x86/kprobes: Fix crash when probing CS CALL instructionsgrafted

When using eBPF to probe CS CALL instructions within a function,
a crash can be triggered.

The eBPF tool probes offset 257 of the __hrtimer_run_queues()
function:

<__hrtimer_run_queues+249>:  nopl   0x0(%rax,%rax,1)
<__hrtimer_run_queues+254>:  mov    %r14,%rdi
<__hrtimer_run_queues+257>:  cs call <__x86_indirect_thunk_r12>
<__hrtimer_run_queues+263>:  mov    %eax,%r12d
<__hrtimer_run_queues+266>:  xchg   %ax,%ax
<__hrtimer_run_queues+268>:  mov    %r13,%rdi

Which triggers this crash:

  BUG: unable to handle page fault for address: 00000000000f41c9
  #PF: supervisor write access in kernel mode
  #PF: error_code(0x0002) - not-present page
  PGD 0 P4D 0
  Oops: 0002 [#1] SMP NOPTI
  CPU: 1 PID: 0 Comm: swapper/1 Kdump: loaded Tainted: P
  RIP: 0010:__hrtimer_run_queues+0x106/0x230

Note that __hrtimer_run_queues+0x106 is __hrtimer_run_queues+262, which is
at the 6th byte of the above CS CALL instruction. Since the CS CALL
instruction occupies 6 bytes, the exception occurred in the middle of that
call instruction.

The root cause is that when using eBPF tools to probe in the middle of a
function, a kprobe with INT3 is used as the underlying implementation.

During single-step emulation of the original CALL instruction,
int3_emulate_call() assumes that the probed CALL instruction is 5 bytes
long. However, the actual CS-prefixed CALL instruction occupies 6 bytes,
so it constructs an incorrect exception return address. When the CPU
returns from the kprobe handler, the next instruction to be executed is at
the address of the last byte of that CS CALL instruction. Coincidentally,
starting from that address, the CPU fetches and decodes a completely
different instruction, which ultimately triggers a kernel crash.

Fix the issue by using the actual instruction length obtained from
the instruction decoder when constructing the exception return
address, rather than relying on the hardcoded CALL_INSN_SIZE macro.

[ mingo: Refined the changelog ]

Fixes: 6256e668b7af ("x86/kprobes: Use int3 instead of debug trap for single-step")
Suggested-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Signed-off-by: Jinke Han <jinkehan@didiglobal.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Acked-by: Yafang Shao <laoar.shao@gmail.com>
Acked-by: Borislav Petkov <bp@alien8.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://patch.msgid.link/20260908073742.GA10517@didi-ThinkCentre-M920t-N000

