Linux on the X220

ALLurGroceries - I tried your line in my grub file. When I ran grub-install i had no issues. And then when i rebooted it starts going through the reboot process (i think there are some error messages but goes a bit fast) then goes to a blank screen with a cursor. I figured there was an error in my grub so i stuck in a live install of Ubuntu, remounted and changed back to the old grub. but i'm still getting the same issue.

I'm running Ubuntu 11.10 with the Kernel 3.0. i haven't had much issues - just wanted to play around and see if i could improve battery - that will teach me!

I was curious if you (or anyone else) has some thoughts on this...

Thanks,

bg

 

Boot errors are from the way you are creating the usb stick. Read the documentation is all I can say about that..

You can get a daily build that will support ethernet (e1000e):
Index of /cdimage/daily-builds/daily/arch-latest/amd64/iso-cd

See: #626220 - Did not support Intel 82579LM Gigabit Ethernet (needs current e1000e) - Debian Bug report logs

or

I assume you have an intel wifi card. To install off of wireless copy the iwlwifi firmware package onto your usb stick:
Debian -- Package Search Results -- firmware-iwlwifi

A blank cursor usually indicates that your MBR has been wiped or that the partition marked active is not bootable.

I have no idea what your partition layout is or how you are booting, so without any details it's impossible to say what happened... but there are many methods of recovery available:

https://help.ubuntu.com/community/Grub2#Reinstalling_GRUB2

 

I'm not sure I follow. I have no problem booting from the USB stick and mounting the computer from within the live cd environment. I just can't do a normal boot.

When I boot up I occasionally get the message:

Invalid Environment Block
Press Any key to continue

then after a bit it continues the boot process going to a blank screen with a cursor...

 

That first reply was to someone else.

Invalid Environment Block means that the small amount of internal data that grub2 stores has gotten corrupted. You can try this:
Ubuntu Forums - View Single Post - error: invalid enviorment block, failed to boot default entries; HELP!!!!!!!
and this:
Invalid Environment Block solution - Ubuntu Forums

 

Thanks for the links - I worked through all the options and finally got it to work by the chroot method...

any thoughts why making that change caused such issues?

bg

 

Yeah the chroot method is the best in my experience (at least for a completely trashed bootloader).

Modifying the arguments to the kernel boot parameters wouldn't do that. To apply those you don't need to run grub-install, just update-grub. Maybe you installed grub to the wrong device or partition at first. Hard to know without knowing your setup in detail.

 

I've had mysterious shutdown/reboot problems on my x220 as well. There were a couple of kernel bugzilla bugs regarding these issues, but of course bugzilla is still down (I find it astonishing how poorly the kernel.org breakin continues to be handled, but that's a matter for a different discussion). Anyway, shutdown would do a reboot for me; I had to do:

Code:

rmmod ehci_hcd

at shutdown to work that around. Another problem was hung at shutdown/reboot, which sounds more like what you're seeing. This one was caused by e1000e for me if eth0 was left up at shutdown. The following would work that around:

Code:

ifconfig eth0 down

I'm writing in past tense because in the mean time I found a different solution that addresses both problems. When you run powertop, at least the old <1.99 version, it offers to enable "Runtime Power Management". What this actually does is:

Code:

for dpcontrol in /sys/bus/{pci,spi,i2c}/devices/*/power/control; do
    echo auto > "$dpcontrol"
done

I found that if various /sys/bus/pci/devices/*/power/control values are set to "auto" rather than the default "on", weird things happen at shutdown. So I modified my shutdown sequence to set them back to "on". All these weird problems disappeared. You can check how the values are set for you by doing:

Code:

cat /sys/bus/pci/devices/*/power/control

 

I've been using Gentoo on the x220 for a month or two now. It works fine for the most part. The biggest problems are:

-The card reader is useless. After it hosed my camera's SD card, I disabled support for it in the kernel and bought a small usb SD card reader

-The Sandy Bridge video driver is atrocious. 2D acceleration works well, but opengl apps routinely crash, have slowdowns, or numerous artifacts.

 

Agreed. It's a shame what a piece of junk Lenovo stuffed in there.

I'm not really using any opengl apps on a daily basis, but I am using a compositing window manager (kwin) without problems and googleearth also seems to work (until it crashes at exit).

I'm on Gentoo too, and I'm using xorg-server-1.10.4-r1, mesa-7.11, libdrm-2.4.26, and xf86-video-intel-2.16.0 (without SNA). Kernel 3.1.4.

 

I have found some tips that improve performance.

Run the xf86-video-intel video driver, the intel-dri driver, and the libva-driver-intel (or vaapi intel) driver (actual names will vary a little depending upon the distro).

