j.p.schulz wrote:
Hy

I don't understand a part in the documentation TDMA.spec .
How calculates the slave the offset?
The calculations in Time Arithmetics (TDMA.spec) needs either T_recv or
t_trans. The slave gets only T_sched and T_xmit by the synchronisation
Frame.

T_recv = T_xmit + t_trans

t_trans is evaluated during the calibration phase. It's the average time a (minimum-sized) packet needs to travel from the master to the slave (NIC hardware delays + transmission time).


I write a benchmark tool for Fusion - how can I measure (near the physical
Layer) t_xmit in Userspace?

The timestamping mechanism for outgoing packets is so far only available for RTnet protocol modules, it is not addressable through the socket API, thus in userspace. One might consider to export this feature also via socket API (e.g. through the msg_control field in the msghdr structure), but this is not yet on the roadmap.

What exactly are to planning to measure? Maybe there are other ways to retrieve the required data.


Another question:
How can i implement a gateway?
The readme says that other ethernet drivers have to be shutdown - so how can
I run the second one - has it to be another driver/card type


You can reload the normal linux driver after the RTnet driver has been initialised. Furthermore, you can use the "cards=" parameter to select which NICs shall be occupied by a real-time driver. Of course, this only applies to systems where multiple NICs of the same type are used. If different drivers serve rt and non-rt NICs, things become much easier.

Jan


-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_id=7477&alloc_id=16492&op=click
_______________________________________________
RTnet-users mailing list
RTnet-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rtnet-users

Reply via email to