* [PATCH] vhost: get 2% performance improved by reducing spin_lock race in vhost_work_queue
@ 2013-05-20 3:06 Qinchuanyu
2013-05-20 3:38 ` Jason Wang
0 siblings, 1 reply; 6+ messages in thread
From: Qinchuanyu @ 2013-05-20 3:06 UTC (permalink / raw)
To: rusty@rustcorp.com.au, mst@redhat.com, dhowells@redhat.com,
jasowang@redhat.com
Cc: (kvm@vger.kernel.org), (netdev@vger.kernel.org)
Right now the wake_up_process func is included in spin_lock/unlock, but it could be done outside the spin_lock.
I have test it with kernel 3.0.27 and guest suse11-sp2, it provide 2%-3% net performance improved.
Signed-off-by: Chuanyu Qin <qinchuanyu@huawei.com>
--- a/drivers/vhost/vhost.c 2013-05-20 10:36:30.000000000 +0800
+++ b/drivers/vhost/vhost.c 2013-05-20 10:36:54.000000000 +0800
@@ -144,9 +144,10 @@
if (list_empty(&work->node)) {
list_add_tail(&work->node, &dev->work_list);
work->queue_seq++;
+ spin_unlock_irqrestore(&dev->work_lock, flags);
wake_up_process(dev->worker);
- }
- spin_unlock_irqrestore(&dev->work_lock, flags);
+ } else
+ spin_unlock_irqrestore(&dev->work_lock, flags);
}
void vhost_poll_queue(struct vhost_poll *poll)
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] vhost: get 2% performance improved by reducing spin_lock race in vhost_work_queue
2013-05-20 3:06 [PATCH] vhost: get 2% performance improved by reducing spin_lock race in vhost_work_queue Qinchuanyu
@ 2013-05-20 3:38 ` Jason Wang
2013-05-20 4:15 ` Lei Li
2013-05-20 4:22 ` Qinchuanyu
0 siblings, 2 replies; 6+ messages in thread
From: Jason Wang @ 2013-05-20 3:38 UTC (permalink / raw)
To: Qinchuanyu
Cc: rusty@rustcorp.com.au, mst@redhat.com, dhowells@redhat.com,
(kvm@vger.kernel.org), (netdev@vger.kernel.org)
On 05/20/2013 11:06 AM, Qinchuanyu wrote:
> Right now the wake_up_process func is included in spin_lock/unlock, but it could be done outside the spin_lock.
> I have test it with kernel 3.0.27 and guest suse11-sp2, it provide 2%-3% net performance improved.
>
> Signed-off-by: Chuanyu Qin <qinchuanyu@huawei.com>
Make sense to me but need generate a patch against net-next.git or
vhost.git in git.kernel.org.
Btw. How did you test this? Care to share the perf numbers?
Thanks
> mu
> --- a/drivers/vhost/vhost.c 2013-05-20 10:36:30.000000000 +0800
> +++ b/drivers/vhost/vhost.c 2013-05-20 10:36:54.000000000 +0800
> @@ -144,9 +144,10 @@
> if (list_empty(&work->node)) {
> list_add_tail(&work->node, &dev->work_list);
> work->queue_seq++;
> + spin_unlock_irqrestore(&dev->work_lock, flags);
> wake_up_process(dev->worker);
> - }
> - spin_unlock_irqrestore(&dev->work_lock, flags);
> + } else
> + spin_unlock_irqrestore(&dev->work_lock, flags);
> }
>
> void vhost_poll_queue(struct vhost_poll *poll)
> --
> To unsubscribe from this list: send the line "unsubscribe netdev" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] vhost: get 2% performance improved by reducing spin_lock race in vhost_work_queue
2013-05-20 3:38 ` Jason Wang
@ 2013-05-20 4:15 ` Lei Li
2013-05-20 4:22 ` Qinchuanyu
1 sibling, 0 replies; 6+ messages in thread
From: Lei Li @ 2013-05-20 4:15 UTC (permalink / raw)
To: Qinchuanyu
Cc: Jason Wang, rusty@rustcorp.com.au, mst@redhat.com,
dhowells@redhat.com, (kvm@vger.kernel.org),
(netdev@vger.kernel.org)
On 05/20/2013 11:38 AM, Jason Wang wrote:
> On 05/20/2013 11:06 AM, Qinchuanyu wrote:
>> Right now the wake_up_process func is included in spin_lock/unlock, but it could be done outside the spin_lock.
