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, 192 insertions, 0 deletions

diff --git a/drivers/xen/privcmd-buf.c b/drivers/xen/privcmd-buf.c
new file mode 100644
index 000000000..4cb66eb54
--- /dev/null
+++ b/drivers/xen/privcmd-buf.c
@@ -0,0 +1,192 @@
+// SPDX-License-Identifier: GPL-2.0 OR MIT
+
+/******************************************************************************
+ * privcmd-buf.c
+ *
+ * Mmap of hypercall buffers.
+ *
+ * Copyright (c) 2018 Juergen Gross
+ */
+
+#define pr_fmt(fmt) "xen:" KBUILD_MODNAME ": " fmt
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/list.h>
+#include <linux/miscdevice.h>
+#include <linux/mm.h>
+#include <linux/slab.h>
+
+#include "privcmd.h"
+
+MODULE_DESCRIPTION("Xen Mmap of hypercall buffers");
+MODULE_LICENSE("GPL");
+
+struct privcmd_buf_private {
+	struct mutex lock;
+	struct list_head list;
+};
+
+struct privcmd_buf_vma_private {
+	struct privcmd_buf_private *file_priv;
+	struct list_head list;
+	unsigned int users;
+	unsigned int n_pages;
+	struct page *pages[];
+};
+
+static int privcmd_buf_open(struct inode *ino, struct file *file)
+{
+	struct privcmd_buf_private *file_priv;
+
+	file_priv = kzalloc_obj(*file_priv);
+	if (!file_priv)
+		return -ENOMEM;
+
+	mutex_init(&file_priv->lock);
+	INIT_LIST_HEAD(&file_priv->list);
+
+	file->private_data = file_priv;
+
+	return 0;
+}
+
+static void privcmd_buf_vmapriv_free(struct privcmd_buf_vma_private *vma_priv)
+{
+	unsigned int i;
+
+	list_del(&vma_priv->list);
+
+	for (i = 0; i < vma_priv->n_pages; i++)
+		__free_page(vma_priv->pages[i]);
+
+	kfree(vma_priv);
+}
+
+static int privcmd_buf_release(struct inode *ino, struct file *file)
+{
+	struct privcmd_buf_private *file_priv = file->private_data;
+	struct privcmd_buf_vma_private *vma_priv;
+
+	mutex_lock(&file_priv->lock);
+
+	while (!list_empty(&file_priv->list)) {
+		vma_priv = list_first_entry(&file_priv->list,
+					    struct privcmd_buf_vma_private,
+					    list);
+		privcmd_buf_vmapriv_free(vma_priv);
+	}
+
+	mutex_unlock(&file_priv->lock);
+
+	kfree(file_priv);
+
+	return 0;
+}
+
+static void privcmd_buf_vma_open(struct vm_area_struct *vma)
+{
+	struct privcmd_buf_vma_private *vma_priv = vma->vm_private_data;
+
+	if (!vma_priv)
+		return;
+
+	mutex_lock(&vma_priv->file_priv->lock);
+	vma_priv->users++;
+	mutex_unlock(&vma_priv->file_priv->lock);
+}
+
+static void privcmd_buf_vma_close(struct vm_area_struct *vma)
+{
+	struct privcmd_buf_vma_private *vma_priv = vma->vm_private_data;
+	struct privcmd_buf_private *file_priv;
+
+	if (!vma_priv)
+		return;
+
+	file_priv = vma_priv->file_priv;
+
+	mutex_lock(&file_priv->lock);
+
+	vma_priv->users--;
+	if (!vma_priv->users)
+		privcmd_buf_vmapriv_free(vma_priv);
+
+	mutex_unlock(&file_priv->lock);
+}
+
+static vm_fault_t privcmd_buf_vma_fault(struct vm_fault *vmf)
+{
+	pr_debug("fault: vma=%p %lx-%lx, pgoff=%lx, uv=%p\n",
+		 vmf->vma, vmf->vma->vm_start, vmf->vma->vm_end,
+		 vmf->pgoff, (void *)vmf->address);
+
+	return VM_FAULT_SIGBUS;
+}
+
+static const struct vm_operations_struct privcmd_buf_vm_ops = {
+	.open = privcmd_buf_vma_open,
+	.close = privcmd_buf_vma_close,
+	.fault = privcmd_buf_vma_fault,
+};
+
+static int privcmd_buf_mmap(struct file *file, struct vm_area_struct *vma)
+{
+	struct privcmd_buf_private *file_priv = file->private_data;
+	struct privcmd_buf_vma_private *vma_priv;
+	unsigned long count = vma_pages(vma);
+	unsigned int i;
+	int ret = 0;
+
+	if (!(vma->vm_flags & VM_SHARED))
+		return -EINVAL;
+
+	vma_priv = kzalloc_flex(*vma_priv, pages, count);
+	if (!vma_priv)
+		return -ENOMEM;
+
+	for (i = 0; i < count; i++) {
+		vma_priv->pages[i] = alloc_page(GFP_KERNEL | __GFP_ZERO);
+		if (!vma_priv->pages[i])
+			break;
+		vma_priv->n_pages++;
+	}
+
+	mutex_lock(&file_priv->lock);
+
+	vma_priv->file_priv = file_priv;
+	vma_priv->users = 1;
+
+	vm_flags_set(vma, VM_IO | VM_DONTEXPAND);
+	vma->vm_ops = &privcmd_buf_vm_ops;
+	vma->vm_private_data = vma_priv;
+
+	list_add(&vma_priv->list, &file_priv->list);
+
+	if (vma_priv->n_pages != count)
+		ret = -ENOMEM;
+	else
+		ret = vm_map_pages_zero(vma, vma_priv->pages,
+						vma_priv->n_pages);
+
+	if (ret)
+		privcmd_buf_vmapriv_free(vma_priv);
+
+	mutex_unlock(&file_priv->lock);
+
+	return ret;
+}
+
+const struct file_operations xen_privcmdbuf_fops = {
+	.owner = THIS_MODULE,
+	.open = privcmd_buf_open,
+	.release = privcmd_buf_release,
+	.mmap = privcmd_buf_mmap,
+};
+EXPORT_SYMBOL_GPL(xen_privcmdbuf_fops);
+
+struct miscdevice xen_privcmdbuf_dev = {
+	.minor = MISC_DYNAMIC_MINOR,
+	.name = "xen/hypercall",
+	.fops = &xen_privcmdbuf_fops,
+};