You can run compiz (make sure desktop cube works as a test).

For the Gnome 3 shell with mutter/clutter wm, do: "export
CLUTTER_PAINT=disable-clipped-redraws:disable-culling" as per the following bug report: https://bugzilla.gnome.org/show_bug.cgi?id=657071#c2

I get poorer performance with Metacity, Kwin or the xfwm4 window managers, or running without a compositing window manager.

 

I've found that to be the case too. xfwm4 seems to run much better when compositing is enabled. Unfortunately, compositing tends to cause some vsync issues with apps I use, such as bsnes.

Coming from nvidia I really expected more from the open source intel drivers, as far as 3D performance goes. In linux, I seem to get about 50% or less 3D performance vs. what I'd get with windows. Wine is especially bad with Intel, to the point where pretty much all 3D apps are unplayable due to performance or crashing issues. Nvidia's drivers may be a pain but they are not that bad.

Fortunately the 2D and compositing window manager performance is really good. I just have to reboot to windows if I'm ever on the go and want to play a 3D game.

 

This problem is actually solved without using these hacks, but we have to wait until kernel 3.3. The patches are already there. See this on phoronix.

 

Hi all,

First time visitor/poster to this forum. I recently purchased an X220 and it is amazing. I'm running Debian testing and most everything is running great. Kernel is 3.1.0-1-amd64 (Debian disto).

One issue that's rather frustrating is that I can't get WebGL working. I get a big NAY from the testing page.

Does My Browser Support WebGL?

I get this from both Chrome and Firefox. My Chrome version is 16.0.912.63 beta. I've tried setting the appropriate flags from about:flags and when I go to about:gpu, I get this....

<code>
Graphics Feature Status
Canvas: Hardware accelerated
HTML Rendering: Hardware accelerated
3D CSS: Hardware accelerated
WebGL: Unavailable. Hardware acceleration disabled.
WebGL multisampling: Hardware accelerated

Problems Detected
WebGL has been disabled, either via about:flags or command line.
Version Information
Data exported Wed Dec 14 2011 17:24:00 GMT-0600 (CST)
Chrome version 16.0.912.63 (Official Build 113337) beta
Operating system Linux 3.1.0-1-amd64
Software rendering list version
ANGLE revision 896
Driver Information
Initialization time 12
Vendor Id 0x8086
Device Id 0x0126
Driver vendor Mesa
Driver version 7.11.2
Driver date
Pixel shader version 1.20
Vertex shader version 1.20
GL version 2.1
GL_VENDOR Tungsten Graphics, Inc
GL_RENDERER Mesa DRI Intel(R) Sandybridge Mobile
GL_VERSION 2.1 Mesa 7.11.2
</code>

What driver should I have loaded? lsmod shows that the i915 driver is installed. Is that what I should be using?

Anyone else get WebGL swinging? The laptop totally rocks and I'd love to get over this hurdle.

Thanks in advance for any help anyone can offer.

 

That WebGL test works for me on Sid.

 

Why are you running the amd64 kernel? Should be the x86_64 kernel I do believe. Could be your problem.

Webpage works fine here (Yay) in Fedora 15 with Firefox 8.0.

Code:

$ uname -a
Linux localhost.localdomain 2.6.41.4-1.fc15.x86_64 #1 SMP Tue Nov 29 11:53:48 UTC 2011 x86_64 x86_64 x86_64 GNU/Linux

 

Arrgh. I'm an idiot. When I was grabbing an image to use with netbootin, I obviously wasn't paying attention. Time for a re-install!

BTW, do I then want to grab the ia64 image from here? Is that the correct 64-bit kernel to use?

Index of /cdimage/weekly-builds

It's probably easiest to re-install the system, right? I can't think of an obvious ``Debian" way to switch kernels like that. I could build my kernel from source like I used to, but I must admit, I've found the Debian kernel maintenance to be quite good the last year or two.

I'll post back here if this does/doesn't fix WebGL.

OK, so some quick searching/educating informs me that the ia64 kernel is for the itanium architecture and the amd64 kernel should work for both Intel and AMD. So does anyone have any ideas how I get a kernel that reads as x86_64? And even more importantly, is there any difference between that and the amd64 kernel I'm using now?

ALLurGroceries, could I trouble you to post the output from about:gpu looks like in Chrome? I'm curious as to what the differences are between Sid (unstable) and Wheezy (testing). Thanks in advance.

 

I doubt your issue has to do with your kernel version (and even less with whether your kernel is called amd64 or x86_64).

