Coppermine, cpufreq
Hi,
I have a laptop (Gericom Webboy) with a PIII Coppermine.
I am working on a 2.4.22-kernel (from kernel.org) and patched the patch-2.4.22-ac4.bz2 from Alan Cox against it.
Because of having a low battery-capacity I tried to enable cpufreq and ACPI-stuff in the kernel (and tried to compile them as modules as well).
If I then boot the machine, I tried the following:
$ cat /proc/cpuinfo processor : 0 vendor_id : GenuineIntel cpu family : 6 model : 8 model name : Pentium III (Coppermine) stepping : 6 cpu MHz : 1000.916 cache size : 256 KB physical id : 0 siblings : 1 fdiv_bug : no hlt_bug : no f00f_bug : no coma_bug : no fpu : yes fpu_exception : yes cpuid level : 2 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 mmx fxsr sse bogomips : 1998.84
$ cat /proc/cpufreq
minimum CPU frequency - maximum CPU frequency - policyroot:/usr/src/linux-2.4.22# insmod speedstep-ich Using /lib/modules/2.4.22-ac4-486/kernel/arch/i386/kernel/speedstep-ich.o /lib/modules/2.4.22-ac4-486/kernel/arch/i386/kernel/speedstep-ich.o: unresolved symbol speedstep_get_processor_frequency /lib/modules/2.4.22-ac4-486/kernel/arch/i386/kernel/speedstep-ich.o: unresolved symbol speedstep_detect_processor /lib/modules/2.4.22-ac4-486/kernel/arch/i386/kernel/speedstep-ich.o: unresolved symbol speedstep_get_freqs
and didn't saw any real frequencies.
At the moment I don't know if my cpu could handle cpufreq-controlling at all, but hope it. I searched the archive of this ml, but except of "Coppermines might cause problems, might not work..." I can't find helpful things.
Thanks a lot and if there is something you need to help get it running, or I should do or try, please tell me. (Well I tried to understand the stuff in Documentation/cpu-freq, but...;))
Best regards,
Matthias
Signature · 4 lines
Freundliche Grüße, ### http://www.stop1984.com Matthias Hannich ### [email protected] ### KeyID: 407A2F69 Fingerprint: 600D 7E07 105C F674 0C4B 12C2 4DE4 A8DF 407A 2F69
Quoted 9 lines · Matthias Hannich
Hi,
I have a laptop (Gericom Webboy) with a PIII Coppermine. I am working on a 2.4.22-kernel (from kernel.org) and patched the patch-2.4.22-ac4.bz2 from Alan Cox against it.
Because of having a low battery-capacity I tried to enable cpufreq and ACPI-stuff in the kernel (and tried to compile them as modules as well).
...
Quoted 9 lines
root:/usr/src/linux-2.4.22# insmod speedstep-ich Using /lib/modules/2.4.22-ac4-486/kernel/arch/i386/kernel/speedstep-ich.o /lib/modules/2.4.22-ac4-486/kernel/arch/i386/kernel/speedstep-ich.o: unresolved symbol speedstep_get_processor_frequency /lib/modules/2.4.22-ac4-486/kernel/arch/i386/kernel/speedstep-ich.o: unresolved symbol speedstep_detect_processor /lib/modules/2.4.22-ac4-486/kernel/arch/i386/kernel/speedstep-ich.o: unresolved symbol speedstep_get_freqs
What about a
modprobe speedstep-ich
instead of insmod?
PS: If modprobe speedstep-ich still do not work
modpbore speedstep-lib
modprobe speedstep-ich
Signature · 4 lines
Ducrot Bruno -- Which is worse: ignorance or apathy? -- Don't know. Don't care.
Am 10/30/2003 03:09 PM schrieb Ducrot Bruno:
Quoted 3 lines
What about a
modprobe speedstep-ich
instead of insmod?
root:/# modprobe speedstep-ich
/lib/modules/2.4.22-ac4-486/kernel/arch/i386/kernel/speedstep-ich.o: init_module: No such device
Hint: insmod errors can be caused by incorrect module parameters, including invalid IO or IRQ parameters.
You may find more information in syslog or the output from dmesg
(can't find anything there)/lib/modules/2.4.22-ac4-486/kernel/arch/i386/kernel/speedstep-ich.o: insmod /lib/modules/2.4.22-ac4-486/kernel/arch/i386/kernel/speedstep-ich.o failed /lib/modules/2.4.22-ac4-486/kernel/arch/i386/kernel/speedstep-ich.o: insmod speedstep-ich failed
Quoted 2 lines
PS: If modprobe speedstep-ich still do not work
modpbore speedstep-lib
No error messages.
Quoted 1 line
modprobe speedstep-ich
The same as above.
Thanks, Matthias
Signature · 4 lines
Freundliche Grüße, ### http://www.stop1984.com Matthias Hannich ### [email protected] ### KeyID: 407A2F69 Fingerprint: 600D 7E07 105C F674 0C4B 12C2 4DE4 A8DF 407A 2F69
Quoted 19 lines · Matthias Hannich
Am 10/30/2003 03:09 PM schrieb Ducrot Bruno:
Quoted 3 lines
What about a
modprobe speedstep-ich
instead of insmod?root:/# modprobe speedstep-ich /lib/modules/2.4.22-ac4-486/kernel/arch/i386/kernel/speedstep-ich.o: init_module: No such device Hint: insmod errors can be caused by incorrect module parameters, including invalid IO or IRQ parameters. You may find more information in syslog or the output from dmesg (can't find anything there)
/lib/modules/2.4.22-ac4-486/kernel/arch/i386/kernel/speedstep-ich.o:
insmod/lib/modules/2.4.22-ac4-486/kernel/arch/i386/kernel/speedstep-ich.ofailed/lib/modules/2.4.22-ac4-486/kernel/arch/i386/kernel/speedstep-ich.o:
insmod speedstep-ich failed
Quoted 7 lines
Quoted 2 lines
PS: If modprobe speedstep-ich still do not work
modpbore speedstep-libNo error messages.
Quoted 1 line
modprobe speedstep-ich
Are you sure you have an ich based system?
If 'lspci -v' show
00:00.0 Host bridge: Intel Corp. 440BX/ZX/DX - 82443BX/ZX/DX Host bridge (rev 03)
...
00:07.0 ISA bridge: Intel Corp. 82371AB/EB/MB PIIX4 ISA (rev 02)
then speedstep-ich will not work.
at first, take a look at:
http://www.poupinou.org/cpufreq/
To resume:
there is 2 drivers, speedstep-piix4 and speedstep-smi.
The two are in CVS, but not in ac tree.
The speedstep-piix4 is 'old', and may be unstable for some people, or even do not work at all. This one is no more recommended, and may be removed very soon now in CVS.
The speedstep-smi is the newer one. It should be a *lot* more stable.
If you want to know if speedstep-smi does work, get first speedstep-detect at ftp://ftp.poupinou.org/cpufreq/speedstep-detect.tar.gz
then
tar xzvfp speedstep-detect.tar.gz; cd speedstep-detect-0.01 make
sudo ./speedstep-detect
...
BIOS DO NOT support GSIC call. Dumping registers anyway:
eax = 0x00008680
ebx = 0x00000000
ecx = 0x00000000
edx = 0x47534943
...
In that case, (and it is normal, it was run actually under a desktop..), speedstep-smi will not work.
But if you see something like:
...
BIOS support GSIC call:
signature: GSIC command port = 0x00b2 command = 0x0080
...
then there is good chance that speedstep-smi will work.
Note also that the others input was only meaningfull for the (old) speedstep-piix4 actually, and there is no more need to reports things to me, and no more need to boot on AC, then on battery even if speedstep-detect will tell you to do so...
It was writen before the speedstep-smi driver.
Some notes:
* a port to ac tree for the speedstep-smi is here:
ftp://ftp.poupinou.org/cpufreq/speedstep-smi-24-22rc2ac3.diff It should apply OK in 2.4.22-ac4.
* 'modprobe speedstep-smi' do not work, try:
modprobe speedstep-smi smi_cmd=0x82
Cheers,
Signature · 4 lines
Ducrot Bruno -- Which is worse: ignorance or apathy? -- Don't know. Don't care.
Am 10/31/2003 08:14 AM schrieb Ducrot Bruno:
Quoted 10 lines
Quoted 8 lines
Quoted 2 lines
PS: If modprobe speedstep-ich still do not work
modpbore speedstep-libNo error messages.
Quoted 1 line
modprobe speedstep-ich
Are you sure you have an ich based system?
00:00.0 Host bridge: Silicon Integrated Systems [SiS] 630 Host (rev 31) 00:01.0 ISA bridge: Silicon Integrated Systems [SiS] 85C503/5513
Sorry, I did completely forgot it.
Quoted 5 lines
BIOS DO NOT support GSIC call. Dumping registers anyway:
eax =0x00008680
ebx =0x00000000
ecx =0x00000000
edx =0x47534943
Was exactly what I got.
Thanks,
Matthias
Signature · 4 lines
Freundliche Gru"?e, ### http://www.stop1984.com Matthias Hannich ### [email protected] ### KeyID: 407A2F69 Fingerprint: 600D 7E07 105C F674 0C4B 12C2 4DE4 A8DF 407A 2F69
Quoted 7 lines · Matthias Hannich
Am 10/31/2003 08:14 AM schrieb Ducrot Bruno:
Quoted 6 lines
Quoted 6 lines
Quoted 2 lines
PS: If modprobe speedstep-ich still do not work
modpbore speedstep-libNo error messages.
Quoted 8 lines
Quoted 4 lines
Quoted 2 lines
Quoted 1 line
modprobe speedstep-ich
Are you sure you have an ich based system?
00:00.0 Host bridge: Silicon Integrated Systems [SiS] 630 Host (rev 31) 00:01.0 ISA bridge: Silicon Integrated Systems [SiS] 85C503/5513
Due to lack of documentation, and since no body (AFAIC) has even started to reverse engeener, the SiS is not supported.
Signature · 4 lines
Ducrot Bruno -- Which is worse: ignorance or apathy? -- Don't know. Don't care.