From: Paolo Bonzini <pbonzini@redhat.com>
To: Avi Kivity <avi@cloudius-systems.com>
Cc: Gleb Natapov <gleb@minantech.com>,
"Zhanghaoyu (A)" <haoyu.zhang@huawei.com>,
Gleb Natapov <gleb@redhat.com>, Avi Kivity <avi.kivity@gmail.com>,
"Huangweidong (C)" <weidong.huang@huawei.com>,
KVM <kvm@vger.kernel.org>, "Michael S. Tsirkin" <mst@redhat.com>,
"Jinxin (F)" <jinxin712@huawei.com>,
Luonengjun <luonengjun@huawei.com>,
"qemu-devel@nongnu.org" <qemu-devel@nongnu.org>,
Zanghongyong <zanghongyong@huawei.com>
Subject: Re: [Qemu-devel] [RFC] create a single workqueue for each vm to update vm irq routing table
Date: Thu, 28 Nov 2013 10:53:37 +0100 [thread overview]
Message-ID: <529712A1.8090207@redhat.com> (raw)
In-Reply-To: <5297118C.3050104@cloudius-systems.com>
Il 28/11/2013 10:49, Avi Kivity ha scritto:
> Linux is safe, it does interrupt migration from within the interrupt
> handler. If you do that before the device-specific EOI, you won't get
> another interrupt until programming the MSI is complete.
>
> Is virtio safe? IIRC it can post multiple interrupts without guest acks.
>
> Using call_rcu() is a better solution than srcu IMO. Less code changes,
> consistently faster.
call_rcu() has the problem of rate limiting, too. It wasn't such a
great idea, I think.
The QRCU I linked would work great latency-wise (it has roughly the same
latency of an rwsem but readers are lock-free). However, the locked
operations in the read path would hurt because of cache misses, so it's
not good either.
Paolo
WARNING: multiple messages have this Message-ID (diff)
From: Paolo Bonzini <pbonzini@redhat.com>
To: Avi Kivity <avi@cloudius-systems.com>
Cc: "Huangweidong (C)" <weidong.huang@huawei.com>,
Gleb Natapov <gleb@redhat.com>, KVM <kvm@vger.kernel.org>,
"Michael S. Tsirkin" <mst@redhat.com>,
"Zhanghaoyu (A)" <haoyu.zhang@huawei.com>,
Luonengjun <luonengjun@huawei.com>,
"qemu-devel@nongnu.org" <qemu-devel@nongnu.org>,
Zanghongyong <zanghongyong@huawei.com>,
Avi Kivity <avi.kivity@gmail.com>,
Gleb Natapov <gleb@minantech.com>,
"Jinxin (F)" <jinxin712@huawei.com>
Subject: Re: [Qemu-devel] [RFC] create a single workqueue for each vm to update vm irq routing table
Date: Thu, 28 Nov 2013 10:53:37 +0100 [thread overview]
Message-ID: <529712A1.8090207@redhat.com> (raw)
In-Reply-To: <5297118C.3050104@cloudius-systems.com>
Il 28/11/2013 10:49, Avi Kivity ha scritto:
> Linux is safe, it does interrupt migration from within the interrupt
> handler. If you do that before the device-specific EOI, you won't get
> another interrupt until programming the MSI is complete.
>
> Is virtio safe? IIRC it can post multiple interrupts without guest acks.
>
> Using call_rcu() is a better solution than srcu IMO. Less code changes,
> consistently faster.
call_rcu() has the problem of rate limiting, too. It wasn't such a
great idea, I think.
The QRCU I linked would work great latency-wise (it has roughly the same
latency of an rwsem but readers are lock-free). However, the locked
operations in the read path would hurt because of cache misses, so it's
not good either.
