Bug 2226558

Summary: bfq + losetup: loop devices disappear
Product: [Fedora] Fedora Reporter: Ed Santiago <santiago>
Component: kernelAssignee: Kernel Maintainer List <kernel-maint>
Status: NEW --- QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: high Docs Contact:
Priority: unspecified    
Version: rawhideCC: acaringi, adscvr, airlied, alciregi, bskeggs, hdegoede, hpa, jarodwilson, josef, kernel-maint, lgoncalv, linville, masami256, mchehab, ptalbert, steved
Target Milestone: ---   
Target Release: ---   
Hardware: Unspecified   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Attachments:
Description Flags
reproducer for bfq-losetup weirdness none

Description Ed Santiago 2023-07-25 23:56:46 UTC 
This is bizarre and I'm not 100% sure I have a true reproducer (see TBD attachment).

Basically: run losetup, use bfq scheduler on it, and future losetup commands cause "losetup -l" to emit no output.

Confirmed on kernels:
  6.4.0-0.rc7.20230623git8a28a0b6f1a1.57.fc39.x86_64
  6.5.0-0.rc2.20230721gitf7e3a1bafdea.20.fc39.x86_64

Reproducible: Always
Comment 1 Ed Santiago 2023-07-25 23:58:10 UTC 
Created attachment 1979984 [details]
reproducer for bfq-losetup weirdness
Comment 2 Ed Santiago 2023-07-26 00:00:46 UTC 
Originally found via:

  # dnf install podman-tests
  # bats /usr/share/podman/test/system/180-blkio.bats
  [works the first time, but gets skipped because /sys/fs/cgroup/blkio/system.slice/blkio.bfq.weight_device is missing ]
  # !!
  [ this time it fails because 'losetup -l' emits no output]