Luis Menina wrote:
> David Rees a écrit :
>> On Fri, Jan 9, 2009 at 12:17 PM, Luis Menina <[email protected]> wrote:
>>> But it seems you're right ! After a modprobe -r forcedeth, the number of
>>> wakeups is divided by 3 !
>> http://ww.linuxpowertop.org/known.php#forcedeth
>>
>> It's a known issue. If you read the comments in the driver, you can
>> see details of why:
>>
>> http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=blob;f=drivers/net/forcedeth.c
>>
>> -Dave
>>
> 
> Thanks for the tip ! I've found a patch proposal to add an option to 
> disable that polling, but it seems it wasn't applied.
> http://linux.derkeiler.com/Mailing-Lists/Kernel/2008-09/msg03838.html
> 
> I'm I overlooking something or the only way to improve the situation ATM 
> is to recompile that module ?

some network cards allow you to tune the number of hardware interrupts per 
second
(rate). Obviously tuning this to a low value will ultimately reduce the amount 
of
wakeups per second, at the cost of some network latency. it depends on the 
driver
and hardware if this is implemented unfortunately, but you could be lucky. I
actually don't know if the forcedeth driver supports this setting.

Auke

_______________________________________________
Power mailing list
[email protected]
http://www.bughost.org/mailman/listinfo/power

Reply via email to