[openbeos] Hybernate and Resume functions

  • From: "Brian Hague" <BrianHague@xxxxxxxxxxxxxxxxxxxx>
  • To: <openbeos@xxxxxxxxxxxxx>
  • Date: Fri, 23 Apr 2004 03:33:50 -0500


I think it would also be a good idea to steal...

Wouldn't it be good if the entire RAM didn't need to be dumped onto the HDD
(like my system) only what's used?
What type of technologies are involved in the apple setup, or the x86?
If the x86 is hardware based, is it proprietary from manufacturer to
Is there any difference between Desktop hardware and Laptop hardware if it's
a hardware issue?
Is it totally software or does the memory and CPU get a trickle of power to
hold the states?
How much code would it require?  (realizing the R1+ factor)
Could there be a shutdown memory dump? (completely bypassing any possible
hardware issues)
How would you do the resume on startup?
Would you have a firstboot bug if the installer was saved into memory on
If you cleared the installer and there was no saved data would that also
cause a bug?
How can you make this work without torturing Axel and his basement "simpsons
milk mice" any further?

these and many more questions in the new thread...

Brian Hague

----- Original Message -----
From: "Geoff Sallee" <spamsk8r@xxxxxxxxxxxxx>
To: <openbeos@xxxxxxxxxxxxx>
Sent: Friday, April 23, 2004 2:15 AM
Subject: [openbeos] Re: x86 boot loader milestone

> Makes good sense to me, that's one of the highlights of my Powerbook.
> Just close the lid and go.  In fact, I wouldn't even consider an
> operating system for a laptop unless it had that feature and it worked
> flawlessly like it does on my Mac.  That would be very excellent were
> it to be implemented in OpenBeOS.
> Just my thoughts,
> Geoff
> On Apr 22, 2004, at 11:04 PM, Sam Persson wrote:
> >
> >
> > Michael Phipps wrote:
> >> The truth is, as fast as BeOS (esp off of compact flash) boots, there
> >> is little benefit to doing this. You might save a whole second or
> >> something. Investigating a faster BIOS (i.e. LinuxBios) makes a whole
> >> lot more sense, from a "speed up the boot process" point of view,
> >> than looking at this. I considered the same things but don't think
> >> that it is worth bothering with. That's just my POV on it.
> >> OTOH, I think that this could be a huge feature if it had nothing to
> >> do with bootup. How cool would it be if you could hit some key combo
> >> and the system would freeze for a few seconds, pouring data onto the
> >> HD. When it was done, you would have a complete, bootable image of
> >> exactly where you where at that moment? Not to save boot time, but
> >> for other things - like saving all of your apps exactly where they
> >> were. This would also work better than a boot setup because you would
> >> restore user processes, not kernel processes. Things like drivers
> >> need to init properly. You can't have an image do that without
> >> special knowledge. OTOH, if you boot to a normal (open)BeOS kernel
> >> with new driver instances and all, then restore the userland
> >> processes, that might work better.
> >
> > Sounds very much like the Hibernate function in WinXP (in my
> > pre-morning-coffee state of mind) and I really do think this one is
> > something really (really really) nice to have. Especially for
> > laptopsusers like myself. One of the most common scenarios at work is
> > when working at a project and somwhere in the middle realizing that
> > "gosh... I have to discuss this with <insert-customer-name>". To be
> > able to just close the lid (tied to Hibernate), get to the customer,
> > open the lid (thus powering on) and simply discuss the code without
> > having to open all the programs and stuff and risking loosing my
> > thoughts while browsing through other code while looking for the
> > 'spot'. That is wonderful. And if Walt... OBOS would be able to do it
> > without choking on removed components (like PCMCIA-cards and
> > USB-devices) the way WinXP does, it'd be just marvellous.
> >
> >
> > Err... did this make sense? (need food... and coffee...)
> >
> > //Sam Persson
> >

Other related posts: