Re: Help with the high res timers

[Nishanth Aravamudan <nacc@us.ibm.com>]

On 04.05.2005 [11:44:56 -0600], Chris Friesen wrote:
> George Anzinger wrote:
> 
> >The, I think, elegant solution to the timer storm problem is to 
> >not restart the timer until the user picks up the prior expiration.  
> >This dynamically adjusts the timer response to the amount of machine 
> >available at the time.
> 
> The disadvantage is that you then lose accuracy since each timer 
> interval is increased by some random amount based on system scheduling. 
>  What about some kind of ulimit-type thing to specify the minimum 
> recurring interval that can be specified?  If root so specifies, you 
> could have 1usec interval timers and the system would hang.  This is 
> conceptually no different than busy-looping in a SCHED_FIFO task.

If I understand your point correctly, I think this is achieved by
TIMERINTERVAL_BITS in my patch (not to claim my patch is function, but
conceptually). No matter what you actually request, the best you can do
is 2^TIMERINTERVAL_BITS nanoseconds, and usually worse because the
tick-rate and timerinterval length do not necessarily line up.

Thanks,
Nish
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/