On Thu, Oct 13, 2016 at 11:23:44AM +0200, Maxime Coquelin wrote:
> I was going to re-run some PVP benchmark with 0% pkt loss, as I had
> some strange results last week.
>
> Problem is that your series no more apply cleanly due to
> next-virtio's master branch history rewrite.
> Any chance you send
On Thu, Oct 13, 2016 at 01:33:24PM +0800, Yuanhan Liu wrote:
> That would be a clean way to optimize vhost mergeable Rx path:
>
> - you don't touch non-mrg Rx path (well, you may could apply the
> shadow_used_ring trick to it as wel)
I meant "to non-mrg Rx as well".
--yliu
On Wed, Oct 12, 2016 at 12:22:08PM +, Wang, Zhihong wrote:
> > > >> > 3. How many percentage of drop are you seeing?
> > > The testing result:
> > > size (bytes) improvement (%)
> > > 64 3.92
> > > 128 11.51
> > > 256 24.16
> > > 512
Hi Thomas,
On 12 October 2016 at 23:31, Thomas Monjalon
wrote:
> Sorry guys, you lost me in the discussion.
>
> Is there some regression only on ARM?
> Does it need some work specifically on memcpy for ARM,
I don't know if there is common way to improve memcpy on different ARM
hardware. Even t
On 10/13/2016 09:54 AM, Maxime Coquelin wrote:
>
>
> On 10/13/2016 08:02 AM, Wang, Zhihong wrote:
>>> > Yes, that's great effort! With your hardwork, we know what the
>>> bottleneck
>>> > is and how it could be improved.
>>> >
>>> > However, you don't have to do code refactor (merge two code path
On 10/13/2016 08:02 AM, Wang, Zhihong wrote:
>> > Yes, that's great effort! With your hardwork, we know what the bottleneck
>> > is and how it could be improved.
>> >
>> > However, you don't have to do code refactor (merge two code path to one)
>> > to apply those improvements. From what I know,
> -Original Message-
> From: Yuanhan Liu [mailto:yuanhan.liu at linux.intel.com]
> Sent: Thursday, October 13, 2016 1:33 PM
> To: Wang, Zhihong
> Cc: Jianbo Liu ; Thomas Monjalon
> ; Maxime Coquelin
> ; dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v3 0/5]
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Wednesday, October 12, 2016 11:31 PM
> To: Wang, Zhihong
> Cc: Yuanhan Liu ; Jianbo Liu
> ; Maxime Coquelin ;
> dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v3 0/5]
Sorry guys, you lost me in the discussion.
Is there some regression only on ARM?
Does it need some work specifically on memcpy for ARM,
or vhost for ARM?
Who can work on ARM optimization?
More comments below.
2016-10-12 12:22, Wang, Zhihong:
> From: Yuanhan Liu [mailto:yuanhan.liu at linux.intel
> -Original Message-
> From: Yuanhan Liu [mailto:yuanhan.liu at linux.intel.com]
> Sent: Wednesday, October 12, 2016 10:53 AM
> To: Wang, Zhihong ; Jianbo Liu linaro.org>
> Cc: Maxime Coquelin ; dev at dpdk.org; Thomas
> Monjalon
> Subject: Re: [dpdk-dev] [PATCH
On Thu, Sep 22, 2016 at 01:47:45PM +0800, Jianbo Liu wrote:
> On 22 September 2016 at 10:29, Yuanhan Liu
> wrote:
> > On Wed, Sep 21, 2016 at 08:54:11PM +0800, Jianbo Liu wrote:
> >> >> > My setup consists of one host running a guest.
