Hi=C2=A0Lukas,<= /span>

>The solution is to acquire a ref on ea= ch device in add_error_device():

I will give try and updat= e if it works .

regards,

Gokul=C2=A0

On Thu, Aug 2, 2018 at 2:16 PM= , Lukas Wunner <lukas@wunner.de> wrote:

On Thu, Aug 02, 2018 at 12:59:18PM +0530, go= kul cg wrote:
> I am suspecting a possible race condition in the kernel between PCI dr= iver
> and AER handling.
>
> Because of the same kernel panic happens from worker thread which hand= les
> bottom half of aer irq.
>
> I am seeing this issue when I suddenly power off PCI card which
> supports/enabled PCIE AER error reporting.
>
> While powering off PCI device, AER driver will get AER IRQ for the dev= ice,
> from AER IRQ handler, it will cache AER error code and schedule worker=
> thread to handle error.
>
> The PCIe device will get removed from PCI tree before worker thread > completes its task and kernel panic is=C2=A0 happening when worker thr= ead tries
> to access PCI device's config space.
>
> #5 [ffff88027469fc70] general_protection at ff= ffffff8176cdf2
>=C2=A0 =C2=A0 =C2=A0[exception RIP: pci_bus_read_config_dword+100]
> #6 [ffff88027469fd50] pci_find_next_ext_capabi= lity at ffffffff81345d7b
> #7 [ffff88027469fd90] pci_find_ext_capability at ffffffff81347225
> #8 [ffff88027469fda0] get_device_error_info at ffffffff81356c4d
> #9 [ffff88027469fdd0] aer_isr at ffffffff81357a38
> #10 [ffff88027469fe28] process_one_work at ffffffff8105d4c0
> #11 [ffff88027469fe70] worker_thread at ffffffff8105e251
> #12 [ffff88027469fed0] kthread at ffffffff81064260
> #13 [ffff88027469ff50] ret_from_fork at ffffffff81773a38
>
> I have tested it on kernel 3.10 . But from sou= rce i could see that this
> case is still relevant for latest Linux source .

I'm not really familiar with the AER driver, but the problem is<= br> actually easy to spot:

find_source_device() walks the hierarchy and saves a pointer to
pci_dev's in an array.=C2=A0 That array is later traversed and the
pci_dev's are accessed.

The solution is to acquire a ref on each device in add_error_device():

-=C2=A0 =C2=A0 =C2=A0 =C2=A0e_info->dev[e_info->error_dev_num] = =3D dev;
+=C2=A0 =C2=A0 =C2=A0 =C2=A0e_info->dev[e_info->error_dev_num] = =3D pci_dev_get(dev);

Then release the ref aer_process_err_devices() by calling pci_dev_put().
I believe there's an ongoing refactoring of the AER driver and the
issue may be addressed in the course of it, but as a quick fix for
an ancient v3.10 kernel, the above should do the trick.

HTH,

Lukas