<?xml version="1.0"?>
<feed xmlns="http://www.w3.org/2005/Atom" xml:lang="en">
	<id>https://www.thinkwiki.org/w/api.php?action=feedcontributions&amp;feedformat=atom&amp;user=DF5JT</id>
	<title>ThinkWiki - User contributions [en]</title>
	<link rel="self" type="application/atom+xml" href="https://www.thinkwiki.org/w/api.php?action=feedcontributions&amp;feedformat=atom&amp;user=DF5JT"/>
	<link rel="alternate" type="text/html" href="https://www.thinkwiki.org/wiki/Special:Contributions/DF5JT"/>
	<updated>2026-04-06T01:37:17Z</updated>
	<subtitle>User contributions</subtitle>
	<generator>MediaWiki 1.31.12</generator>
	<entry>
		<id>https://www.thinkwiki.org/w/index.php?title=Problem_with_CPU_frequency_scaling&amp;diff=49938</id>
		<title>Problem with CPU frequency scaling</title>
		<link rel="alternate" type="text/html" href="https://www.thinkwiki.org/w/index.php?title=Problem_with_CPU_frequency_scaling&amp;diff=49938"/>
		<updated>2010-11-04T22:11:27Z</updated>

		<summary type="html">&lt;p&gt;DF5JT: &lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;{{stub}}&lt;br /&gt;
&lt;br /&gt;
==Symptom==&lt;br /&gt;
If the battery pack is removed and the laptop is powered by AC only, the CPU downclocks to the lowest multiplier and remains locked in low speed. This happens irrespective of the scaling method in use (kernelspace or userspace) and of the frequency governor selected (ondemand, performance, etc.).&lt;br /&gt;
&lt;br /&gt;
Output of cpufreq-info (on AC power without battery):&lt;br /&gt;
&lt;br /&gt;
  analyzing CPU 0:&lt;br /&gt;
  driver: centrino&lt;br /&gt;
  CPUs which need to switch frequency at the same time: 0 1&lt;br /&gt;
  hardware limits: 1000 MHz - 1.83 GHz&lt;br /&gt;
  available frequency steps: 1.83 GHz, 1.33 GHz, 1000 MHz&lt;br /&gt;
  available cpufreq governors: ondemand, userspace, powersave, performance&lt;br /&gt;
  '''current policy: frequency should be within 1000 MHz and 1000 MHz.'''&lt;br /&gt;
                  The governor &amp;quot;ondemand&amp;quot; may decide which speed to use&lt;br /&gt;
                  within this range.&lt;br /&gt;
  current CPU frequency is 1000 MHz (asserted by call to hardware).&lt;br /&gt;
  &lt;br /&gt;
  analyzing CPU 1:&lt;br /&gt;
  driver: centrino&lt;br /&gt;
  CPUs which need to switch frequency at the same time: 0 1&lt;br /&gt;
  hardware limits: 1000 MHz - 1.83 GHz&lt;br /&gt;
  available frequency steps: 1.83 GHz, 1.33 GHz, 1000 MHz&lt;br /&gt;
  available cpufreq governors: ondemand, userspace, powersave, performance&lt;br /&gt;
  '''current policy: frequency should be within 1000 MHz and 1000 MHz.'''&lt;br /&gt;
                  The governor &amp;quot;ondemand&amp;quot; may decide which speed to use&lt;br /&gt;
                  within this range.&lt;br /&gt;
  current CPU frequency is 1000 MHz (asserted by call to hardware).&lt;br /&gt;
&lt;br /&gt;
Output of cpufreq-info (on AC power with battery plugged in):&lt;br /&gt;
&lt;br /&gt;
  analyzing CPU 0:&lt;br /&gt;
  driver: centrino&lt;br /&gt;
  CPUs which need to switch frequency at the same time: 0 1&lt;br /&gt;
  hardware limits: 1000 MHz - 1.83 GHz&lt;br /&gt;
  available frequency steps: 1.83 GHz, 1.33 GHz, 1000 MHz&lt;br /&gt;
  available cpufreq governors: ondemand, userspace, powersave, performance&lt;br /&gt;
  '''current policy: frequency should be within 1000 MHz and 1.83 GHz.'''&lt;br /&gt;
                  The governor &amp;quot;ondemand&amp;quot; may decide which speed to use&lt;br /&gt;
                  within this range.&lt;br /&gt;
  current CPU frequency is 1000 MHz (asserted by call to hardware).&lt;br /&gt;
  &lt;br /&gt;
  analyzing CPU 1:&lt;br /&gt;
  driver: centrino&lt;br /&gt;
  CPUs which need to switch frequency at the same time: 0 1&lt;br /&gt;
  hardware limits: 1000 MHz - 1.83 GHz&lt;br /&gt;
  available frequency steps: 1.83 GHz, 1.33 GHz, 1000 MHz&lt;br /&gt;
  available cpufreq governors: ondemand, userspace, powersave, performance&lt;br /&gt;
  '''current policy: frequency should be within 1000 MHz and 1.83 GHz.'''&lt;br /&gt;
                  The governor &amp;quot;ondemand&amp;quot; may decide which speed to use&lt;br /&gt;
                  within this range.&lt;br /&gt;
  current CPU frequency is 1000 MHz (asserted by call to hardware).&lt;br /&gt;
&lt;br /&gt;
==Solution==&lt;br /&gt;
Other than using the laptop with the battery pack always inserted, no solution is known at this point.&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
'''Update:'''&lt;br /&gt;
&lt;br /&gt;
On my Z61m with a T5500 I had the same problem.&lt;br /&gt;
After several hours of debugging I noticed, that my ACPI BIOS returns another _PPC value (2) than at boot-time with batery (0).&lt;br /&gt;
The Kernel-code says:&lt;br /&gt;
        /*&lt;br /&gt;
         * _PPC indicates the maximum state currently supported by the platform&lt;br /&gt;
         * (e.g. 0 = states 0..n; 1 = states 1..n; etc.&lt;br /&gt;
         */&lt;br /&gt;
Thats mean, that my Bios says, after batery release only state 2 is possible. There is already a (for me unknown and undocumented) Patch for such a worse ACPI BIOS:&lt;br /&gt;
&lt;br /&gt;
http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=623b78c39c4525731f852072edd742cc4fba6786&lt;br /&gt;
&lt;br /&gt;
You only have to add &amp;quot;processor.ignore_ppc=1&amp;quot; to your kernel parameters and everything works fine.&lt;br /&gt;
&lt;br /&gt;
Tested with 2.6.26-rc1 on a Z61m with a 1.67GHz T5500 Processor.&lt;br /&gt;
&lt;br /&gt;
'''Update 2:'''&lt;br /&gt;
&lt;br /&gt;
The Patch above seems to be available since 2.6.25&lt;br /&gt;
&lt;br /&gt;
== Affected Models ==&lt;br /&gt;
{{Z61m}}&amp;lt;br /&amp;gt;&lt;br /&gt;
{{T61}}&amp;lt;br /&amp;gt;&lt;br /&gt;
{{T60p}}&lt;/div&gt;</summary>
		<author><name>DF5JT</name></author>
		
	</entry>
	<entry>
		<id>https://www.thinkwiki.org/w/index.php?title=Problem_with_CPU_frequency_scaling&amp;diff=49937</id>
		<title>Problem with CPU frequency scaling</title>
		<link rel="alternate" type="text/html" href="https://www.thinkwiki.org/w/index.php?title=Problem_with_CPU_frequency_scaling&amp;diff=49937"/>
		<updated>2010-11-04T22:10:58Z</updated>

		<summary type="html">&lt;p&gt;DF5JT: &lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;{{stub}}&lt;br /&gt;
&lt;br /&gt;
==Symptom==&lt;br /&gt;
