Messages in this thread | | | From | "Ulrich Windl" <> | Date | Wed, 17 Mar 1999 08:53:19 +0100 | Subject | Re: [patch] recover losed timer interrupt using the TSC [Re: [patch] kstat change to see how much Linux SMP really scale well] |
| |
On 17 Mar 99, at 2:13, Andrea Arcangeli wrote:
[...] > >> -long tick = (1000000 + HZ/2) / HZ; /* timer interrupt period */ > >> +long tick = 1000000 / HZ; /* timer interrupt period */ > > > >This way the system time will be more behind than before if > >"(1000000 % HZ) >= HZ/2". IMHO the line is correct. After all we are > >not saying "(1000000 + HZ - 1) / HZ". > > ;). Agreed. I also needed such trick that to make the recover of the lost > ticks completly relialable. I just removed such changes from my second
As a side-note: This is from Johan Swenker as a contribution to an NTP FAQ; have you considered it?
--------- Question Every once in a while my system looses 30 msec. How comes?
Answer You are running Linux; you are using automatic module loading using /sbin/kerneld. During module loading the kernel does not take good care of the time. If you can spare the memory, please load the modules at boot time.
Remark to the editor This once occured to me. The sound system was demand loaded. A small application told the time every 15 minutes (saytime running in cron). The ntp logging showed me that something strange happened every 15 minutes. I found out it was the loading of the sound module every 15 minutes. --------
> patch. But thanks for commenting about that. I wanted to hear if I was > missing something of more serious or it was only an attempt to decrease > the error in integer divisions.
I think you are possibly fighting flies when elephants are loose ;-)
Ulrich
- 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/
| |