
-	mask = (FS_IN_IGNORED | FS_UNMOUNT);
+	mask = (FS_UNMOUNT);
 	if (S_ISDIR(inode->i_mode))
 		mask |= FS_EVENT_ON_CHILD;
 
-- 
2.33.0

[PATCH v9 08/31] fsnotify: Add helper to detect overflow_event

[Gabriel Krisman Bertazi]

Similarly to fanotify_is_perm_event and friends, provide a helper
predicate to say whether a mask is of an overflow event.

Suggested-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>
---
 fs/notify/fanotify/fanotify.h    | 3 ++-
 include/linux/fsnotify_backend.h | 5 +++++
 2 files changed, 7 insertions(+), 1 deletion(-)

diff --git a/fs/notify/fanotify/fanotify.h b/fs/notify/fanotify/fanotify.h
index 4a5e555dc3d2..c42cf8fd7d79 100644
--- a/fs/notify/fanotify/fanotify.h
+++ b/fs/notify/fanotify/fanotify.h
@@ -315,7 +315,8 @@ static inline struct path *fanotify_event_path(struct fanotify_event *event)
  */
 static inline bool fanotify_is_hashed_event(u32 mask)
 {
-	return !fanotify_is_perm_event(mask) && !(mask & FS_Q_OVERFLOW);
+	return !(fanotify_is_perm_event(mask) ||
+		 fsnotify_is_overflow_event(mask));
 }
 
 static inline unsigned int fanotify_event_hash_bucket(
diff --git a/include/linux/fsnotify_backend.h b/include/linux/fsnotify_backend.h
index a2db821e8a8f..749bc85e1d1c 100644
--- a/include/linux/fsnotify_backend.h
+++ b/include/linux/fsnotify_backend.h
@@ -510,6 +510,11 @@ static inline void fsnotify_queue_overflow(struct fsnotify_group *group)
 	fsnotify_add_event(group, group->overflow_event, NULL, NULL);
 }
 
+static inline bool fsnotify_is_overflow_event(u32 mask)
+{
+	return mask & FS_Q_OVERFLOW;
+}
+
 static inline bool fsnotify_notify_queue_is_empty(struct fsnotify_group *group)
 {
 	assert_spin_locked(&group->notification_lock);
-- 
2.33.0

[PATCH v9 09/31] fsnotify: Add wrapper around fsnotify_add_event

[Gabriel Krisman Bertazi]

fsnotify_add_event is growing in number of parameters, which in most
case are just passed a NULL pointer.  So, split out a new
fsnotify_insert_event function to clean things up for users who don't
need an insert hook.

Suggested-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>
---
 fs/notify/fanotify/fanotify.c        |  4 ++--
 fs/notify/inotify/inotify_fsnotify.c |  2 +-
 fs/notify/notification.c             | 12 ++++++------
 include/linux/fsnotify_backend.h     | 23 ++++++++++++++++-------
 4 files changed, 25 insertions(+), 16 deletions(-)

diff --git a/fs/notify/fanotify/fanotify.c b/fs/notify/fanotify/fanotify.c
index 310246f8d3f1..f82e20228999 100644
--- a/fs/notify/fanotify/fanotify.c
+++ b/fs/notify/fanotify/fanotify.c
@@ -781,8 +781,8 @@ static int fanotify_handle_event(struct fsnotify_group *group, u32 mask,
 	}
 
 	fsn_event = &event->fse;
-	ret = fsnotify_add_event(group, fsn_event, fanotify_merge,
-				 fanotify_insert_event);
+	ret = fsnotify_insert_event(group, fsn_event, fanotify_merge,
+				    fanotify_insert_event);
 	if (ret) {
 		/* Permission events shouldn't be merged */
 		BUG_ON(ret == 1 && mask & FANOTIFY_PERM_EVENTS);
diff --git a/fs/notify/inotify/inotify_fsnotify.c b/fs/notify/inotify/inotify_fsnotify.c
index d1a64daa0171..a96582cbfad1 100644
--- a/fs/notify/inotify/inotify_fsnotify.c
+++ b/fs/notify/inotify/inotify_fsnotify.c
@@ -116,7 +116,7 @@ int inotify_handle_inode_event(struct fsnotify_mark *inode_mark, u32 mask,
 	if (len)
 		strcpy(event->name, name->name);
 
-	ret = fsnotify_add_event(group, fsn_event, inotify_merge, NULL);
+	ret = fsnotify_add_event(group, fsn_event, inotify_merge);
 	if (ret) {
 		/* Our event wasn't used in the end. Free it. */
 		fsnotify_destroy_event(group, fsn_event);
diff --git a/fs/notify/notification.c b/fs/notify/notification.c
index 32f45543b9c6..44bb10f50715 100644
--- a/fs/notify/notification.c
+++ b/fs/notify/notification.c
@@ -78,12 +78,12 @@ void fsnotify_destroy_event(struct fsnotify_group *group,
  * 2 if the event was not queued - either the queue of events has overflown
  *   or the group is shutting down.
  */
-int fsnotify_add_event(struct fsnotify_group *group,
-		       struct fsnotify_event *event,
-		       int (*merge)(struct fsnotify_group *,
-				    struct fsnotify_event *),
-		       void (*insert)(struct fsnotify_group *,
-				      struct fsnotify_event *))
+int fsnotify_insert_event(struct fsnotify_group *group,
+			  struct fsnotify_event *event,
+			  int (*merge)(struct fsnotify_group *,
+				       struct fsnotify_event *),
+			  void (*insert)(struct fsnotify_group *,
+					 struct fsnotify_event *))
 {
 	int ret = 0;
 	struct list_head *list = &group->notification_list;
diff --git a/include/linux/fsnotify_backend.h b/include/linux/fsnotify_backend.h
index 749bc85e1d1c..b323d0c4b967 100644
--- a/include/linux/fsnotify_backend.h
+++ b/include/linux/fsnotify_backend.h
@@ -498,16 +498,25 @@ extern int fsnotify_fasync(int fd, struct file *file, int on);
 extern void fsnotify_destroy_event(struct fsnotify_group *group,
 				   struct fsnotify_event *event);
 /* attach the event to the group notification queue */
-extern int fsnotify_add_event(struct fsnotify_group *group,
-			      struct fsnotify_event *event,
-			      int (*merge)(struct fsnotify_group *,
-					   struct fsnotify_event *),
-			      void (*insert)(struct fsnotify_group *,
-					     struct fsnotify_event *));
+extern int fsnotify_insert_event(struct fsnotify_group *group,
+				 struct fsnotify_event *event,
+				 int (*merge)(struct fsnotify_group *,
+					      struct fsnotify_event *),
+				 void (*insert)(struct fsnotify_group *,
+						struct fsnotify_event *));
+
+static inline int fsnotify_add_event(struct fsnotify_group *group,
+				     struct fsnotify_event *event,
+				     int (*merge)(struct fsnotify_group *,
+						  struct fsnotify_event *))
+{
+	return fsnotify_insert_event(group, event, merge, NULL);
+}
+
 /* Queue overflow event to a notification group */
 static inline void fsnotify_queue_overflow(struct fsnotify_group *group)
 {
-	fsnotify_add_event(group, group->overflow_event, NULL, NULL);
+	fsnotify_add_event(group, group->overflow_event, NULL);
 }
 
 static inline bool fsnotify_is_overflow_event(u32 mask)
-- 
2.33.0

[PATCH v9 10/31] fsnotify: Retrieve super block from the data field

[Gabriel Krisman Bertazi]

Some file system events (i.e. FS_ERROR) might not be associated with an
inode or directory.  For these, we can retrieve the super block from the
data field.  But, since the super_block is available in the data field
on every event type, simplify the code to always retrieve it from there,
through a new helper.

Suggested-by: Jan Kara <jack@suse.cz>
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>

--
Changes since v6:
  - Always use data field for superblock retrieval
Changes since v5:
  - add fsnotify_data_sb handle to retrieve sb from the data field. (jan)
---
 fs/notify/fsnotify.c             |  7 +++----
 include/linux/fsnotify_backend.h | 15 +++++++++++++++
 2 files changed, 18 insertions(+), 4 deletions(-)

diff --git a/fs/notify/fsnotify.c b/fs/notify/fsnotify.c
index 963e6ce75b96..fde3a1115a17 100644
--- a/fs/notify/fsnotify.c
+++ b/fs/notify/fsnotify.c
@@ -455,16 +455,16 @@ static void fsnotify_iter_next(struct fsnotify_iter_info *iter_info)
  *		@file_name is relative to
  * @file_name:	optional file name associated with event
  * @inode:	optional inode associated with event -
- *		either @dir or @inode must be non-NULL.
- *		if both are non-NULL event may be reported to both.
+ *		If @dir and @inode are both non-NULL, event may be
+ *		reported to both.
  * @cookie:	inotify rename cookie
  */
 int fsnotify(__u32 mask, const void *data, int data_type, struct inode *dir,
 	     const struct qstr *file_name, struct inode *inode, u32 cookie)
 {
 	const struct path *path = fsnotify_data_path(data, data_type);
+	struct super_block *sb = fsnotify_data_sb(data, data_type);
 	struct fsnotify_iter_info iter_info = {};
-	struct super_block *sb;
 	struct mount *mnt = NULL;
 	struct inode *parent = NULL;
 	int ret = 0;
@@ -483,7 +483,6 @@ int fsnotify(__u32 mask, const void *data, int data_type, struct inode *dir,
 		 */
 		parent = dir;
 	}
-	sb = inode->i_sb;
 
 	/*
 	 * Optimization: srcu_read_lock() has a memory barrier which can
diff --git a/include/linux/fsnotify_backend.h b/include/linux/fsnotify_backend.h
index b323d0c4b967..035438fe4a43 100644
--- a/include/linux/fsnotify_backend.h
+++ b/include/linux/fsnotify_backend.h
@@ -289,6 +289,21 @@ static inline const struct path *fsnotify_data_path(const void *data,
 	}
 }
 
+static inline struct super_block *fsnotify_data_sb(const void *data,
+						   int data_type)
+{
+	switch (data_type) {
+	case FSNOTIFY_EVENT_INODE:
+		return ((struct inode *)data)->i_sb;
+	case FSNOTIFY_EVENT_DENTRY:
+		return ((struct dentry *)data)->d_sb;
+	case FSNOTIFY_EVENT_PATH:
+		return ((const struct path *)data)->dentry->d_sb;
+	default:
+		return NULL;
+	}
+}
+
 enum fsnotify_obj_type {
 	FSNOTIFY_OBJ_TYPE_INODE,
 	FSNOTIFY_OBJ_TYPE_PARENT,
-- 
2.33.0

[PATCH v9 11/31] fsnotify: Protect fsnotify_handle_inode_event from no-inode events

[Gabriel Krisman Bertazi]

FAN_FS_ERROR allows events without inodes - i.e. for file system-wide
errors.  Even though fsnotify_handle_inode_event is not currently used
by fanotify, this patch protects other backends from cases where neither
inode or dir are provided.  Also document the constraints of the
interface (inode and dir cannot be both NULL).

Suggested-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>

---
Changes since v8:
  - Convert verifications to WARN_ON
  - Require either inode or dir
  - Protect nfsd backend from !inode.
---
 fs/nfsd/filecache.c              | 3 +++
 fs/notify/fsnotify.c             | 3 +++
 include/linux/fsnotify_backend.h | 1 +
 3 files changed, 7 insertions(+)

diff --git a/fs/nfsd/filecache.c b/fs/nfsd/filecache.c
index be3c1aad50ea..fdf89fcf1a0c 100644
--- a/fs/nfsd/filecache.c
+++ b/fs/nfsd/filecache.c
@@ -602,6 +602,9 @@ nfsd_file_fsnotify_handle_event(struct fsnotify_mark *mark, u32 mask,
 				struct inode *inode, struct inode *dir,
 				const struct qstr *name, u32 cookie)
 {
+	if (WARN_ON_ONCE(!inode))
+		return 0;
+
 	trace_nfsd_file_fsnotify_handle_event(inode, mask);
 
 	/* Should be no marks on non-regular files */
diff --git a/fs/notify/fsnotify.c b/fs/notify/fsnotify.c
index fde3a1115a17..4034ca566f95 100644
--- a/fs/notify/fsnotify.c
+++ b/fs/notify/fsnotify.c
@@ -252,6 +252,9 @@ static int fsnotify_handle_inode_event(struct fsnotify_group *group,
 	if (WARN_ON_ONCE(!ops->handle_inode_event))
 		return 0;
 
+	if (WARN_ON_ONCE(!inode && !dir))
+		return 0;
+
 	if ((inode_mark->mask & FS_EXCL_UNLINK) &&
 	    path && d_unlinked(path->dentry))
 		return 0;
diff --git a/include/linux/fsnotify_backend.h b/include/linux/fsnotify_backend.h
index 035438fe4a43..b71dc788018e 100644
--- a/include/linux/fsnotify_backend.h
+++ b/include/linux/fsnotify_backend.h
@@ -136,6 +136,7 @@ struct mem_cgroup;
  * @dir:	optional directory associated with event -
  *		if @file_name is not NULL, this is the directory that
  *		@file_name is relative to.
+ *		Either @inode or @dir must be non-NULL.
  * @file_name:	optional file name associated with event
  * @cookie:	inotify rename cookie
  *
-- 
2.33.0

Re: [PATCH v9 11/31] fsnotify: Protect fsnotify_handle_inode_event from no-inode events

[Amir Goldstein]

On Mon, Oct 25, 2021 at 10:29 PM Gabriel Krisman Bertazi
<krisman@collabora.com> wrote:
>
> FAN_FS_ERROR allows events without inodes - i.e. for file system-wide
> errors.  Even though fsnotify_handle_inode_event is not currently used
> by fanotify, this patch protects other backends from cases where neither
> inode or dir are provided.  Also document the constraints of the
> interface (inode and dir cannot be both NULL).
>
> Suggested-by: Amir Goldstein <amir73il@gmail.com>
> Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>
>

Reviewed-by: Amir Goldstein <amir73il@gmail.com>

> ---
> Changes since v8:
>   - Convert verifications to WARN_ON
>   - Require either inode or dir
>   - Protect nfsd backend from !inode.
> ---
>  fs/nfsd/filecache.c              | 3 +++
>  fs/notify/fsnotify.c             | 3 +++
>  include/linux/fsnotify_backend.h | 1 +
>  3 files changed, 7 insertions(+)
>
> diff --git a/fs/nfsd/filecache.c b/fs/nfsd/filecache.c
> index be3c1aad50ea..fdf89fcf1a0c 100644
> --- a/fs/nfsd/filecache.c
> +++ b/fs/nfsd/filecache.c
> @@ -602,6 +602,9 @@ nfsd_file_fsnotify_handle_event(struct fsnotify_mark *mark, u32 mask,
>                                 struct inode *inode, struct inode *dir,
>                                 const struct qstr *name, u32 cookie)
>  {
> +       if (WARN_ON_ONCE(!inode))
> +               return 0;
> +
>         trace_nfsd_file_fsnotify_handle_event(inode, mask);
>
>         /* Should be no marks on non-regular files */
> diff --git a/fs/notify/fsnotify.c b/fs/notify/fsnotify.c
> index fde3a1115a17..4034ca566f95 100644
> --- a/fs/notify/fsnotify.c
> +++ b/fs/notify/fsnotify.c
> @@ -252,6 +252,9 @@ static int fsnotify_handle_inode_event(struct fsnotify_group *group,
>         if (WARN_ON_ONCE(!ops->handle_inode_event))
>                 return 0;
>
> +       if (WARN_ON_ONCE(!inode && !dir))
> +               return 0;
> +
>         if ((inode_mark->mask & FS_EXCL_UNLINK) &&
>             path && d_unlinked(path->dentry))
>                 return 0;
> diff --git a/include/linux/fsnotify_backend.h b/include/linux/fsnotify_backend.h
> index 035438fe4a43..b71dc788018e 100644
> --- a/include/linux/fsnotify_backend.h
> +++ b/include/linux/fsnotify_backend.h
> @@ -136,6 +136,7 @@ struct mem_cgroup;
>   * @dir:       optional directory associated with event -
>   *             if @file_name is not NULL, this is the directory that
>   *             @file_name is relative to.
> + *             Either @inode or @dir must be non-NULL.
>   * @file_name: optional file name associated with event
>   * @cookie:    inotify rename cookie
>   *
> --
> 2.33.0
>

