From: Cong Wang <xiyou.wangcong@gmail.com>
To: "Václav Zindulka" <vaclav.zindulka@tlapnet.cz>
Cc: Linux Kernel Network Developers <netdev@vger.kernel.org>
Subject: Re: iproute2: tc deletion freezes whole server
Date: Mon, 4 May 2020 10:46:13 -0700 [thread overview]
Message-ID: <CAM_iQpXjsrraZpU3xhTvQ=owwzSTjAVdx-Aszz-yLitFzE5GsA@mail.gmail.com> (raw)
In-Reply-To: <CANxWus8=CZ8Y1GvqKFJHhdxun9gB8v1SP0XNZ7SMk4oDvkmEww@mail.gmail.com>
On Thu, Apr 30, 2020 at 5:40 AM Václav Zindulka
<vaclav.zindulka@tlapnet.cz> wrote:
>
> On Wed, Apr 15, 2020 at 5:01 PM Václav Zindulka
> <vaclav.zindulka@tlapnet.cz> wrote:
> > > > > The problem is actually more complicated than I thought, although it
> > > > > needs more work, below is the first pile of patches I have for you to
> > > > > test:
> > > > >
> > > > > https://github.com/congwang/linux/commits/qdisc_reset
> > > > >
> > > > > It is based on the latest net-next branch. Please let me know the result.
> > > >
> > > > I have applied all the patches in your four commits to my custom 5.4.6
> > > > kernel source. There was no change in the amount of fq_codel_reset
> > > > calls. Tested on ifb, RJ45 and SFP+ interfaces.
> > >
> > > It is true my patches do not reduce the number of fq_codel_reset() calls,
> > > they are intended to reduce the CPU time spent in each fq_codel_reset().
> > >
> > > Can you measure this? Note, you do not have to add your own printk()
> > > any more, because my patches add a few tracepoints, especially for
> > > qdisc_reset(). So you can obtain the time by checking the timestamps
> > > of these trace events. Of course, you can also use perf trace like you
> > > did before.
> >
> > Sorry for delayed responses. We were moving to a new house so I didn't
> > have much time to test it. I've measured your pile of patches applied
> > vs unpatched kernel. Result is a little bit better, but it is only
> > about 1s faster. Results are here. Do you need any additional reports
> > or measurements of other interfaces?
> > https://github.com/zvalcav/tc-kernel/tree/master/20200415 I've
> > recompiled the kernel without printk which had some overhead too.
>
> Hello Cong,
>
> did you have any time to look at it further? I'm just asking since my
> boss wants me to give him some verdict. I've started to study eBPF and
> XDP in the meantime so we have an alternative in case there won't be a
> solution to this problem.
Sorry for the delay. I lost connection to my dev machine, I am trying
to setup this on my own laptop.
Regarding to your test result above, I think I saw some difference
on my side, I have no idea why you didn't see any difference. Please
let me collect the data once I setup the test environment shortly today.
Thanks!
next prev parent reply other threads:[~2020-05-04 17:46 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-03-22 18:06 iproute2: tc deletion freezes whole server Václav Zindulka
2020-03-23 18:17 ` Cong Wang
2020-03-24 16:27 ` Václav Zindulka
2020-03-24 22:57 ` Cong Wang
2020-03-25 11:27 ` Václav Zindulka
2020-03-25 13:33 ` Václav Zindulka
2020-03-25 17:43 ` Cong Wang
2020-03-25 18:23 ` Václav Zindulka
2020-03-26 14:24 ` Václav Zindulka
2020-03-26 17:07 ` Cong Wang
2020-03-26 17:38 ` Cong Wang
2020-03-27 10:35 ` Václav Zindulka
2020-03-28 13:04 ` Václav Zindulka
2020-03-31 5:59 ` Cong Wang
2020-03-31 12:46 ` Václav Zindulka
2020-04-08 20:18 ` Cong Wang
2020-04-12 20:17 ` Václav Zindulka
2020-04-13 17:28 ` Cong Wang
2020-04-15 15:01 ` Václav Zindulka
2020-04-30 12:40 ` Václav Zindulka
2020-05-04 17:46 ` Cong Wang [this message]
2020-05-04 20:36 ` Cong Wang
2020-05-05 8:46 ` Václav Zindulka
2020-05-07 18:52 ` Cong Wang
2020-05-08 13:59 ` Václav Zindulka
2020-05-17 19:35 ` Cong Wang
2020-05-18 14:16 ` Václav Zindulka
2020-05-18 18:22 ` Cong Wang
2020-05-19 8:04 ` Václav Zindulka
2020-05-19 17:56 ` Cong Wang
2020-05-24 10:03 ` Václav Zindulka
2020-05-26 18:37 ` 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_iQpXjsrraZpU3xhTvQ=owwzSTjAVdx-Aszz-yLitFzE5GsA@mail.gmail.com' \
--to=xiyou.wangcong@gmail.com \
--cc=netdev@vger.kernel.org \
--cc=vaclav.zindulka@tlapnet.cz \
/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).