From: Krishna Kumar <krkumar2@in.ibm.com>
To: davem@davemloft.net
Cc: Jarek Poplawski <jarkao2@gmail.com>,
netdev@vger.kernel.org, herbert@gondor.apana.org.au,
Krishna Kumar <krkumar2@in.ibm.com>,
kaber@trash.net
Subject: [PATCH] Speed-up pfifo_fast lookup using a bitmap
Date: Thu, 13 Aug 2009 12:58:18 +0530 [thread overview]
Message-ID: <20090813072818.7541.77365.sendpatchset@localhost.localdomain> (raw)
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@in.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;
next reply other threads:[~2009-08-13 7:28 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-08-13 7:28 Krishna Kumar [this message]
2009-08-13 10:08 ` [PATCH] Speed-up pfifo_fast lookup using a bitmap Jarek Poplawski
2009-08-13 10:41 ` Krishna Kumar2
2009-08-13 11:27 ` Jarek Poplawski
-- strict thread matches above, loose matches on Subject: below --
2009-08-14 8:19 Krishna Kumar
2009-08-14 11:01 ` Jarek Poplawski
2009-08-14 13:24 Krishna Kumar
2009-08-14 21:36 ` Jarek Poplawski
2009-08-18 2:03 ` David Miller
2009-08-18 16:46 ` Krishna Kumar2
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20090813072818.7541.77365.sendpatchset@localhost.localdomain \
--to=krkumar2@in.ibm.com \
--cc=davem@davemloft.net \
--cc=herbert@gondor.apana.org.au \
--cc=jarkao2@gmail.com \
--cc=kaber@trash.net \
--cc=netdev@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).