Re: Booting - new idea?

2006-06-30 Thread Tim Dijkstra
On Thu, 29 Jun 2006 15:14:07 -0300
Gustavo Franco [EMAIL PROTECTED] wrote:

 On 6/29/06, Christian Perrier [EMAIL PROTECTED] wrote:
   Software suspend which exists in kernel for several years?
 
 
  And works properly in Debian since what? Weeks? Months?:-)
 
 It doesn't really work properly anywhere, does it? :-P

Has worked properly here very well with suspend2, which isn't in the 
stock kernel unfortunately.

Now it also works very well here with vanilla kernel 2.6.17 and the
debian package for uswsusp which I've packaged.

You can try them out if you like (you need a 2.6.17 kernel),

http://www.famdijkstra.org/~tdykstra/debian/uswsusp

But I hope they'll get uploaded very soon.

grts Tim


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: Booting - new idea?

2006-06-30 Thread Gustavo Franco

On 6/30/06, Tim Dijkstra [EMAIL PROTECTED] wrote:

On Thu, 29 Jun 2006 15:14:07 -0300
Gustavo Franco [EMAIL PROTECTED] wrote:

 On 6/29/06, Christian Perrier [EMAIL PROTECTED] wrote:
   Software suspend which exists in kernel for several years?
 
 
  And works properly in Debian since what? Weeks? Months?:-)

 It doesn't really work properly anywhere, does it? :-P

Has worked properly here very well with suspend2, which isn't in the
stock kernel unfortunately.


Oh, ok but my comment was about our kernel really.


Now it also works very well here with vanilla kernel 2.6.17 and the
debian package for uswsusp which I've packaged.

You can try them out if you like (you need a 2.6.17 kernel),

http://www.famdijkstra.org/~tdykstra/debian/uswsusp

But I hope they'll get uploaded very soon.



I'll take a look thanks, but did you see the recent discussion in lkml  ? Linus
came up with a patch to definitely solve some of the problems with the suspend
model. AFAIK, it's not yet into his tree or mm, but i hope to see the
changes merged in time for 2.6.18, really cool stuff there. I think
you can read a summary by Greg at lwn.

regards,
-- stratus


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: Booting - new idea?

2006-06-30 Thread Mike Hommey
On Fri, Jun 30, 2006 at 11:54:42AM +0200, Tim Dijkstra [EMAIL PROTECTED] 
wrote:
 On Thu, 29 Jun 2006 15:14:07 -0300
 Gustavo Franco [EMAIL PROTECTED] wrote:
 
  On 6/29/06, Christian Perrier [EMAIL PROTECTED] wrote:
Software suspend which exists in kernel for several years?
  
  
   And works properly in Debian since what? Weeks? Months?:-)
  
  It doesn't really work properly anywhere, does it? :-P
 
 Has worked properly here very well with suspend2, which isn't in the 
 stock kernel unfortunately.

Has worked properly here very well with whatever is in the debian
kernel, since 2.6.15 at least.

Mike


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: Booting - new idea?

2006-06-30 Thread Gustavo Franco

On 6/30/06, Mike Hommey [EMAIL PROTECTED] wrote:

On Fri, Jun 30, 2006 at 11:54:42AM +0200, Tim Dijkstra [EMAIL PROTECTED] 
wrote:
 On Thu, 29 Jun 2006 15:14:07 -0300
 Gustavo Franco [EMAIL PROTECTED] wrote:

  On 6/29/06, Christian Perrier [EMAIL PROTECTED] wrote:
Software suspend which exists in kernel for several years?
  
  
   And works properly in Debian since what? Weeks? Months?:-)
 
  It doesn't really work properly anywhere, does it? :-P

 Has worked properly here very well with suspend2, which isn't in the
 stock kernel unfortunately.

Has worked properly here very well with whatever is in the debian
kernel, since 2.6.15 at least.



Well Mike, maybe very well for you not for many users and some kernel
developers[0] agreed. Btw, Greg wrote an article for lwn (major
suspend changes),
read it there if you're subscribed. Really interesting content that shows the
current problems with the kernel implementation (up to 2.6.17) that should
be solved soon, since Linus came up with a interesting patch.

[0]  = 
http://thread.gmane.org/gmane.linux.power-management.general/1884/focus=1884

regards,
-- stratus


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: Booting - new idea?

2006-06-30 Thread Mike Hommey
On Fri, Jun 30, 2006 at 01:25:39PM -0300, Gustavo Franco [EMAIL PROTECTED] 
wrote:
 On 6/30/06, Mike Hommey [EMAIL PROTECTED] wrote:
 On Fri, Jun 30, 2006 at 11:54:42AM +0200, Tim Dijkstra 
 [EMAIL PROTECTED] wrote:
  On Thu, 29 Jun 2006 15:14:07 -0300
  Gustavo Franco [EMAIL PROTECTED] wrote:
 
   On 6/29/06, Christian Perrier [EMAIL PROTECTED] wrote:
 Software suspend which exists in kernel for several years?
   