1 files changed, 582 insertions, 0 deletions

diff --git a/drivers/tty/tty_jobctrl.c b/drivers/tty/tty_jobctrl.c
new file mode 100644
index 000000000..37929fb56
--- /dev/null
+++ b/drivers/tty/tty_jobctrl.c
@@ -0,0 +1,582 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ *  Copyright (C) 1991, 1992  Linus Torvalds
+ */
+
+#include <linux/types.h>
+#include <linux/errno.h>
+#include <linux/signal.h>
+#include <linux/sched/signal.h>
+#include <linux/sched/task.h>
+#include <linux/tty.h>
+#include <linux/fcntl.h>
+#include <linux/uaccess.h>
+#include "tty.h"
+
+static int is_ignored(int sig)
+{
+	return (sigismember(&current->blocked, sig) ||
+		current->sighand->action[sig-1].sa.sa_handler == SIG_IGN);
+}
+
+/**
+ *	__tty_check_change	-	check for POSIX terminal changes
+ *	@tty: tty to check
+ *	@sig: signal to send
+ *
+ *	If we try to write to, or set the state of, a terminal and we're
+ *	not in the foreground, send a SIGTTOU.  If the signal is blocked or
+ *	ignored, go ahead and perform the operation.  (POSIX 7.2)
+ *
+ *	Locking: ctrl.lock
+ */
+int __tty_check_change(struct tty_struct *tty, int sig)
+{
+	unsigned long flags;
+	struct pid *pgrp, *tty_pgrp;
+	int ret = 0;
+
+	if (current->signal->tty != tty)
+		return 0;
+
+	rcu_read_lock();
+	pgrp = task_pgrp(current);
+
+	spin_lock_irqsave(&tty->ctrl.lock, flags);
+	tty_pgrp = tty->ctrl.pgrp;
+	spin_unlock_irqrestore(&tty->ctrl.lock, flags);
+
+	if (tty_pgrp && pgrp != tty_pgrp) {
+		if (is_ignored(sig)) {
+			if (sig == SIGTTIN)
+				ret = -EIO;
+		} else if (is_current_pgrp_orphaned())
+			ret = -EIO;
+		else {
+			kill_pgrp(pgrp, sig, 1);
+			set_thread_flag(TIF_SIGPENDING);
+			ret = -ERESTARTSYS;
+		}
+	}
+	rcu_read_unlock();
+
+	if (!tty_pgrp)
+		tty_warn(tty, "sig=%d, tty->pgrp == NULL!\n", sig);
+
+	return ret;
+}
+
+int tty_check_change(struct tty_struct *tty)
+{
+	return __tty_check_change(tty, SIGTTOU);
+}
+EXPORT_SYMBOL(tty_check_change);
+
+void proc_clear_tty(struct task_struct *p)
+{
+	unsigned long flags;
+	struct tty_struct *tty;
+
+	spin_lock_irqsave(&p->sighand->siglock, flags);
+	tty = p->signal->tty;
+	p->signal->tty = NULL;
+	spin_unlock_irqrestore(&p->sighand->siglock, flags);
+	tty_kref_put(tty);
+}
+
+/**
+ * __proc_set_tty -  set the controlling terminal
+ *	@tty: tty structure
+ *
+ * Only callable by the session leader and only if it does not already have
+ * a controlling terminal.
+ *
+ * Caller must hold:  tty_lock()
+ *		      a readlock on tasklist_lock
+ *		      sighand lock
+ */
+static void __proc_set_tty(struct tty_struct *tty)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&tty->ctrl.lock, flags);
+	/*
+	 * The session and fg pgrp references will be non-NULL if
+	 * tiocsctty() is stealing the controlling tty
+	 */
+	put_pid(tty->ctrl.session);
+	put_pid(tty->ctrl.pgrp);
+	tty->ctrl.pgrp = get_pid(task_pgrp(current));
+	tty->ctrl.session = get_pid(task_session(current));
+	spin_unlock_irqrestore(&tty->ctrl.lock, flags);
+	if (current->signal->tty) {
+		tty_debug(tty, "current tty %s not NULL!!\n",
+			  current->signal->tty->name);
+		tty_kref_put(current->signal->tty);
