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, 215 insertions, 0 deletions

diff --git a/drivers/vfio/cdx/intr.c b/drivers/vfio/cdx/intr.c
new file mode 100644
index 000000000..4439481fe
--- /dev/null
+++ b/drivers/vfio/cdx/intr.c
@@ -0,0 +1,215 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2022-2023, Advanced Micro Devices, Inc.
+ */
+
+#include <linux/vfio.h>
+#include <linux/slab.h>
+#include <linux/types.h>
+#include <linux/eventfd.h>
+#include <linux/msi.h>
+#include <linux/interrupt.h>
+
+#include "linux/cdx/cdx_bus.h"
+#include "private.h"
+
+static irqreturn_t vfio_cdx_msihandler(int irq_no, void *arg)
+{
+	struct eventfd_ctx *trigger = arg;
+
+	eventfd_signal(trigger);
+	return IRQ_HANDLED;
+}
+
+static int vfio_cdx_msi_enable(struct vfio_cdx_device *vdev, int nvec)
+{
+	struct cdx_device *cdx_dev = to_cdx_device(vdev->vdev.dev);
+	struct device *dev = vdev->vdev.dev;
+	int msi_idx, ret;
+
+	vdev->cdx_irqs = kzalloc_objs(struct vfio_cdx_irq, nvec);
+	if (!vdev->cdx_irqs)
+		return -ENOMEM;
+
+	ret = cdx_enable_msi(cdx_dev);
+	if (ret)
+		goto err_free;
+
+	/* Allocate cdx MSIs */
+	ret = msi_domain_alloc_irqs(dev, MSI_DEFAULT_DOMAIN, nvec);
+	if (ret)
+		goto err_disable;
+
+	for (msi_idx = 0; msi_idx < nvec; msi_idx++)
+		vdev->cdx_irqs[msi_idx].irq_no = msi_get_virq(dev, msi_idx);
+
+	vdev->msi_count = nvec;
+
+	return 0;
+
+err_disable:
+	cdx_disable_msi(cdx_dev);
+err_free:
+	kfree(vdev->cdx_irqs);
+	vdev->cdx_irqs = NULL;
+	return ret;
+}
+
+static int vfio_cdx_msi_set_vector_signal(struct vfio_cdx_device *vdev,
+					  int vector, int fd)
+{
+	struct eventfd_ctx *trigger;
+	int irq_no, ret;
+
+	if (vector < 0 || vector >= vdev->msi_count)
+		return -EINVAL;
+
+	irq_no = vdev->cdx_irqs[vector].irq_no;
+
+	if (vdev->cdx_irqs[vector].trigger) {
+		free_irq(irq_no, vdev->cdx_irqs[vector].trigger);
+		kfree(vdev->cdx_irqs[vector].name);
+		eventfd_ctx_put(vdev->cdx_irqs[vector].trigger);
+		vdev->cdx_irqs[vector].trigger = NULL;
+	}
+
+	if (fd < 0)
+		return 0;
+
+	vdev->cdx_irqs[vector].name = kasprintf(GFP_KERNEL, "vfio-msi[%d](%s)",
+						vector, dev_name(vdev->vdev.dev));
+	if (!vdev->cdx_irqs[vector].name)
+		return -ENOMEM;
+
+	trigger = eventfd_ctx_fdget(fd);
+	if (IS_ERR(trigger)) {
+		kfree(vdev->cdx_irqs[vector].name);
+		return PTR_ERR(trigger);
+	}
+
+	ret = request_irq(irq_no, vfio_cdx_msihandler, 0,
+			  vdev->cdx_irqs[vector].name, trigger);
+	if (ret) {
+		kfree(vdev->cdx_irqs[vector].name);
+		eventfd_ctx_put(trigger);
+		return ret;
+	}
+
+	vdev->cdx_irqs[vector].trigger = trigger;
+
+	return 0;
+}
+
+static int vfio_cdx_msi_set_block(struct vfio_cdx_device *vdev,
