From: Marcelo Tosatti <mtosatti@redhat.com>
To: Linus Torvalds <torvalds@linux-foundation.org>
Cc: h@amt.cnet, Thomas Gleixner <tglx@linutronix.de>,
Greg KH <gregkh@linuxfoundation.org>,
"Luis R. Rodriguez" <mcgrof@kernel.org>,
Martin Fuzzey <mfuzzey@parkeon.com>,
"Eric W. Biederman" <ebiederm@xmission.com>,
Dmitry Torokhov <dmitry.torokhov@gmail.com>,
Daniel Wagner <wagi@monom.org>,
David Woodhouse <dwmw2@infradead.org>,
jewalt@lgsinnovations.com, rafal@milecki.pl,
Arend Van Spriel <arend.vanspriel@broadcom.com>,
"Rafael J. Wysocki" <rjw@rjwysocki.net>,
"Li, Yi" <yi1.li@linux.intel.com>,
atull@kernel.org, Moritz Fischer <moritz.fischer@ettus.com>,
Petr Mladek <pmladek@suse.com>,
Johannes Berg <johannes.berg@intel.com>,
Emmanuel Grumbach <emmanuel.grumbach@intel.com>,
"Coelho, Luciano" <luciano.coelho@intel.com>,
Kalle Valo <kvalo@codeaurora.org>
Subject: Re: [PATCH 2/4] swait: add the missing killable swaits
Date: Fri, 30 Jun 2017 08:57:14 -0300 [thread overview]
Message-ID: <20170630115714.GD12169@amt.cnet> (raw)
In-Reply-To: <CA+55aFykNULx-b6M6FmUYdK2cn-OJKKfjaPwLN5xZGK+bioGaA@mail.gmail.com>
On Thu, Jun 29, 2017 at 09:03:42PM -0700, Linus Torvalds wrote:
> On Thu, Jun 29, 2017 at 12:15 PM, Marcelo Tosatti <mtosatti@redhat.com> wrote:
> > On Thu, Jun 29, 2017 at 09:13:29AM -0700, Linus Torvalds wrote:
> >>
> >> swait uses special locking and has odd semantics that are not at all
> >> the same as the default wait queue ones. It should not be used without
> >> very strong reasons (and honestly, the only strong enough reason seems
> >> to be "RT").
> >
> > Performance shortcut:
> >
> > https://lkml.org/lkml/2016/2/25/301
>
> Yes, I know why kvm uses it, I just don't think it's necessarily the
> right thing.
>
> That kvm commit is actually a great example: it uses swake_up() from
> an interrupt, and that's in fact the *reason* it uses swake_up().
>
> But that also fundamentally means that it cannot use swake_up_all(),
> so it basically *relies* on there only ever being one single entry
> that needs to be woken up.
>
> And as far as I can tell, it really is because the queue only ever has
> one entry (ie it's per-vcpu, and when the vcpu is blocked, it's
> blocked - so no other user will be waiting there).
Exactly.
>
> So it isn't that you migth queue multiple entries and then just wake
> them up one at a time. There really is just one entry at a time,
> right?
Yes.
> And that means that swait is actuially completely the wrong thing to
> do. It's more expensive and more complex than just saving the single
> process pointer away and just doing "wake_up_process()".
Aha, i see.
>
> Now, it really is entirely possible that I'm missing something, but it
> does look like that to me.
Just drop it -- the optimization is not relevant anymore given VMX
hardware improvements.
> We've had wake_up_process() since pretty much day #1. THAT is the
> fastest and simplest direct wake-up there is, not some "simple
> wait-queue".
>
> Now, admittedly I don't know the code and really may be entirely off,
> but looking at the commit (no need to go to the lkml archives - it's
> commit 8577370fb0cb ("KVM: Use simple waitqueue for vcpu->wq") in
> mainline), I really think the swait() use is simply not correct if
> there can be multiple waiters, exactly because swake_up() only wakes
> up a single entry.
There can't be: its one emulated LAPIC per vcpu. So only one vcpu
waits for that waitqueue.
> So either there is only a single entry, or *all* the code like
>
> dvcpu->arch.wait = 0;
>
> - if (waitqueue_active(&dvcpu->wq))
> - wake_up_interruptible(&dvcpu->wq);
> + if (swait_active(&dvcpu->wq))
> + swake_up(&dvcpu->wq);
>
> is simply wrong. If there are multiple blockers, and you just cleared
> "arch.wait", I think they should *all* be woken up. And that's not
> what swake_up() does.
>
> So I think that kvm_vcpu_block() could easily have instead done
>
> vcpu->process = current;
>
> as the "prepare_to_wait()" part, and "finish_wait()" would be to just
> clear vcpu->process. No wait-queue, just a single pointer to the
> single blocking thread.
>
> (Of course, you still need serialization, so that
> "wake_up_process(vcpu->process)" doesn't end up using a stale value,
> but since processes are already freed with RCU because of other things
> like that, the serialization is very low-cost, you only need to be
> RCU-read safe when waking up).
>
> See what I'm saying?
>
> Note that "wake_up_process()" really is fairly widely used. It's
> widely used because it's fairly obvious, and because that really *is*
> the lowest-possible cost: a single pointer to the sleeping thread, and
> you can often do almost no locking at all.
>
> And unlike swake_up(), it's obvious that you only wake up a single thread.