Debian does not distinguish between AMD or Intel for 64-bit processors. Have a look here: 2.1.Supported Hardware.

But I'm afraid I have no knowledge of WebGL and can't give any further advice.

 

Thanks, mtt1. Was just reading something similar to your post. It's nice to have confirmation on that. Still need to figure out this WebGL issue. Grrr....

 

Just so you know, mattbellis, in Linux 'x86_64' and 'amd64' mean pretty much the same thing as far as hardware compatibility goes.

 

ia64 is indeed for Itanium wich is not an X86 architecture like the one you are using.

the correct version is thus AMD64.

It's called AMD64 because AMD was the first to put an x86 64bit version on the market.

 

I was referring to Iceweasel (firefox) 8.

Chromium is a bit of a different story. I can get it to work using --ignore-gpu-blacklist

See: Issue 76703 - chromium - Chrome Linux Crash in gfx:ixmapGLContext::MakeCurrent - An open-source browser project to help move the web forward. - Google Project Hosting

Code:

Graphics Feature Status
Canvas: Software rendered. Hardware acceleration not enabled.
3D CSS: Hardware accelerated
HTML Rendering: Hardware accelerated
WebGL: Hardware accelerated
WebGL multisampling: Hardware accelerated
Problems Detected
Accelerated 2D canvas has been disabled at the command line
Version Information
Data exported	Thu Dec 15 2011 13:58:47 GMT-0500 (EST)
Chrome version	15.0.874.121 (Developer Build 109964) Debian wheezy/sid
Software rendering list version	1.22
Driver Information
Initialization time	6
Vendor Id	0x8086
Device Id	0x0126
Driver vendor	Mesa Project and SGI
Driver version	7.11.2
Driver date	
Pixel shader version	1.20
Vertex shader version	1.20
GL version	2.1
GL_VENDOR	Tungsten Graphics, Inc
GL_RENDERER	Mesa DRI Intel(R) Sandybridge Mobile
GL_VERSION	2.1 Mesa 7.11.2
GL_EXTENSIONS	GL_ARB_multisample GL_EXT_abgr GL_EXT_bgra GL_EXT_blend_color GL_EXT_blend_logic_op GL_EXT_blend_minmax GL_EXT_blend_subtract GL_EXT_copy_texture GL_EXT_polygon_offset GL_EXT_subtexture GL_EXT_texture_object GL_EXT_vertex_array GL_EXT_compiled_vertex_array GL_EXT_texture GL_EXT_texture3D GL_IBM_rasterpos_clip GL_ARB_point_parameters GL_EXT_draw_range_elements GL_EXT_packed_pixels GL_EXT_point_parameters GL_EXT_rescale_normal GL_EXT_separate_specular_color GL_EXT_texture_edge_clamp GL_SGIS_generate_mipmap GL_SGIS_texture_border_clamp GL_SGIS_texture_edge_clamp GL_SGIS_texture_lod GL_ARB_framebuffer_sRGB GL_ARB_multitexture GL_EXT_framebuffer_sRGB GL_IBM_multimode_draw_arrays GL_IBM_texture_mirrored_repeat GL_3DFX_texture_compression_FXT1 GL_ARB_texture_cube_map GL_ARB_texture_env_add GL_ARB_transpose_matrix GL_EXT_blend_func_separate GL_EXT_fog_coord GL_EXT_multi_draw_arrays GL_EXT_secondary_color GL_EXT_texture_env_add GL_EXT_texture_filter_anisotropic GL_EXT_texture_lod_bias GL_INGR_blend_func_separate GL_NV_blend_square GL_NV_light_max_exponent GL_NV_texgen_reflection GL_NV_texture_env_combine4 GL_SUN_multi_draw_arrays GL_ARB_texture_border_clamp GL_ARB_texture_compression GL_EXT_framebuffer_object GL_EXT_texture_env_dot3 GL_MESA_window_pos GL_NV_packed_depth_stencil GL_NV_texture_rectangle GL_NV_vertex_program GL_ARB_depth_texture GL_ARB_occlusion_query GL_ARB_shadow GL_ARB_texture_env_combine GL_ARB_texture_env_crossbar GL_ARB_texture_env_dot3 GL_ARB_texture_mirrored_repeat GL_ARB_window_pos GL_ATI_envmap_bumpmap GL_EXT_stencil_two_side GL_EXT_texture_cube_map GL_NV_depth_clamp GL_NV_vertex_program1_1 GL_APPLE_client_storage GL_APPLE_packed_pixels GL_APPLE_vertex_array_object GL_ARB_draw_buffers GL_ARB_fragment_program GL_ARB_fragment_shader GL_ARB_shader_objects GL_ARB_vertex_program GL_ARB_vertex_shader GL_ATI_draw_buffers GL_ATI_texture_env_combine3 GL_ATI_texture_float GL_EXT_shadow_funcs GL_EXT_stencil_wrap GL_MESA_pack_invert GL_MESA_ycbcr_texture GL_ARB_depth_clamp GL_ARB_fragment_program_shadow GL_ARB_half_float_pixel GL_ARB_point_sprite GL_ARB_shading_language_100 GL_ARB_sync GL_ARB_texture_non_power_of_two GL_ARB_vertex_buffer_object GL_ATI_blend_equation_separate GL_EXT_blend_equation_separate GL_OES_read_format GL_ARB_color_buffer_float GL_ARB_pixel_buffer_object GL_ARB_texture_compression_rgtc GL_ARB_texture_float GL_ARB_texture_rectangle GL_EXT_pixel_buffer_object GL_EXT_texture_compression_rgtc GL_EXT_texture_rectangle GL_EXT_texture_sRGB GL_ARB_framebuffer_object GL_EXT_framebuffer_blit GL_EXT_framebuffer_multisample GL_EXT_packed_depth_stencil GL_APPLE_object_purgeable GL_ARB_vertex_array_object GL_ATI_separate_stencil GL_EXT_draw_buffers2 GL_EXT_gpu_program_parameters GL_EXT_texture_env_combine GL_EXT_texture_sRGB_decode GL_EXT_timer_query GL_OES_EGL_image GL_ARB_copy_buffer GL_ARB_half_float_vertex GL_ARB_map_buffer_range GL_ARB_texture_rg GL_ARB_texture_swizzle GL_ARB_vertex_array_bgra GL_EXT_separate_shader_objects GL_EXT_texture_swizzle GL_EXT_vertex_array_bgra GL_NV_conditional_render GL_ARB_ES2_compatibility GL_ARB_draw_elements_base_vertex GL_ARB_explicit_attrib_location GL_ARB_fragment_coord_conventions GL_ARB_provoking_vertex GL_ARB_sampler_objects GL_ARB_seamless_cube_map GL_ARB_shader_texture_lod GL_EXT_provoking_vertex GL_EXT_texture_snorm GL_MESA_texture_signed_rgba GL_ARB_robustness
Log Messages
[7791:7791:30163598498:INFO:gpu_child_thread.cc(108)] : gpu_info_collector::CollectGraphicsInfo complete

