net_sched: Add qdisc __NET_XMIT_STOLEN flag
Patrick McHardy <kaber@trash.net> noticed:
"The other problem that affects all qdiscs supporting actions is
TC_ACT_QUEUED/TC_ACT_STOLEN getting mapped to NET_XMIT_SUCCESS
even though the packet is not queued, corrupting upper qdiscs'
qlen counters."
and later explained:
"The reason why it translates it at all seems to be to not increase
the drops counter. Within a single qdisc this could be avoided by
other means easily, upper qdiscs would still increase the counter
when we return anything besides NET_XMIT_SUCCESS though.
This means we need a new NET_XMIT return value to indicate this to
the upper qdiscs. So I'd suggest to introduce NET_XMIT_STOLEN,
return that to upper qdiscs and translate it to NET_XMIT_SUCCESS
in dev_queue_xmit, similar to NET_XMIT_BYPASS."
David Miller <davem@davemloft.net> noticed:
"Maybe these NET_XMIT_* values being passed around should be a set of
bits. They could be composed of base meanings, combined with specific
attributes.
So you could say "NET_XMIT_DROP | __NET_XMIT_NO_DROP_COUNT"
The attributes get masked out by the top-level ->enqueue() caller,
such that the base meanings are the only thing that make their
way up into the stack. If it's only about communication within the
qdisc tree, let's simply code it that way."
This patch is trying to realize these ideas.
Signed-off-by: Jarek Poplawski <jarkao2@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
diff --git a/net/sched/sch_atm.c b/net/sched/sch_atm.c
index 6b517b9..27dd773 100644
--- a/net/sched/sch_atm.c
+++ b/net/sched/sch_atm.c
@@ -415,7 +415,7 @@
case TC_ACT_QUEUED:
case TC_ACT_STOLEN:
kfree_skb(skb);
- return NET_XMIT_SUCCESS;
+ return NET_XMIT_SUCCESS | __NET_XMIT_STOLEN;
case TC_ACT_SHOT:
kfree_skb(skb);
goto drop;
@@ -432,9 +432,11 @@
ret = qdisc_enqueue(skb, flow->q);
if (ret != 0) {
drop: __maybe_unused
- sch->qstats.drops++;
- if (flow)
- flow->qstats.drops++;
+ if (net_xmit_drop_count(ret)) {
+ sch->qstats.drops++;
+ if (flow)
+ flow->qstats.drops++;
+ }
return ret;
}
sch->bstats.bytes += qdisc_pkt_len(skb);
@@ -530,7 +532,7 @@
if (!ret) {
sch->q.qlen++;
sch->qstats.requeues++;
- } else {
+ } else if (net_xmit_drop_count(ret)) {
sch->qstats.drops++;
p->link.qstats.drops++;
}
diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c
index 14954bf..765ae56 100644
--- a/net/sched/sch_cbq.c
+++ b/net/sched/sch_cbq.c
@@ -256,7 +256,7 @@
switch (result) {
case TC_ACT_QUEUED:
case TC_ACT_STOLEN:
- *qerr = NET_XMIT_SUCCESS;
+ *qerr = NET_XMIT_SUCCESS | __NET_XMIT_STOLEN;
case TC_ACT_SHOT:
return NULL;
case TC_ACT_RECLASSIFY:
@@ -397,9 +397,11 @@
return ret;
}
- sch->qstats.drops++;
- cbq_mark_toplevel(q, cl);
- cl->qstats.drops++;
+ if (net_xmit_drop_count(ret)) {
+ sch->qstats.drops++;
+ cbq_mark_toplevel(q, cl);
+ cl->qstats.drops++;
+ }
return ret;
}
@@ -430,8 +432,10 @@
cbq_activate_class(cl);
return 0;
}
- sch->qstats.drops++;
- cl->qstats.drops++;
+ if (net_xmit_drop_count(ret)) {
+ sch->qstats.drops++;
+ cl->qstats.drops++;
+ }
return ret;
}
@@ -664,13 +668,15 @@
q->rx_class = NULL;
if (cl && (cl = cbq_reclassify(skb, cl)) != NULL) {
+ int ret;
cbq_mark_toplevel(q, cl);
q->rx_class = cl;
cl->q->__parent = sch;
- if (qdisc_enqueue(skb, cl->q) == 0) {
+ ret = qdisc_enqueue(skb, cl->q);
+ if (ret == NET_XMIT_SUCCESS) {
sch->q.qlen++;
sch->bstats.packets++;
sch->bstats.bytes += qdisc_pkt_len(skb);
@@ -678,7 +684,8 @@
cbq_activate_class(cl);
return 0;
}
- sch->qstats.drops++;
+ if (net_xmit_drop_count(ret))
+ sch->qstats.drops++;
return 0;
}
diff --git a/net/sched/sch_dsmark.c b/net/sched/sch_dsmark.c
index a935676..7170275 100644
--- a/net/sched/sch_dsmark.c
+++ b/net/sched/sch_dsmark.c
@@ -236,7 +236,7 @@
case TC_ACT_QUEUED:
case TC_ACT_STOLEN:
kfree_skb(skb);
- return NET_XMIT_SUCCESS;
+ return NET_XMIT_SUCCESS | __NET_XMIT_STOLEN;
case TC_ACT_SHOT:
goto drop;
@@ -254,7 +254,8 @@
err = qdisc_enqueue(skb, p->q);
