From: Avi Kivity <avi@redhat.com>
To: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>,
kvm@vger.kernel.org, qemu-devel@nongnu.org
Subject: Re: [patch uq/master 2/2] kvm-all.c: define smp_wmb and use it for coalesced mmio
Date: Mon, 22 Feb 2010 17:08:00 +0200 [thread overview]
Message-ID: <4B829DD0.4010009@redhat.com> (raw)
In-Reply-To: <20100222145739.GC20580@redhat.com>
On 02/22/2010 04:57 PM, Michael S. Tsirkin wrote:
>
>
>>> There is no need (for this case). Older read cannot be reordered with
>>> write, writes are not reordered with other writes, writes by a single
>>> processor are observed in the same order by all processors.
>>>
>>>
>> Well, Linux does use sfence.
>>
> At least on 64 bit it doesnt.
>
Right, I was looking at wmb(), not smp_wmb().
>> I imagine all arches need an instruction. For reads as well.
>>
>> Note, gcc has a __sync_synchronize() builtin that compiles to mfence on
>> x86. We might use that as a baseline for both rmb and wmb, and let each
>> arch override it incrementally.
>>
> This it what my patch did. Note it only works well for recent gcc.
>
Do you know how recent?
--
error compiling committee.c: too many arguments to function
WARNING: multiple messages have this Message-ID (diff)
From: Avi Kivity <avi@redhat.com>
To: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>,
qemu-devel@nongnu.org, kvm@vger.kernel.org
Subject: [Qemu-devel] Re: [patch uq/master 2/2] kvm-all.c: define smp_wmb and use it for coalesced mmio
Date: Mon, 22 Feb 2010 17:08:00 +0200 [thread overview]
Message-ID: <4B829DD0.4010009@redhat.com> (raw)
In-Reply-To: <20100222145739.GC20580@redhat.com>
On 02/22/2010 04:57 PM, Michael S. Tsirkin wrote:
>
>
>>> There is no need (for this case). Older read cannot be reordered with
>>> write, writes are not reordered with other writes, writes by a single
>>> processor are observed in the same order by all processors.
>>>
>>>
>> Well, Linux does use sfence.
>>
> At least on 64 bit it doesnt.
>
Right, I was looking at wmb(), not smp_wmb().
>> I imagine all arches need an instruction. For reads as well.
>>
>> Note, gcc has a __sync_synchronize() builtin that compiles to mfence on
>> x86. We might use that as a baseline for both rmb and wmb, and let each
>> arch override it incrementally.
>>
> This it what my patch did. Note it only works well for recent gcc.
>
Do you know how recent?
--
error compiling committee.c: too many arguments to function
next prev parent reply other threads:[~2010-02-22 15:08 UTC|newest]
Thread overview: 58+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-02-22 13:59 [patch uq/master 0/2] wake iothread on virtio kick / flush_coalesced_mmio smp_wmb Marcelo Tosatti
2010-02-22 13:59 ` [Qemu-devel] " Marcelo Tosatti
2010-02-22 13:59 ` [patch uq/master 1/2] virtio-pci: wake up iothread on VIRTIO_PCI_QUEUE_NOTIFY Marcelo Tosatti
2010-02-22 13:59 ` [Qemu-devel] " Marcelo Tosatti
2010-02-22 14:20 ` Avi Kivity
2010-02-22 14:20 ` [Qemu-devel] " Avi Kivity
2010-02-22 14:29 ` Marcelo Tosatti
2010-02-22 14:29 ` [Qemu-devel] " Marcelo Tosatti
2010-02-22 14:51 ` Avi Kivity
2010-02-22 14:51 ` [Qemu-devel] " Avi Kivity
2010-02-22 15:16 ` Marcelo Tosatti
2010-02-22 15:16 ` [Qemu-devel] " Marcelo Tosatti
2010-02-22 15:29 ` Anthony Liguori
2010-02-22 15:29 ` [Qemu-devel] " Anthony Liguori
2010-02-22 15:32 ` Avi Kivity
2010-02-22 15:32 ` [Qemu-devel] " Avi Kivity
2010-02-22 15:42 ` Anthony Liguori
2010-02-22 15:42 ` [Qemu-devel] " Anthony Liguori
2010-02-22 15:55 ` Avi Kivity
2010-03-12 2:45 ` [patch 0/2] introduce QEMUIOWorker and wake up iothread on virtio-serial-bus notification Marcelo Tosatti
2010-03-12 2:45 ` [Qemu-devel] " Marcelo Tosatti
2010-03-12 2:45 ` [patch 1/2] Pass QEMUIOWorker to qemu_notify_event Marcelo Tosatti
2010-03-12 2:45 ` [Qemu-devel] " Marcelo Tosatti
2010-03-22 21:16 ` Anthony Liguori
2010-03-22 21:16 ` Anthony Liguori
2010-03-25 13:47 ` [Qemu-devel] [patch 0/2] introduce QEMUIOWorker and wake up iothread on virtio-serial-bus notification (v2) Marcelo Tosatti
2010-03-25 13:47 ` [Qemu-devel] [patch 1/2] Pass QEMUIOWorker to qemu_notify_event Marcelo Tosatti
2010-03-25 21:06 ` Paul Brook
2010-03-26 3:55 ` Marcelo Tosatti
2010-03-26 15:23 ` Paul Brook
2010-03-26 15:40 ` Anthony Liguori
2010-03-25 13:47 ` [Qemu-devel] [patch 2/2] virtio-serial-bus: wake up iothread upon guest read notification Marcelo Tosatti
2010-03-12 2:45 ` Marcelo Tosatti
2010-03-12 2:45 ` [Qemu-devel] " Marcelo Tosatti
2010-03-12 5:53 ` Amit Shah
2010-03-12 5:53 ` [Qemu-devel] " Amit Shah
2010-02-22 13:59 ` [patch uq/master 2/2] kvm-all.c: define smp_wmb and use it for coalesced mmio Marcelo Tosatti
2010-02-22 13:59 ` [Qemu-devel] " Marcelo Tosatti
2010-02-22 14:23 ` Avi Kivity
2010-02-22 14:23 ` [Qemu-devel] " Avi Kivity
2010-02-22 14:45 ` Marcelo Tosatti
2010-02-22 14:45 ` [Qemu-devel] " Marcelo Tosatti
2010-02-22 14:57 ` Avi Kivity
2010-02-22 14:57 ` [Qemu-devel] " Avi Kivity
2010-02-22 14:57 ` Michael S. Tsirkin
2010-02-22 14:57 ` [Qemu-devel] " Michael S. Tsirkin
2010-02-22 15:08 ` Avi Kivity [this message]
2010-02-22 15:08 ` Avi Kivity
2010-02-22 15:08 ` Michael S. Tsirkin
2010-02-22 15:08 ` [Qemu-devel] " Michael S. Tsirkin
2010-02-22 15:28 ` Avi Kivity
2010-02-22 15:28 ` [Qemu-devel] " Avi Kivity
2010-02-22 14:44 ` Michael S. Tsirkin
2010-02-22 14:44 ` [Qemu-devel] " Michael S. Tsirkin
2010-02-22 16:57 ` Marcelo Tosatti
2010-02-22 16:57 ` [Qemu-devel] " Marcelo Tosatti
2010-02-22 17:04 ` Avi Kivity
2010-02-22 17:04 ` [Qemu-devel] " Avi Kivity
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=4B829DD0.4010009@redhat.com \
--to=avi@redhat.com \
--cc=kvm@vger.kernel.org \
--cc=mst@redhat.com \
--cc=mtosatti@redhat.com \
--cc=qemu-devel@nongnu.org \
/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.