 

Good suggestion, ALLurGroceries. Chromium does seem to work with WebGL sites, though Chrome does not. Good enough for me. Thanks!

 

Hi everyone,

I just got a new X220 with mSATA Intel 310 80GB. I have a model with i7, IPS screen, 4GB RAM, 6 cell battery.. I'll do a quick round up of my experiences, maybe someone will find them helpful or interesting.

I'm running Debian testing/unstable (via the daily netinstall iso), Debian's 3.1.0-1 kernel, LUKS encrypted LVM.

SSD is great. Boot time is about five seconds, well let it be ten. Even huge programmes like (Libre/Open-)Office Writer or Firefox with 20 saved tabs open instantly.

I did a BIOS update first thing, using the grub4dos method documented at ThinkWiki. Worked without a hitch.

Power comsumption
I haven't done anything out of the ordinary. Enabled power savings in /etc/modules.d/i915-kms.conf, added pcie_aspm=force to the boot parameters, and used powertops suggestions.

I'm getting idle time power usage between 8.5 and 9.5W, more like between 9 and 9.5. While writing this, comsumption is between 10 and 12.8W, occasionally going up to almost 30 when opening a new tab or launching a programme. Overall, reasonable figures it seems to me.

Haven't done any tests on actual battery life so far.

Fan
Without configuration: always on, somewhere between 1950 and 2000 rpms, so barely audible. After changing sensors to /sys/class/hwmon/hwmon0/temp1_input, as suggested in this thread, thinkfan works, switching between level 0 and 1 every so often. Haven't taken a closer look at the configuration yet, though.

Things not working so far
Haven't yet put any effort into getting the touchpad right or enable middle mouse button scrolling. I'll do that when I find some time.

Another thing is the volume up/down keys. They don't work, while mute toggle is working fine (not sure yet whether I like that it's a toggle and not just mute full stop, though). Volume keys were working on a quick Debian install last night on HDD. But I installed the full Desktop suite with Gnome then. I did my own install of X and awesome window manager this time. So I'm sure it's something to do with that, but don't know yet. acpi_listen says the keys are sending "button/volumeup VOLUP" and "button/volumedown VOLDN" signals, so I guess I just have to map them. (Mic mute button is not sending a signal by the way.)

tpb isn't working.

