Random thought. I was reading a fairly old book (Linux Device Drivers
published by O'Reilly) and it mentions that on Intel machines the jiffies
count will wrap sometime after a year of uptime (4 billion / 100 seconds).
It also mentioned that weird stuff may happen at this time (as you may
expect, if anyone is making poor assumptions about jiffies). Has anyone ever
edited the kernel so that jiffies starts at 0xFFF00000 at boot to see what
happens, and fix any ensuing problems?
My theory has always been, if all you ever do is subtract jiffies counts
from each other, then you're fine, as long as no period you're measuring is
> 2^31 units (highly unlikely). I guess checking that no other operations
are made on jiffies counts than ones that make valid assumptions would be
another way of ensuring stability at the wrap point.
Nicholas Vinen
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Thu Mar 23 2000 - 21:00:17 EST