Installing Ubuntu 8.10 (Intrepid Ibex) on a ThinkPad X61

From ThinkWiki
Revision as of 13:58, 21 July 2010 by Jdthood (Talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

I installed Ubuntu 8.10 on a ThinkPad X61 model 7673-CTO.

Some available accessories for this model

DMI info for this model

# dmidecode -s system-manufacturer
LENOVO
# dmidecode -s system-product-name
7673CTO
# dmidecode -s system-version
ThinkPad X61

glxgears output

$ glxgears
3581 frames in 5.0 seconds = 716.195 FPS
3704 frames in 5.0 seconds = 740.681 FPS
3687 frames in 5.0 seconds = 737.260 FPS
3601 frames in 5.0 seconds = 720.102 FPS

tp_smapi and hdaps

The tp_smapi package includes the tp_smapi driver and the customized hdaps driver.

To have the tp_smapi and hdaps_ec modules loaded on boot, add the lines

tp_smapi
hdaps_ec

to the /etc/modules file.

Version 0.37 of tp_smapi is included in Ubuntu Jaunty (in the linux-image-2.6.28-11-generic package).

Version 0.40 is available upstream.

The 0.40 version makes the following /sys files available:

$ cd /sys/devices/platform/smapi
$ ls -aF *
ac_connected  driver@  modalias  smapi_request  subsystem@  uevent

BAT0:
.                     current_avg      first_use_date   inhibit_charge_minutes  power_avg                remaining_running_time      temperature
..                    current_now      force_discharge  installed               power_now                remaining_running_time_now  voltage
barcoding             cycle_count      group0_voltage   last_full_capacity      remaining_capacity       serial
charging_max_current  design_capacity  group1_voltage   manufacture_date        remaining_charging_time  start_charge_thresh
charging_max_voltage  design_voltage   group2_voltage   manufacturer            remaining_percent        state
chemistry             dump             group3_voltage   model                   remaining_percent_error  stop_charge_thresh

BAT1:
.                     current_avg      first_use_date   inhibit_charge_minutes  power_avg                remaining_running_time      temperature
..                    current_now      force_discharge  installed               power_now                remaining_running_time_now  voltage
barcoding             cycle_count      group0_voltage   last_full_capacity      remaining_capacity       serial
charging_max_current  design_capacity  group1_voltage   manufacture_date        remaining_charging_time  start_charge_thresh
charging_max_voltage  design_voltage   group2_voltage   manufacturer            remaining_percent        state
chemistry             dump             group3_voltage   model                   remaining_percent_error  stop_charge_thresh

driver:
.  ..  bind  module  smapi  uevent  unbind

power:
.  ..  wakeup

subsystem:
.  ..  devices  drivers  drivers_autoprobe  drivers_probe  uevent

tp_smapi

tp_smapi allows ThinkPad battery charging to be controlled.

To set charge thresholds that are kinder to lithium ion batteries:

# cd /etc/init.d
# cat <<EOF >tp-smapi
#!/bin/sh
case "$1" in
  start)
	echo 40 > /sys/devices/platform/smapi/BAT0/start_charge_thresh
	echo 70 > /sys/devices/platform/smapi/BAT0/stop_charge_thresh
	;;
esac
EOF
# chmod ugo+x tp-smapi
# update-rc.d tp-smapi start 90 S .

hdaps

Can't load the standard Ubuntu hdaps module

# modprobe hdaps
FATAL: Error inserting hdaps (/lib/modules/2.6.27-11-generic/kernel/drivers/hwmon/hdaps.ko): No such device or address
# rmmod thinkpad_ec
# modprobe hdaps
#

The hdaps and the thinkpad_ec modules shipped in Intrepid and Jaunty (4 March 2009) conflict with each other. Either one can be loaded in the absence of the other. This is mentioned on the tp_smapi page.

With the hdaps module loaded, hdapsd prints an error message:

Setting up hdapsd (1:0.0.20070803-2) ...
 * Not starting hdapsd: /sys/block/hda/queue/protect does not exist,
 please read /usr/share/doc/hdapsd/README.Debian

The mentioned file says:

hdapsd needs the hdaps kernel module and the hdaps_protect patch
in the kernel so it WON'T work out-of-the-box on a Debian machine yet.
Look at http://thinkwiki.org/wiki/HDAPS for additional information.

