ftrace: Synchronize the initialization of ftrace_opsgrafted

There's some internal state that ftrace_ops needs to have set, but since
it can be declared outside of the ftrace.c code, it calls
ftrace_ops_init() on the ops in every global function. The issue is that
if two tasks call it on the same ops at the same time it is possible to
have the initialization of one corrupt the initialization of the other
call.

Create a ops_mutex to use to synchronize every initialization of the
ftrace_ops. The mutex is taken within checking the ftrace_ops flag that
states it was initializied but the flag is checked again after the mutex
has been taken. Checking first outside the mutex allows it to shortcut
having to take the mutex. But then the check needs to be done again after
the mute is taken in case of races.

Cc: stable@vger.kernel.org
Link: https://patch.msgid.link/20260902095501.6b59af20@gandalf.local.home
Fixes: f04f24fb7e48d ("ftrace, kprobes: Fix a deadlock on ftrace_regex_lock")
Reported-by: sashiko-bot@kernel.org
Close: https://lore.kernel.org/all/20260829025528.49A831F000E9@smtp.kernel.org/
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>

1 files changed, 56 insertions, 0 deletions

diff --git a/arch/alpha/lib/ev67-strcat.S b/arch/alpha/lib/ev67-strcat.S
new file mode 100644
index 000000000..f8c7305b1
--- /dev/null
+++ b/arch/alpha/lib/ev67-strcat.S
@@ -0,0 +1,56 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * arch/alpha/lib/ev67-strcat.S
+ * 21264 version contributed by Rick Gorton <rick.gorton@alpha-processor.com>
+ *
+ * Append a null-terminated string from SRC to DST.
+ *
+ * Much of the information about 21264 scheduling/coding comes from:
+ *	Compiler Writer's Guide for the Alpha 21264
+ *	abbreviated as 'CWG' in other comments here
+ *	ftp.digital.com/pub/Digital/info/semiconductor/literature/dsc-library.html
+ * Scheduling notation:
+ *	E	- either cluster
+ *	U	- upper subcluster; U0 - subcluster U0; U1 - subcluster U1
+ *	L	- lower subcluster; L0 - subcluster L0; L1 - subcluster L1
+ * Try not to change the actual algorithm if possible for consistency.
+ * Commentary: It seems bogus to walk the input string twice - once
+ * to determine the length, and then again while doing the copy.
+ * A significant (future) enhancement would be to only read the input
+ * string once.
+ */
+
+#include <linux/export.h>
+	.text
+
+	.align 4
+	.globl strcat
+	.ent strcat
+strcat:
+	.frame $30, 0, $26
+	.prologue 0
+
+	mov	$16, $0		# E : set up return value
+	/* Find the end of the string.  */
+	ldq_u   $1, 0($16)	# L : load first quadword (a0 may be misaligned)
+	lda     $2, -1		# E :
+	insqh   $2, $16, $2	# U :
+
+	andnot  $16, 7, $16	# E :
+	or      $2, $1, $1	# E :
+	cmpbge  $31, $1, $2	# E : bits set iff byte == 0
+	bne     $2, $found	# U :
+
+$loop:	ldq     $1, 8($16)	# L :
+	addq    $16, 8, $16	# E :
+	cmpbge  $31, $1, $2	# E :
+	beq     $2, $loop	# U :
+
+$found:	cttz	$2, $3		# U0 :
+	addq	$16, $3, $16	# E :
+	/* Now do the append.  */
+	mov	$26, $23	# E :
+	br	__stxcpy	# L0 :
+
+	.end strcat
+	EXPORT_SYMBOL(strcat)