Dell Inspiron 6000

[seppelrockt]

Hardware Question regarding the CD-ROM/DVD device:

Today I got a slimline DVD burner NEC 6650 (not from Dell) that is said to be a Master device (or in other words, not "inverse cable select"). For better understanding: slimline devices do not have any jumper thus the are "jumpered" via firmware to "Master" or "Slave". As the notebook vendors do not speek or write much over the used configuration it's not quiete easy to install an of-the-shelf" slimline drive.

The drive works fine with DMA enabled and I can boot from it (if I hab a wrongly jumpered one, it would be PIO only and no booting). Well I'm was quite lucky I thought. Only thing I had to change was the front of the drive and that was not a problem.

OK, then I gave the old drive (original Dell) to a friend to use it in his notebook - and it tuwrned out not to work cause he needs a MASTER and the drive was recognized as SLAVE! Hoopy, Master in my Dell I6000 and slave in his notebook. That's quite impossible usaly, but I have a theory:

When the i915gm in the Dell I6000 is SATA only, but we use standard slimline drives in there, there must be a bridge ship between the DVD 50pol (P)ATA connector and the mainbords SATA connector. Given this bridge chip can handle both master and slave drives than we can use both of them without any problems cause the bridge handles the whole master/slave thing and the SATA interface doesn't care about master / slave at all.

Does that make sense and does anybody have more information how the dvd drive things are done with i915gm?
Happy upgrading! Ahhh, can anybody recomment an slimline dvd burner other than the NEC? I like it but it has very poor error correction for CD-ROM so it can't read a CD-ROM that never was a problem before.

[seppelrockt]

Found something interesting in the device specs for udev on http://www.lanana.org/docs/device-list/devices-2.6+.txt

For SCSI devices, /dev/tape and /dev/cdrom should point to the
``cooked'' devices (/dev/st* and /dev/sr*, respectively), whereas
/dev/cdwriter and /dev/scanner should point to the appropriate generic
SCSI devices (/dev/sg*).

Maybe that's the proper way to go, create a /dev/cdwriter linked to /dev/sg1 and set group of /dev/sg1 to cdrw - the permissions for burning are very confusion atm - why do we have a group cdrw when it's enough to be in cdrom?

I guess following the idea above makes sense, as everything but burning works without permissions for /dev/sg1 and one has the possibility to add fewer people to "cdrw" than to "cdrom" for securety reasons. And it's more transparent - one can guess from the group cdrw that this device note must be important for burning (and not much more).

[VinzC]

Found something interesting in the device specs for udev on http://www.lanana.org/docs/device-list/devices-2.6+.txt

For SCSI devices, /dev/tape and /dev/cdrom should point to the
``cooked'' devices (/dev/st* and /dev/sr*, respectively), whereas
/dev/cdwriter and /dev/scanner should point to the appropriate generic
SCSI devices (/dev/sg*).

And I believed sr meant "SCSI Recorder"... Hence my mistakes: I thought permissions on /dev/sr* had to allow write operations... But you also see /dev/sr* is deprecated in favour of /dev/scd*. That's more logical I think.

[VinzC]

@seppelrockt:

You said you sucessfully used gentoo sources, vesa-tng and a 16/10 resolution for fbsplash (1280x800). Please, would you mind telling me how you did? I've been trying to have fbsplash use my full resolution for months but all I get is either a black or corrupted screen. I've no choice but use vesafb and 1280x1024 consoles instead. It's quite frustrating on a wide screen like this is.

Many many many many many many thanks in advance.

[seppelrockt]

@seppelrockt:

You said you sucessfully used gentoo sources, vesa-tng and a 16/10 resolution for fbsplash (1280x800). Please, would you mind telling me how you did? I've been trying to have fbsplash use my full resolution for months but all I get is either a black or corrupted screen. I've no choice but use vesafb and 1280x1024 consoles instead. It's quite frustrating on a wide screen like this is.

Many many many many many many thanks in advance.

What exactly is the problem when you use the vesafb-tng (or used it the last time)? In case you think of the console mess when switching from X and back, the VBERestore in xorg.conf will do the trick (If it's not in this thread please have a lock in your thread about DRI and i915gm [The patch howto] - I ... hope ... I posted my xorg.conf there). Yes, I'm quite sure it was there and it was about the same question.

