**** BEGIN LOGGING AT Thu Jul 11 03:01:20 2019 **** BEGIN LOGGING AT Thu Jul 11 03:40:59 2019 Jul 11 04:00:06 https://puri.sm/posts/purism-and-the-linux-5-2-kernel/ Jul 11 07:50:49 <3 Jul 11 07:51:42 the best thing about that is that we might even end-up with full proper mainline support for imx8 (instead of a vendor-specific tree) :) Jul 11 07:52:42 I hope they also try merging power-related stuff Jul 11 17:00:11 bencoh: power-related stuff seems to be too arbitrary to ever make a big performance upstream Jul 11 17:00:55 in the end that's where the software gets defined by the schematics of the individual device Jul 11 17:02:08 also the domain where sw devels bail out by the dozen. Obviously a concept of proprietary hardware is uncompatible with the mind of the average sw dev Jul 11 17:06:20 sometimes when you're lucky you may get a huge (set of) driver for a whole PMIC like TWL4030, but even then each design/schematics has a few little idiosyncrasies that are terribly hard or impossible to cover completely and in a nice way by a generic driver Jul 11 17:07:06 IOW no two device designs using same PMIC are so identical that they may use same drivers and completely cover all aspects of the device Jul 11 17:16:30 DocScrutinizer05: cpuidle improvements (and soc-specific idle state) doesn't sound 'arbitrary' to me Jul 11 17:17:08 and pmic drivers are reusable assuming you do it right (yes, it happens) **** ENDING LOGGING AT Fri Jul 12 02:59:57 2019