From a5f7a5bb3b7f28ba7e4fa246775b29a0e5537255 Mon Sep 17 00:00:00 2001 From: Jinke Han Date: Tue, 8 Sep 2026 15:37:42 +0800 Subject: x86/kprobes: Fix crash when probing CS CALL instructions 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) Signed-off-by: Jinke Han Signed-off-by: Ingo Molnar Reviewed-by: Masami Hiramatsu (Google) Acked-by: Yafang Shao Acked-by: Borislav Petkov Cc: Peter Zijlstra Link: https://patch.msgid.link/20260908073742.GA10517@didi-ThinkCentre-M920t-N000 --- drivers/virt/vmgenid.c | 181 +++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 181 insertions(+) create mode 100644 drivers/virt/vmgenid.c (limited to 'drivers/virt/vmgenid.c') diff --git a/drivers/virt/vmgenid.c b/drivers/virt/vmgenid.c new file mode 100644 index 000000000..66135eac3 --- /dev/null +++ b/drivers/virt/vmgenid.c @@ -0,0 +1,181 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (C) 2022 Jason A. Donenfeld . All Rights Reserved. + * + * The "Virtual Machine Generation ID" is exposed via ACPI or DT and changes when a + * virtual machine forks or is cloned. This driver exists for shepherding that + * information to random.c. + */ + +#include +#include +#include +#include +#include +#include + +ACPI_MODULE_NAME("vmgenid"); + +enum { VMGENID_SIZE = 16 }; + +struct vmgenid_state { + u8 *next_id; + u8 this_id[VMGENID_SIZE]; +}; + +static void vmgenid_notify(struct device *device) +{ + struct vmgenid_state *state = device->driver_data; + u8 old_id[VMGENID_SIZE]; + + memcpy(old_id, state->this_id, sizeof(old_id)); + memcpy(state->this_id, state->next_id, sizeof(state->this_id)); + if (!memcmp(old_id, state->this_id, sizeof(old_id))) + return; + add_vmfork_randomness(state->this_id, sizeof(state->this_id)); +} + +static void setup_vmgenid_state(struct vmgenid_state *state, void *virt_addr) +{ + state->next_id = virt_addr; + memcpy(state->this_id, state->next_id, sizeof(state->this_id)); + add_device_randomness(state->this_id, sizeof(state->this_id)); +} + +#ifdef CONFIG_ACPI +static void vmgenid_acpi_handler(acpi_handle __always_unused handle, + u32 __always_unused event, void *dev) +{ + vmgenid_notify(dev); +} + +static int vmgenid_add_acpi(struct device *dev, struct vmgenid_state *state) +{ + struct acpi_device *device = ACPI_COMPANION(dev); + struct acpi_buffer parsed = { ACPI_ALLOCATE_BUFFER }; + union acpi_object *obj; + phys_addr_t phys_addr; + acpi_status status; + void *virt_addr; + int ret = 0; + + status = acpi_evaluate_object(device->handle, "ADDR", NULL, &parsed); + if (ACPI_FAILURE(status)) { + ACPI_EXCEPTION((AE_INFO, status, "Evaluating ADDR")); + return -ENODEV; + } + obj = parsed.pointer; + if (!obj || obj->type != ACPI_TYPE_PACKAGE || obj->package.count != 2 || + obj->package.elements[0].type != ACPI_TYPE_INTEGER || + obj->package.elements[1].type != ACPI_TYPE_INTEGER) { + ret = -EINVAL; + goto out; + } + + phys_addr = (obj->package.elements[0].integer.value << 0) | + (obj->package.elements[1].integer.value << 32); + + virt_addr = devm_memremap(&device->dev, phys_addr, VMGENID_SIZE, MEMREMAP_WB); + if (IS_ERR(virt_addr)) { + ret = PTR_ERR(virt_addr); + goto out; + } + setup_vmgenid_state(state, virt_addr); + + status = acpi_install_notify_handler(device->handle, ACPI_DEVICE_NOTIFY, + vmgenid_acpi_handler, dev); + if (ACPI_FAILURE(status)) { + ret = -ENODEV; + goto out; + } + + dev->driver_data = state; +out: + ACPI_FREE(parsed.pointer); + return ret; +} +#else +static int vmgenid_add_acpi(struct device *dev, struct vmgenid_state *state) +{ + return -EINVAL; +} +#endif + +static irqreturn_t vmgenid_of_irq_handler(int __always_unused irq, void *dev) +{ + vmgenid_notify(dev); + return IRQ_HANDLED; +} + +static int vmgenid_add_of(struct platform_device *pdev, + struct vmgenid_state *state) +{ + void *virt_addr; + int ret; + + virt_addr = devm_platform_get_and_ioremap_resource(pdev, 0, NULL); + if (IS_ERR(virt_addr)) + return PTR_ERR(virt_addr); + + setup_vmgenid_state(state, virt_addr); + + ret = platform_get_irq(pdev, 0); + if (ret < 0) + return ret; + + ret = devm_request_irq(&pdev->dev, ret, vmgenid_of_irq_handler, + IRQF_SHARED, "vmgenid", &pdev->dev); + if (ret < 0) + return ret; + + pdev->dev.driver_data = state; + return 0; +} + +static int vmgenid_add(struct platform_device *pdev) +{ + struct device *dev = &pdev->dev; + struct vmgenid_state *state; + int ret; + + state = devm_kmalloc(dev, sizeof(*state), GFP_KERNEL); + if (!state) + return -ENOMEM; + + if (dev->of_node) + ret = vmgenid_add_of(pdev, state); + else + ret = vmgenid_add_acpi(dev, state); + + if (ret < 0) + devm_kfree(dev, state); + return ret; +} + +static const struct of_device_id vmgenid_of_ids[] = { + { .compatible = "microsoft,vmgenid", }, + { }, +}; +MODULE_DEVICE_TABLE(of, vmgenid_of_ids); + +static const struct acpi_device_id vmgenid_acpi_ids[] = { + { "VMGENCTR", 0 }, + { "VM_GEN_COUNTER", 0 }, + { } +}; +MODULE_DEVICE_TABLE(acpi, vmgenid_acpi_ids); + +static struct platform_driver vmgenid_plaform_driver = { + .probe = vmgenid_add, + .driver = { + .name = "vmgenid", + .acpi_match_table = vmgenid_acpi_ids, + .of_match_table = vmgenid_of_ids, + }, +}; + +module_platform_driver(vmgenid_plaform_driver) + +MODULE_DESCRIPTION("Virtual Machine Generation ID"); +MODULE_LICENSE("GPL v2"); +MODULE_AUTHOR("Jason A. Donenfeld "); -- cgit v1.3.1