I'm not using bluetooth or the fingerprint reader, so I have no idea on those, and haven't tried the cam or mic yet.

Only big thing: I noticed some throttling issue. When running on 65W adapter without battery CPU speed is limited to 800MHz, which seemed to be the normal behaviour before some BIOS update a while ago. I don't know what the opinion on it is now. But the real problem is that, when booted without battery it stays limited to 800MHz even when the battery is plugged in. No idea so far.

 

See what CPU frequency governor you are using:

Code:

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

Change it to performance:

Code:

echo performance > /sys/devices/system/cpu/cpu0/cpufreq/scaling_governor

See what frequency you are running at on processor 0:

Code:

head -n 10 /proc/cpuinfo

and/or

Code:

cpufreq-info | head -n 15

 

Usually I'm using ondemand, changed it to performance for this. But I don't think the problem is related to the governor.

800MHz.

cpufreq-info also says this:

Code:

current policy: frequency should be within 800 MHz and 800 MHz.
                The governor "performance" may decide which speed to use
                within this range.

This is always so when just on AC adapter. When booted with battery it changes to normal (800MHz to 2.7GHz) when I plug in the battery again. When booted without battery, however, it sticks with the "range" 800 to 800MHz no matter whether it's running with or without battery.

I've seen mentions of this problem on the internet but the first couple of leads I tried didn't bring up anything.

 

I can reproduce the 800MHz limit on AC without battery, but after I resume with the battery in I can set the governor like this:

Code:

echo performance | tee /sys/devices/system/cpu/cpu*/cpufreq/scaling_governor

And then:

Code:

current CPU frequency is 2.50 GHz (asserted by call to hardware).
  cpufreq stats: 2.50 GHz:87.62%, 2.50 GHz:0.01%, 2.20 GHz:0.02%, 2.00 GHz:0.01%, 1.80 GHz:0.05%, 1.60 GHz:0.02%, 1.40 GHz:0.00%, 1.20 GHz:0.01%, 1000 MHz:0.18%, 800 MHz:12.09%  (37)

Code:

uptime
 20:08:09 up 11 min,  2 users,  load average: 0.15, 0.25, 0.17

3.2.0-rc6+ and 8DET55WW (1.25 ) here

I seem to remember that the system pulls more power than the AC adapter can give it at maximum load, so the battery gets drained for the extra. Not sure where I read that though.

 

I read that too, probably somewhere in the ginormous x220 owners thread. This is apparently so by design. Also, I think that the problem does not occur if a 90W AC adapter is used rather than the stock 65W; I can test it at work tomorrow if anybody is interested.

 

Yes, that seems to be the case, or at least that was the case until BIOS 1.21 (I'm running 1.25). Have a look here, they mention the issue and for them things got better with BIOS 1.21:

Lenovo X220 i7-2620M CPU/GPU throttling: the full details and [email protected]

I still see the old behaviour, namely 800MHz limit on AC adapter only. While that's far from perfect for how I use my laptop for say half the time (on a desk with battery pulled), I could get used to it (maybe with tp_smapi battery charge thresholds?).

But again, the problem is that the frequency range stays limited to 800MHz, when I booted without battery.

So you have to do a suspend/resume cycle to make it change the frequency limits? s2ram or s2disk?

I'll give it a go with a 90W adapter later.

 

Sorry for double posting, didn't want to make the old post any longer.

I got the volume keys to work with xbindkeys. Apparent the X220 doesn't have the extra way of volume control like older Thinkpads ( Re: ThinkPad X220 volume level unsupported (Linux IBM ACPI)). So my .xbindkeys now reads:

Code:

# Increase volume
"amixer sset Master playback 1+"
    m:0x0 + c:123
    XF86AudioRaiseVolume

# Decrease volume
"amixer sset Master playback 1-"
    m:0x0 + c:122
    XF86AudioLowerVolume

I configured middle mouse button scrolling in /etc/X11/xorg.conf.d/ as described in Thinkwiki. And I'll have to take a more serious look into the clickpad (using ALLurGroceries's advice). Slim login manager is logging some errors about a synaptics device and I wonder if that is what is taking it so long to start.

Only other thing that's taking long is reading the initrd image at boot. Anyone else experiencing this? I need that image because I'm using encryption, but it seems to be taking longer than expected (5-10s).

 

Yeah suspend is what I used. I have a msata and 8GB so I don't even have swap to hibernate.

