Hi Armin, On 11/13/23 20:55, Armin Wolf wrote: > Am 06.11.23 um 07:43 schrieb Armin Wolf: > >> This patch series adds support for an alternative SMM calling >> backend to the dell-smm-hwmon driver. The reason for this is >> that on some modern machines, the legacy SMM calling interface >> does not work anymore and the SMM handler can be called over >> ACPI WMI instead. >> >> The first four patches prepare the driver by allowing to >> specify different SMM calling backends, and by moving most of >> the DMI handling into i8k_init() so that the DMI tables can >> keep their __initconst attributes (the WMI SMM backend driver >> does not probe at module init time). The fifth patch does some >> minor cleanup, while the next three patches implement the new >> WMI SMM calling backend. The last patch adds the machine of >> the user who requested and tested the changes to the fan control >> whitelist. >> >> If the driver does not detect the legacy SMM interface, either >> because the machine is not whitelisted or because the SMM handler >> does not react, it registers an WMI driver which will then bound >> to the WMI SMM interface and do the remaining initialization. >> >> The deprecated procfs interface is not supported when using the >> WMI SMM calling backend for the following reason: the WMI driver >> can potentially be instantiated multiple times while the deprectated >> procfs interface cannot. This should not cause any regressions >> because on machines supporting only the WMI SMM interface, the >> driver would, until now, not load anyway. >> >> All patches where tested on a Dell Inspiron 3505 and a Dell >> OptiPlex 7000. > > Any thoughts on this? I was waiting for the merge window to close before reviewing / merging patches for the next cycle. I plan to review and hopefully merge this and your other series sometime this week. Regards, Hans
