Hello David, I took a look at the logs. The issue appeared in testing big burst in sanity_test() for burst implementation of distributor.
My guess is that the cyclic buffer for holding packets returned from workers might be overflown and dropping some of the packets. It's designed this way, to hold only 127 latest packets. Older, not collected on time with rte_distributor_returned_pkts(), are lost. Although it's just a guess until I can confirm it with some tests on my machine. However the guess is possible, so the question is about the functionality of collecting the returned packets. We can: 1) keep it as is and fix tests not to send new packets until old ones are returned and collected 2) make changes to the distributor library and change behavior of the returned packages queue - however this requires some discussion as any solution will have its drawbacks (e.g. blocking workers until there is a place to store packets or using more space for the queue) Write, what do you think about it. Meanwhile I'll try to reproduce the issue Best regards Lukasz W dniu 11.01.2021 o 09:14, David Marchand pisze: > Hello guys, > > UNH CI is raising failures on a ARM server for the distributor test: > https://protect2.fireeye.com/v1/url?k=ebf50f8c-b46e3693-ebf484c3-000babdfecba-48457a70d8248142&q=1&e=9f688513-9c17-4e71-a095-953f146ca22a&u=https%3A%2F%2Flab.dpdk.org%2Fresults%2Fdashboard%2Fpatchsets%2F15077%2F > > Worker 59 handled 0 packets > Worker 60 handled 0 packets > Worker 61 handled 0 packets > Worker 62 handled 0 packets > Sanity test with non-zero hashes done > === testing big burst (burst) === > line 258: Missing packets, expected 783 > Test Failed > RTE>> > --- stderr --- > > -- Lukasz Wojciechowski Principal Software Engineer Samsung R&D Institute Poland Samsung Electronics Office +48 22 377 88 25 l.wojciec...@partner.samsung.com