For the clickpad progress, kernel bugzilla is still down and there are no commits I can find that have anything remotely to do with clickpads. In terms of the xorg driver there is nothing new, I'm waiting to see what 1.6 brings.

[ANNOUNCE] xf86-input-synaptics 1.5.0

 

Only 4 GB, but no swap either.

So I tried this, to no avail. Booting without battery: 800MHz limit, suspend/resume with battery and AC adapter: the limit's still there. I haven't tried suspend/resume just on battery, might try that later.

Right now running on 90W adapter without battery: no limit. So it's definitely the old throttling problem. I just thought they had solved it with BIOS 1.21. I don't know why I'm still observing it then.

 

This is not the "old throttling problem". As I wrote earlier, this is so by design. x220 is not supposed to run at full frequency without battery with a 65W AC adapter. Either put the battery in or use a stronger adapter. No BIOS update will fix that. This is how the machine was designed.

The "old throttling problem" was about the system becoming permanently stuck at 800 MHz after thermal throttling due to both CPU and GPU being heavily used. The system should allow higher frequencies after it has cooled down a little, but in some cases, at least under Windows, it would not. This is what was fixed with one of the BIOS updates.

 

Oh, thanks for correcting me. I had taken the following quote to mean that while you can't have full performance on 65W AC adapter, you can get more than 800MHz after the BIOS update.

But reading it again, it doesn't mention unplugged battery. So thank you for clearing things up.

On an entirely different note, I updated the kernel earlier today from 3.1.0-1 [3.1.1-1] to 3.1.0-1 [3.1.5-1] (Debian testing). After that wireless wouldn't work anymore. No error messages in the logs, no error messages anywhere actually. Only thing I noticed was that getting a DHCP lease took a bit longer. But 100% packet loss, even when pinging the router.

Since I couldn't see any difference in the output I had no clue where to start, so I went back to the older kernel and everything is back up and running. I've seen some mentions of iwlagn in the Kernel Changelog for 3.1.5, but nothing that would have helped me.

Update:
By now I've found a bit more on it. Apparantly there is a regression in kernel 3.1.5 which breaks iwlagn under certain conditions. There already is a fix, but it's not included in the Debian kernels yet, as far as I could see.
iwlagn regression in v3.1.5
Bug 767173 - kernel 3.1.5-1 iwlagn with Intel 5300 WiFi fails to authenticate

 

I guys Im a newbie on Ubuntu . I recently get my x220 and I already install Ubuntu 11.10 on it. I having some issues.
1- Fan Noise (too loud)
2- Finger Reader (Not working)
3- Microphone key (not working)
4- Touch Pad (Not working properly)
5-Battery Cons

I already read all the previos post but I dont understand how did it. I need a newbie guide for each command for example I already install the thinkfan but I dont a a clue what to do with it.
Please I really need help and I dont wanna go back to Windows XD

I already edit the grub but and change thinkfan but im still getting 15w of use in powertop
Help!!

 

Just thought i would post my experiences pushing the x220. I got a core i7 220 a few months ago and got the extended 19-cell slice battery for it as well. I'm running ubuntu 11.10 - this is both good and bad. Of course i get the crappy battery life but with the slice battery i guess i can suck it up.

I am going to be travelling for a couple of years so i wont have a chance to take my powerful desktop with me so i wanted to trick out this laptop as much as possible.

Good things -
- upgraded to 16 gb ram with no problems using 2 8gb sticks.

- Crucial has very recently released a 7 mm 512 gb ssd drive - it's on newegg and a bunch of other sites - i can't actually find any mention of it on the crucial site itself. I got this drive a couple of days ago and plugged it in and so far everytthing's been working fine.

Bad things -
I wanted to drive an external 30" monitor at 2560 x 1600 with this laptop using display-port. Upgrading to the stock 3.1.2 kernel from ubuntu mainline ppa's worked because the 3.0 kernel wouldn't support this.

But a few days ago i upgraded to the 3.2 mainline kernel and it looks like that's dropped support for whatever it was that fixed the issue in kernel 3.1 - sigh... more digging to do. I love linux and have used it for ages, but sometimes when i have to do real work i wish things like this would just work.

If anyone has any bright ideas about what to do with kernel 3.2 to enable external monitor support over DisplayPort please let me know. The kernel detects the display and i can "miror" displays just fine. But as soon as i try to "unmirror" the displays, the external display just goes blank.

 

Well, to me at least it's acceptable. True, Thinkpads use less power when running under Windows. Still, under Linux a 9-cell easily lasts through a full day of light usage, with powertop showing power consumption of about 8 W. Are you sure you're using the power-saving kernel command line options described earlier in this thread or on phoronix?

