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

3 files changed, 81 insertions, 0 deletions

diff --git a/drivers/reset/tegra/Kconfig b/drivers/reset/tegra/Kconfig
new file mode 100644
index 000000000..4a2d26d12
--- /dev/null
+++ b/drivers/reset/tegra/Kconfig
@@ -0,0 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
+config RESET_TEGRA_BPMP
+	bool "Tegra BPMP Reset Driver" if COMPILE_TEST
+	default TEGRA_BPMP
diff --git a/drivers/reset/tegra/Makefile b/drivers/reset/tegra/Makefile
new file mode 100644
index 000000000..eccba896a
--- /dev/null
+++ b/drivers/reset/tegra/Makefile
@@ -0,0 +1,2 @@
+# SPDX-License-Identifier: GPL-2.0-only
+obj-$(CONFIG_RESET_TEGRA_BPMP) += reset-bpmp.o
diff --git a/drivers/reset/tegra/reset-bpmp.c b/drivers/reset/tegra/reset-bpmp.c
new file mode 100644
index 000000000..4c5bba52b
--- /dev/null
+++ b/drivers/reset/tegra/reset-bpmp.c
@@ -0,0 +1,75 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2016 NVIDIA Corporation
+ */
+
+#include <linux/reset-controller.h>
+
+#include <soc/tegra/bpmp.h>
+#include <soc/tegra/bpmp-abi.h>
+
+static struct tegra_bpmp *to_tegra_bpmp(struct reset_controller_dev *rstc)
+{
+	return container_of(rstc, struct tegra_bpmp, rstc);
+}
+
+static int tegra_bpmp_reset_common(struct reset_controller_dev *rstc,
+				   enum mrq_reset_commands command,
+				   unsigned int id)
+{
+	struct tegra_bpmp *bpmp = to_tegra_bpmp(rstc);
+	struct mrq_reset_request request;
+	struct tegra_bpmp_message msg;
+	int err;
+
+	memset(&request, 0, sizeof(request));
+	request.cmd = command;
+	request.reset_id = id;
+
+	memset(&msg, 0, sizeof(msg));
+	msg.mrq = MRQ_RESET;
+	msg.tx.data = &request;
+	msg.tx.size = sizeof(request);
+
+	err = tegra_bpmp_transfer(bpmp, &msg);
+	if (err)
+		return err;
+	if (msg.rx.ret)
+		return -EINVAL;
+
+	return 0;
+}
+
+static int tegra_bpmp_reset_module(struct reset_controller_dev *rstc,
+				   unsigned long id)
+{
+	return tegra_bpmp_reset_common(rstc, CMD_RESET_MODULE, id);
+}
+
+static int tegra_bpmp_reset_assert(struct reset_controller_dev *rstc,
+				   unsigned long id)
+{
+	return tegra_bpmp_reset_common(rstc, CMD_RESET_ASSERT, id);
+}
+
+static int tegra_bpmp_reset_deassert(struct reset_controller_dev *rstc,
+				     unsigned long id)
+{
+	return tegra_bpmp_reset_common(rstc, CMD_RESET_DEASSERT, id);
+}
+
+static const struct reset_control_ops tegra_bpmp_reset_ops = {
+	.reset = tegra_bpmp_reset_module,
+	.assert = tegra_bpmp_reset_assert,
+	.deassert = tegra_bpmp_reset_deassert,
+};
+
+int tegra_bpmp_init_resets(struct tegra_bpmp *bpmp)
+{
+	bpmp->rstc.ops = &tegra_bpmp_reset_ops;
+	bpmp->rstc.owner = THIS_MODULE;
+	bpmp->rstc.of_node = bpmp->dev->of_node;
+	bpmp->rstc.nr_resets = bpmp->soc->num_resets;
+
+	return devm_reset_controller_register(bpmp->dev, &bpmp->rstc);
+}