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, 164 insertions, 0 deletions

diff --git a/tools/include/nolibc/arch-sh.h b/tools/include/nolibc/arch-sh.h
new file mode 100644
index 000000000..a32378fd6
--- /dev/null
+++ b/tools/include/nolibc/arch-sh.h
@@ -0,0 +1,164 @@
+/* SPDX-License-Identifier: LGPL-2.1 OR MIT */
+/*
+ * SuperH specific definitions for NOLIBC
+ * Copyright (C) 2025 Thomas Weißschuh <linux@weissschuh.net>
+ */
+
+#ifndef _NOLIBC_ARCH_SH_H
+#define _NOLIBC_ARCH_SH_H
+
+#include "compiler.h"
+#include "crt.h"
+
+/*
+ * Syscalls for SuperH:
+ *   - registers are 32bit wide
+ *   - syscall number is passed in r3
+ *   - arguments are in r4, r5, r6, r7, r0, r1, r2
+ *   - the system call is performed by calling trapa #31
+ *   - syscall return value is in r0
+ */
+
+#define __nolibc_syscall0(num)                                                \
+({                                                                            \
+	register long _num __asm__ ("r3") = (num);                            \
+	register long _ret __asm__ ("r0");                                    \
+									      \
+	__asm__ volatile (                                                    \
+		"trapa #31"                                                   \
+		: "=r"(_ret)                                                  \
+		: "r"(_num)                                                   \
+		: "memory", "cc"                                              \
+	);                                                                    \
+	_ret;                                                                 \
+})
+
+#define __nolibc_syscall1(num, arg1)                                          \
+({                                                                            \
+	register long _num __asm__ ("r3") = (num);                            \
+	register long _ret __asm__ ("r0");                                    \
+	register long _arg1 __asm__ ("r4") = (long)(arg1);                    \
+									      \
+	__asm__ volatile (                                                    \
+		"trapa #31"                                                   \
+		: "=r"(_ret)                                                  \
+		: "r"(_num), "r"(_arg1)                                       \
+		: "memory", "cc"                                              \
+	);                                                                    \
+	_ret;                                                                 \
+})
+
+#define __nolibc_syscall2(num, arg1, arg2)                                    \
+({                                                                            \
+	register long _num __asm__ ("r3") = (num);                            \
+	register long _ret __asm__ ("r0");                                    \
+	register long _arg1 __asm__ ("r4") = (long)(arg1);                    \
+	register long _arg2 __asm__ ("r5") = (long)(arg2);                    \
+									      \
+	__asm__ volatile (                                                    \
+		"trapa #31"                                                   \
+		: "=r"(_ret)                                                  \
+		: "r"(_num), "r"(_arg1), "r"(_arg2)                           \
+		: "memory", "cc"                                              \
+	);                                                                    \
+	_ret;                                                                 \
+})
+
+#define __nolibc_syscall3(num, arg1, arg2, arg3)                              \
+({                                                                            \
+	register long _num __asm__ ("r3") = (num);                            \
+	register long _ret __asm__ ("r0");                                    \
+	register long _arg1 __asm__ ("r4") = (long)(arg1);                    \
+	register long _arg2 __asm__ ("r5") = (long)(arg2);                    \
+	register long _arg3 __asm__ ("r6") = (long)(arg3);                    \
+									      \
+	__asm__ volatile (                                                    \
+		"trapa #31"                                                   \
+		: "=r"(_ret)                                                  \
+		: "r"(_num), "r"(_arg1), "r"(_arg2), "r"(_arg3)               \
+		: "memory", "cc"                                              \
+	);                                                                    \
+	_ret;                                                                 \
+})
+
+#define __nolibc_syscall4(num, arg1, arg2, arg3, arg4)                        \
+({                                                                            \
+	register long _num __asm__ ("r3") = (num);                            \
+	register long _ret __asm__ ("r0");                                    \
+	register long _arg1 __asm__ ("r4") = (long)(arg1);                    \
+	register long _arg2 __asm__ ("r5") = (long)(arg2);                    \
+	register long _arg3 __asm__ ("r6") = (long)(arg3);                    \
+	register long _arg4 __asm__ ("r7") = (long)(arg4);                    \
+									      \
+	__asm__ volatile (                                                    \
+		"trapa #31"                                                   \
+		: "=r"(_ret)                                                  \
+		: "r"(_num), "r"(_arg1), "r"(_arg2), "r"(_arg3), "r"(_arg4)   \
+		: "memory", "cc"                                              \
+	);                                                                    \
+	_ret;                                                                 \
+})
+
+#define __nolibc_syscall5(num, arg1, arg2, arg3, arg4, arg5)                  \
+({                                                                            \
+	register long _num __asm__ ("r3") = (num);                            \
+	register long _ret __asm__ ("r0");                                    \
+	register long _arg1 __asm__ ("r4") = (long)(arg1);                    \
+	register long _arg2 __asm__ ("r5") = (long)(arg2);                    \
+	register long _arg3 __asm__ ("r6") = (long)(arg3);                    \
+	register long _arg4 __asm__ ("r7") = (long)(arg4);                    \
+	register long _arg5 __asm__ ("r0") = (long)(arg5);                    \
+									      \
+	__asm__ volatile (                                                    \
+		"trapa #31"                                                   \
+		: "=r"(_ret)                                                  \
+		: "r"(_num), "r"(_arg1), "r"(_arg2), "r"(_arg3), "r"(_arg4),  \
+		  "r"(_arg5)                                                  \
+		: "memory", "cc"                                              \
+	);                                                                    \
+	_ret;                                                                 \
+})
+
+#define __nolibc_syscall6(num, arg1, arg2, arg3, arg4, arg5, arg6)            \
+({                                                                            \
+	register long _num __asm__ ("r3") = (num);                            \
+	register long _ret __asm__ ("r0");                                    \
+	register long _arg1 __asm__ ("r4") = (long)(arg1);                    \
+	register long _arg2 __asm__ ("r5") = (long)(arg2);                    \
+	register long _arg3 __asm__ ("r6") = (long)(arg3);                    \
+	register long _arg4 __asm__ ("r7") = (long)(arg4);                    \
+	register long _arg5 __asm__ ("r0") = (long)(arg5);                    \
+	register long _arg6 __asm__ ("r1") = (long)(arg6);                    \
+									      \
+	__asm__ volatile (                                                    \
+		"trapa #31"                                                   \
+		: "=r"(_ret)                                                  \
+		: "r"(_num), "r"(_arg1), "r"(_arg2), "r"(_arg3), "r"(_arg4),  \
+		  "r"(_arg5), "r"(_arg6)                                      \
+		: "memory", "cc"                                              \
+	);                                                                    \
+	_ret;                                                                 \
+})
+
+#ifndef NOLIBC_NO_RUNTIME
+/* startup code */
+void _start_wrapper(void);
+void __attribute__((weak, noreturn)) __nolibc_entrypoint __nolibc_no_stack_protector _start_wrapper(void)
+{
+	__asm__ volatile (
+		".global _start\n"           /* The C function will have a prologue,         */
+		".type _start, @function\n"  /* corrupting "sp"                              */
+		".weak _start\n"
+		"_start:\n"
+
+		"mov sp, r4\n"               /* save argc pointer to r4, as arg1 of _start_c */
+		"bsr _start_c\n"             /* transfer to c runtime                        */
+		"nop\n"                      /* delay slot                                   */
+
+		".size _start, .-_start\n"
+	);
+	__nolibc_entrypoint_epilogue();
+}
+#endif /* NOLIBC_NO_RUNTIME */
+
+#endif /* _NOLIBC_ARCH_SH_H */