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, 116 insertions, 0 deletions

diff --git a/fs/proc/consoles.c b/fs/proc/consoles.c
new file mode 100644
index 000000000..b7cab1ad9
--- /dev/null
+++ b/fs/proc/consoles.c
@@ -0,0 +1,116 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (c) 2010 Werner Fink, Jiri Slaby
+ */
+
+#include <linux/console.h>
+#include <linux/kernel.h>
+#include <linux/proc_fs.h>
+#include <linux/seq_file.h>
+#include <linux/tty_driver.h>
+
+/*
+ * This is handler for /proc/consoles
+ */
+static int show_console_dev(struct seq_file *m, void *v)
+{
+	static const struct {
+		short flag;
+		char name;
+	} con_flags[] = {
+		{ CON_ENABLED,		'E' },
+		{ CON_CONSDEV,		'C' },
+		{ CON_BOOT,		'B' },
+		{ CON_NBCON,		'N' },
+		{ CON_PRINTBUFFER,	'p' },
+		{ CON_BRL,		'b' },
+		{ CON_ANYTIME,		'a' },
+	};
+	char flags[ARRAY_SIZE(con_flags) + 1];
+	struct console *con = v;
+	unsigned int a;
+	dev_t dev = 0;
+
+	if (con->device) {
+		const struct tty_driver *driver;
+		int index;
+
+		/*
+		 * Take console_lock to serialize device() callback with
+		 * other console operations. For example, fg_console is
+		 * modified under console_lock when switching vt.
+		 */
+		console_lock();
+		driver = con->device(con, &index);
+		console_unlock();
+
+		if (driver) {
+			dev = MKDEV(driver->major, driver->minor_start);
+			dev += index;
+		}
+	}
+
+	for (a = 0; a < ARRAY_SIZE(con_flags); a++)
+		flags[a] = (con->flags & con_flags[a].flag) ?
+			con_flags[a].name : ' ';
+	flags[a] = 0;
+
+	seq_setwidth(m, 21 - 1);
+	seq_printf(m, "%s%d", con->name, con->index);
+	seq_pad(m, ' ');
+	seq_printf(m, "%c%c%c (%s)", con->read ? 'R' : '-',
+		   ((con->flags & CON_NBCON) || con->write) ? 'W' : '-',
+		   con->unblank ? 'U' : '-', flags);
+	if (dev)
+		seq_printf(m, " %4d:%d", MAJOR(dev), MINOR(dev));
+
+	seq_putc(m, '\n');
+	return 0;
+}
+
+static void *c_start(struct seq_file *m, loff_t *pos)
+	__acquires(&console_mutex)
+{
+	struct console *con;
+	loff_t off = 0;
+
+	/*
+	 * Hold the console_list_lock to guarantee safe traversal of the
+	 * console list. SRCU cannot be used because there is no
+	 * place to store the SRCU cookie.
+	 */
+	console_list_lock();
+	for_each_console(con)
+		if (off++ == *pos)
+			break;
+
+	return con;
+}
+
+static void *c_next(struct seq_file *m, void *v, loff_t *pos)
+{
+	struct console *con = v;
+
+	++*pos;
+	return hlist_entry_safe(con->node.next, struct console, node);
+}
+
+static void c_stop(struct seq_file *m, void *v)
+	__releases(&console_mutex)
+{
+	console_list_unlock();
+}
+
+static const struct seq_operations consoles_op = {
+	.start	= c_start,
+	.next	= c_next,
+	.stop	= c_stop,
+	.show	= show_console_dev
+};
+
+static int __init proc_consoles_init(void)
+{
+	proc_create_seq("consoles", 0, NULL, &consoles_op);
+	return 0;
+}
+fs_initcall(proc_consoles_init);