set_user_nice
SYSCALL_DEFINE1_nice!
binder_do_set_priority
call_usermodehelper_exec_async!
create_worker
khugepaged!
ksm_scan_thread!
loop_prepare_queue
normalize_rt_tasks
rescuer_thread!
set_one_prio
watchdog
SYSCALL_DEFINE3_setpriority!
binder_restore_priority
binder_set_priority
get_unbound_pool
loop_set_fd
maybe_create_worker
sysrq_handle_unrt!
task_tick_rt!
workqueue_init
workqueue_prepare_cpu!
alloc_unbound_pwq
binder_thread_read
binder_transaction
binder_transaction_priority
kernel_init_freeable
lo_ioctl
manage_workers
apply_wqattrs_prepare
binder_ioctl_write_read
binder_proc_transaction
binder_thread_read
binder_thread_write
integrity_read_file!
kernel_init!
lo_compat_ioctl!
worker_thread!
wq_update_unbound_numa
apply_workqueue_attrs_locked
binder_ioctl!
binder_ioctl_write_read
binder_transaction
workqueue_apply_unbound_cpumask
workqueue_init
workqueue_offline_cpu!
workqueue_online_cpu!
apply_workqueue_attrs
binder_ioctl!
binder_thread_write
kernel_init_freeable
workqueue_set_unbound_cpumask
wq_cpumask_store!
wq_nice_store!
wq_numa_store!
alloc_and_link_pwqs
binder_ioctl_write_read
integrity_read_file!
kernel_init!
wq_unbound_cpumask_store!
__alloc_workqueue_key
binder_ioctl!
__alloc_workqueue_key!
这个静态分析调用堆栈看起来就方便多了
一个空格表示一层
后面带！表示没有调用者了
从这个分析来看 binder workq loop设备 setpriority 会调整优先级，基本就了解哪些场景会触发调整优先级