Re: [PATCH v9 11/31] fsnotify: Protect fsnotify_handle_inode_event from no-inode events

[Jan Kara]

On Mon 25-10-21 16:27:26, Gabriel Krisman Bertazi wrote:
> FAN_FS_ERROR allows events without inodes - i.e. for file system-wide
> errors.  Even though fsnotify_handle_inode_event is not currently used
> by fanotify, this patch protects other backends from cases where neither
> inode or dir are provided.  Also document the constraints of the
> interface (inode and dir cannot be both NULL).
> 
> Suggested-by: Amir Goldstein <amir73il@gmail.com>
> Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>

Looks good. Feel free to add:

Reviewed-by: Jan Kara <jack@suse.cz>

								Honza

> 
> ---
> Changes since v8:
>   - Convert verifications to WARN_ON
>   - Require either inode or dir
>   - Protect nfsd backend from !inode.
> ---
>  fs/nfsd/filecache.c              | 3 +++
>  fs/notify/fsnotify.c             | 3 +++
>  include/linux/fsnotify_backend.h | 1 +
>  3 files changed, 7 insertions(+)
> 
> diff --git a/fs/nfsd/filecache.c b/fs/nfsd/filecache.c
> index be3c1aad50ea..fdf89fcf1a0c 100644
> --- a/fs/nfsd/filecache.c
> +++ b/fs/nfsd/filecache.c
> @@ -602,6 +602,9 @@ nfsd_file_fsnotify_handle_event(struct fsnotify_mark *mark, u32 mask,
>  				struct inode *inode, struct inode *dir,
>  				const struct qstr *name, u32 cookie)
>  {
> +	if (WARN_ON_ONCE(!inode))
> +		return 0;
> +
>  	trace_nfsd_file_fsnotify_handle_event(inode, mask);
>  
>  	/* Should be no marks on non-regular files */
> diff --git a/fs/notify/fsnotify.c b/fs/notify/fsnotify.c
> index fde3a1115a17..4034ca566f95 100644
> --- a/fs/notify/fsnotify.c
> +++ b/fs/notify/fsnotify.c
> @@ -252,6 +252,9 @@ static int fsnotify_handle_inode_event(struct fsnotify_group *group,
>  	if (WARN_ON_ONCE(!ops->handle_inode_event))
>  		return 0;
>  
> +	if (WARN_ON_ONCE(!inode && !dir))
> +		return 0;
> +
>  	if ((inode_mark->mask & FS_EXCL_UNLINK) &&
>  	    path && d_unlinked(path->dentry))
>  		return 0;
> diff --git a/include/linux/fsnotify_backend.h b/include/linux/fsnotify_backend.h
> index 035438fe4a43..b71dc788018e 100644
> --- a/include/linux/fsnotify_backend.h
> +++ b/include/linux/fsnotify_backend.h
> @@ -136,6 +136,7 @@ struct mem_cgroup;
>   * @dir:	optional directory associated with event -
>   *		if @file_name is not NULL, this is the directory that
>   *		@file_name is relative to.
> + *		Either @inode or @dir must be non-NULL.
>   * @file_name:	optional file name associated with event
>   * @cookie:	inotify rename cookie
>   *
> -- 
> 2.33.0
> 
-- 
Jan Kara <jack@suse.com>
SUSE Labs, CR

[PATCH v9 12/31] fsnotify: Pass group argument to free_event

[Gabriel Krisman Bertazi]

For group-wide mempool backed events, like FS_ERROR, the free_event
callback will need to reference the group's mempool to free the memory.
Wire that argument into the current callers.

Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>
---
 fs/notify/fanotify/fanotify.c        | 3 ++-
 fs/notify/group.c                    | 2 +-
 fs/notify/inotify/inotify_fsnotify.c | 3 ++-
 fs/notify/notification.c             | 2 +-
 include/linux/fsnotify_backend.h     | 2 +-
 5 files changed, 7 insertions(+), 5 deletions(-)

diff --git a/fs/notify/fanotify/fanotify.c b/fs/notify/fanotify/fanotify.c
index f82e20228999..c620b4f6fe12 100644
--- a/fs/notify/fanotify/fanotify.c
+++ b/fs/notify/fanotify/fanotify.c
@@ -835,7 +835,8 @@ static void fanotify_free_name_event(struct fanotify_event *event)
 	kfree(FANOTIFY_NE(event));
 }
 
-static void fanotify_free_event(struct fsnotify_event *fsn_event)
+static void fanotify_free_event(struct fsnotify_group *group,
+				struct fsnotify_event *fsn_event)
 {
 	struct fanotify_event *event;
 
diff --git a/fs/notify/group.c b/fs/notify/group.c
index fb89c351295d..6a297efc4788 100644
--- a/fs/notify/group.c
+++ b/fs/notify/group.c
@@ -88,7 +88,7 @@ void fsnotify_destroy_group(struct fsnotify_group *group)
 	 * that deliberately ignores overflow events.
 	 */
 	if (group->overflow_event)
-		group->ops->free_event(group->overflow_event);
+		group->ops->free_event(group, group->overflow_event);
 
 	fsnotify_put_group(group);
 }
diff --git a/fs/notify/inotify/inotify_fsnotify.c b/fs/notify/inotify/inotify_fsnotify.c
index a96582cbfad1..d92d7b0adc9a 100644
--- a/fs/notify/inotify/inotify_fsnotify.c
+++ b/fs/notify/inotify/inotify_fsnotify.c
@@ -177,7 +177,8 @@ static void inotify_free_group_priv(struct fsnotify_group *group)
 		dec_inotify_instances(group->inotify_data.ucounts);
 }
 
-static void inotify_free_event(struct fsnotify_event *fsn_event)
+static void inotify_free_event(struct fsnotify_group *group,
+			       struct fsnotify_event *fsn_event)
 {
 	kfree(INOTIFY_E(fsn_event));
 }
diff --git a/fs/notify/notification.c b/fs/notify/notification.c
index 44bb10f50715..9022ae650cf8 100644
--- a/fs/notify/notification.c
+++ b/fs/notify/notification.c
@@ -64,7 +64,7 @@ void fsnotify_destroy_event(struct fsnotify_group *group,
 		WARN_ON(!list_empty(&event->list));
 		spin_unlock(&group->notification_lock);
 	}
-	group->ops->free_event(event);
+	group->ops->free_event(group, event);
 }
 
 /*
diff --git a/include/linux/fsnotify_backend.h b/include/linux/fsnotify_backend.h
index b71dc788018e..3a7c31436182 100644
--- a/include/linux/fsnotify_backend.h
+++ b/include/linux/fsnotify_backend.h
@@ -156,7 +156,7 @@ struct fsnotify_ops {
 			    const struct qstr *file_name, u32 cookie);
 	void (*free_group_priv)(struct fsnotify_group *group);
 	void (*freeing_mark)(struct fsnotify_mark *mark, struct fsnotify_group *group);
-	void (*free_event)(struct fsnotify_event *event);
+	void (*free_event)(struct fsnotify_group *group, struct fsnotify_event *event);
 	/* called on final put+free to free memory */
 	void (*free_mark)(struct fsnotify_mark *mark);
 };
-- 
2.33.0

[PATCH v9 13/31] fanotify: Support null inode event in fanotify_dfid_inode

[Gabriel Krisman Bertazi]

FAN_FS_ERROR doesn't support DFID, but this function is still called for
every event.  The problem is that it is not capable of handling null
inodes, which now can happen in case of superblock error events.  For
this case, just returning dir will be enough.

Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>
---
 fs/notify/fanotify/fanotify.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/notify/fanotify/fanotify.c b/fs/notify/fanotify/fanotify.c
index c620b4f6fe12..397ee623ff1e 100644
--- a/fs/notify/fanotify/fanotify.c
+++ b/fs/notify/fanotify/fanotify.c
@@ -452,7 +452,7 @@ static struct inode *fanotify_dfid_inode(u32 event_mask, const void *data,
 	if (event_mask & ALL_FSNOTIFY_DIRENT_EVENTS)
 		return dir;
 
-	if (S_ISDIR(inode->i_mode))
+	if (inode && S_ISDIR(inode->i_mode))
 		return inode;
 
 	return dir;
-- 
2.33.0

[PATCH v9 14/31] fanotify: Allow file handle encoding for unhashed events

[Gabriel Krisman Bertazi]

Allow passing a NULL hash to fanotify_encode_fh and avoid calculating
the hash if not needed.

Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>
---
 fs/notify/fanotify/fanotify.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/fs/notify/fanotify/fanotify.c b/fs/notify/fanotify/fanotify.c
index 397ee623ff1e..ec84fee7ad01 100644
--- a/fs/notify/fanotify/fanotify.c
+++ b/fs/notify/fanotify/fanotify.c
@@ -403,8 +403,12 @@ static int fanotify_encode_fh(struct fanotify_fh *fh, struct inode *inode,
 	fh->type = type;
 	fh->len = fh_len;
 
-	/* Mix fh into event merge key */
-	*hash ^= fanotify_hash_fh(fh);
+	/*
+	 * Mix fh into event merge key.  Hash might be NULL in case of
+	 * unhashed FID events (i.e. FAN_FS_ERROR).
+	 */
+	if (hash)
+		*hash ^= fanotify_hash_fh(fh);
 
 	return FANOTIFY_FH_HDR_LEN + fh_len;
 
-- 
2.33.0

[PATCH v9 15/31] fanotify: Encode empty file handle when no inode is provided

[Gabriel Krisman Bertazi]

Instead of failing, encode an invalid file handle in fanotify_encode_fh
if no inode is provided.  This bogus file handle will be reported by
FAN_FS_ERROR for non-inode errors.

Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>

---
Changes since v6:
  - Use FILEID_ROOT as the internal value (jan)
  - Create an empty FH (jan)

Changes since v5:
  - Preserve flags initialization (jan)
  - Add BUILD_BUG_ON (amir)
  - Require minimum of FANOTIFY_NULL_FH_LEN for fh_len(amir)
  - Improve comment to explain the null FH length (jan)
  - Simplify logic
---
 fs/notify/fanotify/fanotify.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/fs/notify/fanotify/fanotify.c b/fs/notify/fanotify/fanotify.c
index ec84fee7ad01..c64d61b673ca 100644
--- a/fs/notify/fanotify/fanotify.c
+++ b/fs/notify/fanotify/fanotify.c
@@ -370,8 +370,14 @@ static int fanotify_encode_fh(struct fanotify_fh *fh, struct inode *inode,
 	fh->type = FILEID_ROOT;
 	fh->len = 0;
 	fh->flags = 0;
+
+	/*
+	 * Invalid FHs are used by FAN_FS_ERROR for errors not
+	 * linked to any inode. The f_handle won't be reported
+	 * back to userspace.
+	 */
 	if (!inode)
-		return 0;
+		goto out;
 
 	/*
 	 * !gpf means preallocated variable size fh, but fh_len could
@@ -403,6 +409,7 @@ static int fanotify_encode_fh(struct fanotify_fh *fh, struct inode *inode,
 	fh->type = type;
 	fh->len = fh_len;
 
+out:
 	/*
 	 * Mix fh into event merge key.  Hash might be NULL in case of
 	 * unhashed FID events (i.e. FAN_FS_ERROR).
-- 
2.33.0

[PATCH v9 16/31] fanotify: Require fid_mode for any non-fd event

[Gabriel Krisman Bertazi]

Like inode events, FAN_FS_ERROR will require fid mode.  Therefore,
convert the verification during fanotify_mark(2) to require fid for any
non-fd event.  This means fid_mode will not only be required for inode
events, but for any event that doesn't provide a descriptor.

Suggested-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>

---
changes since v5:
  - Fix condition to include FANOTIFY_EVENT_FLAGS. (me)
  - Fix comment identation  (jan)
---
 fs/notify/fanotify/fanotify_user.c | 12 ++++++------
 include/linux/fanotify.h           |  3 +++
 2 files changed, 9 insertions(+), 6 deletions(-)

diff --git a/fs/notify/fanotify/fanotify_user.c b/fs/notify/fanotify/fanotify_user.c
index adeae6d65e35..66ee3c2805c7 100644
--- a/fs/notify/fanotify/fanotify_user.c
+++ b/fs/notify/fanotify/fanotify_user.c
@@ -1458,14 +1458,14 @@ static int do_fanotify_mark(int fanotify_fd, unsigned int flags, __u64 mask,
 		goto fput_and_out;
 
 	/*
-	 * Events with data type inode do not carry enough information to report
-	 * event->fd, so we do not allow setting a mask for inode events unless
-	 * group supports reporting fid.
-	 * inode events are not supported on a mount mark, because they do not
-	 * carry enough information (i.e. path) to be filtered by mount point.
+	 * Events that do not carry enough information to report
+	 * event->fd require a group that supports reporting fid.  Those
+	 * events are not supported on a mount mark, because they do not
+	 * carry enough information (i.e. path) to be filtered by mount
+	 * point.
 	 */
 	fid_mode = FAN_GROUP_FLAG(group, FANOTIFY_FID_BITS);
-	if (mask & FANOTIFY_INODE_EVENTS &&
+	if (mask & ~(FANOTIFY_FD_EVENTS|FANOTIFY_EVENT_FLAGS) &&
 	    (!fid_mode || mark_type == FAN_MARK_MOUNT))
 		goto fput_and_out;
 
diff --git a/include/linux/fanotify.h b/include/linux/fanotify.h
index eec3b7c40811..52d464802d99 100644
--- a/include/linux/fanotify.h
+++ b/include/linux/fanotify.h
@@ -84,6 +84,9 @@ extern struct ctl_table fanotify_table[]; /* for sysctl */
  */
 #define FANOTIFY_DIRENT_EVENTS	(FAN_MOVE | FAN_CREATE | FAN_DELETE)
 
+/* Events that can be reported with event->fd */
+#define FANOTIFY_FD_EVENTS (FANOTIFY_PATH_EVENTS | FANOTIFY_PERM_EVENTS)
+
 /* Events that can only be reported with data type FSNOTIFY_EVENT_INODE */
 #define FANOTIFY_INODE_EVENTS	(FANOTIFY_DIRENT_EVENTS | \
 				 FAN_ATTRIB | FAN_MOVE_SELF | FAN_DELETE_SELF)
-- 
2.33.0

[PATCH v9 17/31] fsnotify: Support FS_ERROR event type

[Gabriel Krisman Bertazi]

Expose a new type of fsnotify event for filesystems to report errors for
userspace monitoring tools.  fanotify will send this type of
notification for FAN_FS_ERROR events.  This also introduce a helper for
generating the new event.

Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>

---
Changes since v6:
  - Add fsnotify_data_error_report

Changes since v5:
  - pass sb inside data field (jan)
Changes since v3:
  - Squash patch ("fsnotify: Introduce helpers to send error_events")
  - Drop reviewed-bys!

