Description of problem: S3 sleep causes X-display on notebook to remain dim (too dim to properly read) after resume. FN-key light controls can't be used any longer (no reaction). Hibernate will not show this effect . it will actually elevate this problem. Tried several hal-qirks but no result. If BIOS seleceted light control via light sensor is used situation with backlight is better, however controls remain unusable after suspend/resume. Kernel upgrade from 2.6.22.9-91.fc7 to 2.6.23.1-10.fc7 causes keyboard to fail in many occasions after suspend/resume causing a full stop (no X-server kill is possible via keyboard). System uses intel video driver (i810 & 915resol. will not work). Version-Release number of selected component (if applicable): libsmbios bin and libs are installed Version 0.13.10-1.fc7. xorg-x11-drv-keyboard-1.1.0-3.fc7 xorg-x11-drivers-7.2-6.fc7 xorg-x11-server-Xorg-1.3.0.0-9.fc7 xorg-x11-server-utils-7.2-1.fc7 xorg-x11-drv-i810-2.0.0-4.fc7 lshal reports: system.firmware.release_date = '08/08/2007' (string) system.firmware.version = 'A04' (string) udi = '/org/freedesktop/Hal/devices/dell_lcd_panel' info.addons = {'hald-addon-dell-backlight'} (string list) info.bus = 'unknown' (string) info.capabilities = {'laptop_panel'} (string list) info.category = 'laptop_panel' (string) info.interfaces = {'org.freedesktop.Hal.Device.LaptopPanel'} (string list) info.parent = '/org/freedesktop/Hal/devices/computer' (string) info.product = 'Dell Laptop Panel' (string) info.subsystem = 'unknown' (string) info.udi = '/org/freedesktop/Hal/devices/dell_lcd_panel' (string) laptop_panel.access_method = 'custom' (string) laptop_panel.brightness_in_hardware = false (bool) laptop_panel.num_levels = 1 (0x1) (int) org.freedesktop.Hal.Device.LaptopPanel.method_argnames = {'brightness_value', ''} (string list) org.freedesktop.Hal.Device.LaptopPanel.method_execpaths = {'hal-system-lcd-set-brightness', 'hal-system-lcd-get-brightn ess'} (string list) org.freedesktop.Hal.Device.LaptopPanel.method_names = {'SetBrightness', 'GetBrightness'} (string list) org.freedesktop.Hal.Device.LaptopPanel.method_signatures = {'i', ''} (string list) lspci shows the graphics to be: 00:02.0 VGA compatible controller: Intel Corporation Mobile Integrated Graphics Controller (rev 0c) and Xorg says in more detail: (--) Chipset 965GM found How reproducible: Backlight dim and unusable FN Keys to control backlight: 100% Keyboard fail after resume: about 60% Steps to Reproduce: 1. suspend notebook, either on panel using power manager or using pm-suspend 2. resume notebook (either powerswitch or opening the lid Actual results: backlight stays dim, dellLcdBrightness will not work setting values, Fn-Keys not working. Sometimes keyboard frozen (only FN-Esc will put the system to suspend mode again). Expected results: backlight at same level as before suspending or at least controllable. dellLcdBrightness should be able to control backlight. Fn-Keys are expected to work. Frozen keyboard should never occur. Additional info: Several notebooks show this behavior, even if installed with an i386 F7 system (eg. running x64_86 is not the problem).
Would some quirk (http://people.freedesktop.org/~hughsient/quirk/quirk-suspend-index.html) help? What I mean is, that you could try to go through http://people.freedesktop.org/~hughsient/quirk/quirk-suspend-debug.html and play with pm-suspend so long as it will be working for you. Please, let us know whether some quirk helps.
I just found out that my D620 needs the following quirks: pm-suspend --quirk-vbe-post --quirk-vbemode-restore These two quirks fix: https://bugzilla.redhat.com/show_bug.cgi?id=358631 and: https://bugzilla.redhat.com/show_bug.cgi?id=358671 Possibly these would help here?
Please, see https://bugzilla.redhat.com/show_bug.cgi?id=358631#c9 and continue accordingly in this bug as well.
I've already played days with pm-suspend and different setups of quirks in hal. Using /usr/share/hal/fdi/policy/10osvendor/10-dell-laptop-brightness.fdi and /usr/share/hal/fdi/information/10freedesktop/20-video-quirk-pm-dell.fdi and the modes described by Michael. I did also try power_management.quirk.no_fb, power_management.quirk.s3_bios, power_management.quirk.s3_mode, the latter both together with appropriate change to the kernel. No success. 64bit debugging is unfortunately not possible (as described on qirk-suspend-debug homepage). AFAIK D620 and D830 have _different_ video hardware (945 vs 965 chipsets). Quirks working to D620 need not neccessarily work.
OK, I give up -- could you please attach to this bug as separate uncompressed attachments your /etc/X11/xorg.conf and /var/log/Xorg.0.log and I will pass this bug to developers to sort it out.
Created attachment 250011 [details] The xorg.conf
Created attachment 250021 [details] Xorg.0.log Just contains hibernates.
Attached both requested files. I will capture another Xorg.0.log for with just one suspend in it. This one has several hibernates in it. Thank you for help!
I just came back to my D830. First I saw lots of updates including new kernel 2.6.23.1-21.fc7. I yum'ed them and tried suspending. While the suspend through PowerManager (2.18.3) did not let the system resume correctly (see above) it did do the suspend using pm-suspend --quirk-vbe-post --quirk-vbemode-restore as described by Michael. I don't overview what's effectively new, but a fully up to date Fedora 7 will work using this. I think this can be used to program hal. I'm satified. This issue can be closed by now. Thank you all, Aron