* [PATCH net] net_sched: sch_fq: clear q->band_pkt_count[] in fq_reset()
@ 2026-03-04 1:56 Eric Dumazet
2026-03-04 2:09 ` Neal Cardwell
2026-03-05 2:20 ` patchwork-bot+netdevbpf
0 siblings, 2 replies; 4+ messages in thread
From: Eric Dumazet @ 2026-03-04 1:56 UTC (permalink / raw)
To: David S . Miller, Jakub Kicinski, Paolo Abeni
Cc: Simon Horman, Jamal Hadi Salim, Jiri Pirko, netdev, eric.dumazet,
Eric Dumazet, Praveen Kaligineedi, Willem de Bruijn
When/if a NIC resets, queues are deactivated by dev_deactivate_many(),
then reactivated when the reset operation completes.
fq_reset() removes all the skbs from various queues.
If we do not clear q->band_pkt_count[], these counters keep growing
and can eventually reach sch->limit, preventing new packets to be queued.
Many thanks to Praveen for discovering the root cause.
Fixes: 29f834aa326e ("net_sched: sch_fq: add 3 bands and WRR scheduling")
Diagnosed-by: Praveen Kaligineedi <pkaligineedi@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Willem de Bruijn <willemb@google.com>
---
net/sched/sch_fq.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/net/sched/sch_fq.c b/net/sched/sch_fq.c
index 80235e85f8440ee83032f171cf28df6f161473db..05084c9af48e1c7254a695b30293806a8ac78632 100644
--- a/net/sched/sch_fq.c
+++ b/net/sched/sch_fq.c
@@ -827,6 +827,7 @@ static void fq_reset(struct Qdisc *sch)
for (idx = 0; idx < FQ_BANDS; idx++) {
q->band_flows[idx].new_flows.first = NULL;
q->band_flows[idx].old_flows.first = NULL;
+ q->band_pkt_count[idx] = 0;
}
q->delayed = RB_ROOT;
q->flows = 0;
--
2.53.0.473.g4a7958ca14-goog
^ permalink raw reply related [flat|nested] 4+ messages in thread* Re: [PATCH net] net_sched: sch_fq: clear q->band_pkt_count[] in fq_reset()
2026-03-04 1:56 [PATCH net] net_sched: sch_fq: clear q->band_pkt_count[] in fq_reset() Eric Dumazet
@ 2026-03-04 2:09 ` Neal Cardwell
2026-03-04 2:26 ` Willem de Bruijn
2026-03-05 2:20 ` patchwork-bot+netdevbpf
1 sibling, 1 reply; 4+ messages in thread
From: Neal Cardwell @ 2026-03-04 2:09 UTC (permalink / raw)
To: Eric Dumazet
Cc: David S . Miller, Jakub Kicinski, Paolo Abeni, Simon Horman,
Jamal Hadi Salim, Jiri Pirko, netdev, eric.dumazet,
Praveen Kaligineedi, Willem de Bruijn
On Tue, Mar 3, 2026 at 5:56 PM Eric Dumazet <edumazet@google.com> wrote:
>
> When/if a NIC resets, queues are deactivated by dev_deactivate_many(),
> then reactivated when the reset operation completes.
>
> fq_reset() removes all the skbs from various queues.
>
> If we do not clear q->band_pkt_count[], these counters keep growing
> and can eventually reach sch->limit, preventing new packets to be queued.
>
> Many thanks to Praveen for discovering the root cause.
>
> Fixes: 29f834aa326e ("net_sched: sch_fq: add 3 bands and WRR scheduling")
> Diagnosed-by: Praveen Kaligineedi <pkaligineedi@google.com>
> Signed-off-by: Eric Dumazet <edumazet@google.com>
> Cc: Willem de Bruijn <willemb@google.com>
> ---
Reviewed-by: Neal Cardwell <ncardwell@google.com>
Thanks, Eric and Praveen!
neal
^ permalink raw reply [flat|nested] 4+ messages in thread* Re: [PATCH net] net_sched: sch_fq: clear q->band_pkt_count[] in fq_reset()
2026-03-04 2:09 ` Neal Cardwell
@ 2026-03-04 2:26 ` Willem de Bruijn
0 siblings, 0 replies; 4+ messages in thread
From: Willem de Bruijn @ 2026-03-04 2:26 UTC (permalink / raw)
To: Neal Cardwell, Eric Dumazet
Cc: David S . Miller, Jakub Kicinski, Paolo Abeni, Simon Horman,
Jamal Hadi Salim, Jiri Pirko, netdev, eric.dumazet,
Praveen Kaligineedi, Willem de Bruijn
Neal Cardwell wrote:
> On Tue, Mar 3, 2026 at 5:56 PM Eric Dumazet <edumazet@google.com> wrote:
> >
> > When/if a NIC resets, queues are deactivated by dev_deactivate_many(),
> > then reactivated when the reset operation completes.
> >
> > fq_reset() removes all the skbs from various queues.
> >
> > If we do not clear q->band_pkt_count[], these counters keep growing
> > and can eventually reach sch->limit, preventing new packets to be queued.
> >
> > Many thanks to Praveen for discovering the root cause.
> >
> > Fixes: 29f834aa326e ("net_sched: sch_fq: add 3 bands and WRR scheduling")
> > Diagnosed-by: Praveen Kaligineedi <pkaligineedi@google.com>
> > Signed-off-by: Eric Dumazet <edumazet@google.com>
> > Cc: Willem de Bruijn <willemb@google.com>
> > ---
>
> Reviewed-by: Neal Cardwell <ncardwell@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH net] net_sched: sch_fq: clear q->band_pkt_count[] in fq_reset()
2026-03-04 1:56 [PATCH net] net_sched: sch_fq: clear q->band_pkt_count[] in fq_reset() Eric Dumazet
2026-03-04 2:09 ` Neal Cardwell
@ 2026-03-05 2:20 ` patchwork-bot+netdevbpf
1 sibling, 0 replies; 4+ messages in thread
From: patchwork-bot+netdevbpf @ 2026-03-05 2:20 UTC (permalink / raw)
To: Eric Dumazet
Cc: davem, kuba, pabeni, horms, jhs, jiri, netdev, eric.dumazet,
pkaligineedi, willemb
Hello:
This patch was applied to netdev/net.git (main)
by Jakub Kicinski <kuba@kernel.org>:
On Wed, 4 Mar 2026 01:56:40 +0000 you wrote:
> When/if a NIC resets, queues are deactivated by dev_deactivate_many(),
> then reactivated when the reset operation completes.
>
> fq_reset() removes all the skbs from various queues.
>
> If we do not clear q->band_pkt_count[], these counters keep growing
> and can eventually reach sch->limit, preventing new packets to be queued.
>
> [...]
Here is the summary with links:
- [net] net_sched: sch_fq: clear q->band_pkt_count[] in fq_reset()
https://git.kernel.org/netdev/net/c/a4c2b8be2e53
You are awesome, thank you!
--
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2026-03-05 2:20 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-03-04 1:56 [PATCH net] net_sched: sch_fq: clear q->band_pkt_count[] in fq_reset() Eric Dumazet
2026-03-04 2:09 ` Neal Cardwell
2026-03-04 2:26 ` Willem de Bruijn
2026-03-05 2:20 ` patchwork-bot+netdevbpf
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox