aarch64 内核配置开KASAN，宕机出vmcore，crash工具无法正常显示bt

【标题描述】内核配置开KASAN，宕机出vmcore，crash工具无法正常显示bt
【环境信息】
硬件信息：
1）	裸机场景提供出问题的硬件信息；
2）	虚机场景提供虚机XML文件或者配置信息
硬件无关，aarch64架构
鲲鹏920/飞腾S2500/虚拟机均可复现

软件信息：
1）	openEuler-20.03-LTS-SP3  aarch64
2）	kernel-4.19.90-2203.1.0.0139.oe1.aarch64  内核配置开启 CONFIG_KASAN=y 重编译
3）	crash-7.2.8-4.oe1.aarch64
如果有特殊组网，请提供网络拓扑图
【问题复现步骤】
1. aarch64 内核配置开CONFIG_KASAN=y，重编内核rpm包
2. 使用kasan内核启动设备，开启kdump服务
3. 主动触发内核宕机
   echo c >/proc/sysrq-trigger 
4. 待重启后使用crash工具加载 vmlinux 分析宕机vmcore
5. crash工具中使用bt查看触发宕机的线程栈，无法正常显示

必现
【预期结果】
crash> bt
PID: 1465   TASK: ffff80015c100000  CPU: 3   COMMAND: "bash"
 #0 [ffff800102e675e0] machine_kexec at ffff2000200c0460
 #1 [ffff800102e676b0] __crash_kexec at ffff20002031c548
 #2 [ffff800102e67860] crash_kexec at ffff20002031c688
 #3 [ffff800102e67890] die at ffff20002009a214
 #4 [ffff800102e678e0] die_kernel_fault at ffff2000200cdfec
 #5 [ffff800102e67910] __do_kernel_fault at ffff2000200cdbf8
 #6 [ffff800102e67940] do_translation_fault at ffff200021c64060
 #7 [ffff800102e67970] do_mem_abort at ffff20002008155c
 #8 [ffff800102e67bb0] el1_ia at ffff20002008390c
     PC: ffff2000210fa3b0  [sysrq_handle_crash+88]
     LR: ffff2000210fb708  [__handle_sysrq+320]
     SP: ffff800102e67bc0  PSTATE: 80400005
    X29: ffff800102e67bc0  X28: 1ffff0002b107d14  X27: ffff80015883e8a0
    X26: ffff800122118388  X25: 1ffff0002b107d18  X24: ffff200023435000
    X23: 0000000000000063  X22: 0000000000000004  X21: 0000000000000000
    X20: ffff200023154000  X19: ffff20002310f000  X18: 0000000000000000
    X17: 0000000000000000  X16: ffff2000201a5d30  X15: 0000000000000000
    X14: 0000000000000000  X13: 1ffff0002842cde1  X12: 0000000000000000
    X11: 0000000000000000  X10: 0000000000000001   X9: 1ffff0002ab30024
     X8: 0000000000000000   X7: ffff20002420a420   X6: 0000000000000000
     X5: 0000000000000001   X4: 0000000000000000   X3: 1fffe40004621fbc
     X2: 0000000000000004   X1: 0000000000000000   X0: dfff200000000000
 #9 [ffff800102e67bc0] sysrq_handle_crash at ffff2000210fa3ac
#10 [ffff800102e67be0] __handle_sysrq at ffff2000210fb704
#11 [ffff800102e67c50] write_sysrq_trigger at ffff2000210fc5d0
#12 [ffff800102e67c80] proc_reg_write at ffff2000207ccbe4
#13 [ffff800102e67cc0] __vfs_write at ffff200020677ee8
#14 [ffff800102e67d00] vfs_write at ffff200020678360
#15 [ffff800102e67d60] ksys_write at ffff200020678a80
#16 [ffff800102e67e20] __arm64_sys_write at ffff200020678bf4
#17 [ffff800102e67e50] el0_svc_common at ffff2000200a9a90
#18 [ffff800102e67ea0] el0_svc_handler at ffff2000200a9c5c
#19 [ffff800102e67ff0] el0_svc at ffff200020084884
     PC: 0000ffff187d7dd4   LR: 0000ffff18782cbc   SP: 0000ffffc9f8ab60
    X29: 0000ffffc9f8ab60  X28: 0000000000000000  X27: 0000aaab3d094130
    X26: 0000aaab3d0f7000  X25: 0000ffff18891630  X24: 0000000000000002
    X23: 0000aaab76bc8f10  X22: 0000000000000002  X21: 0000ffff18891548
    X20: 0000aaab76bc8f10  X19: 0000000000000001  X18: 0000ffffc9f89a4e
    X17: 0000ffff1877f5d0  X16: 0000aaab3d13e828  X15: 0000000000000003
    X14: 0000000000000000  X13: 0000000000000000  X12: 0000000000000020
    X11: 0000ffffc9f8ab08  X10: 0000000000000000   X9: 0000ffff187832e0
     X8: 0000000000000040   X7: 0000000000000001   X6: 0000000000000063
     X5: 0000000155510004   X4: 0000000000000000   X3: 0000ffff18895a70
     X2: 0000000000000002   X1: 0000aaab76bc8f10   X0: 0000000000000001
    ORIG_X0: 0000000000000001  SYSCALLNO: 40  PSTATE: 20000000
【实际结果】
crash> bt
PID: 1465   TASK: ffff80015c100000  CPU: 3   COMMAND: "bash"
 #0 [ffff2000224b0cb0] machine_kexec_prepare at ffff2000200bff4c
     PC: 0000ffff187d7dd4   LR: 0000ffff18782cbc   SP: 0000ffffc9f8ab60
    X29: 0000ffffc9f8ab60  X28: 0000000000000000  X27: 0000aaab3d094130
    X26: 0000aaab3d0f7000  X25: 0000ffff18891630  X24: 0000000000000002
    X23: 0000aaab76bc8f10  X22: 0000000000000002  X21: 0000ffff18891548
    X20: 0000aaab76bc8f10  X19: 0000000000000001  X18: 0000ffffc9f89a4e
    X17: 0000ffff1877f5d0  X16: 0000aaab3d13e828  X15: 0000000000000003
    X14: 0000000000000000  X13: 0000000000000000  X12: 0000000000000020
    X11: 0000ffffc9f8ab08  X10: 0000000000000000   X9: 0000ffff187832e0
     X8: 0000000000000040   X7: 0000000000000001   X6: 0000000000000063
     X5: 0000000155510004   X4: 0000000000000000   X3: 0000ffff18895a70
     X2: 0000000000000002   X1: 0000aaab76bc8f10   X0: 0000000000000001
    ORIG_X0: 0000000000000001  SYSCALLNO: 40  PSTATE: 20000000
crash> 
【附件信息】
x86 正常内核与KASAN内核 bt显示无问题
aarch64 正常内核 bt显示无问题
aarch64 KASAN内核 bt无法正常回溯堆栈

此问题已有相关补丁，正在验证修复
https://github.com/crash-utility/crash/commit/1a1fd21c625cb2ca335e626eb50426f13c4160f7

不是同一个问题，你发的patch修的是段错误；当前issue是bt只显示一层栈，而且是错的栈
此问题的补丁正在推upstream
https://listman.redhat.com/archives/crash-utility/2022-December/010288.html

补丁已进入upstream
https://github.com/crash-utility/crash/commit/5f27639196c3240810fbf30d367da0063a6612ff

src-openEuler/crash

内容风险标识

评论 (4)

src-openEuler/crash .gitee-modal { width: 500px !important; }

内容风险标识