On 04/21/2017 11:39 AM, Bart Van Assche wrote: > On Fri, 2017-04-21 at 11:07 -0600, Jens Axboe wrote: >> Let me know what that bisection yields. > > Hello Jens, > > Although I still have to double-check the result, this is what git bisect > reported: > > b00c53e8f411a0b2be036c41852c6858541afbf7 is the first bad commit > commit b00c53e8f411a0b2be036c41852c6858541afbf7 > Author: Jens Axboe <[email protected]> > Date: Thu Apr 20 16:40:36 2017 -0600 > > blk-mq: fix schedule-while-atomic with scheduler attached > > I have attached my kernel config to this e-mail. Please let me know if you > need more information.
That would have been my prime suspect as well, since it is fiddling with the preempt counts. I'll double check the patch. -- Jens Axboe
