On 2019/4/29 10:29, Tangnianyao (ICT) wrote:
Hi, Marc
> On 2019/4/23 18:00, Marc Zyngier wrote:
>
> Hi, Marc
>
>> On Tue, 23 Apr 2019 08:44:17 +0100,
>> "Tangnianyao (ICT)" wrote:
>>>
>>> Hi, Marc
>>
>> [...]
>>
>>> I've learned that there's some implementation problem for the PCIe
>>> cont
On 2019/4/23 18:00, Marc Zyngier wrote:
Hi, Marc
> On Tue, 23 Apr 2019 08:44:17 +0100,
> "Tangnianyao (ICT)" wrote:
>>
>> Hi, Marc
>
> [...]
>
>> I've learned that there's some implementation problem for the PCIe
>> controller of Hi1616, the processor of D05. The PCIe ACS was not
>> impleme
On Tue, 23 Apr 2019 08:44:17 +0100,
"Tangnianyao (ICT)" wrote:
>
> Hi, Marc
[...]
> I've learned that there's some implementation problem for the PCIe
> controller of Hi1616, the processor of D05. The PCIe ACS was not
> implemented properly and D05 doesn't support VM using pcie vf.
My D05 comp
Hi, Marc
On 2019/4/4 18:59, Marc Zyngier wrote:
> On Thu, 04 Apr 2019 11:07:59 +0100,
> "Tangnianyao (ICT)" wrote:
>>
>>
>>
>> On 2019/3/30 17:43, Marc Zyngier wrote:
>>
>> Hi, Marc
>>
>>> On Sat, 30 Mar 2019 08:42:58 +,
>>> "Tangnianyao (ICT)" wrote:
Hi, Marc
On 2019/3/2
On Thu, 04 Apr 2019 11:07:59 +0100,
"Tangnianyao (ICT)" wrote:
>
>
>
> On 2019/3/30 17:43, Marc Zyngier wrote:
>
> Hi, Marc
>
> > On Sat, 30 Mar 2019 08:42:58 +,
> > "Tangnianyao (ICT)" wrote:
> >>
> >> Hi, Marc
> >>
> >> On 2019/3/21 1:02, Marc Zyngier Wrote:
> >>> On Tue, 19 Mar 2019 2
On 2019/3/30 17:43, Marc Zyngier wrote:
Hi, Marc
> On Sat, 30 Mar 2019 08:42:58 +,
> "Tangnianyao (ICT)" wrote:
>>
>> Hi, Marc
>>
>> On 2019/3/21 1:02, Marc Zyngier Wrote:
>>> On Tue, 19 Mar 2019 21:25:47 +0800
>>> "Tangnianyao (ICT)" wrote:
>>>
Hi, all
Using gicv4, when g
On Tue, 19 Mar 2019 21:25:47 +0800
"Tangnianyao (ICT)" wrote:
> Hi, all
>
> Using gicv4, when guest is waiting for irq, it sends wfi and traps to kvm.
> When vlpi is forwarded to PE after its_vpe_deschedule, before halt_poll in
> kvm_vcpu_block, halt_poll may increase latency for this vlpi getti
+cc Marc and Christoffer...
On 2019/3/19 21:25, Tangnianyao (ICT) wrote:
Hi, all
Using gicv4, when guest is waiting for irq, it sends wfi and traps to kvm.
When vlpi is forwarded to PE after its_vpe_deschedule, before halt_poll in
kvm_vcpu_block, halt_poll may increase latency for this vlpi ge
Hi, all
Using gicv4, when guest is waiting for irq, it sends wfi and traps to kvm.
When vlpi is forwarded to PE after its_vpe_deschedule, before halt_poll in
kvm_vcpu_block, halt_poll may increase latency for this vlpi getting to guest.
In halt_poll process, it checks if there's pending irq for vc