I've been worried about this small bug where hwclock waits for timeout in while 
with a message:

select() to /dev/rtc to wait for clock tick timed out. 

The machine is a MacBook, and I see references to this message on
dual-core Intel processors, so it looks like a generic problem.

People seem to be working around by 

1. disabling APIC / ACPI? 

2. adding --directisa to hwclock.

of course, there is another real option

3. making the kernel work.

I've noticed that this message appears when I am using 'rtc.ko' for controlling 
There are apparently three drivers for RTC in Linux kernel, 

1. rtc (the old rtc driver)
2. genrtc (generic RTC emulation)
3. rtc-dev (new RTC class framework)

unloading rtc and loading either of genrtc/rtc-dev allowed hwclock to
work on my system.

The question I have is how to accomplish this.  It looks like hotplug
is loading stuff at boot-time, and it's not quite clear as how to make
it not load rtc.

I have tried not loading rtc with blacklisting, but apparently that
makes my system hang after hotplug does its job. Modules set to load
in modconf seems to get loaded very much later than hotplug.

Anyone suffering the same problem?

[EMAIL PROTECTED],netfort.gr.jp}   Debian Project

with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Reply via email to