On Wed, 9 May 2007, Rafael J. Wysocki wrote: > On Tuesday, 8 May 2007 23:50, Tomas Pospisek's Mailing Lists wrote: >> On Tue, 8 May 2007, Rafael J. Wysocki wrote: >> >>> On Tuesday, 8 May 2007 22:58, Tomas Pospisek's Mailing Lists wrote: >>>> Hello Rafael >>>> >>>> On Tue, 8 May 2007, Rafael J. Wysocki wrote: >>>> >>>>> On Tuesday, 8 May 2007 22:02, Tomas Pospisek's Mailing Lists wrote: >>>>>> On Tue, 1 May 2007, Rafael J. Wysocki wrote: >>>>>> >>>>>>> On Tuesday, 1 May 2007 14:13, Tomas Pospisek's Mailing Lists wrote: >>>>>>>> On Tue, 1 May 2007, Rafael J. Wysocki wrote: >>>>>>>> >>>>>>>>> On Tuesday, 1 May 2007 10:13, Tomas Pospisek's Mailing Lists wrote: >>>>>>>>>> On the machine here: >>>>>>>>>> >>>>>>>>>> $ s2ram -i >>>>>>>>>> sys_vendor = "Fujitsu Siemens" >>>>>>>>>> sys_product = "P6VAP-AP" >>>>>>>>>> sys_version = " " >>>>>>>>>> bios_version = "6.00 PG" >>>>>>>>>> >>>>>>>>>> s2ram works just fine whether run from X or from the console. >>>>>>>>>> s2disk works also, *but only as long as X is not started*. >>>>>>>>> >>>>>>>>> Can you please check if you are able to reproduce the symptoms with >>>>>>>>> the >>>>>>>>> built-in swsusp, ie.: >>>>>>>>> >>>>>>>>> # echo 8 > /proc/sys/kernel/printk >>>>>>>>> # echo disk > /sys/power/state >>>>>>>> >>>>>>>> This method doesn't work at all. What I can see is that the following >>>>>>>> line is displayed: >>>>>>>> >>>>>>>> swsusp: Cannot find swapdevice, try swapon -a >>>>>>>> >>>>>>>> "swapon -a" works just fine. "swapon -s" shows me that the swap >>>>>>>> partition >>>>>>>> is in use. >>>>>>>> >>>>>>>> The effect of the suggested commands is that the command just returns >>>>>>>> back to the console with the following message: >>>>>>>> >>>>>>>> # echo 8 > /proc/sys/kernel/printk >>>>>>>> # echo disk > /sys/power/state >>>>>>>> bash: echo: write error: No such device >>>>>>>> # >>>>>>> >>>>>>> For this to work you need to append resume=<path_to_your_resume_device> >>>>>>> to >>>>>>> the kernel's command line and reboot (<path_to_your_resume_device> is >>>>>>> the >>>>>>> same as in the s2disk's configuration file). >>>>>> >>>>>> Thanks. >>>>>> >>>>>> It would be nice if there was an error message suggesting this. That >>>>>> would >>>>>> help the user. >>>>>> >>>>>> Now, I did what you suggested and the result is the same as with s2disk: >>>>>> the system goes dead. That is, it is still on in the sense that the >>>>>> computer leds, fan etc. are on, but the video card signal goes off (as my >>>>>> monitor tells me) and the PC takes no more input nor gives it any output. >>>>>> >>>>>> After a reset the filesystem is a bit corrupted, thus I get an FS check >>>>>> and things seems to work fine again. >>>>> >>>>> Really strange. >>>>> >>>>> Can you please check if doing >>>>> >>>>> # echo shutdown > /sys/power/disk >>>>> >>>>> before the 'echo disk > /sys/power/state' changes anything? >>>> >>>> Yes, it does seem to change something (however to be certain, I'd need to >>>> get half a dozen samples/tries I'd say). >>>> >>>> Now when I execute 'echo disk > /sys/power/state' the system switches to >>>> console mode (that is to VGA resolution) as it did previously, but now I >>>> see some activity. The systems tells me it's suspending tasks and I see a >>>> progress bar of sorts. But after that, same as before, the machine goes >>>> dumb. >>>> >>>> This time I pinged the machine too after it went dumb: no reply. I guess >>>> it's only electrically alive but the system is crashed on some level >>>> (possibly far) below userspace. >>> >>> Which X driver do you use? >> >> nvidia-kernel-legacy-2.6.18-4-686, Version: 1.0.7184+6 >> >> Thus Nvidia's own binary blob driver together with Xorg 1:7.1.0-18. >> >> (just for perspective and memory refresh: s2ram works with the driver and >> all) > > Yes, I remember. Still, this looks like a problem with X interacting with the > kernel in a wrong way and it's quite possible that the NVidia driver causes it > to happen. I'm afraid I can't help you any more with that.
Oh well, thanks a lot helping me to get onto the right track. From Nvidia's docu [1] for "legacy" Nvidia cards: "(app-s) APPENDIX S: POWER MANAGEMENT SUPPORT ____________________________________________________________________ This release includes support for APM based power management. This means that our driver will support suspend and resume, but will not support standby." "For ACPI, only S3 "Suspend to Ram" is currently supported. This means that S4 "Suspend to Disk", otherwise known as "Software Suspend" or "swsusp" does not currently work reliably." So there we have it: no s2disk with older Nvidia cards. Maybe there's hope for the future, but right now it doesn't work: "We hope to finish this support in the near future." And there's a HOWTO from SuSE which I'll try out [2]. Thanks again for going through this with me, *t [1] http://us.download.nvidia.com/XFree86/Linux-x86/1.0-7185/README/readme.txt [2] http://en.opensuse.org/NVidia_Suspend_HOWTO -- ----------------------------------------------------------- Tomas Pospisek http://sourcepole.com - Linux & Open Source Solutions ----------------------------------------------------------- ------------------------------------------------------------------------- This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ _______________________________________________ Suspend-devel mailing list Suspend-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/suspend-devel
