Date:	Thu, 13 Aug 2009 12:58:18 +0530
From:	Krishna Kumar <krkumar2@...ibm.com>
To:	davem@...emloft.net
Cc:	Jarek Poplawski <jarkao2@...il.com>, netdev@...r.kernel.org,
	herbert@...dor.apana.org.au, Krishna Kumar <krkumar2@...ibm.com>,
	kaber@...sh.net
Subject: [PATCH] Speed-up pfifo_fast lookup using a bitmap

Maintain a per-qdisc bitmap indicating availability of skbs for
each band. This helps in faster lookup for a skb when there are
no high priority skbs. Also, it helps in (rare) cases where there
are no skbs on the list where an immediate lookup helps rather
than iterating through the three bands.

Another option I considered was to create a private qdisc pointer
and avoid touching Qdisc structure:
	struct pfifo_fast_priv {
		unsigned long bitmap;
		struct sk_buff_head q[PFIFO_FAST_BANDS];
	};
but the test numbers came a little less, since it takes a few more
memory references on enqueue/dequeue.

By keeping the bitmap in Qdisc, it is possible to implement the
lookup for other schedulers, maybe sch_prio which goes through 16
bands?

The BW numbers are average across 5 iterations for multiple
netperf sessions (1-12 on x86_64, and 1-32 on P6) tested with
Chelsio 10 gbps cards over a 2 hour run:

-------------------------------------------------------------------------
     |         x86_64 (Mb/s)          |               P6 (Mb/s)
--------------------------------------|----------------------------------
Size |  ORG BW          NEW BW        |    ORG BW          NEW BW       
-----|--------------------------------|----------------------------------
16K  |  157700          158237        |    153876          156696
64K  |  155916          157882        |    154176          155987
128K |  155122          155628        |    154983          155904
256K |  154808          158913        |    153898          155164
-------------------------------------------------------------------------

Thanks,

- KK

Signed-off-by: Krishna Kumar <krkumar2@...ibm.com>
---

 include/net/sch_generic.h |    1 
 net/sched/sch_generic.c   |   46 +++++++++++++++++++++++-------------
 2 files changed, 31 insertions(+), 16 deletions(-)

diff -ruNp org/include/net/sch_generic.h new/include/net/sch_generic.h
--- org/include/net/sch_generic.h	2009-08-07 12:05:43.000000000 +0530
+++ new/include/net/sch_generic.h	2009-08-07 19:35:16.000000000 +0530
@@ -72,6 +72,7 @@ struct Qdisc
 	 * For performance sake on SMP, we put highly modified fields at the end
 	 */
 	unsigned long		state;
+	unsigned long		bitmap;
 	struct sk_buff_head	q;
 	struct gnet_stats_basic bstats;
 	struct gnet_stats_queue	qstats;
diff -ruNp org/net/sched/sch_generic.c new/net/sched/sch_generic.c
--- org/net/sched/sch_generic.c	2009-08-07 12:05:43.000000000 +0530
+++ new/net/sched/sch_generic.c	2009-08-13 11:57:54.000000000 +0530
@@ -406,18 +406,29 @@ static const u8 prio2band[TC_PRIO_MAX+1]
 
 #define PFIFO_FAST_BANDS 3
 
-static inline struct sk_buff_head *prio2list(struct sk_buff *skb,
-					     struct Qdisc *qdisc)
+/*
+ * Convert a bitmap to the first band number where an skb is queue'd, where:
+ * 	bitmap=0 means there are no skbs for any bands
+ * 	bitmap=1 means there is a skb on band 0
+ *	bitmap=7 means there are skbs on all 3 bands, etc.
+ */
+static const int bitmap2band[] =
+	{-1, 0, 1, 0, 2, 0, 1, 0};
+
+static inline struct sk_buff_head *band2list(struct Qdisc *qdisc, int band)
 {
 	struct sk_buff_head *list = qdisc_priv(qdisc);
-	return list + prio2band[skb->priority & TC_PRIO_MAX];
+
+	return list + band;
 }
 
 static int pfifo_fast_enqueue(struct sk_buff *skb, struct Qdisc* qdisc)
 {
-	struct sk_buff_head *list = prio2list(skb, qdisc);
+	int band = prio2band[skb->priority & TC_PRIO_MAX];
+	struct sk_buff_head *list = band2list(qdisc, band);
 
 	if (skb_queue_len(list) < qdisc_dev(qdisc)->tx_queue_len) {
+		qdisc->bitmap |= (1 << band);
 		qdisc->q.qlen++;
 		return __qdisc_enqueue_tail(skb, qdisc, list);
 	}
@@ -427,14 +438,17 @@ static int pfifo_fast_enqueue(struct sk_
 
 static struct sk_buff *pfifo_fast_dequeue(struct Qdisc* qdisc)
 {
-	int prio;
-	struct sk_buff_head *list = qdisc_priv(qdisc);
+	int band = bitmap2band[qdisc->bitmap];
 
-	for (prio = 0; prio < PFIFO_FAST_BANDS; prio++) {
-		if (!skb_queue_empty(list + prio)) {
-			qdisc->q.qlen--;
-			return __qdisc_dequeue_head(qdisc, list + prio);
-		}
+	if (likely(band >= 0)) {
+		struct sk_buff_head *list = qdisc_priv(qdisc);
+		struct sk_buff *skb = __qdisc_dequeue_head(qdisc, list + band);
+
+		qdisc->q.qlen--;
+		if (skb_queue_empty(list + band))
+			qdisc->bitmap &= ~(1 << band);
+
+		return skb;
 	}
 
 	return NULL;
@@ -442,12 +456,12 @@ static struct sk_buff *pfifo_fast_dequeu
 
 static struct sk_buff *pfifo_fast_peek(struct Qdisc* qdisc)
 {
-	int prio;
-	struct sk_buff_head *list = qdisc_priv(qdisc);
+	int band = bitmap2band[qdisc->bitmap];
+
+	if (band >= 0) {
+		struct sk_buff_head *list = qdisc_priv(qdisc);
 
-	for (prio = 0; prio < PFIFO_FAST_BANDS; prio++) {
-		if (!skb_queue_empty(list + prio))
-			return skb_peek(list + prio);
+		return skb_peek(list + band);
 	}
 
 	return NULL;
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Powered by blists - more mailing lists