On Mon, Feb 04, 2019 at 03:03:49PM +0300, Dmitry Osipenko wrote:
04.02.2019 14:05, Thierry Reding пишет:
On Mon, Feb 04, 2019 at 09:53:32AM +0000, Jon Hunter wrote:
On 04/02/2019 08:45, Thierry Reding wrote:
...
The idea was, as I was saying below, to reuse dev_pm_ops even if !CONFIG_PM. So pm_runtime_enable() could be something like this:
pm_runtime_enable(dev) { if (!CONFIG_PM) if (dev->pm_ops->resume) dev->pm_ops->resume(dev);
...
}
But that's admittedly somewhat of a stretch. This could of course be made somewhat nicer by adding an explicit variant, say:
pm_runtime_enable_foo(dev) { if (!CONFIG_PM && dev->pm_ops->resume) return dev->pm_ops->resume(dev);
return 0;
}
Maybe the fact that I couldn't come up with a good name is a good indication that this is a bad idea...
How about some new APIs called ...
pm_runtime_enable_get() pm_runtime_enable_get_sync() pm_runtime_put_disable() (implies a put_sync)
... and in these APIs we add ...
pm_runtime_enable_get(dev) { if (!CONFIG_PM && dev->pm_ops->resume) return dev->pm_ops->resume(dev);
pm_runtime_enable(dev); return pm_runtime_get(dev); }
Yeah, those sound sensible. I'm still a bit torn between this and just enforcing PM. At least on the display side I think we already require PM and with all the power domain work that you did we'd be much better off if we could just get rid of the !PM workarounds.
This would be somewhat tricky because drivers usually use SET_RUNTIME_PM_OPS to populate the struct dev_pm_ops and that would result in an empty structure if !CONFIG_PM, but we could probably work around that by adding a __SET_RUNTIME_PM_OPS that would never be compiled out for this kind of case. Or such drivers could even manually set .runtime_suspend and .runtime_resume to make sure they're always populated.
Another way out of this would be to make sure we never run into the case where runtime PM is disabled. If we always "select PM" on Tegra, then PM should always be available. But is it guaranteed that runtime PM for the devices is functional in that case? From a cursory look at the code it would seem that way.
If you select PM, then all of the requisite code should be there.
We do this on 64-bit ARM, but there had been some pushback when we had proposed to do the same thing on 32-bit ARM. I think there were two concerns:
select PM would force the setting for all platforms on multi- platforms builds
prevents anyone from disabling PM for debugging purposes
no longer seems to be valid because Rockchip already selects PM
unconditionally. I'm not sure if 2) is valid anymore either. I haven't run a build with !PM in a very long time and I wouldn't be surprised if that was completely broken.
Maybe we need to try this again since a couple of years have elapsed and runtime PM support on Tegra is much more mature at this point.
Alternatively, you can make the driver depend on PM.
That's probably the easiest way out, but to be honest I think I'd prefer to just enforce PM and keep things simple.
Jon, any objections?
None, but seems overkill just for this case.
But that's precisely the point. It's not just about this case. We've already got some drivers where we do a similar dance just to be able to support the, let's admit it, exotic case where somebody turns off PM. I think supporting !PM might have made sense at a point where we had no support for power management at all. But I think we've come a long way since then, and while we may still have some ways to go in some areas, we do fairly decent runtime PM most of the time, to the point where I no longer see any benefits in !PM.
I'm requesting PM_DEBUG_ALWAYS_ON option then! Disabling PM is a useful debug feature, it can't just go away.
What is it about disabling PM that you consider useful? I can understand why you'd want that option if power management is broken, but as far as I can tell, power management on Tegra is in pretty good state, and it's more likely that !PM would actually be broken (though I haven't built a configuration like that in a couple of years, so I'm speculating). We already can't disable PM on 64-bit ARM, so I don't understand why 32-bit ARM should be treated any differently.
Thierry