If the battery pack is removed and the laptop is powered by AC only, the CPU downclocks to the lowest multiplier and remains locked in low speed. This happens irrespective of the scaling method in use (kernelspace or userspace) and of the frequency governor selected (ondemand, performance, etc.).&lt;br /&gt;
&lt;br /&gt;
Output of cpufreq-info (on AC power without battery):&lt;br /&gt;
&lt;br /&gt;
  analyzing CPU 0:&lt;br /&gt;
  driver: centrino&lt;br /&gt;
  CPUs which need to switch frequency at the same time: 0 1&lt;br /&gt;
  hardware limits: 1000 MHz - 1.83 GHz&lt;br /&gt;
  available frequency steps: 1.83 GHz, 1.33 GHz, 1000 MHz&lt;br /&gt;
  available cpufreq governors: ondemand, userspace, powersave, performance&lt;br /&gt;
  '''current policy: frequency should be within 1000 MHz and 1000 MHz.'''&lt;br /&gt;
                  The governor &amp;quot;ondemand&amp;quot; may decide which speed to use&lt;br /&gt;
                  within this range.&lt;br /&gt;
  current CPU frequency is 1000 MHz (asserted by call to hardware).&lt;br /&gt;
  &lt;br /&gt;
  analyzing CPU 1:&lt;br /&gt;
  driver: centrino&lt;br /&gt;
  CPUs which need to switch frequency at the same time: 0 1&lt;br /&gt;
  hardware limits: 1000 MHz - 1.83 GHz&lt;br /&gt;
  available frequency steps: 1.83 GHz, 1.33 GHz, 1000 MHz&lt;br /&gt;
  available cpufreq governors: ondemand, userspace, powersave, performance&lt;br /&gt;
  '''current policy: frequency should be within 1000 MHz and 1000 MHz.'''&lt;br /&gt;
                  The governor &amp;quot;ondemand&amp;quot; may decide which speed to use&lt;br /&gt;
                  within this range.&lt;br /&gt;
  current CPU frequency is 1000 MHz (asserted by call to hardware).&lt;br /&gt;
&lt;br /&gt;
Output of cpufreq-info (on AC power with battery plugged in):&lt;br /&gt;
&lt;br /&gt;
  analyzing CPU 0:&lt;br /&gt;
  driver: centrino&lt;br /&gt;
  CPUs which need to switch frequency at the same time: 0 1&lt;br /&gt;
  hardware limits: 1000 MHz - 1.83 GHz&lt;br /&gt;
  available frequency steps: 1.83 GHz, 1.33 GHz, 1000 MHz&lt;br /&gt;
  available cpufreq governors: ondemand, userspace, powersave, performance&lt;br /&gt;
  '''current policy: frequency should be within 1000 MHz and 1.83 GHz.'''&lt;br /&gt;
                  The governor &amp;quot;ondemand&amp;quot; may decide which speed to use&lt;br /&gt;
                  within this range.&lt;br /&gt;
  current CPU frequency is 1000 MHz (asserted by call to hardware).&lt;br /&gt;
  &lt;br /&gt;
  analyzing CPU 1:&lt;br /&gt;
  driver: centrino&lt;br /&gt;
  CPUs which need to switch frequency at the same time: 0 1&lt;br /&gt;
  hardware limits: 1000 MHz - 1.83 GHz&lt;br /&gt;
  available frequency steps: 1.83 GHz, 1.33 GHz, 1000 MHz&lt;br /&gt;
  available cpufreq governors: ondemand, userspace, powersave, performance&lt;br /&gt;
  '''current policy: frequency should be within 1000 MHz and 1.83 GHz.'''&lt;br /&gt;
                  The governor &amp;quot;ondemand&amp;quot; may decide which speed to use&lt;br /&gt;
                  within this range.&lt;br /&gt;
  current CPU frequency is 1000 MHz (asserted by call to hardware).&lt;br /&gt;
&lt;br /&gt;
==Solution==&lt;br /&gt;
Other than using the laptop with the battery pack always inserted, no solution is known at this point.&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
'''Update:'''&lt;br /&gt;
&lt;br /&gt;
On my Z61m with a T5500 I had the same problem.&lt;br /&gt;
After several hours of debugging I noticed, that my ACPI BIOS returns another _PPC value (2) than at boot-time with batery (0).&lt;br /&gt;
The Kernel-code says:&lt;br /&gt;
        /*&lt;br /&gt;
         * _PPC indicates the maximum state currently supported by the platform&lt;br /&gt;
         * (e.g. 0 = states 0..n; 1 = states 1..n; etc.&lt;br /&gt;
         */&lt;br /&gt;
Thats mean, that my Bios says, after batery release only state 2 is possible. There is already a (for me unknown and undocumented) Patch for such a worse ACPI BIOS:&lt;br /&gt;
&lt;br /&gt;
http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=623b78c39c4525731f852072edd742cc4fba6786&lt;br /&gt;
&lt;br /&gt;
You only have to add &amp;quot;processor.ignore_ppc=1&amp;quot; to your kernel parameters and everything works fine.&lt;br /&gt;
&lt;br /&gt;
Tested with 2.6.26-rc1 on a Z61m with a 1.67GHz T5500 Processor.&lt;br /&gt;
&lt;br /&gt;
'''Update 2:'''&lt;br /&gt;
&lt;br /&gt;
The Patch above seems to be available since 2.6.25&lt;br /&gt;
&lt;br /&gt;
== Affected Models ==&lt;br /&gt;
{{Z61m}}&amp;lt;br /&amp;gt;&lt;br /&gt;
{{T61}}{{T60p}}&lt;/div&gt;</summary>
		<author><name>DF5JT</name></author>
		
	</entry>
	<entry>
		<id>https://www.thinkwiki.org/w/index.php?title=ATI_Mobility_FireGL_V5200&amp;diff=42639</id>
		<title>ATI Mobility FireGL V5200</title>
		<link rel="alternate" type="text/html" href="https://www.thinkwiki.org/w/index.php?title=ATI_Mobility_FireGL_V5200&amp;diff=42639"/>
		<updated>2009-04-12T09:17:17Z</updated>

		<summary type="html">&lt;p&gt;DF5JT: /* Linux X.Org driver */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;__NOTOC__&lt;br /&gt;
{| width=&amp;quot;100%&amp;quot;&lt;br /&gt;
|style=&amp;quot;vertical-align:top&amp;quot; |&lt;br /&gt;
&amp;lt;div style=&amp;quot;margin: 0; margin-right:10px; border: 1px solid #dfdfdf; padding: 0em 1em 1em 1em; background-color:#F8F8FF; align:right;&amp;quot;&amp;gt;&lt;br /&gt;
&lt;br /&gt;
This is an ATI video adapter&lt;br /&gt;
&lt;br /&gt;
== Features ==&lt;br /&gt;
* Chipset: ATI M56 - Similar to Radeon X1600&lt;br /&gt;
* PCI ID: 1002:71c4&lt;br /&gt;
* PCI Express x16&lt;br /&gt;
* 256 MB GDDR3-SDRAM video memory&lt;br /&gt;
&amp;lt;/div&amp;gt;&lt;br /&gt;
|style=&amp;quot;vertical-align:top&amp;quot; |&lt;br /&gt;
|}&lt;br /&gt;
&lt;br /&gt;
=== Linux X.Org driver ===&lt;br /&gt;
Initial support included in the stable branch of the Xorg [[radeon]] driver (2D only). XVideo acceleration and 3D support have been added by the xorg driver developers and the driver is used by all majos distributions. The driver can be considered stable and fully supports Suspend-To-Disk and Suspend-To-RAM as well as XVideo acceleration in combination with composite extensions (compiz).&lt;br /&gt;
&lt;br /&gt;
In addition to the radeon driver the card is also supported by the [[radeonhd]] driver. A Debian package already exists in the experimental branch. 3d is still under development and does not work satisfactorily on the T60p.&lt;br /&gt;
&lt;br /&gt;
