On 7/18/24 6:21 PM, devel-request@lists.crash-utility.osci.io wrote:

> Date: Thu, 18 Jul 2024 07:26:02 -0000
> From:qiwu.chen@transsion.com
> Subject: [Crash-utility] Re: [PATCH] arm64: fix a potential segfault
>       in arm64_unwind_frame
> To:devel@lists.crash-utility.osci.io
> Message-ID:<20240718072602.21739.62283@lists.crash-utility.osci.io>
> Content-Type: text/plain; charset="utf-8"
>
> Hi Lianbo,
>
> 1. The current issue can be reproduced with arm64_unwind_frame_v2():

Thank you for the confirmation, qiwu.

If so, the same changes should be done in the arm64_unwind_frame_v2().
What do you think?

Thanks

Lianbo


> crash> bt
> [Detaching after fork from child process 4778]
>
> Thread 1 "crash" received signal SIGSEGV, Segmentation fault.
> 0x0000555555826dae in arm64_unwind_frame_v2 (bt=0x7fffffffd8f0,
> frame=0x7fffffffd060, ofp=0x555559909970) at arm64.c:3048
> 3048            frame->pc = GET_STACK_ULONG(fp + 8);
> (gdb) bt
> #0  0x0000555555826dae in arm64_unwind_frame_v2 (bt=0x7fffffffd8f0,
> frame=0x7fffffffd060, ofp=0x555559909970) at arm64.c:3048
> #1  0x0000555555827d99 in arm64_back_trace_cmd_v2 (bt=0x7fffffffd8f0) at
>   arm64.c:3426
> #2  0x00005555557df95e in back_trace (bt=0x7fffffffd8f0) at
> kernel.c:3240
> #3  0x00005555557dd8b8 in cmd_bt () at kernel.c:2881
> #4  0x000055555573696b in exec_command () at main.c:893
> #5  0x000055555573673e in main_loop () at main.c:840
> #6  0x0000555555aa4a61 in captured_main (data=<optimized out>) at
> main.c:1284
> #7  gdb_main (args=<optimized out>) at main.c:1313
> #8  0x0000555555aa4ae0 in gdb_main_entry (argc=<optimized out>,
> argv=<optimized out>) at main.c:1338
> #9  0x00005555558021df in gdb_main_loop (argc=2, argv=0x7fffffffe248) at
>   gdb_interface.c:81
> #10 0x0000555555736401 in main (argc=3, argv=0x7fffffffe248) at
> main.c:721
> (gdb) p/x *(struct arm64_stackframe *)0x7fffffffd060
> $1 = {fp = 0xffffffc008003f50, sp = 0xffffffc008003f40, pc =
> 0xffffffdfd669447c}
> (gdb) p/x *(struct bt_info *)0x7fffffffd8f0
> $2 = {task = 0xffffff8118012500, flags = 0x0, instptr =
> 0xffffffdfd669447c, stkptr = 0xffffffc008003f40, bptr = 0x0, stackbase =
>   0xffffffc01b5b0000, stacktop = 0xffffffc01b5b4000,
>    stackbuf = 0x555556117a80, tc = 0x55557a3b3480, hp = 0x0, textlist =
> 0x0, ref = 0x0, frameptr = 0xffffffc008003f50, call_target = 0x0,
> machdep = 0x0, debug = 0x0, eframe_ip = 0x0, radix = 0x0,
>    cpumask = 0x0}
>
> 2. The issue can be easily reproduced by "echo c >
> /proc/sysrq-trigger" on Andriod GKI-5.10 platform.
> >From the reproduced dump we can see, the current fp/sp of crashing
> cpu1 is out of range task's stack, but located in the irq stack of cpu0
>       KERNEL: vmlinux  [TAINTED]
>      DUMPFILE: SYS_COREDUMP
>          CPUS: 8 [OFFLINE: 7]
>     MACHINE: aarch64  (unknown Mhz)
>        MEMORY: 8 GB
>      PANIC: "Kernel panic - not syncing: sysrq triggered crash"
>           PID: 9089
>       COMMAND: "sh"
>          TASK: ffffff8118012500  [THREAD_INFO: ffffff8118012500]
