My system clock runs a little bit too slow. I have no idea, what the problem cause might be. The problem did not occur with my old kernel 2.6.10. I do not know, what exact version introduced this bug. Linux stefanspc 2.6.18-4-amd64 #1 SMP Thu May 10 01:01:58 UTC 2007 x86_64 GNU/Linux stefanspc:~# cat /proc/cpuinfo processor : 0 vendor_id : AuthenticAMD cpu family : 15 model : 4 model name : AMD Athlon(tm) 64 Processor 3000+ stepping : 8 cpu MHz : 800.000 cache size : 1024 KB 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 lm 3dnowext 3dnow up bogomips : 1597.06 TLB size : 1024 4K pages clflush size : 64 cache_alignment : 64 address sizes : 40 bits physical, 48 bits virtual power management: ts fid vid ttp stefanspc:~# dmesg | grep tim time.c: Using 3.579545 MHz WALL PM GTOD PIT/TSC timer. time.c: Detected 1794.962 MHz processor. Calibrating delay using timer specific routine.. 3593.39 BogoMIPS (lpj=7186780) Using local APIC timer interrupts. Detected 12.465 MHz APIC timer. PCI: Setting latency timer of device 0000:00:01.0 to 64 PCI: Setting latency timer of device 0000:00:0b.0 to 64 PCI: Setting latency timer of device 0000:00:0b.1 to 64 PCI: Setting latency timer of device 0000:00:11.6 to 64 PCI: Setting latency timer of device 0000:00:11.5 to 64 Your time source seems to be instable or some driver is hogging interupts stefanspc:~#
RedHat recommends the boot option notsc on system with PowerNow Capable Processors (like my AMD). The warning message about "instable timer source" does not occure anymore. I need to monitor this a while to see if the system clock speed is correct now.
2.6.18 is very old, and it looks also like this is a distro kernel, not a mainline kernel, in which case you need to file it with the distro.
I assume, the kernel version does not matter. Anyway, I opened already a bug report for my debian distribution. Redhat writes that the tsc timer is broken on PowerNow capableCPU's, see http://kbase.redhat.com/faq/FAQ_85_6993.shtm
using the "notsc" boot option (as RedHat recommends), does not help. I also tried to the "noapic" option but this causes a dramatic slowdown during boot until freezes.
The problem does not occur anymore on my machine with a very recent 2.26 kernel.