On 2016/5/24 9:16, David Matlack wrote:
On Mon, May 23, 2016 at 6:13 PM, Yang Zhang wrote:
On 2016/5/24 2:04, David Matlack wrote:
On Sun, May 22, 2016 at 6:26 PM, Yang Zhang
wrote:
On 2016/5/21 2:37, David Matlack wrote:
It's not
On 2016/5/24 9:16, David Matlack wrote:
On Mon, May 23, 2016 at 6:13 PM, Yang Zhang wrote:
On 2016/5/24 2:04, David Matlack wrote:
On Sun, May 22, 2016 at 6:26 PM, Yang Zhang
wrote:
On 2016/5/21 2:37, David Matlack wrote:
It's not obvious to me why polling for a timer interrupt would
2016-05-19 21:57 GMT+08:00 Paolo Bonzini :
>
>
> On 19/05/2016 15:27, Wanpeng Li wrote:
>> From: Wanpeng Li
>>
>> If an emulated lapic timer will fire soon(in the scope of 10us the
>> base of dynamic halt-polling, lower-end of message passing workload
2016-05-19 21:57 GMT+08:00 Paolo Bonzini :
>
>
> On 19/05/2016 15:27, Wanpeng Li wrote:
>> From: Wanpeng Li
>>
>> If an emulated lapic timer will fire soon(in the scope of 10us the
>> base of dynamic halt-polling, lower-end of message passing workload
>> latency TCP_RR's poll time < 10us) we can
On Mon, May 23, 2016 at 6:13 PM, Yang Zhang wrote:
> On 2016/5/24 2:04, David Matlack wrote:
>>
>> On Sun, May 22, 2016 at 6:26 PM, Yang Zhang
>> wrote:
>>>
>>> On 2016/5/21 2:37, David Matlack wrote:
It's not obvious to me why
On Mon, May 23, 2016 at 6:13 PM, Yang Zhang wrote:
> On 2016/5/24 2:04, David Matlack wrote:
>>
>> On Sun, May 22, 2016 at 6:26 PM, Yang Zhang
>> wrote:
>>>
>>> On 2016/5/21 2:37, David Matlack wrote:
It's not obvious to me why polling for a timer interrupt would improve
On 2016/5/24 2:04, David Matlack wrote:
On Sun, May 22, 2016 at 6:26 PM, Yang Zhang wrote:
On 2016/5/21 2:37, David Matlack wrote:
It's not obvious to me why polling for a timer interrupt would improve
context switch latency. Can you explain a bit more?
We have a
On 2016/5/24 2:04, David Matlack wrote:
On Sun, May 22, 2016 at 6:26 PM, Yang Zhang wrote:
On 2016/5/21 2:37, David Matlack wrote:
It's not obvious to me why polling for a timer interrupt would improve
context switch latency. Can you explain a bit more?
We have a workload which using high
On Sun, May 22, 2016 at 6:26 PM, Yang Zhang wrote:
> On 2016/5/21 2:37, David Matlack wrote:
>>
>> It's not obvious to me why polling for a timer interrupt would improve
>> context switch latency. Can you explain a bit more?
>
>
> We have a workload which using high
On Sun, May 22, 2016 at 6:26 PM, Yang Zhang wrote:
> On 2016/5/21 2:37, David Matlack wrote:
>>
>> It's not obvious to me why polling for a timer interrupt would improve
>> context switch latency. Can you explain a bit more?