Changes since v2:
  - FAN_ERROR->FAN_FS_ERROR (Amir)

Changes since v1:
  - Overload FS_ERROR with FS_IN_IGNORED
  - Implement support for this type on fsnotify_data_inode (Amir)
---
 include/linux/fsnotify.h         | 13 +++++++++++++
 include/linux/fsnotify_backend.h | 32 +++++++++++++++++++++++++++++++-
 2 files changed, 44 insertions(+), 1 deletion(-)

diff --git a/include/linux/fsnotify.h b/include/linux/fsnotify.h
index 1e5f7435a4b5..787545e87eeb 100644
--- a/include/linux/fsnotify.h
+++ b/include/linux/fsnotify.h
@@ -339,4 +339,17 @@ static inline void fsnotify_change(struct dentry *dentry, unsigned int ia_valid)
 		fsnotify_dentry(dentry, mask);
 }
 
+static inline int fsnotify_sb_error(struct super_block *sb, struct inode *inode,
+				    int error)
+{
+	struct fs_error_report report = {
+		.error = error,
+		.inode = inode,
+		.sb = sb,
+	};
+
+	return fsnotify(FS_ERROR, &report, FSNOTIFY_EVENT_ERROR,
+			NULL, NULL, NULL, 0);
+}
+
 #endif	/* _LINUX_FS_NOTIFY_H */
diff --git a/include/linux/fsnotify_backend.h b/include/linux/fsnotify_backend.h
index 3a7c31436182..00dbaafbcf95 100644
--- a/include/linux/fsnotify_backend.h
+++ b/include/linux/fsnotify_backend.h
@@ -42,6 +42,12 @@
 
 #define FS_UNMOUNT		0x00002000	/* inode on umount fs */
 #define FS_Q_OVERFLOW		0x00004000	/* Event queued overflowed */
+#define FS_ERROR		0x00008000	/* Filesystem Error (fanotify) */
+
+/*
+ * FS_IN_IGNORED overloads FS_ERROR.  It is only used internally by inotify
+ * which does not support FS_ERROR.
+ */
 #define FS_IN_IGNORED		0x00008000	/* last inotify event here */
 
 #define FS_OPEN_PERM		0x00010000	/* open event in an permission hook */
@@ -95,7 +101,8 @@
 #define ALL_FSNOTIFY_EVENTS (ALL_FSNOTIFY_DIRENT_EVENTS | \
 			     FS_EVENTS_POSS_ON_CHILD | \
 			     FS_DELETE_SELF | FS_MOVE_SELF | FS_DN_RENAME | \
-			     FS_UNMOUNT | FS_Q_OVERFLOW | FS_IN_IGNORED)
+			     FS_UNMOUNT | FS_Q_OVERFLOW | FS_IN_IGNORED | \
+			     FS_ERROR)
 
 /* Extra flags that may be reported with event or control handling of events */
 #define ALL_FSNOTIFY_FLAGS  (FS_EXCL_UNLINK | FS_ISDIR | FS_IN_ONESHOT | \
@@ -250,6 +257,13 @@ enum fsnotify_data_type {
 	FSNOTIFY_EVENT_PATH,
 	FSNOTIFY_EVENT_INODE,
 	FSNOTIFY_EVENT_DENTRY,
+	FSNOTIFY_EVENT_ERROR,
+};
+
+struct fs_error_report {
+	int error;
+	struct inode *inode;
+	struct super_block *sb;
 };
 
 static inline struct inode *fsnotify_data_inode(const void *data, int data_type)
@@ -261,6 +275,8 @@ static inline struct inode *fsnotify_data_inode(const void *data, int data_type)
 		return d_inode(data);
 	case FSNOTIFY_EVENT_PATH:
 		return d_inode(((const struct path *)data)->dentry);
+	case FSNOTIFY_EVENT_ERROR:
+		return ((struct fs_error_report *)data)->inode;
 	default:
 		return NULL;
 	}
@@ -300,6 +316,20 @@ static inline struct super_block *fsnotify_data_sb(const void *data,
 		return ((struct dentry *)data)->d_sb;
 	case FSNOTIFY_EVENT_PATH:
 		return ((const struct path *)data)->dentry->d_sb;
+	case FSNOTIFY_EVENT_ERROR:
+		return ((struct fs_error_report *) data)->sb;
+	default:
+		return NULL;
+	}
+}
+
+static inline struct fs_error_report *fsnotify_data_error_report(
+							const void *data,
+							int data_type)
+{
+	switch (data_type) {
+	case FSNOTIFY_EVENT_ERROR:
+		return (struct fs_error_report *) data;
 	default:
 		return NULL;
 	}
-- 
2.33.0

[PATCH v9 18/31] fanotify: Reserve UAPI bits for FAN_FS_ERROR

[Gabriel Krisman Bertazi]

FAN_FS_ERROR allows reporting of event type FS_ERROR to userspace, which
is a mechanism to report file system wide problems via fanotify.  This
commit preallocate userspace visible bits to match the FS_ERROR event.

Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>
---
 fs/notify/fanotify/fanotify.c | 1 +
 include/uapi/linux/fanotify.h | 1 +
 2 files changed, 2 insertions(+)

diff --git a/fs/notify/fanotify/fanotify.c b/fs/notify/fanotify/fanotify.c
index c64d61b673ca..8f152445d75c 100644
--- a/fs/notify/fanotify/fanotify.c
+++ b/fs/notify/fanotify/fanotify.c
@@ -752,6 +752,7 @@ static int fanotify_handle_event(struct fsnotify_group *group, u32 mask,
 	BUILD_BUG_ON(FAN_ONDIR != FS_ISDIR);
 	BUILD_BUG_ON(FAN_OPEN_EXEC != FS_OPEN_EXEC);
 	BUILD_BUG_ON(FAN_OPEN_EXEC_PERM != FS_OPEN_EXEC_PERM);
+	BUILD_BUG_ON(FAN_FS_ERROR != FS_ERROR);
 
 	BUILD_BUG_ON(HWEIGHT32(ALL_FANOTIFY_EVENT_BITS) != 19);
 
diff --git a/include/uapi/linux/fanotify.h b/include/uapi/linux/fanotify.h
index 64553df9d735..2990731ddc8b 100644
--- a/include/uapi/linux/fanotify.h
+++ b/include/uapi/linux/fanotify.h
@@ -20,6 +20,7 @@
 #define FAN_OPEN_EXEC		0x00001000	/* File was opened for exec */
 
 #define FAN_Q_OVERFLOW		0x00004000	/* Event queued overflowed */
+#define FAN_FS_ERROR		0x00008000	/* Filesystem error */
 
 #define FAN_OPEN_PERM		0x00010000	/* File open in perm check */
 #define FAN_ACCESS_PERM		0x00020000	/* File accessed in perm check */
-- 
2.33.0

[PATCH v9 19/31] fanotify: Pre-allocate pool of error events

[Gabriel Krisman Bertazi]

Pre-allocate slots for file system errors to have greater chances of
succeeding, since error events can happen in GFP_NOFS context.  This
patch introduces a group-wide mempool of error events, shared by all
FAN_FS_ERROR marks in this group.

For now, just allocate 128 positions.  A future patch allows this
array to be dynamically resized when a new mark is added.

Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>

---
Changes since v8:
  - FANOTIFY_DEFAULT_MAX_FEE_POOL -> FANOTIFY_DEFAULT_FEE_POOL_SIZE
  - Reduce limit to 32. (Jan)
Changes since v7:
  - Expand limit to 128. (Amir)
---
 fs/notify/fanotify/fanotify.c      |  3 +++
 fs/notify/fanotify/fanotify.h      | 11 +++++++++++
 fs/notify/fanotify/fanotify_user.c | 26 +++++++++++++++++++++++++-
 include/linux/fsnotify_backend.h   |  2 ++
 4 files changed, 41 insertions(+), 1 deletion(-)

diff --git a/fs/notify/fanotify/fanotify.c b/fs/notify/fanotify/fanotify.c
index 8f152445d75c..01d68dfc74aa 100644
--- a/fs/notify/fanotify/fanotify.c
+++ b/fs/notify/fanotify/fanotify.c
@@ -819,6 +819,9 @@ static void fanotify_free_group_priv(struct fsnotify_group *group)
 	if (group->fanotify_data.ucounts)
 		dec_ucount(group->fanotify_data.ucounts,
 			   UCOUNT_FANOTIFY_GROUPS);
+
+	if (mempool_initialized(&group->fanotify_data.error_events_pool))
+		mempool_exit(&group->fanotify_data.error_events_pool);
 }
 
 static void fanotify_free_path_event(struct fanotify_event *event)
diff --git a/fs/notify/fanotify/fanotify.h b/fs/notify/fanotify/fanotify.h
index c42cf8fd7d79..a577e87fac2b 100644
--- a/fs/notify/fanotify/fanotify.h
+++ b/fs/notify/fanotify/fanotify.h
@@ -141,6 +141,7 @@ enum fanotify_event_type {
 	FANOTIFY_EVENT_TYPE_PATH,
 	FANOTIFY_EVENT_TYPE_PATH_PERM,
 	FANOTIFY_EVENT_TYPE_OVERFLOW, /* struct fanotify_event */
+	FANOTIFY_EVENT_TYPE_FS_ERROR, /* struct fanotify_error_event */
 	__FANOTIFY_EVENT_TYPE_NUM
 };
 
@@ -196,6 +197,16 @@ FANOTIFY_NE(struct fanotify_event *event)
 	return container_of(event, struct fanotify_name_event, fae);
 }
 
+struct fanotify_error_event {
+	struct fanotify_event fae;
+};
+
+static inline struct fanotify_error_event *
+FANOTIFY_EE(struct fanotify_event *event)
+{
+	return container_of(event, struct fanotify_error_event, fae);
+}
+
 static inline __kernel_fsid_t *fanotify_event_fsid(struct fanotify_event *event)
 {
 	if (event->type == FANOTIFY_EVENT_TYPE_FID)
diff --git a/fs/notify/fanotify/fanotify_user.c b/fs/notify/fanotify/fanotify_user.c
index 66ee3c2805c7..2f4182b754b2 100644
--- a/fs/notify/fanotify/fanotify_user.c
+++ b/fs/notify/fanotify/fanotify_user.c
@@ -30,6 +30,7 @@
 #define FANOTIFY_DEFAULT_MAX_EVENTS	16384
 #define FANOTIFY_OLD_DEFAULT_MAX_MARKS	8192
 #define FANOTIFY_DEFAULT_MAX_GROUPS	128
+#define FANOTIFY_DEFAULT_FEE_POOL_SIZE	32
 
 /*
  * Legacy fanotify marks limits (8192) is per group and we introduced a tunable
@@ -1054,6 +1055,15 @@ static struct fsnotify_mark *fanotify_add_new_mark(struct fsnotify_group *group,
 	return ERR_PTR(ret);
 }
 
+static int fanotify_group_init_error_pool(struct fsnotify_group *group)
+{
+	if (mempool_initialized(&group->fanotify_data.error_events_pool))
+		return 0;
+
+	return mempool_init_kmalloc_pool(&group->fanotify_data.error_events_pool,
+					 FANOTIFY_DEFAULT_FEE_POOL_SIZE,
+					 sizeof(struct fanotify_error_event));
+}
 
 static int fanotify_add_mark(struct fsnotify_group *group,
 			     fsnotify_connp_t *connp, unsigned int type,
@@ -1062,6 +1072,7 @@ static int fanotify_add_mark(struct fsnotify_group *group,
 {
 	struct fsnotify_mark *fsn_mark;
 	__u32 added;
+	int ret = 0;
 
 	mutex_lock(&group->mark_mutex);
 	fsn_mark = fsnotify_find_mark(connp, group);
@@ -1072,13 +1083,26 @@ static int fanotify_add_mark(struct fsnotify_group *group,
 			return PTR_ERR(fsn_mark);
 		}
 	}
+
+	/*
+	 * Error events are pre-allocated per group, only if strictly
+	 * needed (i.e. FAN_FS_ERROR was requested).
+	 */
+	if (!(flags & FAN_MARK_IGNORED_MASK) && (mask & FAN_FS_ERROR)) {
+		ret = fanotify_group_init_error_pool(group);
+		if (ret)
+			goto out;
+	}
+
 	added = fanotify_mark_add_to_mask(fsn_mark, mask, flags);
 	if (added & ~fsnotify_conn_mask(fsn_mark->connector))
 		fsnotify_recalc_mask(fsn_mark->connector);
+
+out:
 	mutex_unlock(&group->mark_mutex);
 
 	fsnotify_put_mark(fsn_mark);
-	return 0;
+	return ret;
 }
 
 static int fanotify_add_vfsmount_mark(struct fsnotify_group *group,
diff --git a/include/linux/fsnotify_backend.h b/include/linux/fsnotify_backend.h
index 00dbaafbcf95..51ef2b079bfa 100644
--- a/include/linux/fsnotify_backend.h
+++ b/include/linux/fsnotify_backend.h
@@ -19,6 +19,7 @@
 #include <linux/atomic.h>
 #include <linux/user_namespace.h>
 #include <linux/refcount.h>
+#include <linux/mempool.h>
 
 /*
  * IN_* from inotfy.h lines up EXACTLY with FS_*, this is so we can easily
@@ -246,6 +247,7 @@ struct fsnotify_group {
 			int flags;           /* flags from fanotify_init() */
 			int f_flags; /* event_f_flags from fanotify_init() */
 			struct ucounts *ucounts;
+			mempool_t error_events_pool;
 		} fanotify_data;
 #endif /* CONFIG_FANOTIFY */
 	};
-- 
2.33.0

Re: [PATCH v9 19/31] fanotify: Pre-allocate pool of error events

[Amir Goldstein]

On Mon, Oct 25, 2021 at 10:30 PM Gabriel Krisman Bertazi
<krisman@collabora.com> wrote:
>
> Pre-allocate slots for file system errors to have greater chances of
> succeeding, since error events can happen in GFP_NOFS context.  This
> patch introduces a group-wide mempool of error events, shared by all
> FAN_FS_ERROR marks in this group.
>
> For now, just allocate 128 positions.  A future patch allows this
> array to be dynamically resized when a new mark is added.

This phrase is out dated.
No need to re-post entire v10 series just for that.
You may suggest the re-phrase here.

Thanks,
Amir.