I run nothing special here, gentoo-sources-2.6.12-r4, xorg-6.99.15 (with your patch) and BIOS A07. I hope this helps - if not ww can dig even deeper (to see how deep the rabbit whole goes;)

The VBRestore messes the screen for some seconds e.g. when I close my X session with KDE - that mees it doesn't run 100% perfect but is useable of course (maybe we can make it perfect in the future?).

[VinzC]

What exactly is the problem when you use the vesafb-tng (or used it the last time)? In case you think of the console mess when switching from X and back, the VBERestore in xorg.conf will do the trick (If it's not in this thread please have a lock in your thread about DRI and i915gm [The patch howto] - I ... hope ... I posted my xorg.conf there). Yes, I'm quite sure it was there and it was about the same question.

I run nothing special here, gentoo-sources-2.6.12-r4, xorg-6.99.15 (with your patch) and BIOS A07. I hope this helps - if not ww can dig even deeper (to see how deep the rabbit whole goes;)

The VBRestore messes the screen for some seconds e.g. when I close my X session with KDE - that mees it doesn't run 100% perfect but is useable of course (maybe we can make it perfect in the future?).

Yes, I know about the VBERestore trick and I've used it. In fact what I get is less worse than without. But the console screen is still unusable when switch from X back to text. The mess remains for ever. I can wait as long as I want, consoles remain unusable.

All my console screens look like 640x480 and the text is truncated halfway down. But the text is green instead of grey. Besides it looks as if on an interlaced screen. As for framebuffer consoles the image is like inverted (bitwise not). To put it all together, the top half of the console text is displayed on a centered 640x480 rectangle.

Finally all my attempts to use a 1680x1050 framebuffer console results in a black screen. X starts properly though. Only text consoles are unusable.

[seppelrockt]

Hmmm,

first you might have noticed that I only hav WXGA (1280x800) so it might be a problem of your high screen resolution.
Second did you play with your BIOS setting for "interplation" of the screen (when I resolutions below the default resolution of the screen)? I have interpolation off, seeing black borders around a centred screen with 1024x768 resolution. Don't know if this is related.

What else might it be? I have no idea ... If you want my kernel .config or something, please let me know!

Good Luck,
seppelrockt

[seppelrockt]

As a little "sunday gift" I found out today that alsa-1.0.9 silently found it's way into the kernel in the last weeks - now it does software mixing out of the box. This means I can play wesnoth and listen to ogg music at the same time - no more apps blocking the sound device. GREAT!

The downside is that I have many noise in the playback from apps that use oss(-emulation) . Any ideas how to fix this?

[VinzC]

first you might have noticed that I only hav WXGA (1280x800) so it might be a problem of your high screen resolution.
Second did you play with your BIOS setting for "interplation" of the screen (when I resolutions below the default resolution of the screen)? I have interpolation off, seeing black borders around a centred screen with 1024x768 resolution. Don't know if this is related.

Yep, I have noticed the difference. But I'd be quite surprised if it was the only cause. I indeed disabled screen expansion so I have black borders at lower resolutions.

What else might it be? I have no idea ... If you want my kernel .config or something, please let me know!

IIRC you've posted your kernel config at some places, didn't you? I'll try finding it back first. Unless you have added/removed features. I'm currently playing around with 2.6.12-suspend2-r4.

EDIT: also using BIOS A07 and Xorg 6.8.99-15.

[seppelrockt]

I guess the problems might be solved in gentoo-sources then? Worth a try maybe!

I just managed to get easy access to my webspace on the server of the university (with kio_sftp from Konqueror - works great!) so I can finaly present you and the world out there my xorg.conf and kernel config!

[VinzC]

I guess the problems might be solved in gentoo-sources then? Worth a try maybe!

I tried Gentoo Sources 2.6.12-r8. Same problems. I'll try your kernel config maybe.

[seppelrockt]

Has anybody a working speedstep with the Sonoma Pentium-M on Dell I6000? I've tried the powersaving-HowTo but:

Code: Select all

cpufreq-info
cpufrequtils 0.3: cpufreq-info (C) Dominik Brodowski 2004
Report errors and bugs to linux@brodo.de, please.
analyzing CPU 0:
  no or unknown cpufreq driver is active on this CPU

In my kernel config there is:

