Re: [PATCH v3 0/3] blk-mq: Don't complete in IRQ, use llist_head

2021-02-10 Thread Jens Axboe
On 1/23/21 1:10 PM, Sebastian Andrzej Siewior wrote: > Patch 2+3 were applied and then dropped by Jens due to a NOHZ+softirq > related warning [0]. Turns out a successful wakeup via > set_nr_if_polling() will not process any softirqs and the CPU may go > back to idle. This is addressed by patch #1.

Re: [PATCH v3 0/3] blk-mq: Don't complete in IRQ, use llist_head

2021-01-24 Thread Jens Axboe
On 1/23/21 1:10 PM, Sebastian Andrzej Siewior wrote: > Patch 2+3 were applied and then dropped by Jens due to a NOHZ+softirq > related warning [0]. Turns out a successful wakeup via > set_nr_if_polling() will not process any softirqs and the CPU may go > back to idle. This is addressed by patch #1.

[PATCH v3 0/3] blk-mq: Don't complete in IRQ, use llist_head

2021-01-23 Thread Sebastian Andrzej Siewior
Patch 2+3 were applied and then dropped by Jens due to a NOHZ+softirq related warning [0]. Turns out a successful wakeup via set_nr_if_polling() will not process any softirqs and the CPU may go back to idle. This is addressed by patch #1. smpcfd_dying_cpu() will also invoke SMP-functions calls via