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, 241 insertions, 0 deletions

diff --git a/drivers/mailbox/rockchip-mailbox.c b/drivers/mailbox/rockchip-mailbox.c
new file mode 100644
index 000000000..bc2f9dd79
--- /dev/null
+++ b/drivers/mailbox/rockchip-mailbox.c
@@ -0,0 +1,241 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (c) 2015, Fuzhou Rockchip Electronics Co., Ltd
+ */
+
+#include <linux/clk.h>
+#include <linux/interrupt.h>
+#include <linux/io.h>
+#include <linux/kernel.h>
+#include <linux/mailbox_controller.h>
+#include <linux/of.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+
+#define MAILBOX_A2B_INTEN		0x00
+#define MAILBOX_A2B_STATUS		0x04
+#define MAILBOX_A2B_CMD(x)		(0x08 + (x) * 8)
+#define MAILBOX_A2B_DAT(x)		(0x0c + (x) * 8)
+
+#define MAILBOX_B2A_INTEN		0x28
+#define MAILBOX_B2A_STATUS		0x2C
+#define MAILBOX_B2A_CMD(x)		(0x30 + (x) * 8)
+#define MAILBOX_B2A_DAT(x)		(0x34 + (x) * 8)
+
+struct rockchip_mbox_msg {
+	u32 cmd;
+	int rx_size;
+};
+
+struct rockchip_mbox_data {
+	int num_chans;
+};
+
+struct rockchip_mbox_chan {
+	int idx;
+	int irq;
+	struct rockchip_mbox_msg *msg;
+	struct rockchip_mbox *mb;
+};
+
+struct rockchip_mbox {
+	struct mbox_controller mbox;
+	void __iomem *mbox_base;
+
+	/* The maximum size of buf for each channel */
+	u32 buf_size;
+
+	struct rockchip_mbox_chan chans[];
+};
+
+static int rockchip_mbox_send_data(struct mbox_chan *chan, void *data)
+{
+	struct rockchip_mbox *mb = dev_get_drvdata(chan->mbox->dev);
+	struct rockchip_mbox_msg *msg = data;
+	struct rockchip_mbox_chan *chans = mb->chans;
+
+	if (!msg)
+		return -EINVAL;
+
+	if (msg->rx_size > mb->buf_size) {
+		dev_err(mb->mbox.dev, "Transmit size over buf size(%d)\n",
+			mb->buf_size);
+		return -EINVAL;
+	}
+
+	dev_dbg(mb->mbox.dev, "Chan[%d]: A2B message, cmd 0x%08x\n",
+		chans->idx, msg->cmd);
+
+	mb->chans[chans->idx].msg = msg;
+
+	writel_relaxed(msg->cmd, mb->mbox_base + MAILBOX_A2B_CMD(chans->idx));
+	writel_relaxed(msg->rx_size, mb->mbox_base +
+		       MAILBOX_A2B_DAT(chans->idx));
+
+	return 0;
+}
+
+static int rockchip_mbox_startup(struct mbox_chan *chan)
+{
+	struct rockchip_mbox *mb = dev_get_drvdata(chan->mbox->dev);
+
+	/* Enable all B2A interrupts */
+	writel_relaxed((1 << mb->mbox.num_chans) - 1,
+		       mb->mbox_base + MAILBOX_B2A_INTEN);
+
+	return 0;
+}
+
+static void rockchip_mbox_shutdown(struct mbox_chan *chan)
+{
+	struct rockchip_mbox *mb = dev_get_drvdata(chan->mbox->dev);
+	struct rockchip_mbox_chan *chans = mb->chans;
+
+	/* Disable all B2A interrupts */
+	writel_relaxed(0, mb->mbox_base + MAILBOX_B2A_INTEN);
+
+	mb->chans[chans->idx].msg = NULL;
+}
+
+static const struct mbox_chan_ops rockchip_mbox_chan_ops = {
+	.send_data	= rockchip_mbox_send_data,
+	.startup	= rockchip_mbox_startup,
+	.shutdown	= rockchip_mbox_shutdown,
+};
+
+static irqreturn_t rockchip_mbox_irq(int irq, void *dev_id)
+{
+	int idx;
+	struct rockchip_mbox *mb = (struct rockchip_mbox *)dev_id;
+	u32 status = readl_relaxed(mb->mbox_base + MAILBOX_B2A_STATUS);
+
+	for (idx = 0; idx < mb->mbox.num_chans; idx++) {
+		if ((status & (1 << idx)) && (irq == mb->chans[idx].irq)) {
+			/* Clear mbox interrupt */
+			writel_relaxed(1 << idx,
+				       mb->mbox_base + MAILBOX_B2A_STATUS);