+	}
+	put_pid(current->signal->tty_old_pgrp);
+	current->signal->tty = tty_kref_get(tty);
+	current->signal->tty_old_pgrp = NULL;
+}
+
+static void proc_set_tty(struct tty_struct *tty)
+{
+	spin_lock_irq(&current->sighand->siglock);
+	__proc_set_tty(tty);
+	spin_unlock_irq(&current->sighand->siglock);
+}
+
+/*
+ * Called by tty_open() to set the controlling tty if applicable.
+ */
+void tty_open_proc_set_tty(struct file *filp, struct tty_struct *tty)
+{
+	read_lock(&tasklist_lock);
+	spin_lock_irq(&current->sighand->siglock);
+	if (current->signal->leader &&
+	    !current->signal->tty &&
+	    tty->ctrl.session == NULL) {
+		/*
+		 * Don't let a process that only has write access to the tty
+		 * obtain the privileges associated with having a tty as
+		 * controlling terminal (being able to reopen it with full
+		 * access through /dev/tty, being able to perform pushback).
+		 * Many distributions set the group of all ttys to "tty" and
+		 * grant write-only access to all terminals for setgid tty
+		 * binaries, which should not imply full privileges on all ttys.
+		 *
+		 * This could theoretically break old code that performs open()
+		 * on a write-only file descriptor. In that case, it might be
+		 * necessary to also permit this if
+		 * inode_permission(inode, MAY_READ) == 0.
+		 */
+		if (filp->f_mode & FMODE_READ)
+			__proc_set_tty(tty);
+	}
+	spin_unlock_irq(&current->sighand->siglock);
+	read_unlock(&tasklist_lock);
+}
+
+struct tty_struct *get_current_tty(void)
+{
+	struct tty_struct *tty;
+	unsigned long flags;
+
+	spin_lock_irqsave(&current->sighand->siglock, flags);
+	tty = tty_kref_get(current->signal->tty);
+	spin_unlock_irqrestore(&current->sighand->siglock, flags);
+	return tty;
+}
+EXPORT_SYMBOL_GPL(get_current_tty);
+
+/*
+ * Called from tty_release().
+ */
+void session_clear_tty(struct pid *session)
+{
+	struct task_struct *p;
+
+	do_each_pid_task(session, PIDTYPE_SID, p) {
+		proc_clear_tty(p);
+	} while_each_pid_task(session, PIDTYPE_SID, p);
+}
+
+/**
+ *	tty_signal_session_leader	- sends SIGHUP to session leader
+ *	@tty: controlling tty
+ *	@exit_session: if non-zero, signal all foreground group processes
+ *
+ *	Send SIGHUP and SIGCONT to the session leader and its process group.
+ *	Optionally, signal all processes in the foreground process group.
+ *
+ *	Returns the number of processes in the session with this tty
+ *	as their controlling terminal. This value is used to drop
+ *	tty references for those processes.
+ */
+int tty_signal_session_leader(struct tty_struct *tty, int exit_session)
+{
+	struct task_struct *p;
+	int refs = 0;
+	struct pid *tty_pgrp = NULL;
+
+	read_lock(&tasklist_lock);
+	if (tty->ctrl.session) {
+		do_each_pid_task(tty->ctrl.session, PIDTYPE_SID, p) {
+			spin_lock_irq(&p->sighand->siglock);
+			if (p->signal->tty == tty) {
+				p->signal->tty = NULL;
+				/*
+				 * We defer the dereferences outside of
+				 * the tasklist lock.
+				 */
+				refs++;
+			}
+			if (!p->signal->leader) {
+				spin_unlock_irq(&p->sighand->siglock);
+				continue;
+			}
+			send_signal_locked(SIGHUP, SEND_SIG_PRIV, p, PIDTYPE_TGID);
