Just as an update, I've updated my DSDT for the latest BIOS (1.16) from Lenovo. You can find the latest copy of my DSDT here:
pbui / abs-pnutzh4x0r / source — bitbucket.org
Also, I would like to note that the latest proprietary ATI graphics driver (catalyst-10.4) works great with the HD 3200 and allows for the machine to run cooler and have longer battery life. Moreover, it no longer requires xorg-1.6... yay! Additionally, the realtek wireless drivers are more stable now, especially if you disable power management, but I still get a weak signal and disconnects from at my school... sucks, but everything mostly is fine.
-
-
-
Offtopic: Really love the laptop but the battery issue is really bothering me. Ubuntu Lucid is reporting 48% capacity. The battery now lasts for only 40-50 minutes before shutting down. -
Confirmed, it works.
Thanks for the post. had to practice the kernel build (used git) in order to make this work, it seemed to complicated three weeks ago when I first purchased the system but the battery was draining like crazy. the graphics driver also helps in lowering power consumption.
However, still could not figure out how to apply the recent bios update by lenovo (which is designed for solely for windows it seems!), help would be appreciated -
After the kernel update (including the dsdt.hex file into the kernel) and loading the ati catalyst driver things were looking good. However, I closed the lid, went to sleep, next day when I opened the lid the computer did not wake up
I pressed the power button, it woke up however, computer froze!
Then I rebooted this time my brightness control using the function keys became inactivated (although in the upper right corner I could see the brightness level was changing as I played with the keys). no brigthness control.
at the end of the output of dmesg, I get several lines like
Code:[fglrx:firegl_acpi_video_event] *ERROR* Could not find private acpi context by busid: LCD
I inactivated the tp_smapi module, that did not work either...
Any ideas are appreciated! -
The glossy screen kills any interest I might have had. Why even put it in the Thinkpad line?
-
Hi,
I just got my edge today and spent the day with installing archlinux and xmonad on it. I really like it and it feels very solid, but the lack of cpu-scaling is really annoying so I thought that I maybe also could modify my DSDT, but I wanted to know, if that is somehow dangerous to the hardware or could void the warranty. And I also want to know, if I'll have to change something in the DSDT for my System, because I have the Version with 2048 MB Ram and the 160GB harddrive.
Regarding battery: I tested it today without cpu-scaling and I had it running for 3h20min. In that time I was surfing with firefox, configuring the new system, testing the webcam via vlc (it worked out of the box). And in the midtime I switched from the opensource drivers to the catalyst and not only had a performance boost, but the power consumption was also about 4 W lower (powertop).
@pnutzh4x0r: thx for your great work in this thread, i'm allready using your pkgbuild for the wlan from aur!
/edit: Yesterday the 8192 driver worked pretty well, but today I lose connection every 5 minutes. I wonder if it could be because I don't have the 8192 chipset, but the 8172 chipset (according to lspci), but I can't find any drivers for 8172. Or is it the same Chipset?
Do you know if Realtek will improve the quality of the driver in near future?
And another question: Do you think lenovo will maybe release a BIOS update, so that cpu-scaling will be possible without overriding the DSDT? -
the bios upgrade using grub2 works... it is described here BIOS Upgrade - ThinkWiki
-
wouldn't it also work just using dd to copy the image on an usb-disk?
another question. What values do you get from sensors?
My output looks like this: sensors
k8temp-pci-00c3
Adapter: PCI adapter
Core0 Temp: +69.0°C
Core0 Temp: +66.0°C
Core1 Temp: +65.0°C
Core1 Temp: +68.0°C
acpitz-virtual-0
Adapter: Virtual device
temp1: +26.8°C (crit = +100.0°C)
But I can't believe that's true. -
well yes the adapter temperature does not look real, besides I've got exactly the same temperature as yours
$ sensors
acpitz-virtual-0
Adapter: Virtual device
temp1: +26.8°C (crit = +100.0°C)
k8temp-pci-00c3
Adapter: PCI adapter
Core0 Temp: +65.0°C
Core0 Temp: +58.0°C
Core1 Temp: +71.0°C
Core1 Temp: +67.0°C
Some extra observations about performance. For the last three weeks, I've been running an in-house kernel compiled to include the dsdt. file provided here. My bios version was 1.12-1.04. Without the dsdt.hex file being externally loaded, I got an error after I logged in complaining about not being able to use CPU frequency scaling. The dsdt.hex got rid of that and I could see the change in CPU freq in idle times thru the panel CPU monitor.
However, that really did not have an impact on battery life, at least not as much as I thought it would. I am confident on that because after I upgraded my BIOS firmware yesterday to version 1.18-1.09, I switched back to kernel without dsdt.hex override, thinking that the 1.18 upgrade may have implemented a fix in the dsdt table. Nope, its still not fixed cause I got the CPU freq scaling support error.
The interesting thing is without CPU freq scaling I get a power consumption in the same level, around 21W, as that when I had the kernel with dsdt override.
Additionally, when I had the dsdt override my backlight adjustment keys did not work, i.e. I was unable to adjust the screen brightness using the F7 and F8 keys (which was uncomfortable).
Another thing is when I close the lid and the machine goes to suspend, it wakes back up but is totally unresponsive and requires a forced reboot by pressing the power button. This occurs whether the dsdt override is used or not. So I have to suspend the machine, not by closing the lid, but by using the suspend option in the top panel (on screen). Then wake up has no problems.
the battery drains real fast it feels, with or without the dsdt override. It takes less than two hours, at a rate of approx 21 W.
