mlxbf-gige: Fix kernel panic after reboot

[Asmaa Mnebhi]

SRU Justification:

[Impact]

Previously we fixed 2022370 by modifying the shutdown() function
but that creates a bug which needs to be reverted. It doesnt deinitialize the oob_net0 interface properly after reboot and causes issues for the UEFI OOB driver.

[Fix]

* Revert 2022370 changes
* Instead, just check that "priv" is not NULL.

[Test Case]

* Run the reboot test 1000 times
* Check that there is no kernel panic after reboot
* Check that the UEFI OOB interface gets an ip

[Regression Potential]

* We could still see a kernel panic since it is very intermittent. QA reproduced it once every 350 reboots.

[Bartlomiej Zolnierkiewicz (bzolnier)]

This bug is awaiting verification that the linux-bluefield/5.15.0-1022.24 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-jammy' to 'verification-done-jammy'. If the problem still exists, change the tag 'verification-needed-jammy' to 'verification-failed-jammy'.

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!

[Feysel Mohammed (feyselm)]

After performing 1000 reboots, oob interface has an ip and no kernel panic have been discovered.

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

This bug is awaiting verification that the linux-bluefield/5.4.0-1070.76 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-linux-bluefield' to 'verification-done-focal-linux-bluefield'. If the problem still exists, change the tag 'verification-needed-focal-linux-bluefield' to 'verification-failed-focal-linux-bluefield'.

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!

[Feysel Mohammed (feyselm)]

After performing 1000 reboots, oob interface has an ip and no kernel panic have been discovered.

[Launchpad Janitor (janitor)]

This bug was fixed in the package linux-bluefield - 5.4.0-1070.76

---------------
linux-bluefield (5.4.0-1070.76) focal; urgency=medium

  * focal/linux-bluefield: 5.4.0-1070.76 -proposed tracker (LP: #2030986)

  * EFI pstore not cleared on boot (LP: #1978079)
    - [Config] bluefield: Enable CONFIG_EFI_VARS_PSTORE as build-in
    - [Config] bluefield: Remove efi-pstore from modules list

  * Focal update: v5.4.246 upstream stable release (LP: #2028981)
    - [Config] bluefield: updateconfigs for SCSI_DPT_I2O
    - SAUCE: xfrm: Check if_id in xfrm full offload

  * rshim console truncates dmesg output due to tmfifo issue (LP: #2028197)
    - SAUCE: mlxbf-tmfifo.c: Fix rhsim console w/ truncated dmesg output

  * mlxbf-gige: Fix kernel panic after reboot (LP: #2030765)
    - SAUCE: mlxbf-gige: Fix kernel panic after reboot (part 1/2)
    - SAUCE: mlxbf-gige: Fix kernel panic after reboot (part 2/2)

  [ Ubuntu: 5.4.0-162.179 ]

  * focal/linux: 5.4.0-162.179 -proposed tracker (LP: #2031128)
  * libgnutls report "trap invalid opcode" when trying to install packages over
    https (LP: #2031093)
    - [Config]: disable CONFIG_GDS_FORCE_MITIGATION

  [ Ubuntu: 5.4.0-160.177 ]

  * focal/linux: 5.4.0-160.177 -proposed tracker (LP: #2031017)
  * Fix boot test warning for log_check "CPU: 0 PID: 0 at
    arch/x86/kernel/fpu/xstate.c:878 get_xsave_addr+0x98/0xb0" (LP: #2031022)
    - x86/pkeys: Revert a5eff7259790 ("x86/pkeys: Add PKRU value to init_fpstate")

  [ Ubuntu: 5.4.0-158.175 ]

  * focal/linux: 5.4.0-158.175 -proposed tracker (LP: #2030466)
  * CVE-2022-40982
    - x86/mm: Initialize text poking earlier
    - x86/mm: fix poking_init() for Xen PV guests
    - x86/mm: Use mm_alloc() in poking_init()
    - mm: Move mm_cachep initialization to mm_init()
    - init: Provide arch_cpu_finalize_init()
    - x86/cpu: Switch to arch_cpu_finalize_init()
    - ARM: cpu: Switch to arch_cpu_finalize_init()
    - sparc/cpu: Switch to arch_cpu_finalize_init()
    - um/cpu: Switch to arch_cpu_finalize_init()
    - init: Remove check_bugs() leftovers
    - init: Invoke arch_cpu_finalize_init() earlier
    - init, x86: Move mem_encrypt_init() into arch_cpu_finalize_init()
    - x86/fpu: Remove cpuinfo argument from init functions
    - x86/fpu: Mark init functions __init
    - x86/fpu: Move FPU initialization into arch_cpu_finalize_init()
    - x86/xen: Fix secondary processors' FPU initialization
    - x86/speculation: Add Gather Data Sampling mitigation
    - x86/speculation: Add force option to GDS mitigation
    - x86/speculation: Add Kconfig option for GDS
    - KVM: Add GDS_NO support to KVM
    - Documentation/x86: Fix backwards on/off logic about YMM support
    - [Config]: Enable CONFIG_ARCH_HAS_CPU_FINALIZE_INIT and
      CONFIG_GDS_FORCE_MITIGATION
  * CVE-2023-3609
    - net/sched: cls_u32: Fix reference counter leak leading to overflow
  * CVE-2023-20593
    - x86/cpu/amd: Move the errata checking functionality up
    - x86/cpu/amd: Add a Zenbleed fix
  * CVE-2023-3611
    - net/sched: sch_qfq: account for stab overhead in qfq_enqueue
  * stacked overlay file system mounts that have chroot() called against them
    appear to be getting locked (...