Recent releases of the ATI proprietary driver do not support this card anymore. Ever since ATI has been bought by AMD the specs for this card have been released to driver developers.&lt;br /&gt;
&lt;br /&gt;
==== ThinkPad LCD ====&lt;br /&gt;
&lt;br /&gt;
All current distributions offer full support for the 1600x1200 resolution of the screen.&lt;br /&gt;
&lt;br /&gt;
==== External VGA port ====&lt;br /&gt;
Yes&lt;br /&gt;
&lt;br /&gt;
==== SVideo port ====&lt;br /&gt;
No&lt;br /&gt;
&lt;br /&gt;
==== DVI port ====&lt;br /&gt;
No (yes on Docking Station, working with [[fglrx]] driver&lt;br /&gt;
&lt;br /&gt;
=== Proprietary ATI driver ===&lt;br /&gt;
&lt;br /&gt;
'''The following information is outdated only relevant if you use an older distribution.&lt;br /&gt;
'''&lt;br /&gt;
&lt;br /&gt;
Proprietary [[fglrx]] driver works since version 8.25.18, version 8.24.8 works by adding &amp;quot;ChipID 0x71c5&amp;quot; to the xorg.conf &amp;quot;Device&amp;quot; statement and fooling it into thinking it's a X1600 mobility (do this only as a last resort or if you're sure you have an older driver; it is fatal to version 8.25.18).&lt;br /&gt;
&lt;br /&gt;
The new driver 8.25.18 seems to support the card, see&lt;br /&gt;
https://support.ati.com/ics/support/KBAnswer.asp?questionID=1643&lt;br /&gt;
(successfully tested with the provided autorun installation script on OpenSuSE 10.1, T60p with 15&amp;quot; display).&lt;br /&gt;
&lt;br /&gt;
(K)ubuntu 6.06 comes with this driver and it works out of the box. However, it seems to have problems with VT Switching (Alt+Fn). Instead of switching to a console, X resets and goes back to the log in screen (This doesn't happen when using the [[vesa]] driver). The switching issue might be resolvable by setting ENABLE_LAPTOP_MODE=true in /etc/default/acpi-support, but it doesn't seem to help on all systems. &lt;br /&gt;
&lt;br /&gt;
{{NOTE|On Ubuntu 6.10 (Edgy) with 'splash' on the command line, you cannot switch to VTs. This is a Thinkpad T60. The screen remains graphical, but is very badly garbled.  There is no text. X does not reset, and you can switch back to your X session. The solution is to include 'vga-791' or to remove 'splash' from the grub or lilo command line. You don't need to do both.  (It's been a few days, so I don't remember where I found this...I just chanced upon this wiki page for the video card I'm trying to get working with xgl/AIGLX and Beryl).}}&lt;br /&gt;
   &lt;br /&gt;
Also, the pointer gets garbled sometimes in X, but this is only a minor annoyance. Dual Head (Xinerama) also works with this driver, but don't use KDE control center to enable it (use the ATi tool or edit xorg.conf manually)  &lt;br /&gt;
&lt;br /&gt;
I have a feeling that this card gets in the way when trying to suspend to RAM (resuming with a complete blank screen, lights are blinking, but that's it), but this can also be caused by other means.&lt;br /&gt;
&lt;br /&gt;
=== Proprietary ATI driver on Ubuntu Feisty ===&lt;br /&gt;
I had a hard time getting resume from suspend-to-ram to work with Feisty on a Thinkpad T60p. When resuming, the screen just went black with X hanging. &lt;br /&gt;
&lt;br /&gt;
It works like this:&lt;br /&gt;
# In &amp;lt;code&amp;gt;/etc/acpi/sleep.sh&amp;lt;/code&amp;gt;, as second line, add the line &amp;lt;code&amp;gt;sudo chvt 1&amp;lt;/code&amp;gt;. This forces a switch to the text console before standby.&lt;br /&gt;
# Likewise, in add to the very end of the file &amp;lt;code&amp;gt;/etc/acpi/resume.sh&amp;lt;/code&amp;gt; the line &amp;lt;code&amp;gt;sudo chvt 7&amp;lt;/code&amp;gt;.&lt;br /&gt;
# Finally, the perhaps most important change goes into &amp;lt;code&amp;gt;/etc/default/acpi-support&amp;lt;/code&amp;gt;. Change the line &amp;lt;code&amp;gt;POST_VIDEO=true&amp;lt;/code&amp;gt; to read &amp;lt;code&amp;gt;POST_VIDEO=&amp;lt;/code&amp;gt;. This was the point when it started working on my system.&lt;br /&gt;
&lt;br /&gt;
:'''NOTE:'''  Step 3. above was the only change '''I''' needed to make.  YMMV. --[[User:Brianfinley|Brian E. Finley]] 23:26, 30 March 2007 (CEST)&lt;br /&gt;
&lt;br /&gt;
:'''NOTE:'''  With only Step 3. above, my T60p would not even suspend when docked.  Step 3. above was sufficient to make resume work for me when my laptop was not in its docking station.  However, when I subsequently docked the machine I saw something I thought was banished with my upgrade to a 2.6.20 kernel: it uncermoniously powered off the instant it was docked, almost as though someone threw a hardware switch.   --[[User:dave_abrahams|Dave Abrahams]] Thu Apr 12 11:18:22 EDT 2007&lt;br /&gt;
&lt;br /&gt;
:'''UPDATE:''' The problems mentioned above related to docking seem to happen when I use the gnome-power-manager applet menu but not when I use the suspend function in the window that results from clicking the &amp;quot;Log Out Button&amp;quot; applet.  When I use the latter, suspend-to-RAM seems to work right, consistently.  I'm [https://bugs.launchpad.net/ubuntu/+bug/107674 following up with Ubuntu] on this point.   --[[User:dave_abrahams|Dave Abrahams]] Thu Apr 19 11:53 GMT 2007&lt;br /&gt;
&lt;br /&gt;
:'''NOTE:''' Editing &amp;lt;code&amp;gt;/etc/acpi/sleep.sh&amp;lt;/code&amp;gt; and &amp;lt;code&amp;gt;/etc/acpi/resume.sh&amp;lt;/code&amp;gt; is not necessary, as there is an option in &amp;lt;code&amp;gt;/etc/default/acpi-support&amp;lt;/code&amp;gt; for ''that very feature'': Just set the commented out DOUBLE_CONSOLE_SWITCH=true, and it will do the console switch suggested automatically. --[[User:Webograph|Webograph]] 12:43, 29 August 2007 (UTC)&lt;br /&gt;
&lt;br /&gt;
:'''NOTE:'''  I also had success doing nothing more than setting &amp;lt;code&amp;gt;POST_VIDEO=false&amp;lt;/code&amp;gt; (step 3).  Using a T60p (2007-83U) with fglrx 8.3 (8.47.3-3~bpo40 from Debian's etch-backports) --[[User:sihde|Steve Ihde]] 31 May 2008&lt;br /&gt;
&lt;br /&gt;
For completeness, here are the complete settings of this file:&lt;br /&gt;
&amp;lt;ol&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;ACPI_SLEEP=true&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;ACPI_HIBERNATE=true&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;ACPI_SLEEP_MODE=mem&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;MODULES=&amp;quot;&amp;quot;&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;MODULE_WHITELIST=&amp;quot;&amp;quot;&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;SAVE_VBE_STATE=true&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;VBE_STATE=/var/lib/acpi_support/vbestate&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;POST_VIDEO=&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;USE_DPMS=true&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;# RADEON_LIGHT=true&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;# DOUBLE_CONSOLE_SWITCH=true&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;HIBERNATE_MODE=shutdown&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;LOCK_SCREEN=true&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;# DISABLE_DMA=true&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;# RESET_DRIVE=true&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;STOP_SERVICES=&amp;quot;mysql &amp;quot;&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;RESTART_IRDA=false&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;ENABLE_LAPTOP_MODE=true&amp;lt;/li&amp;gt;&lt;br /&gt;
&amp;lt;/ol&amp;gt;&lt;br /&gt;
&lt;br /&gt;
Point 11 may eventually replace the chvt settings above, but I didn't try.&lt;br /&gt;
&lt;br /&gt;
=== Ubuntu Hardy ===&lt;br /&gt;
Suspend/hibernate seems to work in Hardy with latest Catalyst driver out of box (no need to edit any conf files!). The only confusing thing is that it goes to sleep and hibernate pretty fast, but it takes a long time to resume from both, especially from hibernate. The first impression that it hangs, but it's not.&lt;br /&gt;
&lt;br /&gt;
=== Linux kernel Framebuffer driver ===&lt;br /&gt;
&lt;br /&gt;
Supported by the [http://dev.gentoo.org/~spock/projects/vesafb-tng/ vesafb-tng] project, but requires the boot parameter {{bootparm|video|vesafb:nocrtc}}.&lt;br /&gt;
&lt;br /&gt;
Alternatively, you may use classic vesafb with {{bootparm|vga|838}} for a 1400x1050 screen with 32 bit color depth.&lt;br /&gt;
&lt;br /&gt;
{{NOTE|The [[fglrx]] driver may cause distortion/corruption of the framebuffer after use at high resolutions, making the text difficult to read.}}&lt;br /&gt;
&lt;br /&gt;
=== Ubuntu 8.10 (Intrepid) using RadeonHD driver ===&lt;br /&gt;
&lt;br /&gt;
The RadeonHD driver is coming along and is very usable for 2D graphics, and now with the v1.2.3 release of the driver (not yet in the Intrepid repos), 3D support is there (XVideo extenstion works)!  I have experienced no suspend/resume issues at all with this driver, so no caveats there.&lt;br /&gt;
&lt;br /&gt;
The package you need to install is xserver-xorg-video-radeonhd.  You probably have this installed already if you've installed Intrepid.  However, as of this writing, the version included is 1.2.1 of the driver which doesn't have the 3D support in it.  I have a '''[http://www.acmeps.com/kernel/debs/xserver-xorg-video-radeonhd_1.2.3-1~pre1_i386.deb build of v1.2.3 available]''' (actually git commit 29cd7382, built by dropping the latest bits into the xserver-xorg-video-radeon source package and building) to use until Intrepid has updated packages.&lt;br /&gt;
&lt;br /&gt;
Just a few configuration nits.&lt;br /&gt;
&lt;br /&gt;
/etc/X11/xorg.conf snippet:&lt;br /&gt;
&lt;br /&gt;
 Section &amp;quot;Device&amp;quot;&lt;br /&gt;
        Identifier  &amp;quot;LCD ATI Radeon Graphics Adapter&amp;quot;&lt;br /&gt;
        Driver      &amp;quot;radeonhd&amp;quot;&lt;br /&gt;
        Option      &amp;quot;AccelMethod&amp;quot;         &amp;quot;exa&amp;quot;&lt;br /&gt;
        Option      &amp;quot;DRI&amp;quot;                 &amp;quot;true&amp;quot;&lt;br /&gt;
        BusID       &amp;quot;PCI:1:0:0&amp;quot;&lt;br /&gt;
 EndSection&lt;br /&gt;
&lt;br /&gt;
You have to at least turn on DRI so the right plumbing gets hooked up for the 3D graphics.  DRI in the driver is still off by default.  And you can choose to not include the AccelMethod option, I've just found EXA to be faster than XAA (the default).&lt;br /&gt;
&lt;br /&gt;
In previous versions of the driver you had to turn off both Composite and AIGLX.  It appears you no longer need to do that, so if you have those in your Extensions or ServerFlags sections of xorg.conf, you can remove them.&lt;br /&gt;
&lt;br /&gt;
There is a caveat, however, to get the kernel (2.6.27) to behave properly with a graphics console (i.e., what you get if you hit CTRL-ALT-F1).  Similar to the above comment about using a framebuffer, you have to edit /boot/grub/menu.lst and change the defoptions line to something like:&lt;br /&gt;
&lt;br /&gt;
 # defoptions=video=vesafb vga=835&lt;br /&gt;
&lt;br /&gt;
Then you type &amp;quot;update-grub&amp;quot; as root to update the kernel boot params with this new info.  Mode 835 is 1400x1050 at 256 color depth.  There are lots of other options, use &amp;quot;vga=ask&amp;quot; instead and reboot to find out what they are for your specific screen.&lt;br /&gt;
&lt;br /&gt;
Note that you can't have &amp;quot;splash&amp;quot; on this line as the splash boot screen will mess up the console so that it becomes unusable, that is, a blank screen if you switch to any console.  (Other suggestions for ways to deal with the boot console welcome.)&lt;br /&gt;
&lt;br /&gt;
=== ThinkPads this chip may be found in ===&lt;br /&gt;
* {{T60p}}&lt;br /&gt;
* {{Z61p}}&lt;br /&gt;
&lt;br /&gt;
[[Category:Components]]&lt;/div&gt;</summary>
		<author><name>DF5JT</name></author>
		
	</entry>
	<entry>
		<id>https://www.thinkwiki.org/w/index.php?title=ATI_Mobility_FireGL_V5200&amp;diff=42638</id>
		<title>ATI Mobility FireGL V5200</title>
		<link rel="alternate" type="text/html" href="https://www.thinkwiki.org/w/index.php?title=ATI_Mobility_FireGL_V5200&amp;diff=42638"/>
		<updated>2009-04-12T09:14:14Z</updated>

		<summary type="html">&lt;p&gt;DF5JT: /* Proprietary ATI driver */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;__NOTOC__&lt;br /&gt;
{| width=&amp;quot;100%&amp;quot;&lt;br /&gt;
|style=&amp;quot;vertical-align:top&amp;quot; |&lt;br /&gt;
&amp;lt;div style=&amp;quot;margin: 0; margin-right:10px; border: 1px solid #dfdfdf; padding: 0em 1em 1em 1em; background-color:#F8F8FF; align:right;&amp;quot;&amp;gt;&lt;br /&gt;
&lt;br /&gt;
This is an ATI video adapter&lt;br /&gt;
&lt;br /&gt;
== Features ==&lt;br /&gt;
* Chipset: ATI M56 - Similar to Radeon X1600&lt;br /&gt;
* PCI ID: 1002:71c4&lt;br /&gt;
* PCI Express x16&lt;br /&gt;
* 256 MB GDDR3-SDRAM video memory&lt;br /&gt;
&amp;lt;/div&amp;gt;&lt;br /&gt;
|style=&amp;quot;vertical-align:top&amp;quot; |&lt;br /&gt;
|}&lt;br /&gt;
&lt;br /&gt;