>> I have test it with kernel 3.0.27 and guest suse11-sp2, it provide 2%-3% net performance improved.
>>
>> Signed-off-by: Chuanyu Qin <qinchuanyu@huawei.com>
> Make sense to me but need generate a patch against net-next.git or
> vhost.git in git.kernel.org.
>
> Btw. How did you test this? Care to share the perf numbers?
I wonder how did you get the performance improved number 2%?
>
> Thanks
>> mu
>> --- a/drivers/vhost/vhost.c 2013-05-20 10:36:30.000000000 +0800
>> +++ b/drivers/vhost/vhost.c 2013-05-20 10:36:54.000000000 +0800
>> @@ -144,9 +144,10 @@
>> if (list_empty(&work->node)) {
>> list_add_tail(&work->node, &dev->work_list);
>> work->queue_seq++;
>> + spin_unlock_irqrestore(&dev->work_lock, flags);
>> wake_up_process(dev->worker);
>> - }
>> - spin_unlock_irqrestore(&dev->work_lock, flags);
>> + } else
>> + spin_unlock_irqrestore(&dev->work_lock, flags);
>> }
>>
>> void vhost_poll_queue(struct vhost_poll *poll)
>> --
>> To unsubscribe from this list: send the line "unsubscribe netdev" in
>> the body of a message to majordomo@vger.kernel.org
>> More majordomo info at http://vger.kernel.org/majordomo-info.html
> --
> To unsubscribe from this list: send the line "unsubscribe kvm" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
>
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] vhost: get 2% performance improved by reducing spin_lock race in vhost_work_queue
2013-05-20 3:38 ` Jason Wang
2013-05-20 4:15 ` Lei Li
@ 2013-05-20 4:22 ` Qinchuanyu
2013-05-20 4:46 ` Jason Wang
1 sibling, 1 reply; 6+ messages in thread
From: Qinchuanyu @ 2013-05-20 4:22 UTC (permalink / raw)
To: Jason Wang
Cc: rusty@rustcorp.com.au, mst@redhat.com, dhowells@redhat.com,
(kvm@vger.kernel.org), (netdev@vger.kernel.org), Heguansen
The patch below is base on
https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git/tree/drivers/vhost/vhost.c?id=refs/tags/next-20130517
Signed-off-by: Chuanyu Qin <qinchuanyu@huawei.com>
--- a/drivers/vhost/vhost.c 2013-05-20 11:47:05.000000000 +0800
+++ b/drivers/vhost/vhost.c 2013-05-20 11:48:24.000000000 +0800
@@ -154,9 +154,10 @@
if (list_empty(&work->node)) {
list_add_tail(&work->node, &dev->work_list);
work->queue_seq++;
+ spin_unlock_irqrestore(&dev->work_lock, flags);
wake_up_process(dev->worker);
- }
- spin_unlock_irqrestore(&dev->work_lock, flags);
+ } else
+ spin_unlock_irqrestore(&dev->work_lock, flags);
}
void vhost_poll_queue(struct vhost_poll *poll)
I did the test by using iperf in 10G environment, the test num as below:
orignal modified
thread_num tp(Gbps) vhost(%) | tp(Gbps) vhost(%)
1 9.59 28.82 | 9.59 27.49
8 9.61 32.92 | 9.62 26.77
64 9.58 46.48 | 9.55 38.99
256 9.6 63.7 | 9.6 52.59
The cost of vhost reduced while the throughput is almost unchanged.
On 05/20/2013 11:06 AM, Qinchuanyu wrote:
> Right now the wake_up_process func is included in spin_lock/unlock, but it could be done outside the spin_lock.
> I have test it with kernel 3.0.27 and guest suse11-sp2, it provide 2%-3% net performance improved.
>
> Signed-off-by: Chuanyu Qin <qinchuanyu@huawei.com>
Make sense to me but need generate a patch against net-next.git or
vhost.git in git.kernel.org.
Btw. How did you test this? Care to share the perf numbers?
