Theyre giving me hard time because Im using the same physical PC. You stand no chance to send something over one physical ethernet interface and receiving it at the same end. Im trying to do a simple loopback test between the two. Either your network is f*cked up or you use techniques like Spanning Tree to explicitely prevent your signal to be ever routed back to you.Īs you can see, the internal data routing is essential. broadcast packages) lead to serious trouble in the network.Ĭase #3: You have a complicated network setup where loops are possible. It does not make sense to do it and would in some cases (e.g. The switch will in its routing automatically never send the signal back to the source. The NIC has no interest in reading its own signal while sending it.Ĭase #2: The NIC sends the data to a switch. Nobody will talk back to the NIC about the signal. Triple-Speed Ethernet Intel® FPGA IP User Guide Archives 12. However I doubt you will ever be able to test the PHY without a second receiver.Ĭase #1: The NIC sends the data to a hub (do these still exist?). * Ping packet (echo-request) generation and parsing (echo-reply) isĪlready present in u-boot.As you may have guessed, this is not a bug but a feature (and an essential one). Standard packet and may be useful elsewhere. Responder return a "ping" response (ICMP echo-reply) this will be a If you sent a "ping" (ICMP echo-request) packet and had your loopback I have not looked at the code, so I don't know how Switched into loopback mode, but i agree that those non-standard packetsĬould be problematic when sent to a network. ![]() Generated packet would remain on the board because PHY or EMAC are I introduced that packet type toĪchieve a proper separtion from all other packet types. The PROT_IP part of the NetReceive function examines the complete IP I don't like such non-standard solutions. I also don't understand why specific ethernet packet types => mii w 0 0 8000 <- PHY reset after test is completedĪn embedded and charset-unspecified text was scrubbed. I've tested this on our PPC405 custom board, but it should also work on all other systems because packet generation is independent of the underlying ethernet hardware. Write to the outgoing socket with sendto, specifying the port and IP address. Bind the incoming socket to a specific port. Create two sockets in UDP mode (SOCKDGRAM) Bind both sockets to the specific interface being tested. It is necessary to set the ethernet line into loopback mode either by programming the PHY or using a loopback plug. This is essentially what the test is currently doing: Bring up the interface and make sure it has a valid IP address. The looped back package is compared with the original one and a message is displayed. The functions EthLoopStart, EthLoopSend and EthLoopHandler generate a simple ethernet frame with NetOurEther as destination address and packet type PROT_TEST. The patch is against SF CVS from 20050307. To achieve this, a new ethernet packet type (PROT_TEST, 0x0808, AFAIK this is not used otherwise) is introduced in net/net.c. ![]() There are 3 amber lights on TenG1/0/1 ,TenG1/0/7 and TenG1/0/15. After the switch finish the test and reboot I run the command : Show dianostic post there are 3 port failed. I run the command : diagnostic start switch 1 test all. ![]() ![]() The switch is stand anlone and is not connect to network. The patch attached provides a command called 'ethloop' which allows to perform loopback testing on the current eth device. We have a WS-C3850-24XU failed on PHY Loopback: loopback Test.
0 Comments
Leave a Reply. |
Details
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |