On 28.05.2012, at 12:40, Avi Kivity wrote:
> On 05/25/2012 06:12 AM, Benjamin Herrenschmidt wrote:
>>
>> BTW. This is a qemu patch, and that hypercall isn't KVM related at all,
>> ie, it's implemented in qemu and is used with or without KVM, so
>> documenting it in the kernel tree makes little sense. Same goes with
>> H_RTAS.
>>
>> I'll add a doc to qemu in my next spin of it.
>>
>
> Depends. How do you detect it exists? Are you detecting kvm, or qemu,
> or the hypercall itself?
The hypercall itself. SLOF is the only user. QEMU provides SLOF. SLOF calls the hypercall. If the hypercall returns "I don't exist", it doesn't exist. :)
> I'd hate us to find ourselves in a maze of disconnected documentation
> with no clear guidelines on when a feature is available and when it is not.
Yeah, but semantically these hypercalls are on the same layer as fw_cfg. So they clearly belong to QEMU. In fact, they're also used when running with emulation.
Alex
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
[KVM ARM]
[KVM ia64]
[KVM ppc]
[Spice Development]
[Libvirt]
[Libvirt Users]
[Linux USB Devel]
[Video for Linux]
[Linux Audio Users]
[Photo]
[Yosemite News]
[Yosemite Photos]
[Linux Kernel]
[Linux SCSI]
[XFree86]