Code: Select all

#
# CPU Frequency scaling
#
CONFIG_CPU_FREQ=y
CONFIG_CPU_FREQ_TABLE=y
CONFIG_CPU_FREQ_DEBUG=y
CONFIG_CPU_FREQ_STAT=y
CONFIG_CPU_FREQ_STAT_DETAILS=y
# CONFIG_CPU_FREQ_DEFAULT_GOV_PERFORMANCE is not set
CONFIG_CPU_FREQ_DEFAULT_GOV_USERSPACE=y
CONFIG_CPU_FREQ_GOV_PERFORMANCE=y
CONFIG_CPU_FREQ_GOV_POWERSAVE=y
CONFIG_CPU_FREQ_GOV_USERSPACE=y
CONFIG_CPU_FREQ_GOV_ONDEMAND=y
# CONFIG_CPU_FREQ_GOV_CONSERVATIVE is not set

#
# CPUFreq processor drivers
#
CONFIG_X86_ACPI_CPUFREQ=m
# CONFIG_X86_POWERNOW_K6 is not set
# CONFIG_X86_POWERNOW_K7 is not set
# CONFIG_X86_POWERNOW_K8 is not set
# CONFIG_X86_GX_SUSPMOD is not set
CONFIG_X86_SPEEDSTEP_CENTRINO=y
CONFIG_X86_SPEEDSTEP_CENTRINO_TABLE=y
# CONFIG_X86_SPEEDSTEP_ICH is not set
# CONFIG_X86_SPEEDSTEP_SMI is not set
# CONFIG_X86_P4_CLOCKMOD is not set
# CONFIG_X86_CPUFREQ_NFORCE2 is not set
# CONFIG_X86_LONGRUN is not set
# CONFIG_X86_LONGHAUL is not set

#
# shared options
#
# CONFIG_X86_ACPI_CPUFREQ_PROC_INTF is not set
# CONFIG_X86_SPEEDSTEP_LIB is not set

[VinzC]

Has anybody a working speedstep with the Sonoma Pentium-M on Dell I6000?

Code: Select all

$ zgrep -Ei '^config_cpu|^config_x86' /proc/config.gz
CONFIG_X86=y
CONFIG_X86_PC=y
CONFIG_X86_CMPXCHG=y
CONFIG_X86_XADD=y
CONFIG_X86_L1_CACHE_SHIFT=6
CONFIG_X86_WP_WORKS_OK=y
CONFIG_X86_INVLPG=y
CONFIG_X86_BSWAP=y
CONFIG_X86_POPAD_OK=y
CONFIG_X86_GOOD_APIC=y
CONFIG_X86_INTEL_USERCOPY=y
CONFIG_X86_USE_PPRO_CHECKSUM=y
CONFIG_X86_TSC=y
CONFIG_X86_MCE=y
CONFIG_X86_MSR=y
CONFIG_X86_PM_TIMER=y
CONFIG_CPU_FREQ=y
CONFIG_CPU_FREQ_TABLE=y
CONFIG_CPU_FREQ_STAT=y
CONFIG_CPU_FREQ_DEFAULT_GOV_PERFORMANCE=y
CONFIG_CPU_FREQ_GOV_PERFORMANCE=y
CONFIG_CPU_FREQ_GOV_POWERSAVE=y
CONFIG_CPU_FREQ_GOV_USERSPACE=y
CONFIG_CPU_FREQ_GOV_CONSERVATIVE=y
CONFIG_X86_SPEEDSTEP_CENTRINO=y
CONFIG_X86_SPEEDSTEP_CENTRINO_ACPI=y
CONFIG_X86_SPEEDSTEP_CENTRINO_TABLE=y
CONFIG_X86_SPEEDSTEP_ICH=y
CONFIG_X86_SPEEDSTEP_SMI=y
CONFIG_X86_SPEEDSTEP_LIB=y
CONFIG_X86_BIOS_REBOOT=y

Speedstepping works perfectly on mine.

[seppelrockt]

Yes, I compiled the ACPI stuff into the kernel and added some more and now it works.

What are you using as deamon? I have cpufreqd here and realized that it produces 15 - 30 % load itself! That's why it never got below 1.33 Ghz yet with conservative governor. I will change the settings but this high load is very bad!

EDIT: I had to set acpi_workaround=1 in /etc/cpufreqd.conf to discrease load of the cpufreqd to ~ 6%. Still quite much though but at least a temporary workaround.

[seppelrockt]

Hi all,

now that speedstep works on my Dell I enabled i8k and in 800Mhz mode with the fan of I have max. 44°C - Great Silence!
I found a way to get i8k working as modul (to lasy to recompile the kernel again):

create the file in /etc/modules.d/i8k with the following line in it:

Code: Select all

options i8k force=1

And you are done! Of cause don't forget to put i8k into /etc/modules.autoload/kernel-2.6, too.

