社区专家好，我是华为四川鲲鹏生态创新中心的朱凯。现企业在开展openEuler适配 时，遇到以下issue问题，详情见链接： https://gitee.com/openeuler/community-issue/issues/I5YDUT?from=project-issue 。
请求社区专家给予技术支持。
简要如下：
【标题描述】在鲲鹏服务器（aarch64）上 qemu6.2 的colo 备机虚拟机异常退出 【环境信息】 硬件信息： 1）鲲鹏920 软件信息： 1） 系统版本为：openEuler-22.03-LTS-aarch64 2） 内核信息：Linux vServerOs-11 5.10.0-60.18.0.50.oe2203.aarch64 3） qemu-6.2.0-49.aarch64 【问题复现步骤】 具体操作步骤 出现概率 100%出现 【预期结果】 能正常建立COLO复制 【实际结果】 备机虚拟机断言后退出（x86_64的机器上没问题） 【附件信息】 qemu日志： Unexpected error in kvm_device_access() at ../accel/kvm/kvm-all.c:3121: 2022-10-29 15:28:01 qemu-kvm: KVM_SET_DEVICE_ATTR failed: Group 8 attr 0x0000000000000090: Device or resource busy 2022-10-29 07:28:02.408+0000: shutting down, reason=crashed
堆栈如下： (gdb) bt #0 0x0000ffff9da74bd0 in ?? () from /usr/lib64/libc.so.6 #1 0x0000ffff9da30bfc in raise () from /usr/lib64/libc.so.6 #2 0x0000ffff9da1ec70 in abort () from /usr/lib64/libc.so.6 #3 0x0000aaaad1a759a0 in error_handle_fatal (err=err@entry=0xfffd580f9620, errp=0xaaaad216fde8 <error_abort>) at ../util/error.c:40 #4 0x0000aaaad1a75e2c in error_handle_fatal (err=0xfffd580f9620, errp=0xaaaad216fde8 <error_abort>) at ../util/error.c:44 #5 error_setv (errp=errp@entry=0xaaaad216fde8 <error_abort>, src=src@entry=0xaaaad1c437b8 "../accel/kvm/kvm-all.c", line=line@entry=3121, func=func@entry=0xaaaad1c45000 <func.15> "kvm_device_access", err_class=err_class@entry=ERROR_CLASS_GENERIC_ERROR, fmt=fmt@entry=0xaaaad1c44a38 "KVM_%s_DEVICE_ATTR failed: Group %d attr 0x%016lx", ap=..., suffix=0xffff9db3a220 "Device or resource busy") at ../util/error.c:73 #6 0x0000aaaad1a76058 in error_setg_errno_internal (errp=errp@entry=0xaaaad216fde8 <error_abort>, src=src@entry=0xaaaad1c437b8 "../accel/kvm/kvm-all.c", line=line@entry=3121, func=func@entry=0xaaaad1c45000 <func.15> "kvm_device_access", os_errno=16, fmt=fmt@entry=0xaaaad1c44a38 "KVM_%s_DEVICE_ATTR failed: Group %d attr 0x%016lx") at ../util/error.c:109 #7 0x0000aaaad19144e4 in kvm_device_access (fd=, group=group@entry=8, attr=attr@entry=144, val=val@entry=0xaaaadc5c20d0, write=write@entry=true, errp=errp@entry=0xaaaad216fde8 <error_abort>) at ../accel/kvm/kvm-all.c:3121 #8 0x0000aaaad18541d8 in kvm_arm_its_post_load (s=0xaaaadc5c1a70) at ../hw/intc/arm_gicv3_its_kvm.c:177 #9 0x0000aaaad13e7498 in gicv3_its_post_load (opaque=0xaaaadc5c1a70, version_id=) at ../hw/intc/arm_gicv3_its_common.c:46 #10 0x0000aaaad12b7078 in vmstate_load_state (f=0xfffd5809c6d0, vmsd=0xaaaad1e93288 <vmstate_its>, opaque=0xaaaadc5c1a70, version_id=0) at ../migration/vmstate.c:168 #11 0x0000aaaad12d87fc in qemu_loadvm_section_start_full (f=f@entry=0xfffd5809c6d0, mis=0xaaaadc39de20) at ../migration/savevm.c:2506 #12 0x0000aaaad12dbfc8 in qemu_loadvm_state_main (f=f@entry=0xfffd5809c6d0, mis=0xaaaadc39de20) at ../migration/savevm.c:2741 #13 0x0000aaaad12dd920 in qemu_load_device_state (f=f@entry=0xfffd5809c6d0) at ../migration/savevm.c:2913 #14 0x0000aaaad12c360c in colo_incoming_process_checkpoint (errp=0xfffe7eafc6b8, history_device_state_buff=0x0, bioc=0xfffd58039410, fb=0xfffd5809c6d0, mis=0xaaaadc39de20) at ../migration/colo.c:1376 #15 colo_wait_handle_message (errp=0xfffe7eafc6b8, history_device_state_buff=0x0, bioc=0xfffd58039410, fb=0xfffd5809c6d0, mis=0xaaaadc39de20) at ../migration/colo.c:1490 #16 colo_process_incoming_thread (opaque=opaque@entry=0xaaaadc39de20) at ../migration/colo.c:1637 #17 0x0000aaaad1a72258 in qemu_thread_start (args=) at ../util/qemu-thread-posix.c:556