Bottom line: The newer version of the hdaps driver furnished by the tp_smapi project is needed but neither Intrepid nor Jaunty (4 March 2009) includes this driver.

Bug reported: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/297213

Workaround

Install the tp_smapi and hdaps drivers built from the upstream sources. First install the Linux headers and necessary kernel build tools and download the tp_smapi tarball. Then:

$ tar xzvf tp_smapi-0.40.tgz
$ cd tp_smapi-0.40
$ sudo make load HDAPS=1
$ lsmod|head -4
Module                  Size  Used by
hdaps                  18648  0 
tp_smapi               29712  0 
thinkpad_ec            15000  2 hdaps,tp_smapi

hdaps at work

The image below shows hdaps-gl correctly portraying the spatial orientation of the laptop.

HDAPS working on ThinkPad X61.jpg

$ cd /sys/devices/platform/hdaps
$ ls -aF
./         invert             oversampling_ratio        sampling_rate
../        keyboard_activity  position                  subsystem@
calibrate  modalias           power/                    temp1
driver@    mouse_activity     running_avg_filter_order  uevent
$ cat invert
6

Docking

I had hoped that with Ubuntu I would be able to dock and undock the X61, but I was disappointed.

Hard freeze on undock from UltraBase

The ThinkPad X61 can be docked in the UltraBase X6 which has an Ultrabay Slim into which an optical drive can be installed such as the Ultrabay Slim Multi-Burner III Drive.

If the X61 is removed from the UltraBase then the X61 freezes, presumably because the kernel can no longer see the optical drive.

A sufficient solution is to "delete" the optical drive before undocking.

echo 1 >  /sys/class/scsi_device/1:0:0:0/device/delete

It doesn't seem to be necessary or even helpful to write to other files such as /sys/devices/platform/dock.0 or /proc/acpi/ibm/bay as some suggest.

However, it does seem to be necessary to delete the optical drive before pressing the "undock" button on the UltraBase X6.

After reattaching the UltraBase, this:

echo 0 0 0 > /sys/class/scsi_host/host1/scan

causes the kernel to notice the optical drive again.

Note that it is possible that the device number be something other than 1:0:0:0. Do, e.g.,

ls -l /sys/class/scsi_device/*:0:0:0/device/delete

to check.

Refs:

Bug reported: https://bugs.launchpad.net/ubuntu/+bug/242638

No /proc/acpi/ibm/dock

The file /proc/acpi/ibm/dock does not exist even though the thinkpad-acpi module is loaded and the X61 is booted while docked in the UltraBase X6. So, instructions in the thinkpad-acpi README and elsewhere that make reference to this file cannot be followed.

/sys/devices/platform/dock.0

There is a /sys/devices/platform/dock.0 directory with:

# cd /sys/devices/platform/dock.0
# ls -al
total 0
drwxr-xr-x  3 root root    0 2009-03-01 17:22 .
drwxr-xr-x 13 root root    0 2009-03-01 17:22 ..
-r--r--r--  1 root root 4096 2009-03-01 17:22 docked
-r--r--r--  1 root root 4096 2009-03-01 17:28 flags
-r--r--r--  1 root root 4096 2009-03-01 17:28 modalias
drwxr-xr-x  2 root root    0 2009-03-01 17:28 power
lrwxrwxrwx  1 root root    0 2009-03-01 17:22 subsystem -> ../../../bus/platform
-rw-r--r--  1 root root 4096 2009-03-01 17:22 uevent
-r--r--r--  1 root root 4096 2009-03-01 17:28 uid
--w-------  1 root root 4096 2009-03-01 17:28 undock

When not docked:

# cat /sys/devices/platform/dock.0/docked
0

When docked:

# cat /sys/devices/platform/dock.0/docked
1

In Jaunty at least, doing

echo 1 > /sys/devices/platform/dock.0/undock

does strange things. The first time I tried it it logged me out of GNOME but did not disconnect the external monitor connected to the dock. After I logged back in, the same command did cause the disconnection of the external monitor.

