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, 108 insertions, 0 deletions

diff --git a/drivers/md/dm-vdo/thread-utils.c b/drivers/md/dm-vdo/thread-utils.c
new file mode 100644
index 000000000..826afc952
--- /dev/null
+++ b/drivers/md/dm-vdo/thread-utils.c
@@ -0,0 +1,108 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright 2023 Red Hat
+ */
+
+#include "thread-utils.h"
+
+#include <asm/current.h>
+#include <linux/delay.h>
+#include <linux/kthread.h>
+#include <linux/mutex.h>
+#include <linux/types.h>
+
+#include "errors.h"
+#include "logger.h"
+#include "memory-alloc.h"
+
+static struct hlist_head thread_list;
+static struct mutex thread_mutex;
+
+struct thread {
+	void (*thread_function)(void *thread_data);
+	void *thread_data;
+	struct hlist_node thread_links;
+	struct task_struct *thread_task;
+	struct completion thread_done;
+};
+
+void vdo_initialize_threads_mutex(void)
+{
+	mutex_init(&thread_mutex);
+}
+
+static int thread_starter(void *arg)
+{
+	struct registered_thread allocating_thread;
+	struct thread *thread = arg;
+
+	thread->thread_task = current;
+	mutex_lock(&thread_mutex);
+	hlist_add_head(&thread->thread_links, &thread_list);
+	mutex_unlock(&thread_mutex);
+	vdo_register_allocating_thread(&allocating_thread, NULL);
+	thread->thread_function(thread->thread_data);
+	vdo_unregister_allocating_thread();
+	complete(&thread->thread_done);
+	return 0;
+}
+
+int vdo_create_thread(void (*thread_function)(void *), void *thread_data,
+		      const char *name, struct thread **new_thread)
+{
+	char *name_colon = strchr(name, ':');
+	char *my_name_colon = strchr(current->comm, ':');
+	struct task_struct *task;
+	struct thread *thread;
+	int result;
+
+	result = vdo_allocate(1, __func__, &thread);
+	if (result != VDO_SUCCESS) {
+		vdo_log_warning("Error allocating memory for %s", name);
+		return result;
+	}
+
+	thread->thread_function = thread_function;
+	thread->thread_data = thread_data;
+	init_completion(&thread->thread_done);
+	/*
+	 * Start the thread, with an appropriate thread name.
+	 *
+	 * If the name supplied contains a colon character, use that name. This causes uds module
+	 * threads to have names like "uds:callbackW" and the main test runner thread to be named
+	 * "zub:runtest".
+	 *
+	 * Otherwise if the current thread has a name containing a colon character, prefix the name
+	 * supplied with the name of the current thread up to (and including) the colon character.
+	 * Thus when the "kvdo0:dedupeQ" thread opens an index session, all the threads associated
+	 * with that index will have names like "kvdo0:foo".
+	 *
+	 * Otherwise just use the name supplied. This should be a rare occurrence.
+	 */
+	if ((name_colon == NULL) && (my_name_colon != NULL)) {
+		task = kthread_run(thread_starter, thread, "%.*s:%s",
+				   (int) (my_name_colon - current->comm), current->comm,
+				   name);
+	} else {
+		task = kthread_run(thread_starter, thread, "%s", name);
+	}
+
+	if (IS_ERR(task)) {
+		vdo_free(thread);
+		return PTR_ERR(task);
+	}
+
+	*new_thread = thread;
+	return VDO_SUCCESS;
+}
+
+void vdo_join_threads(struct thread *thread)
+{
+	while (wait_for_completion_interruptible(&thread->thread_done))
+		fsleep(1000);
+
+	mutex_lock(&thread_mutex);
+	hlist_del(&thread->thread_links);
+	mutex_unlock(&thread_mutex);
+	vdo_free(thread);
+}