+			return IRQ_WAKE_THREAD;
+		}
+	}
+
+	return IRQ_NONE;
+}
+
+static irqreturn_t rockchip_mbox_isr(int irq, void *dev_id)
+{
+	int idx;
+	struct rockchip_mbox_msg *msg = NULL;
+	struct rockchip_mbox *mb = (struct rockchip_mbox *)dev_id;
+
+	for (idx = 0; idx < mb->mbox.num_chans; idx++) {
+		if (irq != mb->chans[idx].irq)
+			continue;
+
+		msg = mb->chans[idx].msg;
+		if (!msg) {
+			dev_err(mb->mbox.dev,
+				"Chan[%d]: B2A message is NULL\n", idx);
+			break; /* spurious */
+		}
+
+		mbox_chan_received_data(&mb->mbox.chans[idx], msg);
+		mb->chans[idx].msg = NULL;
+
+		dev_dbg(mb->mbox.dev, "Chan[%d]: B2A message, cmd 0x%08x\n",
+			idx, msg->cmd);
+
+		break;
+	}
+
+	return IRQ_HANDLED;
+}
+
+static const struct rockchip_mbox_data rk3368_drv_data = {
+	.num_chans = 4,
+};
+
+static const struct of_device_id rockchip_mbox_of_match[] = {
+	{ .compatible = "rockchip,rk3368-mailbox", .data = &rk3368_drv_data},
+	{ },
+};
+MODULE_DEVICE_TABLE(of, rockchip_mbox_of_match);
+
+static int rockchip_mbox_probe(struct platform_device *pdev)
+{
+	struct rockchip_mbox *mb;
+	const struct rockchip_mbox_data *drv_data;
+	struct resource *res;
+	struct clk *pclk;
+	int ret, irq, i;
+
+	if (!pdev->dev.of_node)
+		return -ENODEV;
+
+	drv_data = (const struct rockchip_mbox_data *) device_get_match_data(&pdev->dev);
+
+	mb = devm_kzalloc(&pdev->dev, struct_size(mb, chans, drv_data->num_chans), GFP_KERNEL);
+	if (!mb)
+		return -ENOMEM;
+
+	mb->mbox.chans = devm_kcalloc(&pdev->dev, drv_data->num_chans,
+				      sizeof(*mb->mbox.chans), GFP_KERNEL);
+	if (!mb->mbox.chans)
+		return -ENOMEM;
+
+	platform_set_drvdata(pdev, mb);
+
+	mb->mbox.dev = &pdev->dev;
+	mb->mbox.num_chans = drv_data->num_chans;
+	mb->mbox.ops = &rockchip_mbox_chan_ops;
+	mb->mbox.txdone_irq = true;
+
+	mb->mbox_base = devm_platform_get_and_ioremap_resource(pdev, 0, &res);
+	if (IS_ERR(mb->mbox_base))
+		return PTR_ERR(mb->mbox_base);
+
+	/* Each channel has two buffers for A2B and B2A */
+	mb->buf_size = (size_t)resource_size(res) / (drv_data->num_chans * 2);
+
+	pclk = devm_clk_get_enabled(&pdev->dev, "pclk_mailbox");
+	if (IS_ERR(pclk))
+		return dev_err_probe(&pdev->dev, PTR_ERR(pclk),
+				     "failed to get and enable pclk_mailbox clock\n");
+
+	for (i = 0; i < mb->mbox.num_chans; i++) {
+		irq = platform_get_irq(pdev, i);
+		if (irq < 0)
+			return irq;
+
+		ret = devm_request_threaded_irq(&pdev->dev, irq,
+						rockchip_mbox_irq,
+						rockchip_mbox_isr, IRQF_ONESHOT,
+						dev_name(&pdev->dev), mb);
+		if (ret < 0)
+			return ret;
+
+		mb->chans[i].idx = i;
+		mb->chans[i].irq = irq;
+		mb->chans[i].mb = mb;
+		mb->chans[i].msg = NULL;
+	}
+
+	ret = devm_mbox_controller_register(&pdev->dev, &mb->mbox);
+	if (ret < 0)
+		dev_err(&pdev->dev, "Failed to register mailbox: %d\n", ret);
+
+	return ret;
+}
+
+static struct platform_driver rockchip_mbox_driver = {
+	.probe	= rockchip_mbox_probe,
+	.driver = {
+		.name = "rockchip-mailbox",
+		.of_match_table = rockchip_mbox_of_match,
+	},
+};
+
+module_platform_driver(rockchip_mbox_driver);
+
+MODULE_DESCRIPTION("Rockchip mailbox: communicate between CPU cores and MCU");
+MODULE_AUTHOR("Addy Ke <addy.ke@rock-chips.com>");
+MODULE_AUTHOR("Caesar Wang <wxt@rock-chips.com>");