The kernel hangs up where a task is calling throttle_direct_reclaim() and looping indefinitely

Issue

• A large number of blocked tasks appear:

INFO: task python3:1670971 blocked for more than 120 seconds.
      Tainted: G           OE     -------- -  - 4.18.0-553.el8_10.aarch64 #1
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
task:python3         state:D stack:0     pid:1670971 ppid:1667117 flags:0x00800080
Call trace:
 __switch_to+0xd0/0x120
 __schedule+0x340/0xac8
 schedule+0x68/0x118
 schedule_preempt_disabled+0x18/0x20
 rwsem_down_read_slowpath+0x2d4/0x4b8
 down_read+0x50/0xc0
 do_page_fault+0x120/0x3f0
 do_translation_fault+0xa0/0xb0
 do_mem_abort+0x54/0xb0
 el0_da+0x40/0x78
 el0t_64_sync_handler+0x60/0xb0
 el0t_64_sync+0x148/0x14c

• All of these blocked tasks have been in a long-term TASK_UNINTERRUPTIBLE state, waiting to acquire the mm->mmap_sem.
• The mmap_sem owner is calling throttle_direct_reclaim() and looping indefinitely, leading to the numerous blocked tasks and the severe system hangup.

        PID: 1667291  TASK: ffff00901a361100  CPU: 12   COMMAND: "python3"
         #0 [ffff80002cb6f8d0] __switch_to at ffff8000080095ac
         #1 [ffff80002cb6f900] __schedule at ffff800008abbd1c
         #2 [ffff80002cb6f990] schedule at ffff800008abc50c
         #3 [ffff80002cb6f9b0] throttle_direct_reclaim at ffff800008273550
         #4 [ffff80002cb6fa20] try_to_free_pages at ffff800008277b68
         #5 [ffff80002cb6fae0] __alloc_pages_nodemask at ffff8000082c4660
         #6 [ffff80002cb6fc50] alloc_pages_vma at ffff8000082e4a98
         #7 [ffff80002cb6fca0] do_anonymous_page at ffff80000829f5a8
         #8 [ffff80002cb6fce0] __handle_mm_fault at ffff8000082a5974
         #9 [ffff80002cb6fd90] handle_mm_fault at ffff8000082a5bd4
                ...