=== Linux X.Org driver ===&lt;br /&gt;
Initial support included in the stable branch of the Xorg [[radeon]] driver (2D only). XVideo acceleration and 3D support have been added by the xorg driver developers and the driver is used by all majos distributions.&lt;br /&gt;
&lt;br /&gt;
It is also supported by the [[radeonhd]] driver. A Debian package already exists in the experimental branch. 3d is still under development.&lt;br /&gt;
There is also a Gentoo ebuild recently added to the tree (x11-drivers/xf86-video-radeonhd), which seems to be working without any problems.&lt;br /&gt;
&lt;br /&gt;
Recent releases of the ATI proprietary driver do not support this card anymore. Ever since ATI has been bought by AMD the specs for this card have been released to driver developers.&lt;br /&gt;
&lt;br /&gt;
==== ThinkPad LCD ====&lt;br /&gt;
&lt;br /&gt;
All current distributions offer full support for the 1600x1200 resolution of the screen.&lt;br /&gt;
&lt;br /&gt;
==== External VGA port ====&lt;br /&gt;
Yes&lt;br /&gt;
&lt;br /&gt;
==== SVideo port ====&lt;br /&gt;
No&lt;br /&gt;
&lt;br /&gt;
==== DVI port ====&lt;br /&gt;
No (yes on Docking Station, working with [[fglrx]] driver&lt;br /&gt;
&lt;br /&gt;
=== Proprietary ATI driver ===&lt;br /&gt;
&lt;br /&gt;
'''The following information is outdated only relevant if you use an older distribution.&lt;br /&gt;
'''&lt;br /&gt;
&lt;br /&gt;
Proprietary [[fglrx]] driver works since version 8.25.18, version 8.24.8 works by adding &amp;quot;ChipID 0x71c5&amp;quot; to the xorg.conf &amp;quot;Device&amp;quot; statement and fooling it into thinking it's a X1600 mobility (do this only as a last resort or if you're sure you have an older driver; it is fatal to version 8.25.18).&lt;br /&gt;
&lt;br /&gt;
The new driver 8.25.18 seems to support the card, see&lt;br /&gt;
https://support.ati.com/ics/support/KBAnswer.asp?questionID=1643&lt;br /&gt;
(successfully tested with the provided autorun installation script on OpenSuSE 10.1, T60p with 15&amp;quot; display).&lt;br /&gt;
&lt;br /&gt;
(K)ubuntu 6.06 comes with this driver and it works out of the box. However, it seems to have problems with VT Switching (Alt+Fn). Instead of switching to a console, X resets and goes back to the log in screen (This doesn't happen when using the [[vesa]] driver). The switching issue might be resolvable by setting ENABLE_LAPTOP_MODE=true in /etc/default/acpi-support, but it doesn't seem to help on all systems. &lt;br /&gt;
&lt;br /&gt;
{{NOTE|On Ubuntu 6.10 (Edgy) with 'splash' on the command line, you cannot switch to VTs. This is a Thinkpad T60. The screen remains graphical, but is very badly garbled.  There is no text. X does not reset, and you can switch back to your X session. The solution is to include 'vga-791' or to remove 'splash' from the grub or lilo command line. You don't need to do both.  (It's been a few days, so I don't remember where I found this...I just chanced upon this wiki page for the video card I'm trying to get working with xgl/AIGLX and Beryl).}}&lt;br /&gt;
   &lt;br /&gt;
Also, the pointer gets garbled sometimes in X, but this is only a minor annoyance. Dual Head (Xinerama) also works with this driver, but don't use KDE control center to enable it (use the ATi tool or edit xorg.conf manually)  &lt;br /&gt;
&lt;br /&gt;
I have a feeling that this card gets in the way when trying to suspend to RAM (resuming with a complete blank screen, lights are blinking, but that's it), but this can also be caused by other means.&lt;br /&gt;
&lt;br /&gt;
=== Proprietary ATI driver on Ubuntu Feisty ===&lt;br /&gt;
I had a hard time getting resume from suspend-to-ram to work with Feisty on a Thinkpad T60p. When resuming, the screen just went black with X hanging. &lt;br /&gt;
&lt;br /&gt;
It works like this:&lt;br /&gt;
# In &amp;lt;code&amp;gt;/etc/acpi/sleep.sh&amp;lt;/code&amp;gt;, as second line, add the line &amp;lt;code&amp;gt;sudo chvt 1&amp;lt;/code&amp;gt;. This forces a switch to the text console before standby.&lt;br /&gt;
# Likewise, in add to the very end of the file &amp;lt;code&amp;gt;/etc/acpi/resume.sh&amp;lt;/code&amp;gt; the line &amp;lt;code&amp;gt;sudo chvt 7&amp;lt;/code&amp;gt;.&lt;br /&gt;
# Finally, the perhaps most important change goes into &amp;lt;code&amp;gt;/etc/default/acpi-support&amp;lt;/code&amp;gt;. Change the line &amp;lt;code&amp;gt;POST_VIDEO=true&amp;lt;/code&amp;gt; to read &amp;lt;code&amp;gt;POST_VIDEO=&amp;lt;/code&amp;gt;. This was the point when it started working on my system.&lt;br /&gt;
&lt;br /&gt;
:'''NOTE:'''  Step 3. above was the only change '''I''' needed to make.  YMMV. --[[User:Brianfinley|Brian E. Finley]] 23:26, 30 March 2007 (CEST)&lt;br /&gt;
&lt;br /&gt;
:'''NOTE:'''  With only Step 3. above, my T60p would not even suspend when docked.  Step 3. above was sufficient to make resume work for me when my laptop was not in its docking station.  However, when I subsequently docked the machine I saw something I thought was banished with my upgrade to a 2.6.20 kernel: it uncermoniously powered off the instant it was docked, almost as though someone threw a hardware switch.   --[[User:dave_abrahams|Dave Abrahams]] Thu Apr 12 11:18:22 EDT 2007&lt;br /&gt;
&lt;br /&gt;
:'''UPDATE:''' The problems mentioned above related to docking seem to happen when I use the gnome-power-manager applet menu but not when I use the suspend function in the window that results from clicking the &amp;quot;Log Out Button&amp;quot; applet.  When I use the latter, suspend-to-RAM seems to work right, consistently.  I'm [https://bugs.launchpad.net/ubuntu/+bug/107674 following up with Ubuntu] on this point.   --[[User:dave_abrahams|Dave Abrahams]] Thu Apr 19 11:53 GMT 2007&lt;br /&gt;
&lt;br /&gt;
:'''NOTE:''' Editing &amp;lt;code&amp;gt;/etc/acpi/sleep.sh&amp;lt;/code&amp;gt; and &amp;lt;code&amp;gt;/etc/acpi/resume.sh&amp;lt;/code&amp;gt; is not necessary, as there is an option in &amp;lt;code&amp;gt;/etc/default/acpi-support&amp;lt;/code&amp;gt; for ''that very feature'': Just set the commented out DOUBLE_CONSOLE_SWITCH=true, and it will do the console switch suggested automatically. --[[User:Webograph|Webograph]] 12:43, 29 August 2007 (UTC)&lt;br /&gt;
&lt;br /&gt;
:'''NOTE:'''  I also had success doing nothing more than setting &amp;lt;code&amp;gt;POST_VIDEO=false&amp;lt;/code&amp;gt; (step 3).  Using a T60p (2007-83U) with fglrx 8.3 (8.47.3-3~bpo40 from Debian's etch-backports) --[[User:sihde|Steve Ihde]] 31 May 2008&lt;br /&gt;
