Thanks for the response Henrique. I do have the laptop setup not to
suspend when I close the lid.
So 60C / ~2.0GHz sounds right to you if I have it set to "balanced" for
battery power, as opposed to the 80C / ~2.9GHz when I run on battery with
the lid open? I do have a couple options available through the power
management GUI in Linux Mint, but it's standard stuff like suspend, screen
dimming, etc. I will poke around and see if maybe there are some other
options under the hood that look related.
I haven't noticed anything weird with the fan controls. But I can be
running it with the lid closed and get these messages periodically even
when the temperature is low enough that the fan is at 0 RPM, around 45
degrees. I also noticed that Lenovo did release another firmware update,
so I'll give that a try and see if anything changes. In any case,
whatever's in control of this appears to be blowing off my settings in the
BIOS, because the clock speeds and temps don't seem to change even if I set
everything to Max Performance.
If I wanted to change the code of thinkpad-acpi.c to "handle" this event,
should I treat it the same way as the normal lid open/close events in the
hotkey_notify_usrevent section? Is this something that could be put in the
code for later versions?
Thanks,
Neil
On Wed, May 10, 2017 at 1:09 PM, Henrique de Moraes Holschuh <h...@hmh.eng.br
> wrote:
> On Sat, 22 Apr 2017, neil k wrote:
> > I spent some more time testing this today. It seems like it only happens
> > when I close the lid while *running on battery*. If I close the lid with
>
> Yeah, that event is some sort of hint to the Lenovo Windows drivers to
> change their behavior, or to reprogram something in other windows
> drivers.
>
> I assume you are closing the lid **without** suspending, right?
>
> Probably, the Lenovo windows drivers kick everything to idle-mode
> forcibly (or maybe kicks the GPU into low-power mode, disables the
> discrete GPU, whatever).
>
> thinkpad-acpi is not doing much other than warning you of an event we
> don't know about, blacklisting it won't help, unless it is actually
> being used to mess with fan control. Do note thinkpad-acpi fan-control
> is opt-in, someone must have forced it to be enabled using a module
> parameter, if it is active.
>
> > the laptop plugged in, THEN unplug it, I don't see these temperature
> > messages happen during idling / light use. Also notable is that unhandled
> > HKEY event 0x6032 doesn't show up if I close the lid on A/C power. This
>
> Yeah, firmware implementation did not cover all possibilities :p
>
> > issue still happens if I blacklist the thinkpad-acpi module, but I'm
> > wondering if it's something Lenovo specific that isn't being handled and
> > triggers the issue.
>
> Exactly.
>
> > Thermal
> > AC – max performance
> > Battery – Balanced
>
> Check if it is not messing with the fan when you close the lid. That
> might be it...
>
> > I also tested to see how the CPU behaved under load and found something
> > else interesting. If I close the lid with the laptop unplugged (the
> > condition that causes this issue), the processor runs at a much lower
> clock
> > speed (~2.0 GHz) and much lower temperature (60C) when under load. This
>
> Yeah, that's the UEFI balanced mode properly working.
>
> > makes me think something about the lid being closed on battery power
> causes
> > an issue with Intel Speedstep / p states / c states. I don't really know
> > much about how that works though, unfortunately.
>
> Either that, or some stupid crap is touching knobs it shouldn't. Usual
> suspects are the power management stuff from pm-tools, systemd, or
> laptop-mode.
>
> > converting video with lid closed plugged in will hover at 96-98C / 3.48
> GHz
> > <-- 3.5GHz is the maximum turbo boost for this processor
>
> Yeah, this is too much if the fan slows down or the box does non-trivial
> dissipation through the keyboard or palm-rest, and it will overheat.
>
> Do ensure you keep the firmware of that box up-to-date, please. You
> should get a new UEFI and EC update from Lenovo every so often on such a
> new machine. If you didn't check their site this week yet, please do so
> ;-)
>
> --
> Henrique Holschuh
>
------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
ibm-acpi-devel mailing list
ibm-acpi-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ibm-acpi-devel
