mbox series

[v7,0/1] PCI: Handle conflict between hotplug and error handling

Message ID 20180805225136.5800-1-okaya@kernel.org (mailing list archive)
Headers show
Series PCI: Handle conflict between hotplug and error handling | expand

Message

Sinan Kaya Aug. 5, 2018, 10:51 p.m. UTC
Lukas:
'Your patch "[PATCH v6 1/1] PCI: pciehp: Ignore link events when there
is a fatal error pending" only checks once for a pending fatal error,
it should poll until either the fatal error is gone or a timeout is
hit.  If the fatal error is gone and the link is up, you can just return
from pciehp_handle_presence_or_link_change().  Else (in the timeout case)
fall back to the normal handling of a Link Down, i.e. let it bring down
the slot.'


Sinan Kaya (1):
  PCI: pciehp: Ignore link events when there is a fatal error pending

 drivers/pci/hotplug/pciehp_ctrl.c | 19 ++++++++++
 drivers/pci/pci.h                 |  2 ++
 drivers/pci/pcie/err.c            | 60 +++++++++++++++++++++++++++++++
 3 files changed, 81 insertions(+)

Comments

Bjorn Helgaas Sept. 4, 2018, 6:18 p.m. UTC | #1
On Sun, Aug 05, 2018 at 03:51:35PM -0700, Sinan Kaya wrote:
> Lukas:
> 'Your patch "[PATCH v6 1/1] PCI: pciehp: Ignore link events when there
> is a fatal error pending" only checks once for a pending fatal error,
> it should poll until either the fatal error is gone or a timeout is
> hit.  If the fatal error is gone and the link is up, you can just return
> from pciehp_handle_presence_or_link_change().  Else (in the timeout case)
> fall back to the normal handling of a Link Down, i.e. let it bring down
> the slot.'
> 
> 
> Sinan Kaya (1):
>   PCI: pciehp: Ignore link events when there is a fatal error pending
> 
>  drivers/pci/hotplug/pciehp_ctrl.c | 19 ++++++++++
>  drivers/pci/pci.h                 |  2 ++
>  drivers/pci/pcie/err.c            | 60 +++++++++++++++++++++++++++++++
>  3 files changed, 81 insertions(+)

Dropping for now because it sounds like you intend to mask out the
error for hotplug ports.
Sinan Kaya Sept. 4, 2018, 6:28 p.m. UTC | #2
On 9/4/2018 11:18 AM, Bjorn Helgaas wrote:
> On Sun, Aug 05, 2018 at 03:51:35PM -0700, Sinan Kaya wrote:
>> Lukas:
>> 'Your patch "[PATCH v6 1/1] PCI: pciehp: Ignore link events when there
>> is a fatal error pending" only checks once for a pending fatal error,
>> it should poll until either the fatal error is gone or a timeout is
>> hit.  If the fatal error is gone and the link is up, you can just return
>> from pciehp_handle_presence_or_link_change().  Else (in the timeout case)
>> fall back to the normal handling of a Link Down, i.e. let it bring down
>> the slot.'
>>
>>
>> Sinan Kaya (1):
>>    PCI: pciehp: Ignore link events when there is a fatal error pending
>>
>>   drivers/pci/hotplug/pciehp_ctrl.c | 19 ++++++++++
>>   drivers/pci/pci.h                 |  2 ++
>>   drivers/pci/pcie/err.c            | 60 +++++++++++++++++++++++++++++++
>>   3 files changed, 81 insertions(+)
> 
> Dropping for now because it sounds like you intend to mask out the
> error for hotplug ports.
> 

Yes, I handed this work off-list to Keith and Lucas. I don't have hardware
to test.