&lt;br /&gt;
For completeness, here are the complete settings of this file:&lt;br /&gt;
&amp;lt;ol&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;ACPI_SLEEP=true&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;ACPI_HIBERNATE=true&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;ACPI_SLEEP_MODE=mem&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;MODULES=&amp;quot;&amp;quot;&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;MODULE_WHITELIST=&amp;quot;&amp;quot;&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;SAVE_VBE_STATE=true&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;VBE_STATE=/var/lib/acpi_support/vbestate&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;POST_VIDEO=&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;USE_DPMS=true&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;# RADEON_LIGHT=true&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;# DOUBLE_CONSOLE_SWITCH=true&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;HIBERNATE_MODE=shutdown&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;LOCK_SCREEN=true&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;# DISABLE_DMA=true&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;# RESET_DRIVE=true&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;STOP_SERVICES=&amp;quot;mysql &amp;quot;&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;RESTART_IRDA=false&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;ENABLE_LAPTOP_MODE=true&amp;lt;/li&amp;gt;&lt;br /&gt;
&amp;lt;/ol&amp;gt;&lt;br /&gt;
&lt;br /&gt;
Point 11 may eventually replace the chvt settings above, but I didn't try.&lt;br /&gt;
&lt;br /&gt;
=== Ubuntu Hardy ===&lt;br /&gt;
Suspend/hibernate seems to work in Hardy with latest Catalyst driver out of box (no need to edit any conf files!). The only confusing thing is that it goes to sleep and hibernate pretty fast, but it takes a long time to resume from both, especially from hibernate. The first impression that it hangs, but it's not.&lt;br /&gt;
&lt;br /&gt;
=== Linux kernel Framebuffer driver ===&lt;br /&gt;
&lt;br /&gt;
Supported by the [http://dev.gentoo.org/~spock/projects/vesafb-tng/ vesafb-tng] project, but requires the boot parameter {{bootparm|video|vesafb:nocrtc}}.&lt;br /&gt;
&lt;br /&gt;
Alternatively, you may use classic vesafb with {{bootparm|vga|838}} for a 1400x1050 screen with 32 bit color depth.&lt;br /&gt;
&lt;br /&gt;
{{NOTE|The [[fglrx]] driver may cause distortion/corruption of the framebuffer after use at high resolutions, making the text difficult to read.}}&lt;br /&gt;
&lt;br /&gt;
=== Ubuntu 8.10 (Intrepid) using RadeonHD driver ===&lt;br /&gt;
&lt;br /&gt;
The RadeonHD driver is coming along and is very usable for 2D graphics, and now with the v1.2.3 release of the driver (not yet in the Intrepid repos), 3D support is there (XVideo extenstion works)!  I have experienced no suspend/resume issues at all with this driver, so no caveats there.&lt;br /&gt;
&lt;br /&gt;
The package you need to install is xserver-xorg-video-radeonhd.  You probably have this installed already if you've installed Intrepid.  However, as of this writing, the version included is 1.2.1 of the driver which doesn't have the 3D support in it.  I have a '''[http://www.acmeps.com/kernel/debs/xserver-xorg-video-radeonhd_1.2.3-1~pre1_i386.deb build of v1.2.3 available]''' (actually git commit 29cd7382, built by dropping the latest bits into the xserver-xorg-video-radeon source package and building) to use until Intrepid has updated packages.&lt;br /&gt;
&lt;br /&gt;
Just a few configuration nits.&lt;br /&gt;
&lt;br /&gt;
/etc/X11/xorg.conf snippet:&lt;br /&gt;
&lt;br /&gt;
 Section &amp;quot;Device&amp;quot;&lt;br /&gt;
        Identifier  &amp;quot;LCD ATI Radeon Graphics Adapter&amp;quot;&lt;br /&gt;
        Driver      &amp;quot;radeonhd&amp;quot;&lt;br /&gt;
        Option      &amp;quot;AccelMethod&amp;quot;         &amp;quot;exa&amp;quot;&lt;br /&gt;
        Option      &amp;quot;DRI&amp;quot;                 &amp;quot;true&amp;quot;&lt;br /&gt;
        BusID       &amp;quot;PCI:1:0:0&amp;quot;&lt;br /&gt;
 EndSection&lt;br /&gt;
&lt;br /&gt;
You have to at least turn on DRI so the right plumbing gets hooked up for the 3D graphics.  DRI in the driver is still off by default.  And you can choose to not include the AccelMethod option, I've just found EXA to be faster than XAA (the default).&lt;br /&gt;
&lt;br /&gt;
In previous versions of the driver you had to turn off both Composite and AIGLX.  It appears you no longer need to do that, so if you have those in your Extensions or ServerFlags sections of xorg.conf, you can remove them.&lt;br /&gt;
&lt;br /&gt;
There is a caveat, however, to get the kernel (2.6.27) to behave properly with a graphics console (i.e., what you get if you hit CTRL-ALT-F1).  Similar to the above comment about using a framebuffer, you have to edit /boot/grub/menu.lst and change the defoptions line to something like:&lt;br /&gt;
&lt;br /&gt;
 # defoptions=video=vesafb vga=835&lt;br /&gt;
&lt;br /&gt;
Then you type &amp;quot;update-grub&amp;quot; as root to update the kernel boot params with this new info.  Mode 835 is 1400x1050 at 256 color depth.  There are lots of other options, use &amp;quot;vga=ask&amp;quot; instead and reboot to find out what they are for your specific screen.&lt;br /&gt;
&lt;br /&gt;
Note that you can't have &amp;quot;splash&amp;quot; on this line as the splash boot screen will mess up the console so that it becomes unusable, that is, a blank screen if you switch to any console.  (Other suggestions for ways to deal with the boot console welcome.)&lt;br /&gt;
&lt;br /&gt;
=== ThinkPads this chip may be found in ===&lt;br /&gt;
* {{T60p}}&lt;br /&gt;
* {{Z61p}}&lt;br /&gt;
&lt;br /&gt;
[[Category:Components]]&lt;/div&gt;</summary>
		<author><name>DF5JT</name></author>
		
	</entry>
	<entry>
		<id>https://www.thinkwiki.org/w/index.php?title=ATI_Mobility_FireGL_V5200&amp;diff=42637</id>
		<title>ATI Mobility FireGL V5200</title>
		<link rel="alternate" type="text/html" href="https://www.thinkwiki.org/w/index.php?title=ATI_Mobility_FireGL_V5200&amp;diff=42637"/>
		<updated>2009-04-12T09:12:37Z</updated>

		<summary type="html">&lt;p&gt;DF5JT: /* Linux X.Org driver */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;__NOTOC__&lt;br /&gt;
{| width=&amp;quot;100%&amp;quot;&lt;br /&gt;