>
> Reviewed-by: Amir Goldstein <amir73il@gmail.com>
> Reviewed-by: Jan Kara <jack@suse.cz>
> Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>
>
> ---
> Changes since v8:
>   - FANOTIFY_DEFAULT_MAX_FEE_POOL -> FANOTIFY_DEFAULT_FEE_POOL_SIZE
>   - Reduce limit to 32. (Jan)
> Changes since v7:
>   - Expand limit to 128. (Amir)
> ---
>  fs/notify/fanotify/fanotify.c      |  3 +++
>  fs/notify/fanotify/fanotify.h      | 11 +++++++++++
>  fs/notify/fanotify/fanotify_user.c | 26 +++++++++++++++++++++++++-
>  include/linux/fsnotify_backend.h   |  2 ++
>  4 files changed, 41 insertions(+), 1 deletion(-)
>
> diff --git a/fs/notify/fanotify/fanotify.c b/fs/notify/fanotify/fanotify.c
> index 8f152445d75c..01d68dfc74aa 100644
> --- a/fs/notify/fanotify/fanotify.c
> +++ b/fs/notify/fanotify/fanotify.c
> @@ -819,6 +819,9 @@ static void fanotify_free_group_priv(struct fsnotify_group *group)
>         if (group->fanotify_data.ucounts)
>                 dec_ucount(group->fanotify_data.ucounts,
>                            UCOUNT_FANOTIFY_GROUPS);
> +
> +       if (mempool_initialized(&group->fanotify_data.error_events_pool))
> +               mempool_exit(&group->fanotify_data.error_events_pool);
>  }
>
>  static void fanotify_free_path_event(struct fanotify_event *event)
> diff --git a/fs/notify/fanotify/fanotify.h b/fs/notify/fanotify/fanotify.h
> index c42cf8fd7d79..a577e87fac2b 100644
> --- a/fs/notify/fanotify/fanotify.h
> +++ b/fs/notify/fanotify/fanotify.h
> @@ -141,6 +141,7 @@ enum fanotify_event_type {
>         FANOTIFY_EVENT_TYPE_PATH,
>         FANOTIFY_EVENT_TYPE_PATH_PERM,
>         FANOTIFY_EVENT_TYPE_OVERFLOW, /* struct fanotify_event */
> +       FANOTIFY_EVENT_TYPE_FS_ERROR, /* struct fanotify_error_event */
>         __FANOTIFY_EVENT_TYPE_NUM
>  };
>
> @@ -196,6 +197,16 @@ FANOTIFY_NE(struct fanotify_event *event)
>         return container_of(event, struct fanotify_name_event, fae);
>  }
>
> +struct fanotify_error_event {
> +       struct fanotify_event fae;
> +};
> +
> +static inline struct fanotify_error_event *
> +FANOTIFY_EE(struct fanotify_event *event)
> +{
> +       return container_of(event, struct fanotify_error_event, fae);
> +}
> +
>  static inline __kernel_fsid_t *fanotify_event_fsid(struct fanotify_event *event)
>  {
>         if (event->type == FANOTIFY_EVENT_TYPE_FID)
> diff --git a/fs/notify/fanotify/fanotify_user.c b/fs/notify/fanotify/fanotify_user.c
> index 66ee3c2805c7..2f4182b754b2 100644
> --- a/fs/notify/fanotify/fanotify_user.c
> +++ b/fs/notify/fanotify/fanotify_user.c
> @@ -30,6 +30,7 @@
>  #define FANOTIFY_DEFAULT_MAX_EVENTS    16384
>  #define FANOTIFY_OLD_DEFAULT_MAX_MARKS 8192
>  #define FANOTIFY_DEFAULT_MAX_GROUPS    128
> +#define FANOTIFY_DEFAULT_FEE_POOL_SIZE 32
>
>  /*
>   * Legacy fanotify marks limits (8192) is per group and we introduced a tunable
> @@ -1054,6 +1055,15 @@ static struct fsnotify_mark *fanotify_add_new_mark(struct fsnotify_group *group,
>         return ERR_PTR(ret);
>  }
>
> +static int fanotify_group_init_error_pool(struct fsnotify_group *group)
> +{
> +       if (mempool_initialized(&group->fanotify_data.error_events_pool))
> +               return 0;
> +
> +       return mempool_init_kmalloc_pool(&group->fanotify_data.error_events_pool,
> +                                        FANOTIFY_DEFAULT_FEE_POOL_SIZE,
> +                                        sizeof(struct fanotify_error_event));
> +}
>
>  static int fanotify_add_mark(struct fsnotify_group *group,
>                              fsnotify_connp_t *connp, unsigned int type,
> @@ -1062,6 +1072,7 @@ static int fanotify_add_mark(struct fsnotify_group *group,
>  {
>         struct fsnotify_mark *fsn_mark;
>         __u32 added;
> +       int ret = 0;
>
>         mutex_lock(&group->mark_mutex);
>         fsn_mark = fsnotify_find_mark(connp, group);
> @@ -1072,13 +1083,26 @@ static int fanotify_add_mark(struct fsnotify_group *group,
>                         return PTR_ERR(fsn_mark);
>                 }
>         }
> +
> +       /*
> +        * Error events are pre-allocated per group, only if strictly
> +        * needed (i.e. FAN_FS_ERROR was requested).
> +        */
> +       if (!(flags & FAN_MARK_IGNORED_MASK) && (mask & FAN_FS_ERROR)) {
> +               ret = fanotify_group_init_error_pool(group);
> +               if (ret)
> +                       goto out;
> +       }
> +
>         added = fanotify_mark_add_to_mask(fsn_mark, mask, flags);
>         if (added & ~fsnotify_conn_mask(fsn_mark->connector))
>                 fsnotify_recalc_mask(fsn_mark->connector);
> +
> +out:
>         mutex_unlock(&group->mark_mutex);
>
>         fsnotify_put_mark(fsn_mark);
> -       return 0;
> +       return ret;
>  }
>
>  static int fanotify_add_vfsmount_mark(struct fsnotify_group *group,
> diff --git a/include/linux/fsnotify_backend.h b/include/linux/fsnotify_backend.h
> index 00dbaafbcf95..51ef2b079bfa 100644
> --- a/include/linux/fsnotify_backend.h
> +++ b/include/linux/fsnotify_backend.h
> @@ -19,6 +19,7 @@
>  #include <linux/atomic.h>
>  #include <linux/user_namespace.h>
>  #include <linux/refcount.h>
> +#include <linux/mempool.h>
>
>  /*
>   * IN_* from inotfy.h lines up EXACTLY with FS_*, this is so we can easily
> @@ -246,6 +247,7 @@ struct fsnotify_group {
>                         int flags;           /* flags from fanotify_init() */
>                         int f_flags; /* event_f_flags from fanotify_init() */
>                         struct ucounts *ucounts;
> +                       mempool_t error_events_pool;
>                 } fanotify_data;
>  #endif /* CONFIG_FANOTIFY */
>         };
> --
> 2.33.0
>

[PATCH v9 20/31] fanotify: Support enqueueing of error events

[Gabriel Krisman Bertazi]

Once an error event is triggered, enqueue it in the notification group,
similarly to what is done for other events.  FAN_FS_ERROR is not
handled specially, since the memory is now handled by a preallocated
mempool.

For now, make the event unhashed.  A future patch implements merging of
this kind of event.

Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>

---
Changes since v7:
  - WARN_ON -> WARN_ON_ONCE (Amir)
---
 fs/notify/fanotify/fanotify.c | 35 +++++++++++++++++++++++++++++++++++
 fs/notify/fanotify/fanotify.h |  6 ++++++
 2 files changed, 41 insertions(+)

diff --git a/fs/notify/fanotify/fanotify.c b/fs/notify/fanotify/fanotify.c
index 01d68dfc74aa..1f195c95dfcd 100644
--- a/fs/notify/fanotify/fanotify.c
+++ b/fs/notify/fanotify/fanotify.c
@@ -574,6 +574,27 @@ static struct fanotify_event *fanotify_alloc_name_event(struct inode *id,
 	return &fne->fae;
 }
 
+static struct fanotify_event *fanotify_alloc_error_event(
+						struct fsnotify_group *group,
+						__kernel_fsid_t *fsid,
+						const void *data, int data_type)
+{
+	struct fs_error_report *report =
+			fsnotify_data_error_report(data, data_type);
+	struct fanotify_error_event *fee;
+
+	if (WARN_ON_ONCE(!report))
+		return NULL;
+
+	fee = mempool_alloc(&group->fanotify_data.error_events_pool, GFP_NOFS);
+	if (!fee)
+		return NULL;
+
+	fee->fae.type = FANOTIFY_EVENT_TYPE_FS_ERROR;
+
+	return &fee->fae;
+}
+
 static struct fanotify_event *fanotify_alloc_event(struct fsnotify_group *group,
 						   u32 mask, const void *data,
 						   int data_type, struct inode *dir,
@@ -641,6 +662,9 @@ static struct fanotify_event *fanotify_alloc_event(struct fsnotify_group *group,
 
 	if (fanotify_is_perm_event(mask)) {
 		event = fanotify_alloc_perm_event(path, gfp);
+	} else if (fanotify_is_error_event(mask)) {
+		event = fanotify_alloc_error_event(group, fsid, data,
+						   data_type);
 	} else if (name_event && (file_name || child)) {
 		event = fanotify_alloc_name_event(id, fsid, file_name, child,
 						  &hash, gfp);
@@ -850,6 +874,14 @@ static void fanotify_free_name_event(struct fanotify_event *event)
 	kfree(FANOTIFY_NE(event));
 }
 
+static void fanotify_free_error_event(struct fsnotify_group *group,
+				      struct fanotify_event *event)
+{
+	struct fanotify_error_event *fee = FANOTIFY_EE(event);
+
+	mempool_free(fee, &group->fanotify_data.error_events_pool);
+}
+
 static void fanotify_free_event(struct fsnotify_group *group,
 				struct fsnotify_event *fsn_event)
 {
@@ -873,6 +905,9 @@ static void fanotify_free_event(struct fsnotify_group *group,
 	case FANOTIFY_EVENT_TYPE_OVERFLOW:
 		kfree(event);
 		break;
+	case FANOTIFY_EVENT_TYPE_FS_ERROR:
+		fanotify_free_error_event(group, event);
+		break;
 	default:
 		WARN_ON_ONCE(1);
 	}
diff --git a/fs/notify/fanotify/fanotify.h b/fs/notify/fanotify/fanotify.h
index a577e87fac2b..ebef952481fa 100644
--- a/fs/notify/fanotify/fanotify.h
+++ b/fs/notify/fanotify/fanotify.h
@@ -298,6 +298,11 @@ static inline struct fanotify_event *FANOTIFY_E(struct fsnotify_event *fse)
 	return container_of(fse, struct fanotify_event, fse);
 }
 
+static inline bool fanotify_is_error_event(u32 mask)
+{
+	return mask & FAN_FS_ERROR;
+}
+
 static inline bool fanotify_event_has_path(struct fanotify_event *event)
 {
 	return event->type == FANOTIFY_EVENT_TYPE_PATH ||
@@ -327,6 +332,7 @@ static inline struct path *fanotify_event_path(struct fanotify_event *event)
 static inline bool fanotify_is_hashed_event(u32 mask)
 {
 	return !(fanotify_is_perm_event(mask) ||
+		 fanotify_is_error_event(mask) ||
 		 fsnotify_is_overflow_event(mask));
 }
 
-- 
2.33.0

[PATCH v9 21/31] fanotify: Support merging of error events

[Gabriel Krisman Bertazi]

Error events (FAN_FS_ERROR) against the same file system can be merged
by simply iterating the error count.  The hash is taken from the fsid,
without considering the FH.  This means that only the first error object
is reported.

Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>

---
Changes since v7:
  - Move fee->fsid assignment here (Amir)
  - Open code error event merge logic in fanotify_merge (Jan)
---
 fs/notify/fanotify/fanotify.c | 26 ++++++++++++++++++++++++--
 fs/notify/fanotify/fanotify.h |  4 +++-
 2 files changed, 27 insertions(+), 3 deletions(-)

diff --git a/fs/notify/fanotify/fanotify.c b/fs/notify/fanotify/fanotify.c
index 1f195c95dfcd..cedcb1546804 100644
--- a/fs/notify/fanotify/fanotify.c
+++ b/fs/notify/fanotify/fanotify.c
@@ -111,6 +111,16 @@ static bool fanotify_name_event_equal(struct fanotify_name_event *fne1,
 	return fanotify_info_equal(info1, info2);
 }
 
+static bool fanotify_error_event_equal(struct fanotify_error_event *fee1,
+				       struct fanotify_error_event *fee2)
+{
+	/* Error events against the same file system are always merged. */
+	if (!fanotify_fsid_equal(&fee1->fsid, &fee2->fsid))
+		return false;
+
+	return true;
+}
+
 static bool fanotify_should_merge(struct fanotify_event *old,
 				  struct fanotify_event *new)
 {
@@ -141,6 +151,9 @@ static bool fanotify_should_merge(struct fanotify_event *old,
 	case FANOTIFY_EVENT_TYPE_FID_NAME:
 		return fanotify_name_event_equal(FANOTIFY_NE(old),
 						 FANOTIFY_NE(new));
+	case FANOTIFY_EVENT_TYPE_FS_ERROR:
+		return fanotify_error_event_equal(FANOTIFY_EE(old),
+						  FANOTIFY_EE(new));
 	default:
 		WARN_ON_ONCE(1);
 	}
@@ -176,6 +189,10 @@ static int fanotify_merge(struct fsnotify_group *group,
 			break;
 		if (fanotify_should_merge(old, new)) {
 			old->mask |= new->mask;
+
+			if (fanotify_is_error_event(old->mask))
+				FANOTIFY_EE(old)->err_count++;
+
 			return 1;
 		}
 	}
@@ -577,7 +594,8 @@ static struct fanotify_event *fanotify_alloc_name_event(struct inode *id,
 static struct fanotify_event *fanotify_alloc_error_event(
 						struct fsnotify_group *group,
 						__kernel_fsid_t *fsid,
-						const void *data, int data_type)
+						const void *data, int data_type,
+						unsigned int *hash)
 {
 	struct fs_error_report *report =
 			fsnotify_data_error_report(data, data_type);
@@ -591,6 +609,10 @@ static struct fanotify_event *fanotify_alloc_error_event(
 		return NULL;
 
 	fee->fae.type = FANOTIFY_EVENT_TYPE_FS_ERROR;
+	fee->err_count = 1;
+	fee->fsid = *fsid;
+
+	*hash ^= fanotify_hash_fsid(fsid);
 
 	return &fee->fae;
 }
@@ -664,7 +686,7 @@ static struct fanotify_event *fanotify_alloc_event(struct fsnotify_group *group,
 		event = fanotify_alloc_perm_event(path, gfp);
 	} else if (fanotify_is_error_event(mask)) {
 		event = fanotify_alloc_error_event(group, fsid, data,
-						   data_type);
+						   data_type, &hash);
 	} else if (name_event && (file_name || child)) {
 		event = fanotify_alloc_name_event(id, fsid, file_name, child,
 						  &hash, gfp);
diff --git a/fs/notify/fanotify/fanotify.h b/fs/notify/fanotify/fanotify.h
index ebef952481fa..2b032b79d5b0 100644
--- a/fs/notify/fanotify/fanotify.h
+++ b/fs/notify/fanotify/fanotify.h
@@ -199,6 +199,9 @@ FANOTIFY_NE(struct fanotify_event *event)
 
 struct fanotify_error_event {
 	struct fanotify_event fae;
+	u32 err_count; /* Suppressed errors count */
+
+	__kernel_fsid_t fsid; /* FSID this error refers to. */
 };
 
 static inline struct fanotify_error_event *
@@ -332,7 +335,6 @@ static inline struct path *fanotify_event_path(struct fanotify_event *event)
 static inline bool fanotify_is_hashed_event(u32 mask)
 {
 	return !(fanotify_is_perm_event(mask) ||
-		 fanotify_is_error_event(mask) ||
 		 fsnotify_is_overflow_event(mask));
 }
 
-- 
2.33.0

[PATCH v9 22/31] fanotify: Wrap object_fh inline space in a creator macro

[Gabriel Krisman Bertazi]

fanotify_error_event would duplicate this sequence of declarations that
already exist elsewhere with a slight different size.  Create a helper
macro to avoid code duplication.

Suggested-by: Jan Kara <jack@suse.cz>
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>

---
Changes since v8:
  - Drop comment (Amir)
  - Pass fanotify_fh object name to constructor (Jan)
---
 fs/notify/fanotify/fanotify.h | 12 +++++++++---
 1 file changed, 9 insertions(+), 3 deletions(-)

diff --git a/fs/notify/fanotify/fanotify.h b/fs/notify/fanotify/fanotify.h
index 2b032b79d5b0..3510d06654ed 100644
--- a/fs/notify/fanotify/fanotify.h
+++ b/fs/notify/fanotify/fanotify.h
@@ -171,12 +171,18 @@ static inline void fanotify_init_event(struct fanotify_event *event,
 	event->pid = NULL;
 }
 
+#define FANOTIFY_INLINE_FH(name, size)					\
+struct {								\
+	struct fanotify_fh (name);					\
+	/* Space for object_fh.buf[] - access with fanotify_fh_buf() */	\
+	unsigned char _inline_fh_buf[(size)];				\
+}
+
 struct fanotify_fid_event {
 	struct fanotify_event fae;
 	__kernel_fsid_t fsid;
-	struct fanotify_fh object_fh;
-	/* Reserve space in object_fh.buf[] - access with fanotify_fh_buf() */
-	unsigned char _inline_fh_buf[FANOTIFY_INLINE_FH_LEN];
+
+	FANOTIFY_INLINE_FH(object_fh, FANOTIFY_INLINE_FH_LEN);
 };
 
 static inline struct fanotify_fid_event *
-- 
2.33.0

[PATCH v9 23/31] fanotify: Add helpers to decide whether to report FID/DFID

[Gabriel Krisman Bertazi]

Now that there is an event that reports FID records even for a zeroed
file handle, wrap the logic that deides whether to issue the records
into helper functions.  This shouldn't have any impact on the code, but
simplifies further patches.

Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>

---
Changes since v8:
  - Simplify constructs (Amir)
---
 fs/notify/fanotify/fanotify.h      | 10 ++++++++++
 fs/notify/fanotify/fanotify_user.c | 13 +++++++------
 2 files changed, 17 insertions(+), 6 deletions(-)

diff --git a/fs/notify/fanotify/fanotify.h b/fs/notify/fanotify/fanotify.h
index 3510d06654ed..80af269eebb8 100644
--- a/fs/notify/fanotify/fanotify.h
+++ b/fs/notify/fanotify/fanotify.h
@@ -264,6 +264,16 @@ static inline int fanotify_event_dir_fh_len(struct fanotify_event *event)
 	return info ? fanotify_info_dir_fh_len(info) : 0;
 }
 
+static inline bool fanotify_event_has_object_fh(struct fanotify_event *event)
+{
+	return fanotify_event_object_fh_len(event) > 0;
+}
+
+static inline bool fanotify_event_has_dir_fh(struct fanotify_event *event)
+{
+	return fanotify_event_dir_fh_len(event) > 0;
+}
+
 struct fanotify_path_event {
 	struct fanotify_event fae;
 	struct path path;
diff --git a/fs/notify/fanotify/fanotify_user.c b/fs/notify/fanotify/fanotify_user.c
index 2f4182b754b2..a9b5c36ee49e 100644
--- a/fs/notify/fanotify/fanotify_user.c
+++ b/fs/notify/fanotify/fanotify_user.c
@@ -140,10 +140,9 @@ static size_t fanotify_event_len(unsigned int info_mode,
 		return event_len;
 
 	info = fanotify_event_info(event);
-	dir_fh_len = fanotify_event_dir_fh_len(event);
-	fh_len = fanotify_event_object_fh_len(event);
 
-	if (dir_fh_len) {
+	if (fanotify_event_has_dir_fh(event)) {
+		dir_fh_len = fanotify_event_dir_fh_len(event);
 		event_len += fanotify_fid_info_len(dir_fh_len, info->name_len);
 	} else if ((info_mode & FAN_REPORT_NAME) &&
 		   (event->mask & FAN_ONDIR)) {
@@ -157,8 +156,10 @@ static size_t fanotify_event_len(unsigned int info_mode,
 	if (info_mode & FAN_REPORT_PIDFD)
 		event_len += FANOTIFY_PIDFD_INFO_HDR_LEN;
 
-	if (fh_len)
+	if (fanotify_event_has_object_fh(event)) {
+		fh_len = fanotify_event_object_fh_len(event);
 		event_len += fanotify_fid_info_len(fh_len, dot_len);
+	}
 
 	return event_len;
 }
@@ -451,7 +452,7 @@ static int copy_info_records_to_user(struct fanotify_event *event,
 	/*
 	 * Event info records order is as follows: dir fid + name, child fid.
 	 */
-	if (fanotify_event_dir_fh_len(event)) {
+	if (fanotify_event_has_dir_fh(event)) {
 		info_type = info->name_len ? FAN_EVENT_INFO_TYPE_DFID_NAME :
 					     FAN_EVENT_INFO_TYPE_DFID;
 		ret = copy_fid_info_to_user(fanotify_event_fsid(event),
@@ -467,7 +468,7 @@ static int copy_info_records_to_user(struct fanotify_event *event,
 		total_bytes += ret;
 	}
 
-	if (fanotify_event_object_fh_len(event)) {
+	if (fanotify_event_has_object_fh(event)) {
 		const char *dot = NULL;
 		int dot_len = 0;
 
-- 
2.33.0

Re: [PATCH v9 23/31] fanotify: Add helpers to decide whether to report FID/DFID

[Amir Goldstein]

On Mon, Oct 25, 2021 at 10:30 PM Gabriel Krisman Bertazi
<krisman@collabora.com> wrote:
>
> Now that there is an event that reports FID records even for a zeroed
> file handle, wrap the logic that deides whether to issue the records
> into helper functions.  This shouldn't have any impact on the code, but
> simplifies further patches.
>
> Reviewed-by: Jan Kara <jack@suse.cz>
> Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>
>
Reviewed-by: Amir Goldstein <amir73il@gmail.com>

> ---
> Changes since v8:
>   - Simplify constructs (Amir)
> ---
>  fs/notify/fanotify/fanotify.h      | 10 ++++++++++
>  fs/notify/fanotify/fanotify_user.c | 13 +++++++------
>  2 files changed, 17 insertions(+), 6 deletions(-)
>
> diff --git a/fs/notify/fanotify/fanotify.h b/fs/notify/fanotify/fanotify.h
> index 3510d06654ed..80af269eebb8 100644
> --- a/fs/notify/fanotify/fanotify.h
> +++ b/fs/notify/fanotify/fanotify.h
> @@ -264,6 +264,16 @@ static inline int fanotify_event_dir_fh_len(struct fanotify_event *event)
>         return info ? fanotify_info_dir_fh_len(info) : 0;
>  }
>
> +static inline bool fanotify_event_has_object_fh(struct fanotify_event *event)
> +{
> +       return fanotify_event_object_fh_len(event) > 0;
> +}
> +
> +static inline bool fanotify_event_has_dir_fh(struct fanotify_event *event)
> +{
> +       return fanotify_event_dir_fh_len(event) > 0;
> +}
> +
>  struct fanotify_path_event {
>         struct fanotify_event fae;
>         struct path path;
> diff --git a/fs/notify/fanotify/fanotify_user.c b/fs/notify/fanotify/fanotify_user.c
> index 2f4182b754b2..a9b5c36ee49e 100644
> --- a/fs/notify/fanotify/fanotify_user.c
> +++ b/fs/notify/fanotify/fanotify_user.c
> @@ -140,10 +140,9 @@ static size_t fanotify_event_len(unsigned int info_mode,
>                 return event_len;
>
>         info = fanotify_event_info(event);
> -       dir_fh_len = fanotify_event_dir_fh_len(event);
> -       fh_len = fanotify_event_object_fh_len(event);
>
> -       if (dir_fh_len) {
> +       if (fanotify_event_has_dir_fh(event)) {
> +               dir_fh_len = fanotify_event_dir_fh_len(event);
>                 event_len += fanotify_fid_info_len(dir_fh_len, info->name_len);
>         } else if ((info_mode & FAN_REPORT_NAME) &&
>                    (event->mask & FAN_ONDIR)) {
> @@ -157,8 +156,10 @@ static size_t fanotify_event_len(unsigned int info_mode,
>         if (info_mode & FAN_REPORT_PIDFD)
>                 event_len += FANOTIFY_PIDFD_INFO_HDR_LEN;
>
> -       if (fh_len)
> +       if (fanotify_event_has_object_fh(event)) {
> +               fh_len = fanotify_event_object_fh_len(event);
>                 event_len += fanotify_fid_info_len(fh_len, dot_len);
> +       }
>
>         return event_len;
>  }
> @@ -451,7 +452,7 @@ static int copy_info_records_to_user(struct fanotify_event *event,
>         /*
>          * Event info records order is as follows: dir fid + name, child fid.
>          */
> -       if (fanotify_event_dir_fh_len(event)) {
> +       if (fanotify_event_has_dir_fh(event)) {
>                 info_type = info->name_len ? FAN_EVENT_INFO_TYPE_DFID_NAME :
>                                              FAN_EVENT_INFO_TYPE_DFID;
>                 ret = copy_fid_info_to_user(fanotify_event_fsid(event),
> @@ -467,7 +468,7 @@ static int copy_info_records_to_user(struct fanotify_event *event,
>                 total_bytes += ret;
>         }
>
> -       if (fanotify_event_object_fh_len(event)) {
> +       if (fanotify_event_has_object_fh(event)) {
>                 const char *dot = NULL;
>                 int dot_len = 0;
>
> --
> 2.33.0
>

[PATCH v9 24/31] fanotify: Report fid entry even for zero-length file_handle

[Gabriel Krisman Bertazi]

Non-inode errors will reported with an empty file_handle.  In
preparation for that, allow some events to print the FID record even if
there isn't any file_handle encoded

Even though FILEID_ROOT is used internally, make zero-length file
handles be reported as FILEID_INVALID.

Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>

---
Changes since v8:
  - Move fanotify_event_has_object_fh check here (jan)
---
 fs/notify/fanotify/fanotify.h      | 3 +++
 fs/notify/fanotify/fanotify_user.c | 8 +++++---
 2 files changed, 8 insertions(+), 3 deletions(-)

diff --git a/fs/notify/fanotify/fanotify.h b/fs/notify/fanotify/fanotify.h
index 80af269eebb8..f51ab6e556e8 100644
--- a/fs/notify/fanotify/fanotify.h
+++ b/fs/notify/fanotify/fanotify.h
@@ -266,6 +266,9 @@ static inline int fanotify_event_dir_fh_len(struct fanotify_event *event)
 
 static inline bool fanotify_event_has_object_fh(struct fanotify_event *event)
 {
+	/* For error events, even zeroed fh are reported. */
+	if (event->type == FANOTIFY_EVENT_TYPE_FS_ERROR)
+		return true;
 	return fanotify_event_object_fh_len(event) > 0;
 }
 
diff --git a/fs/notify/fanotify/fanotify_user.c b/fs/notify/fanotify/fanotify_user.c
index a9b5c36ee49e..ff4a7373f7a5 100644
--- a/fs/notify/fanotify/fanotify_user.c
+++ b/fs/notify/fanotify/fanotify_user.c
@@ -339,9 +339,6 @@ static int copy_fid_info_to_user(__kernel_fsid_t *fsid, struct fanotify_fh *fh,
 	pr_debug("%s: fh_len=%zu name_len=%zu, info_len=%zu, count=%zu\n",
 		 __func__, fh_len, name_len, info_len, count);
 
-	if (!fh_len)
-		return 0;
-
 	if (WARN_ON_ONCE(len < sizeof(info) || len > count))
 		return -EFAULT;
 
@@ -376,6 +373,11 @@ static int copy_fid_info_to_user(__kernel_fsid_t *fsid, struct fanotify_fh *fh,
 
 	handle.handle_type = fh->type;
 	handle.handle_bytes = fh_len;
+
+	/* Mangle handle_type for bad file_handle */
+	if (!fh_len)
+		handle.handle_type = FILEID_INVALID;
+
 	if (copy_to_user(buf, &handle, sizeof(handle)))
 		return -EFAULT;
 
-- 
2.33.0

Re: [PATCH v9 24/31] fanotify: Report fid entry even for zero-length file_handle

[Amir Goldstein]

On Mon, Oct 25, 2021 at 10:30 PM Gabriel Krisman Bertazi
<krisman@collabora.com> wrote:
>
> Non-inode errors will reported with an empty file_handle.  In
> preparation for that, allow some events to print the FID record even if
> there isn't any file_handle encoded
>
> Even though FILEID_ROOT is used internally, make zero-length file
> handles be reported as FILEID_INVALID.
>
> Reviewed-by: Amir Goldstein <amir73il@gmail.com>
> Reviewed-by: Jan Kara <jack@suse.cz>
> Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>
>
> ---
> Changes since v8:
>   - Move fanotify_event_has_object_fh check here (jan)

Logically, this move is wrong, because after this patch,
copy_fid_info_to_user() can theoretically be called with NULL fh in the
existing construct of:
  if (fanotify_event_has_object_fh(event)) {
  ...
    ret = copy_fid_info_to_user(fanotify_event_fsid(event),

fanotify_event_object_fh(event),

The thing that prevents this case in effect is that FAN_FS_ERROR
is not yet wired, but I am not sure if leaving this theoretic bisect
issue is a good idea.

Anyway, that's a very minor theoretic issue and I am sure Jan can
decide whether to deal with it and how (no need to post v10 IMO).

Thanks,
Amir.

Re: [PATCH v9 24/31] fanotify: Report fid entry even for zero-length file_handle

[Jan Kara]

On Tue 26-10-21 12:09:19, Amir Goldstein wrote:
> On Mon, Oct 25, 2021 at 10:30 PM Gabriel Krisman Bertazi
> <krisman@collabora.com> wrote:
> >
> > Non-inode errors will reported with an empty file_handle.  In
> > preparation for that, allow some events to print the FID record even if
> > there isn't any file_handle encoded
> >
> > Even though FILEID_ROOT is used internally, make zero-length file
> > handles be reported as FILEID_INVALID.
> >
> > Reviewed-by: Amir Goldstein <amir73il@gmail.com>
> > Reviewed-by: Jan Kara <jack@suse.cz>
> > Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>
> >
> > ---
> > Changes since v8:
> >   - Move fanotify_event_has_object_fh check here (jan)
> 
> Logically, this move is wrong, because after this patch,
> copy_fid_info_to_user() can theoretically be called with NULL fh in the
> existing construct of:
>   if (fanotify_event_has_object_fh(event)) {
>   ...
>     ret = copy_fid_info_to_user(fanotify_event_fsid(event),
> 
> fanotify_event_object_fh(event),
> 
> The thing that prevents this case in effect is that FAN_FS_ERROR
> is not yet wired, but I am not sure if leaving this theoretic bisect
> issue is a good idea.
> 
> Anyway, that's a very minor theoretic issue and I am sure Jan can
> decide whether to deal with it and how (no need to post v10 IMO).

Hum, correct. I guess I'll just fold this patch into patch 26. Logically
they are very close anyway.

								Honza
-- 
Jan Kara <jack@suse.com>
SUSE Labs, CR

[PATCH v9 25/31] fanotify: WARN_ON against too large file handles

[Gabriel Krisman Bertazi]

struct fanotify_error_event, at least, is preallocated and isn't able to
to handle arbitrarily large file handles.  Future-proof the code by
complaining loudly if a handle larger than MAX_HANDLE_SZ is ever found.

Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>
---
 fs/notify/fanotify/fanotify.c | 12 +++++++++++-
 1 file changed, 11 insertions(+), 1 deletion(-)

diff --git a/fs/notify/fanotify/fanotify.c b/fs/notify/fanotify/fanotify.c
index cedcb1546804..45df610debbe 100644
--- a/fs/notify/fanotify/fanotify.c
+++ b/fs/notify/fanotify/fanotify.c
@@ -360,13 +360,23 @@ static u32 fanotify_group_event_mask(struct fsnotify_group *group,
 static int fanotify_encode_fh_len(struct inode *inode)
 {
 	int dwords = 0;
+	int fh_len;
 
 	if (!inode)
 		return 0;
 
 	exportfs_encode_inode_fh(inode, NULL, &dwords, NULL);
+	fh_len = dwords << 2;
 
-	return dwords << 2;
+	/*
+	 * struct fanotify_error_event might be preallocated and is
+	 * limited to MAX_HANDLE_SZ.  This should never happen, but
+	 * safeguard by forcing an invalid file handle.
+	 */
+	if (WARN_ON_ONCE(fh_len > MAX_HANDLE_SZ))
+		return 0;
+
+	return fh_len;
 }
 
 /*
-- 
2.33.0

[PATCH v9 26/31] fanotify: Report fid info for file related file system errors

[Gabriel Krisman Bertazi]

Plumb the pieces to add a FID report to error records.  Since all error
event memory must be pre-allocated, we pre-allocate the maximum file
handle size possible, such that it should always fit.

For errors that don't expose a file handle report it with an invalid
FID.

Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>

---
Changes since v8:
  - Drop comment (Amir)
  - Move report dereference till after null check (jan)
  - Move has_object_fh hunk to earlier patch (jan)
Changes since v7:
  - Move WARN_ON to separate patch (Amir)
  - Avoid duplication in the structure definition (Amir)
Changes since v6:
  - pass fsid from handle_events
Changes since v5:
  - Use preallocated MAX_HANDLE_SZ FH buffer
  - Report superblock errors with a zerolength INVALID FID (jan, amir)
---
 fs/notify/fanotify/fanotify.c | 11 +++++++++++
 fs/notify/fanotify/fanotify.h |  6 ++++++
 2 files changed, 17 insertions(+)

diff --git a/fs/notify/fanotify/fanotify.c b/fs/notify/fanotify/fanotify.c
index 45df610debbe..465f07e70e6d 100644
--- a/fs/notify/fanotify/fanotify.c
+++ b/fs/notify/fanotify/fanotify.c
@@ -609,7 +609,9 @@ static struct fanotify_event *fanotify_alloc_error_event(
 {
 	struct fs_error_report *report =
 			fsnotify_data_error_report(data, data_type);
+	struct inode *inode;
 	struct fanotify_error_event *fee;
+	int fh_len;
 
 	if (WARN_ON_ONCE(!report))
 		return NULL;
@@ -622,6 +624,15 @@ static struct fanotify_event *fanotify_alloc_error_event(
 	fee->err_count = 1;
 	fee->fsid = *fsid;
 
+	inode = report->inode;
+	fh_len = fanotify_encode_fh_len(inode);
+
+	/* Bad fh_len. Fallback to using an invalid fh. Should never happen. */
+	if (!fh_len && inode)
+		inode = NULL;
+
+	fanotify_encode_fh(&fee->object_fh, inode, fh_len, NULL, 0);
+
 	*hash ^= fanotify_hash_fsid(fsid);
 
 	return &fee->fae;
diff --git a/fs/notify/fanotify/fanotify.h b/fs/notify/fanotify/fanotify.h
index f51ab6e556e8..edd7587adcc5 100644
--- a/fs/notify/fanotify/fanotify.h
+++ b/fs/notify/fanotify/fanotify.h
@@ -208,6 +208,8 @@ struct fanotify_error_event {
 	u32 err_count; /* Suppressed errors count */
 
 	__kernel_fsid_t fsid; /* FSID this error refers to. */
+
+	FANOTIFY_INLINE_FH(object_fh, MAX_HANDLE_SZ);
 };
 
 static inline struct fanotify_error_event *
@@ -222,6 +224,8 @@ static inline __kernel_fsid_t *fanotify_event_fsid(struct fanotify_event *event)
 		return &FANOTIFY_FE(event)->fsid;
 	else if (event->type == FANOTIFY_EVENT_TYPE_FID_NAME)
 		return &FANOTIFY_NE(event)->fsid;
+	else if (event->type == FANOTIFY_EVENT_TYPE_FS_ERROR)
+		return &FANOTIFY_EE(event)->fsid;
 	else
 		return NULL;
 }
@@ -233,6 +237,8 @@ static inline struct fanotify_fh *fanotify_event_object_fh(
 		return &FANOTIFY_FE(event)->object_fh;
 	else if (event->type == FANOTIFY_EVENT_TYPE_FID_NAME)
 		return fanotify_info_file_fh(&FANOTIFY_NE(event)->info);
+	else if (event->type == FANOTIFY_EVENT_TYPE_FS_ERROR)
+		return &FANOTIFY_EE(event)->object_fh;
 	else
 		return NULL;
 }
-- 
2.33.0

[PATCH v9 27/31] fanotify: Emit generic error info for error event

[Gabriel Krisman Bertazi]

The error info is a record sent to users on FAN_FS_ERROR events
documenting the type of error.  It also carries an error count,
documenting how many errors were observed since the last reporting.

Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>

---
Changes since v6:
  - Rebase on top of pidfd patches
Changes since v5:
  - Move error code here
---
 fs/notify/fanotify/fanotify.c      |  1 +
 fs/notify/fanotify/fanotify.h      |  1 +
 fs/notify/fanotify/fanotify_user.c | 35 ++++++++++++++++++++++++++++++
 include/uapi/linux/fanotify.h      |  7 ++++++
 4 files changed, 44 insertions(+)

diff --git a/fs/notify/fanotify/fanotify.c b/fs/notify/fanotify/fanotify.c
index 465f07e70e6d..af61425e6e3b 100644
--- a/fs/notify/fanotify/fanotify.c
+++ b/fs/notify/fanotify/fanotify.c
@@ -621,6 +621,7 @@ static struct fanotify_event *fanotify_alloc_error_event(
 		return NULL;
 
 	fee->fae.type = FANOTIFY_EVENT_TYPE_FS_ERROR;
+	fee->error = report->error;
 	fee->err_count = 1;
 	fee->fsid = *fsid;
 
diff --git a/fs/notify/fanotify/fanotify.h b/fs/notify/fanotify/fanotify.h
index edd7587adcc5..d25f500bf7e7 100644
--- a/fs/notify/fanotify/fanotify.h
+++ b/fs/notify/fanotify/fanotify.h
@@ -205,6 +205,7 @@ FANOTIFY_NE(struct fanotify_event *event)
 
 struct fanotify_error_event {
 	struct fanotify_event fae;
+	s32 error; /* Error reported by the Filesystem. */
 	u32 err_count; /* Suppressed errors count */
 
 	__kernel_fsid_t fsid; /* FSID this error refers to. */
diff --git a/fs/notify/fanotify/fanotify_user.c b/fs/notify/fanotify/fanotify_user.c
index ff4a7373f7a5..97ec5b005b50 100644
--- a/fs/notify/fanotify/fanotify_user.c
+++ b/fs/notify/fanotify/fanotify_user.c
@@ -115,6 +115,8 @@ struct kmem_cache *fanotify_perm_event_cachep __read_mostly;
 	(sizeof(struct fanotify_event_info_fid) + sizeof(struct file_handle))
 #define FANOTIFY_PIDFD_INFO_HDR_LEN \
 	sizeof(struct fanotify_event_info_pidfd)
+#define FANOTIFY_ERROR_INFO_LEN \
+	(sizeof(struct fanotify_event_info_error))
 
 static int fanotify_fid_info_len(int fh_len, int name_len)
 {
@@ -139,6 +141,9 @@ static size_t fanotify_event_len(unsigned int info_mode,
 	if (!info_mode)
 		return event_len;
 
+	if (fanotify_is_error_event(event->mask))
+		event_len += FANOTIFY_ERROR_INFO_LEN;
+
 	info = fanotify_event_info(event);
 
 	if (fanotify_event_has_dir_fh(event)) {
@@ -324,6 +329,28 @@ static int process_access_response(struct fsnotify_group *group,
 	return -ENOENT;
 }
 
+static size_t copy_error_info_to_user(struct fanotify_event *event,
+				      char __user *buf, int count)
+{
+	struct fanotify_event_info_error info;
+	struct fanotify_error_event *fee = FANOTIFY_EE(event);
+
+	info.hdr.info_type = FAN_EVENT_INFO_TYPE_ERROR;
+	info.hdr.pad = 0;
+	info.hdr.len = FANOTIFY_ERROR_INFO_LEN;
+
+	if (WARN_ON(count < info.hdr.len))
+		return -EFAULT;
+
+	info.error = fee->error;
+	info.error_count = fee->err_count;
+
+	if (copy_to_user(buf, &info, sizeof(info)))
+		return -EFAULT;
+
+	return info.hdr.len;
+}
+
 static int copy_fid_info_to_user(__kernel_fsid_t *fsid, struct fanotify_fh *fh,
 				 int info_type, const char *name,
 				 size_t name_len,
@@ -530,6 +557,14 @@ static int copy_info_records_to_user(struct fanotify_event *event,
 		total_bytes += ret;
 	}
 
+	if (fanotify_is_error_event(event->mask)) {
+		ret = copy_error_info_to_user(event, buf, count);
+		if (ret < 0)
+			return ret;
+		buf += ret;
+		count -= ret;
+	}
+
 	return total_bytes;
 }
 
diff --git a/include/uapi/linux/fanotify.h b/include/uapi/linux/fanotify.h
index 2990731ddc8b..bd1932c2074d 100644
--- a/include/uapi/linux/fanotify.h
+++ b/include/uapi/linux/fanotify.h
@@ -126,6 +126,7 @@ struct fanotify_event_metadata {
 #define FAN_EVENT_INFO_TYPE_DFID_NAME	2
 #define FAN_EVENT_INFO_TYPE_DFID	3
 #define FAN_EVENT_INFO_TYPE_PIDFD	4
+#define FAN_EVENT_INFO_TYPE_ERROR	5
 
 /* Variable length info record following event metadata */
 struct fanotify_event_info_header {
@@ -160,6 +161,12 @@ struct fanotify_event_info_pidfd {
 	__s32 pidfd;
 };
 
+struct fanotify_event_info_error {
+	struct fanotify_event_info_header hdr;
+	__s32 error;
+	__u32 error_count;
+};
+
 struct fanotify_response {
 	__s32 fd;
 	__u32 response;
-- 
2.33.0

Re: [PATCH v9 27/31] fanotify: Emit generic error info for error event

[Jan Kara]

On Mon 25-10-21 16:27:42, Gabriel Krisman Bertazi wrote:
> The error info is a record sent to users on FAN_FS_ERROR events
> documenting the type of error.  It also carries an error count,
> documenting how many errors were observed since the last reporting.
> 
> Reviewed-by: Amir Goldstein <amir73il@gmail.com>
> Reviewed-by: Jan Kara <jack@suse.cz>
> Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>

...

> @@ -530,6 +557,14 @@ static int copy_info_records_to_user(struct fanotify_event *event,
>  		total_bytes += ret;
>  	}
>  
> +	if (fanotify_is_error_event(event->mask)) {
> +		ret = copy_error_info_to_user(event, buf, count);
> +		if (ret < 0)
> +			return ret;
> +		buf += ret;
> +		count -= ret;
> +	}
> +
>  	return total_bytes;
>  }

This is currently harmless but we should add

total_bytes += ret;

here as well.

								Honza
-- 
Jan Kara <jack@suse.com>
SUSE Labs, CR

[PATCH v9 28/31] fanotify: Allow users to request FAN_FS_ERROR events

[Gabriel Krisman Bertazi]

Wire up the FAN_FS_ERROR event in the fanotify_mark syscall, allowing
user space to request the monitoring of FAN_FS_ERROR events.

These events are limited to filesystem marks, so check it is the
case in the syscall handler.

Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>

---
Changes since v7:
  - Move the verification closer to similar code (Amir)
---
 fs/notify/fanotify/fanotify.c      | 2 +-
 fs/notify/fanotify/fanotify_user.c | 4 ++++
 include/linux/fanotify.h           | 6 +++++-
 3 files changed, 10 insertions(+), 2 deletions(-)

diff --git a/fs/notify/fanotify/fanotify.c b/fs/notify/fanotify/fanotify.c
index af61425e6e3b..b6091775aa6e 100644
--- a/fs/notify/fanotify/fanotify.c
+++ b/fs/notify/fanotify/fanotify.c
@@ -822,7 +822,7 @@ static int fanotify_handle_event(struct fsnotify_group *group, u32 mask,
 	BUILD_BUG_ON(FAN_OPEN_EXEC_PERM != FS_OPEN_EXEC_PERM);
 	BUILD_BUG_ON(FAN_FS_ERROR != FS_ERROR);
 
-	BUILD_BUG_ON(HWEIGHT32(ALL_FANOTIFY_EVENT_BITS) != 19);
+	BUILD_BUG_ON(HWEIGHT32(ALL_FANOTIFY_EVENT_BITS) != 20);
 
 	mask = fanotify_group_event_mask(group, iter_info, mask, data,
 					 data_type, dir);
diff --git a/fs/notify/fanotify/fanotify_user.c b/fs/notify/fanotify/fanotify_user.c
index 97ec5b005b50..77a77064f3c9 100644
--- a/fs/notify/fanotify/fanotify_user.c
+++ b/fs/notify/fanotify/fanotify_user.c
@@ -1519,6 +1519,10 @@ static int do_fanotify_mark(int fanotify_fd, unsigned int flags, __u64 mask,
 	    group->priority == FS_PRIO_0)
 		goto fput_and_out;
 
+	if (mask & FAN_FS_ERROR &&
+	    mark_type != FAN_MARK_FILESYSTEM)
+		goto fput_and_out;
+
 	/*
 	 * Events that do not carry enough information to report
 	 * event->fd require a group that supports reporting fid.  Those
diff --git a/include/linux/fanotify.h b/include/linux/fanotify.h
index 52d464802d99..616af2ea20f3 100644
--- a/include/linux/fanotify.h
+++ b/include/linux/fanotify.h
@@ -91,9 +91,13 @@ extern struct ctl_table fanotify_table[]; /* for sysctl */
 #define FANOTIFY_INODE_EVENTS	(FANOTIFY_DIRENT_EVENTS | \
 				 FAN_ATTRIB | FAN_MOVE_SELF | FAN_DELETE_SELF)
 
+/* Events that can only be reported with data type FSNOTIFY_EVENT_ERROR */
+#define FANOTIFY_ERROR_EVENTS	(FAN_FS_ERROR)
+
 /* Events that user can request to be notified on */
 #define FANOTIFY_EVENTS		(FANOTIFY_PATH_EVENTS | \
-				 FANOTIFY_INODE_EVENTS)
+				 FANOTIFY_INODE_EVENTS | \
+				 FANOTIFY_ERROR_EVENTS)
 
 /* Events that require a permission response from user */
 #define FANOTIFY_PERM_EVENTS	(FAN_OPEN_PERM | FAN_ACCESS_PERM | \
-- 
2.33.0

[PATCH v9 29/31] ext4: Send notifications on error

[Gabriel Krisman Bertazi]

Send a FS_ERROR message via fsnotify to a userspace monitoring tool
whenever a ext4 error condition is triggered.  This follows the existing
error conditions in ext4, so it is hooked to the ext4_error* functions.

Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>

---
Changes since v8:
  - Report always report non-zero errno (Jan, Amir, Ted)
Changes since v6:
  - Report ext4_std_errors agains superblock (jan)
---
 fs/ext4/super.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index 88d5d274a868..1a766c68a55e 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -46,6 +46,7 @@
 #include <linux/part_stat.h>
 #include <linux/kthread.h>
 #include <linux/freezer.h>
+#include <linux/fsnotify.h>
 
 #include "ext4.h"
 #include "ext4_extents.h"	/* Needed for trace points definition */
@@ -759,6 +760,8 @@ void __ext4_error(struct super_block *sb, const char *function,
 		       sb->s_id, function, line, current->comm, &vaf);
 		va_end(args);
 	}
+	fsnotify_sb_error(sb, NULL, error ? error : EFSCORRUPTED);
+
 	ext4_handle_error(sb, force_ro, error, 0, block, function, line);
 }
 
@@ -789,6 +792,8 @@ void __ext4_error_inode(struct inode *inode, const char *function,
 			       current->comm, &vaf);
 		va_end(args);
 	}
+	fsnotify_sb_error(inode->i_sb, inode, error ? error : EFSCORRUPTED);
+
 	ext4_handle_error(inode->i_sb, false, error, inode->i_ino, block,
 			  function, line);
 }
@@ -827,6 +832,8 @@ void __ext4_error_file(struct file *file, const char *function,
 			       current->comm, path, &vaf);
 		va_end(args);
 	}
+	fsnotify_sb_error(inode->i_sb, inode, EFSCORRUPTED);
+
 	ext4_handle_error(inode->i_sb, false, EFSCORRUPTED, inode->i_ino, block,
 			  function, line);
 }
@@ -894,6 +901,7 @@ void __ext4_std_error(struct super_block *sb, const char *function,
 		printk(KERN_CRIT "EXT4-fs error (device %s) in %s:%d: %s\n",
 		       sb->s_id, function, line, errstr);
 	}
+	fsnotify_sb_error(sb, NULL, errno ? errno : EFSCORRUPTED);
 
 	ext4_handle_error(sb, false, -errno, 0, 0, function, line);
 }
-- 
2.33.0

Re: [PATCH v9 29/31] ext4: Send notifications on error

[Theodore Ts'o]

On Mon, Oct 25, 2021 at 04:27:44PM -0300, Gabriel Krisman Bertazi wrote:
> Send a FS_ERROR message via fsnotify to a userspace monitoring tool
> whenever a ext4 error condition is triggered.  This follows the existing
> error conditions in ext4, so it is hooked to the ext4_error* functions.
> 
> Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>

Acked-by: Theodore Ts'o <tytso@mit.edu>

Thanks!

