[Pkg-acpi-devel] Bug#524986: acpi-support: /etc/acpi/sleep.sh should not call prepare.sh unconditionally

[Bart Samwel]

Bjørn Mork wrote:
> Package: acpi-support
> Version: 0.114-1
> Severity: normal
> 
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> One of the changes between 0.109-11 and 0.114-1 is that sleep.sh calls
> prepare.sh, which does
> 
> for SCRIPT in /etc/acpi/suspend.d/*.sh; do
>     if [ -x $SCRIPT ] ; then
>         . $SCRIPT
>     fi
> done
> 
> This should not happen unless we're using the legacy suspend support.
> Other methods have their own list of suspend quirks, which are more likely
> to be up-to-date than the scripts in /etc/acpi/suspend.d/*.sh
> 
> The same goes for the hdparm DMA fiddling:  acpi-support should leave this
> up to the actual suspend method unless legacy suspend is used.
> 
> On my Lenovo Thinkpad X301, running the default scripts in 
> /etc/acpi/suspend.d/*.sh results in 
> - - non-functional wireless network (module iwlagn unloaded)
> - - non-functional display (needs VT switch)
> + possibly other minor errors
> 
> Calling "/usr/share/acpi-support/suspendorhibernate suspend" directly
> still works as expected (using pm-utils as suspend method).

Hmmm, that's probably caused by the merge of upstream 0.114 from Ubuntu 
-- Ubuntu does not support suspend methods, this has been added to 
Debian only. (The ubuntu package did at some point mention suspend 
methods in the config files though, probably an over-enthousiastic merge 
from Debian. Don't know if that's still there.)

Cheers,
Bart