>
> Linus
Feel free to drop the KVM usage... agreed the interface is a special
case and a generic one which handles multiple waiters
and has debugging etc should be preferred to avoid bugs
Not sure if other people are using it (swait).
next prev parent reply other threads:[~2017-06-30 11:57 UTC|newest]
Thread overview: 38+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-06-14 22:20 [PATCH 0/4] firmware: fix fallback mechanism by ignoring SIGCHLD Luis R. Rodriguez
2017-06-14 22:20 ` [PATCH 1/4] test_firmware: add test case for SIGCHLD on sync fallback Luis R. Rodriguez
2017-06-14 22:20 ` [PATCH 2/4] swait: add the missing killable swaits Luis R. Rodriguez
[not found] ` <20170614222017.14653-3-mcgrof-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
2017-06-29 12:54 ` Greg KH
[not found] ` <20170629125402.GH26046-U8xfFu+wG4EAvxtiuMwx3w@public.gmane.org>
2017-06-29 13:05 ` Thomas Gleixner
2017-06-29 13:35 ` Greg KH
2017-06-29 13:46 ` Thomas Gleixner
2017-06-29 16:13 ` Linus Torvalds
2017-06-29 16:31 ` Matthew Wilcox
2017-06-29 17:29 ` Luis R. Rodriguez
2017-06-29 17:40 ` Davidlohr Bueso
2017-06-29 17:57 ` Linus Torvalds
2017-06-29 18:33 ` Davidlohr Bueso
[not found] ` <20170629183339.GD3954-3dK4OQgjB4rH06JGZaSw0A@public.gmane.org>
2017-06-29 18:59 ` Linus Torvalds
[not found] ` <CA+55aFz8Mhx+A-g-5yOG-O1ZLRUR_fpeeA4iBNGH8EnDBZEdpA-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2017-06-29 19:40 ` Luis R. Rodriguez
2017-06-29 19:44 ` Luis R. Rodriguez
[not found] ` <20170629194455.GR21846-B4tOwbsTzaBolqkO4TVVkw@public.gmane.org>
2017-06-29 20:58 ` Jakub Kicinski
2017-06-29 22:50 ` Luis R. Rodriguez
2017-06-29 22:53 ` Jakub Kicinski
2017-06-29 23:00 ` Luis R. Rodriguez
2017-06-29 23:06 ` Jakub Kicinski
[not found] ` <20170629225003.GU21846-B4tOwbsTzaBolqkO4TVVkw@public.gmane.org>
2017-07-12 21:33 ` Luis R. Rodriguez
[not found] ` <20170629194015.GQ21846-B4tOwbsTzaBolqkO4TVVkw@public.gmane.org>
2017-06-29 20:57 ` Linus Torvalds
2017-07-05 2:06 ` Davidlohr Bueso
[not found] ` <20170705020635.GD11168-3dK4OQgjB4rH06JGZaSw0A@public.gmane.org>
2017-07-07 19:58 ` Linus Torvalds
2017-07-07 22:27 ` Davidlohr Bueso
2017-07-07 22:48 ` Linus Torvalds
2017-06-29 19:15 ` Marcelo Tosatti
[not found] ` <20170629191506.GB12368-I4X2Mt4zSy4@public.gmane.org>
2017-06-30 4:03 ` Linus Torvalds
2017-06-30 11:55 ` Marcelo Tosatti
2017-06-30 11:57 ` Marcelo Tosatti [this message]
2017-06-30 17:30 ` Krister Johansen
2017-06-14 22:20 ` [PATCH 3/4] firmware: avoid invalid fallback aborts by using killable swait Luis R. Rodriguez
2017-06-14 22:20 ` [PATCH 4/4] firmware: send -EINTR on signal abort on fallback mechanism Luis R. Rodriguez
2017-06-15 7:49 ` [PATCH 0/4] firmware: fix fallback mechanism by ignoring SIGCHLD Martin Fuzzey
[not found] ` <20170614222017.14653-1-mcgrof-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
2017-06-26 21:19 ` Luis R. Rodriguez
2017-06-29 15:14 ` Greg KH
[not found] ` <20170629151442.GA4880-U8xfFu+wG4EAvxtiuMwx3w@public.gmane.org>
2017-06-29 17:29 ` Luis R. Rodriguez
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=20170630115714.GD12169@amt.cnet \
--to=mtosatti@redhat.com \
--cc=arend.vanspriel@broadcom.com \
--cc=atull@kernel.org \
--cc=dmitry.torokhov@gmail.com \
--cc=dwmw2@infradead.org \
--cc=ebiederm@xmission.com \
--cc=emmanuel.grumbach@intel.com \
--cc=gregkh@linuxfoundation.org \
--cc=h@amt.cnet \
--cc=jewalt@lgsinnovations.com \
--cc=johannes.berg@intel.com \
--cc=kvalo@codeaurora.org \
--cc=luciano.coelho@intel.com \
--cc=mcgrof@kernel.org \
--cc=mfuzzey@parkeon.com \
--cc=moritz.fischer@ettus.com \
--cc=pmladek@suse.com \
--cc=rafal@milecki.pl \
--cc=rjw@rjwysocki.net \
--cc=tglx@linutronix.de \
--cc=torvalds@linux-foundation.org \
--cc=wagi@monom.org \
--cc=yi1.li@linux.intel.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).