>
>
> We have a workload which using high resolution timer(less than
On 2016/5/21 2:37, David Matlack wrote:
On Thu, May 19, 2016 at 7:04 PM, Yang Zhang wrote:
On 2016/5/20 2:36, David Matlack wrote:
On Thu, May 19, 2016 at 11:01 AM, David Matlack
wrote:
On Thu, May 19, 2016 at 6:27 AM, Wanpeng Li
On 2016/5/21 2:37, David Matlack wrote:
On Thu, May 19, 2016 at 7:04 PM, Yang Zhang wrote:
On 2016/5/20 2:36, David Matlack wrote:
On Thu, May 19, 2016 at 11:01 AM, David Matlack
wrote:
On Thu, May 19, 2016 at 6:27 AM, Wanpeng Li wrote:
From: Wanpeng Li
If an emulated lapic timer
On Thu, May 19, 2016 at 7:04 PM, Yang Zhang wrote:
> On 2016/5/20 2:36, David Matlack wrote:
>>
>> On Thu, May 19, 2016 at 11:01 AM, David Matlack
>> wrote:
>>>
>>> On Thu, May 19, 2016 at 6:27 AM, Wanpeng Li wrote:
On Thu, May 19, 2016 at 7:04 PM, Yang Zhang wrote:
> On 2016/5/20 2:36, David Matlack wrote:
>>
>> On Thu, May 19, 2016 at 11:01 AM, David Matlack
>> wrote:
>>>
>>> On Thu, May 19, 2016 at 6:27 AM, Wanpeng Li wrote:
From: Wanpeng Li
If an emulated lapic timer will fire
2016-05-20 10:04 GMT+08:00 Yang Zhang :
> On 2016/5/20 2:36, David Matlack wrote:
>>
>> On Thu, May 19, 2016 at 11:01 AM, David Matlack
>> wrote:
>>>
>>> On Thu, May 19, 2016 at 6:27 AM, Wanpeng Li wrote:
From: Wanpeng
2016-05-20 10:04 GMT+08:00 Yang Zhang :
> On 2016/5/20 2:36, David Matlack wrote:
>>
>> On Thu, May 19, 2016 at 11:01 AM, David Matlack
>> wrote:
>>>
>>> On Thu, May 19, 2016 at 6:27 AM, Wanpeng Li wrote:
From: Wanpeng Li
If an emulated lapic timer will fire soon(in the
On 2016/5/20 2:36, David Matlack wrote:
On Thu, May 19, 2016 at 11:01 AM, David Matlack wrote:
On Thu, May 19, 2016 at 6:27 AM, Wanpeng Li wrote:
From: Wanpeng Li
If an emulated lapic timer will fire soon(in the scope of
On 2016/5/20 2:36, David Matlack wrote:
On Thu, May 19, 2016 at 11:01 AM, David Matlack wrote:
On Thu, May 19, 2016 at 6:27 AM, Wanpeng Li wrote:
From: Wanpeng Li
If an emulated lapic timer will fire soon(in the scope of 10us the
base of dynamic halt-polling, lower-end of message passing
On Thu, May 19, 2016 at 11:01 AM, David Matlack wrote:
> On Thu, May 19, 2016 at 6:27 AM, Wanpeng Li wrote:
>> From: Wanpeng Li
>>
>> If an emulated lapic timer will fire soon(in the scope of 10us the
>> base of dynamic
On Thu, May 19, 2016 at 11:01 AM, David Matlack wrote:
> On Thu, May 19, 2016 at 6:27 AM, Wanpeng Li wrote:
>> From: Wanpeng Li
>>
>> If an emulated lapic timer will fire soon(in the scope of 10us the
>> base of dynamic halt-polling, lower-end of message passing workload
>> latency TCP_RR's
On Thu, May 19, 2016 at 6:27 AM, Wanpeng Li wrote:
> From: Wanpeng Li
>
> If an emulated lapic timer will fire soon(in the scope of 10us the
> base of dynamic halt-polling, lower-end of message passing workload
> latency TCP_RR's poll time < 10us) we
On Thu, May 19, 2016 at 6:27 AM, Wanpeng Li wrote:
> From: Wanpeng Li
>
> If an emulated lapic timer will fire soon(in the scope of 10us the
> base of dynamic halt-polling, lower-end of message passing workload
> latency TCP_RR's poll time < 10us) we can treat it as a short halt,
> and poll to
On 05/19/2016 05:06 PM, Paolo Bonzini wrote:
>
>
> On 19/05/2016 17:03, Christian Borntraeger wrote:
Would this work too and be simpler?
Hmm, your patch does only fiddle with the grow/shrink logic (which might
be a good idea independently of this change), but the original patch
On 05/19/2016 05:06 PM, Paolo Bonzini wrote:
>
>
> On 19/05/2016 17:03, Christian Borntraeger wrote:
Would this work too and be simpler?
Hmm, your patch does only fiddle with the grow/shrink logic (which might
be a good idea independently of this change), but the original patch
On 19/05/2016 17:03, Christian Borntraeger wrote:
> > > Would this work too and be simpler?
