On 10/13/2010 11:46 AM, Wolfgang Grandegger wrote:
>>> I'm following the discussion and realized that it will not work for
>>> devices doing the local loopback in hardware, like the flexcan
>>> controller. 
>>
>> What is actually the advantage of using hardware loopback? One can
>> always quit using it and use Tx interrupt for self reception ...

...from my other mail:

I think we (the socketcan people) originally thought it's a better
solution to loop the CAN frames back in hardware (if available), as it
represents what's happening on the wire far better than software
loopback.

> Hey, we care about overhead ;-). CAN can cause high interrupt loads,
> especially at 1 MB/s.

On the flexcan, we use (currently) the hardware loopback. So a TX'ed CAN
frame generates a TX-complete interrupt, which we need for the
netif_wake_queue(). And it also generates a RX-interrupt due to the
hardware loopback.

If the replace hardware loopback by software loopback we can actually
save the RX interrupt.

cheers, Marc

-- 
Pengutronix e.K.                  | Marc Kleine-Budde           |
Industrial Linux Solutions        | Phone: +49-231-2826-924     |
Vertretung West/Dortmund          | Fax:   +49-5121-206917-5555 |
Amtsgericht Hildesheim, HRA 2686  | http://www.pengutronix.de   |

Attachment: signature.asc
Description: OpenPGP digital signature

_______________________________________________
Socketcan-users mailing list
[email protected]
https://lists.berlios.de/mailman/listinfo/socketcan-users

Reply via email to