Anders Johansson wrote:
On Wednesday 02 February 2005 15:14, Mario Ohnewald wrote:
Hello List,
i have a K7S8X ASROCK motherboard, 256MB ram and an AMD Duron 1.6Ghz. When i try to install suse9.2 from a FTP-DVD, i get this error on tty4 and it freezes: kernel: Losing some ticks...checking if CPU frequency changed.
I have already tried apci=off and SafeSettings. Then i loaded the default Bios settings, and took out my PCI modem, disabled onboard Audio and Lan. Memtest ran 12h and its ok. Has anyone ideas how i could solve this?! FTP or NFS install freezes with the same error.
It doesn't have to be dangerous. In the kernel source we find
/* sanity check to ensure we're not always losing ticks */ if (lost_count++ > 100) { printk(KERN_WARNING "Losing too many ticks!\n"); printk(KERN_WARNING "TSC cannot be used as a timesource. \n"); printk(KERN_WARNING "Possible reasons for this are: \n"); printk(KERN_WARNING " You're running with Speedstep, \n" ); printk(KERN_WARNING " You don't have DMA enabled for yo ur hard disk (see hdparm),\n"); printk(KERN_WARNING " Incorrect TSC synchronization on an SMP system (see dmesg).\n"); printk(KERN_WARNING "Falling back to a sane timesource n ow.\n");
clock_fallback(); } /* ... but give the TSC a fair chance */ if (lost_count > 25) cpufreq_delayed_get();
and cpufreq_delayed_get is what prints the message you see. It could be from many possible sources, not all of them involving fried hardware
I see this message on x86_64 laptop with 9.2 x86_64 installed, but the time is rock solid on 2.6.10-ac8, but on AMD XP2800+/XP3000+ boxes with 2.6.10, I'm seeing quite bad time drift without the message. Playing around with values of tickadj can minimise the drift, but it's very hit and miss. Regards Sid. -- Sid Boyce .... Hamradio G3VBV and Keen Flyer =====ALMOST ALL LINUX USED HERE, Solaris 10 SPARC is just for play=====