Fedora Core 6 on the Dell X1 - discussion

Yep, I’ve installed the Fedora Core 6 on my Dell Latitude X1. Check out my Fedora Core 6 page (FC6) on the sidebar. Here we can discuss how things work out. Before asking here, you might want to check out my other post about Linux on the X1, that is mostly related to FC4 and FC5.

When available, I will of course move on to Fedora Core 6 stable.

29 Responses to “Fedora Core 6 on the Dell X1 - discussion”

  1. john Moylan Says:

    Hi Martin,

    I think you install Flash twice. Aren’t Adobe flash and Macromedia flash the same thing now?

  2. Martin Says:

    Thanks, I guess you’re right - I fixed it now.

  3. Austin Says:

    Heads up for those of you using the libsmbios packages from Dell to wake your backlight back up: the stock fedora kernel does not have the dcdbas module built in. You will have to recompile the kernel to add these features, I think.

  4. Mark Says:

    Hi Martin

    like you, im a fedora fan and i only use KDE (gnome is sooo boring)

    ive set the box up, tried to run the commands and ended up without window decorations (like what happened to you) BUT didnt end up with a working compiz stuff…

    i am on the verge of switching to mandriva actually… at work ive installed a laptop with mandriva free 2007.0 which has compiz enabled for KDE (wow)

    looks like im gonna switch distros for the siomple use of compiz, or even still on the window manager philosophy of the distro

    KDE for ever

  5. Martin Says:

    Hi Mark - yes KDE rules :)
    Regarding the compiz stuff, I expect it to be much more polished when the Beryl project release som rpms. I no longer need to run the ‘gnome-window-decorator --replace‘ manually as I have put a desktop link to that in my ~/.kde/Autorun folder. Are you running FC6 now? Have you checked that you have all the relevant lines in your ‘/etc/X11/xorg.conf‘ file? No error messages from compiz?

    I have also considered changing distro - I have even installed Kubuntu on a partition, but felt a bit uncomfortable with Debian. Remember that the kde-redhat yum repositories are live and kicking - KDE 3.5.5 will be available pretty soon according to their mailing list :)

  6. Mark Says:

    Well… im responding to this thread through my newly installed Mandriva 2007 box :-D

    I tried everything.. your suggestion, and even the one in this link http://liquidat.wordpress.com/2006/10/27/using-compiz-in-kde-on-fedora-core-6/

    all i get is this:
    - no window decorations
    - no pager
    - no functionality which is compiz-like
    - frustration

    I also tried to run into the dreaded gnome on fc6 and compiz didnt work either… so i moved my /home files and reformatted /boot and / with a Mandriva dvd… and it’s working like a dream

    debian is fine too, i have a virtual machine of it somewhere, but its slightly sluggish with old (but stable) software… but im more on the cutting edge

    you mentioned yum… well for debian there is apt, and for mandriva there is urpmi, so when kde3.5.5 komes out, ill definitely kdownload it :-D

  7. John Says:

    I have just setup FC6 with compiz on a (whitebox) Athlon 2800 PC with an Nvidia MX440 - the trick was to use the beta Nvida drivers because they have support for GLX_EXT

  8. Dennis Says:

    Hi Martin,

    i’ve set up FC6 (Zod) on my X1 (upgraded from FC5 actually) and my cpuspeed doesn’t work. It did work in FC5, like a charm. When booting, I get e message saying the module speedstep_centrino couldn’t be found. The funny thing is that in /etc/cpuspeed.conf, the module is written as speedstep-centrino. With a hyphen, instead of an underscore. If I change the hyphen to an underscore in /etc/cpuspeed.conf, the same message appears.

    I’ve also noticed, after the upgrade from FC5 to FC6, the i686 kernel-devel package was installed. Weird, cause the i586 kernel was installed. I ran into this problem when I was compiling my Cisco VPN client. The build ran fine, but insmodding the module didn’t work. I was wondering if I need to run the i686 or i586 kernel package in order to get cpuspeed working again. What kernel arch are you running on? (I upgraded using the DVD, not the yum method).

    Thanks for sharing your experience

  9. Martin Says:

    Hi Dennis.
    I do not use cpuspeed myself (I don’t have the service running); I use the kernel space cpu governors. You can list the available ones like this:

    # cat /sys/devices/system/cpu/cpu0/cpufreq/scaling_available_governors

    I have:

    powersave ondemand userspace performance

    Depending on which governors that are compiled into the kernel, you might need to do a “modprobe cpufreq-powersave” etc. List the active governor with:

    # cat /sys/devices/system/cpu/cpu0/cpufreq/scaling_governor

    and change the governor with:

    # echo -n ondemand > /sys/devices/system/cpu/cpu0/cpufreq/scaling_governor

    etc.

    By the way, from “man modprobe“:

    note that for convenience, there is no difference between _ and - in module names.

    About the kernel vs. kernel-devel mismatch I’ve experience the same on another system, but have not had any problems on my X1:

    $ rpm -q --qf '%{ARCH}' kernel
    i686i686
    $ rpm -q --qf '%{ARCH}' kernel-devel
    i686i686
    $ uname -a
    Linux x1 2.6.18-1.2798.fc6 #1 SMP Mon Oct 16 14:37:32 EDT 2006 i686 i686 i386 GNU/Linux

    If a normal ‘yum update‘ doesn’t help, I think you can obtain the correct kernel arch by doing (on your own risk - be sure the kernel is available in the repo first…):

    # rpm -e kernel --nodeps
    # yum install kernel.i686
    # rpm -e kernel-devel --nodeps
    # yum install kernel-devel.i686

    Good luck :)

  10. Martin Says:

    Mark: Thanks for the tip on “liquidat”’s blog - I’ve updated my page with info on Berel. It works for me on the Dell X1, but it is not as smooth as Compiz.

  11. Ian Thomas Says:

    Hi Martin,

    I’ve managed to get rid of the “libGL warning: 3D driver claims to not support visual 0×5b” on my Inspiron 1300 (i810 driver) by disabling AIGLX. To do this add the following to your xorg.conf (taken from fedoraproject.org/wiki):

    Section “ServerFlags”
    Option “AIGLX” “off”
    EndSection
    Section “Extensions”
    Option “Composite” “Disable”
    EndSection

    I assume the underlying problem is some incompatibility between the i810 driver and AIGLX that will be ironed out eventually. In the mean time, if you don’t disable AIGLX you may get some strange OpenGL errors. I was getting GLXBadContext errors when trying to render to an offscreen buffer, and this now works OK.

  12. Dennis Says:

    Hi again Martin,

    removing the kernel and reinstalling it the way you described it was a bad idea… the new kernel was supposed to copy information needed for the grub config. If no kernel is installed, any newly installed kernel can’t be booted because of the missing grub config. But hey, I managed to get my Linux installation booted again (and I learned yet another important lesson -> don’t remove a kernel before you’ve installed another one, kernel-xen for instance).

    I now have my X1 running the i686 kernel and cpu scaling works! Out of the box. Must have been that issue I mentioned earlier.

    BTW, everything else seems to be running fine. I also moved from the compiled Cisco VPN client (the official one) to vpnc together with NetworkManager-vpnc, works like a charm and I don’t need to recompile again whenever there is a kernel update

  13. Martin Says:

    Hi all. Just checkin’ in to say that I’ve updated some stuff on my FC6 page (speciallly to you that are subscribing to this post).
    I’ve made a “Open BUGS” section where I list bugs that I believe we can get rid of if we join forces and vote for them, plus provide more info to the developers. Registering at http://bugs.kde.org/ only takes a few minutes, and your contribution is what makes open source and free software powerful - isn’t it?

    Have a nice day!

  14. mark Says:

    Hi Everyone,

    I just found out that an up to date FC6 system includes the i810 “modesetting” driver. To enable it, just change “i810″ to “intel” in xorg.conf.

    This driver doesn’t use the VBIOS modes, so there is no need to run that 915resolution program or VBERestore. 1280×768 works well, and survives both a suspend-to-ram and suspend-to-disk. Also, the text consoles (alt-F1 ) still work after suspending and resuming. Some of you had this working before, but none of it worked for me until I switched from the i810 driver to the modesetting intel driver.

