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, 117 insertions, 0 deletions

diff --git a/lib/kunit/user_alloc.c b/lib/kunit/user_alloc.c
new file mode 100644
index 000000000..b8cac765e
--- /dev/null
+++ b/lib/kunit/user_alloc.c
@@ -0,0 +1,117 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * KUnit userspace memory allocation resource management.
+ */
+#include <kunit/resource.h>
+#include <kunit/test.h>
+#include <linux/kthread.h>
+#include <linux/mm.h>
+
+struct kunit_vm_mmap_resource {
+	unsigned long addr;
+	size_t size;
+};
+
+/* vm_mmap() arguments */
+struct kunit_vm_mmap_params {
+	struct file *file;
+	unsigned long addr;
+	unsigned long len;
+	unsigned long prot;
+	unsigned long flag;
+	unsigned long offset;
+};
+
+int kunit_attach_mm(void)
+{
+	struct mm_struct *mm;
+
+	if (current->mm)
+		return 0;
+
+	/* arch_pick_mmap_layout() is only sane with MMU systems. */
+	if (!IS_ENABLED(CONFIG_MMU))
+		return -EINVAL;
+
+	mm = mm_alloc();
+	if (!mm)
+		return -ENOMEM;
+
+	/* Define the task size. */
+	mm->task_size = TASK_SIZE;
+
+	/* Make sure we can allocate new VMAs. */
+	arch_pick_mmap_layout(mm, &current->signal->rlim[RLIMIT_STACK]);
+
+	/* Attach the mm. It will be cleaned up when the process dies. */
+	kthread_use_mm(mm);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(kunit_attach_mm);
+
+static int kunit_vm_mmap_init(struct kunit_resource *res, void *context)
+{
+	struct kunit_vm_mmap_params *p = context;
+	struct kunit_vm_mmap_resource vres;
+	int ret;
+
+	ret = kunit_attach_mm();
+	if (ret)
+		return ret;
+
+	vres.size = p->len;
+	vres.addr = vm_mmap(p->file, p->addr, p->len, p->prot, p->flag, p->offset);
+	if (!vres.addr)
+		return -ENOMEM;
+	res->data = kmemdup(&vres, sizeof(vres), GFP_KERNEL);
+	if (!res->data) {
+		vm_munmap(vres.addr, vres.size);
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
+static void kunit_vm_mmap_free(struct kunit_resource *res)
+{
+	struct kunit_vm_mmap_resource *vres = res->data;
+
+	/*
+	 * Since this is executed from the test monitoring process,
+	 * the test's mm has already been torn down. We don't need
+	 * to run vm_munmap(vres->addr, vres->size), only clean up
+	 * the vres.
+	 */
+
+	kfree(vres);
+	res->data = NULL;
+}
+
+unsigned long kunit_vm_mmap(struct kunit *test, struct file *file,
+			    unsigned long addr, unsigned long len,
+			    unsigned long prot, unsigned long flag,
+			    unsigned long offset)
+{
+	struct kunit_vm_mmap_params params = {
+		.file = file,
+		.addr = addr,
+		.len = len,
+		.prot = prot,
+		.flag = flag,
+		.offset = offset,
+	};
+	struct kunit_vm_mmap_resource *vres;
+
+	vres = kunit_alloc_resource(test,
+				    kunit_vm_mmap_init,
+				    kunit_vm_mmap_free,
+				    GFP_KERNEL,
+				    &params);
+	if (vres)
+		return vres->addr;
+	return 0;
+}
+EXPORT_SYMBOL_GPL(kunit_vm_mmap);
+
+MODULE_IMPORT_NS("EXPORTED_FOR_KUNIT_TESTING");