An recursive locking in igb and igc drivers

[AceLan Kao]

[Impact]
The rtnl_lock() has been called in rtnetlink_rcv_msg(), and then in
__dev_open() it calls pm_runtime_resume() to resume devices, and in
some devices' resume function(igb_resum,igc_resume) they calls rtnl_lock()
again. That leads to a recursive lock.

This issue is introduced by below commit since v5.9-rc1
bd869245a3dc net: core: try to runtime-resume detached device in __dev_open

[Fix]
This commit currently is still under discussion fixes the issue.
https://lkml.org/lkml/2021/4/20/123

[Test]
Verified on Dell Precision workstations.

[Where problems could occur]
The patch is pretty straightforward, no problems could occur by this change.

[Ubuntu Kernel Bot (ubuntu-kernel-bot)]

This bug is awaiting verification that the kernel in -proposed solves the problem. Please test the kernel and update this bug with the results. If the problem is solved, change the tag 'verification-needed-focal' to 'verification-done-focal'. If the problem still exists, change the tag 'verification-needed-focal' to 'verification-failed-focal'.

If verification is not done by 5 working days from today, this fix will be dropped from the source code, and this bug will be closed.

See https://wiki.ubuntu.com/Testing/EnableProposed for documentation how to enable and use -proposed. Thank you!

[Launchpad Janitor (janitor)]

This bug was fixed in the package linux-oem-5.10 - 5.10.0-1029.30

---------------
linux-oem-5.10 (5.10.0-1029.30) focal; urgency=medium

  * focal/linux-oem-5.10: 5.10.0-1029.30 -proposed tracker (LP: #1930076)

  * CVE-2021-33200
    - bpf: Wrap aux data inside bpf_sanitize_info container
    - bpf: Fix mask direction swap upon off reg sign change
    - bpf: No need to simulate speculative domain for immediates

linux-oem-5.10 (5.10.0-1028.29) focal; urgency=medium

  * focal/linux-oem-5.10: 5.10.0-1028.29 -proposed tracker (LP: #1929167)

  * Packaging resync (LP: #1786013)
    - update dkms package versions

  * TGL-H system NV GPU fallen off the bus after resuming from s2idle with the
    external display connected via docking station (LP: #1929166)
    - SAUCE: ACPI: avoid NVIDIA GPU fallen with an _OSI string

  * AX201 BT will cause system could not enter S0i3 (LP: #1928047)
    - drm/i915: Tweaked Wa_14010685332 for all PCHs

  * Realtek USB hubs in Dell WD19SC/DC/TB fail to work after exiting s2idle
    (LP: #1928242)
    - USB: Verify the port status when timeout happens during port suspend
    - Revert "USB: Add reset-resume quirk for WD19's Realtek Hub"

  * Support mic-mute on Dell's platform (LP: #1928750)
    - ASoC: rt715: add main capture switch and main capture volume
    - ASoC: rt715: remove kcontrols which no longer be used
    - ASoC: rt715: modification for code simplicity
    - platform/x86: Move all dell drivers to their own subdirectory
    - SAUCE: platform/x86: dell-privacy: Add support for Dell hardware privacy
    - SAUCE: ASoC: rt715:add micmute led state control supports
    - [Config] Update configs for Dell's E-Privacy

linux-oem-5.10 (5.10.0-1027.28) focal; urgency=medium

  * focal/linux-oem-5.10: 5.10.0-1027.28 -proposed tracker (LP: #1927620)

  * Introduce the 465 driver series, fabric-manager, and libnvidia-nscq
    (LP: #1925522)
    - debian/dkms-versions -- add NVIDIA 465 and migrate 450 to 460

  * Packaging resync (LP: #1786013)
    - update dkms package versions

  * On TGL platforms screen shows garbage when browsing website by scrolling
    mouse (LP: #1926579)
    - SAUCE: drm/i915/display: Disable PSR2 if TGL Display stepping is B1 from A0

  * Add s2idle support on AMD Renoir and Cezanne (LP: #1927067)
    - drm/amd/display: setup system context in dm_init
    - drm/amd/display: add S/G support for Renoir
    - drm/amdgpu: drop extra drm_kms_helper_poll_enable/disable calls
    - drm/amdgpu: use runpm flag rather than fbcon for kfd runtime suspend (v2)
    - drm/amdgpu: reset runpm flag if device suspend fails
    - drm/amdgpu: add s0i3 capacity check for s0i3 routine (v2)
    - drm/amdgpu: add amdgpu_gfx_state_change_set() set gfx power change entry
      (v2)
    - drm/amdgpu: update amdgpu device suspend/resume sequence for s0i3 support
    - drm/amd/pm: add gfx_state_change_set() for rn gfx power switch (v2)
    - drm/amdgpu: add judgement for suspend/resume sequence
    - drm/amdgpu/pm: no need GPU status set since
      mmnbif_gpu_BIF_DOORBELL_FENCE_CNTL added in FSDL
    - drm/amdgpu: fix shutdown and poweroff process failed with s0ix
    - drm/amdgpu: Only check for S0ix if AMD_PMC ...

[Timo Aaltonen (tjaalton)]

this is not upstream as-is, but is there something else?

[AceLan Kao (acelankao)]

This issue has been fixed by below commit which is included in v5.16
ac8c58f5b535 igb: fix deadlock caused by taking RTNL in RPM resume path