if (err != NET_XMIT_SUCCESS) {
- sch->qstats.drops++;
+ if (net_xmit_drop_count(err))
+ sch->qstats.drops++;
return err;
}
@@ -321,7 +322,8 @@
err = p->q->ops->requeue(skb, p->q);
if (err != NET_XMIT_SUCCESS) {
- sch->qstats.drops++;
+ if (net_xmit_drop_count(err))
+ sch->qstats.drops++;
return err;
}
diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c
index 0ae7d19..5cf9ae7 100644
--- a/net/sched/sch_hfsc.c
+++ b/net/sched/sch_hfsc.c
@@ -1166,7 +1166,7 @@
switch (result) {
case TC_ACT_QUEUED:
case TC_ACT_STOLEN:
- *qerr = NET_XMIT_SUCCESS;
+ *qerr = NET_XMIT_SUCCESS | __NET_XMIT_STOLEN;
case TC_ACT_SHOT:
return NULL;
}
@@ -1586,8 +1586,10 @@
err = qdisc_enqueue(skb, cl->qdisc);
if (unlikely(err != NET_XMIT_SUCCESS)) {
- cl->qstats.drops++;
- sch->qstats.drops++;
+ if (net_xmit_drop_count(err)) {
+ cl->qstats.drops++;
+ sch->qstats.drops++;
+ }
return err;
}
diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c
index 75a4095..538d79b 100644
--- a/net/sched/sch_htb.c
+++ b/net/sched/sch_htb.c
@@ -221,7 +221,7 @@
switch (result) {
case TC_ACT_QUEUED:
case TC_ACT_STOLEN:
- *qerr = NET_XMIT_SUCCESS;
+ *qerr = NET_XMIT_SUCCESS | __NET_XMIT_STOLEN;
case TC_ACT_SHOT:
return NULL;
}
@@ -572,9 +572,11 @@
kfree_skb(skb);
return ret;
#endif
- } else if (qdisc_enqueue(skb, cl->un.leaf.q) != NET_XMIT_SUCCESS) {
- sch->qstats.drops++;
- cl->qstats.drops++;
+ } else if ((ret = qdisc_enqueue(skb, cl->un.leaf.q)) != NET_XMIT_SUCCESS) {
+ if (net_xmit_drop_count(ret)) {
+ sch->qstats.drops++;
+ cl->qstats.drops++;
+ }
return NET_XMIT_DROP;
} else {
cl->bstats.packets +=
@@ -615,10 +617,12 @@
kfree_skb(skb);
return ret;
#endif
- } else if (cl->un.leaf.q->ops->requeue(skb, cl->un.leaf.q) !=
+ } else if ((ret = cl->un.leaf.q->ops->requeue(skb, cl->un.leaf.q)) !=
NET_XMIT_SUCCESS) {
- sch->qstats.drops++;
- cl->qstats.drops++;
+ if (net_xmit_drop_count(ret)) {
+ sch->qstats.drops++;
+ cl->qstats.drops++;
+ }
return NET_XMIT_DROP;
} else
htb_activate(q, cl);
diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c
index a590857..6cd6f2b 100644
--- a/net/sched/sch_netem.c
+++ b/net/sched/sch_netem.c
@@ -240,8 +240,9 @@
sch->q.qlen++;
sch->bstats.bytes += qdisc_pkt_len(skb);
sch->bstats.packets++;
- } else
+ } else if (net_xmit_drop_count(ret)) {
sch->qstats.drops++;
+ }
pr_debug("netem: enqueue ret %d\n", ret);
return ret;
diff --git a/net/sched/sch_prio.c b/net/sched/sch_prio.c
index f849243..adb1a52 100644
--- a/net/sched/sch_prio.c
+++ b/net/sched/sch_prio.c
@@ -45,7 +45,7 @@
switch (err) {
case TC_ACT_STOLEN:
case TC_ACT_QUEUED:
- *qerr = NET_XMIT_SUCCESS;
+ *qerr = NET_XMIT_SUCCESS | __NET_XMIT_STOLEN;
case TC_ACT_SHOT:
return NULL;
}
@@ -88,7 +88,8 @@
sch->q.qlen++;
return NET_XMIT_SUCCESS;
}
- sch->qstats.drops++;
+ if (net_xmit_drop_count(ret))
+ sch->qstats.drops++;
return ret;
}
@@ -114,7 +115,8 @@
sch->qstats.requeues++;
return 0;
}
- sch->qstats.drops++;
+ if (net_xmit_drop_count(ret))
+ sch->qstats.drops++;
return NET_XMIT_DROP;
}
diff --git a/net/sched/sch_red.c b/net/sched/sch_red.c
index 3f2d1d7..5da0583 100644
--- a/net/sched/sch_red.c
+++ b/net/sched/sch_red.c
@@ -97,7 +97,7 @@
sch->bstats.bytes += qdisc_pkt_len(skb);
sch->bstats.packets++;
sch->q.qlen++;
- } else {
+ } else if (net_xmit_drop_count(ret)) {
q->stats.pdrop++;
sch->qstats.drops++;
}
diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c
index 8589da6..3a456e1 100644
--- a/net/sched/sch_sfq.c
+++ b/net/sched/sch_sfq.c
@@ -178,7 +178,7 @@
switch (result) {
case TC_ACT_STOLEN:
case TC_ACT_QUEUED:
- *qerr = NET_XMIT_SUCCESS;
+ *qerr = NET_XMIT_SUCCESS | __NET_XMIT_STOLEN;
case TC_ACT_SHOT:
return 0;
}
diff --git a/net/sched/sch_tbf.c b/net/sched/sch_tbf.c
index b296672..7d3b7ff 100644
--- a/net/sched/sch_tbf.c
+++ b/net/sched/sch_tbf.c
@@ -135,7 +135,8 @@
ret = qdisc_enqueue(skb, q->qdisc);
if (ret != 0) {
- sch->qstats.drops++;
+ if (net_xmit_drop_count(ret))
+ sch->qstats.drops++;
return ret;
}