On Thu, Apr 18, 2013 at 9:04 PM, Tejun Heo <tj@xxxxxxxxxx> wrote: > So, the thing is there is no actual deadlock. If you're okay with releasing > w/ spurious lockdep warning, leaving things alone should be fine. An issue > with mst's patch is that it actually changes the behavior to avoid a > spurious warning. An alternative course would be leaving it alone now and > remove the spurious warning during the coming devel cycle and mark it w/ > -stable. If I understand correctly, you need v3.9-rc1 or later, CONFIG_PROVE_LOCKING=y, and an SR-IOV device to see the warning. I like the idea of fixing the spurious warning for v3.10 and backporting to -stable. It sounds like there's a cleaner fix in the works that needs a bit more polishing. If we need a quick fix sooner, we'll still have this one in our back pocket. Bjorn -- To unsubscribe from this list: send the line "unsubscribe linux-pci" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
