Merge tag 'trace-v7.3-2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-tracegrafted

Pull tracing fixes from Steven Rostedt:

 - Fix error output of boot instance creation failure

   Currently if a boot instance creation fails, instead of printing out
   the name of the instance that failed, it prints "(null)". That is
   because it prints "cur_str" that had already been processed by
   strsep(). Print the saved name instead.

   While at it, print the error code of the failure.

 - Fix use-after-free for same named historgrams

   Histograms can be named so that they can be used in multiple events.
   But if the named histogram has a variable attached, the second event
   that uses the named histogram which duplicates it and needs to free
   the original after duplication leaves the old variable in place and
   still visible. If another histogram uses than variable, it will use
   the stale one which will try to reference the freed duplicate
   histogram and crash the kernel.

   Free the duplicate variables along with the duplicated histogram
   data.

 - Check return value of kthread_run() in event self test

   The events self tests uses a kthread for testing but does not check
   if it succeeded in creating a kthread. If the kthread creation were
   to fail, the code will still try to call kthread_stop() on the error
   returned.

 - Fix race between reading trace_pipe and updating subbuffer size

   If a user is reading the trace_pipe file at the same time they update
   the ring buffer sub-buffer size, can cause the trace_pipe read to
   read stale data. Add trace_access_lock() around updating the ring
   buffer sub-buffer size.

 - Fix eventfs_inode on failure path in creation of the events directory

   In the creation of the "events" directory, if after allocating the
   eventfs_inode a failure is detected, it calls cleanup_ei() which
   calls free_ei(). The free_ei() will test if eventfs_inode being freed
   has no children. It is a bug if it does. But on the failure case of
   the creation of the "events" directory, the children lists have not
   yet been initialized and the free will trigger a warning because
   list_empty() on an uninitialized list returns false.

   Move the initialization into init_ei() where it makes more sense and
   makes sure that a created eventfs_inode has its lists initialized
   upon creation.

 - Check return value of kthread_run() in ftrace direct sample code

   The sample code that shows how to use the ftrace direct calls does
   not test the return of kthread_run() to see if it succeeds. Return a
   failure if the kthread_run() doesn't succeed.

 - Clear user events state on fork in case of alloc failure

   On fork, the child gets a pointer to the parent's user events state.
   It makes a copy of it then updates the child's pointer to it. But if
   the allocation fails, the duplication function leaves the child with
   a pointer to its parent's descriptor. When the child cleans up its
   data, it will free the parent's descriptor while the parent is still
   using it.

   In the duplication function, set the child's user_event_mm to NULL
   before testing if the allocation succeeded, and when it exits it will
   not free the parent's descriptor.

 - Fix retry exhaustion in simple ring buffer reader swap

   simple_ring_buffer_swap_reader_page() starts with retry set to 8 and
   post-decrements it only after a failed link replacement. On the final
   attempt, a successful replacement leaves retry at zero, while a
   failed replacement leaves it at -1.

   But the check for success expects the retry value to be non-zero and
   exits with an error on zero. This is the opposite result. Fix it.

 - Fail nicely when the remote swap_reader_page() returns an error

   Currently, if the swap_reader_page() of a remote buffer fails, it
   triggers a WARN_ON_ONCE() and continues normally. Instead, have it
   exit with an error and a pr_warn() print instead of a full WARNING.

* tag 'trace-v7.3-2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace:
  ring-buffer: Stop remote reader update when page swap fails
  tracing: Fix retry exhaustion in simple ring buffer reader swap
  tracing/user_events: Clear copied tracing state before fork duplication
  samples/ftrace: Fix kthread_stop() on ERR_PTR in ftrace-direct-multi-modify
  samples/ftrace: Fix kthread_stop() on ERR_PTR in ftrace-direct-modify
  eventfs: Initialize ei->children and ei->list in init_ei()
  tracing: Fix use-after-free in trace_pipe read on sub-buffer order change
  tracing: Fix crash passing ERR_PTR to kthread_stop()
  tracing: Fix use-after-free with same-name named triggers
  tracing: Fix logged instance name on creation failure

