On 2014-03-16 20:08, Peter Dufault wrote:

On Mar 15, 2014, at 18:01 , Joel Sherrill <joel.sherr...@oarcorp.com
<mailto:joel.sherr...@oarcorp.com>> wrote:

There are three pieces and I may have confused you.

+ current stack in tree
+ some add on drivers with porting kit
+ new dual mode stack

I do not think there is an smc driver in the add on kit. I just thought if
you decided to use a new driver with the old stack, it would be a good 
reference.

The new stack may be an option. It would have the latest drivers but has not
been tuned at all. We know it is larger than the current stack but I do not
know if the performance has been measured.

I am not recommending a solution. Just offering more possibilities.



This is something I'll be using in production in the June timeframe, so moving
to the new stack isn't an option.  I'll look at improving the performance of
the current driver.  I think that on the MPC5554 it just gets overloaded and
stops responding, causing re-transmissions.  I still have at least two 
questions:

I cannot say much to the SMC91111. The new network stack however will not improve the performance in its current state.

On a MPC5674F running at 264MHz we had approx. 5MByte/second upstream and downstream with the SMSC LAN9218i using the eDMA. The limit was the memcpy() and in_cksum() load.

--
Sebastian Huber, embedded brains GmbH

Address : Dornierstr. 4, D-82178 Puchheim, Germany
Phone   : +49 89 189 47 41-16
Fax     : +49 89 189 47 41-09
E-Mail  : sebastian.hu...@embedded-brains.de
PGP     : Public key available on request.

Diese Nachricht ist keine geschäftliche Mitteilung im Sinne des EHUG.
_______________________________________________
rtems-devel mailing list
rtems-devel@rtems.org
http://www.rtems.org/mailman/listinfo/rtems-devel

Reply via email to