+				  unsigned int start, unsigned int count,
+				  int32_t *fds)
+{
+	int i, j, ret = 0;
+
+	if (start >= vdev->msi_count || start + count > vdev->msi_count)
+		return -EINVAL;
+
+	for (i = 0, j = start; i < count && !ret; i++, j++) {
+		int fd = fds ? fds[i] : -1;
+
+		ret = vfio_cdx_msi_set_vector_signal(vdev, j, fd);
+	}
+
+	if (ret) {
+		for (--j; j >= (int)start; j--)
+			vfio_cdx_msi_set_vector_signal(vdev, j, -1);
+	}
+
+	return ret;
+}
+
+static void vfio_cdx_msi_disable(struct vfio_cdx_device *vdev)
+{
+	struct cdx_device *cdx_dev = to_cdx_device(vdev->vdev.dev);
+	struct device *dev = vdev->vdev.dev;
+
+	vfio_cdx_msi_set_block(vdev, 0, vdev->msi_count, NULL);
+
+	if (!vdev->cdx_irqs)
+		return;
+
+	msi_domain_free_irqs_all(dev, MSI_DEFAULT_DOMAIN);
+	cdx_disable_msi(cdx_dev);
+	kfree(vdev->cdx_irqs);
+
+	vdev->cdx_irqs = NULL;
+	vdev->msi_count = 0;
+}
+
+static int vfio_cdx_set_msi_trigger(struct vfio_cdx_device *vdev,
+				    unsigned int index, unsigned int start,
+				    unsigned int count, u32 flags,
+				    void *data)
+{
+	struct cdx_device *cdx_dev = to_cdx_device(vdev->vdev.dev);
+	int i;
+
+	if (start + count > cdx_dev->num_msi)
+		return -EINVAL;
+
+	guard(mutex)(&vdev->cdx_irqs_lock);
+
+	if (!count && (flags & VFIO_IRQ_SET_DATA_NONE)) {
+		vfio_cdx_msi_disable(vdev);
+		return 0;
+	}
+
+	if (flags & VFIO_IRQ_SET_DATA_EVENTFD) {
+		s32 *fds = data;
+		int ret;
+
+		if (vdev->cdx_irqs)
+			return vfio_cdx_msi_set_block(vdev, start, count,
+						  fds);
+		ret = vfio_cdx_msi_enable(vdev, cdx_dev->num_msi);
+		if (ret)
+			return ret;
+
+		ret = vfio_cdx_msi_set_block(vdev, start, count, fds);
+		if (ret)
+			vfio_cdx_msi_disable(vdev);
+
+		return ret;
+	}
+
+	if (!vdev->cdx_irqs)
+		return -EINVAL;
+
+	for (i = start; i < start + count; i++) {
+		if (!vdev->cdx_irqs[i].trigger)
+			continue;
+		if (flags & VFIO_IRQ_SET_DATA_NONE) {
+			eventfd_signal(vdev->cdx_irqs[i].trigger);
+		} else if (flags & VFIO_IRQ_SET_DATA_BOOL) {
+			u8 *bools = data;
+
+			if (bools[i - start])
+				eventfd_signal(vdev->cdx_irqs[i].trigger);
+		}
+	}
+
+	return 0;
+}
+
+int vfio_cdx_set_irqs_ioctl(struct vfio_cdx_device *vdev,
+			    u32 flags, unsigned int index,
+			    unsigned int start, unsigned int count,
+			    void *data)
+{
+	if (flags & VFIO_IRQ_SET_ACTION_TRIGGER)
+		return vfio_cdx_set_msi_trigger(vdev, index, start,
+			  count, flags, data);
+	else
+		return -EINVAL;
+}
+
+/* Free All IRQs for the given device */
+void vfio_cdx_irqs_cleanup(struct vfio_cdx_device *vdev)
+{
+	vfio_cdx_set_msi_trigger(vdev, 0, 0, 0, VFIO_IRQ_SET_DATA_NONE, NULL);
+}