Re: Making powerd=YES default

[Jukka Ruohonen <jruohonen%iki.fi@localhost>]

On Mon, Jul 25, 2011 at 03:12:40PM +0100, Jean-Yves Migeon wrote:
> When we'll start having questions like "I tried an xm save but the 
> command froze afterwards, and domU is still executing, what am I doing 
> wrong?", it won't be that funny to answer them all. And I'd like to 
> avoid the route of "just do everything in kernel it's easier", or 
> "foobar allows doing this and that, if you are unsure say [Y] to the 
> question."

Indeed. People have already had to answer comparable questions for years.
And as we have discussed previously, it is presumably a bad design choice to
even consider the option of "doing everything in the kernel", given the
inherited complexity of power management and "platform specific events".

On Mon, Jul 25, 2011 at 06:11:39PM +0000, Emmanuel Dreyfus wrote:
> I think the sysctl telling that power management is avaiable, and
> having powerd start or not depending on it, is a good idea: the
> syscl could be usefull for other purposes.

Like Jean-Yves pointed out, powerd(8) is not strictly related to "power
management" but rather to generic hardware support. I don't see any reason
for more magic, undocumented, buttons.

On Mon, Jul 25, 2011 at 03:34:19PM +0000, Jonathan A. Kollasch wrote:
> I'd do it like we handle wscons=YES on x86. That is, have sysinst do it
> without asking.  Sysinst will at least know what port it's installing, 
> and if it is probably appropriate to enable it.

I think this might be a reasonable compromise.

- Jukka.