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, 146 insertions, 0 deletions

diff --git a/net/ceph/auth_none.c b/net/ceph/auth_none.c
new file mode 100644
index 000000000..99e1f3e10
--- /dev/null
+++ b/net/ceph/auth_none.c
@@ -0,0 +1,146 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <linux/ceph/ceph_debug.h>
+
+#include <linux/err.h>
+#include <linux/module.h>
+#include <linux/random.h>
+#include <linux/slab.h>
+
+#include <linux/ceph/decode.h>
+#include <linux/ceph/auth.h>
+
+#include "auth_none.h"
+
+static void reset(struct ceph_auth_client *ac)
+{
+	struct ceph_auth_none_info *xi = ac->private;
+
+	xi->starting = true;
+}
+
+static void destroy(struct ceph_auth_client *ac)
+{
+	kfree(ac->private);
+	ac->private = NULL;
+}
+
+static int is_authenticated(struct ceph_auth_client *ac)
+{
+	struct ceph_auth_none_info *xi = ac->private;
+
+	return !xi->starting;
+}
+
+static int should_authenticate(struct ceph_auth_client *ac)
+{
+	struct ceph_auth_none_info *xi = ac->private;
+
+	return xi->starting;
+}
+
+static int ceph_auth_none_build_authorizer(struct ceph_auth_client *ac,
+					   struct ceph_none_authorizer *au)
+{
+	void *p = au->buf;
+	void *const end = p + sizeof(au->buf);
+	int ret;
+
+	ceph_encode_8_safe(&p, end, 1, e_range);
+	ret = ceph_auth_entity_name_encode(ac->name, &p, end);
+	if (ret < 0)
+		return ret;
+
+	ceph_encode_64_safe(&p, end, ac->global_id, e_range);
+	au->buf_len = p - (void *)au->buf;
+	dout("%s built authorizer len %d\n", __func__, au->buf_len);
+	return 0;
+
+e_range:
+	return -ERANGE;
+}
+
+static int build_request(struct ceph_auth_client *ac, void *buf, void *end)
+{
+	return 0;
+}
+
+/*
+ * the generic auth code decode the global_id, and we carry no actual
+ * authenticate state, so nothing happens here.
+ */
+static int handle_reply(struct ceph_auth_client *ac, u64 global_id,
+			void *buf, void *end, u8 *session_key,
+			int *session_key_len, u8 *con_secret,
+			int *con_secret_len)
+{
+	struct ceph_auth_none_info *xi = ac->private;
+
+	xi->starting = false;
+	ceph_auth_set_global_id(ac, global_id);
+	return 0;
+}
+
+static void ceph_auth_none_destroy_authorizer(struct ceph_authorizer *a)
+{
+	kfree(a);
+}
+
+/*
+ * build an 'authorizer' with our entity_name and global_id.  it is
+ * identical for all services we connect to.
+ */
+static int ceph_auth_none_create_authorizer(
+	struct ceph_auth_client *ac, int peer_type,
+	struct ceph_auth_handshake *auth)
+{
+	struct ceph_none_authorizer *au;
+	int ret;
+
+	au = kmalloc_obj(*au, GFP_NOFS);
+	if (!au)
+		return -ENOMEM;
+
+	au->base.destroy = ceph_auth_none_destroy_authorizer;
+
+	ret = ceph_auth_none_build_authorizer(ac, au);
+	if (ret) {
+		kfree(au);
+		return ret;
+	}
+
+	auth->authorizer = (struct ceph_authorizer *) au;
+	auth->authorizer_buf = au->buf;
+	auth->authorizer_buf_len = au->buf_len;
+	auth->authorizer_reply_buf = NULL;
+	auth->authorizer_reply_buf_len = 0;
+
+	return 0;
+}
+
+static const struct ceph_auth_client_ops ceph_auth_none_ops = {
+	.reset = reset,
+	.destroy = destroy,
+	.is_authenticated = is_authenticated,
+	.should_authenticate = should_authenticate,
+	.build_request = build_request,
+	.handle_reply = handle_reply,
+	.create_authorizer = ceph_auth_none_create_authorizer,
+};
+
+int ceph_auth_none_init(struct ceph_auth_client *ac)
+{
+	struct ceph_auth_none_info *xi;
+
+	dout("ceph_auth_none_init %p\n", ac);
+	xi = kzalloc_obj(*xi, GFP_NOFS);
+	if (!xi)
+		return -ENOMEM;
+
+	xi->starting = true;
+
+	ac->protocol = CEPH_AUTH_NONE;
+	ac->private = xi;
+	ac->ops = &ceph_auth_none_ops;
+	return 0;
+}