From: Paolo Bonzini <pbonzini@redhat.com>
To: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>,
Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>,
mingo@redhat.com, kvm@vger.kernel.org,
linux-kernel@vger.kernel.org,
Thomas Gleixner <tglx@linutronix.de>
Subject: Re: [patch 3/3] x86: kvm guest side support for KVM_HC_RT_PRIO hypercall\
Date: Fri, 29 Sep 2017 19:05:41 +0200 [thread overview]
Message-ID: <e3f4e6b0-4974-f1e2-b8a2-6b8cd1fb759e@redhat.com> (raw)
In-Reply-To: <20170929164006.GC29391@amt.cnet>
On 29/09/2017 18:40, Marcelo Tosatti wrote:
>> If you know you have this kind disk workload, you must use virtio-blk or
>> virtio-scsi with iothreads and place the iothreads on their own physical
>> CPUs.
>>
>> Among "run arbitrary workloads", "run real-time workloads", "pack stuff
>> into as few physical CPUs as possible", you can only pick two.
>
> Thats not the state of things (userspace in vcpu-0 is not specially tailored
> to not violate latencies in vcpu-1): that is not all user triggered
> actions can be verified.
>
> Think "updatedb", and so on...
_Which_ spinlock is it that can cause unwanted latency while running
updatedb on VCPU0 and a real-time workload on VCPU1, and only so on virt
because of the emulator thread? Is this still broken if you set up
priorities for the emulator thread correctly and use PI mutexes in QEMU?
And if so, what is the cause of interruptions in the emulator thread
and how are these interruptions causing the jitter?
Priorities and priority inheritance (or lack of them) is a _known_
issue. Jan was doing his KVM-RT things in 2009 and he was talking about
priorities[1] back then. The effect of correct priorities is to _lower_
jitter, not to make it worse, and anyway certainly not worse than
SCHED_NORMAL I/O thread. Once that's fixed, we can look at other problems.
Paolo
[1] http://static.lwn.net/images/conf/rtlws11/papers/proc/p18.pdf which
also mentions pv scheduling
next prev parent reply other threads:[~2017-09-29 17:05 UTC|newest]
Thread overview: 52+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-09-21 11:38 [patch 0/3] KVM KVM_HC_RT_PRIO hypercall support Marcelo Tosatti
2017-09-21 11:38 ` [patch 1/3] KVM: x86: add per-vcpu option to set guest vcpu -RT priority Marcelo Tosatti
2017-09-21 11:38 ` [patch 2/3] KVM: x86: KVM_HC_RT_PRIO hypercall (host-side) Marcelo Tosatti
2017-09-21 13:32 ` Konrad Rzeszutek Wilk
2017-09-21 13:49 ` Paolo Bonzini
2017-09-22 1:08 ` Marcelo Tosatti
2017-09-22 7:23 ` Paolo Bonzini
2017-09-22 12:24 ` Marcelo Tosatti
2017-09-21 11:38 ` [patch 3/3] x86: kvm guest side support for KVM_HC_RT_PRIO hypercall Marcelo Tosatti
2017-09-21 13:36 ` Konrad Rzeszutek Wilk
2017-09-21 14:06 ` Peter Zijlstra
2017-09-22 1:10 ` Marcelo Tosatti
2017-09-22 10:00 ` Peter Zijlstra
2017-09-22 10:56 ` Peter Zijlstra
2017-09-22 12:33 ` Marcelo Tosatti
2017-09-22 12:55 ` Peter Zijlstra
2017-09-23 10:56 ` Paolo Bonzini
2017-09-23 13:41 ` Peter Zijlstra
2017-09-24 13:05 ` Paolo Bonzini
2017-09-25 2:57 ` Marcelo Tosatti
2017-09-25 9:13 ` Peter Zijlstra
2017-09-25 15:12 ` Paolo Bonzini
2017-09-26 22:49 ` [patch 3/3] x86: kvm guest side support for KVM_HC_RT_PRIO hypercall\ Marcelo Tosatti
2017-09-27 9:37 ` Paolo Bonzini
2017-09-28 0:44 ` Marcelo Tosatti
2017-09-28 7:22 ` Paolo Bonzini
2017-09-28 21:35 ` Marcelo Tosatti
2017-09-28 21:41 ` Marcelo Tosatti
2017-09-29 8:18 ` Paolo Bonzini
2017-09-29 16:40 ` Marcelo Tosatti
2017-09-29 17:05 ` Paolo Bonzini [this message]
2017-09-29 20:17 ` Marcelo Tosatti
2017-10-02 12:30 ` Paolo Bonzini
2017-10-02 12:48 ` Peter Zijlstra
2017-09-26 23:22 ` [patch 3/3] x86: kvm guest side support for KVM_HC_RT_PRIO hypercall Marcelo Tosatti
2017-09-25 16:20 ` Konrad Rzeszutek Wilk
2017-09-22 12:16 ` Marcelo Tosatti
2017-09-22 12:31 ` Peter Zijlstra
2017-09-22 12:36 ` Marcelo Tosatti
2017-09-22 12:59 ` Peter Zijlstra
2017-09-25 1:52 ` Marcelo Tosatti
2017-09-25 8:35 ` Peter Zijlstra
2017-09-22 12:40 ` [patch 3/3] x86: kvm guest side support for KVM_HC_RT_PRIO hypercall\ Marcelo Tosatti
2017-09-22 13:01 ` Peter Zijlstra
2017-09-25 2:22 ` Marcelo Tosatti
2017-09-25 8:58 ` Peter Zijlstra
2017-09-25 10:41 ` Thomas Gleixner
2017-09-25 18:28 ` Jan Kiszka
2017-09-21 17:45 ` [patch 0/3] KVM KVM_HC_RT_PRIO hypercall support Jan Kiszka
2017-09-22 1:19 ` Marcelo Tosatti
2017-09-22 6:23 ` Jan Kiszka
2017-09-26 23:59 ` Marcelo Tosatti
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=e3f4e6b0-4974-f1e2-b8a2-6b8cd1fb759e@redhat.com \
--to=pbonzini@redhat.com \
--cc=konrad.wilk@oracle.com \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=mtosatti@redhat.com \
--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 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.