[VinzC]

create the file in /etc/modules.d/i8k with the following line in it:

Code: Select all

options i8k force=1

And you are done! Of course don't forget to put i8k into /etc/modules.autoload/kernel-2.6, too.

Good. I used stormer's trick, i.e. kernel argument i8k.force. But this is even nicer. Thanks for the info.

[VinzC]

Something that might be of interrest too.

I've got CD burning working again with my lap'. First I had to change /etc/udev/rules.d/50-udev.rules:

Code: Select all

# grep sr /etc/udev/rules.d/50-udev.rules | head -n 1
BUS=="scsi", KERNEL="sr[0-9]*", PROGRAM="/etc/udev/scripts/cdsymlinks.sh %k", SYMLINK+="%c{1} %c{2} %c{3} %c{4} %c{5} %c{6}", GROUP:="cdrw"

.
Then I added the following custom rules, to have /dev/sg0 and /dev/sg1 readable and writeable respectively:

Code: Select all

# cat /etc/udev/rules.d/10-local.rules
BUS=="scsi", KERNEL="sg0", PROGRAM="/etc/udev/scripts/cdsymlinks.sh %k", SYMLINK+="%c{1} %c{2} %c{3} %c{4} %c{5} %c{6}", GROUP:="cdrom", MODE="0440"
BUS=="scsi", KERNEL="sg1", PROGRAM="/etc/udev/scripts/cdsymlinks.sh %k", SYMLINK+="%c{1} %c{2} %c{3} %c{4} %c{5} %c{6}", GROUP:="cdrw", MODE="0660"

Note the groups. Finally I installed nautilus-cd-burner-2.10.2 (masked) with USE flag hal:

Code: Select all

# equery u nautilus-cd-burner
...
[ Found these USE variables for gnome-extra/nautilus-cd-burner-2.10.2 ]
 + + cdr    : Adds support for CD writer hardware
 + + hal    : Enable Hardware Abstraction Layer (HAL) support
...

I don't know however whether the masked version is necessary or not. I think the hal flag should be enough. But I've waited too long for my CD burner to be useable again. Note: changing UDEV rules won't help enough for Nautilus CD burner tries to get something with /dev/sg0 and I don't know why; /dev/sg0 is the SATA hard disk.

After adding my rules and changing the default ones about SCSI disks, here are the access rights I expected:

Code: Select all

 # ls -l /dev/sg*
cr--r-----  1 root cdrom 21, 0 Aug 15 18:32 /dev/sg0
crw-rw----  1 root cdrw  21, 1 Aug 15 18:32 /dev/sg1

Code: Select all

# ls -l /dev/sr*
brw-rw----  1 root cdrw 11, 0 Aug 15 18:32 /dev/sr0

I have now burnt my first CD-RW since UDEV gave me headaches .

EDIT: is it because I now can burn CDs or because I'm using suspend2-sources-2.6.12-r4 instead of suspend2-sources-2.6.12-r4? I'm now experiencing random lockups like some of you . And I didn't before.

[seppelrockt]

Sorry, can't help you out here - I did only have two (random) KDE freezes in the last weeks with my system and I changed my udev rules quite similar to your.

Why did you change the first rule direcly in 50-udev-rules instead of adding a rule to 10-local-rules? NOTE the syntax! The ":=" tells udev not to overwrite a setting, e.g. if you set the group of /dev/sr[0-9]* to :=cdrw in 10-local-rules it will not be overwritten from 50-udev-rules (I guess you then should have = instead of := in 50-udev-rules). Might be wise to "fix" things in trhe right places!

Are you sure /dev/sg0 is the harddrive? I wonder why this is needed for burning (same here)?

For everybody how is interessted I've tweaked i8kmon for my needs (automatic fan control). I have added i8k to my default runlevel (if you don't want the i8kbutton deamon to start as well then comment out all button stuff in /etc/conf.d/ik8). Then I put this config file for i8kmon to /etc/i8kmon:

EDIT: Don't use the config, it may cause damage cause the fan doesn't start at all !!! Everything works great without any config. [File removed]!

[VinzC]

Why did you change the first rule direcly in 50-udev-rules instead of adding a rule to 10-local-rules? NOTE the syntax! The ":=" tells udev not to overwrite a setting, e.g. if you set the group of /dev/sr[0-9]* to :=cdrw in 10-local-rules it will not be overwritten from 50-udev-rules (I guess you then should have = instead of := in 50-udev-rules). Might be wise to "fix" things in the right places!