/proc/acpi/ibm/bay

When the thinkpad_acpi module is loaded there is a /proc/acpi/ibm/bay file with the following contents (with the X61 docked in the UltraBase X6 with a Ultrabay Slim Multi-Burner III Drive in the Ultrabay Slim port).

# cat bay
status:         occupied
commands:       eject

Doing

echo eject > /proc/acpi/ibm/bay

after

echo 1 >  /sys/class/scsi_device/1:0:0:0/device/delete

causes the light on the drive in the Ultrabay Slim to go out.

Doing the "eject" without first doing the "delete" mentioned earlier causes a system hang.

dock and bay modules

In Intrepid the lsmod command reveals that the dock and bay modules are loaded.

In Jaunty these modules are no longer present.

  • In Linux 2.6.28 the bay driver code was incorporated into the dock driver.
  • In Jaunty the dock driver is built integrally rather than modularly.
$ cd /boot
$ grep DOCK config*
config-2.6.27-11-generic:CONFIG_ACPI_DOCK=m
config-2.6.28-11-generic:CONFIG_ACPI_DOCK=y

Do something on undock and dock

To arrange for some action to be taken on undock and dock, add an /etc/udev/rules.d/55-thinkpad-local.rules file containing the following

KERNEL=="dock.0", ATTR{docked}=="1", RUN+="/etc/thinkpad/dock.sh 1"
KERNEL=="dock.0", ATTR{docked}=="0", RUN+="/etc/thinkpad/dock.sh 0"

and add an /etc/thinkpad/dock.sh file containing something like the following:

#!/bin/sh
DOCKED=$(cat /sys/devices/platform/dock.0/docked)
echo Docked: $DOCKED | logger -t $0
case "$DOCKED" in
        "0")
                # Do something
                ;;
        "1")
                # Do something else
                ;;
esac

Ref: http://marc.info/?l=linux-thinkpad&m=123416226107039&w=2

Keys

ThinkVantage and Zoom keys not seen by X

The ThinkVantage and Zoom keys are seen and reported by ACPI.

$ acpi_listen
ibm/hotkey HKEY 00000080 00001018
ibm/hotkey HKEY 00000080 00001014

They are seen and reported by input-events as KEY_VENDOR and KEY_ZOOM which according to /usr/include/linux/input.h have the values 0x168 and 0x174, respectively.

They are seen by showkey as having keycodes 0x168 (360 in decimal) and 0x174 (372 in decimal) respectively.

But, in Intrepid, they aren't seen by xev.

In Jaunty (4 March 2009) the ThinkVantage key is seen by xev but the Zoom key still is not.

Steve Langasek wrote on Ubuntu bug report log 217504:
X can only handly 256 distinct keycodes due to a protocol limitation, and due to evdev the keycodes are now standardized to match the Linux input layer - with no keycode below 256 that could reasonably be substituted for KEY_ZOOM. As a workaround, you can add your own fdi file to /etc/hal/fdi/information that maps this key to a different keycode (see /usr/share/hal/fdi/information/10freedesktop/30-keymap-module-thinkpad-acpi.fdi for an example of how this is currently done).

Refs:

Bug reported: https://bugs.launchpad.net/ubuntu/hardy/+source/linux/+bug/217504/

Fn-F5 emits KEY_RADIO instead of KEY_WLAN

Bug reported: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=504643

Table of keycodes for special keys and key combinations

While investigating the above problem I researched what key codes the various special keys and key combinations generate on this ThinkPad using various tools. The observations (which haven't been double-checked yet) are summarized in the following table. The ThinkVantage key is only seen by xev in Jaunty, not in Intrepid.