|style=&amp;quot;vertical-align:top&amp;quot; |&lt;br /&gt;
&amp;lt;div style=&amp;quot;margin: 0; margin-right:10px; border: 1px solid #dfdfdf; padding: 0em 1em 1em 1em; background-color:#F8F8FF; align:right;&amp;quot;&amp;gt;&lt;br /&gt;
&lt;br /&gt;
This is an ATI video adapter&lt;br /&gt;
&lt;br /&gt;
== Features ==&lt;br /&gt;
* Chipset: ATI M56 - Similar to Radeon X1600&lt;br /&gt;
* PCI ID: 1002:71c4&lt;br /&gt;
* PCI Express x16&lt;br /&gt;
* 256 MB GDDR3-SDRAM video memory&lt;br /&gt;
&amp;lt;/div&amp;gt;&lt;br /&gt;
|style=&amp;quot;vertical-align:top&amp;quot; |&lt;br /&gt;
|}&lt;br /&gt;
&lt;br /&gt;
=== Linux X.Org driver ===&lt;br /&gt;
Initial support included in the stable branch of the Xorg [[radeon]] driver (2D only). XVideo acceleration and 3D support have been added by the xorg driver developers and the driver is used by all majos distributions.&lt;br /&gt;
&lt;br /&gt;
It is also supported by the [[radeonhd]] driver. A Debian package already exists in the experimental branch. 3d is still under development.&lt;br /&gt;
There is also a Gentoo ebuild recently added to the tree (x11-drivers/xf86-video-radeonhd), which seems to be working without any problems.&lt;br /&gt;
&lt;br /&gt;
Recent releases of the ATI proprietary driver do not support this card anymore. Ever since ATI has been bought by AMD the specs for this card have been released to driver developers.&lt;br /&gt;
&lt;br /&gt;
==== ThinkPad LCD ====&lt;br /&gt;
&lt;br /&gt;
All current distributions offer full support for the 1600x1200 resolution of the screen.&lt;br /&gt;
&lt;br /&gt;
==== External VGA port ====&lt;br /&gt;
Yes&lt;br /&gt;
&lt;br /&gt;
==== SVideo port ====&lt;br /&gt;
No&lt;br /&gt;
&lt;br /&gt;
==== DVI port ====&lt;br /&gt;
No (yes on Docking Station, working with [[fglrx]] driver&lt;br /&gt;
&lt;br /&gt;
=== Proprietary ATI driver ===&lt;br /&gt;
Proprietary [[fglrx]] driver works since version 8.25.18, version 8.24.8 works by adding &amp;quot;ChipID 0x71c5&amp;quot; to the xorg.conf &amp;quot;Device&amp;quot; statement and fooling it into thinking it's a X1600 mobility (do this only as a last resort or if you're sure you have an older driver; it is fatal to version 8.25.18).&lt;br /&gt;
&lt;br /&gt;
The new driver 8.25.18 seems to support the card, see&lt;br /&gt;
https://support.ati.com/ics/support/KBAnswer.asp?questionID=1643&lt;br /&gt;
(successfully tested with the provided autorun installation script on OpenSuSE 10.1, T60p with 15&amp;quot; display).&lt;br /&gt;
&lt;br /&gt;
(K)ubuntu 6.06 comes with this driver and it works out of the box. However, it seems to have problems with VT Switching (Alt+Fn). Instead of switching to a console, X resets and goes back to the log in screen (This doesn't happen when using the [[vesa]] driver). The switching issue might be resolvable by setting ENABLE_LAPTOP_MODE=true in /etc/default/acpi-support, but it doesn't seem to help on all systems. &lt;br /&gt;
&lt;br /&gt;
{{NOTE|On Ubuntu 6.10 (Edgy) with 'splash' on the command line, you cannot switch to VTs. This is a Thinkpad T60. The screen remains graphical, but is very badly garbled.  There is no text. X does not reset, and you can switch back to your X session. The solution is to include 'vga-791' or to remove 'splash' from the grub or lilo command line. You don't need to do both.  (It's been a few days, so I don't remember where I found this...I just chanced upon this wiki page for the video card I'm trying to get working with xgl/AIGLX and Beryl).}}&lt;br /&gt;
   &lt;br /&gt;
Also, the pointer gets garbled sometimes in X, but this is only a minor annoyance. Dual Head (Xinerama) also works with this driver, but don't use KDE control center to enable it (use the ATi tool or edit xorg.conf manually)  &lt;br /&gt;
&lt;br /&gt;
I have a feeling that this card gets in the way when trying to suspend to RAM (resuming with a complete blank screen, lights are blinking, but that's it), but this can also be caused by other means.&lt;br /&gt;
&lt;br /&gt;
=== Proprietary ATI driver on Ubuntu Feisty ===&lt;br /&gt;
I had a hard time getting resume from suspend-to-ram to work with Feisty on a Thinkpad T60p. When resuming, the screen just went black with X hanging. &lt;br /&gt;
&lt;br /&gt;
It works like this:&lt;br /&gt;
# In &amp;lt;code&amp;gt;/etc/acpi/sleep.sh&amp;lt;/code&amp;gt;, as second line, add the line &amp;lt;code&amp;gt;sudo chvt 1&amp;lt;/code&amp;gt;. This forces a switch to the text console before standby.&lt;br /&gt;
# Likewise, in add to the very end of the file &amp;lt;code&amp;gt;/etc/acpi/resume.sh&amp;lt;/code&amp;gt; the line &amp;lt;code&amp;gt;sudo chvt 7&amp;lt;/code&amp;gt;.&lt;br /&gt;
# Finally, the perhaps most important change goes into &amp;lt;code&amp;gt;/etc/default/acpi-support&amp;lt;/code&amp;gt;. Change the line &amp;lt;code&amp;gt;POST_VIDEO=true&amp;lt;/code&amp;gt; to read &amp;lt;code&amp;gt;POST_VIDEO=&amp;lt;/code&amp;gt;. This was the point when it started working on my system.&lt;br /&gt;
&lt;br /&gt;
:'''NOTE:'''  Step 3. above was the only change '''I''' needed to make.  YMMV. --[[User:Brianfinley|Brian E. Finley]] 23:26, 30 March 2007 (CEST)&lt;br /&gt;
&lt;br /&gt;
:'''NOTE:'''  With only Step 3. above, my T60p would not even suspend when docked.  Step 3. above was sufficient to make resume work for me when my laptop was not in its docking station.  However, when I subsequently docked the machine I saw something I thought was banished with my upgrade to a 2.6.20 kernel: it uncermoniously powered off the instant it was docked, almost as though someone threw a hardware switch.   --[[User:dave_abrahams|Dave Abrahams]] Thu Apr 12 11:18:22 EDT 2007&lt;br /&gt;
&lt;br /&gt;
:'''UPDATE:''' The problems mentioned above related to docking seem to happen when I use the gnome-power-manager applet menu but not when I use the suspend function in the window that results from clicking the &amp;quot;Log Out Button&amp;quot; applet.  When I use the latter, suspend-to-RAM seems to work right, consistently.  I'm [https://bugs.launchpad.net/ubuntu/+bug/107674 following up with Ubuntu] on this point.   --[[User:dave_abrahams|Dave Abrahams]] Thu Apr 19 11:53 GMT 2007&lt;br /&gt;
&lt;br /&gt;
