no speedstep for PIII Latitude C800

Bug #36604 reported by Jean Privat
14
This bug affects 1 person
Affects Status Importance Assigned to Milestone
powernowd (Ubuntu)
Invalid
Medium
Unassigned

Bug Description

As per the instructions on https://launchpad.net/distros/ubuntu/+source/powernowd/+bug/10813
I've added a new bug since speedstep_smi fails to load on my laptop. I'm running Dapper updated as of 20060325. speedstep_smi fails to load on my Dell Latitude C800.

$ sudo modprobe speedstep-smi

FATAL: Error inserting speedstep_smi (/lib/modules/2.6.15-19-686/kernel/arch/i386/kernel/cpu/cpufreq/speedstep-smi.ko): No such device

$ sudo modprobe -r speedstep-lib speedstep-smi
$ sudo modprobe speedstep-lib relaxed_check=1
$ sudo modprobe speedstep-smi

FATAL: Error inserting speedstep_smi (/lib/modules/2.6.15-19-686/kernel/arch/i386/kernel/cpu/cpufreq/speedstep-smi.ko): No such device

Revision history for this message
Jean Privat (privat) wrote : cpuinfo

$ cat /proc/cpuinfo

Changed in powernowd:
assignee: nobody → sladen
Revision history for this message
Jean Privat (privat) wrote : dmesg

$ dmesg

Revision history for this message
Jean Privat (privat) wrote : dmi

$ sudo dmidecode

Revision history for this message
Paul Sladen (sladen) wrote :

Do you have an ICH chipset? If so the driver would be:

  sudo modprobe speedstep-ich

the simple answer is that these coppermine CPUs do not report that they are capable of CPU scaling. If you can work out how to get a module loaded we can special case it to work out of the box (but you need to work out what it is first! :-)

No answers, but Googling turned up:

  http://lists.linuxcoding.com/rhl/2004/msg65300.html

Revision history for this message
Jean Privat (privat) wrote :

$ sudo modprobe speedstep-ich

FATAL: Error inserting speedstep_ich (/lib/modules/2.6.15-19-686/kernel/arch/i386/kernel/cpu/cpufreq/speedstep-ich.ko): No such device

Ok, I will wait there are some news. However, if you need I try or test something, just ask me.

Revision history for this message
John Dong (jdong) wrote :

I don't recommend speedstepping these guys. When the state changes, it causes the input devices (mouse, keyboard, and touchpad) to lose sync, halting input. It adds up to extremely annoying dropped keys and stuck mices every couple of seconds. It's primarily a Dell PIII laptop problem.

Revision history for this message
Darik Horn (dajhorn) wrote :

SpeedStep in a similar Dell C800 cannot be enabled in Edgy. Most of the CPUs that were shipped with the C800 do not have SpeedStep built-in, so the 1 GHz Coppermine CPU is a corner case in this laptop model.

Given that SpeedStep is not properly supported on the Dell C800 even in Microsoft Windows, this ticket is a non-bug that should be closed. The issue is just a hardware limitation.

Paul Sladen (sladen)
Changed in powernowd:
assignee: sladen → nobody
Revision history for this message
Nico (nico-rdo) wrote :

OK... non-bug.

Great.

Is there a way to detect this Dell implementation in a code placed in the startup script, and not start speedstep if it's going to be bad?

There are still *a lot* of those machines around.

Nico

Revision history for this message
Paul Sladen (sladen) wrote :

Okay, I'm going to jump and finally Reject this one---based on the fact that it's a hardware limitation.

Nico: Yeah, because this is such a corner case, it's going to be a little hard. Yes, it's possible to match machines based on the 'DMI' data (use 'dmidecode' or look in /var/cache/acpi-support/*-* for the saved output).

One possibility from being able to construct a cast-iron match is to apply this directly into the kernel module.

Changed in powernowd:
status: Unconfirmed → Rejected
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.