I wanted to keep this for a separate message: My experiences so far haven't been successful. /proc/acpi seems to be populated with all the right information, as best I can tell. However, with an empty /etc/acpi/actions and an empty (apart from sample.conf) /etc/acpi/events I guess that is not that surprising. I did, whilst in X try shutting the lid and opening it again to see what happened. This was the result: The machine didn't suspend -- the PCMCIA Wi-Fi card remained on and I could see network activity from the little flashing green activity indicator, I could hear some disk activity, and the fan remained on. However, the screen did blank, and when I opened the lid again it was a bit of a struggle to get it to come on again (had to hit CRT/LCD a couple of times, and ctrl-alt-F1 ctrl-alt-F7). When it did come on there was some unpleasant flickering. As I was unsure of what other things the ACPI lid event might have done I ended up rebooting. I also, whilst in X, tried as root to echo -n "standby" > /sys/power/state which momentarily dropped the machine out of X, flashed something on the screen that I didn't have a chance to read, and then returned me to X. Thereafter the FireWire drive was unavailable. Eventually I rebooted. Best, Darren -- ===================================================================== D. D. Brierton darren(a)dzr-web.com www.dzr-web.com Trying is the first step towards failure (Homer Simpson) =====================================================================

I have a dell inspiron 8600. Fairly new. ACPI seems to mostly work, except for suspend/resume. Little detail there. kernel-src/Documentation/power/video.txt seems to match the problem. I have not played with the various settings. (The one I did try once didn't work) The video card is a Nvidia GeForce varient. I'm hoping the next round of propriatary drivers will support suspend/resume. I have gotten _software_ suspend/resume mostly working (USB has issues. After a resume, unplugging a USB device after a resume will cause the machine to run like the turbo switch was accidentially hit.) but the most recent kernel updates appear to have removed some assembler bits that it needs. Advice: If your laptop supports APM, continue to use it. acpid is much more programmable, but since the video won't come back after an ACPI suspend/resume cycle unless you are lucky (see the video.txt file mentioned above), APM suspend will actually WORK. Which is probably more important. Doug -- Doug Kilpatrick kilpatds(a)oppositelock.org

Michal Jaegermann wrote: I have such a computer, an HP ze4315us model. It is good that re-enabled within the FC2 kernels. Battery status and thermals work with apci. I never tried to suspend this machine. Jim -- It's lucky you're going so slowly, because you're going in the wrong direction.