1 files changed, 194 insertions, 0 deletions

diff --git a/fs/fserror.c b/fs/fserror.c
new file mode 100644
index 000000000..1e4d11fd9
--- /dev/null
+++ b/fs/fserror.c
@@ -0,0 +1,194 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Copyright (c) 2025 Oracle.  All Rights Reserved.
+ * Author: Darrick J. Wong <djwong@kernel.org>
+ */
+#include <linux/fs.h>
+#include <linux/fsnotify.h>
+#include <linux/mempool.h>
+#include <linux/fserror.h>
+
+#define FSERROR_DEFAULT_EVENT_POOL_SIZE		(32)
+
+static struct mempool fserror_events_pool;
+
+void fserror_mount(struct super_block *sb)
+{
+	/*
+	 * The pending error counter is biased by 1 so that we don't wake_var
+	 * until we're actually trying to unmount.
+	 */
+	refcount_set(&sb->s_pending_errors, 1);
+}
+
+void fserror_unmount(struct super_block *sb)
+{
+	/*
+	 * If we don't drop the pending error count to zero, then wait for it
+	 * to drop below 1, which means that the pending errors cleared and
+	 * hopefully we didn't saturate with 1 billion+ concurrent events.
+	 */
+	if (!refcount_dec_and_test(&sb->s_pending_errors))
+		wait_var_event(&sb->s_pending_errors,
+			       refcount_read(&sb->s_pending_errors) < 1);
+}
+
+static inline void fserror_pending_dec(struct super_block *sb)
+{
+	if (refcount_dec_and_test(&sb->s_pending_errors))
+		wake_up_var(&sb->s_pending_errors);
+}
+
+static inline void fserror_free_event(struct fserror_event *event)
+{
+	fserror_pending_dec(event->sb);
+	mempool_free(event, &fserror_events_pool);
+}
+
+static void fserror_worker(struct work_struct *work)
+{
+	struct fserror_event *event =
+			container_of(work, struct fserror_event, work);
+	struct super_block *sb = event->sb;
+
+	if (sb->s_flags & SB_ACTIVE) {
+		struct fs_error_report report = {
+			/* send positive error number to userspace */
+			.error = -event->error,
+			.inode = event->inode,
+			.sb = event->sb,
+		};
+
+		if (sb->s_op->report_error)
+			sb->s_op->report_error(event);
+
+		fsnotify(FS_ERROR, &report, FSNOTIFY_EVENT_ERROR, NULL, NULL,
+			 NULL, 0);
+	}
+
+	iput(event->inode);
+	fserror_free_event(event);
+}
+
+static inline struct fserror_event *fserror_alloc_event(struct super_block *sb,
+							gfp_t gfp_flags)
+{
+	struct fserror_event *event = NULL;
+
+	/*
+	 * If pending_errors already reached zero or is no longer active,
+	 * the superblock is being deactivated so there's no point in
+	 * continuing.
+	 *
+	 * The order of the check of s_pending_errors and SB_ACTIVE are
+	 * mandated by order of accesses in generic_shutdown_super and
+	 * fserror_unmount.  Barriers are implicitly provided by the refcount
+	 * manipulations in this function and fserror_unmount.
+	 */
+	if (!refcount_inc_not_zero(&sb->s_pending_errors))
+		return NULL;
+	if (!(sb->s_flags & SB_ACTIVE))
+		goto out_pending;
+
+	event = mempool_alloc(&fserror_events_pool, gfp_flags);
+	if (!event)
+		goto out_pending;
+
+	/* mempool_alloc doesn't support GFP_ZERO */
+	memset(event, 0, sizeof(*event));
+	event->sb = sb;
+	INIT_WORK(&event->work, fserror_worker);
+
+	return event;
+
+out_pending:
+	fserror_pending_dec(sb);
+	return NULL;
+}
+
+/**
+ * fserror_report - report a filesystem error of some kind
+ *
+ * @sb:		superblock of the filesystem
+ * @inode:	inode within that filesystem, if applicable
+ * @type:	type of error encountered
+ * @pos:	start of inode range affected, if applicable
+ * @len:	length of inode range affected, if applicable
+ * @error:	error number encountered, must be negative
+ * @gfp:	memory allocation flags for conveying the event to a worker,
+ *		since this function can be called from atomic contexts
+ *
+ * Report details of a filesystem error to the super_operations::report_error
+ * callback if present; and to fsnotify for distribution to userspace.  @sb,
+ * @gfp, @type, and @error must all be specified.  For file I/O errors, the
+ * @inode, @pos, and @len fields must also be specified.  For file metadata
+ * errors, @inode must be specified.  If @inode is not NULL, then @inode->i_sb
+ * must point to @sb.
+ *
+ * Reporting work is deferred to a workqueue to ensure that ->report_error is
+ * called from process context without any locks held.  An active reference to
+ * the inode is maintained until event handling is complete, and unmount will
+ * wait for queued events to drain.
+ */
+void fserror_report(struct super_block *sb, struct inode *inode,
+		    enum fserror_type type, loff_t pos, u64 len, int error,
+		    gfp_t gfp)
+{
+	struct fserror_event *event;
+
+	/* sb and inode must be from the same filesystem */
+	WARN_ON_ONCE(inode && inode->i_sb != sb);
+
+	/* error number must be negative */
+	WARN_ON_ONCE(error >= 0);
+
+	event = fserror_alloc_event(sb, gfp);
+	if (!event)
+		goto lost;
+
+	event->type = type;
+	event->pos = pos;
+	event->len = len;
+	event->error = error;
+
+	/*
+	 * Can't iput from non-sleeping context, so grabbing another reference
+	 * to the inode must be the last thing before submitting the event.
+	 */
+	if (inode) {
+		event->inode = igrab(inode);
+		if (!event->inode)
+			goto lost_event;
+	}
+
+	/*
+	 * Use schedule_work here even if we're already in process context so
+	 * that fsnotify and super_operations::report_error implementations are
+	 * guaranteed to run in process context without any locks held.  Since
+	 * errors are supposed to be rare, the overhead shouldn't kill us any
+	 * more than the failing device will.
+	 */
+	schedule_work(&event->work);
+	return;
+
+lost_event:
+	fserror_free_event(event);
+lost:
+	if (inode)
+		pr_err_ratelimited(
+ "%s: lost file I/O error report for ino %llu type %u pos 0x%llx len 0x%llx error %d",
+		       sb->s_id, inode->i_ino, type, pos, len, error);
+	else
+		pr_err_ratelimited(
+ "%s: lost filesystem error report for type %u error %d",
+		       sb->s_id, type, error);
+}
+EXPORT_SYMBOL_GPL(fserror_report);
+
+static int __init fserror_init(void)
+{
+	return mempool_init_kmalloc_pool(&fserror_events_pool,
+					 FSERROR_DEFAULT_EVENT_POOL_SIZE,
+					 sizeof(struct fserror_event));
+}
+fs_initcall(fserror_init);