Patchwork sched: fix virtual-start-time update in QFQ

login
register
mail settings
Submitter Paolo Valente
Date Sept. 15, 2012, 10:41 a.m.
Message ID <20120915104134.GA29862@paolo-ThinkPad-W520>
Download mbox | patch
Permalink /patch/184076/
State Accepted
Delegated to: David Miller
Headers show

Comments

Paolo Valente - Sept. 15, 2012, 10:41 a.m.
If the old timestamps of a class, say cl, are stale when the class
becomes active, then QFQ may assign to cl a much higher start time
than the maximum value allowed. This may happen when QFQ assigns to
the start time of cl the finish time of a group whose classes are
characterized by a higher value of the ratio
max_class_pkt/weight_of_the_class with respect to that of
cl. Inserting a class with a too high start time into the bucket list
corrupts the data structure and may eventually lead to crashes.
This patch limits the maximum start time assigned to a class.

Signed-off-by: Paolo Valente <paolo.valente@unimore.it>
---
 net/sched/sch_qfq.c |    5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
David Miller - Sept. 19, 2012, 8:25 p.m.
From: Paolo Valente <paolo.valente@unimore.it>
Date: Sat, 15 Sep 2012 12:41:35 +0200

> If the old timestamps of a class, say cl, are stale when the class
> becomes active, then QFQ may assign to cl a much higher start time
> than the maximum value allowed. This may happen when QFQ assigns to
> the start time of cl the finish time of a group whose classes are
> characterized by a higher value of the ratio
> max_class_pkt/weight_of_the_class with respect to that of
> cl. Inserting a class with a too high start time into the bucket list
> corrupts the data structure and may eventually lead to crashes.
> This patch limits the maximum start time assigned to a class.
> 
> Signed-off-by: Paolo Valente <paolo.valente@unimore.it>

Applied and queued up for -stable, thanks.

Please use "pkt_sched" instead of just plain "sched" in the prefixes
of your subject lines for packet scheduler changes so such changes
can be distinguished from process scheduler changes.

I fixed it up for you this time.

Thanks.
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Patch

diff --git a/net/sched/sch_qfq.c b/net/sched/sch_qfq.c
index e4723d3..211a212 100644
--- a/net/sched/sch_qfq.c
+++ b/net/sched/sch_qfq.c
@@ -865,7 +865,10 @@  static void qfq_update_start(struct qfq_sched *q, struct qfq_class *cl)
 		if (mask) {
 			struct qfq_group *next = qfq_ffs(q, mask);
 			if (qfq_gt(roundedF, next->F)) {
-				cl->S = next->F;
+				if (qfq_gt(limit, next->F))
+					cl->S = next->F;
+				else /* preserve timestamp correctness */
+					cl->S = limit;
 				return;
 			}
 		}