>           CPU: 1
>         STATE: TASK_RUNNING (PANIC)
> crash> help -m |grep irq
>          irq_stack_size: 16384
>           irq_stacks[0]: ffffffc008000000
>           irq_stacks[1]: ffffffc008008000
>           irq_stacks[2]: ffffffc008010000
>           irq_stacks[3]: ffffffc008018000
>           irq_stacks[4]: ffffffc008020000
>           irq_stacks[5]: ffffffc008028000
>           irq_stacks[6]: ffffffc008030000
>           irq_stacks[7]: ffffffc008038000
> crash> task_struct.thread -x  ffffff8118012500
>    thread = {
>      cpu_context = {
>        x19 = 0xffffff80c01ea500,
>        x20 = 0xffffff8118012500,
>        x21 = 0xffffff8118012500,
>        x22 = 0xffffff80c01ea500,
>        x23 = 0xffffff8118012500,
>        x24 = 0xffffff81319ac270,
>        x25 = 0xffffffdfd8f87000,
>        x26 = 0xffffff8118012500,
>        x27 = 0xffffffdfd88ea180,
>        x28 = 0xffffffdfd7e1b4b8,
>        fp = 0xffffffc01b5b3a10,
>        sp = 0xffffffc01b5b3a10,
>        pc = 0xffffffdfd667b89c
>      },
> crash> bt -S 0xffffffc01b5b3a10
> PID: 9089     TASK: ffffff8118012500  CPU: 1    COMMAND: "sh"
>   #0 [ffffffc008003f50] local_cpu_stop at ffffffdfd6694478
> crash> bt -S ffffffc008003f50
> PID: 9089     TASK: ffffff8118012500  CPU: 1    COMMAND: "sh"
> bt: non-process stack address for this task: ffffffc008003f50
>      (valid range: ffffffc01b5b0000 - ffffffc01b5b4000)
>
> The second frame begins to switch to irq stack of cpu0.
> crash> rd ffffffc008003f50 2
> ffffffc008003f50:  ffffffc008003f70 ffffffdfd68125d0   p?.......%......
> crash> dis -x ffffffdfd68125d0
> 0xffffffdfd68125d0 <handle_percpu_devid_fasteoi_ipi+0xb0>:      b
>        0xffffffdfd6812730 <handle_percpu_devid_fasteoi_ipi+0x210>
> crash> rd ffffffc008003f70 2
> ffffffc008003f70:  ffffffc008003fb0 ffffffdfd680352c   .?......,5......
> crash> dis ffffffdfd680352c -x
> 0xffffffdfd680352c <__handle_domain_irq+0x114>: bl
> 0xffffffdfd673d204 <__irq_exit_rcu>
> crash> rd ffffffc008003fb0 2
> ffffffc008003fb0:  ffffffc008003fe0 ffffffdfd6610380   .?........a.....
> crash> dis -x ffffffdfd6610380
> 0xffffffdfd6610380 <gic_handle_irq.30555+0x6c>: cbz     w0,
> 0xffffffdfd6610348 <gic_handle_irq.30555+0x34>
> crash> rd ffffffc008003fe0 2
> ffffffc008003fe0:  ffffffdfd8d83e20 ffffffdfd6612624
> >......$&a.....
> crash> dis -x ffffffdfd6612624
> 0xffffffdfd6612624 <el1_irq+0xe4>:      mov     sp, x19
> crash> rd ffffffdfd8d83e20 2
> ffffffdfd8d83e20:  ffffffdfd8d83e80 ffffffdfd768c690
> .>........h.....
> crash> dis -x ffffffdfd768c690
> 0xffffffdfd768c690 <cpuidle_enter_state+0x3a4>: tbnz    w19, #31,
> 0xffffffdfd768c720 <cpuidle_enter_state+0x434>
> crash> rd ffffffdfd8d83e80 2
> ffffffdfd8d83e80:  ffffffdfd8d83ef0 ffffffdfd67ab4f4
> .>........z.....
> crash> dis -x ffffffdfd67ab4f4
> 0xffffffdfd67ab4f4 <do_idle+0x308>:     str     xzr, [x19, #8]
> crash> rd ffffffdfd8d83ef0 2
> ffffffdfd8d83ef0:  ffffffdfd8d83f50 ffffffdfd67ab7e4   P?........z.....
> crash> dis -x ffffffdfd67ab7e4
