On Fri, 06 May 2022 12:11:56 +0200 Rolf Eike Beer wrote: > Am Freitag, 6. Mai 2022, 11:42:50 CEST schrieb Yang Yingliang: > > Fix the missing pci_disable_device() before return > > from tulip_init_one() in the error handling case. > > I would suggest removing the pci_disable_device() from tulip_remove_one() > instead and using pcim_enable_device(), i.e. devres, and let the driver core > handle all these things. Of course more of the used functions could be > converted them, e.g. using devm_alloc_etherdev() and so on. Let's not rewrite the error handling in this dinosaur of a driver any more than absolutely necessary, please.