					- Ted

Re: [PATCH v9 29/31] ext4: Send notifications on error

[Amir Goldstein]

On Mon, Oct 25, 2021 at 10:31 PM Gabriel Krisman Bertazi
<krisman@collabora.com> wrote:
>
> Send a FS_ERROR message via fsnotify to a userspace monitoring tool
> whenever a ext4 error condition is triggered.  This follows the existing
> error conditions in ext4, so it is hooked to the ext4_error* functions.
>
> Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>
>
Reviewed-by: Amir Goldstein <amir73il@gmail.com>

> ---
> Changes since v8:
>   - Report always report non-zero errno (Jan, Amir, Ted)
> Changes since v6:
>   - Report ext4_std_errors agains superblock (jan)
> ---
>  fs/ext4/super.c | 8 ++++++++
>  1 file changed, 8 insertions(+)
>
> diff --git a/fs/ext4/super.c b/fs/ext4/super.c
> index 88d5d274a868..1a766c68a55e 100644
> --- a/fs/ext4/super.c
> +++ b/fs/ext4/super.c
> @@ -46,6 +46,7 @@
>  #include <linux/part_stat.h>
>  #include <linux/kthread.h>
>  #include <linux/freezer.h>
> +#include <linux/fsnotify.h>
>
>  #include "ext4.h"
>  #include "ext4_extents.h"      /* Needed for trace points definition */
> @@ -759,6 +760,8 @@ void __ext4_error(struct super_block *sb, const char *function,
>                        sb->s_id, function, line, current->comm, &vaf);
>                 va_end(args);
>         }
> +       fsnotify_sb_error(sb, NULL, error ? error : EFSCORRUPTED);
> +
>         ext4_handle_error(sb, force_ro, error, 0, block, function, line);
>  }
>
> @@ -789,6 +792,8 @@ void __ext4_error_inode(struct inode *inode, const char *function,
>                                current->comm, &vaf);
>                 va_end(args);
>         }
> +       fsnotify_sb_error(inode->i_sb, inode, error ? error : EFSCORRUPTED);
> +
>         ext4_handle_error(inode->i_sb, false, error, inode->i_ino, block,
>                           function, line);
>  }
> @@ -827,6 +832,8 @@ void __ext4_error_file(struct file *file, const char *function,
>                                current->comm, path, &vaf);
>                 va_end(args);
>         }
> +       fsnotify_sb_error(inode->i_sb, inode, EFSCORRUPTED);
> +
>         ext4_handle_error(inode->i_sb, false, EFSCORRUPTED, inode->i_ino, block,
>                           function, line);
>  }
> @@ -894,6 +901,7 @@ void __ext4_std_error(struct super_block *sb, const char *function,
>                 printk(KERN_CRIT "EXT4-fs error (device %s) in %s:%d: %s\n",
>                        sb->s_id, function, line, errstr);
>         }
> +       fsnotify_sb_error(sb, NULL, errno ? errno : EFSCORRUPTED);
>
>         ext4_handle_error(sb, false, -errno, 0, 0, function, line);
>  }
> --
> 2.33.0
>

Re: [PATCH v9 29/31] ext4: Send notifications on error

[Jan Kara]

On Mon 25-10-21 16:27:44, Gabriel Krisman Bertazi wrote:
> Send a FS_ERROR message via fsnotify to a userspace monitoring tool
> whenever a ext4 error condition is triggered.  This follows the existing
> error conditions in ext4, so it is hooked to the ext4_error* functions.
> 
> Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>

Looks good. Feel free to add:

Reviewed-by: Jan Kara <jack@suse.cz>

								Honza

> ---
> Changes since v8:
>   - Report always report non-zero errno (Jan, Amir, Ted)
> Changes since v6:
>   - Report ext4_std_errors agains superblock (jan)
> ---
>  fs/ext4/super.c | 8 ++++++++
>  1 file changed, 8 insertions(+)
> 
> diff --git a/fs/ext4/super.c b/fs/ext4/super.c
> index 88d5d274a868..1a766c68a55e 100644
> --- a/fs/ext4/super.c
> +++ b/fs/ext4/super.c
> @@ -46,6 +46,7 @@
>  #include <linux/part_stat.h>
>  #include <linux/kthread.h>
>  #include <linux/freezer.h>
> +#include <linux/fsnotify.h>
>  
>  #include "ext4.h"
>  #include "ext4_extents.h"	/* Needed for trace points definition */
> @@ -759,6 +760,8 @@ void __ext4_error(struct super_block *sb, const char *function,
>  		       sb->s_id, function, line, current->comm, &vaf);
>  		va_end(args);
>  	}
> +	fsnotify_sb_error(sb, NULL, error ? error : EFSCORRUPTED);
> +
>  	ext4_handle_error(sb, force_ro, error, 0, block, function, line);
>  }
>  
> @@ -789,6 +792,8 @@ void __ext4_error_inode(struct inode *inode, const char *function,
>  			       current->comm, &vaf);
>  		va_end(args);
>  	}
> +	fsnotify_sb_error(inode->i_sb, inode, error ? error : EFSCORRUPTED);
> +
>  	ext4_handle_error(inode->i_sb, false, error, inode->i_ino, block,
>  			  function, line);
>  }
> @@ -827,6 +832,8 @@ void __ext4_error_file(struct file *file, const char *function,
>  			       current->comm, path, &vaf);
>  		va_end(args);
>  	}
> +	fsnotify_sb_error(inode->i_sb, inode, EFSCORRUPTED);
> +
>  	ext4_handle_error(inode->i_sb, false, EFSCORRUPTED, inode->i_ino, block,
>  			  function, line);
>  }
> @@ -894,6 +901,7 @@ void __ext4_std_error(struct super_block *sb, const char *function,
>  		printk(KERN_CRIT "EXT4-fs error (device %s) in %s:%d: %s\n",
>  		       sb->s_id, function, line, errstr);
>  	}
> +	fsnotify_sb_error(sb, NULL, errno ? errno : EFSCORRUPTED);
>  
>  	ext4_handle_error(sb, false, -errno, 0, 0, function, line);
>  }
> -- 
> 2.33.0
> 
-- 
Jan Kara <jack@suse.com>
SUSE Labs, CR

[PATCH v9 30/31] samples: Add fs error monitoring example

[Gabriel Krisman Bertazi]

Introduce an example of a FAN_FS_ERROR fanotify user to track filesystem
errors.

Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>

---
Changes since v4:
  - Protect file_handle defines with ifdef guards

Changes since v1:
  - minor fixes
---
 samples/Kconfig               |   9 +++
 samples/Makefile              |   1 +
 samples/fanotify/Makefile     |   5 ++
 samples/fanotify/fs-monitor.c | 142 ++++++++++++++++++++++++++++++++++
 4 files changed, 157 insertions(+)
 create mode 100644 samples/fanotify/Makefile
 create mode 100644 samples/fanotify/fs-monitor.c

diff --git a/samples/Kconfig b/samples/Kconfig
index b0503ef058d3..88353b8eac0b 100644
--- a/samples/Kconfig
+++ b/samples/Kconfig
@@ -120,6 +120,15 @@ config SAMPLE_CONNECTOR
 	  with it.
 	  See also Documentation/driver-api/connector.rst
 
+config SAMPLE_FANOTIFY_ERROR
+	bool "Build fanotify error monitoring sample"
+	depends on FANOTIFY
+	help
+	  When enabled, this builds an example code that uses the
+	  FAN_FS_ERROR fanotify mechanism to monitor filesystem
+	  errors.
+	  See also Documentation/admin-guide/filesystem-monitoring.rst.
+
 config SAMPLE_HIDRAW
 	bool "hidraw sample"
 	depends on CC_CAN_LINK && HEADERS_INSTALL
diff --git a/samples/Makefile b/samples/Makefile
index 087e0988ccc5..931a81847c48 100644
--- a/samples/Makefile
+++ b/samples/Makefile
@@ -5,6 +5,7 @@ subdir-$(CONFIG_SAMPLE_AUXDISPLAY)	+= auxdisplay
 subdir-$(CONFIG_SAMPLE_ANDROID_BINDERFS) += binderfs
 obj-$(CONFIG_SAMPLE_CONFIGFS)		+= configfs/
 obj-$(CONFIG_SAMPLE_CONNECTOR)		+= connector/
+obj-$(CONFIG_SAMPLE_FANOTIFY_ERROR)	+= fanotify/
 subdir-$(CONFIG_SAMPLE_HIDRAW)		+= hidraw
 obj-$(CONFIG_SAMPLE_HW_BREAKPOINT)	+= hw_breakpoint/
 obj-$(CONFIG_SAMPLE_KDB)		+= kdb/
diff --git a/samples/fanotify/Makefile b/samples/fanotify/Makefile
new file mode 100644
index 000000000000..e20db1bdde3b
--- /dev/null
+++ b/samples/fanotify/Makefile
@@ -0,0 +1,5 @@
+# SPDX-License-Identifier: GPL-2.0-only
+userprogs-always-y += fs-monitor
+
+userccflags += -I usr/include -Wall
+
diff --git a/samples/fanotify/fs-monitor.c b/samples/fanotify/fs-monitor.c
new file mode 100644
index 000000000000..a0e44cd31e6f
--- /dev/null
+++ b/samples/fanotify/fs-monitor.c
@@ -0,0 +1,142 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright 2021, Collabora Ltd.
+ */
+
+#define _GNU_SOURCE
+#include <errno.h>
+#include <err.h>
+#include <stdlib.h>
+#include <stdio.h>
+#include <fcntl.h>
+#include <sys/fanotify.h>
+#include <sys/types.h>
+#include <unistd.h>
+#include <sys/types.h>
+
+#ifndef FAN_FS_ERROR
+#define FAN_FS_ERROR		0x00008000
+#define FAN_EVENT_INFO_TYPE_ERROR	5
+
+struct fanotify_event_info_error {
+	struct fanotify_event_info_header hdr;
+	__s32 error;
+	__u32 error_count;
+};
+#endif
+
+#ifndef FILEID_INO32_GEN
+#define FILEID_INO32_GEN	1
+#endif
+
+#ifndef FILEID_INVALID
+#define	FILEID_INVALID		0xff
+#endif
+
+static void print_fh(struct file_handle *fh)
+{
+	int i;
+	uint32_t *h = (uint32_t *) fh->f_handle;
+
+	printf("\tfh: ");
+	for (i = 0; i < fh->handle_bytes; i++)
+		printf("%hhx", fh->f_handle[i]);
+	printf("\n");
+
+	printf("\tdecoded fh: ");
+	if (fh->handle_type == FILEID_INO32_GEN)
+		printf("inode=%u gen=%u\n", h[0], h[1]);
+	else if (fh->handle_type == FILEID_INVALID && !fh->handle_bytes)
+		printf("Type %d (Superblock error)\n", fh->handle_type);
+	else
+		printf("Type %d (Unknown)\n", fh->handle_type);
+
+}
+
+static void handle_notifications(char *buffer, int len)
+{
+	struct fanotify_event_metadata *event =
+		(struct fanotify_event_metadata *) buffer;
+	struct fanotify_event_info_header *info;
+	struct fanotify_event_info_error *err;
+	struct fanotify_event_info_fid *fid;
+	int off;
+
+	for (; FAN_EVENT_OK(event, len); event = FAN_EVENT_NEXT(event, len)) {
+
+		if (event->mask != FAN_FS_ERROR) {
+			printf("unexpected FAN MARK: %llx\n", event->mask);
+			goto next_event;
+		}
+
+		if (event->fd != FAN_NOFD) {
+			printf("Unexpected fd (!= FAN_NOFD)\n");
+			goto next_event;
+		}
+
+		printf("FAN_FS_ERROR (len=%d)\n", event->event_len);
+
+		for (off = sizeof(*event) ; off < event->event_len;
+		     off += info->len) {
+			info = (struct fanotify_event_info_header *)
+				((char *) event + off);
+
+			switch (info->info_type) {
+			case FAN_EVENT_INFO_TYPE_ERROR:
+				err = (struct fanotify_event_info_error *) info;
+
+				printf("\tGeneric Error Record: len=%d\n",
+				       err->hdr.len);
+				printf("\terror: %d\n", err->error);
+				printf("\terror_count: %d\n", err->error_count);
+				break;
+
+			case FAN_EVENT_INFO_TYPE_FID:
+				fid = (struct fanotify_event_info_fid *) info;
+
+				printf("\tfsid: %x%x\n",
+				       fid->fsid.val[0], fid->fsid.val[1]);
+				print_fh((struct file_handle *) &fid->handle);
+				break;
+
+			default:
+				printf("\tUnknown info type=%d len=%d:\n",
+				       info->info_type, info->len);
+			}
+		}
+next_event:
+		printf("---\n\n");
+	}
+}
+
+int main(int argc, char **argv)
+{
+	int fd;
+
+	char buffer[BUFSIZ];
+
+	if (argc < 2) {
+		printf("Missing path argument\n");
+		return 1;
+	}
+
+	fd = fanotify_init(FAN_CLASS_NOTIF|FAN_REPORT_FID, O_RDONLY);
+	if (fd < 0)
+		errx(1, "fanotify_init");
+
+	if (fanotify_mark(fd, FAN_MARK_ADD|FAN_MARK_FILESYSTEM,
+			  FAN_FS_ERROR, AT_FDCWD, argv[1])) {
+		errx(1, "fanotify_mark");
+	}
+
+	while (1) {
+		int n = read(fd, buffer, BUFSIZ);
+
+		if (n < 0)
+			errx(1, "read");
+
+		handle_notifications(buffer, n);
+	}
+
+	return 0;
+}
-- 
2.33.0

[PATCH v9 31/31] docs: Document the FAN_FS_ERROR event

[Gabriel Krisman Bertazi]

Document the FAN_FS_ERROR event for user administrators and user space
developers.

Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>

---
Changes Since v8:
  - Replace fs-error specific errno bits with generic errno. (Jan)
  - Explain event order guarantees and point to example parser (Jan)
Changes Since v7:
  - Update semantics
Changes Since v6:
  - English fixes (jan)
  - Proper document error field (jan)
Changes Since v4:
  - Update documentation about reporting non-file error.
Changes Since v3:
  - Move FAN_FS_ERROR notification into a subsection of the file.
Changes Since v2:
  - NTR
Changes since v1:
  - Drop references to location record
  - Explain that the inode field is optional
  - Explain we are reporting only the first error
---
 .../admin-guide/filesystem-monitoring.rst     | 74 +++++++++++++++++++
 Documentation/admin-guide/index.rst           |  1 +
 2 files changed, 75 insertions(+)
 create mode 100644 Documentation/admin-guide/filesystem-monitoring.rst

diff --git a/Documentation/admin-guide/filesystem-monitoring.rst b/Documentation/admin-guide/filesystem-monitoring.rst
new file mode 100644
index 000000000000..5a3c84e60095
--- /dev/null
+++ b/Documentation/admin-guide/filesystem-monitoring.rst
@@ -0,0 +1,74 @@
+.. SPDX-License-Identifier: GPL-2.0
+
+====================================
+File system Monitoring with fanotify
+====================================
+
+File system Error Reporting
+===========================
+
+Fanotify supports the FAN_FS_ERROR event type for file system-wide error
+reporting.  It is meant to be used by file system health monitoring
+daemons, which listen for these events and take actions (notify
+sysadmin, start recovery) when a file system problem is detected.
+
+By design, a FAN_FS_ERROR notification exposes sufficient information
+for a monitoring tool to know a problem in the file system has happened.
+It doesn't necessarily provide a user space application with semantics
+to verify an IO operation was successfully executed.  That is out of
+scope for this feature.  Instead, it is only meant as a framework for
+early file system problem detection and reporting recovery tools.
+
+When a file system operation fails, it is common for dozens of kernel
+errors to cascade after the initial failure, hiding the original failure
+log, which is usually the most useful debug data to troubleshoot the
+problem.  For this reason, FAN_FS_ERROR tries to report only the first
+error that occurred for a file system since the last notification, and
+it simply counts additional errors.  This ensures that the most
+important pieces of information are never lost.
+
+FAN_FS_ERROR requires the fanotify group to be setup with the
+FAN_REPORT_FID flag.
+
+At the time of this writing, the only file system that emits FAN_FS_ERROR
+notifications is Ext4.
+
+A FAN_FS_ERROR Notification has the following format::
+
+  [ Notification Metadata (Mandatory) ]
+  [ Generic Error Record  (Mandatory) ]
+  [ FID record            (Mandatory) ]
+
+The order of records is not guaranteed, and new records might be added
+in the future.  Therefore, applications must not rely on the order and
+must be prepared to skip over unknown records. Please refer to
+``samples/fanotify/fs-monitor.c`` for an example parser.
+
+Generic error record
+--------------------
+
+The generic error record provides enough information for a file system
+agnostic tool to learn about a problem in the file system, without
+providing any additional details about the problem.  This record is
+identified by ``struct fanotify_event_info_header.info_type`` being set
+to FAN_EVENT_INFO_TYPE_ERROR.
+
+  struct fanotify_event_info_error {
+	struct fanotify_event_info_header hdr;
+	__s32 error;
+	__u32 error_count;
+  };
+
+The `error` field identifies the type of error using errno values.
+`error_count` tracks the number of errors that occurred and were
+suppressed to preserve the original error information, since the last
+notification.
+
+FID record
+----------
+
+The FID record can be used to uniquely identify the inode that triggered
+the error through the combination of fsid and file handle.  A file system
+specific application can use that information to attempt a recovery
+procedure.  Errors that are not related to an inode are reported with an
+empty file handle of type FILEID_INVALID.
diff --git a/Documentation/admin-guide/index.rst b/Documentation/admin-guide/index.rst
index dc00afcabb95..1bedab498104 100644
--- a/Documentation/admin-guide/index.rst
+++ b/Documentation/admin-guide/index.rst
@@ -82,6 +82,7 @@ configure specific aspects of kernel behavior to your liking.
    edid
    efi-stub
    ext4
+   filesystem-monitoring
    nfs/index
    gpio/index
    highuid
-- 
2.33.0

Re: [PATCH v9 31/31] docs: Document the FAN_FS_ERROR event

[Jan Kara]

On Mon 25-10-21 16:27:46, Gabriel Krisman Bertazi wrote:
> Document the FAN_FS_ERROR event for user administrators and user space
> developers.
> 
> Reviewed-by: Amir Goldstein <amir73il@gmail.com>
> Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>