A correction about power consumption:
after switching to ati catalyst driver the power rate dropped to between 12-14 W from around 21 W. -
Yes, the power consumption with catalyst is way lower.
I realized some freezes now and then, but can't determine the reason. Did you also realize freezes with no obvious reason? Does someone know how to find out, what causes that freezes? They're not very often, but too often for a Linux-System.
Could it be the temperature? Does the BIOS-Update provide monitoring for Temperature?
/edit:
I just tried to run glxgears to test the temperatur. Sensors is still reporting strange values, but I noticed that the temperatures from sensors went up after I startet glxgears. But I have a problem do determine the gpu-temperature because aticonfig --adapter=0 --od-gettemperature says:
ERROR - Get temperature failed for Apapter 0 - ATI Radeon HD 3200 Graphics
Do you also have that problem?
/edit2:
I just stoped glxgears and the temperatures dropped from ~80 to 66°C, so something in sensors seems to work, but I don't think that, the temperatures are correct. -
I had freezes with Edge 13 0197-6GG. My computer frooze at wake up from suspend by closing the lid (suspending from menu did not lead to such a probem).
I have upgraded my bios (to 1.18) and have not experienced any freezes ever since... -
Could you tell me how you updated the BIOS? Do you have an external CD-Drive? Because I tried to copy the .iso to a usb-disk with dd and also with unetbootin, but i couldn't get it work. I also tried to extract the .exe with cabextract, but that also didn't work.
-
the bios upgrade using grub2 works... it is described here BIOS Upgrade - ThinkWiki
see the section: GRUB2 boot cd image
I had to dig and find a file called memdisk from Index of /pub/linux/utils/boot/syslinux/3.xx
Use this file, not the one in /usr/lib/syslinux/... -
I just updated the BIOS using Grub, it also works.
I copied memdisk from my local install:
dolph ~ $ pacman -Qs syslinux
local/syslinux 4.01-1
And put both in /boot
Then I edited the /boot/grub/menu.lst, to look like this:
# (2) BIOS Update
title BIOS Update
root (hd0,0)
kernel /memdisk iso raw
initrd /biosupdate.iso
The update really takes about a minute. That was a really long minute.After the update was finished it made a really loud and long beep, just to warn others, who might try this.
I can't say if the updated has fixed the problem with system freezes, because I've done it just an hour ago, but after a few days I should know it. -
Could it be, that the wlan-driver causes system-freezes? Today I had another two freezes and I just looked at the messages.log and found this:
Code:Jul 4 14:52:47 edge NetworkManager: <info> Activation (wlan0) starting connection 'Auto Sabrina' Jul 4 14:52:47 edge NetworkManager: <info> (wlan0): device state change: 3 -> 4 (reason 0) Jul 6 20:14:57 edge kernel: ===>rtl8192se_link_change():ieee->iw_mode is 2 Jul 6 20:14:57 edge kernel: rtl8192_update_cap(): WLAN_CAPABILITY_SHORT_PREAMBLE Jul 6 20:15:03 edge kernel: ===>rtl819x_watchdog_wqcallback(): AP is power off,chan:5, connect another one Jul 6 20:15:03 edge kernel: ===========>RemovePeerTS,00:1d:19:0d:fe:13 Jul 6 20:15:03 edge kernel: ====>remove Tx_TS_admin_list Jul 6 20:15:03 edge kernel: ===>rtl8192se_link_change():ieee->iw_mode is 2 Jul 6 20:15:03 edge kernel: notify_wx_assoc_event(): Tell user space disconnected Jul 6 20:15:03 edge NetworkManager: <info> (wlan0): supplicant connection state: completed -> disconnected Jul 6 20:15:03 edge kernel: =====>rtl8192_set_chan()====ch:1 Jul 6 20:15:03 edge NetworkManager: <info> (wlan0): supplicant connection state: disconnected -> scanning Jul 6 20:15:03 edge kernel: =====>rtl8192_set_chan()====ch:2 Jul 6 20:15:03 edge kernel: =====>rtl8192_set_chan()====ch:3 Jul 6 20:15:03 edge kernel: =====>rtl8192_set_chan()====ch:4 Jul 6 20:15:04 edge kernel: =====>rtl8192_set_chan()====ch:5 Jul 6 20:15:04 edge kernel: =====>rtl8192_set_chan()====ch:6 Jul 6 20:15:04 edge kernel: =====>rtl8192_set_chan()====ch:7 Jul 6 20:15:04 edge kernel: =====>rtl8192_set_chan()====ch:8 Jul 6 20:15:04 edge keJul 6 20:18:45 edge kernel: Initializing cgroup subsys cpuset Jul 6 20:18:45 edge kernel: Initializing cgroup subsys cpu
Code:Jul 6 20:15:04 edge keJul 6 20:18:45 edge kernel: Initializing cgroup subsys cpuset
It was a new type of freeze. Before the BIOS-Update I had some freezes, where the LCD just went black I had to hard reset. But this time it doesn't goe really black, but just faded to black, so there was still light.
Could that have something to do with me having done modprobe -r r8192se_pci and modprobe r8192se_pci to get my wireless working again? At the freeze before I have done the same. -
Just a little note from my side.
It seems that the sensors-module is working now!
Thinkpad Edge AMD impressions
Discussion in 'Lenovo' started by pnutzh4x0r, Feb 3, 2010.