+			send_signal_locked(SIGCONT, SEND_SIG_PRIV, p, PIDTYPE_TGID);
+			put_pid(p->signal->tty_old_pgrp);  /* A noop */
+			spin_lock(&tty->ctrl.lock);
+			tty_pgrp = get_pid(tty->ctrl.pgrp);
+			if (tty->ctrl.pgrp)
+				p->signal->tty_old_pgrp =
+					get_pid(tty->ctrl.pgrp);
+			spin_unlock(&tty->ctrl.lock);
+			spin_unlock_irq(&p->sighand->siglock);
+		} while_each_pid_task(tty->ctrl.session, PIDTYPE_SID, p);
+	}
+	read_unlock(&tasklist_lock);
+
+	if (tty_pgrp) {
+		if (exit_session)
+			kill_pgrp(tty_pgrp, SIGHUP, exit_session);
+		put_pid(tty_pgrp);
+	}
+
+	return refs;
+}
+
+/**
+ *	disassociate_ctty	-	disconnect controlling tty
+ *	@on_exit: true if exiting so need to "hang up" the session
+ *
+ *	This function is typically called only by the session leader, when
+ *	it wants to disassociate itself from its controlling tty.
+ *
+ *	It performs the following functions:
+ *	(1)  Sends a SIGHUP and SIGCONT to the foreground process group
+ *	(2)  Clears the tty from being controlling the session
+ *	(3)  Clears the controlling tty for all processes in the
+ *		session group.
+ *
+ *	The argument on_exit is set to 1 if called when a process is
+ *	exiting; it is 0 if called by the ioctl TIOCNOTTY.
+ *
+ *	Locking:
+ *		BTM is taken for hysterical raisons, and held when
+ *		  called from no_tty().
+ *		  tty_mutex is taken to protect tty
+ *		  ->siglock is taken to protect ->signal/->sighand
+ *		  tasklist_lock is taken to walk process list for sessions
+ *		    ->siglock is taken to protect ->signal/->sighand
+ */
+void disassociate_ctty(int on_exit)
+{
+	struct tty_struct *tty;
+
+	if (!current->signal->leader)
+		return;
+
+	tty = get_current_tty();
+	if (tty) {
+		if (on_exit && tty->driver->type != TTY_DRIVER_TYPE_PTY) {
+			tty_vhangup_session(tty);
+		} else {
+			struct pid *tty_pgrp = tty_get_pgrp(tty);
+
+			if (tty_pgrp) {
+				kill_pgrp(tty_pgrp, SIGHUP, on_exit);
+				if (!on_exit)
+					kill_pgrp(tty_pgrp, SIGCONT, on_exit);
+				put_pid(tty_pgrp);
+			}
+		}
+		tty_kref_put(tty);
+
+	} else if (on_exit) {
+		struct pid *old_pgrp;
+
+		spin_lock_irq(&current->sighand->siglock);
+		old_pgrp = current->signal->tty_old_pgrp;
+		current->signal->tty_old_pgrp = NULL;
+		spin_unlock_irq(&current->sighand->siglock);
+		if (old_pgrp) {
+			kill_pgrp(old_pgrp, SIGHUP, on_exit);
+			kill_pgrp(old_pgrp, SIGCONT, on_exit);
+			put_pid(old_pgrp);
+		}
+		return;
+	}
+
+	tty = get_current_tty();
+	if (tty) {
+		unsigned long flags;
+
+		tty_lock(tty);
+		spin_lock_irqsave(&tty->ctrl.lock, flags);
+		put_pid(tty->ctrl.session);
+		put_pid(tty->ctrl.pgrp);
+		tty->ctrl.session = NULL;
+		tty->ctrl.pgrp = NULL;
+		spin_unlock_irqrestore(&tty->ctrl.lock, flags);
+		tty_unlock(tty);
+		tty_kref_put(tty);
+	}
+
+	/* If tty->ctrl.pgrp is not NULL, it may be assigned to
+	 * current->signal->tty_old_pgrp in a race condition, and
+	 * cause pid memleak. Release current->signal->tty_old_pgrp
+	 * after tty->ctrl.pgrp set to NULL.
+	 */
+	spin_lock_irq(&current->sighand->siglock);
+	put_pid(current->signal->tty_old_pgrp);
+	current->signal->tty_old_pgrp = NULL;
