[PATCH] ethdev: document that stats reset APIs are not thread-safe

[Morten Brørup]

> From: Mattias Rönnblom [mailto:hofors at lysator.liu.se]
> Sent: Sunday, 28 April 2024 17.53
> 
> On 2024-04-26 17:17, Morten Brørup wrote:
> >> From: Stephen Hemminger [mailto:stephen at networkplumber.org]
> >> Sent: Friday, 26 April 2024 17.14
> >>
> >> On Fri, 26 Apr 2024 14:20:01 +0200
> >> Morten Brørup <mb at smartsharesystems.com> wrote:
> >>
> >>>> From: Ferruh Yigit [mailto:ferruh.yigit at amd.com]
> >>>> Sent: Thursday, 25 April 2024 18.53
> >>>>
> >>>> Making 'rte_eth_stats_reset()' and 'rte_eth_xstats_reset()' APIs thread
> >>>> safe has performance impact on datapath.
> >>>>
> >>>> Instead document APIs as not thread safe and add condition for reliable
> >>>> stats reset functionality, forwarding should be stopped.
> >>>
> >>> I'm not sure stopping forwarding suffices.
> >>> NIC hardware counters will keep progressing unless RX and TX is stopped at
> >> NIC level.
> >>>
> >>> I don't have any suggestions for a better wording, though. :-(
> >>>
> >>> Anyway, better with the patch than without...
> >>> Acked-by: Morten Brørup <mb at smartsharesystems.com>
> >>>
> >>
> >> The safest option would be:
> >> 	rte_eth_dev_stop
> >> 	rte_eth_stats_reset
> >> 	rte_eth_dev_start
> >
> > Yes, but this will cause packet loss.
> > Network admins should be able to clear the counters without causing packet
> loss.
> >
> 
> For sure, and they would do it via some O&M interface, which would in
> turn talk the control plane, which in turn would talk to the data plane
> (including <rte_ethdev.h>).
> 
> Either the control plane or the O&M layer could keep track of a reset
> offset. Doesn't have to be on the PMD level.

Exactly.

Snapshotting the "reset" offset in the control plane has many advantages.

Let's promote that design pattern, rather than trying to work around different quirks required by different drivers.