From: Eric Dumazet Date: Tue, 15 Dec 2015 17:43:12 +0000 (-0800) Subject: net_sched: make qdisc_tree_decrease_qlen() work for non mq X-Git-Tag: v4.1.16~13 X-Git-Url: http://git.hungrycats.org/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=262ee9759e5f7b0db137fd6ee5780f92e119ff6e;p=linux net_sched: make qdisc_tree_decrease_qlen() work for non mq [ Upstream commit 225734de70cd0a9e0b978f3583a4a87939271d5e ] Stas Nichiporovich reported a regression in his HFSC qdisc setup on a non multi queue device. It turns out I mistakenly added a TCQ_F_NOPARENT flag on all qdisc allocated in qdisc_create() for non multi queue devices, which was rather buggy. I was clearly mislead by the TCQ_F_ONETXQUEUE that is also set here for no good reason, since it only matters for the root qdisc. Fixes: 4eaf3b84f288 ("net_sched: fix qdisc_tree_decrease_qlen() races") Reported-by: Stas Nichiporovich Tested-by: Stas Nichiporovich Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index d4b6f3682c14..68c599a5e1d1 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -950,7 +950,7 @@ qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue, } lockdep_set_class(qdisc_lock(sch), &qdisc_tx_lock); if (!netif_is_multiqueue(dev)) - sch->flags |= TCQ_F_ONETXQUEUE | TCQ_F_NOPARENT; + sch->flags |= TCQ_F_ONETXQUEUE; } sch->handle = handle;