Thanks
> mu
> --- a/drivers/vhost/vhost.c 2013-05-20 10:36:30.000000000 +0800
> +++ b/drivers/vhost/vhost.c 2013-05-20 10:36:54.000000000 +0800
> @@ -144,9 +144,10 @@
> if (list_empty(&work->node)) {
> list_add_tail(&work->node, &dev->work_list);
> work->queue_seq++;
> + spin_unlock_irqrestore(&dev->work_lock, flags);
> wake_up_process(dev->worker);
> - }
> - spin_unlock_irqrestore(&dev->work_lock, flags);
> + } else
> + spin_unlock_irqrestore(&dev->work_lock, flags);
> }
>
> void vhost_poll_queue(struct vhost_poll *poll)
> --
> To unsubscribe from this list: send the line "unsubscribe netdev" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] vhost: get 2% performance improved by reducing spin_lock race in vhost_work_queue
2013-05-20 4:22 ` Qinchuanyu
@ 2013-05-20 4:46 ` Jason Wang
2013-05-21 2:40 ` Qinchuanyu
0 siblings, 1 reply; 6+ messages in thread
From: Jason Wang @ 2013-05-20 4:46 UTC (permalink / raw)
To: Qinchuanyu
Cc: rusty@rustcorp.com.au, mst@redhat.com, dhowells@redhat.com,
(kvm@vger.kernel.org), (netdev@vger.kernel.org), Heguansen
On 05/20/2013 12:22 PM, Qinchuanyu wrote:
> The patch below is base on
> https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git/tree/drivers/vhost/vhost.c?id=refs/tags/next-20130517
>
> Signed-off-by: Chuanyu Qin <qinchuanyu@huawei.com>
> --- a/drivers/vhost/vhost.c 2013-05-20 11:47:05.000000000 +0800
> +++ b/drivers/vhost/vhost.c 2013-05-20 11:48:24.000000000 +0800
> @@ -154,9 +154,10 @@
> if (list_empty(&work->node)) {
> list_add_tail(&work->node, &dev->work_list);
> work->queue_seq++;
> + spin_unlock_irqrestore(&dev->work_lock, flags);
> wake_up_process(dev->worker);
> - }
> - spin_unlock_irqrestore(&dev->work_lock, flags);
> + } else
> + spin_unlock_irqrestore(&dev->work_lock, flags);
> }
>
> void vhost_poll_queue(struct vhost_poll *poll)
>
> I did the test by using iperf in 10G environment, the test num as below:
> orignal modified
> thread_num tp(Gbps) vhost(%) | tp(Gbps) vhost(%)
> 1 9.59 28.82 | 9.59 27.49
> 8 9.61 32.92 | 9.62 26.77
> 64 9.58 46.48 | 9.55 38.99
> 256 9.6 63.7 | 9.6 52.59
>
> The cost of vhost reduced while the throughput is almost unchanged.
Thanks, and please generate a formal patch based on
Documentation/SubmittingPatches (put the description and perf numbers in
the commit log). Then resubmit it to let the maintainer apply it.
>
> On 05/20/2013 11:06 AM, Qinchuanyu wrote:
>> Right now the wake_up_process func is included in spin_lock/unlock, but it could be done outside the spin_lock.
>> I have test it with kernel 3.0.27 and guest suse11-sp2, it provide 2%-3% net performance improved.
>>
>> Signed-off-by: Chuanyu Qin <qinchuanyu@huawei.com>
> Make sense to me but need generate a patch against net-next.git or
> vhost.git in git.kernel.org.
>
> Btw. How did you test this? Care to share the perf numbers?
>
> Thanks
>> mu
>> --- a/drivers/vhost/vhost.c 2013-05-20 10:36:30.000000000 +0800
>> +++ b/drivers/vhost/vhost.c 2013-05-20 10:36:54.000000000 +0800
>> @@ -144,9 +144,10 @@
>> if (list_empty(&work->node)) {
>> list_add_tail(&work->node, &dev->work_list);
>> work->queue_seq++;
>> + spin_unlock_irqrestore(&dev->work_lock, flags);
>> wake_up_process(dev->worker);
>> - }
>> - spin_unlock_irqrestore(&dev->work_lock, flags);
>> + } else
>> + spin_unlock_irqrestore(&dev->work_lock, flags);
>> }
>>
>> void vhost_poll_queue(struct vhost_poll *poll)
>> --
>> To unsubscribe from this list: send the line "unsubscribe netdev" in
>> the body of a message to majordomo@vger.kernel.org
>> More majordomo info at http://vger.kernel.org/majordomo-info.html
> N�����r��y���b�X��ǧv�^�){.n�+���z�^�)���w*\x1fjg���\x1e�����ݢj/���z�ޖ��2�ޙ���&�)ߡ�a��\x7f��\x1e�G���h�\x0f�j:+v���w�٥
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] vhost: get 2% performance improved by reducing spin_lock race in vhost_work_queue
2013-05-20 4:46 ` Jason Wang
@ 2013-05-21 2:40 ` Qinchuanyu
0 siblings, 0 replies; 6+ messages in thread
From: Qinchuanyu @ 2013-05-21 2:40 UTC (permalink / raw)
To: mst@redhat.com, Jason Wang
Cc: rusty@rustcorp.com.au, dhowells@redhat.com, (kvm@vger.kernel.org),
(netdev@vger.kernel.org), Heguansen
From: Chuanyu Qin <qinchuanyu@huawei.com>
Subject: [PATCH] get 2% or more performance improved by reducing spin_lock race in vhost_work_queue
the wake_up_process func is included by spin_lock/unlock in vhost_work_queue,
but it could be done outside the spin_lock.