Looks good. Feel free to add:

Reviewed-by: Jan Kara <jack@suse.cz>

								Honza

> 
> ---
> Changes Since v8:
>   - Replace fs-error specific errno bits with generic errno. (Jan)
>   - Explain event order guarantees and point to example parser (Jan)
> Changes Since v7:
>   - Update semantics
> Changes Since v6:
>   - English fixes (jan)
>   - Proper document error field (jan)
> Changes Since v4:
>   - Update documentation about reporting non-file error.
> Changes Since v3:
>   - Move FAN_FS_ERROR notification into a subsection of the file.
> Changes Since v2:
>   - NTR
> Changes since v1:
>   - Drop references to location record
>   - Explain that the inode field is optional
>   - Explain we are reporting only the first error
> ---
>  .../admin-guide/filesystem-monitoring.rst     | 74 +++++++++++++++++++
>  Documentation/admin-guide/index.rst           |  1 +
>  2 files changed, 75 insertions(+)
>  create mode 100644 Documentation/admin-guide/filesystem-monitoring.rst
> 
> diff --git a/Documentation/admin-guide/filesystem-monitoring.rst b/Documentation/admin-guide/filesystem-monitoring.rst
> new file mode 100644
> index 000000000000..5a3c84e60095
> --- /dev/null
> +++ b/Documentation/admin-guide/filesystem-monitoring.rst
> @@ -0,0 +1,74 @@
> +.. SPDX-License-Identifier: GPL-2.0
> +
> +====================================
> +File system Monitoring with fanotify
> +====================================
> +
> +File system Error Reporting
> +===========================
> +
> +Fanotify supports the FAN_FS_ERROR event type for file system-wide error
> +reporting.  It is meant to be used by file system health monitoring
> +daemons, which listen for these events and take actions (notify
> +sysadmin, start recovery) when a file system problem is detected.
> +
> +By design, a FAN_FS_ERROR notification exposes sufficient information
> +for a monitoring tool to know a problem in the file system has happened.
> +It doesn't necessarily provide a user space application with semantics
> +to verify an IO operation was successfully executed.  That is out of
> +scope for this feature.  Instead, it is only meant as a framework for
> +early file system problem detection and reporting recovery tools.
> +
> +When a file system operation fails, it is common for dozens of kernel
> +errors to cascade after the initial failure, hiding the original failure
> +log, which is usually the most useful debug data to troubleshoot the
> +problem.  For this reason, FAN_FS_ERROR tries to report only the first
> +error that occurred for a file system since the last notification, and
> +it simply counts additional errors.  This ensures that the most
> +important pieces of information are never lost.
> +
> +FAN_FS_ERROR requires the fanotify group to be setup with the
> +FAN_REPORT_FID flag.
> +
> +At the time of this writing, the only file system that emits FAN_FS_ERROR
> +notifications is Ext4.
> +
> +A FAN_FS_ERROR Notification has the following format::
> +
> +  [ Notification Metadata (Mandatory) ]
> +  [ Generic Error Record  (Mandatory) ]
> +  [ FID record            (Mandatory) ]
> +
> +The order of records is not guaranteed, and new records might be added
> +in the future.  Therefore, applications must not rely on the order and
> +must be prepared to skip over unknown records. Please refer to
> +``samples/fanotify/fs-monitor.c`` for an example parser.
> +
> +Generic error record
> +--------------------
> +
> +The generic error record provides enough information for a file system
> +agnostic tool to learn about a problem in the file system, without
> +providing any additional details about the problem.  This record is
> +identified by ``struct fanotify_event_info_header.info_type`` being set
> +to FAN_EVENT_INFO_TYPE_ERROR.
> +
> +  struct fanotify_event_info_error {
> +	struct fanotify_event_info_header hdr;
> +	__s32 error;
> +	__u32 error_count;
> +  };
> +
> +The `error` field identifies the type of error using errno values.
> +`error_count` tracks the number of errors that occurred and were
> +suppressed to preserve the original error information, since the last
> +notification.
> +
> +FID record
> +----------
> +
> +The FID record can be used to uniquely identify the inode that triggered
> +the error through the combination of fsid and file handle.  A file system
> +specific application can use that information to attempt a recovery
> +procedure.  Errors that are not related to an inode are reported with an
> +empty file handle of type FILEID_INVALID.
> diff --git a/Documentation/admin-guide/index.rst b/Documentation/admin-guide/index.rst
> index dc00afcabb95..1bedab498104 100644
> --- a/Documentation/admin-guide/index.rst
> +++ b/Documentation/admin-guide/index.rst
> @@ -82,6 +82,7 @@ configure specific aspects of kernel behavior to your liking.
>     edid
>     efi-stub
>     ext4
> +   filesystem-monitoring
>     nfs/index
>     gpio/index
>     highuid
> -- 
> 2.33.0
> 
-- 
Jan Kara <jack@suse.com>
SUSE Labs, CR

Re: [PATCH v9 00/31] file system-wide error monitoring

[Amir Goldstein]

On Mon, Oct 25, 2021 at 10:27 PM Gabriel Krisman Bertazi
<krisman@collabora.com> wrote:
>
> Hi,
>
> This is the 9th version of this patch series.  Thank you, Amir, Jan and
> Ted, for the feedback in the previous versions.
>
> The main difference in this version is that the pool is no longer
> resizeable nor limited in number of marks, even though we only
> pre-allocate 32 slots.  In addition, ext4 was modified to always return
> non-zero errno, and the documentation was fixed accordingly (No longer
> suggests we return EXT4_ERR* values.
>
> I also droped the Reviewed-by tags from the ext4 patch, due to the
> changes above.
>
> Please let me know what you think.
>

All good on my end.
I've made a couple of minor comments that
could be addressed on commit if no other issues are found.

Thanks,
Amir.

Re: [PATCH v9 00/31] file system-wide error monitoring

[Jan Kara]

On Tue 26-10-21 12:12:38, Amir Goldstein wrote:
> On Mon, Oct 25, 2021 at 10:27 PM Gabriel Krisman Bertazi
> <krisman@collabora.com> wrote:
> >
> > Hi,
> >
> > This is the 9th version of this patch series.  Thank you, Amir, Jan and
> > Ted, for the feedback in the previous versions.
> >
> > The main difference in this version is that the pool is no longer
> > resizeable nor limited in number of marks, even though we only
> > pre-allocate 32 slots.  In addition, ext4 was modified to always return
> > non-zero errno, and the documentation was fixed accordingly (No longer
> > suggests we return EXT4_ERR* values.
> >
> > I also droped the Reviewed-by tags from the ext4 patch, due to the
> > changes above.
> >
> > Please let me know what you think.
> >
> 
> All good on my end.
> I've made a couple of minor comments that
> could be addressed on commit if no other issues are found.

All good on my end as well. I've applied all the minor updates, tested the
result and pushed it out to fsnotify branch of my tree. WRT to your new
FS_ERROR LTP tests, I've noticed that the testcases 1 and 3 from test
fanotify20 fail for me. After a bit of debugging this seems to be a bug in
ext4 where it calls ext4_abort() with EXT4_ERR_ESHUTDOWN instead of plain
ESHUTDOWN. Not sure if you have that fixed or how come the tests passed for
you. After fixing that ext4 bug everything passes for me.

								Honza
-- 
Jan Kara <jack@suse.com>
SUSE Labs, CR

Re: [PATCH v9 00/31] file system-wide error monitoring

[Amir Goldstein]

On Wed, Oct 27, 2021 at 2:22 PM Jan Kara <jack@suse.cz> wrote:
>
> On Tue 26-10-21 12:12:38, Amir Goldstein wrote:
> > On Mon, Oct 25, 2021 at 10:27 PM Gabriel Krisman Bertazi
> > <krisman@collabora.com> wrote:
> > >
> > > Hi,
> > >
> > > This is the 9th version of this patch series.  Thank you, Amir, Jan and
> > > Ted, for the feedback in the previous versions.
> > >
> > > The main difference in this version is that the pool is no longer
> > > resizeable nor limited in number of marks, even though we only
> > > pre-allocate 32 slots.  In addition, ext4 was modified to always return
> > > non-zero errno, and the documentation was fixed accordingly (No longer
> > > suggests we return EXT4_ERR* values.
> > >
> > > I also droped the Reviewed-by tags from the ext4 patch, due to the
> > > changes above.
> > >
> > > Please let me know what you think.
> > >
> >
> > All good on my end.
> > I've made a couple of minor comments that
> > could be addressed on commit if no other issues are found.
>
> All good on my end as well. I've applied all the minor updates, tested the
> result and pushed it out to fsnotify branch of my tree. WRT to your new
> FS_ERROR LTP tests, I've noticed that the testcases 1 and 3 from test
> fanotify20 fail for me. After a bit of debugging this seems to be a bug in
> ext4 where it calls ext4_abort() with EXT4_ERR_ESHUTDOWN instead of plain
> ESHUTDOWN. Not sure if you have that fixed or how come the tests passed for
> you. After fixing that ext4 bug everything passes for me.
>

Gabriel mentioned that bug in the cover letter of the LTP series :-)
https://lore.kernel.org/linux-ext4/20211026173302.84000-1-krisman@collabora.com/T/#u

Thanks,
Amir.

Re: [PATCH v9 00/31] file system-wide error monitoring

[Gabriel Krisman Bertazi]

Amir Goldstein <amir73il@gmail.com> writes:

> On Wed, Oct 27, 2021 at 2:22 PM Jan Kara <jack@suse.cz> wrote:
>>
>> On Tue 26-10-21 12:12:38, Amir Goldstein wrote:
>> > On Mon, Oct 25, 2021 at 10:27 PM Gabriel Krisman Bertazi
>> > <krisman@collabora.com> wrote:
>> > >
>> > > Hi,
>> > >
>> > > This is the 9th version of this patch series.  Thank you, Amir, Jan and
>> > > Ted, for the feedback in the previous versions.
>> > >
>> > > The main difference in this version is that the pool is no longer
>> > > resizeable nor limited in number of marks, even though we only
>> > > pre-allocate 32 slots.  In addition, ext4 was modified to always return
>> > > non-zero errno, and the documentation was fixed accordingly (No longer
>> > > suggests we return EXT4_ERR* values.
>> > >
>> > > I also droped the Reviewed-by tags from the ext4 patch, due to the
>> > > changes above.
>> > >
>> > > Please let me know what you think.
>> > >
>> >
>> > All good on my end.
>> > I've made a couple of minor comments that
>> > could be addressed on commit if no other issues are found.
>>
>> All good on my end as well. I've applied all the minor updates, tested the
>> result and pushed it out to fsnotify branch of my tree. WRT to your new
>> FS_ERROR LTP tests, I've noticed that the testcases 1 and 3 from test
>> fanotify20 fail for me. After a bit of debugging this seems to be a bug in
>> ext4 where it calls ext4_abort() with EXT4_ERR_ESHUTDOWN instead of plain
>> ESHUTDOWN. Not sure if you have that fixed or how come the tests passed for
>> you. After fixing that ext4 bug everything passes for me.
>>
>
> Gabriel mentioned that bug in the cover letter of the LTP series :-)
> https://lore.kernel.org/linux-ext4/20211026173302.84000-1-krisman@collabora.com/T/#u

Yes :)

Also, thank you both for the extensive review and ideas during the
development of this series.  It was really appreciated!

I'm sending out the new version for tests + man pages today.

-- 
Gabriel Krisman Bertazi

Re: [PATCH v9 00/31] file system-wide error monitoring

[Amir Goldstein]

> Also, thank you both for the extensive review and ideas during the
> development of this series.  It was really appreciated!
>

Thank you for your appreciated effort!
It was a wild journey through some interesting experiments, but
you survived it well ;-)

Would you be interested in pursuing FAN_WB_ERROR after a due rest
and after all the dust on FAN_FS_ERROR has settled?

FAN_WB_ERROR can use the same info record and same internal
error event struct as FAN_FS_ERROR.

A call to fsnotify_wb_error(sb, inode, err) can be placed inside
mapping_set_error() and maybe for other sporadic callers of errseq_set().

For wb error, we can consider storing a snapshot of errseq of the sb/inode
in the sb/inode mark and compute error_count from the errseq diff
instead of counting it when merging events. This will keep a more accurate
report even when error reports are dropped due to allocation failure or event
queue overflow.

I have a feeling that the Postgres project would find this
functionality useful (?).

Thanks,
Amir.

Re: [PATCH v9 00/31] file system-wide error monitoring

[Gabriel Krisman Bertazi]

Amir Goldstein <amir73il@gmail.com> writes:

>> Also, thank you both for the extensive review and ideas during the
>> development of this series.  It was really appreciated!
>>
>
> Thank you for your appreciated effort!
> It was a wild journey through some interesting experiments, but
> you survived it well ;-)
>
> Would you be interested in pursuing FAN_WB_ERROR after a due rest
> and after all the dust on FAN_FS_ERROR has settled?

I think it would make sense for me to continue working on it, yes.  But,
before that, I think I still have some support to add to FAN_FS_ERROR,
like a detailed, fs-specific, info record, and an error location info
record, which has a use-case in Google Cloud environments.  I have to
discuss priorities internally, but we (collabora) do have an interest in
supporting WB_ERROR too.

For the detailed error report, fanotify could have a new info record
that carries a structure sent out by the file system.  fanotify could
handle the lifetime of this object, by keeping a larger mempool, or
delegate its allocation/destruction to the filesystem.

Like I proposed in an earlier version of FAN_FS_ERROR, the format could
be as simple as:

struct fanotify_error_data_info {
   struct fanotify_event_info_header hdr;
   char data[];
}

I think xfs, at least, would be able to make good use of this record with
xfs_scrub, as the xfs maintainers mentioned.

-- 
Gabriel Krisman Bertazi

Re: [PATCH v9 00/31] file system-wide error monitoring

[Amir Goldstein]

On Sat, Oct 30, 2021 at 1:24 AM Gabriel Krisman Bertazi
<krisman@collabora.com> wrote:
>
> Amir Goldstein <amir73il@gmail.com> writes:
>
> >> Also, thank you both for the extensive review and ideas during the
> >> development of this series.  It was really appreciated!
> >>
> >
> > Thank you for your appreciated effort!
> > It was a wild journey through some interesting experiments, but
> > you survived it well ;-)
> >
> > Would you be interested in pursuing FAN_WB_ERROR after a due rest
> > and after all the dust on FAN_FS_ERROR has settled?
>
> I think it would make sense for me to continue working on it, yes.  But,
> before that, I think I still have some support to add to FAN_FS_ERROR,
> like a detailed, fs-specific, info record, and an error location info
> record, which has a use-case in Google Cloud environments.  I have to
> discuss priorities internally, but we (collabora) do have an interest in
> supporting WB_ERROR too.
>
> For the detailed error report, fanotify could have a new info record
> that carries a structure sent out by the file system.  fanotify could
> handle the lifetime of this object, by keeping a larger mempool, or
> delegate its allocation/destruction to the filesystem.
>

Before you try anything radical, please check the size of prospect
fs-specific data.
My hunch says that in most cases fs-specific data could fit cozy along side the
file handle within MAX_HANDLE_SZ and if this is true, then we do not need to
worry about extreme cases right now.
If there comes a time when we have a justified case of a filesystem that needs
to report much bigger fs-specific data, we can consider it then.
Until that time, we simply drop the over sized fs-specific data same as we do
if filesystem passed in a file handle larger than MAX_HANDLE_SZ.

> Like I proposed in an earlier version of FAN_FS_ERROR, the format could
> be as simple as:
>
> struct fanotify_error_data_info {
>    struct fanotify_event_info_header hdr;
>    char data[];
> }
>

We can add char data[] field to the end of struct fanotify_event_info_error.
It does not change the layout nor size of the structure and the info record
is variable size per definition anyway.

I know Jan didn't like this so much at the time and contemplated a
separate info record for filename, but eventually, fanotify_event_info_fid
also has an optional name following the unsigned char handle[].

> I think xfs, at least, would be able to make good use of this record with
> xfs_scrub, as the xfs maintainers mentioned.

I am not sure if that was the final conclusion.
xfs_scrub is proactive and should have no problem reporting its own findings,
but I have no objections to fs-specific details in FS_ERROR event.

Thanks,
Amir.