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, 173 insertions, 0 deletions

diff --git a/samples/livepatch/livepatch-shadow-fix1.c b/samples/livepatch/livepatch-shadow-fix1.c
new file mode 100644
index 000000000..cbf68ca40
--- /dev/null
+++ b/samples/livepatch/livepatch-shadow-fix1.c
@@ -0,0 +1,173 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Copyright (C) 2017 Joe Lawrence <joe.lawrence@redhat.com>
+ */
+
+/*
+ * livepatch-shadow-fix1.c - Shadow variables, livepatch demo
+ *
+ * Purpose
+ * -------
+ *
+ * Fixes the memory leak introduced in livepatch-shadow-mod through the
+ * use of a shadow variable.  This fix demonstrates the "extending" of
+ * short-lived data structures by patching its allocation and release
+ * functions.
+ *
+ *
+ * Usage
+ * -----
+ *
+ * This module is not intended to be standalone.  See the "Usage"
+ * section of livepatch-shadow-mod.c.
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/livepatch.h>
+#include <linux/slab.h>
+
+/* Shadow variable enums */
+#define SV_LEAK		1
+
+/* Allocate new dummies every second */
+#define ALLOC_PERIOD	1
+/* Check for expired dummies after a few new ones have been allocated */
+#define CLEANUP_PERIOD	(3 * ALLOC_PERIOD)
+/* Dummies expire after a few cleanup instances */
+#define EXPIRE_PERIOD	(4 * CLEANUP_PERIOD)
+
+struct dummy {
+	struct list_head list;
+	unsigned long jiffies_expire;
+};
+
+/*
+ * The constructor makes more sense together with klp_shadow_get_or_alloc().
+ * In this example, it would be safe to assign the pointer also to the shadow
+ * variable returned by klp_shadow_alloc().  But we wanted to show the more
+ * complicated use of the API.
+ */
+static int shadow_leak_ctor(void *obj, void *shadow_data, void *ctor_data)
+{
+	int **shadow_leak = shadow_data;
+	int **leak = ctor_data;
+
+	if (!ctor_data)
+		return -EINVAL;
+
+	*shadow_leak = *leak;
+	return 0;
+}
+
+static struct dummy *livepatch_fix1_dummy_alloc(void)
+{
+	struct dummy *d;
+	int *leak;
+	int **shadow_leak;
+
+	d = kzalloc(sizeof(*d), GFP_KERNEL);
+	if (!d)
+		return NULL;
+
+	d->jiffies_expire = jiffies + secs_to_jiffies(EXPIRE_PERIOD);
+
+	/*
+	 * Patch: save the extra memory location into a SV_LEAK shadow
+	 * variable.  A patched dummy_free routine can later fetch this
+	 * pointer to handle resource release.
+	 */
+	leak = kzalloc(sizeof(*leak), GFP_KERNEL);
+	if (!leak)
+		goto err_leak;
+
+	shadow_leak = klp_shadow_alloc(d, SV_LEAK, sizeof(leak), GFP_KERNEL,
+				       shadow_leak_ctor, &leak);
+	if (!shadow_leak) {
+		pr_err("%s: failed to allocate shadow variable for the leaking pointer: dummy @ %p, leak @ %p\n",
+		       __func__, d, leak);
+		goto err_shadow;
+	}
+
+	pr_info("%s: dummy @ %p, expires @ %lx\n",
+		__func__, d, d->jiffies_expire);
+
+	return d;
+
+err_shadow:
+	kfree(leak);
+err_leak:
+	kfree(d);
+	return NULL;
+}
+
+static void livepatch_fix1_dummy_leak_dtor(void *obj, void *shadow_data)
+{
+	void *d = obj;
+	int **shadow_leak = shadow_data;
+
+	pr_info("%s: dummy @ %p, prevented leak @ %p\n",
+			 __func__, d, *shadow_leak);
+	kfree(*shadow_leak);
+}
+
+static void livepatch_fix1_dummy_free(struct dummy *d)
+{
+	int **shadow_leak;
+
+	/*
+	 * Patch: fetch the saved SV_LEAK shadow variable, detach and
+	 * free it.  Note: handle cases where this shadow variable does
+	 * not exist (ie, dummy structures allocated before this livepatch
+	 * was loaded.)
+	 */
+	shadow_leak = klp_shadow_get(d, SV_LEAK);
+	if (shadow_leak)
+		klp_shadow_free(d, SV_LEAK, livepatch_fix1_dummy_leak_dtor);
+	else
+		pr_info("%s: dummy @ %p leaked!\n", __func__, d);
+
+	kfree(d);
+}
+
+static struct klp_func funcs[] = {
+	{
+		.old_name = "dummy_alloc",
+		.new_func = livepatch_fix1_dummy_alloc,
+	},
+	{
+		.old_name = "dummy_free",
+		.new_func = livepatch_fix1_dummy_free,
+	}, { }
+};
+
+static struct klp_object objs[] = {
+	{
+		.name = "livepatch_shadow_mod",
+		.funcs = funcs,
+	}, { }
+};
+
+static struct klp_patch patch = {
+	.mod = THIS_MODULE,
+	.objs = objs,
+};
+
+static int livepatch_shadow_fix1_init(void)
+{
+	return klp_enable_patch(&patch);
+}
+
+static void livepatch_shadow_fix1_exit(void)
+{
+	/* Cleanup any existing SV_LEAK shadow variables */
+	klp_shadow_free_all(SV_LEAK, livepatch_fix1_dummy_leak_dtor);
+}
+
+module_init(livepatch_shadow_fix1_init);
+module_exit(livepatch_shadow_fix1_exit);
+MODULE_DESCRIPTION("Live patching demo for shadow variables");
+MODULE_LICENSE("GPL");
+MODULE_INFO(livepatch, "Y");