13.09.2021 20:14, Sasha Levin пишет: > On Thu, Sep 09, 2021 at 02:04:50PM +0100, Mark Brown wrote: >> On Thu, Sep 09, 2021 at 03:45:45PM +0300, Dmitry Osipenko wrote: >>> 09.09.2021 15:37, Mark Brown пишет: >> >>> > This feels new featureish to me - it'll give you runtime PM where >>> > previously there was none. >> >>> Apparently all patches which have a word 'fix' in commit message are >>> auto-selected. I agree that it's better not to port this patch. >> >> Yeah, it's a fairly common source of false positives :/ > > And some of that falls on me: if it's obvious that the "fix" isn't a > real fix, I won't take it. In cases like this it's not clear to me > whether it's purely a better behaviour, or whether the devices staying > on/off/etc causes an actual problem. > You made the right choice. It's indeed not obvious from commit message what is achieved with this change. This patch shouldn't have any visible effect on older kernels which don't support power management in a case of Tegra SoCs. It will have effect using the upcoming kernels, once the full set of in-progress power management patches will be merged.