On 2025-09-12 11:04:24 [+0200], Kurt Kanzenbach wrote:
…
> I did run the same test as you mentioned here. But, my numbers are
> completely different. Especially the number of hardware TX timestamps
> are significantly lower overall.
…

Using the command line, I see hardly any difference over 5 runs. One
thing that made me curious:

| NTP packets received       : 1061901
| NTP daemon TX timestamps   : 565892
| NTP kernel TX timestamps   : 327905
| NTP hardware TX timestamps : 168104
| tx_hwstamp:395778

tx_hwstamp is a counter in igb_ptp_tx_tstamp_event() keeping track how
many packets it processed. So it processed ~395k packets but "NTP
hardware TX" says 168k. Reading the timestamp directly or via the
worker, it looks mostly like noise. I see on ntpperf side ~ 45% - 55%
loss.

If I do
| ntpperf -i X … -I -r 1000 -t 2

then there is no loss and on other side I see

| NTP packets received       : 2201
| NTP timestamps held        : 2101
| NTP daemon TX timestamps   : 200
| NTP kernel TX timestamps   : 901
| NTP hardware TX timestamps : 1100
| tx_hwstamp:2101

Here the tx_hwstamp counter colorates with "NTP timestamps held". Does
it this make any sense? I don't see this matching with the "larger" runs
where ntpperf reports loss.

> Thanks,
> Kurt

Sebastian

Reply via email to