[2.6.0-test8] Difference between Software Suspend and Suspend-to-disk?

[2.6.0-test8] Difference between Software Suspend and Suspend-to-disk?

[Jan De Luyck]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello all,

I'm puzzled by two options in the menuconfig which seem to do the same thing:

Software Suspend (EXPERIMENTAL)
Suspend-to-Disk Support

Do i need to have the second to actually suspend? Or does the first one by 
itself do everything I need to suspend?

Thanks.

Jan
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.3 (GNU/Linux)

iD8DBQE/lRVsUQQOfidJUwQRArkdAJ4/XEyzNle9XxPy1wkDAtTp9Hf0aQCeMOBL
qkvnb8YPymSID7AKaYtJ3Ss=
=/Nbf
-----END PGP SIGNATURE-----

Re: [2.6.0-test8] Difference between Software Suspend and Suspend-to-disk?

[Tomas Szepe]

On Oct-21 2003, Tue, 13:15 +0200
Jan De Luyck <lkml@kcore.org> wrote:

> Software Suspend (EXPERIMENTAL)
> Suspend-to-Disk Support

They're competing implementations of the same mechanism.

-- 
Tomas Szepe <szepe@pinerecords.com>

Re: [2.6.0-test8] Difference between Software Suspend and Suspend-to-disk?

[Måns Rullgård]

Tomas Szepe <szepe@pinerecords.com> writes:

>> Software Suspend (EXPERIMENTAL)
>> Suspend-to-Disk Support
>
> They're competing implementations of the same mechanism.

And neither one works reliably, I might add.  They both appear to save
the current state to disk, but no matter what I try, I can't make it
resume properly.

-- 
Måns Rullgård
mru@kth.se

Re: [2.6.0-test8] Difference between Software Suspend and Suspend-to-disk?

[Jan De Luyck]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Tuesday 21 October 2003 13:40, Måns Rullgård wrote:
> Tomas Szepe <szepe@pinerecords.com> writes:
> >> Software Suspend (EXPERIMENTAL)
> >> Suspend-to-Disk Support
> >
> > They're competing implementations of the same mechanism.
>
> And neither one works reliably, I might add.  They both appear to save
> the current state to disk, but no matter what I try, I can't make it
> resume properly.

Well, I could suspend/resume correctly first time I tried (runlevel 2 and 
disabling nearly everything to minimize loss of data on crash).

The second test in X resulted in a 'double fault'...

Jan
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.3 (GNU/Linux)

iD8DBQE/lSOrUQQOfidJUwQRAsV8AJ0WGVwtPCvoRU+nYzp+1rEPGjHrDwCffiZz
wEG77k6o4VjLOb3dMvyLopg=
=uta/
-----END PGP SIGNATURE-----

Re: [2.6.0-test8] Difference between Software Suspend and Suspend-to-disk?

[Felipe Alfaro Solana]

On Tue, 2003-10-21 at 13:40, Måns Rullgård wrote:
> Tomas Szepe <szepe@pinerecords.com> writes:
> 
> >> Software Suspend (EXPERIMENTAL)
> >> Suspend-to-Disk Support
> >
> > They're competing implementations of the same mechanism.
> 
> And neither one works reliably, I might add.  They both appear to save
> the current state to disk, but no matter what I try, I can't make it
> resume properly.

Yep! I must say I cannot resume my system from disk. The kernel always
complains about failing when trying to resume from disk. Also, after
suspending to disk, the swap partition is completely valid, and I don't
need to "mkswap" it. On previous releases of the kernel, when STD
worked, trying to skip resume from disk left the swap partition unusable
(as expected after dumping memory to the swap file).

Re: [2.6.0-test8] Difference between Software Suspend and Suspend-to-disk?

[Måns Rullgård]

Felipe Alfaro Solana <felipe_alfaro@linuxmail.org> writes:

>> >> Software Suspend (EXPERIMENTAL)
>> >> Suspend-to-Disk Support
>> >
>> > They're competing implementations of the same mechanism.
>> 
>> And neither one works reliably, I might add.  They both appear to save
>> the current state to disk, but no matter what I try, I can't make it
>> resume properly.
>
> Yep! I must say I cannot resume my system from disk. The kernel always
> complains about failing when trying to resume from disk. Also, after
> suspending to disk, the swap partition is completely valid, and I don't
> need to "mkswap" it. On previous releases of the kernel, when STD
> worked, trying to skip resume from disk left the swap partition unusable
> (as expected after dumping memory to the swap file).

Same here.  I recall seeing a patch that did something to sync the
disks before powering off.  Maybe it's related.  I'll dig out the
patch and see what it's all about.

-- 
Måns Rullgård
mru@kth.se

Re: [2.6.0-test8] Difference between Software Suspend and Suspend-to-disk?

[Stefan Smietanowski]

Måns Rullgård wrote:
> Tomas Szepe <szepe@pinerecords.com> writes:
> 
> 
>>>Software Suspend (EXPERIMENTAL)
>>>Suspend-to-Disk Support
>>
>>They're competing implementations of the same mechanism.
> 
> 
> And neither one works reliably, I might add.  They both appear to save
> the current state to disk, but no matter what I try, I can't make it
> resume properly.
> 

That's why it's called Software Suspend and Suspend-to-Disk.
The matching components Software Resume and Resume-from-Disk
don't work yet :)

// Stefan

[2.6.0-test8] Difference between Software Suspend and Suspend-to-disk?

[John Mock]

    > > They're competing implementations of the same mechanism.
    >
    > And neither one works reliably, I might add. They both appear to save
    > the current state to disk, but no matter what I try, I can't make it
    > resume properly.

    Well, I could suspend/resume correctly first time I tried (runlevel 2 and
    disabling nearly everything to minimize loss of data on crash).

    The second test in X resulted in a 'double fault'...

What happens if you turn off X and/or run X in some kind of framebuffer
mode (probably VESA mode)?  That's the only way it will work for me.

The native mode X drivers like to hack the hardware (look at all the PCI
analysis in XFree86.0.log if you want to see what i'm talking about).
If X sets up the hardware in a special way, then if LINUX gets suspend
and resumed (re-booted, from the standpoint of the display hardware),
then any state that X set up for itself will be hopelessly lost.

So i'm not sure X in DRI mode can work at all, at least, not the version
that i'm running (4.2.1-11).
				  -- JM