BUG: bfq scheduler results in soft lockup

Hey, I recently tried enabling the bfq scheduler by booting with scsi_mod.use_blk_mq=1 and selecting the scheduler using the following /etc/udev/rules.d/60-scheduler.rules: ACTION=="add|change", KERNEL=="sd[a-z]", ATTR{queue/scheduler}="bfq" I’m using an INTEL SSDSC2BP48 disk. The system boots

PROBLEM: kvm processes lock up when writing to LVM snapshots

Occasionally, my machine gets into a state where one or more kvm processes are locked up (in state D, not killable). I see the following in dmesg: [2432494.119477] INFO: task kvm:1823 blocked for more than 120 seconds. [2432494.119670] Not tainted 3.16.0-0.bpo.4-amd64 #1 [2432494.119821] "ec