Yes, I know that but my lap' froze as I was trying the new UDEV rules as I put them into 10-local.rules. I thought the latter was responsible for the freeze. So I removed them from the file and changed the default rules. I'll probably revert to the initial situation then.

Are you sure /dev/sg0 is the harddrive? I wonder why this is needed for burning (same here)?

Code: Select all

# dmesg | egrep 'scsi'
scsi0 : ata_piix
scsi1 : ata_piix
Attached scsi disk sda at scsi0, channel 0, id 0, lun 0
sr0: scsi3-mmc drive: 24x/24x writer cd/rw xa/form2 cdda tray
Attached scsi CD-ROM sr0 at scsi1, channel 0, id 0, lun 0
Attached scsi generic sg0 at scsi0, channel 0, id 0, lun 0,  type 0
Attached scsi generic sg1 at scsi1, channel 0, id 0, lun 0,  type 5

[VinzC]

For everybody how is interessted I've tweaked i8kmon for my needs (automatic fan control). I have added i8k to my default runlevel (if you don't want the i8kbutton deamon to start as well then comment out all button stuff in /etc/conf.d/ik8). Then I put this config file for i8kmon to /etc/i8kmon:

Code: Select all

# Kernel I8K status file
set config(proc_i8k)    /proc/i8k

# Kernel APM status file
#set config(proc_apm)    /proc/apm

# Kernel ACPI status file
set config(proc_acpi)   /proc/acpi/ac_adapter/AC/status

# External program to control the fans
set config(i8kfan)      /usr/bin/i8kfan

# Applet geometry, override with --geometry option
set config(geometry)    {}

# Run as daemon, override with --daemon option
set config(daemon)      1

# Automatic fan control, override with --noauto option
set config(auto)        1

# Report status on stdout, override with --verbose option
set config(verbose)     0

# Status check timeout (seconds), override with --timeout option
set config(timeout)     5

# Temperature display unit (C/F), override with --unit option
set config(unit)        C

# Temperature threshold at which the temperature is displayed in red
set config(t_high)      80

# Minimum expected fan speed
set config(min_speed)   0

# Temperature thresholds: {fan_speeds low_ac high_ac low_batt high_batt}
set config(0)   {{0 0}  -1  65  -1  65}
set config(1)   {{1 0}  60  85  60  85}
set config(2)   {{2 0}  75 128  75 128}

Note that it seems not to work when you have "-" instead of "0" for the second fan (which doesn't exist on I6000) in the thresholds settings as proposed in the docs.

With this config I have the fan of for my everyday work (I have working speedstep) with ~ 45°C max on 800Mhz and when the temp reaches 65 °C the fan will change to "1" (slow) - this is what you would have all the time with BIOS fan control. Works as expected so far.

I don't have such a file but the fan does start at about the same temperature. As soon as module i8k is loaded (or activated for I have it compiled into the kernel) I notice the fan stops. Are you sure the fan can't do without this?

[VinzC]

Has anyone succeeded in patching recent vanilla or suspend2 sources with these patches that were mentionned by Anterion and mhanski? I can get nothing but rejects when I attempt to use 00_kernel-2.6.12.4.patch. Which ones do I have to apply in fact? All or just the first one?

[seppelrockt]

Any pointers how to get suspend-to-RAM to work? I'm very new on this topic... I don't care about suspend-to-disk cause I don't have enough space on my HD for an enormous swap and it's a securety problem as well. Suspend-to-Ram give me a black screen after reboot.

[VinzC]

Any pointers how to get suspend-to-RAM to work? I'm very new on this topic... I don't care about suspend-to-disk cause I don't have enough space on my HD for an enormous swap and it's a securety problem as well.

Enormous? You should at least have the same amount of RAM plus 30% for suspend-2-disk. Or don't you have any swap space at all?

You also have LZH compression if you want to keep your swap space. But I think you should at least have the same amount of RAM in your swap. As for security you also have encryption if you want to try it. I've tried LZH on a Thinkpad R40 and that worked perfectly.

Suspend-to-Ram give me a black screen after reboot.

Welcome .