Paolo
next prev parent reply other threads:[~2013-11-28 9:53 UTC|newest]
Thread overview: 120+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-11-26 12:40 [RFC] create a single workqueue for each vm to update vm irq routing table Zhanghaoyu (A)
2013-11-26 12:40 ` [Qemu-devel] " Zhanghaoyu (A)
2013-11-26 12:47 ` Paolo Bonzini
2013-11-26 12:47 ` [Qemu-devel] " Paolo Bonzini
2013-11-26 12:56 ` Gleb Natapov
2013-11-26 12:56 ` [Qemu-devel] " Gleb Natapov
2013-11-26 13:13 ` Paolo Bonzini
2013-11-26 13:13 ` [Qemu-devel] " Paolo Bonzini
2013-11-28 3:46 ` Zhanghaoyu (A)
2013-11-28 3:46 ` [Qemu-devel] " Zhanghaoyu (A)
2013-11-26 16:05 ` Michael S. Tsirkin
2013-11-26 16:05 ` [Qemu-devel] " Michael S. Tsirkin
2013-11-26 16:14 ` Gleb Natapov
2013-11-26 16:14 ` [Qemu-devel] " Gleb Natapov
2013-11-26 16:24 ` Michael S. Tsirkin
2013-11-26 16:24 ` [Qemu-devel] " Michael S. Tsirkin
2013-11-30 2:46 ` Zhanghaoyu (A)
2013-11-30 2:46 ` [Qemu-devel] " Zhanghaoyu (A)
2013-11-26 13:18 ` Avi Kivity
2013-11-26 13:18 ` [Qemu-devel] " Avi Kivity
2013-11-26 13:47 ` Paolo Bonzini
2013-11-26 13:47 ` [Qemu-devel] " Paolo Bonzini
2013-11-26 14:36 ` Avi Kivity
2013-11-26 14:36 ` [Qemu-devel] " Avi Kivity
2013-11-26 14:46 ` Paolo Bonzini
2013-11-26 14:46 ` Paolo Bonzini
2013-11-26 14:54 ` Avi Kivity
2013-11-26 14:54 ` Avi Kivity
2013-11-26 15:03 ` Gleb Natapov
2013-11-26 15:03 ` Gleb Natapov
2013-11-26 15:20 ` Paolo Bonzini
2013-11-26 15:20 ` Paolo Bonzini
2013-11-26 15:25 ` Avi Kivity
2013-11-26 15:25 ` Avi Kivity
2013-11-26 15:28 ` Paolo Bonzini
2013-11-26 15:28 ` Paolo Bonzini
2013-11-26 15:35 ` Avi Kivity
2013-11-26 15:35 ` Avi Kivity
2013-11-26 15:58 ` Paolo Bonzini
2013-11-26 15:58 ` Paolo Bonzini
2013-11-26 16:06 ` Avi Kivity
2013-11-26 16:06 ` Avi Kivity
2013-11-26 16:11 ` Michael S. Tsirkin
2013-11-26 16:11 ` Michael S. Tsirkin
2013-11-26 16:21 ` Avi Kivity
2013-11-26 16:21 ` Avi Kivity
2013-11-26 16:42 ` Paolo Bonzini
2013-11-26 16:42 ` Paolo Bonzini
2013-11-26 16:08 ` Michael S. Tsirkin
2013-11-26 16:08 ` Michael S. Tsirkin
2013-11-26 16:24 ` Gleb Natapov
2013-11-26 16:24 ` Gleb Natapov
2013-11-26 16:27 ` Avi Kivity
2013-11-26 16:27 ` Avi Kivity
2013-11-26 16:38 ` Gleb Natapov
2013-11-26 16:38 ` Gleb Natapov
2013-11-26 16:28 ` Paolo Bonzini
2013-11-26 16:28 ` Paolo Bonzini
2013-11-26 16:29 ` Avi Kivity
2013-11-26 16:29 ` Avi Kivity
2013-11-26 16:37 ` Gleb Natapov
2013-11-26 16:37 ` Gleb Natapov
