All of lore.kernel.org
 help / color / mirror / Atom feed
From: Cong Wang <xiyou.wangcong@gmail.com>
To: Eric Dumazet <eric.dumazet@gmail.com>
Cc: David Miller <davem@davemloft.net>,
	Linux Kernel Network Developers <netdev@vger.kernel.org>,
	Jamal Hadi Salim <jhs@mojatatu.com>,
	John Fastabend <john.fastabend@gmail.com>,
	Hadar Hen Zion <hadarh@mellanox.com>,
	Amir Vadai <amirva@mellanox.com>
Subject: Re: [PATCH net] net_sched: act_mirred: full rcu conversion
Date: Sun, 11 Sep 2016 22:46:02 -0700	[thread overview]
Message-ID: <CAM_iQpXeNY1Kxo1MvZPW5JqB+wEvUXoD0oh-LoB2CdOL=qh9CQ@mail.gmail.com> (raw)
In-Reply-To: <1473423837.18970.46.camel@edumazet-glaptop3.roam.corp.google.com>

On Fri, Sep 9, 2016 at 5:23 AM, Eric Dumazet <eric.dumazet@gmail.com> wrote:
> On Thu, 2016-09-08 at 22:24 -0700, Cong Wang wrote:
>> On Thu, Sep 8, 2016 at 8:35 AM, Eric Dumazet <eric.dumazet@gmail.com> wrote:
>> > From: Eric Dumazet <edumazet@google.com>
>> >
>> > As reported by Cong Wang, I was lazy when I did initial RCU conversion
>> > of tc_mirred, as I thought I could avoid allocation/freeing of a
>> > parameter block.
>>
>> Quote from Eric Dumazet:
>>
>> https://www.mail-archive.com/netdev@vger.kernel.org/msg115482.html
>>
>> <Quote>
>> Well, I added a READ_ONCE() to read tcf_action once.
>>
>> Adding rcu here would mean adding a pointer and extra cache line, to
>> deref the values.
>>
>> IMHO the race here has no effect . You either read the old or new value.
>> </Quote>
>>
>> Me with facepalm... ;-)
>
>
> Point is still valid. Show me a real case where it was a serious
> problem, instead of simply theoretical.
>
> tc_mirred + ifb patches allowed us to reach a milestone, removing the
> last contended spinlocks, and you are catching up with this one year
> later.
>
> I wont backport this fix in Google prod kernels, because there is
> absolutely no way we need it, and the extra memory cache line might hurt
> latencies.
>
> Since you did not write a fix on your side since June 17th, I presume
> you do not care that much.

Sounds like I were the author of this patch.... Why are you questioning
your own patch? Did I ask you to care about it? ;-)

Please drop this patch.

Thanks.

  reply	other threads:[~2016-09-12  5:46 UTC|newest]

Thread overview: 37+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-09-02  5:57 [RFC Patch net-next 0/6] net_sched: really switch to RCU for tc actions Cong Wang
2016-09-02  5:57 ` [RFC Patch net-next 1/6] net_sched: use RCU for action hash table Cong Wang
2016-09-06 12:47   ` Jamal Hadi Salim
2016-09-06 22:37     ` Cong Wang
2016-09-02  5:57 ` [RFC Patch net-next 2/6] net_sched: introduce tcf_hash_replace() Cong Wang
2016-09-06 12:52   ` Jamal Hadi Salim
2016-09-06 22:34     ` Cong Wang
2016-09-02  5:57 ` [RFC Patch net-next 3/6] net_sched: return NULL in tcf_hash_check() Cong Wang
2016-09-02  5:57 ` [RFC Patch net-next 4/6] net_sched: introduce tcf_hash_copy() Cong Wang
2016-09-02  5:57 ` [RFC Patch net-next 5/6] net_sched: use rcu in fast path Cong Wang
2016-09-06 14:52   ` Eric Dumazet
2016-09-08  6:04     ` John Fastabend
2016-09-08 13:28       ` Eric Dumazet
2016-09-08 15:35         ` [PATCH net] net_sched: act_mirred: full rcu conversion Eric Dumazet
2016-09-08 15:47           ` John Fastabend
2016-09-08 15:51             ` Eric Dumazet
2016-09-09  5:26               ` Cong Wang
2016-09-09 12:23                 ` Eric Dumazet
2016-09-09 15:52                 ` John Fastabend
2016-09-12  6:12                   ` Cong Wang
2016-09-12 15:34                     ` John Fastabend
2016-09-09  5:24           ` Cong Wang
2016-09-09  5:48             ` Alexei Starovoitov
2016-09-09  5:59               ` Cong Wang
2016-09-09 12:25                 ` Eric Dumazet
2016-09-09 12:23             ` Eric Dumazet
2016-09-12  5:46               ` Cong Wang [this message]
2016-09-08 15:49         ` [RFC Patch net-next 5/6] net_sched: use rcu in fast path John Fastabend
2016-09-09  5:54           ` Cong Wang
2016-09-09 15:25             ` John Fastabend
2016-09-09  5:49     ` Cong Wang
2016-09-02  5:57 ` [RFC Patch net-next 6/6] net_sched: switch to RCU API for act_mirred Cong Wang
2016-09-02  7:09 ` [RFC Patch net-next 0/6] net_sched: really switch to RCU for tc actions Jiri Pirko
2016-09-02 19:44   ` Cong Wang
2016-09-07 16:23 ` John Fastabend
2016-09-08  6:05   ` John Fastabend
2016-09-09  5:22   ` Cong Wang

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='CAM_iQpXeNY1Kxo1MvZPW5JqB+wEvUXoD0oh-LoB2CdOL=qh9CQ@mail.gmail.com' \
    --to=xiyou.wangcong@gmail.com \
    --cc=amirva@mellanox.com \
    --cc=davem@davemloft.net \
    --cc=eric.dumazet@gmail.com \
    --cc=hadarh@mellanox.com \
    --cc=jhs@mojatatu.com \
    --cc=john.fastabend@gmail.com \
    --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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.