    It fixes one other thing for me: I recently got a wide screen external monitor (also 1280×768), and couldn’t get the X1 to drive it. Other people reported similar problems: driving an external monitor with a mode that had to be set by 915resolution caused a lot of us trouble. This bug describes the problem

    https://bugs.freedesktop.org/show_bug.cgi?id=643

    and reports these issues resolved with the modesetting driver.

  15. Martin Says:

    mark: Thanks - finally it works! I’ve updated my FC6 page with your suggestions, and can confirm that it works:

    * No 915resolution needed

    * Suspend to RAM works.

    * Suspend to disk works (hibernate)

    * Virtual consoles works

    * Works together with Beryl.

    This is great.

  16. Martin Says:

    Hold your horses… I just did a clean restart, and I guess some newly updated packages kicked in (maybe also a new kernel), and now my acpi scripts no longer work… The keyboard doesn’t respond on resume, and the screen stays black

    mark: Do you unload anything before you suspend? Do you not use vbetool at all? Are you using any extra options in your xorg.conf?

    I tried to suspend in a text console, without X running at all, and it locked up, so I think it might be something wrong with my /etc/acpi/actions/sleep.sh script. Maybe the unloading of modules.

    Anyway, you might want to skip my scripts for now. Suggestions are appreciated.

  17. Martin Says:

    Well, I put back my working acpi scripts in the /etc/ folder (that use 915resolution, vbetool, the i810 driver, etc, and works with compiz/beryl).

    Not sure what went wrong with the ‘intel’ driver though. If anyone have a working xorg.conf with Driver "intel" in it, that survives suspend and hibernate, please let us know :)

  18. mark Says:

    Hi Martin,

    Sorry to send you downt that dead end!

    It is still working well for me, but I notice that yum is telling me I now have 20 packages to update. Now I’m afraid to update them :-) Once I do I’ll report any problems.

    I took out all the extra options from my xorg.conf (VBERestore, ForceBIOS). I dont run the standalone vbetool either. For suspend scripts, I use the stock gnome ones, unmodified (I think /usr/sbin/pm-action). But I dont have my X1 with me so I cant check what pm-action is doing.

    Mark

  19. mark Says:

    Hi Martin,

    I completely updated my fc6 system a couple of days ago, and everything is still working well (suspend-to-ram, suspend-to-disk and text consoles) with the “intel” driver. So I dont know what could explain the problems you saw.

    Here’s my rather simple xorg.conf:

    Section “ServerLayout”
    Identifier “Default Layout”
    Screen 0 “Screen0″ 0 0
    InputDevice “Keyboard0″ “CoreKeyboard”
    InputDevice “Synaptics” “CorePointer”
    EndSection

    Section “InputDevice”
    Identifier “Keyboard0″
    Driver “kbd”
    Option “XkbModel” “pc105″
    Option “XkbLayout” “us”
    EndSection

    Section “InputDevice”
    Identifier “Synaptics”
    Driver “synaptics”
    Option “Device” “/dev/input/mice”
    Option “Protocol” “auto-dev”
    Option “Emulate3Buttons” “yes”
    EndSection

    Section “Device”
    Identifier “Videocard0″
    #Driver “i810″
    Driver “intel”
    Option “Clone” “True”
    Option “MonitorLayout” “CRT,LFP”
    Option “CloneRefresh” “60″
    EndSection

    Section “Screen”
    Identifier “Screen0″
    Device “Videocard0″
    DefaultDepth 24
    SubSection “Display”
    Viewport 0 0
    Depth 24
    Modes “1280×768″ “1024×768″ “960×576″
    EndSubSection
    EndSection

    # add this if you get strange opengl errors:
    #
    #Section “ServerFlags”
    # Option “AIGLX” “off”
    #EndSection
    #Section “Extensions”
    # Option “Composite” “Disable”
    #EndSection
    #

  20. Martin Says:

    Mark: Thanks for posting your xorg.conf — I will try this and report back when I have the time.

    Cheers :)

  21. Justin Says:

    Heya! Very useful information. Google listed your page as I searched for bluetooth information (working on a Toshiba Qosmio here). Thanks for listing the info on kbluetoothd!

