[lkml]   [2003]   [Oct]   [24]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
    SubjectRe: Wow. Suspend to disk works for me in test8. :)
    On Friday 24 October 2003 02:56, Pavel Machek wrote:
    > Hi!
    > > > > A couple of down sides I've noticed: I have to run "hwclock
    > > > > --hctosys" after a resume because the time you saved at is the time
    > > > > the system thinks it is when you resume (ouch). And because of that,
    > > > > things that should time out and renew themselves (like dhcp leases)
    > > > > have to be thumped manually.
    > > >
    > > > I sent fix for that yesterday... but you'd need to fix swsusp.c's
    > > > sysdev handling and mtrr-s => better wait.
    > > > Pavel
    > >
    > > It's largely working for me. My laptop's backed up regularly, so I'm not
    > > risking too much data. It reliably fails trying to suspend if I close
    > > the lid, and if I don't close the lid every once in a while the power
    > > down step won't power down immediately and the sucker will boot back up
    > > to the desktop and inform me that my dhcp lease file is corrupt, and then
    > > suddenly power down right from the desktop. (I reboot and force a full
    > > fsck in this circumstance.)
    > Well, this looks like ACPI problems to me. You might want to set it to
    > reboot and hit powerswitch manually.

    Why should it be an ACPI problem? Shutdown has never had trouble powering
    down since I got the laptop. Before Patrick forked off a suspend that
    actually worked for me, I'd shut down every time I wanted to move the laptop.
    (Ctrl-alt-backspace, ctrl-alt-delete was the fast way, I tweaked inittab to
    make that call shutdown -h.) I'd start a shutdown, and while red hat was
    running it's 8 gazillion unnecessary scripts I'd close the lid, put the thing
    in my backpack, and forget about it. It shut down every time, all the way
    back to at least -test3 (first version I installed on this laptop).

    Now I have to leave the lid open until it's finished suspending...

    > > I've also had it just hang there, on both suspend and resume, for upwards
    > > of 30 seconds doing nothing I can see until I start holding the power
    > > button down: after ten seconds it'll hard power off, but after two or
    > > three it suddenly wakes up and continues with the suspend or resume.
    > > (Suspend usually hangs in "snapshotting memory" or something like that.
    > > Resume hangs printing ........::::::::] at the end of the boot log, right
    > > before it would otherwise clear the screen and rerun the end of the power
    > > down phase.
    > Not sure what is going on there.

    Me neither. If I get some time this weekend I'll stick beep calls into the
    code and try to force it to fail. If I open the lid after it's hung the
    display is powered off, so I can't see any printout. No serial port on this
    machine to rig up a serial console. The "march of progress". Sigh...

    > I have similar hangs on omnibook xe3 when I do not load ohci driver
    > (but they happen during regular operation)....

    Nah. My usb scanner works on this puppy just fine when I bother to modprobe
    scanner. But I haven't plugged that in in a week. (And that has nothing to
    do with closing the lid, if USB was a problem you'd think it would happen
    when it was open.)

    I only had it fail to suspend (booting back up to the desktop with the
    filesystem horked) exactly once, and that was when I let the battery drop
    down to 0% charged. (It'll keep going for 15 minutes when it's like that,
    there's a reserve for suspend-to-ram. That might have been an ACPI problem
    somehow, yet if I'd told it to shutdown it would have (it always does, I've
    done it a lot, yes with the battery at 0%), and presumably ACPI is only being
    used to power the puppy off, the rest should just be a variant of module
    unloading, checkpointing process state, and flushing memory to swap...

    > Pavel

    To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
    the body of a message to
    More majordomo info at
    Please read the FAQ at

     \ /
      Last update: 2005-03-22 13:58    [W:0.025 / U:17.320 seconds]
    ©2003-2017 Jasper Spaans. hosted at Digital OceanAdvertise on this site