> >> >> > The guest generates as much 64bytes packets as possi
k.org
>> Subject: Re: [dpdk-dev] [PATCH v3 0/5] vhost: optimize enqueue
>>
>> On 10 October 2016 at 10:44, Yuanhan Liu
>> wrote:
>> > On Sun, Oct 09, 2016 at 12:09:07PM +, Wang, Zhihong wrote:
>> >> > > > Tested with testpmd, host: txonly, g
On 10 October 2016 at 10:44, Yuanhan Liu wrote:
> On Sun, Oct 09, 2016 at 12:09:07PM +, Wang, Zhihong wrote:
>> > > > Tested with testpmd, host: txonly, guest: rxonly
>> > > > size (bytes) improvement (%)
>> > > > 644.12
>> > > > 128 6
>> > > > 256
On Sun, Oct 09, 2016 at 12:09:07PM +, Wang, Zhihong wrote:
> > > > Tested with testpmd, host: txonly, guest: rxonly
> > > > size (bytes) improvement (%)
> > > > 644.12
> > > > 128 6
> > > > 256 2.65
> > > > 512 -1.12
> -Original Message-
> From: Jianbo Liu [mailto:jianbo.liu at linaro.org]
> Sent: Monday, October 10, 2016 2:58 PM
> To: Wang, Zhihong
> Cc: Yuanhan Liu ; Maxime Coquelin
> ; dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v3 0/5] vhost: optimize enqueue
>
&g
> -Original Message-
> From: Jianbo Liu [mailto:jianbo.liu at linaro.org]
> Sent: Monday, October 10, 2016 1:32 PM
> To: Yuanhan Liu
> Cc: Wang, Zhihong ; Maxime Coquelin
> ; dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v3 0/5] vhost: optimize enqueue
>
&g
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Wang, Zhihong
> Sent: Wednesday, September 28, 2016 12:45 AM
> To: Yuanhan Liu ; Jianbo Liu
>
> Cc: Maxime Coquelin ; dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v3 0/5] vh
6 1:48 PM
> >> To: Yuanhan Liu
> >> Cc: Wang, Zhihong ; Maxime Coquelin
> >> ; dev at dpdk.org
> >> Subject: Re: [dpdk-dev] [PATCH v3 0/5] vhost: optimize enqueue
> >>
> >> On 22 September 2016 at 10:29, Yuanhan Liu
> >> wrote:
&
> -Original Message-
> From: Yuanhan Liu [mailto:yuanhan.liu at linux.intel.com]
> Sent: Tuesday, September 27, 2016 6:21 PM
> To: Jianbo Liu
> Cc: Wang, Zhihong ; Maxime Coquelin
> ; dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v3 0/5] vhost: optimize enqueu
On 26 September 2016 at 13:37, Luke Gorrie wrote:
> On 22 September 2016 at 11:01, Jianbo Liu wrote:
>>
>> Tested with testpmd, host: txonly, guest: rxonly
>> size (bytes) improvement (%)
>> 644.12
>> 128 6
>> 256 2.65
>> 512
;> Liu ; Maxime Coquelin
>>
>> Subject: Re: [dpdk-dev] [PATCH v3 0/5] vhost: optimize enqueue
>>
>> On 25 September 2016 at 13:41, Wang, Zhihong
>> wrote:
>> >
>> >
>> >> -Original Message-
>> >> From: Thomas
On 25 September 2016 at 13:41, Wang, Zhihong wrote:
>
>
>> -Original Message-
>> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
>> Sent: Friday, September 23, 2016 9:41 PM
>> To: Jianbo Liu
>> Cc: dev at dpdk.org; Wang, Zhihong ; Yuanhan Liu
>> ; Maxime Coquelin
>>
> Th
Hi Thomas,
On 23 September 2016 at 21:41, Thomas Monjalon
wrote:
> 2016-09-23 18:41, Jianbo Liu:
>> On 23 September 2016 at 10:56, Wang, Zhihong
>> wrote:
>> .
>> > This is expected because the 2nd patch is just a baseline and all
>> > optimization
>> > patches are organized in the rest of
On 22 September 2016 at 11:01, Jianbo Liu wrote:
> Tested with testpmd, host: txonly, guest: rxonly
> size (bytes) improvement (%)
> 644.12
> 128 6
> 256 2.65
> 512 -1.12
> 1024 -7.02
>
Have you conside
> -Original Message-
> From: Jianbo Liu [mailto:jianbo.liu at linaro.org]
> Sent: Monday, September 26, 2016 1:39 PM
> To: Wang, Zhihong
> Cc: Thomas Monjalon ; dev at dpdk.org; Yuanhan
> Liu ; Maxime Coquelin
>
> Subject: Re: [dpdk-dev] [PATCH v3 0/5] vhost:
> -Original Message-
> From: Jianbo Liu [mailto:jianbo.liu at linaro.org]
> Sent: Monday, September 26, 2016 1:13 PM
> To: Wang, Zhihong
> Cc: Thomas Monjalon ; dev at dpdk.org; Yuanhan
> Liu ; Maxime Coquelin
>
> Subject: Re: [dpdk-dev] [PATCH v3 0/5] vhost:
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Friday, September 23, 2016 9:41 PM
> To: Jianbo Liu
> Cc: dev at dpdk.org; Wang, Zhihong ; Yuanhan Liu
> ; Maxime Coquelin
>
> Subject: Re: [dpdk-dev] [PATCH v3 0/5]
On 23 September 2016 at 10:56, Wang, Zhihong wrote:
.
