Re: [RFC][PATCH] PM: Use suspend_console in swsusp and make it configureable

[Pavel Machek <pavel@suse.cz>]

Hi!

> > > The appended patch does the following:
> > > 
> > > 1) Adds suspend_console() and resume_console() to the suspend-to-disk code
> > > paths so that people using netconsole are safe with swsusp.
> 
> So I assume this one is OK.
> 
> > > 2) Adds a Kconfig option allowing us to disable suspend_/resume_console()
> > > if need be.
> > 
> > Slightly ugly, but I guess that is the way to go.
> 
> It also seems to be needed to add a 2 sec. dealy in suspend_console() so
> that eg. the network console can send the messages before we try to suspend
> the device.
> 
> > > 3) Marks CONFIG_PM_TRACE as dangerous.
> > 
> > I do not think that is enough. "(WILL TRASH YOUR CMOS)" would be more
> > suitable. Dangerous is "may cause problems to you". This is "will
> > cause problems to you". And for this to be useful, people have to edit
> > sources, anyway.
> > 
> > 								Pavel
> > 
> > Can we just delete the config option?
> 
> OK, I'll do that.
> 
> I have divided the changes into three individual patches that will follow as
> replies to this message.

Ack on all three... and thanks.

-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html