Hello Andreas, On Fri, 2025-07-25 at 11:00 +0200, Andreas Tille wrote: > The only relevant part here is: > > new: > lrwxrwxrwx root/root /etc/power/event.d/laptop-mode -> > ../scripts.d/laptop- > mode > > > > old: > -rwxr-xr-x root/root /etc/power/event.d/laptop- > mode > > > > > which boils down to the following diff inside the upstream source: > > > laptop-mode-tools(master) $ diff -u etc/power/event.d/laptop-mode > etc/power/scripts.d/laptop-mode > --- etc/power/event.d/laptop-mode 2025-07-25 07:45:31.913904502 > +0200 > +++ etc/power/scripts.d/laptop-mode 2025-07-25 07:45:31.913904502 > +0200 > @@ -27,6 +27,9 @@ > performance) > $LMODE "stop" > ;; > + resume) > + $LMODE "auto" "force" > + ;; > esac > elif [ -x logger ]; then > logger -p daemon.error -t laptop-mode "Laptop-mode support > missing in kernel." > > > I think this is an intended change inside install.sh and the other > laptop-mode file might be just some regression due to using some > competing installer procedure in d/rules in the old package. In any > case I would like to catch your attraction onto this matter since you > are upstream as well.
It has been a while since this change. My recollection about it is about how the Linux kernel behaves on device plug/unplug vs device swsusp/resume. I recollect an $LMMODE "auto" being invoked for device [un]plug while a swsusp/resume required a more forced [re]initialization. This was a change introduced somewhere around the time we were concluding systemd as the default init in Debian. So, around then, I put a lot of restructuring into how all could LMT be invoked (init, systemd, udev) -- Ritesh Raj Sarraf | http://people.debian.org/~rrs Debian - The Universal Operating System
signature.asc
Description: This is a digitally signed message part
