Re: [PATCH 1/1] x86/apic: Introduce paravirq irq_domain

[Jan Kiszka]

On 2016-08-17 18:58, Alexander Popov wrote:
> On 17.08.2016 17:36, Jan Kiszka wrote:
>> On 2016-08-15 14:37, Paolo Bonzini wrote:
>>> On 15/08/2016 13:51, Alexander Popov wrote:
>>>> It seems to me that the idea of an irq_domain for interrupts injected
>>>> by a hypervisor is quite generic.
>>>
>>> True, but all of Xen, KVM and VMware use PCI devices for this.
>>
>> So does Jailhouse. We have to have the code anyway because we need to
>> keep Linux alive after taking over control. Thus it is actually easier
>> to reuse the same logic for para-virtualized domains (non-root cells).
> 
> Hello, Jan! Yes, I see.
> 
> I can only say that Xen, KVM, VMware and Jailhouse happily use hypercalls,
> which are a valid interface between a hypervisor and its guests.
> 
> Positive Technologies hypervisor called Gvandra (named after a big Caucasus
> mountain) tries to use only the hypercalls and avoid PCI device emulation
> to become slimmer.

[Hmm, naming something that's supposed to be slim after something that's
rather big...]

BTW, is there a user of this interface already publicly available? You
didn't reference anything in your posting. Generally, infrastructure
extensions without in-tree users aren't well received (in the best case).

Jan

-- 
Siemens AG, Corporate Technology, CT RDA ITP SES-DE
Corporate Competence Center Embedded Linux