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, 193 insertions, 0 deletions

diff --git a/drivers/bluetooth/btmrvl_debugfs.c b/drivers/bluetooth/btmrvl_debugfs.c
new file mode 100644
index 000000000..9b3e768c0
--- /dev/null
+++ b/drivers/bluetooth/btmrvl_debugfs.c
@@ -0,0 +1,193 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Marvell Bluetooth driver: debugfs related functions
+ *
+ * Copyright (C) 2009, Marvell International Ltd.
+ **/
+
+#include <linux/debugfs.h>
+#include <linux/slab.h>
+
+#include <net/bluetooth/bluetooth.h>
+#include <net/bluetooth/hci_core.h>
+
+#include "btmrvl_drv.h"
+
+struct btmrvl_debugfs_data {
+	struct dentry *config_dir;
+	struct dentry *status_dir;
+};
+
+static ssize_t btmrvl_hscfgcmd_write(struct file *file,
+			const char __user *ubuf, size_t count, loff_t *ppos)
+{
+	struct btmrvl_private *priv = file->private_data;
+	long result, ret;
+
+	ret = kstrtol_from_user(ubuf, count, 10, &result);
+	if (ret)
+		return ret;
+
+	priv->btmrvl_dev.hscfgcmd = result;
+
+	if (priv->btmrvl_dev.hscfgcmd) {
+		btmrvl_prepare_command(priv);
+		wake_up_interruptible(&priv->main_thread.wait_q);
+	}
+
+	return count;
+}
+
+static ssize_t btmrvl_hscfgcmd_read(struct file *file, char __user *userbuf,
+						size_t count, loff_t *ppos)
+{
+	struct btmrvl_private *priv = file->private_data;
+	char buf[16];
+	int ret;
+
+	ret = snprintf(buf, sizeof(buf) - 1, "%d\n",
+						priv->btmrvl_dev.hscfgcmd);
+
+	return simple_read_from_buffer(userbuf, count, ppos, buf, ret);
+}
+
+static const struct file_operations btmrvl_hscfgcmd_fops = {
+	.read	= btmrvl_hscfgcmd_read,
+	.write	= btmrvl_hscfgcmd_write,
+	.open	= simple_open,
+	.llseek = default_llseek,
+};
+
+static ssize_t btmrvl_pscmd_write(struct file *file, const char __user *ubuf,
+						size_t count, loff_t *ppos)
+{
+	struct btmrvl_private *priv = file->private_data;
+	long result, ret;
+
+	ret = kstrtol_from_user(ubuf, count, 10, &result);
+	if (ret)
+		return ret;
+
+	priv->btmrvl_dev.pscmd = result;
+
+	if (priv->btmrvl_dev.pscmd) {
+		btmrvl_prepare_command(priv);
+		wake_up_interruptible(&priv->main_thread.wait_q);
+	}
+
+	return count;
+
+}
+
+static ssize_t btmrvl_pscmd_read(struct file *file, char __user *userbuf,
+						size_t count, loff_t *ppos)
+{
+	struct btmrvl_private *priv = file->private_data;
+	char buf[16];
+	int ret;
+
+	ret = snprintf(buf, sizeof(buf) - 1, "%d\n", priv->btmrvl_dev.pscmd);
+
+	return simple_read_from_buffer(userbuf, count, ppos, buf, ret);
+}
+
+static const struct file_operations btmrvl_pscmd_fops = {
+	.read = btmrvl_pscmd_read,
+	.write = btmrvl_pscmd_write,
+	.open = simple_open,
+	.llseek = default_llseek,
+};
+
+static ssize_t btmrvl_hscmd_write(struct file *file, const char __user *ubuf,
+						size_t count, loff_t *ppos)
+{
+	struct btmrvl_private *priv = file->private_data;
+	long result, ret;
+
+	ret = kstrtol_from_user(ubuf, count, 10, &result);
+	if (ret)
+		return ret;
+
+	priv->btmrvl_dev.hscmd = result;
+	if (priv->btmrvl_dev.hscmd) {
+		btmrvl_prepare_command(priv);
+		wake_up_interruptible(&priv->main_thread.wait_q);
+	}
+
+	return count;
+}
+
+static ssize_t btmrvl_hscmd_read(struct file *file, char __user *userbuf,
+						size_t count, loff_t *ppos)
+{
+	struct btmrvl_private *priv = file->private_data;
+	char buf[16];
+	int ret;
+
+	ret = snprintf(buf, sizeof(buf) - 1, "%d\n", priv->btmrvl_dev.hscmd);
+
+	return simple_read_from_buffer(userbuf, count, ppos, buf, ret);
+}
+
+static const struct file_operations btmrvl_hscmd_fops = {
+	.read	= btmrvl_hscmd_read,
+	.write	= btmrvl_hscmd_write,
+	.open	= simple_open,
+	.llseek = default_llseek,
+};
+
+void btmrvl_debugfs_init(struct hci_dev *hdev)
+{
+	struct btmrvl_private *priv = hci_get_drvdata(hdev);
+	struct btmrvl_debugfs_data *dbg;
+
+	if (!hdev->debugfs)
+		return;
+
+	dbg = kzalloc_obj(*dbg);
+	priv->debugfs_data = dbg;
+
+	if (!dbg) {
+		BT_ERR("Can not allocate memory for btmrvl_debugfs_data.");
+		return;
+	}
+
+	dbg->config_dir = debugfs_create_dir("config", hdev->debugfs);
+
+	debugfs_create_u8("psmode", 0644, dbg->config_dir,
+			  &priv->btmrvl_dev.psmode);
+	debugfs_create_file("pscmd", 0644, dbg->config_dir,
+			    priv, &btmrvl_pscmd_fops);
+	debugfs_create_x16("gpiogap", 0644, dbg->config_dir,
+			   &priv->btmrvl_dev.gpio_gap);
+	debugfs_create_u8("hsmode", 0644, dbg->config_dir,
+			  &priv->btmrvl_dev.hsmode);
+	debugfs_create_file("hscmd", 0644, dbg->config_dir,
+			    priv, &btmrvl_hscmd_fops);
+	debugfs_create_file("hscfgcmd", 0644, dbg->config_dir,
+			    priv, &btmrvl_hscfgcmd_fops);
+
+	dbg->status_dir = debugfs_create_dir("status", hdev->debugfs);
+	debugfs_create_u8("curpsmode", 0444, dbg->status_dir,
+			  &priv->adapter->psmode);
+	debugfs_create_u8("psstate", 0444, dbg->status_dir,
+			  &priv->adapter->ps_state);
+	debugfs_create_u8("hsstate", 0444, dbg->status_dir,
+			  &priv->adapter->hs_state);
+	debugfs_create_u8("txdnldready", 0444, dbg->status_dir,
+			  &priv->btmrvl_dev.tx_dnld_rdy);
+}
+
+void btmrvl_debugfs_remove(struct hci_dev *hdev)
+{
+	struct btmrvl_private *priv = hci_get_drvdata(hdev);
+	struct btmrvl_debugfs_data *dbg = priv->debugfs_data;
+
+	if (!dbg)
+		return;
+
+	debugfs_remove_recursive(dbg->config_dir);
+	debugfs_remove_recursive(dbg->status_dir);
+
+	kfree(dbg);
+}