:'''NOTE:''' Editing &amp;lt;code&amp;gt;/etc/acpi/sleep.sh&amp;lt;/code&amp;gt; and &amp;lt;code&amp;gt;/etc/acpi/resume.sh&amp;lt;/code&amp;gt; is not necessary, as there is an option in &amp;lt;code&amp;gt;/etc/default/acpi-support&amp;lt;/code&amp;gt; for ''that very feature'': Just set the commented out DOUBLE_CONSOLE_SWITCH=true, and it will do the console switch suggested automatically. --[[User:Webograph|Webograph]] 12:43, 29 August 2007 (UTC)&lt;br /&gt;
&lt;br /&gt;
:'''NOTE:'''  I also had success doing nothing more than setting &amp;lt;code&amp;gt;POST_VIDEO=false&amp;lt;/code&amp;gt; (step 3).  Using a T60p (2007-83U) with fglrx 8.3 (8.47.3-3~bpo40 from Debian's etch-backports) --[[User:sihde|Steve Ihde]] 31 May 2008&lt;br /&gt;
&lt;br /&gt;
For completeness, here are the complete settings of this file:&lt;br /&gt;
&amp;lt;ol&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;ACPI_SLEEP=true&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;ACPI_HIBERNATE=true&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;ACPI_SLEEP_MODE=mem&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;MODULES=&amp;quot;&amp;quot;&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;MODULE_WHITELIST=&amp;quot;&amp;quot;&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;SAVE_VBE_STATE=true&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;VBE_STATE=/var/lib/acpi_support/vbestate&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;POST_VIDEO=&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;USE_DPMS=true&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;# RADEON_LIGHT=true&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;# DOUBLE_CONSOLE_SWITCH=true&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;HIBERNATE_MODE=shutdown&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;LOCK_SCREEN=true&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;# DISABLE_DMA=true&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;# RESET_DRIVE=true&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;STOP_SERVICES=&amp;quot;mysql &amp;quot;&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;RESTART_IRDA=false&amp;lt;/li&amp;gt;&lt;br /&gt;
  &amp;lt;li&amp;gt;ENABLE_LAPTOP_MODE=true&amp;lt;/li&amp;gt;&lt;br /&gt;
&amp;lt;/ol&amp;gt;&lt;br /&gt;
&lt;br /&gt;
Point 11 may eventually replace the chvt settings above, but I didn't try.&lt;br /&gt;
&lt;br /&gt;
=== Ubuntu Hardy ===&lt;br /&gt;
Suspend/hibernate seems to work in Hardy with latest Catalyst driver out of box (no need to edit any conf files!). The only confusing thing is that it goes to sleep and hibernate pretty fast, but it takes a long time to resume from both, especially from hibernate. The first impression that it hangs, but it's not.&lt;br /&gt;
&lt;br /&gt;
=== Linux kernel Framebuffer driver ===&lt;br /&gt;
&lt;br /&gt;
Supported by the [http://dev.gentoo.org/~spock/projects/vesafb-tng/ vesafb-tng] project, but requires the boot parameter {{bootparm|video|vesafb:nocrtc}}.&lt;br /&gt;
&lt;br /&gt;
Alternatively, you may use classic vesafb with {{bootparm|vga|838}} for a 1400x1050 screen with 32 bit color depth.&lt;br /&gt;
&lt;br /&gt;
{{NOTE|The [[fglrx]] driver may cause distortion/corruption of the framebuffer after use at high resolutions, making the text difficult to read.}}&lt;br /&gt;
&lt;br /&gt;
=== Ubuntu 8.10 (Intrepid) using RadeonHD driver ===&lt;br /&gt;
&lt;br /&gt;
The RadeonHD driver is coming along and is very usable for 2D graphics, and now with the v1.2.3 release of the driver (not yet in the Intrepid repos), 3D support is there (XVideo extenstion works)!  I have experienced no suspend/resume issues at all with this driver, so no caveats there.&lt;br /&gt;
&lt;br /&gt;
The package you need to install is xserver-xorg-video-radeonhd.  You probably have this installed already if you've installed Intrepid.  However, as of this writing, the version included is 1.2.1 of the driver which doesn't have the 3D support in it.  I have a '''[http://www.acmeps.com/kernel/debs/xserver-xorg-video-radeonhd_1.2.3-1~pre1_i386.deb build of v1.2.3 available]''' (actually git commit 29cd7382, built by dropping the latest bits into the xserver-xorg-video-radeon source package and building) to use until Intrepid has updated packages.&lt;br /&gt;
&lt;br /&gt;
Just a few configuration nits.&lt;br /&gt;
&lt;br /&gt;
/etc/X11/xorg.conf snippet:&lt;br /&gt;
&lt;br /&gt;
 Section &amp;quot;Device&amp;quot;&lt;br /&gt;
        Identifier  &amp;quot;LCD ATI Radeon Graphics Adapter&amp;quot;&lt;br /&gt;
        Driver      &amp;quot;radeonhd&amp;quot;&lt;br /&gt;
        Option      &amp;quot;AccelMethod&amp;quot;         &amp;quot;exa&amp;quot;&lt;br /&gt;
        Option      &amp;quot;DRI&amp;quot;                 &amp;quot;true&amp;quot;&lt;br /&gt;
        BusID       &amp;quot;PCI:1:0:0&amp;quot;&lt;br /&gt;
 EndSection&lt;br /&gt;
&lt;br /&gt;
You have to at least turn on DRI so the right plumbing gets hooked up for the 3D graphics.  DRI in the driver is still off by default.  And you can choose to not include the AccelMethod option, I've just found EXA to be faster than XAA (the default).&lt;br /&gt;
&lt;br /&gt;
In previous versions of the driver you had to turn off both Composite and AIGLX.  It appears you no longer need to do that, so if you have those in your Extensions or ServerFlags sections of xorg.conf, you can remove them.&lt;br /&gt;
&lt;br /&gt;
There is a caveat, however, to get the kernel (2.6.27) to behave properly with a graphics console (i.e., what you get if you hit CTRL-ALT-F1).  Similar to the above comment about using a framebuffer, you have to edit /boot/grub/menu.lst and change the defoptions line to something like:&lt;br /&gt;
&lt;br /&gt;
 # defoptions=video=vesafb vga=835&lt;br /&gt;
&lt;br /&gt;
Then you type &amp;quot;update-grub&amp;quot; as root to update the kernel boot params with this new info.  Mode 835 is 1400x1050 at 256 color depth.  There are lots of other options, use &amp;quot;vga=ask&amp;quot; instead and reboot to find out what they are for your specific screen.&lt;br /&gt;
&lt;br /&gt;
Note that you can't have &amp;quot;splash&amp;quot; on this line as the splash boot screen will mess up the console so that it becomes unusable, that is, a blank screen if you switch to any console.  (Other suggestions for ways to deal with the boot console welcome.)&lt;br /&gt;
&lt;br /&gt;
=== ThinkPads this chip may be found in ===&lt;br /&gt;
* {{T60p}}&lt;br /&gt;
* {{Z61p}}&lt;br /&gt;
&lt;br /&gt;
[[Category:Components]]&lt;/div&gt;</summary>
		<author><name>DF5JT</name></author>
		
	</entry>
</feed>