+	spin_unlock_irq(&current->sighand->siglock);
+
+	/* Now clear signal->tty under the lock */
+	read_lock(&tasklist_lock);
+	session_clear_tty(task_session(current));
+	read_unlock(&tasklist_lock);
+}
+
+/*
+ *
+ *	no_tty	- Ensure the current process does not have a controlling tty
+ */
+void no_tty(void)
+{
+	/*
+	 * FIXME: Review locking here. The tty_lock never covered any race
+	 * between a new association and proc_clear_tty but possibly we need
+	 * to protect against this anyway.
+	 */
+	struct task_struct *tsk = current;
+
+	disassociate_ctty(0);
+	proc_clear_tty(tsk);
+}
+
+/**
+ *	tiocsctty	-	set controlling tty
+ *	@tty: tty structure
+ *	@file: file structure used to check permissions
+ *	@arg: user argument
+ *
+ *	This ioctl is used to manage job control. It permits a session
+ *	leader to set this tty as the controlling tty for the session.
+ *
+ *	Locking:
+ *		Takes tty_lock() to serialize proc_set_tty() for this tty
+ *		Takes tasklist_lock internally to walk sessions
+ *		Takes ->siglock() when updating signal->tty
+ */
+static int tiocsctty(struct tty_struct *tty, struct file *file, int arg)
+{
+	int ret = 0;
+
+	tty_lock(tty);
+	read_lock(&tasklist_lock);
+
+	if (current->signal->leader &&
+			task_session(current) == tty->ctrl.session)
+		goto unlock;
+
+	/*
+	 * The process must be a session leader and
+	 * not have a controlling tty already.
+	 */
+	if (!current->signal->leader || current->signal->tty) {
+		ret = -EPERM;
+		goto unlock;
+	}
+
+	if (tty->ctrl.session) {
+		/*
+		 * This tty is already the controlling
+		 * tty for another session group!
+		 */
+		if (arg == 1 && capable(CAP_SYS_ADMIN)) {
+			/*
+			 * Steal it away
+			 */
+			session_clear_tty(tty->ctrl.session);
+		} else {
+			ret = -EPERM;
+			goto unlock;
+		}
+	}
+
+	/* See the comment in tty_open_proc_set_tty(). */
+	if ((file->f_mode & FMODE_READ) == 0 && !capable(CAP_SYS_ADMIN)) {
+		ret = -EPERM;
+		goto unlock;
+	}
+
+	proc_set_tty(tty);
+unlock:
+	read_unlock(&tasklist_lock);
+	tty_unlock(tty);
+	return ret;
+}
+
+/**
+ *	tty_get_pgrp	-	return a ref counted pgrp pid
+ *	@tty: tty to read
+ *
+ *	Returns a refcounted instance of the pid struct for the process
+ *	group controlling the tty.
+ */
+struct pid *tty_get_pgrp(struct tty_struct *tty)
+{
+	unsigned long flags;
+	struct pid *pgrp;
+
+	spin_lock_irqsave(&tty->ctrl.lock, flags);
+	pgrp = get_pid(tty->ctrl.pgrp);
+	spin_unlock_irqrestore(&tty->ctrl.lock, flags);
+
+	return pgrp;
+}
+EXPORT_SYMBOL_GPL(tty_get_pgrp);
+
+/*
+ * This checks not only the pgrp, but falls back on the pid if no
+ * satisfactory pgrp is found. I dunno - gdb doesn't work correctly
+ * without this...
+ *
+ * The caller must hold rcu lock or the tasklist lock.
+ */
+static struct pid *session_of_pgrp(struct pid *pgrp)
+{
+	struct task_struct *p;
+	struct pid *sid = NULL;
+
+	p = pid_task(pgrp, PIDTYPE_PGID);
+	if (p == NULL)
+		p = pid_task(pgrp, PIDTYPE_PID);
+	if (p != NULL)
+		sid = task_session(p);
+
+	return sid;
+}
+
+/**
+ *	tiocgpgrp		-	get process group
+ *	@tty: tty passed by user
+ *	@real_tty: tty side of the tty passed by the user if a pty else the tty