    Also, I didn’t know KDE had created its own Yum repo. I’ve been using Gnome, but noted some apps I wanted from KDE, but the Fedora repos I had weren’t picking them up. Thanks for listing how to configure KDE’s repo.

    Cheers

  22. Martin Says:

    Glad I could help :)

  23. Dennis Says:

    Hi Martin,

    I followed Mark’s suggestions and I can confirm they work like a charm for me! Resuming from suspended or hibernated state doesn’t screw up my resolution anymore!

  24. Pat Says:

    Hi Martin,
    I’ve been following instructions on your website for quite some time now, but I have a recurring problem with my wireless which I don’t understand, nor could I get anybody better with linux than myself to understand. Basically, I haven’t been able to use wireless since last summer. When I’m in public places and try to connect, knetworkmanager will work for half a second and then die. At school, which we’ve been trying to set-up this week, they use wpa and I’ll paste parts of my wpa.log file below. It seems that the access point is not being recognized as using wpa. I wander if people on this list have had similar problems and if anybody has a clue as to how I can get my wireless to work.
    Thanks for your time, and also for your awesome web site.
    Pat

    Initializing interface ‘eth1′ conf ‘/etc/wpa_supplicant.conf’ driver ‘wext’ ctrl_interface ‘N/A’
    Configuration file ‘/etc/wpa_supplicant.conf’ -> ‘/etc/wpa_supplicant.conf’
    Reading configuration file ‘/etc/wpa_supplicant.conf’
    update_config=1
    ctrl_interface=’/var/run/wpa_supplicant’
    ctrl_interface_group=0
    eapol_version=1
    ap_scan=1
    fast_reauth=1
    Line: 664 - start of a new network block
    ssid - hexdump_ascii(len=3):
    75 76 61 uva
    scan_ssid=1 (0×1)
    key_mgmt: 0×8
    eap methods - hexdump(len=2): 15 00
    anonymous_identity - hexdump_ascii(len=16):
    61 6e 6f 6e 79 6d 6f 75 73 40 75 76 61 2e 6e 6c anonymous@uva.nl
    phase2 - hexdump_ascii(len=8):
    61 75 74 68 3d 50 41 50 auth=PAP
    identity - hexdump_ascii(len=14):
    70 67 69 72 61 72 64 40 75 76 61 2e 6e 6c pgirard@uva.nl
    password - hexdump_ascii(len=9): [REMOVED]
    ca_cert - hexdump_ascii(len=37):
    2f 65 74 63 2f 31 78 2f 63 65 72 74 73 2f 53 55 /etc/1x/certs/SU
    52 46 6e 65 74 2d 50 43 41 2d 52 6f 6f 74 2d 43 RFnet-PCA-Root-C
    41 2e 70 65 6d A.pem
    Line: 678 - start of a new network block
    ssid - hexdump_ascii(len=7):
    65 64 75 72 6f 61 6d eduroam
    scan_ssid=1 (0×1)
    key_mgmt: 0×8
    eap methods - hexdump(len=2): 15 00
    anonymous_identity - hexdump_ascii(len=16):
    61 6e 6f 6e 79 6d 6f 75 73 40 75 76 61 2e 6e 6c anonymous@uva.nl
    phase2 - hexdump_ascii(len=8):
    61 75 74 68 3d 50 41 50 auth=PAP
    identity - hexdump_ascii(len=14):
    70 67 69 72 61 72 64 40 75 76 61 2e 6e 6c pgirard@uva.nl
    password - hexdump_ascii(len=9): [REMOVED]
    ca_cert - hexdump_ascii(len=37):
    2f 65 74 63 2f 31 78 2f 63 65 72 74 73 2f 53 55 /etc/1x/certs/SU
    52 46 6e 65 74 2d 50 43 41 2d 52 6f 6f 74 2d 43 RFnet-PCA-Root-C
    41 2e 70 65 6d A.pem
    Priority group 0
    id=0 ssid=’uva’
    id=1 ssid=’eduroam’
    Initializing interface (2) ‘eth1′
    EAPOL: SUPP_PAE entering state DISCONNECTED
    EAPOL: KEY_RX entering state NO_KEY_RECEIVE
    EAPOL: SUPP_BE entering state INITIALIZE