> This is expected because the 2nd patch is just a baseline and all optimization
> patches are organized in the rest of this patch set.
>
> I think you can do bottleneck analysis on ARM to see what's slowing down the
> perf, there might be
2016-09-23 18:41, Jianbo Liu:
> On 23 September 2016 at 10:56, Wang, Zhihong
> wrote:
> .
> > This is expected because the 2nd patch is just a baseline and all
> > optimization
> > patches are organized in the rest of this patch set.
> >
> > I think you can do bottleneck analysis on ARM to s
> -Original Message-
> From: Jianbo Liu [mailto:jianbo.liu at linaro.org]
> Sent: Thursday, September 22, 2016 10:42 PM
> To: Wang, Zhihong
> Cc: Yuanhan Liu ; Maxime Coquelin
> ; dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v3 0/5] vhost: optimize enqueue
>
linaro.org]
>> >> Sent: Thursday, September 22, 2016 1:48 PM
>> >> To: Yuanhan Liu
>> >> Cc: Wang, Zhihong ; Maxime Coquelin
>> >> ; dev at dpdk.org
>> >> Subject: Re: [dpdk-dev] [PATCH v3 0/5] vhost: optimize enqueue
>> >>
>>
dev at dpdk.org
>> Subject: Re: [dpdk-dev] [PATCH v3 0/5] vhost: optimize enqueue
>>
>> On 22 September 2016 at 10:29, Yuanhan Liu
>> wrote:
>> > On Wed, Sep 21, 2016 at 08:54:11PM +0800, Jianbo Liu wrote:
>> >> >> > My setup consists of on
On 22 September 2016 at 10:29, Yuanhan Liu
wrote:
> On Wed, Sep 21, 2016 at 08:54:11PM +0800, Jianbo Liu wrote:
>> >> > My setup consists of one host running a guest.
>> >> > The guest generates as much 64bytes packets as possible using
>> >>
>> >> Have you tested with other different packet size
On Wed, Sep 21, 2016 at 08:54:11PM +0800, Jianbo Liu wrote:
> >> > My setup consists of one host running a guest.
> >> > The guest generates as much 64bytes packets as possible using
> >>
> >> Have you tested with other different packet size?
> >> My testing shows that performance is dropping when
> -Original Message-
> From: Jianbo Liu [mailto:jianbo.liu at linaro.org]
> Sent: Thursday, September 22, 2016 5:02 PM
> To: Wang, Zhihong
> Cc: Yuanhan Liu ; Maxime Coquelin
> ; dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v3 0/5] vhost: optimize enqueue
>
> -Original Message-
> From: Jianbo Liu [mailto:jianbo.liu at linaro.org]
> Sent: Thursday, September 22, 2016 1:48 PM
> To: Yuanhan Liu
> Cc: Wang, Zhihong ; Maxime Coquelin
> ; dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v3 0/5] vhost: optimize enqueue
>
> -Original Message-
> From: Jianbo Liu [mailto:jianbo.liu at linaro.org]
> Sent: Wednesday, September 21, 2016 8:54 PM
> To: Wang, Zhihong
> Cc: Maxime Coquelin ; dev at dpdk.org;
> yuanhan.liu at linux.intel.com
> Subject: Re: [dpdk-dev] [PATCH v3 0/5] vhost: opt
yuanhan.liu at linux.intel.com
>> Subject: Re: [dpdk-dev] [PATCH v3 0/5] vhost: optimize enqueue
>>
>> Hi Maxime,
>>
>> On 22 August 2016 at 16:11, Maxime Coquelin
>> wrote:
>> > Hi Zhihong,
>> >
>> > On 08/19/2016 07:43 AM, Zhihong
Hi Maxime,
On 22 August 2016 at 16:11, Maxime Coquelin
wrote:
> Hi Zhihong,
>
> On 08/19/2016 07:43 AM, Zhihong Wang wrote:
>>
>> This patch set optimizes the vhost enqueue function.