I have test it with kernel 3.0.27 and guest suse11-sp2 using iperf, the num as below.
orignal modified
thread_num tp(Gbps) vhost(%) | tp(Gbps) vhost(%)
1 9.59 28.82 | 9.59 27.49
8 9.61 32.92 | 9.62 26.77
64 9.58 46.48 | 9.55 38.99
256 9.6 63.7 | 9.6 52.59
Signed-off-by: Chuanyu Qin <qinchuanyu@huawei.com>
---
drivers/vhost/vhost.c | 5 +++--
1 files changed, 3 insertions(+), 2 deletions(-)
diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c
index 94dbd25..8bee109 100644
--- a/drivers/vhost/vhost.c
+++ b/drivers/vhost/vhost.c
@@ -146,9 +146,10 @@ static inline void vhost_work_queue(struct vhost_dev *dev,
if (list_empty(&work->node)) {
list_add_tail(&work->node, &dev->work_list);
work->queue_seq++;
+ spin_unlock_irqrestore(&dev->work_lock, flags);
wake_up_process(dev->worker);
- }
- spin_unlock_irqrestore(&dev->work_lock, flags);
+ } else
+ spin_unlock_irqrestore(&dev->work_lock, flags);
}
void vhost_poll_queue(struct vhost_poll *poll)
--
1.7.3.1.msysgit.0
> On 05/20/2013 12:22 PM, Qinchuanyu wrote:
> > The patch below is base on
> > https://git.kernel.org/cgit/linux/kernel/git/next/linux-
> next.git/tree/drivers/vhost/vhost.c?id=refs/tags/next-20130517
> >
> > Signed-off-by: Chuanyu Qin <qinchuanyu@huawei.com>
> > --- a/drivers/vhost/vhost.c 2013-05-20 11:47:05.000000000 +0800
> > +++ b/drivers/vhost/vhost.c 2013-05-20 11:48:24.000000000 +0800
> > @@ -154,9 +154,10 @@
> > if (list_empty(&work->node)) {
> > list_add_tail(&work->node, &dev->work_list);
> > work->queue_seq++;
> > + spin_unlock_irqrestore(&dev->work_lock, flags);
> > wake_up_process(dev->worker);
> > - }
> > - spin_unlock_irqrestore(&dev->work_lock, flags);
> > + } else
> > + spin_unlock_irqrestore(&dev->work_lock, flags);
> > }
> >
> > void vhost_poll_queue(struct vhost_poll *poll)
> >
> > I did the test by using iperf in 10G environment, the test num as
> below:
> > orignal modified
> > thread_num tp(Gbps) vhost(%) | tp(Gbps) vhost(%)
> > 1 9.59 28.82 | 9.59 27.49
> > 8 9.61 32.92 | 9.62 26.77
> > 64 9.58 46.48 | 9.55 38.99
> > 256 9.6 63.7 | 9.6 52.59
> >
> > The cost of vhost reduced while the throughput is almost unchanged.
>
> Thanks, and please generate a formal patch based on
> Documentation/SubmittingPatches (put the description and perf numbers
> in the commit log). Then resubmit it to let the maintainer apply it.
^ permalink raw reply related [flat|nested] 6+ messages in thread
end of thread, other threads:[~2013-05-21 2:41 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2013-05-20 3:06 [PATCH] vhost: get 2% performance improved by reducing spin_lock race in vhost_work_queue Qinchuanyu
2013-05-20 3:38 ` Jason Wang
2013-05-20 4:15 ` Lei Li
2013-05-20 4:22 ` Qinchuanyu
2013-05-20 4:46 ` Jason Wang
2013-05-21 2:40 ` Qinchuanyu
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).