Hi, Andrew and Kirill, Sorry for late! It's a shame that I misunderstood your mails. @Kirill I'm very sorry for misunderstood before! Now, I got your points. Yes, the v1 patch is buggy in preemption case. I also tried your suggestion, it's works! Please help to review the new patch again. Thanks very much for you kindly guidance! Shile Changelog: v1: https://lore.kernel.org/lkml/20200110082510.172517-1-shile.zhang@xxxxxxxxxxxxxxxxx/ Shile Zhang (1): mm: fix interrupt disabled long time inside deferred_init_memmap() mm/page_alloc.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) -- 2.24.0.rc2