2013-11-28 6:27 ` Zhanghaoyu (A)
2013-11-28 6:27 ` Zhanghaoyu (A)
2013-11-28 8:55 ` Paolo Bonzini
2013-11-28 8:55 ` Paolo Bonzini
2013-11-28 9:19 ` Gleb Natapov
2013-11-28 9:19 ` Gleb Natapov
2013-11-28 9:29 ` Paolo Bonzini
2013-11-28 9:29 ` Paolo Bonzini
2013-11-28 9:43 ` Gleb Natapov
2013-11-28 9:43 ` Gleb Natapov
2013-11-28 9:49 ` Avi Kivity
2013-11-28 9:49 ` Avi Kivity
2013-11-28 9:53 ` Paolo Bonzini [this message]
2013-11-28 9:53 ` Paolo Bonzini
2013-11-28 10:16 ` Avi Kivity
2013-11-28 10:16 ` Avi Kivity
2013-11-28 10:40 ` Paolo Bonzini
2013-11-28 10:40 ` Paolo Bonzini
2013-11-28 10:47 ` Avi Kivity
2013-11-28 10:47 ` Avi Kivity
2013-11-28 11:09 ` Gleb Natapov
2013-11-28 11:09 ` Gleb Natapov
2013-11-28 11:10 ` Paolo Bonzini
2013-11-28 11:10 ` Paolo Bonzini
2013-11-28 11:16 ` Avi Kivity
2013-11-28 11:16 ` Avi Kivity
2013-11-28 11:23 ` Gleb Natapov
2013-11-28 11:23 ` Gleb Natapov
2013-11-28 11:31 ` Paolo Bonzini
2013-11-28 11:31 ` Paolo Bonzini
2013-11-28 11:33 ` Avi Kivity
2013-11-28 11:33 ` Avi Kivity
2013-11-28 10:11 ` Gleb Natapov
2013-11-28 10:11 ` Gleb Natapov
2013-11-28 10:12 ` Avi Kivity
2013-11-28 10:12 ` Avi Kivity
2013-11-28 11:02 ` Gleb Natapov
2013-11-28 11:02 ` Gleb Natapov
2013-11-28 11:18 ` Avi Kivity
2013-11-28 11:18 ` Avi Kivity
2013-11-28 11:22 ` Gleb Natapov
2013-11-28 11:22 ` Gleb Natapov
2013-11-28 11:30 ` Avi Kivity
2013-11-28 11:30 ` [Qemu-devel] " Avi Kivity
2013-11-28 11:33 ` Michael S. Tsirkin
2013-11-28 11:33 ` Michael S. Tsirkin
2013-11-28 11:33 ` Gleb Natapov
2013-11-28 11:33 ` Gleb Natapov
2013-11-26 15:24 ` Avi Kivity
2013-11-26 15:24 ` Avi Kivity
2013-11-28 9:14 ` Zhanghaoyu (A)
2013-11-28 9:14 ` Zhanghaoyu (A)
2013-11-28 9:20 ` Gleb Natapov
2013-11-28 9:20 ` Gleb Natapov
2013-11-26 12:48 ` Gleb Natapov
2013-11-26 12:48 ` [Qemu-devel] " Gleb Natapov
2013-11-26 12:50 ` Gleb Natapov
2013-11-26 12:50 ` [Qemu-devel] " Gleb Natapov
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=529712A1.8090207@redhat.com \
--to=pbonzini@redhat.com \
--cc=avi.kivity@gmail.com \
--cc=avi@cloudius-systems.com \
--cc=gleb@minantech.com \
--cc=gleb@redhat.com \
--cc=haoyu.zhang@huawei.com \
--cc=jinxin712@huawei.com \
--cc=kvm@vger.kernel.org \
--cc=luonengjun@huawei.com \
--cc=mst@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=weidong.huang@huawei.com \
--cc=zanghongyong@huawei.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 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.