> 0xffffffdfd67ab7e4 <cpu_startup_entry+0x84>:    b
> 0xffffffdfd67ab7e0 <cpu_startup_entry+0x80>
>
> It's unreasonable cpu1 is in cpu0's irq context, which is far away from
> the backtrace showed by "bt -T", so we must avoid this case.
> crash> bt -T
> PID: 9089     TASK: ffffff8118012500  CPU: 1    COMMAND: "sh"
>    [ffffffc01b5b3238] vsnprintf at ffffffdfd7075c10
>    [ffffffc01b5b32b8] sprintf at ffffffdfd707b9e4
>    [ffffffc01b5b3398] __sprint_symbol at ffffffdfd68abff4
>    [ffffffc01b5b33c8] symbol_string at ffffffdfd70774ac
>    [ffffffc01b5b33d8] symbol_string at ffffffdfd7077510
>    [ffffffc01b5b34c8] string at ffffffdfd70767a8
>    [ffffffc01b5b34d8] vsnprintf at ffffffdfd7075c2c
>    [ffffffc01b5b34e8] vsnprintf at ffffffdfd7075fdc
>    [ffffffc01b5b3518] vscnprintf at ffffffdfd707b8b4
>    [ffffffc01b5b3558] ktime_get_ts64 at ffffffdfd686a2f8
>    [ffffffc01b5b3598] data_alloc at ffffffdfd68009b4
>    [ffffffc01b5b35d8] prb_reserve at ffffffdfd68011b4
>    [ffffffc01b5b35e8] prb_reserve at ffffffdfd68010a0
>    [ffffffc01b5b3648] log_store at ffffffdfd67fb024
>    [ffffffc01b5b3698] number at ffffffdfd7076ea4
>    [ffffffc01b5b36d8] number at ffffffdfd7076ea4
>    [ffffffc01b5b3738] vsnprintf at ffffffdfd7075c10
>    [ffffffc01b5b3778] number at ffffffdfd7076ea4
>    [ffffffc01b5b37c8] number at ffffffdfd7076ea4
>    [ffffffc01b5b3828] vsnprintf at ffffffdfd7075c10
>    [ffffffc01b5b3868] vsnprintf at ffffffdfd7075c2c
>    [ffffffc01b5b3888] number at ffffffdfd7076ea4
>    [ffffffc01b5b38e8] vsnprintf at ffffffdfd7075c10
>    [ffffffc01b5b3928] vsnprintf at ffffffdfd7075c2c
>    [ffffffc01b5b3968] aee_nested_printf at ffffffdfd3d05d7c [mrdump]
>    [ffffffc01b5b3a48] mrdump_common_die at ffffffdfd3d05a98 [mrdump]
>    [ffffffc01b5b3ac8] ipanic at ffffffdfd3d06078 [mrdump]
>    [ffffffc01b5b3ae8] __typeid__ZTSFiP14notifier_blockmPvE_global_addr at
>   ffffffdfd7e3c118
>    [ffffffc01b5b3af0] ipanic.cfi_jt at ffffffdfd3d0ab40 [mrdump]
>    [ffffffc01b5b3b18] atomic_notifier_call_chain at ffffffdfd678236c
>    [ffffffc01b5b3b28] panic at ffffffdfd672aa28
>    [ffffffc01b5b3bf8] rcu_read_unlock.34874 at ffffffdfd718ddfc
>    [ffffffc01b5b3c58] __handle_sysrq at ffffffdfd718d3c0
>    [ffffffc01b5b3c68] write_sysrq_trigger at ffffffdfd718ea20
>    [ffffffc01b5b3cb0] __typeid__ZTSFlP4filePKcmPxE_global_addr at
> ffffffdfd7e2fa88
>    [ffffffc01b5b3cc8] proc_reg_write at ffffffdfd6c56f44
>    [ffffffc01b5b3cd8] file_start_write at ffffffdfd6b3f884
>    [ffffffc01b5b3d08] vfs_write at ffffffdfd6b403b4
>    [ffffffc01b5b3da8] ksys_write at ffffffdfd6b40240
>    [ffffffc01b5b3df8] __arm64_sys_write at ffffffdfd6b401b4
>    [ffffffc01b5b3e20] __typeid__ZTSFlPK7pt_regsE_global_addr at
> ffffffdfd7e25240
>    [ffffffc01b5b3e38] el0_svc_common at ffffffdfd6695980
>    [ffffffc01b5b3e48] el0_da at ffffffdfd7d43e30
>    [ffffffc01b5b3e58] el0_svc at ffffffdfd7d43d9c
>    [ffffffc01b5b3e98] el0_sync_handler at ffffffdfd7d43d10
>    [ffffffc01b5b3ea8] el0_sync at ffffffdfd66128b8
>
> Thanks