On Sat, Oct 26, 2024 at 12:26:13PM +0100, Jonathan Cameron wrote: > On Thu, 24 Oct 2024 22:05:00 +0300 > Andy Shevchenko <andriy.shevchenko@xxxxxxxxxxxxxxx> wrote: > > > Instead of having a enum and keeping IDs as driver data pointers, > > just have a chip_info struct per supported device. > I'm not keen longer term on acpi_type, as the various bits dependent on that > should probably be done via optional callbacks in the kx_chipset_info > structure, but this is a sensible intermediate step. Yeah, I decided to postpone that because the main point of this series is to clean up acpi_match_device(). And indeed I was thinking about callbacks and other things (flags, string literals, etc) to be moved to chip_info. > I see the chipset one goes away later hence no comment on that. > > I did a bit of white space massaging whilst applying this. > Hopefully that won't make me mess up applying the following patches. Thanks! (OTOH I'm not sure what spacing you meant because in the result it still seems slightly inconsistent.) -- With Best Regards, Andy Shevchenko