I know exactly what you mean. I spent the better part of last weekend debugging connectivity problems to a new WiFi access point, only to realize in the end that I was hitting the bug described by mtt1 three messages back (thanks!).

I only "mirror" displays, and that has worked fine for me in all recent kernels, including 3.2. Well, there is a long-standing kernel bug with inconsistent DPMS properties ( https://bugzilla.kernel.org/show_bug.cgi?id=24982) which results in no output in spite of xrandr being correctly configured. Switching from X to a text console and back enables the display. Why don't you try if that helps? If not, I would file a bug at freedesktop's bugzilla. Anything interesting in /var/log/Xorg.0.log as you unmirror the displays?

 

Running 3.1.9 here and I am not having any problem using two displays.

This is the output from my xrandr

Code:

Screen 0: minimum 320 x 200, current 3286 x 1200, maximum 8192 x 8192
LVDS1 connected 1366x768+1920+0 (normal left inverted right x axis y axis) 277mm x 156mm
   1366x768       60.0*+
   1024x768       60.0 
   800x600        60.3     56.2 
   640x480        59.9 
VGA1 disconnected (normal left inverted right x axis y axis)
HDMI1 disconnected (normal left inverted right x axis y axis)
DP1 connected 1920x1200+0+0 (normal left inverted right x axis y axis) 518mm x 324mm
   1920x1200      60.0*+
   1920x1080      60.0 
   1600x1200      60.0 
   1280x1024      75.0     60.0 
   1152x864       75.0 
   1024x768       75.1     60.0 
   800x600        75.0     60.3 
   640x480        75.0     60.0 
   720x400        70.1 
HDMI2 disconnected (normal left inverted right x axis y axis)
HDMI3 disconnected (normal left inverted right x axis y axis)
DP2 disconnected (normal left inverted right x axis y axis)
DP3 disconnected (normal left inverted right x axis y axis)

I have two scripts to switch from monitor + laptop screen to just the monitor. These are just xrandr commands and you can run them from the terminal.

Code:

#!/bin/sh
# monitor plugged in
xrandr --output DP3 --off --output DP2 --off --output DP1 --mode 1920x1200 --pos 0x0 --rotate normal --output HDMI3 --off --output HDMI2 --off --output HDMI1 --off --output LVDS1 --mode 1366x768 --pos 1920x0 --rotate normal --output VGA1 --off

Code:

#!/bin/sh
# laptop only
xrandr --output DP3 --off --output DP2 --off --output DP1 --off --output HDMI3 --off --output HDMI2 --off --output HDMI1 --off --output LVDS1 --mode 1366x768 --pos 0x0 --rotate normal --output VGA1 --off

There is a gui for xrandr called arandr that I used and it created these scripts, I made an alias in my .bashrc and call them as I need them.

I did have problems using both displays in gnome under ubuntu but not having any problems so far in Arch Linux with xmonad or awesome window managers.

 

grandr is another gui you might try.

 

Just wanted to add to this, I have also tried it in debian wheezy, xrandr works fine for using both displays.

X220 running Debian Wheezy with Xmonad

 

I'm at my wit's end with this, I'll probably have to give up on Linux. No matter what distribution I use, I ALWAYS end up getting the video corruption bug (random blocks and characters getting corrupted on various elements of the desktop), almost always accompanied by crashes. I've searched for a solution for months but to no avail.

This is the worst laptop I have ever used as far as Linux support goes.

 

yeah!If you can test how the fan and battery holds up in ubuntu it would be helpful. Also maybe the special media buttons?

alcohol tests

 

This need not be the case, it certainly isn't for me, even with i915 rc6 enabled. What boot options are you using, and what kernels on what distros?

 

I have used the following:

Ubuntu 11.04
Ubuntu 11.10
Ubuntu 12.04 alpha
Lubuntu 11.04
Lubuntu 11.10
Lubuntu 12.04 alpha
Fedora 15
Fedora 16
Fedora 17 alpha
OpenSUSE 12.1
Mint 11
Mint LXDE
Mint 12

And a few others I'm not sure of anymore.

Many distros like Mageia, Debian, CentOS/Scientific Linux don't detect my wireless network card.

Are you saying I should change something in my bios?

I don't know what to do anymore to stop these glitches/crashes. At some point on Ubuntu 11.10 I disabled sync on v-blank in compiz and it looked like it reduced them greatly, but ultimately they just occured again.

I think this happens because of Intel's -poor video drivers for the video card in the Lenovo.

 

Any distro with a kernel lower than 3.1 is going to be pretty crappy. So that eliminates most of your list.

