On 4/19/2018 4:26 PM, Jason Gunthorpe wrote:
> On Thu, Apr 19, 2018 at 03:56:23PM -0400, Sinan Kaya wrote:
>> The infiniband adapter might be connected to a PCI hotplug slot. Performing
>> secondary bus reset on a hotplug slot causes PCI link up/down interrupts.
>>
>> Hotplug driver removes the device from system when a link down interrupt
>> is observed and performs re-enumeration when link up interrupt is observed.
>>
>> This conflicts with what this code is trying to do. Try secondary bus reset
>> only if pci_reset_slot() fails/unsupported.
>>
>> Signed-off-by: Sinan Kaya <okaya@xxxxxxxxxxxxxx>
>> drivers/infiniband/hw/hfi1/pcie.c | 3 ++-
>> 1 file changed, 2 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/infiniband/hw/hfi1/pcie.c b/drivers/infiniband/hw/hfi1/pcie.c
>> index 83d66e8..75f49e3 100644
>> +++ b/drivers/infiniband/hw/hfi1/pcie.c
>> @@ -908,7 +908,8 @@ static int trigger_sbr(struct hfi1_devdata *dd)
>
> The code above this hunk is:
>
> /*
> * Trigger a secondary bus reset (SBR) on ourselves using our parent.
> *
> * Based on pci_parent_bus_reset() which is not exported by the
> * kernel core.
> */
> static int trigger_sbr(struct hfi1_devdata *dd)
> {
>
> [..]
>
> This really seems like something the PCI core should be helping with,
> drivers shouldn't be doing stuff like this. I get the feeling this
> should be a common need if drivers support various error recovery
> schemes?
pci_parent_bus_reset() still doesn't deal with hotplug. We need to call
a variation of pci_slot_reset() before calling pci_parent_bus_reset().
rc = pci_dev_reset_slot_function(dev, 0);
if (rc != -ENOTTY)
return rc;
return pci_parent_bus_reset(dev, 0);
VFIO driver does this.
/* User has access, do the reset */
ret = slot ? pci_try_reset_slot(vdev->pdev->slot) :
pci_try_reset_bus(vdev->pdev->bus);
I assumed the responsibility is at the driver to call the right API it likes.
--
Sinan Kaya
Qualcomm Datacenter Technologies, Inc. as an affiliate of Qualcomm Technologies, Inc.
Qualcomm Technologies, Inc. is a member of the Code Aurora Forum, a Linux Foundation Collaborative Project.
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
