Creating a new bug to track this, as the old one was initially a different problem and has already been closed as a dupe. +++ This bug was initially created as a clone of Bug #215813 +++ Description of problem: the module acpi-cpufreq.ko is not present and to the cpuspeed it is need . I was wayting for new kernel and now with the 2.6.18-1.2849.fc6 kernel it is the same. For Notbook it is a big problem because it make it very hot Version-Release number of selected component (if applicable): 2.6.18-1.2798.fc6 2.6.18-1.2849.fc6 How reproducible: evry time Steps to Reproduce: #/etc/init.d/cpuspeed FATAL: Error inserting acpi_cpufreq (/lib/modules/2.6.18-1.2849.fc6/kernel/arch/i386/kernel/cpu/cpufreq/acpi-cpufreq.ko): No such device Actual results: Expected results: Additional info: Was Find in FC5 -- Additional comment from delaigue.cyril on 2006-11-16 10:39 EST -- It was a bad version of the kernel. Change the version of the kernel (i586 to i686) kernel and it work find. i have done : yum install yum-utils yumdownloader kernel-2.6.18-1.2849.fc6.i686 rpm -Uvh --replacefiles --replacepkgs kernel-2.6.18-1.2849.fc6.i686.rpm *** This bug has been marked as a duplicate of 211941 *** -- Additional comment from mark on 2006-12-05 15:54 EST -- I do not think this is related to the kernel version problem with anaconda. I have fixed my kernel from i586 to i686 and still am seeing this problem. [root@pacific ~]# /etc/init.d/cpuspeed start FATAL: Error inserting acpi_cpufreq (/lib/modules/2.6.18-1.2849.fc6/kernel/arch/i386/kernel/cpu/cpufreq/acpi-cpufreq.ko): No such device [root@pacific ~]# uname -a Linux pacific.mjfrazer.org 2.6.18-1.2849.fc6 #1 SMP Fri Nov 10 12:45:28 EST 2006 i686 athlon i386 GNU/Linux [root@pacific ~]# find /lib/modules -name acpi-cpufreq.ko /lib/modules/2.6.18-1.2849.fc6/kernel/arch/i386/kernel/cpu/cpufreq/acpi-cpufreq.ko /lib/modules/2.6.18-1.2798.fc6/kernel/arch/i386/kernel/cpu/cpufreq/acpi-cpufreq.ko [root@pacific ~]# [root@pacific ~]# rpm -q --qf '%{name} %{version} %{release} %{arch}\n' kernel kernel 2.6.18 1.2798.fc6 i586 kernel 2.6.18 1.2849.fc6 i686 -- Additional comment from jwilson on 2006-12-05 16:23 EST -- Please try updating cpuspeed to the version in updates-testing, this should stop acpi-cpufreq from trying to load on systems where it isn't properly supported. -- Additional comment from mark on 2006-12-05 16:27 EST -- Indeed it does. Are athlon processors no longer supported for cpu frequency scalnig? -- Additional comment from jwilson on 2006-12-05 16:46 EST -- Athlon procs aren't supported by acpi-cpufreq, but they are supported by powernow-k7. Edit /etc/cpuspeed.conf and set DRIVER accordingly (I believe all you should have to do is uncomment a line already in there), then restart cpuspeed and I think you should be good to go. (most?) k7 systems don't properly populate sysfs with cpufreq info via /sys/devices/system/cpu/cpu*/cpufreq/scaling_driver). I suppose we could fall back to looking elsewhere (like /proc/cpuinfo) and try loading powernow-k7, but I dunno if there are potential pitfalls w/doing something like that... -- Additional comment from jwilson on 2006-12-05 16:57 EST -- Eep. I don't know what I'm thinking... Don't set DRIVER, because powernow-k7 is built into the kernel, its not a module... Okay, so I'm not sure what the correct solution is at the moment. I'll get back to ya on this one, poking at an athlon box now...
Okay, I've talked this over a bit internally... Basically, AMD just didn't get it wrt cpu frequency scaling until the AMD64 line. Only a few Athlon K7 systems, primarily laptops, actually have functional support for cpu frequency scaling. The chip has to properly support it (most mobile athlons do, most desktop athon procs do not), but the motherboard bios must also support it, so putting a mobile chip in a desktop won't necessarily help. If your system does properly support cpu frequency scaling, your dmesg output upon boot should have a message prefixed "powernow-k7: " that gives some indication of this, and /sys/devices/system/cpu/cpu*/cpufreq/scaling_driver *should* be present... If you've got one or the other, there may be something we need to fix, but otherwise, this is a hardware problem that we can't fix. If you could attach dmesg output from just after boot and show the output of 'ls -l /sys/devices/system/cpu/cpu*/cpufreq/', it should tell us whether or not there's anything that can be done here.
[root@pacific ~]# dmesg | grep -i k7 [root@pacific ~]# grep -i k7 /var/log/messages [root@pacific ~]# ls /sys/devices/system/cpu/cpu? cache crash_notes topology [root@pacific ~]# So, not supported. And the new cpuspeed package from updates-testing gets rid of the nag message.
Okay, closing bug as CANTFIX accordingly, since this is a hardware limitation.