Key combination Extra function symbol Scan code
(showkey -s)
(hex)
Key code
(showkey -k)
Input event
(input-events)
ACPI event
(acpi_listen)
X event
(xev)
Keyboard Shortcuts name
⟦◁̸ (loudspeaker icon with stroke) e0 20 e0 a0 113 (none) (none) 121 XF86AudioMute
⟦◁▾ (loudspeaker icon, down arrow) e0 2e e0 ae 114 (none) (none) 122 XF86AudioLowerVolume
⟦◁▴ (loudspeaker icon, up arrow) e0 30 e0 b0 115 (none) (none) 123 XF86AudioRaiseVolume
ThinkVantage e0 1f e0 9f 360 KEY_VENDOR ibm/hotkey HKEY 00000080 00001018 156 XF86Launch1
Fn e0 63 e0 e3 143 (none) 151 XF86WakeUp
Fn-F1 (none) 466 KEY_FN_F1 ibm/hotkey HKEY 00000080 00001001 (none) (none)
Fn-F2 padlock e0 12 e0 92 152 KEY_SCREENLOCK ibm/hotkey HKEY 00000080 00001002 (none) XF86ScreenSaver
Fn-F3 battery e0 71 e0 f1 236 KEY_BATTERY ibm/hotkey HKEY 00000080 00001003 (none) (none)
Fn-F4 e0 5f e0 df 142; 142 KEY_SLEEP ibm/hotkey HKEY 00000080 00001004 (none) XF86Sleep
Fn-F5 radiating computer (none) 385 KEY_RADIO ibm/hotkey HKEY 00000080 00001005 (none) (none)
Fn-F6 (none) 471 KEY_FN_F6 ibm/hotkey HKEY 00000080 00001006 (none) (none)
Fn-F7 screen, line, computer e0 56 e0 d6 227 KEY_SWITCHVIDEOMODE ibm/hotkey HKEY 00000080 00001007 (none) (none)
Fn-F8 trackpoint, line, trackpad e0 79 e0 f9 192 KEY_F22 ibm/hotkey HKEY 00000080 00001008 (none) (none)
Fn-F9 computer, ⏏ 6f ef 194 KEY_F24 ibm/hotkey HKEY 00000080 00001009 (none) (none)
Fn-F10 e0 63 143 (none) (none) (none) XF86WakeUp
Fn-F11 (none) 476 KEY_FN_F11 ibm/hotkey HKEY 00000080 0000100b (none) (none)
Fn-F12 screen, arrow, diskpack e0 25 e0 a5 205 KEY_SUSPEND ibm/hotkey HKEY 00000080 0000100c (none) (none)
Fn-Home ☀▴ e0 54 e0 d4 e0 54 e0 d4 225; 225 KEY_BRIGHTNESSUP ibm/hotkey HKEY 00000080 00001010, video LCD0 00000086 00000000 (none) XF86MonBrightnessUp
Fn-End ☀▾ e0 4c e0 cc e0 4c e0 cc 224; 224 KEY_BRIGHTNESSDOWN ibm/hotkey HKEY 00000080 00001011, video LCD0 00000087 00000000 (none) XF86MonBrightnessDown
Fn-PgUp illuminated overhead lamp (none) (none) (none) (none) (none)
Fn-Space screen with magnifying glass ("zoom" symbol) (none) 372 KEY_ZOOM ibm/hotkey HKEY 00000080 00001014 (none) (none)
Fn- barred right arrowhead ("next" symbol) e0 19 e0 99 163 (none) (none) 171 XF86AudioNext
Fn- right arrowhead, double bar ("play/pause" symbol) e0 63 e0 e3 164 (none) (none) 172 XF86AudioPlay
Fn- barred left arrowhead ("prev" symbol) e0 10 e0 90 165 (none) (none) 173 XF86AudioPrev
Fn- ∎ ("stop play" symbol) e0 24 e0 a4 166 (none) (none) 174 XF86AudioStop
◂⎗ (left arrowhead, page icon) e0 6a e0 ea 158 (none) (none) 166 XF86Back
⎘▸ (page icon, right arrowhead) e0 69 e0 e9 159 (none) (none) 167 XF86Forward

Hotkey mask when the above table was compiled:

$ cat /sys/devices/platform/thinkpad_acpi/hotkey_mask
0x00fdffff
$ cat /sys/devices/platform/thinkpad_acpi/hotkey_all_mask
0x00ffffff
$ cat /sys/devices/platform/thinkpad_acpi/hotkey_recommended_mask
0x008c7fff
$ cat /sys/devices/platform/thinkpad_acpi/hotkey_report_mode
1

Compare Default meanings of special keys.

Suspend and resume

I had hoped that with Ubuntu I would be able to suspend and resume the X61, but I was disappointed. Presumably because of flaws in the ath5k driver, discussed below, the machine often hangs after resume.

