I have a PCI device that is sitting behind a bridge.

Under certain circumstances the PCI device will become inactive. Reading the PCI config space returns all 0xFFFFFFFF.

The bridge appears to still be functional. Reading the status from the bridge I see a Fatal Error due to a Surprise Down event.

I am trying to figure out how to bring the device back online.

I tried toggling the secondary bus reset bit of the Bridge Control Register but it doesn't appear to make any difference. I still see 0xFFFFFFFF in the device config space.

I provided a pci_error_handler but the error_detected() function is not getting called.

Given that these two methods are not helping me out what other choices do I have to either reset the PCI device or hot-plug the device from a kernel driver. Or some other method of bring the device back to life.

Note that I am running Linux 4.8 in dom0 on Xen (if that makes a difference).

Thanks,
Kelly

Having any form of virtualization in your path adds the question: "Is it the hardware or the virtualization screwing up?"

ie. Try the hardware without the virtualization and see if life improves.

>>"Is it the hardware or the virtualization screwing up?"

I haven't tried without virtualization yet but I am pretty sure that it is the hardware as I am giving it a nasty whack as part of stress testing.

Hardware is sometimes timing sensitive, so adding in a bunch of cycles delay through virtualization can cause sorrow.

Sometimes rmmod'ing the driver and then modprobe'ing it works to revive sad hardware.

If you can't rmmod it, you may have to remove the entire stack of mods that depends on it first.

Sometimes way up the /sys/devices tree you can find useful things to toggle as well.

1 members found this post helpful.

I tried rmmod but it hard hung the system. Thinking back however my driver assumes that the device is still active when it does an rmmod. During rmmod it tries to "uninitialized" the device. Touching the hardware hung the system.

I am going to add a check in my driver to take a "quick-path" out if the hardware is not responsive.

Does rmmod/modprobe do anything to the hardware (ie reset) or does it just unload/load the driver?

Vague memory that it runs some hook code at unload time, whether it gets lower layer modules like the pci driver to do anything sensible I don't know.

I sort of would expect the ability to hard reset / power on / power off a device to would reside with the bus layer.