qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Paolo Bonzini <pbonzini@redhat.com>
To: "Michael S. Tsirkin" <mst@redhat.com>
Cc: "Huangweidong (C)" <weidong.huang@huawei.com>,
	"gleb@redhat.com" <gleb@redhat.com>,
	Radim Krcmar <rkrcmar@redhat.com>,
	"qemu-devel@nongnu.org" <qemu-devel@nongnu.org>,
	"Gonglei (Arei)" <arei.gonglei@huawei.com>,
	"avi.kivity@gmail.com" <avi.kivity@gmail.com>,
	"Herongguang (Stephen)" <herongguang.he@huawei.com>
Subject: Re: [Qemu-devel] [RFC] vhost: Can we change synchronize_rcu to call_rcu in vhost_set_memory() in vhost kernel module?
Date: Mon, 12 May 2014 15:02:03 +0200	[thread overview]
Message-ID: <5370C64B.9080007@redhat.com> (raw)
In-Reply-To: <20140512125323.GA16846@redhat.com>

Il 12/05/2014 14:53, Michael S. Tsirkin ha scritto:
>> > In any case, whether writes synchronize with RCU or bypass it
>> > doesn't change the picture.  In either case, writes are ordered
>> > against each other but not against reads.  RCU does nothing except
>> > preventing dangling pointer accesses.
>> >
>> > Paolo
> This is the only part I don't get.
> RCU will make sure no VCPUs are running, won't it?
> So it's a kind of full barrier.

The actual point where the new value becomes visible is where the write 
happens, not where you do synchronize_rcu.  This is the same for both 
full-copy of the routing table or overwriting the entry.

However, the delay in servicing an older irqfd write can be arbitrary if 
the scheduler decides not to run the irqfd_inject thread.  Such an older 
write might definitely read a newer routing entry.  And that's the 
invalid scenario according to the PCI spec.

Paolo

  reply	other threads:[~2014-05-12 13:02 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-05-09  1:57 [Qemu-devel] [RFC] vhost: Can we change synchronize_rcu to call_rcu in vhost_set_memory() in vhost kernel module? Gonglei (Arei)
2014-05-09  8:14 ` Paolo Bonzini
2014-05-09  9:04   ` Gonglei (Arei)
2014-05-09  9:53     ` Paolo Bonzini
2014-05-12  9:28       ` Gonglei (Arei)
2014-05-12  9:57         ` Paolo Bonzini
2014-05-12 10:08           ` Michael S. Tsirkin
2014-05-12 10:14             ` Paolo Bonzini
2014-05-12 10:18               ` Michael S. Tsirkin
2014-05-12 10:25                 ` Paolo Bonzini
2014-05-12 11:07                   ` Michael S. Tsirkin
2014-05-12 11:46                     ` Paolo Bonzini
2014-05-12 12:12                       ` Michael S. Tsirkin
2014-05-12 12:46                         ` Paolo Bonzini
2014-05-12 12:53                           ` Michael S. Tsirkin
2014-05-12 13:02                             ` Paolo Bonzini [this message]
2014-05-12 10:30           ` Michael S. Tsirkin
2014-05-13  7:03             ` Gonglei (Arei)
2014-05-13  8:21               ` Michael S. Tsirkin
2014-05-13  7:01           ` Gonglei (Arei)

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=5370C64B.9080007@redhat.com \
    --to=pbonzini@redhat.com \
    --cc=arei.gonglei@huawei.com \
    --cc=avi.kivity@gmail.com \
    --cc=gleb@redhat.com \
    --cc=herongguang.he@huawei.com \
    --cc=mst@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=rkrcmar@redhat.com \
    --cc=weidong.huang@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 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).