Networking

Atheros AR5212 801.11 abg Wi-Fi card doesn't resume

Workaround for Intrepid: Add to /usr/lib/pm-utils/sleep.d/10NetworkManager:

 	thaw|resume)
+		ifconfig wifi0 up
 		resume_nm

This file disappears on upgrade to Jaunty.

In Jaunty the above problem does not occur. But there is a new problem. Sometimes the Atheros is braindead after resume. Reloading ath5k doesn't help. One has to cold-reboot.

Bugs reported:

Getting the Wi-Fi LED to flicker (with Atheros AR5212 801.11 abg Wi-Fi)

For Intrepid, add the following lines to /etc/sysctl.conf and reboot:

dev.wifi0.ledpin=1
dev.wifi0.softled=1

This does not work for Jaunty. In Jaunty there is no longer a Wi-Fi interface named wifi0; now there are interfaces wlan0 and wmaster0. Solution still being sought....

Network-manager slow to reconnect

Present in Intrepid.

Seemingly fixed in Jaunty.

Bug reported: https://bugs.launchpad.net/ubuntu/+source/network-manager/+bug/247381

Sound and video

No beep

Solution:

  • Open Volume Control (e.g., right click on loudspeaker icon in the panel and select "Open Volume Control").
  • In Preferences, select "Beep" in "tracks to be visible" list. Close.
  • On Playback tab, unmute Beep and raise its volume control.

Sound too quiet

The problem of sound being too quiet in Hardy does not appear to be present in Intrepid.

Sound crackles

In Jaunty (4 March 2009) the welcome jingle crackles.

This problem did not affect Intrepid.

Forgets screen resolution

In Jaunty (4 March 2009) when one logs back in, dual heads are always mirrored even if they have been configured previously. Resetting or restoring the configuration does not always fully succeed and the display is not always redrawn correctly.

Suspending and resuming sometimes fixes the problem.

Bug reported: https://bugs.launchpad.net/ubuntu/+source/screen-resolution-extra/+bug/337427

This problem seems (21 April 2009) to have been fixed during the runup to the release of Jaunty.

compiz and dual-head

The Intel Graphics Media Accelerator X3100 is limited to a maximum 3D texture size of 2048 × 2048.

$ glxinfo -l 2>/dev/null | grep GL_MAX_TEXTURE_SIZE | sed 's/.*=[^0-9]//g'
2048

This means that compiz visual effects can only be enabled if the virtual screen area is 2048 wide/high or less. This means that if one has an external display with a width greater than 1024 pixels, it can't be located beside the laptop display if you want wobbly windows and such. E.g., a 1680 × 1050 monitor beside the X61's 1024 × 768 LCD panel would require a virtual screen that is at least 2704 pixels wide.

A wide external display can be placed above or below the laptop display, however. E.g., the above monitor placed "above" the laptop panel requires a virtual screen height of 1818 pixels. ("Above" here of course just refers to the relative position set in System | Preferences | Screen Resolution.)

compiz-check

jdthood@triffid:/usr/local/sbin$ wget http://www.blogage.de/files/9124/download -O compiz-check
jdthood@triffid:/usr/local/sbin$  ./compiz-check 

Gathering information about your system...

 Distribution:          Ubuntu 9.04
 Desktop environment:   GNOME
 Graphics chip:         Intel Corporation Mobile GM965/GL960 Integrated Graphics Controller (rev 0c)
 Driver in use:         intel
 Rendering method:      AIGLX

Checking if it's possible to run Compiz on your system...

 Checking for texture_from_pixmap...               [ OK ]
 Checking for non power of two support...          [ OK ]
 Checking for composite extension...               [ OK ]
 Checking for FBConfig...                          [ OK ]
 Checking for hardware/setup problems...           [WARN]

Something potential problematic has been detected with your setup:
 Warning: PCI ID 8086:2a02 detected. 

Would you like to know more? (Y/n) y

 Your particular graphics chip may be blacklisted on certain distributions.
 However that does not necessarily mean you will not be able to run Compiz.

 You can skip this blacklist -- but keep in mind that you did so.
 Do not complain if you encounter any problems with Compiz afterwards. 

Do you want to skip blacklist checks by Compiz? (y/N) 

Movies jitter

Ref: http://marc.info/?l=linux-thinkpad&m=123180105228016&w=2