This is the mail archive of the ecos-patches@sources.redhat.com mailing list for the eCos project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: AT91 serial driver


On Fri, 24 Oct 2003 00:09:57 +0200
Thomas Koeller <thomas.koeller@baslerweb.com> wrote:

> 
> Hi Laurent,
> 
> I tested your driver with my application and of course it works. 

Thank you for testing.

> However,
> it slows down its operation. I am receiving large amounts of data that
> is sent in records.  My primary interest is in bringing the total time
> required to transfer a given amount of data to a minimum by maximizing
> the overlap between data transfer and processing. Your driver defeats
> this by delaying notification of the application until after an entire
> data record has been received. 

True. My driver lacks of configuration. For your application, I suggest
you to try to reduce the Rx_PDC buffer size from 32 to 8 for example. If
your records are sent with some "inter-record" delay between each of
them, I rather you to reduce the time out value programmed in the
TIMEOUT register from 40 bits to 16 for example.

> Put another way, your driver may reduce overall system load by
> reducing the number of interrupts to process, but it also prevents me
> from making any use of the resources thus freed - it is only the idle
> thread that benefits. If you have to use lower baud rates for some
> reason, it gets even worse.

Is this behaviour specific to your application ? Your thread sleeps on
reading one record at time, or several ??

> 
> However, you do of course have a point - minimizing system load is
> always good, and depending on the nature of the particular application
> this advantage may outweigh the speed penalty. Is there any reason why
> we can't (or shouldn't)
> have two at91 serial drivers in ecos?

I am not against. On the contrary, we could take advantage of the
powerfull configuration framework of eCos to let the user choose if he
prefers responsiveness (yours) or lowest IT rate (mine).
We all agree on that point: the old driver should disappear !

-- 
GONZALEZ Laurent
18 rue George SAND
38100 GRENOBLE


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]