[seppelrockt]

For everybody how is interessted I've tweaked i8kmon for my needs (automatic fan control). I have added i8k to my default runlevel (if you don't want the i8kbutton deamon to start as well then comment out all button stuff in /etc/conf.d/ik8). Then I put this config file for i8kmon to /etc/i8kmon:

Code: Select all

# Kernel I8K status file
set config(proc_i8k)    /proc/i8k

# Kernel APM status file
#set config(proc_apm)    /proc/apm

# Kernel ACPI status file
set config(proc_acpi)   /proc/acpi/ac_adapter/AC/status

# External program to control the fans
set config(i8kfan)      /usr/bin/i8kfan

# Applet geometry, override with --geometry option
set config(geometry)    {}

# Run as daemon, override with --daemon option
set config(daemon)      1

# Automatic fan control, override with --noauto option
set config(auto)        1

# Report status on stdout, override with --verbose option
set config(verbose)     0

# Status check timeout (seconds), override with --timeout option
set config(timeout)     5

# Temperature display unit (C/F), override with --unit option
set config(unit)        C

# Temperature threshold at which the temperature is displayed in red
set config(t_high)      80

# Minimum expected fan speed
set config(min_speed)   0

# Temperature thresholds: {fan_speeds low_ac high_ac low_batt high_batt}
set config(0)   {{0 0}  -1  65  -1  65}
set config(1)   {{1 0}  60  85  60  85}
set config(2)   {{2 0}  75 128  75 128}

Note that it seems not to work when you have "-" instead of "0" for the second fan (which doesn't exist on I6000) in the thresholds settings as proposed in the docs.

With this config I have the fan of for my everyday work (I have working speedstep) with ~ 45°C max on 800Mhz and when the temp reaches 65 °C the fan will change to "1" (slow) - this is what you would have all the time with BIOS fan control. Works as expected so far.

I don't have such a file but the fan does start at about the same temperature. As soon as module i8k is loaded (or activated for I have it compiled into the kernel) I notice the fan stops. Are you sure the fan can't do without this?

I have removed the file because it caused serious problems - without the config everythings OK now - THANKS!

[seppelrockt]

Does anybody have experiences with the conservative governor? As it serves it purpose quite good when on AC (Scaling processor speed depending on CPU load without to fast "jumps") it stays at 800 Mhz when on battery.

cpufreq-info looks like this then (NOTE "current policy") - that makes no sense compared to my cpufreqd.conf!

Code: Select all

# cpufreq-info
cpufrequtils 0.3: cpufreq-info (C) Dominik Brodowski 2004
Report errors and bugs to linux@brodo.de, please.
analyzing CPU 0:
  driver: centrino
  CPUs which need to switch frequency at the same time: 0
  hardware limits: 800 MHz - 1.60 GHz
  available frequency steps: 1.60 GHz, 1.33 GHz, 1.07 GHz, 800 MHz
  available cpufreq governors: conservative, ondemand, powersave, userspace, performance
  current policy: frequency should be within 800 MHz and 800 MHz.
                  The governor "conservative" may decide which speed to use
                  within this range.
  current CPU frequency is 800 MHz (asserted by call to hardware).

Code: Select all

# cat /etc/cpufreqd.conf
# An example cpufreq.conf can be found in /etc/cpufreqd.conf.example
# This cpufreq.conf is from www.gentoo.org/doc/en/power-management-guide.xml

[General]
pidfile=/var/run/cpufreqd.pid
poll_interval=3
pm_type=acpi
acpi_workaround=1
#verbosity=5

[Profile]
name=ondemand
minfreq=800000
#maxfreq=1600000
maxfreq=1067000
policy=conservative

[Profile]
name=powersave
minfreq=800000
maxfreq=1600000
policy=powersave

[Profile]
name=performance
minfreq=800000
maxfreq=1600000
#policy=performance
policy=conservative

[Rule]
name=battery
ac=off
profile=ondemand

[Rule]
name=battery_low
ac=off
battery_interval=0-20
profile=powersave

[Rule]
name=ac
ac=on
profile=performance

EDIT: I guess it's the conservative governor itself that sets the freqency limit on battery to 800 Mhz cause I can change them manualy with echo "1600000" > /sys/devices/system/cpu/cpu0/cpufreq/scaling_max_freq and then the CPU freq rises depending on load. However when I switch back to AC and battery again it's 800Mhz again ...

Can somebody help me changing this behaviour of conservative governor with a small script or something or maybe directly in the kernel source. If I could set the battery default max to 1067000 in the source I could even get rit of cpufreqd.