On 2019-01-04 15:33:21 [-0500], Scott Wood wrote: > + /* > + * If sem->rtmutex blocks, the function sched_submit_work will not > + * call blk_schedule_flush_plug (because tsk_is_pi_blocked would be > + * true). We must call blk_schedule_flush_plug here; if we don't > + * call it, an I/O deadlock may occur. > + */ I kept the original comment plus I replaced device-mapper with I/O (the patch said already that it is not device-mapper specific and may occur with ext4 as well). Sebastian
