CVE-2024-38608 kernel: net/mlx5e: Fix netif state handling

JIRA: https://issues.redhat.com/browse/RHEL-43870
CVE: CVE-2024-38608

commit 3d5918477f94e4c2f064567875c475468e264644
Author: Shay Drory <shayd@nvidia.com>
Date:   Thu May 9 14:29:47 2024 +0300

    net/mlx5e: Fix netif state handling

    mlx5e_suspend cleans resources only if netif_device_present() returns
    true. However, mlx5e_resume changes the state of netif, via
    mlx5e_nic_enable, only if reg_state == NETREG_REGISTERED.
    In the below case, the above leads to NULL-ptr Oops[1] and memory
    leaks:

    mlx5e_probe
     _mlx5e_resume
      mlx5e_attach_netdev
       mlx5e_nic_enable  <-- netdev not reg, not calling netif_device_attach()
      register_netdev <-- failed for some reason.
    ERROR_FLOW:
     _mlx5e_suspend <-- netif_device_present return false, resources aren't freed :(

    Hence, clean resources in this case as well.

    [1]
    BUG: kernel NULL pointer dereference, address: 0000000000000000
    PGD 0 P4D 0
    Oops: 0010 [#1] SMP
    CPU: 2 PID: 9345 Comm: test-ovs-ct-gen Not tainted 6.5.0_for_upstream_min_debug_2023_09_05_16_01 #1
    Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014
    RIP: 0010:0x0
    Code: Unable to access opcode bytes at0xffffffffffffffd6.
    RSP: 0018:ffff888178aaf758 EFLAGS: 00010246
    Call Trace:
     <TASK>
     ? __die+0x20/0x60
     ? page_fault_oops+0x14c/0x3c0
     ? exc_page_fault+0x75/0x140
     ? asm_exc_page_fault+0x22/0x30
     notifier_call_chain+0x35/0xb0
     blocking_notifier_call_chain+0x3d/0x60
     mlx5_blocking_notifier_call_chain+0x22/0x30 [mlx5_core]
     mlx5_core_uplink_netdev_event_replay+0x3e/0x60 [mlx5_core]
     mlx5_mdev_netdev_track+0x53/0x60 [mlx5_ib]
     mlx5_ib_roce_init+0xc3/0x340 [mlx5_ib]
     __mlx5_ib_add+0x34/0xd0 [mlx5_ib]
     mlx5r_probe+0xe1/0x210 [mlx5_ib]
     ? auxiliary_match_id+0x6a/0x90
     auxiliary_bus_probe+0x38/0x80
     ? driver_sysfs_add+0x51/0x80
     really_probe+0xc9/0x3e0
     ? driver_probe_device+0x90/0x90
     __driver_probe_device+0x80/0x160
     driver_probe_device+0x1e/0x90
     __device_attach_driver+0x7d/0x100
     bus_for_each_drv+0x80/0xd0
     __device_attach+0xbc/0x1f0
     bus_probe_device+0x86/0xa0
     device_add+0x637/0x840
     __auxiliary_device_add+0x3b/0xa0
     add_adev+0xc9/0x140 [mlx5_core]
     mlx5_rescan_drivers_locked+0x22a/0x310 [mlx5_core]
     mlx5_register_device+0x53/0xa0 [mlx5_core]
     mlx5_init_one_devl_locked+0x5c4/0x9c0 [mlx5_core]
     mlx5_init_one+0x3b/0x60 [mlx5_core]
     probe_one+0x44c/0x730 [mlx5_core]
     local_pci_probe+0x3e/0x90
     pci_device_probe+0xbf/0x210
     ? kernfs_create_link+0x5d/0xa0
     ? sysfs_do_create_link_sd+0x60/0xc0
     really_probe+0xc9/0x3e0
     ? driver_probe_device+0x90/0x90
     __driver_probe_device+0x80/0x160
     driver_probe_device+0x1e/0x90
     __device_attach_driver+0x7d/0x100
     bus_for_each_drv+0x80/0xd0
     __device_attach+0xbc/0x1f0
     pci_bus_add_device+0x54/0x80
     pci_iov_add_virtfn+0x2e6/0x320
     sriov_enable+0x208/0x420
     mlx5_core_sriov_configure+0x9e/0x200 [mlx5_core]
     sriov_numvfs_store+0xae/0x1a0
     kernfs_fop_write_iter+0x10c/0x1a0
     vfs_write+0x291/0x3c0
     ksys_write+0x5f/0xe0
     do_syscall_64+0x3d/0x90
     entry_SYSCALL_64_after_hwframe+0x46/0xb0
     CR2: 0000000000000000
     ---[ end trace 0000000000000000  ]---

    Fixes: 2c3b5beec46a ("net/mlx5e: More generic netdev management API")
    Signed-off-by: Shay Drory <shayd@nvidia.com>
    Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
    Reviewed-by: Simon Horman <horms@kernel.org>
    Link: https://lore.kernel.org/r/20240509112951.590184-2-tariqt@nvidia.com
    Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Depends on
b1a33e651347 net/mlx5e: Add wrapping for auxiliary_driver ops and remove unused args (v6.8-rc1)
To avoid this dependency, the same changes to mlx5e_suspend() would have to be introduced.

Signed-off-by: Benjamin Poirier bpoirier@redhat.com