On 9/23/20 1:07 AM, Maxime Coquelin wrote:
Hi David, Could you please post a v2 with Chenbo's comments taken into account?
Sorry, been out of the office for a bit but now working through my backlog. I did make the change but discovered there were additional dependencies between stats and xstats that weren't addressed in the v1 patch, specifically per queue packet/byte counters:
$ diff before.txt after.txt: < tx_good_packets: 24769339 --- > tx_good_packets: 0 6c6 < tx_good_bytes: 1501282012967 --- > tx_good_bytes: 0 14,15c14,15 < tx_q0packets: 24769339 < tx_q0bytes: 1501282012967 --- > tx_q0packets: 0 > tx_q0bytes: 0 51c51 < tx_good_packets: 2653125 --- > tx_good_packets: 0 53c53 < tx_good_bytes: 167404266 --- > tx_good_bytes: 0 61,62c61,62 < tx_q0packets: 2653125 < tx_q0bytes: 167404266 --- > tx_q0packets: 0 > tx_q0bytes: 0 I'm looking at how to address these as well for a v2 patch soon. Dave
