"Ronciak, John" <[email protected]> writes: >Hi Dan, > >What are you using to send the packets to the receiving system? Is it >some sort of packet generator? Are the systems connected through a >switch or back-to-back? If it's a switch, what kind? Regarding the >daemon, is this the Sourceforge V2 PTPd or something else? Did you make >modification to the 3.0.22 so that it could be used with the SF PTPd? > >Has this been tried using the latest (2.6.39) kernel where 1588 is >supported by the kernel and the driver that is included in that kernel? >If so what were those results? > >We have not seen a situation using the 82580 HW where packets did not get >a timestamp applied. This would most likely not be a driver issue as the >driver does not touch the timestamp until after the packet has been DMA'd >to the host, long after it has been received and timestamped by the HW. > >Cheers, >John
Hi John, Yes we are using the Sourceforge V2 PTPd but we modified it to use hardware timestamps, but I was not aware of any modification to the 3.0.22 driver so it could be used with the SF PTPd. I would appreciate any information on the modifications needed to be done in order to get that working. The tests were ran both ways, connected to the Cisco IE30008TC (A) switch at a 100mbps connection and then they were also connected back-to-back and both tests yielded the same results of the second packet not receiving a timestamp because they were received merely nanoseconds apart from the transmitter. We were thinking it could be a hardware problem but you said you have not seen a situation where this hardware did not apply a timestamp, so it is possibly all in the 3.0.22 driver modification. No we have no had the chance to try this with the latest 2.6.39 kernel, but that may be a possibility. Thanks, Dan ------------------------------------------------------------------------------ All of the data generated in your IT infrastructure is seriously valuable. Why? It contains a definitive record of application performance, security threats, fraudulent activity, and more. Splunk takes this data and makes sense of it. IT sense. And common sense. http://p.sf.net/sfu/splunk-d2d-c2 _______________________________________________ E1000-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/e1000-devel To learn more about Intel® Ethernet, visit http://communities.intel.com/community/wired