+ *	@p: returned pid
+ *
+ *	Obtain the process group of the tty. If there is no process group
+ *	return an error.
+ *
+ *	Locking: none. Reference to current->signal->tty is safe.
+ */
+static int tiocgpgrp(struct tty_struct *tty, struct tty_struct *real_tty, pid_t __user *p)
+{
+	struct pid *pid;
+	int ret;
+	/*
+	 * (tty == real_tty) is a cheap way of
+	 * testing if the tty is NOT a master pty.
+	 */
+	if (tty == real_tty && current->signal->tty != real_tty)
+		return -ENOTTY;
+	pid = tty_get_pgrp(real_tty);
+	ret =  put_user(pid_vnr(pid), p);
+	put_pid(pid);
+	return ret;
+}
+
+/**
+ *	tiocspgrp		-	attempt to set process group
+ *	@tty: tty passed by user
+ *	@real_tty: tty side device matching tty passed by user
+ *	@p: pid pointer
+ *
+ *	Set the process group of the tty to the session passed. Only
+ *	permitted where the tty session is our session.
+ *
+ *	Locking: RCU, ctrl lock
+ */
+static int tiocspgrp(struct tty_struct *tty, struct tty_struct *real_tty, pid_t __user *p)
+{
+	struct pid *pgrp;
+	pid_t pgrp_nr;
+	int retval = tty_check_change(real_tty);
+
+	if (retval == -EIO)
+		return -ENOTTY;
+	if (retval)
+		return retval;
+
+	if (get_user(pgrp_nr, p))
+		return -EFAULT;
+	if (pgrp_nr < 0)
+		return -EINVAL;
+
+	guard(spinlock_irq)(&real_tty->ctrl.lock);
+	if (!current->signal->tty ||
+	    (current->signal->tty != real_tty) ||
+	    (real_tty->ctrl.session != task_session(current)))
+		return -ENOTTY;
+
+	guard(rcu)();
+	pgrp = find_vpid(pgrp_nr);
+	if (!pgrp)
+		return -ESRCH;
+	if (session_of_pgrp(pgrp) != task_session(current))
+		return -EPERM;
+
+	put_pid(real_tty->ctrl.pgrp);
+	real_tty->ctrl.pgrp = get_pid(pgrp);
+
+	return 0;
+}
+
+/**
+ *	tiocgsid		-	get session id
+ *	@tty: tty passed by user
+ *	@real_tty: tty side of the tty passed by the user if a pty else the tty
+ *	@p: pointer to returned session id
+ *
+ *	Obtain the session id of the tty. If there is no session
+ *	return an error.
+ */
+static int tiocgsid(struct tty_struct *tty, struct tty_struct *real_tty, pid_t __user *p)
+{
+	pid_t sid;
+
+	/*
+	 * (tty == real_tty) is a cheap way of
+	 * testing if the tty is NOT a master pty.
+	 */
+	if (tty == real_tty && current->signal->tty != real_tty)
+		return -ENOTTY;
+
+	scoped_guard(spinlock_irqsave, &real_tty->ctrl.lock) {
+		if (!real_tty->ctrl.session)
+			return -ENOTTY;
+		sid = pid_vnr(real_tty->ctrl.session);
+	}
+
+	return put_user(sid, p);
+}
+
+/*
+ * Called from tty_ioctl(). If tty is a pty then real_tty is the slave side,
+ * if not then tty == real_tty.
+ */
+long tty_jobctrl_ioctl(struct tty_struct *tty, struct tty_struct *real_tty,
+		       struct file *file, unsigned int cmd, unsigned long arg)
+{
+	void __user *p = (void __user *)arg;
+
+	switch (cmd) {
+	case TIOCNOTTY:
+		if (current->signal->tty != tty)
+			return -ENOTTY;
+		no_tty();
+		return 0;
+	case TIOCSCTTY:
+		return tiocsctty(real_tty, file, arg);
+	case TIOCGPGRP:
+		return tiocgpgrp(tty, real_tty, p);
+	case TIOCSPGRP:
+		return tiocspgrp(tty, real_tty, p);
+	case TIOCGSID:
+		return tiocgsid(tty, real_tty, p);
+	}
+	return -ENOIOCTLCMD;
+}