Hiho, On Jun 19, 2004, at 4:08, D. D. Brierton wrote: [...] My Hardware: IBM X31, Type 2672 (Cisco WLAN, no Bluetooth) My Software: FC2, kernel 2.6.6-1.435, acpid-1.0.2-6, no FAT partition for S2D, using supplied radeon kernel module for video, linuxant driverloader-1.71 for WLAN, radeontool-1.5 What works: - suspend to RAM works by simply closing lid. - laptop_mode works. Caveats: this setup requires pressing the power button (2-3 sec) to wake the machine up. Be careful not to press it too long, lest you execute /etc/acpi/events/sample.conf. Remove the latter if you do not want that risk. What does not work: - suspend to disk does not work, no surprise as I do not have the necessary FAT partition for the BIOS to handle it, and CONFIG_SOFTWARE_SUSPEND and CONFIG_PM_DISK are not set. (search the list if you want to enable one of the latter) - So far I've only once had video corruption on wake from suspend2ram (both X and vt), this was cured by a further suspend cycle. What is not tested: - the lightwatch tool from radeontool, I just close the lid when I want to save power. - test runs to see if removing the USB modules is still necessary, I just did not bother, adapt /etc/acpi/actions/suspend2ram.sh if you want to test URLs that were helpful when setting this up: (as the instructions are for Debian at php.co.co and not necessarily for FC2 at the other URLs, use common sense when following them, not all things apply the same way) - http://www.php.co.ba/X31/ - All X31 pages at http://www.linux-on-laptops.com/ibm.html The files: [/etc/acpi/events/lid] event=button/lid.* action=/etc/acpi/actions/lid.sh [end] [/etc/acpi/events/ac_adapter] event=ac_adapter.* action=/etc/acpi/actions/battery.sh [end] [/etc/acpi/events/sample.conf unmodified file from acppid-1.0.2-6] # This is a sample ACPID configuration event=button/power.* action=/sbin/shutdown -h now [end] [/etc/acpi/actions/lid.sh] #!/bin/sh # lid event handler status=`awk '/^state: / { print $2 }' /proc/acpi/button/lid/LID/state` case $status in "closed") echo "Bringing down vpnc" /usr/share/doc/vpnc-0.2rm+zomb.1/vpnc-disconnect echo "Suspending" # suspend2ram works on X31 # with kernel 2.6.6-1.435 # pcfe, 2004-06-17 /etc/acpi/actions/suspend2ram.sh exit 0 ;; *) echo "Lid event $status not handeled" ;; esac [end] [/etc/acpi/actions/suspend2ram.sh as you can see it is still crawling with echo statements for debug, once you feel confident sleep works as you want it, comment the echo lines out] #!/bin/sh # adaptation of: ############################ ## Author: mirza ## ## /etc/scripts/standby ## ## ## ## TestEnvironment: ## ## - IBM X31 ## ## - X-driver=VESA ## ## - kernel 2.6.5 ## ############################ # Changes made for Fedora Core 2 use by pcfe on 2004-06-17 ############################ ## start SUSPEND TO RAM ## ############################ # change to VT8 chvt 8 # Unmount any NFS or SMB filesystems: echo "unmounting NFS and SMB filesystems" umount -a -r -t nfs,smbfs # bluetooth off # just left in here in case I plug a Bluetooth dongle in and use it echo "bluetooth off" hciconfig hci0 down # remove modules echo "remove modules" #rmmod ipw2100 ## WLAN dldrstop ## stop linuxant driverloader #rmmod psmouse ## mouse rmmod ehci-hcd ## USB module rmmod uhci-hcd ## -||- #rmmod ohci-hcd ## -||- # update the disk super block echo "sync" sync # wait a second sleep 1 # turn the LCD off echo "backlight off" radeontool light off # suspend to ram echo "suspend to ram" echo -n mem > /sys/power/state ############################ ## resume from SUSPEND ## ############################ # wait 2 seconds to avoid double # or longe pressing of power button sleep 2 # set system time to hardware clock echo "set system time to hardware clock" hwclock --hctosys # load modules #modprobe psmouse # mouse module # don't load ipw2100. I'll load # it manually if I need it # modprobe ipw2100 # WLAN module # this would be driverloader in my case anyway # turn the LCD on echo "turn the LCD on" radeontool light on # go back to VT7 (X) echo "back to VT7" chvt 7 # /tmp/standby_state file will be created to be able # to avoid catching "LOCK" state by lightwatch.pl # see lightwatch.pl for details. if /tmp/standby_state # exists, lightwatch.pl will not react on this command touch /tmp/standby_state xscreensaver-command -lock 2> /dev/null 1> /dev/null rm /tmp/standby_state # start laptop mode if ac-adapter was # plugged out during the suspend mode # although linux should also do this # during standby mode #. /etc/acpi/acpi_handler.sh battery BAT0 2> /dev/null 1> /dev/null . /etc/acpi/actions/battery.sh 2> /dev/null 1> /dev/null ########### END ########### [end] [/etc/acpi/actions/battery.sh] #!/bin/sh # cpu throttling # cat /proc/acpi/processor/CPU0/throttling for more info ACAD_THR=0 BATT_THR=2 # spindown time for HD (man hdparm for valid values) # I prefer 2 hours for acad and 60 seconds for batt ACAD_HD=244 BATT_HD=12 # Power management level # 255 (off) on AC # 128 (medium on batt ACAD_PM=255 BATT_PM=64 # ac/battery event handler status=`awk '/^state: / { print $2 }' /proc/acpi/ac_adapter/AC/state` case $status in "on-line") echo "Setting HD spindown for AC mode." /sbin/laptop_mode stop /sbin/hdparm -S $ACAD_HD /dev/hda > /dev/null 2>&1 /sbin/hdparm -B $ACAD_PM /dev/hda > /dev/null 2>&1 #echo -n $ACAD_CPU:$ACAD_THR > /proc/acpi/processor/CPU0/limit exit 0 ;; "off-line") echo "Setting HD spindown for battery mode." /sbin/laptop_mode start /sbin/hdparm -S $BATT_HD /dev/hda > /dev/null 2>&1 /sbin/hdparm -B $BATT_PM /dev/hda > /dev/null 2>&1 #echo -n $BATT_CPU:$BATT_THR > /proc/acpi/processor/CPU0/limit exit 0 ;; esac [end] [add at the bottom of /etc/rc.local] # we run the detect AC on or off and detect laptop mode scrip once at boot # acpid will do this for us on AC changes /etc/acpi/actions/battery.sh [end] Plus you must set up the scripts you can extract from the file laptop-mode.txt in either the kernel doc RPM or the kernel source. Be sure to use the latest version. [...] That is normal, ACPI overrides APM. Both services can be on. ACPI will only activate if the machine supports it. APM will only activate if ACPI is not already on and the machine supports it. No need to change anything. The necessary patches are disabled (see above and search the list). If you decide to enable them, swap = ram should be enough as AFAIK the X31 uses shared memory for video. On laptops that do not you must add up as you suggested. Given today's disk sizes, I set up my swap to be RAM*2 on this machine anyway (1 Gig swap) You'll need the scripts from above Set up as above for S2R, make a FAT partition and a suspend file in there with phdisk under DOS or lphdisk under linux. (http://www.procyon.com/~pda/lphdisk/), as stated above, I have not tested this personally. RU PCFE -- ---------------------- http://www.pcfe.net/ ---------------------- Patrick C. F. Ernzer | e-mail: pcfe(a)pcfe.net -go-and-make-your-own-disclaimer-+-------------------------------- Some people assert that many hackers have a mild form of autism or Asperger's Syndrome, and are actually missing some of the brain circuitry that lubricates `normal' human social interaction. This may or may not be true. -- How To Ask Questions The Smart Way by Eric Steven Raymond

My experiences with a Thinkpad X40 and both ACPI and APM are in Bugzilla #125458: https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=125458 The basic synopsis is: ACPI busted (but less busted with the 2.6.5-1.358 kernel), APM busted unless you use the 2.6.5-1.358 i586 kernel, in which case it works some of the time. I haven't tried the latest 2.6.6-1.435 kernel as I have only sporadic access to the laptop now. - J<