   
And works properly in Debian since what? Weeks? Months?:-)
  
   It doesn't really work properly anywhere, does it? :-P
 
  Has worked properly here very well with suspend2, which isn't in the
  stock kernel unfortunately.
 
 Has worked properly here very well with whatever is in the debian
 kernel, since 2.6.15 at least.
 
 
 Well Mike, maybe very well for you not for many users and some kernel
 developers[0] agreed. Btw, Greg wrote an article for lwn (major
 suspend changes),
 read it there if you're subscribed. Really interesting content that shows 
 the
 current problems with the kernel implementation (up to 2.6.17) that should
 be solved soon, since Linus came up with a interesting patch.
 
 [0]  = 
 http://thread.gmane.org/gmane.linux.power-management.general/1884/focus=1884

Hum, this is about suspend to ram, not suspend to disk. Though suspend
to ram has been working almost flawlessly for me since 2.6.9 or 2.6.10.
(sometimes it freezes at wakeup time). I don't really care if the kernel
developpers say it's badly done. It just works. That's all I'm asking.

Mike


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: Booting - new idea?

2006-06-30 Thread Tim Dijkstra
On Fri, 30 Jun 2006 13:25:39 -0300
Gustavo Franco [EMAIL PROTECTED] wrote:

 On 6/30/06, Mike Hommey [EMAIL PROTECTED] wrote:
  On Fri, Jun 30, 2006 at 11:54:42AM +0200, Tim Dijkstra
  [EMAIL PROTECTED] wrote:
   On Thu, 29 Jun 2006 15:14:07 -0300
   Gustavo Franco [EMAIL PROTECTED] wrote:
  
On 6/29/06, Christian Perrier [EMAIL PROTECTED] wrote:
  Software suspend which exists in kernel for several years?


 And works properly in Debian since what? Weeks? Months?:-)
   
It doesn't really work properly anywhere, does it? :-P
  
   Has worked properly here very well with suspend2, which isn't in
   the stock kernel unfortunately.
 
  Has worked properly here very well with whatever is in the debian
  kernel, since 2.6.15 at least.
 
 
 Well Mike, maybe very well for you not for many users and some kernel
 developers[0] agreed. Btw, Greg wrote an article for lwn (major
 suspend changes),
 read it there if you're subscribed. Really interesting content that
 shows the current problems with the kernel implementation (up to
 2.6.17) that should be solved soon, since Linus came up with a
 interesting patch.
 
 [0]  =
 http://thread.gmane.org/gmane.linux.power-management.general/1884/focus=1884

I didn't read the whole thread, but AFAICS it only talks about a
(clever) hack to make debugging suspend-to-ram easier...

grts Tim


signature.asc
Description: PGP signature


Booting - new idea?

2006-06-29 Thread Arto Inkala
Hello,

booting takes usually tens of seconds and it's based on starting and
running different programs. Is it possible to skip this part of booting?

Particularly, is it possible to save the state of hardware and memories
during shutdown and upload them during boot time? I suppose, this would
minimize booting time, if it is technically possible to realize and
start programs this way?

If there is already this kind of project going on, I'd like to hear
about it.

Regards,
Arto


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: Booting - new idea?

2006-06-29 Thread Michal Čihař
Hi

On Thu, 29 Jun 2006 13:57:09 +0300
Arto Inkala [EMAIL PROTECTED] wrote:

 booting takes usually tens of seconds and it's based on starting and
 running different programs. Is it possible to skip this part of booting?
 
 Particularly, is it possible to save the state of hardware and memories
 during shutdown and upload them during boot time? I suppose, this would
 minimize booting time, if it is technically possible to realize and
 start programs this way?
 
 If there is already this kind of project going on, I'd like to hear
 about it.

Software suspend which exists in kernel for several years?

-- 
Michal Čihař | http://cihar.com | http://blog.cihar.com


signature.asc
Description: PGP signature


Re: Booting - new idea?

2006-06-29 Thread Yves-Alexis Perez
On Thu, 2006-06-29 at 13:57 +0300, Arto Inkala wrote:
 Hello,
 
 booting takes usually tens of seconds and it's based on starting and
 running different programs. Is it possible to skip this part of booting?
 
 Particularly, is it possible to save the state of hardware and memories
 during shutdown and upload them during boot time? I suppose, this would
 minimize booting time, if it is technically possible to realize and
 start programs this way?

You mean suspend to disk ?

-- 
Yves-Alexis Perez



Re: Booting - new idea?

2006-06-29 Thread Christian Perrier
 Software suspend which exists in kernel for several years?


And works properly in Debian since what? Weeks? Months? :-)




signature.asc
Description: Digital signature


Re: Booting - new idea?

2006-06-29 Thread Gustavo Franco

On 6/29/06, Christian Perrier [EMAIL PROTECTED] wrote:

 Software suspend which exists in kernel for several years?


And works properly in Debian since what? Weeks? Months?:-)


It doesn't really work properly anywhere, does it? :-P

regards,
-- stratus


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]