On Thu, Jan 09, 2025 at 08:50:02AM -0800, Jakub Kicinski wrote: > On Thu, 9 Jan 2025 17:27:03 +0100 Oleksij Rempel wrote: > > > > This index is not used in the series, I see later on you'll add power > > > > evaluation strategy but that also seems to be within a domain not > > > > device? > > > > > > > > Doesn't it make sense to move patches 11-14 to the next series? > > > > The other 11 patches seem to my untrained eye to reshuffle existing > > > > stuff, so they would make sense as a cohesive series. > > > > > > Indeed PSE index is used only as user information but there is nothing > > > correlated. You are right maybe we can add PSE index when we have something > > > usable for it. > > Oh, maybe you want to do the devlink-y thing then? > Devlink identifies its devices by what I'd call "sysfs name" - > bus name and device name. > This is more meaningful to user than an artificial ID. > Downside is it won't work well if you have multiple objects > under a single struct device, or multiple struct device for > one ASIC (e.g. multiple PCIe PFs on one PCIe card) > > > No user, means, it is not exposed to the user space, it is not about > > actual user space users. > > Can't parse this :) OK, please forget it :) -- Pengutronix e.K. | | Steuerwalder Str. 21 | http://www.pengutronix.de/ | 31137 Hildesheim, Germany | Phone: +49-5121-206917-0 | Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 |
