ACPI suspend weirdness on Dell Inspiron 2650

ACPI suspend weirdness on Dell Inspiron 2650

[Luke Hammer]

Hi,

I've finallyjust managed to get some kind of suspend state
working on my Dell Inspiron 2650 laptop, using the 2.4.21-pre3 kernel
and the ACPI-20030125-2.4.21-pre3 patch (under RedHat 7.3). I can
get some kind of suspend state working using 'echo 1 >/proc/acpi/sleep'
but I don't think it's the genuine suspend-to-RAM state. For one
thing, as soon as I put it into this state, one of the fans comes on
and stays on (not very good when you're trying to conserve battery
juice, but this is probably a Dell BIOS problem); and secondly, the screen
doesn't actually seem to be turned off - I can still make out an image 
on it,
it just looks like the backlighting has been turned off.

Nothing at all seems to happen if I try to echo 3, 4, 4b or 5 to 
/proc/acpi/sleep,
which is strange since the ChangeLog for the latest ACPI patch mentions
being able to change settings for the S3 suspend, and it also claims to
incorporate the s4bios patch. If I cat /proc/acpi/sleep I get "S0 S1 S3 
S4 S5", so
I'm not sure if I'm doing something wrong, or if ACPI suspend-to-RAM support
isn't fully functional yet.

So I'm wondering if anyone can give me any hints about getting these
real suspend states to work? Any information would be much appreciated.

-Luke




-------------------------------------------------------
This SF.NET email is sponsored by:
SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
http://www.vasoftware.com

Re: ACPI suspend weirdness on Dell Inspiron 2650

[Jean-Pierre Schwickerath]

Luke Hammer <lukeh-rdE3vRTOlSC6c6uEtOJ/EA@public.gmane.org> wrote:

> Hi,
> 
> I've finallyjust managed to get some kind of suspend state
> working on my Dell Inspiron 2650 laptop, using the 2.4.21-pre3 kernel
> and the ACPI-20030125-2.4.21-pre3 patch (under RedHat 7.3). I can
> get some kind of suspend state working using 'echo 1
> >/proc/acpi/sleep'
> but I don't think it's the genuine suspend-to-RAM state. For one

You're right.

> Nothing at all seems to happen if I try to echo 3, 4, 4b or 5 to 
> /proc/acpi/sleep,
> which is strange since the ChangeLog for the latest ACPI patch
> mentions being able to change settings for the S3 suspend, and it also
> claims to incorporate the s4bios patch. If I cat /proc/acpi/sleep I
> get "S0 S1 S3 S4 S5", so
> I'm not sure if I'm doing something wrong, or if ACPI suspend-to-RAM
> support isn't fully functional yet.

Suspend-to-RAM is S3 as far as I know and someone posted here that it'll
probably no be working on 2.4 that soon. Rather try kernel 2.5.x for
this. 

Jean-Pierre
-- 
Powered by Linux From Scratch - http://schwicky.net/
PGP Key ID: 0xEE6F49B4 - AIM/Jabber: Schwicky - ICQ: 4690141

Nothing is impossible... Everything is relative!


-------------------------------------------------------
This SF.NET email is sponsored by:
SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
http://www.vasoftware.com

Re: ACPI suspend weirdness on Dell Inspiron 2650

[Ducrot Bruno]

On Tue, Feb 11, 2003 at 04:00:17PM +0000, Luke Hammer wrote:
> Hi,
> 
> I've finallyjust managed to get some kind of suspend state
> working on my Dell Inspiron 2650 laptop, using the 2.4.21-pre3 kernel
> and the ACPI-20030125-2.4.21-pre3 patch (under RedHat 7.3). I can
> get some kind of suspend state working using 'echo 1 >/proc/acpi/sleep'
> but I don't think it's the genuine suspend-to-RAM state. For one

Correct.  It is 'system standby'.

> thing, as soon as I put it into this state, one of the fans comes on
> and stays on (not very good when you're trying to conserve battery
> juice, but this is probably a Dell BIOS problem); and secondly, the screen
> doesn't actually seem to be turned off - I can still make out an image 
> on it,

How?  Could you explain ?  If yes, then there is a problem.  The
processor is supposed to have stopped execution.

> it just looks like the backlighting has been turned off.

It sound like, but definitely you should some power saving in
this state.  It is not the best, though, and S3 is really preferable.

> 
> Nothing at all seems to happen if I try to echo 3, 4, 4b or 5 to 
> /proc/acpi/sleep,

'3' is disabled explicitely.  It is in active developpement in 2.5.
'4' is not supported.  Actually it is defaulted to '4b'.
    You have to get swsusp (or a 2.5 tree) to get it.
'4b' is still under developpement for 2.4.  For example, nothing is
     done yet to restore a fbconsole (especially the svga one).  It
     is certainly not SMP safe, not to mention that  it can not be preempted.


> which is strange since the ChangeLog for the latest ACPI patch mentions
> being able to change settings for the S3 suspend,

Actually this is only for 2.5.  Perhaps a miss here (I have not read carrefully
the ChangeLog, sorry)

> and it also claims to
> incorporate the s4bios patch. If I cat /proc/acpi/sleep I get "S0 S1 S3 

But only for 2.4.

> S4 S5", so

You should have 'S4bios' if you cat /proc/acpi/sleep.  It is
somehow different from S4.  Remember, S4bios
work _only_ if your bios support it.  If not, I can not do so much for
you.  Please note also that S4bios is _not_ apm suspend-to-disk, so
your laptop can support the apm one, but not S4bios (even though it
sound like strange in my point of view)...

> I'm not sure if I'm doing something wrong, or if ACPI suspend-to-RAM support
> isn't fully functional yet.

S3 would not backported to 2.4.  I don't say this is not
possible, though, but sound like nobody will try to go for 2.4.

Cheers,

-- 
Ducrot Bruno

--  Which is worse:  ignorance or apathy?
--  Don't know.  Don't care.


-------------------------------------------------------
This SF.NET email is sponsored by:
SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
http://www.vasoftware.com