[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Proper creation of /dev/apm_bios

John Goerzen <jgoerzen@complete.org> wrote:
> On certain (many?) machines, the new X packages will cause the
> console to completely hang after waking back up from being
> put to sleep.  This can be fixed by having /dev/apm_bios and
> appropriate kernel support.  The powermgmt-base package does
> this, but on PowerPC, it is not installed by default, leaving
> PowerPC laptops in a situation where they are easily hung by
> default.

That does sound like an unacceptable situation on powerpc.
We don't want machines hanging mysteriously.

If it were simply a matter of checking for the presence of
/dev/apm_bios, then X could do that at startup time
and exit if the node was absent.

However, it sounds to me as if the problem isn't one that 
can be fixed by X.  Without APM support, X doesn't know when
a suspend-and-resume cycle occurs, and so doesn't know when it
should reinitialize hardware registers.

So the solution must be to configure APM support on machines
that are capable of suspending and resuming.

If you don't have APM support then you shouldn't suspend the
machine.  If you go ahead and suspend the machine anyway then
you should be prepared for failures.

It might be fair to ask that X print a message at startup
when it can't find APM support: it could warn that, in the
absence of APM support, suspending and resuming an X session
may have unpredictable results.

Thomas Hood

Reply to: