Garrett D'Amore wrote: > Steven Stallion wrote: >> Results attached. >> >> I should also mention that the results for the NEW revision are 'worst >> case'. I have had several runs with NICDRV that reports PASS in all >> categories that were not UNSUPPORTED. >> >> Cheers, >> >> Steve > Actually, from my own experience, NICDRV is rather fragile. > > Did you use a single back-to-back cable for the testing? What was the > test peer?
I did. The test peer is a Solaris 10u5 install. The device is a bge. > If you can collect results for individual tests where you have PASS > reported (just re-run the failing tests to get a PASS result), then I > can submit the RTI. I've been working on getting these tests to PASS again, but for some reason after the aggr tests fire off, the peer link drops - this was not the case prior to somewhat recent ON update on the test machine. Unfortunately the only other peer I can use at this point is an eri on a Netra T4. > I agree that it looks like this is an improvement in any case. > > Test 13 failures look like a bad assumption in the NICDRV test suite.... > clearly your driver can't support a 9000 byte mtu (the hardware can't do > it!) Works for me. I'm focusing on getting test08 thru test10 to PASS for the RTI. My only worry at this point is dealing with latent bugs in the dnet source. Any ideas? Steve _______________________________________________ driver-discuss mailing list driver-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/driver-discuss
