After cleanly building hardened 2.6.25, /proc/cpuinfo is reporting an incorrect value for the CPU speed when using the ondemand governor. Reproducible: Always Steps to Reproduce: 1. Install hardened sources 2.6.25 2. Build kernel (will attach .config) 3. cat /proc/cpuinfio Actual Results: $ cat /proc/cpuinfo processor : 0 vendor_id : GenuineIntel cpu family : 15 model : 1 model name : Intel(R) Pentium(R) 4 CPU 1.70GHz stepping : 2 cpu MHz : 3450.000 cache size : 256 KB 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 mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pebs bts bogomips : 3437.14 clflush size : 64 Expected Results: Should be something like 425Mhz when there's no load on the box. e.g. cpu MHz : 425.000 Cheers.
More info: $ cat /sys/devices/system/cpu/cpu0/cpufreq/scaling_cur_freq 3450000 $ cat /sys/devices/system/cpu/cpu0/cpufreq/scaling_min_freq 3450000 $ cat /sys/devices/system/cpu/cpu0/cpufreq/scaling_max_freq 13800000 $ cat /sys/devices/system/cpu/cpu0/cpufreq/scaling_governor ondemand
Created attachment 158171 [details] Config for 2.6.25 hardened Attached config.
Please test with vanilla and confirm if problem still exists. If it does not then please reopen bug and leave assigned to hardened. If it does then reopen and assign to kernel@ I can't see this being limited to hardened or a user error in any such way as any patch we add should have 0 to do with such areas of code in the kernel.
I have another box using ~x86 on an AMD, it is running gentoo-sources 2.6.25-r5 with the ondemand governor and it reports correctly: $ uname -a Linux starbuck 2.6.25-gentoo-r5 #1 Mon Jun 23 18:10:51 EST 2008 i686 AMD Athlon(tm) 64 Processor 3700+ AuthenticAMD GNU/Linux $ cat /proc/cpuinfo processor : 0 vendor_id : AuthenticAMD cpu family : 15 model : 39 model name : AMD Athlon(tm) 64 Processor 3700+ stepping : 1 cpu MHz : 1000.000 cache size : 1024 KB fdiv_bug : no hlt_bug : no f00f_bug : no coma_bug : no fpu : yes fpu_exception : yes cpuid level : 1 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 syscall nx mmxext fxsr_opt lm 3dnowext 3dnow pni lahf_lm ts fid vid ttp tm stc bogomips : 2059.20 clflush size : 64 Will build vanilla on the Intel box if that is your suggestion.
Latest vanilla for ~x86 is: vanilla-sources-2.6.26_rc7 Is this what you want me to build?
Ignore last comment, building a stock 2.6.25 from the tarball in my distfiles directory.
Confirmed, vanilla 2.6.25 shows same dodgy info: $ uname -a Linux nexus 2.6.25 #1 Tue Jun 24 13:35:56 EST 2008 i686 Intel(R) Pentium(R) 4 CPU 1.70GHz GenuineIntel GNU/Linux [pezz@nexus]-[13:41:53]-[~] $ cat /proc/cpuinfo processor : 0 vendor_id : GenuineIntel cpu family : 15 model : 1 model name : Intel(R) Pentium(R) 4 CPU 1.70GHz stepping : 2 cpu MHz : 3450.000 cache size : 256 KB 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 clflush dts acpi mmx fxsr sse sse2 ss ht tm pebs bts bogomips : 3436.77 clflush size : 64
Ok so confirmed.. This has is not related to hardened and thus this bug is mis-assigned. Reassign to kernel.
Looks like the p4-clockmod driver wasn't helping me power consumption like I thought it was: http://www.gossamer-threads.com/lists/linux/kernel/927570 I think I should just forget this and get a newer CPU... :) Feel free to close unless you feel the issue is worth addressing. Cheers.
Heh, interesting. If you're interested in following up, you should file a bug at http://bugzilla.kernel.org, and feel free to let us know if patches become available.