    EAP: EAP entering state DISABLED
    EAPOL: External notification - portEnabled=0
    EAPOL: External notification - portValid=0
    SIOCGIWRANGE: WE(compiled)=21 WE(source)=18 enc_capa=0xf
    capabilities: key_mgmt 0xf enc 0xf
    Own MAC address: 00:12:f0:0c:43:8a
    wpa_driver_wext_set_wpa
    wpa_driver_wext_set_key: alg=0 key_idx=0 set_tx=0 seq_len=0 key_len=0
    wpa_driver_wext_set_key: alg=0 key_idx=1 set_tx=0 seq_len=0 key_len=0
    wpa_driver_wext_set_key: alg=0 key_idx=2 set_tx=0 seq_len=0 key_len=0
    wpa_driver_wext_set_key: alg=0 key_idx=3 set_tx=0 seq_len=0 key_len=0
    wpa_driver_wext_set_countermeasures
    wpa_driver_wext_set_drop_unencrypted
    Setting scan request: 0 sec 100000 usec
    Added interface eth1
    Wireless event: cmd=0×8b06 len=8
    RTM_NEWLINK, IFLA_IFNAME: Interface ‘eth1′ added
    RTM_NEWLINK, IFLA_IFNAME: Interface ‘eth1′ added
    Wireless event: cmd=0×8b19 len=8
    Received 3415 bytes of scan results (18 BSSes)
    Scan results: 18
    Selecting BSS from priority group 0
    0: 00:11:f5:dc:8b:3f ssid=’SpeedTouchDCE6E9′ wpa_ie_len=26 rsn_ie_len=0 caps=0×11
    skip - SSID mismatch
    skip - SSID mismatch
    1: 00:14:a9:d5:a0:40 ssid=’uva’ wpa_ie_len=0 rsn_ie_len=0 caps=0×11
    skip - no WPA/RSN IE
    2: 00:13:1a:96:7f:f0 ssid=’eduroam’ wpa_ie_len=0 rsn_ie_len=0 caps=0×11
    skip - no WPA/RSN IE
    3: 00:13:1a:96:7f:f0 ssid=’uva’ wpa_ie_len=0 rsn_ie_len=0 caps=0×11
    skip - no WPA/RSN IE
    4: 00:13:1a:40:3a:90 ssid=’eduroam’ wpa_ie_len=0 rsn_ie_len=0 caps=0×11
    skip - no WPA/RSN IE
    5: 00:13:1a:40:3a:90 ssid=’uva’ wpa_ie_len=0 rsn_ie_len=0 caps=0×11
    skip - no WPA/RSN IE
    6: 00:13:1a:96:78:90 ssid=’eduroam’ wpa_ie_len=0 rsn_ie_len=0 caps=0×11
    skip - no WPA/RSN IE
    7: 00:01:71:0c:59:26 ssid=’wifi-home-boven’ wpa_ie_len=0 rsn_ie_len=0 caps=0×11
    skip - no WPA/RSN IE
    8: 00:14:a9:d5:a0:40 ssid=’eduroam’ wpa_ie_len=0 rsn_ie_len=0 caps=0×11
    skip - no WPA/RSN IE
    9: 00:13:1a:40:4f:30 ssid=’uva’ wpa_ie_len=0 rsn_ie_len=0 caps=0×11
    skip - no WPA/RSN IE
    10: 00:13:1a:40:4f:30 ssid=’eduroam’ wpa_ie_len=0 rsn_ie_len=0 caps=0×11
    skip - no WPA/RSN IE
    11: 00:13:1a:96:78:90 ssid=’uva’ wpa_ie_len=0 rsn_ie_len=0 caps=0×11
    skip - no WPA/RSN IE
    12: 00:12:17:77:38:01 ssid=’sadako01′ wpa_ie_len=0 rsn_ie_len=0 caps=0×1
    skip - no WPA/RSN IE
    13: 02:0c:f1:71:fb:98 ssid=’hpsetup’ wpa_ie_len=0 rsn_ie_len=0 caps=0×2
    skip - no WPA/RSN IE
    14: 00:12:17:77:37:6e ssid=’sadako09′ wpa_ie_len=0 rsn_ie_len=0 caps=0×1
    skip - no WPA/RSN IE
    15: 00:12:17:77:37:cb ssid=’sadako07′ wpa_ie_len=0 rsn_ie_len=0 caps=0×1
    skip - no WPA/RSN IE
    16: 00:12:17:77:37:65 ssid=’sadako10′ wpa_ie_len=0 rsn_ie_len=0 caps=0×1
    skip - no WPA/RSN IE
    17: 00:12:17:77:32:fa ssid=’sadako06′ wpa_ie_len=0 rsn_ie_len=0 caps=0×1
    skip - no WPA/RSN IE
    selected non-WPA AP 00:14:a9:d5:a0:40 ssid=’uva’
    Trying to associate with 00:14:a9:d5:a0:40 (SSID=’uva’ freq=0 MHz)
    Cancelling scan request
    WPA: clearing own WPA/RSN IE
    Automatic auth_alg selection: 0×1
    WPA: clearing AP WPA IE
    WPA: clearing AP RSN IE
    WPA: clearing own WPA/RSN IE
    No keys have been configured - skip key clearing
    wpa_driver_wext_set_drop_unencrypted
    State: DISCONNECTED -> ASSOCIATING

