Again, thanks for all the input. Stefan my machine is new. I bought it less than two years ago and it was a very expensive laptop ;-> Its a Sony A190 (the ones with the giant screens that are completely unportable) and in Windows resume is pretty fast (its been a while since i tested it but it was < 5 secs I'd say). I'll check later but I also think windows doesn't consume battery nearly as quickly when suspending to ram (this is probably the issue Pavel brought up so I'll try and look into it).
When I wake the machine from a suspend the console comes back up instantly telling me that its "Restarting tasks" but then it sits there for about 30 seconds (I've timed it many times and its pretty consistent) before finally going to X. I've gone through the logs to try and find out where its spending the time but the log times don't seem reliable to me because I think suspend messages from before and after suspending often get mixed into the log. In any case its not clear to me from the log whether the messages are tasks being restarted after X is already up (such as dhcp, etc...) or while I still see the console. I also modified the 'restore_after_sleep' function in /usr/lib/powersave/scripts to print out the time before each message but it seems like all the things happening in that script happen almost instantaneously. It gets called from restore_after_suspend_to_ram which basically doesn't do anything else.
I don't really know how powersave is setup but I presume if I just call s2ram manually from the console that this does not run any scripts or anything on resume right? But when I try s2ram manually it still takes ~30 seconds to get me back to X windows. I see the console immediately but it just sits there saying 'Restarting tasks' so it seems to me like whatever is happening is happening in the kernel.
Any suggestions as to how I can debug this issue?
Many thanks,
Sheer
On 8/9/06, Stefan Seyfried <
[EMAIL PROTECTED]> wrote:
Hi,
On Wed, Aug 09, 2006 at 09:47:18AM +0200, Pavel Machek wrote:
> On Wed 2006-08-09 08:50:07, Stefan Seyfried wrote:
> > I think this highly depends on the specific machine's BIOS.
> > I have machines here that resume in less than 2 seconds and others that
> > take 20 seconds and more.
>
> Which take > 20 sec? Anything recent? I believe M$ specified they want
> it up in less than 2 sec, and most manufacturers follow that...
I think it was something old, SHARP PC-AR10 probably, nothing new.
It might also be, that the 20 seconds are mis-estimated :-)
Some machines seem to do pretty heavy DSDT processing on resume (armada
e500 has nice debug messages in its AML code :-) which could probably,
depending on the performance of the ACPI interpreter, also take quite
some time.
If i see a recent machine that takes long to resume, i will report it :-)
--
Stefan Seyfried | "Please, just tell people
QA / R&D Team Mobile Devices | to use KDE."
SUSE LINUX Products GmbH, Nürnberg | -- Linus Torvalds
------------------------------------------------------------------------- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________ Suspend-devel mailing list Suspend-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/suspend-devel
