On 3/16/25 4:27 PM, chia-yu.chang@nokia-bell-labs.com wrote:
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 0dbfe069a6e3..987686e91919 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -30,6 +30,7 @@ #include <asm/byteorder.h> #include <asm/local.h> +#include <linux/netdev_features.h> #include <linux/percpu.h> #include <linux/rculist.h> #include <linux/workqueue.h>
I guess this chunck is a left-over... in any case it should be dropped.
[...]
+struct dualpi2_sched_data {
- struct Qdisc *l_queue; /* The L4S Low latency queue (L-queue) */
- struct Qdisc *sch; /* The Classic queue (C-queue) */
- /* Registered tc filters */
- struct {
struct tcf_proto __rcu *filters;struct tcf_block *block;- } tcf;
This usage of anonimous struct is quite unconventional, the preferred way of scoping fields is to add apprepriate prefix to the field name.
[...]
+static bool must_drop(struct Qdisc *sch, struct dualpi2_sched_data *q,
struct sk_buff *skb)+{
- u64 local_l_prob;
- u32 prob;
- bool overload;
Please respect the reverse christmas tree order in variable declaration (many other occurrences below).
[...]
+static void drop_and_retry(struct dualpi2_sched_data *q, struct sk_buff *skb,
struct Qdisc *sch)+{
- ++q->deferred_drops.cnt;
- q->deferred_drops.len += qdisc_pkt_len(skb);
- consume_skb(skb);
Possibly you want to use kfree_skb_reason() here.
I think this patch is still too huge to get a reasonable review. You should try to split it further, e.g. struct definition and parsing in pX, dump in pX+1, enqueue related ops in p+2, ...
Thanks,
Paolo