Re: [mm] e52271917f: BUG:sleeping_function_called_from_invalid_context_at_mm/slab.h

[Roman Gushchin]

On Thu, May 23, 2019 at 08:58:58AM +0800, kernel test robot wrote:
> FYI, we noticed the following commit (built with gcc-7):
> 
> commit: e52271917f9f5159c791eda8ba748a66d659c27e ("[PATCH v4 5/7] mm: rework non-root kmem_cache lifecycle management")
> url: https://github.com/0day-ci/linux/commits/Roman-Gushchin/mm-reparent-slab-memory-on-cgroup-removal/20190517-173841
> 
> 
> in testcase: nvml
> with following parameters:
> 
> 	group: obj
> 	test: non-pmem
> 
> 
> 
> on test machine: qemu-system-x86_64 -enable-kvm -cpu SandyBridge -smp 2 -m 8G
> 
> caused below changes (please refer to attached dmesg/kmsg for entire log/backtrace):
> 
> 
> +------------------------------------------------------------------------------+------------+------------+
> |                                                                              | ff756a15f3 | e52271917f |
> +------------------------------------------------------------------------------+------------+------------+
> | boot_successes                                                               | 5          | 4          |
> | boot_failures                                                                | 861        | 852        |
> | BUG:kernel_reboot-without-warning_in_test_stage                              | 738        | 163        |
> | BUG:kernel_hang_in_boot_stage                                                | 120        | 122        |
> | BUG:soft_lockup-CPU##stuck_for#s                                             | 4          | 1          |
> | RIP:free_unref_page                                                          | 1          |            |
> | Kernel_panic-not_syncing:softlockup:hung_tasks                               | 4          | 1          |
> | RIP:free_reserved_area                                                       | 3          | 1          |
> | BUG:sleeping_function_called_from_invalid_context_at_mm/slab.h               | 0          | 560        |
> | BUG:scheduling_while_atomic                                                  | 0          | 561        |
> | WARNING:at_lib/usercopy.c:#_copy_to_user                                     | 0          | 116        |
> | RIP:_copy_to_user                                                            | 0          | 116        |
> | WARNING:at_arch/x86/kernel/fpu/signal.c:#copy_fpstate_to_sigframe            | 0          | 534        |
> | RIP:copy_fpstate_to_sigframe                                                 | 0          | 532        |
> | WARNING:at_arch/x86/kernel/signal.c:#do_signal                               | 0          | 527        |
> | RIP:do_signal                                                                | 0          | 526        |
> | WARNING:at_lib/usercopy.c:#_copy_from_user                                   | 0          | 389        |
> | RIP:_copy_from_user                                                          | 0          | 388        |
> | kernel_BUG_at_mm/vmalloc.c                                                   | 0          | 304        |
> | invalid_opcode:#[##]                                                         | 0          | 304        |
> | RIP:__get_vm_area_node                                                       | 0          | 301        |
> | Kernel_panic-not_syncing:Fatal_exception_in_interrupt                        | 0          | 294        |
> | Kernel_panic-not_syncing:Aiee,killing_interrupt_handler                      | 0          | 155        |
> | WARNING:at_fs/read_write.c:#vfs_write                                        | 0          | 15         |
> | RIP:vfs_write                                                                | 0          | 15         |
> | BUG:sleeping_function_called_from_invalid_context_at_kernel/locking/rwsem.c  | 0          | 101        |
> | BUG:sleeping_function_called_from_invalid_context_at_include/linux/uaccess.h | 0          | 54         |
> | Kernel_panic-not_syncing:Attempted_to_kill_init!exitcode=                    | 0          | 47         |
> | BUG:sleeping_function_called_from_invalid_context_at_lib/iov_iter.c          | 0          | 1          |
> | BUG:sleeping_function_called_from_invalid_context_at_fs/dcache.c             | 0          | 57         |
> | BUG:sleeping_function_called_from_invalid_context_at_mm/memory.c             | 0          | 1          |
> | BUG:sleeping_function_called_from_invalid_context_at_kernel/locking/mutex.c  | 0          | 104        |
> | BUG:kernel_hang_in_test_stage                                                | 0          | 5          |
> | WARNING:at_arch/x86/include/asm/uaccess.h:#strncpy_from_user                 | 0          | 4          |
> | RIP:strncpy_from_user                                                        | 0          | 4          |
> | WARNING:at_fs/read_write.c:#vfs_read                                         | 0          | 4          |
> | RIP:vfs_read                                                                 | 0          | 4          |
> | BUG:sleeping_function_called_from_invalid_context_at_mm/filemap.c            | 0          | 3          |
> | BUG:sleeping_function_called_from_invalid_context_at_mm/page_alloc.c         | 0          | 8          |
> | BUG:sleeping_function_called_from_invalid_context_at_mm/gup.c                | 0          | 1          |
> | BUG:sleeping_function_called_from_invalid_context_at_include/linux/freezer.h | 0          | 1          |
> | BUG:sleeping_function_called_from_invalid_context_at/kb                      | 0          | 1          |
> +------------------------------------------------------------------------------+------------+------------+
> 
> 
> If you fix the issue, kindly add following tag
> Reported-by: kernel test robot <rong.a.chen@xxxxxxxxx>

Hi!

It seems that it's caused by unbalanced rcu_read_lock(),
which already has been fixed in v5.

Thanks!