From: John Fastabend <john.fastabend@gmail.com>
To: Cong Wang <xiyou.wangcong@gmail.com>
Cc: Jamal Hadi Salim <jhs@mojatatu.com>,
Linux Kernel Network Developers <netdev@vger.kernel.org>,
David Miller <davem@davemloft.net>,
Eric Dumazet <eric.dumazet@gmail.com>
Subject: Re: [RFC PATCH 14/15] net: sched: make qstats per cpu
Date: Wed, 30 Apr 2014 15:29:14 -0700 [thread overview]
Message-ID: <5361793A.7000201@gmail.com> (raw)
In-Reply-To: <CAM_iQpUMJcP3guRvCPBGqfzTbVn+t4uZ8eETGVnwk5Y88ZJi_A@mail.gmail.com>
On 04/30/2014 10:08 AM, Cong Wang wrote:
> On Wed, Apr 30, 2014 at 9:40 AM, John Fastabend
> <john.fastabend@gmail.com> wrote:
>> diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
>> index 1a55e19..6d67a27 100644
>> --- a/include/net/sch_generic.h
>> +++ b/include/net/sch_generic.h
>> @@ -92,7 +92,10 @@ struct Qdisc {
>> struct gnet_stats_basic_cpu __percpu *cpu_bstats;
>> } bstats_qdisc;
>> unsigned int __state;
>> - struct gnet_stats_queue qstats;
>> + union {
>> + struct gnet_stats_queue qstats;
>> + struct gnet_stats_queue __percpu *cpu_qstats;
>> + } qstats_qdisc;
>> struct rcu_head rcu_head;
>> int padded;
>> atomic_t refcnt;
>
>
> So only ingress qdisc uses ->cpu_qstats, the rest still uses
> ->qstats. Hmm, doesn't this mean ingress qdisc should init
> ->cpu_qstats in its ->init()?
>
> [...]
>
>> diff --git a/net/sched/sch_ingress.c b/net/sched/sch_ingress.c
>> index 5700c27..2423d7b 100644
>> --- a/net/sched/sch_ingress.c
>> +++ b/net/sched/sch_ingress.c
>> @@ -58,6 +58,7 @@ static struct tcf_proto **ingress_find_tcf(struct Qdisc *sch, unsigned long cl)
>> static int ingress_enqueue(struct sk_buff *skb, struct Qdisc *sch)
>> {
>> struct ingress_qdisc_data *p = qdisc_priv(sch);
>> + struct gnet_stats_queue *qstats;
>> struct tcf_result res;
>> struct tcf_proto *fl = rcu_dereference_bh(p->filter_list);
>> int result;
>> @@ -68,7 +69,8 @@ static int ingress_enqueue(struct sk_buff *skb, struct Qdisc *sch)
>> switch (result) {
>> case TC_ACT_SHOT:
>> result = TC_ACT_SHOT;
>> - sch->qstats.drops++;
>> + qstats = this_cpu_ptr(sch->qstats_qdisc.cpu_qstats);
>> + qstats->drops++;
>> break;
>> case TC_ACT_STOLEN:
>> case TC_ACT_QUEUED:
>
> I don't see you call alloc_percpu() and free_percpu() in sch_ingress.c.
>
Right I missed this update in the patch set its included now. Although
its part of sch_api.c and triggers off the TCQ_F_LLQDISC flag.
Thanks,
John
--
John Fastabend Intel Corporation
next prev parent reply other threads:[~2014-04-30 22:29 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-04-30 16:34 [RFC PATCH 00/15] remove qdisc lock from ingress_qdisc John Fastabend
2014-04-30 16:35 ` [RFC PATCH 01/15] net: qdisc: use rcu prefix and silence sparse warnings John Fastabend
2014-04-30 17:00 ` Eric Dumazet
2014-04-30 22:25 ` John Fastabend
2014-04-30 23:29 ` Eric Dumazet
2014-05-01 15:20 ` John Fastabend
2014-05-14 19:39 ` John Fastabend
2014-05-15 20:41 ` Paul E. McKenney
2014-05-15 21:11 ` John Fastabend
2014-05-15 20:43 ` David Miller
2014-04-30 16:35 ` [RFC PATCH 02/15] net: rcu-ify tcf_proto John Fastabend
2014-04-30 16:36 ` [RFC PATCH 03/15] net: sched: cls_basic use RCU John Fastabend
2014-04-30 16:36 ` [RFC PATCH 04/15] net: sched: cls_cgroup " John Fastabend
2014-04-30 16:36 ` [RFC PATCH 05/15] net: sched: cls_flow " John Fastabend
2014-04-30 16:37 ` [RFC PATCH 06/15] net: sched: fw " John Fastabend
2014-04-30 16:37 ` [RFC PATCH 07/15] net: sched: RCU cls_route John Fastabend
2014-04-30 16:38 ` [RFC PATCH 08/15] net: sched: RCU cls_tcindex John Fastabend
2014-04-30 16:38 ` [RFC PATCH 09/15] net: sched: make cls_u32 lockless John Fastabend
2014-04-30 16:39 ` [RFC PATCH 10/15] net: sched: rcu'ify cls_rsvp John Fastabend
2014-04-30 16:39 ` [RFC PATCH 11/15] net: make cls_bpf rcu safe John Fastabend
2014-04-30 16:39 ` [RFC PATCH 12/15] net: sched: make tc_action safe to walk under RCU John Fastabend
2014-04-30 16:40 ` [RFC PATCH 13/15] net: sched: make bstats per cpu and estimator RCU safe John Fastabend
2014-04-30 16:40 ` [RFC PATCH 14/15] net: sched: make qstats per cpu John Fastabend
2014-04-30 17:08 ` Cong Wang
2014-04-30 22:29 ` John Fastabend [this message]
2014-04-30 16:41 ` [RFC PATCH 15/15] net: sched: drop ingress qdisc lock John Fastabend
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=5361793A.7000201@gmail.com \
--to=john.fastabend@gmail.com \
--cc=davem@davemloft.net \
--cc=eric.dumazet@gmail.com \
--cc=jhs@mojatatu.com \
--cc=netdev@vger.kernel.org \
--cc=xiyou.wangcong@gmail.com \
/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).