On Wed, 1 Jul 2009, 배영부 wrote: > I'm still evaluating RSS feature of Intel 82598EB AF 10G ethernet card. > I think this is my last problem. > that is computed hash value of NIC. > Before the evaluation, I wrote a simulation code to simulate ComputeHash > function according to datasheet. > and I checked it by RSS verification suit of datasheet. It's correct. > But when I read rss value from descriptor(of cause I use advanced > descriptor format) of received packet, > It is not the same with result of simulating code. > Of cause, I've used same source and destination IP address and port. But > it's not. > and My simulating code also use the same seed key with device driver code. > I think they should be the same. aren't they?
They should be the same, but you must have set up something incorrectly in your simulation, are you converting the seed key correctly to little endian 32 bit values? > (Can you understand me? I'm afraid of my poor English.) It seems I do. > Why? Is there another ComputeHash function differ from datasheet? I can say that our hardware's RSS algorithm passes microsoft's certification and so is proved correct. I cannot easily say what is wrong with your code, I guess thats up to you to figure out. Good luck with your evaluation, Jesse ------------------------------------------------------------------------------ _______________________________________________ E1000-devel mailing list E1000-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/e1000-devel
