On Tue, Jul 16, 2024 at 8:02 PM Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx> wrote: > > On Tue, 16 Jul 2024 at 01:04, Manivannan Sadhasivam > <manivannan.sadhasivam@xxxxxxxxxx> wrote: > > > > - if (IS_ENABLED(CONFIG_OF) && pci_is_bridge(dev)) { > > + if (IS_ENABLED(CONFIG_OF) && dev_of_node(&dev->dev) && pci_is_bridge(dev)) { > > retval = of_platform_populate(dev->dev.of_node, NULL, NULL, While at it: you could replace the dev.of_node here with dev_of_node() too. I seem to have missed this helper. Bartosz > > &dev->dev); > > I think you should just drop the IS_ENABLED(CONFIG_OF) check entirely. > > afaik, dev_of_node() already returns NULL if CONFIG_OF isn't set. > > So the bug was literally that you based the decision on something > pointless that shouldn't be there at all. > > Linus