  25. Martin Says:

    Hi Pat. Sorry, but I’ve only used WEP with my wireless, and I must say it now works very well. My only annoyance is that it sometimes connects to open untrusted networks (KNetworkManager bug 134402)

    These are my current versions:

    x1:~:$ /sbin/modinfo ipw2200|grep ^version
    version: 1.1.4kmprq
    x1:~:$ rpm -q knetworkmanager
    knetworkmanager-0.1-0.6.svn20061113.fc6

    Does anybody have any clues for Pat - please post’em :)

  26. Matt Rose Says:

    Hi all. I’ve just done a fresh install of FC6 and have cobbled together a working xorg config; suspend works, after I’ve pressed fn+up, but suspend to disk doesn’t work at all. It suspends ok, but when it comes back up, all I get is a white line in the top right of the screen; any ideas anyone?
    Any help would be much appreciated. I installed fC6 because I was fed up of XP taking anything from 5 to 45 minutes to wake up from hybernate.
    Thanks again

  27. Martin Says:

    Mark: You probably need to do the 915resolution trick right before the X server is brought back on wakeup from hibernate. Take a look at my sleep.sh script in my /etc/ folder (direct link: sleep.sh). Essentially:

    echo disk > /sys/power/state
    # Restore Vbios info before the xserver gets back
    /usr/sbin/915resolution 3c 1280 768

    (if you used “3c” in your /etc/rc.local file in the first place.

    Hm, my hibernate section on my FC6 page is a bit vague and outdated… sorry. I’m actually still not using the intel driver in my xorg.conf, although people have reported success using it. I just haven’t had time to try it yet ;)

  28. Matt Rose Says:

    Hi again. I fixed my blank screen after hybernate problem. To all using the the intel video driver, all I did to get round my problem was to add:
    Option “VBERestore” “true”
    to my xorg.conf. My X1 now hibernates and wakes up again while keeping 1280×768.
    I played with the 915resolution way of doing things, but the X1 woke up after suspend in 1024×768.
    Hope this helps. :)

  29. Morten Lind Says:

    [Original comment by Morten Lind 2007-10-18, reposted by webmaster bacause it was lost when the site was moved to the new location. Thanks for the comment by the way :) -Martin]

    Angående Skandiabanken og Konqueror har det lenge vært et problem for meg at jeg ikke kunne få lastet ned og installert sertifikat. (Konqueror er min favoritt nettleser.) Det fungerte, som du foreslog, ved å eksportere fra annen browser; i mitt tilfelde IceWeasel. Takk for den hjelpen!!!!

    Du nevner der etter at du har problem java script, og jeg eksperimenterte med nettleseridentifikasjon. Setter du identifikasjonen til FireFox 1.5.0.4, da fungerer det greit.

    Vennligst,
    Morten Lind.

Leave a Reply