On Fri, 10 May 2024 04:17:04 +0000 Joe Damato wrote:
> Add functions to support the netdev-genl per queue stats API.
>
> ./cli.py --spec netlink/specs/netdev.yaml \
> --dump qstats-get --json '{"scope": "queue"}'
>
> ...snip
>
> {'ifindex': 7,
> 'queue-id': 62,
> 'queue-type': 'rx',
> 'rx-alloc-fail': 0,
> 'rx-bytes': 105965251,
> 'rx-packets': 179790},
> {'ifindex': 7,
> 'queue-id': 0,
> 'queue-type': 'tx',
> 'tx-bytes': 9402665,
> 'tx-packets': 17551},
>
> ...snip
>
> Also tested with the script tools/testing/selftests/drivers/net/stats.py
> in several scenarios to ensure stats tallying was correct:
>
> - on boot (default queue counts)
> - adjusting queue count up or down (ethtool -L eth0 combined ...)
> - adding mqprio TCs
>
> Signed-off-by: Joe Damato <[email protected]>
Tariq, could you take a look? Is it good enough to make 6.10?
Would be great to have it..
On Mon, May 13, 2024 at 07:58:27AM -0700, Jakub Kicinski wrote:
> On Fri, 10 May 2024 04:17:04 +0000 Joe Damato wrote:
> > Add functions to support the netdev-genl per queue stats API.
> >
> > ./cli.py --spec netlink/specs/netdev.yaml \
> > --dump qstats-get --json '{"scope": "queue"}'
> >
> > ...snip
> >
> > {'ifindex': 7,
> > 'queue-id': 62,
> > 'queue-type': 'rx',
> > 'rx-alloc-fail': 0,
> > 'rx-bytes': 105965251,
> > 'rx-packets': 179790},
> > {'ifindex': 7,
> > 'queue-id': 0,
> > 'queue-type': 'tx',
> > 'tx-bytes': 9402665,
> > 'tx-packets': 17551},
> >
> > ...snip
> >
> > Also tested with the script tools/testing/selftests/drivers/net/stats.py
> > in several scenarios to ensure stats tallying was correct:
> >
> > - on boot (default queue counts)
> > - adjusting queue count up or down (ethtool -L eth0 combined ...)
> > - adding mqprio TCs
> >
> > Signed-off-by: Joe Damato <[email protected]>
>
> Tariq, could you take a look? Is it good enough to make 6.10?
> Would be great to have it..
Thanks Jakub.
FYI: I've also sent a v5 of the mlx4 patches which is only a very minor
change from the v4 as suggested by Tariq (see the changelog in that cover
letter).
I am not trying to "rush" either in, to to speak, but if they both made it
to 6.10 it would be great to have the same support on both drivers in the
same kernel release :)
On Mon, May 13, 2024 at 10:33:28AM -0700, Joe Damato wrote:
> On Mon, May 13, 2024 at 07:58:27AM -0700, Jakub Kicinski wrote:
> > On Fri, 10 May 2024 04:17:04 +0000 Joe Damato wrote:
> > > Add functions to support the netdev-genl per queue stats API.
> > >
> > > ./cli.py --spec netlink/specs/netdev.yaml \
> > > --dump qstats-get --json '{"scope": "queue"}'
> > >
> > > ...snip
> > >
> > > {'ifindex': 7,
> > > 'queue-id': 62,
> > > 'queue-type': 'rx',
> > > 'rx-alloc-fail': 0,
> > > 'rx-bytes': 105965251,
> > > 'rx-packets': 179790},
> > > {'ifindex': 7,
> > > 'queue-id': 0,
> > > 'queue-type': 'tx',
> > > 'tx-bytes': 9402665,
> > > 'tx-packets': 17551},
> > >
> > > ...snip
> > >
> > > Also tested with the script tools/testing/selftests/drivers/net/stats.py
> > > in several scenarios to ensure stats tallying was correct:
> > >
> > > - on boot (default queue counts)
> > > - adjusting queue count up or down (ethtool -L eth0 combined ...)
> > > - adding mqprio TCs
> > >
> > > Signed-off-by: Joe Damato <[email protected]>
> >
> > Tariq, could you take a look? Is it good enough to make 6.10?
> > Would be great to have it..
>
> Thanks Jakub.
>
> FYI: I've also sent a v5 of the mlx4 patches which is only a very minor
> change from the v4 as suggested by Tariq (see the changelog in that cover
> letter).
>
> I am not trying to "rush" either in, to to speak, but if they both made it
> to 6.10 it would be great to have the same support on both drivers in the
> same kernel release :)
Err, sorry, just going through emails now and saw that net-next was closed
just before I sent the v5.
My apologies for missing that announcement.
Do I need to re-send after net-next re-opens or will it automatically be in
the queue for net-next?
On Mon, 13 May 2024 11:46:57 -0700 Joe Damato wrote:
> > FYI: I've also sent a v5 of the mlx4 patches which is only a very minor
> > change from the v4 as suggested by Tariq (see the changelog in that cover
> > letter).
> >
> > I am not trying to "rush" either in, to to speak, but if they both made it
> > to 6.10 it would be great to have the same support on both drivers in the
> > same kernel release :)
>
> Err, sorry, just going through emails now and saw that net-next was closed
> just before I sent the v5.
>
> My apologies for missing that announcement.
>
> Do I need to re-send after net-next re-opens or will it automatically be in
> the queue for net-next?
Right, unless it somehow magically gets into our 6.10 PR - you'll most
likely have to make a fresh posting after the merge window :)