From: Paolo Abeni <pabeni@redhat.com>
To: Jakub Kicinski <kuba@kernel.org>,
peterz@infradead.org, tglx@linutronix.de
Cc: jstultz@google.com, edumazet@google.com, netdev@vger.kernel.org,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH 0/3] softirq: uncontroversial change
Date: Thu, 20 Apr 2023 19:24:02 +0200 [thread overview]
Message-ID: <305d7742212cbe98621b16be782b0562f1012cb6.camel@redhat.com> (raw)
In-Reply-To: <20221222221244.1290833-1-kuba@kernel.org>
Hi all,
On Thu, 2022-12-22 at 14:12 -0800, Jakub Kicinski wrote:
> Catching up on LWN I run across the article about softirq
> changes, and then I noticed fresh patches in Peter's tree.
> So probably wise for me to throw these out there.
>
> My (can I say Meta's?) problem is the opposite to what the RT
> sensitive people complain about. In the current scheme once
> ksoftirqd is woken no network processing happens until it runs.
>
> When networking gets overloaded - that's probably fair, the problem
> is that we confuse latency tweaks with overload protection. We have
> a needs_resched() in the loop condition (which is a latency tweak)
> Most often we defer to ksoftirqd because we're trying to be nice
> and let user space respond quickly, not because there is an
> overload. But the user space may not be nice, and sit on the CPU
> for 10ms+. Also the sirq's "work allowance" is 2ms, which is
> uncomfortably close to the timer tick, but that's another story.
>
> We have a sirq latency tracker in our prod kernel which catches
> 8ms+ stalls of net Tx (packets queued to the NIC but there is
> no NAPI cleanup within 8ms) and with these patches applied
> on 5.19 fully loaded web machine sees a drop in stalls from
> 1.8 stalls/sec to 0.16/sec. I also see a 50% drop in outgoing
> TCP retransmissions and ~10% drop in non-TLP incoming ones.
> This is not a network-heavy workload so most of the rtx are
> due to scheduling artifacts.
>
> The network latency in a datacenter is somewhere around neat
> 1000x lower than scheduling granularity (around 10us).
>
> These patches (patch 2 is "the meat") change what we recognize
> as overload. Instead of just checking if "ksoftirqd is woken"
> it also caps how long we consider ourselves to be in overload,
> a time limit which is different based on whether we yield due
> to real resource exhaustion vs just hitting that needs_resched().
>
> I hope the core concept is not entirely idiotic. It'd be great
> if we could get this in or fold an equivalent concept into ongoing
> work from others, because due to various "scheduler improvements"
> every time we upgrade the production kernel this problem is getting
> worse :(
Please allow me to revive this old thread.
My understanding is that we want to avoid adding more heuristics here,
preferring a consistent refactor.
I would like to propose a revert of:
4cd13c21b207 softirq: Let ksoftirqd do its job
the its follow-ups:
3c53776e29f8 Mark HI and TASKLET softirq synchronous
0f50524789fc softirq: Don't skip softirq execution when softirq thread is parking
The problem originally addressed by 4cd13c21b207 can now be tackled
with the threaded napi, available since:
29863d41bb6e net: implement threaded-able napi poll loop support
Reverting the mentioned commit should address the latency issues
mentioned by Jakub - I verified it solves a somewhat related problem in
my setup - and reduces the layering of heuristics in this area.
A refactor introducing uniform overload detection and proper resource
control will be better, but I admit it's beyond me and anyway it could
still land afterwards.
Any opinion more then welcome!
Thanks,
Paolo
next prev parent reply other threads:[~2023-04-20 17:25 UTC|newest]
Thread overview: 38+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-12-22 22:12 [PATCH 0/3] softirq: uncontroversial change Jakub Kicinski
2022-12-22 22:12 ` [PATCH 1/3] softirq: rename ksoftirqd_running() -> ksoftirqd_should_handle() Jakub Kicinski
2022-12-22 22:12 ` [PATCH 2/3] softirq: avoid spurious stalls due to need_resched() Jakub Kicinski
2023-01-31 22:32 ` Jakub Kicinski
2023-03-03 13:30 ` Thomas Gleixner
2023-03-03 15:18 ` Thomas Gleixner
2023-03-03 21:31 ` Jakub Kicinski
2023-03-03 22:37 ` Paul E. McKenney
2023-03-03 23:25 ` Dave Taht
2023-03-04 1:14 ` Paul E. McKenney
2023-03-03 23:36 ` Paul E. McKenney
2023-03-03 23:44 ` Jakub Kicinski
2023-03-04 1:25 ` Paul E. McKenney
2023-03-04 1:39 ` Jakub Kicinski
2023-03-04 3:11 ` Paul E. McKenney
2023-03-04 20:48 ` Paul E. McKenney
2023-03-05 20:43 ` Thomas Gleixner
2023-03-05 22:42 ` Paul E. McKenney
2023-03-05 23:00 ` Frederic Weisbecker
2023-03-06 4:30 ` Paul E. McKenney
2023-03-06 11:22 ` Frederic Weisbecker
2023-03-06 9:13 ` David Laight
2023-03-06 11:57 ` Frederic Weisbecker
2023-03-06 14:57 ` Paul E. McKenney
2023-03-07 0:51 ` Jakub Kicinski
2022-12-22 22:12 ` [PATCH 3/3] softirq: don't yield if only expedited handlers are pending Jakub Kicinski
2023-01-09 9:44 ` Peter Zijlstra
2023-01-09 10:16 ` Eric Dumazet
2023-01-09 19:12 ` Jakub Kicinski
2023-03-03 11:41 ` Thomas Gleixner
2023-03-03 14:17 ` Thomas Gleixner
2023-04-20 17:24 ` Paolo Abeni [this message]
2023-04-20 17:41 ` [PATCH 0/3] softirq: uncontroversial change Eric Dumazet
2023-04-20 20:23 ` Paolo Abeni
2023-04-21 2:48 ` Jason Xing
2023-04-21 9:33 ` Paolo Abeni
2023-04-21 9:46 ` Jason Xing
2023-05-09 19:56 ` [tip: irq/core] Revert "softirq: Let ksoftirqd do its job" tip-bot2 for Paolo Abeni
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=305d7742212cbe98621b16be782b0562f1012cb6.camel@redhat.com \
--to=pabeni@redhat.com \
--cc=edumazet@google.com \
--cc=jstultz@google.com \
--cc=kuba@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=netdev@vger.kernel.org \
--cc=peterz@infradead.org \
--cc=tglx@linutronix.de \
/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).