> > > Hmm, your patch does only fiddle with the grow/shrink logic (which might
> > > be a good idea independently of this change), but the original patch
> > > actually takes into account that we have a
On 19/05/2016 17:03, Christian Borntraeger wrote:
> > > Would this work too and be simpler?
> > > Hmm, your patch does only fiddle with the grow/shrink logic (which might
> > > be a good idea independently of this change), but the original patch
> > > actually takes into account that we have a
On 05/19/2016 04:56 PM, Paolo Bonzini wrote:
>
>
> On 19/05/2016 16:52, Christian Borntraeger wrote:
Would this work too and be simpler?
>> Hmm, your patch does only fiddle with the grow/shrink logic (which might
>> be a good idea independently of this change), but the original patch
>>
On 05/19/2016 04:56 PM, Paolo Bonzini wrote:
>
>
> On 19/05/2016 16:52, Christian Borntraeger wrote:
Would this work too and be simpler?
>> Hmm, your patch does only fiddle with the grow/shrink logic (which might
>> be a good idea independently of this change), but the original patch
>>
On 19/05/2016 16:52, Christian Borntraeger wrote:
>> > Would this work too and be simpler?
> Hmm, your patch does only fiddle with the grow/shrink logic (which might
> be a good idea independently of this change), but the original patch
> actually takes into account that we have a guaranteed
On 19/05/2016 16:52, Christian Borntraeger wrote:
>> > Would this work too and be simpler?
> Hmm, your patch does only fiddle with the grow/shrink logic (which might
> be a good idea independently of this change), but the original patch
> actually takes into account that we have a guaranteed
On 05/19/2016 03:57 PM, Paolo Bonzini wrote:
>
>
> On 19/05/2016 15:27, Wanpeng Li wrote:
>> From: Wanpeng Li
>>
>> If an emulated lapic timer will fire soon(in the scope of 10us the
>> base of dynamic halt-polling, lower-end of message passing workload
>> latency
On 05/19/2016 03:57 PM, Paolo Bonzini wrote:
>
>
> On 19/05/2016 15:27, Wanpeng Li wrote:
>> From: Wanpeng Li
>>
>> If an emulated lapic timer will fire soon(in the scope of 10us the
>> base of dynamic halt-polling, lower-end of message passing workload
>> latency TCP_RR's poll time < 10us) we
On 19/05/2016 15:27, Wanpeng Li wrote:
> From: Wanpeng Li
>
> If an emulated lapic timer will fire soon(in the scope of 10us the
> base of dynamic halt-polling, lower-end of message passing workload
> latency TCP_RR's poll time < 10us) we can treat it as a short halt,
>
On 19/05/2016 15:27, Wanpeng Li wrote:
> From: Wanpeng Li
>
> If an emulated lapic timer will fire soon(in the scope of 10us the
> base of dynamic halt-polling, lower-end of message passing workload
> latency TCP_RR's poll time < 10us) we can treat it as a short halt,
> and poll to wait it
On 05/19/2016 03:27 PM, Wanpeng Li wrote:
> From: Wanpeng Li
>
> If an emulated lapic timer will fire soon(in the scope of 10us the
> base of dynamic halt-polling, lower-end of message passing workload
> latency TCP_RR's poll time < 10us) we can treat it as a short halt,
On 05/19/2016 03:27 PM, Wanpeng Li wrote:
> From: Wanpeng Li
>
> If an emulated lapic timer will fire soon(in the scope of 10us the
> base of dynamic halt-polling, lower-end of message passing workload
> latency TCP_RR's poll time < 10us) we can treat it as a short halt,
> and poll to wait it
From: Wanpeng Li
If an emulated lapic timer will fire soon(in the scope of 10us the
base of dynamic halt-polling, lower-end of message passing workload
latency TCP_RR's poll time < 10us) we can treat it as a short halt,
and poll to wait it fire, the fire callback
From: Wanpeng Li
If an emulated lapic timer will fire soon(in the scope of 10us the
base of dynamic halt-polling, lower-end of message passing workload
latency TCP_RR's poll time < 10us) we can treat it as a short halt,
and poll to wait it fire, the fire callback apic_timer_fn() will set
38 matches
Mail list logo