From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eric Dumazet Subject: [PATCH net-next-2.6] sfq: fix sfq stats handling Date: Wed, 22 Dec 2010 07:14:59 +0100 Message-ID: <1292998499.4317.13.camel@edumazet-laptop> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit Cc: netdev , Jarek Poplawski To: David Miller Return-path: Received: from mail-wy0-f174.google.com ([74.125.82.174]:55737 "EHLO mail-wy0-f174.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751221Ab0LVGPF (ORCPT ); Wed, 22 Dec 2010 01:15:05 -0500 Received: by wyb28 with SMTP id 28so4699774wyb.19 for ; Tue, 21 Dec 2010 22:15:03 -0800 (PST) Sender: netdev-owner@vger.kernel.org List-ID: David, since this one fixes a bug, could you apply it before the pending SFQ patch (sch_sfq: allow big packets and be fair), I'll respin this one if necessary. Thanks [PATCH net-next-2.6] sfq: fix sfq class stats handling sfq_walk() runs without qdisc lock. By the time it selects a non empty hash slot and sfq_dump_class_stats() is run (with lock held), slot might have been freed : We then access q->slots[SFQ_EMPTY_SLOT], out of bounds, and crash in slot_queue_walk() On previous kernels, bug is here but out of bounds qs[SFQ_DEPTH] and allot[SFQ_DEPTH] are located in struct sfq_sched_data, so no illegal memory access happens, only possibly wrong data reported to user. Also, slot_dequeue_tail() should make sure slot skb chain is correctly terminated, or sfq_dump_class_stats() can access freed skbs. Signed-off-by: Eric Dumazet Cc: Jarek Poplawski --- net/sched/sch_sfq.c | 16 +++++++++++----- 1 files changed, 11 insertions(+), 5 deletions(-) diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c index 13322e8..6a2f88f 100644 --- a/net/sched/sch_sfq.c +++ b/net/sched/sch_sfq.c @@ -281,6 +281,7 @@ static inline struct sk_buff *slot_dequeue_tail(struct sfq_slot *slot) struct sk_buff *skb = slot->skblist_prev; slot->skblist_prev = skb->prev; + skb->prev->next = (struct sk_buff *)slot; skb->next = skb->prev = NULL; return skb; } @@ -608,14 +609,19 @@ static int sfq_dump_class_stats(struct Qdisc *sch, unsigned long cl, struct gnet_dump *d) { struct sfq_sched_data *q = qdisc_priv(sch); - const struct sfq_slot *slot = &q->slots[q->ht[cl - 1]]; - struct gnet_stats_queue qs = { .qlen = slot->qlen }; - struct tc_sfq_xstats xstats = { .allot = slot->allot }; + sfq_index idx = q->ht[cl - 1]; + struct gnet_stats_queue qs = { 0 }; + struct tc_sfq_xstats xstats = { 0 }; struct sk_buff *skb; - slot_queue_walk(slot, skb) - qs.backlog += qdisc_pkt_len(skb); + if (idx != SFQ_EMPTY_SLOT) { + const struct sfq_slot *slot = &q->slots[idx]; + xstats.allot = slot->allot; + qs.qlen = slot->qlen; + slot_queue_walk(slot, skb) + qs.backlog += qdisc_pkt_len(skb); + } if (gnet_stats_copy_queue(d, &qs) < 0) return -1; return gnet_stats_copy_app(d, &xstats, sizeof(xstats));