>>
...
>
> My setup consists of one host running a guest.
> The guest generates as much 64bytes packets as possi
> -Original Message-
> From: Jianbo Liu [mailto:jianbo.liu at linaro.org]
> Sent: Wednesday, September 21, 2016 4:50 PM
> To: Maxime Coquelin
> Cc: Wang, Zhihong ; dev at dpdk.org;
> yuanhan.liu at linux.intel.com
> Subject: Re: [dpdk-dev] [PATCH v3 0/5] vhost: opt
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Monday, August 22, 2016 6:35 PM
> To: Maxime Coquelin ; Wang, Zhihong
> ; yuanhan.liu at linux.intel.com
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v3 0/5] vh
On Tue, Aug 23, 2016 at 10:43:36AM +, Wang, Zhihong wrote:
> > > I forgot to add that before this series, I think we should first fix the
> > > windows
> > bug.
> > > Else we will need a dedicated fix for the stable branch.
> >
> > Okay I'll try to fix it, though I can't make any promises at
On 08/23/2016 12:43 PM, Wang, Zhihong wrote:
>
>
>> -Original Message-
>> From: Wang, Zhihong
>> Sent: Tuesday, August 23, 2016 10:31 AM
>> To: Maxime Coquelin ; dev at dpdk.org
>> Cc: yuanhan.liu at linux.intel.com
>> Subject: RE: [PATCH v3 0/5] vhost: optimize enqueue
>>
>>
>>
>>> -
> -Original Message-
> From: Wang, Zhihong
> Sent: Tuesday, August 23, 2016 10:31 AM
> To: Maxime Coquelin ; dev at dpdk.org
> Cc: yuanhan.liu at linux.intel.com
> Subject: RE: [PATCH v3 0/5] vhost: optimize enqueue
>
>
>
> > -Original Message-
> > From: Maxime Coquelin [mailto
> -Original Message-
> From: Maxime Coquelin [mailto:maxime.coquelin at redhat.com]
> Sent: Monday, August 22, 2016 6:02 PM
> To: Wang, Zhihong ; dev at dpdk.org
> Cc: yuanhan.liu at linux.intel.com
> Subject: Re: [PATCH v3 0/5] vhost: optimize enqueue
>
>
> On 08/22/2016 10:11 AM, Maxi
> Subject: Re: [PATCH v3 0/5] vhost: optimize enqueue
>
> Hi Zhihong,
>
[...]
> > The main optimization techniques are:
> >
> > 1. Reorder code to reduce CPU pipeline stall cycles.
> >
> > 2. Batch update the used ring for better efficiency.
> >
> > 3. Prefetch descriptor to hide cache latency
2016-08-22 12:01, Maxime Coquelin:
> I forgot to add that before this series, I think we should first fix the
> windows bug.
> Else we will need a dedicated fix for the stable branch.
This is a funny situation :)
If Zhihong had reworked the code without mentioning it is fixing a scenario
with Win
On 08/22/2016 10:11 AM, Maxime Coquelin wrote:
> Hi Zhihong,
>
> On 08/19/2016 07:43 AM, Zhihong Wang wrote:
> > This patch set optimizes the vhost enqueue function.
> >
> > It implements the vhost logic from scratch into a single function
> > designed
> > for high performance and good maintainabi
Hi Zhihong,
On 08/19/2016 07:43 AM, Zhihong Wang wrote:
> This patch set optimizes the vhost enqueue function.
>
> It implements the vhost logic from scratch into a single function designed
> for high performance and good maintainability, and improves CPU efficiency
> significantly by optimizing c
This patch set optimizes the vhost enqueue function.
It implements the vhost logic from scratch into a single function designed
for high performance and good maintainability, and improves CPU efficiency
significantly by optimizing cache access, which means:
* For fast frontends (eg. DPDK virtio
50 matches
Mail list logo
