Re: [Xenomai-help] Big differences in analogy interrupt handling time

[Anders Blomdell <anders.blomdell@domain.hid>]

On 12/16/2011 12:57 PM, Fernando Herrero Carrón wrote:
>   [...]
>
>> Nice that somebody has already tested this.
>>
>
>
>> Nobody that has any insight on what might slow down execution with 20%
>> (curious mind wants to know)?
>>
>
> Dear Anders,
>
> I am afraid I don't understand your question. 20% with respect to what?
> Indeed, the "sync_dma" function performs some copy between kernel/user
> space. The old code performs exactly the same copy ten times, which
> multiplies the time spent in the interrupt handler tenfold, and adds 10us
> in idle wait. Besides that, I believe that the most time spent in an
> interrupt handler is in the i-pipe dispatching mechanism itself.
20% difference in execution time in identical code paths (attachment 'A' 
and 'B'), where executing the same codepath in 'B' takes 20% longer time 
than in 'A'

>
> There are for sure other people in the list more knowledgeable than me on
> this subject.
And me, obviously.


-- 
Anders Blomdell                  Email: anders.blomdell@domain.hid
Department of Automatic Control
Lund University                  Phone:    +46 46 222 4625
P.O. Box 118                     Fax:      +46 46 138118
SE-221 00 Lund, Sweden