Try a 3.1 or later kernel on a bleeding edge distro for the best sandy bridge graphics support. If you still have problems, you can disable i915 options like this at the grub boot prompt (note that depending on what distro and kernel you run, these may be enabled or disabled by default):

Code:

i915.i915_enable_fbc=0

and/or

Code:

i915.i915_enable_rc6=0

The reason Debian doesn't see your wireless card is because you need to install proprietary firmware for it to work.

WiFi - Debian Wiki

iwlwifi - Debian Wiki

 

A few of these distros had fairly recent kernels, the Lubuntu and Ubuntu betas for example had 3.2. From what I can recall, although it seems to diminish these problems, it definitely doesn't eliminate them entirely.

I might try with your command line options. I don't understand what it so hard with supporting Sandy Bridge properly.

 

I am currently using Arch XFCE with Compiz, Fedora 16 Gnome with a clutter environment variable to enable vsync for smooth video playback.

https://bugzilla.gnome.org/show_bug.cgi?id=657071#c2

I also use Mageia 1 KDE with Compiz. Compiz in Arch XFCE and Mageia KDE enable vsync.

Really not having any problems with artifacts in any of the three distros I am using.

 

Sandy Bridge introduced some major GPU hardware changes. IIRC one of those changes was a change to the hardware used by the driver to figure out when the vblank interval is.

The real answer to the question is that Intel didn't seem to really work on the necessary i915 changes until way late in the hardware release cycle. They also had a pretty lousy record with being too aggressive in terms of defaulting to use code with tons of recent churn (the 2.6.39 i915 drama was an example of this.) They seem to have learned their lesson though -- this time around with Ivy Bridge, Intel's been adding mainline support before the hardware hits the market.

If you're running i915 on an SB-equipped ThinkPad, I'd bet that most of your fun driver issues will come in the form of rings missing semaphores. On the plus side, it's easy to avoid (i915.semaphores is the module option you want.) On the minus side, disabling semaphores can negatively impact driver stability. Wait, did I just say that semaphores can hurt stability either way? Yeah. Fun, isn't it? Fortunately, one of the two settings is likely to provide better results on your particular hardware... still, it's pretty annoying.

RC6, FBC, and LVDS downclocking *should* be pretty stable as of 3.2.x. IIRC there used to be some odd issues with LVDS downclocking and changing refresh rate settings while Xorg was up (some nice seizure-inducing visual artifacts,) but I think those have been solved with recent kernel/Xorg combinations.

 

A new version of the xorg synaptics driver came out, 1.5.99.901: [ANNOUNCE] xf86-input-synaptics 1.5.99.901

There is supposed to be support for click-and-drag on clickpads with this patch: [synaptics,v5,13/18] Enable clickpad click and drag with two fingers - Patchwork

Unfortunately I haven't been able to get that to work. Click and drag doesn't work at all when I enable the clickpad parameter. I have tried it both with and without two finger scrolling enabled. Two finger scrolling also feels sloppy/laggy with this new release. I'm going to poke around more to hopefully find some clues.

It took me some time, but I found reasonable coordinates for the X220 clickpad. Here is the incantation for this new synaptics driver:

Code:

synclient ClickPad=1 RightButtonAreaLeft=3700 RightButtonAreaRight=5650 MiddleButtonAreaRight=3600 MiddleButtonAreaLeft=3400

Note: MiddleButtonAreaLeft has to come before right or there will be a BadValue/out of range error.

Update: I bulit and packaged 1.5.99.902 but still no luck.

 

Received my X220 yesterday and needless to say win7 didn't last long
Running Archlinux with gnome 3 and Cairo-dock and this thing runs cool, quiet and fast, had some initial problems with ssd's though. Seems this laptop does not like ahci setting with intel X25-M G1 and G2, it causes COMRESET -16 error's on startup maybe it's a bios problem with 1.28. Kinda put the latest on straight away when i still had windows on it so meh.

Either way since for now i'm using a G1 the ahci setting is not that important because this drive does not support trim. Even still i hope that when i get a Samsung 830 in a couple of weeks it will work ok.

Other then that i really like this laptop, using some settings from this thread powertop says i'm using between 7.5 and 10w on average which is great

 

Very interesting thread. After some unsuccessful tries I finally installed ubuntu 12.04 beta in UEFI on my ssd and enabled trim (haven't checked if it works yet).

The AutoScroll and Chromium Wheel Smooth Scroller extensions made my life a lot easier in Chromium but I'm mainly using an external keyboard and mouse. Actually my touchpad is buggy high left click does the same thing as the middle click for some reason.