Re: [PATCH net-next 2/2] net/sched: fq_codel: local packets no longer count against memory limit

[Paolo Abeni <pabeni@redhat.com>]

On 5/12/26 11:48 AM, Eric Dumazet wrote:
> Commit 95b58430abe7 ("fq_codel: add memory limitation per queue")
> claimed that the 32Mb default was "reasonable even for heavy duty usages."
> 
> In practice, this is not the case.
> 
> Packets that are associated with local sockets sk_wmem_alloc
> do not really need additional memory control.
> 
> Signed-off-by: Eric Dumazet <edumazet@google.com>
> ---
>  net/sched/sch_fq_codel.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/net/sched/sch_fq_codel.c b/net/sched/sch_fq_codel.c
> index 1b1de693d4c64a1f5f4e9e788371829dea91740e..71107dc52be799a14f370f2ad74d2eadd93992c1 100644
> --- a/net/sched/sch_fq_codel.c
> +++ b/net/sched/sch_fq_codel.c
> @@ -212,7 +212,7 @@ static int fq_codel_enqueue(struct sk_buff *skb, struct Qdisc *sch,
>  		q->new_flow_count++;
>  		WRITE_ONCE(flow->deficit, q->quantum);
>  	}
> -	get_codel_cb(skb)->mem_usage = skb->truesize;
> +	get_codel_cb(skb)->mem_usage = is_skb_wmem(skb) ? 0 : skb->truesize;
>  	q->memory_usage += get_codel_cb(skb)->mem_usage;
>  	memory_limited = q->memory_usage > q->memory_limit;
>  	qdisc_qlen_inc(sch);

I *think* this patch is causing TDC self-tests failures:

# not ok 704 a4bb - Test FQ_CODEL with HTB parent - force packet drop with empty queue
# Could not match regex pattern. Verify command output:
# qdisc fq_codel 10: parent 1:10 limit 10240p flows 1 quantum 1514 target 99us interval 999us memory_limit 1b ecn drop_batch 64 
#  Sent 490 bytes 5 pkt (dropped 0, overlimits 0 requeues 0) 
#  backlog 0b 0p requeues 0
#   maxpacket 98 drop_overlimit 0 new_flow_count 1 ecn_mark 0
#   new_flows_len 1 old_flows_len 0

whole log at:

https://github.com/p4tc-dev/tc-executor/tree/storage/artifacts/645644/1-tdc-sh

Looking at the test code, I think the patch broke it, but the failures
started a little afterward the patch merge, so I'm possibly
misunderstanding something here.

Thanks,

Paolo